Section by W. Paul Cockshott and Allin F. Cottrell
The most influential philospher of the last two centuries was Karl Marx. Among philosophers he was unusual both in being a communist and in being an expert in economics. The fact that Marx wrote extensively on economics has led some to call him an economist; but this is misleading. The two principal works on the subject that he published during his lifetime (Marx, 1971  and Marx, 1970 ) were both titled `critiques of political economy.’ They were demolition jobs on political economy, from the standpoint of communist philosophy, which sought to reveal how the categories used by the political economists-money, price, profit, capital-were not fundamental. Instead, Marx argued, these were expressions of historically specific forms of social relations, which, with changes in social relations, could disappear.
Following the counter-revolutions in Eastern Europe and the former Soviet Union, bourgeois thinkers proclaimed that Marx had been proved wrong in practice-that money, price, profit and the rest have indeed been shown to be fundamentals of any modern society. For communists to respond to to this by claiming that the USSR was never socialist does not get us very far: if the USSR did not have a socialist economy, what do we mean by a socialist economy?
An alternative response is to say, Yes, the USSR was a socialist economy of sorts-and as such had great strengths which it is currently unfashionable to emphasize-but there are other possible sorts of socialism which are both more in keeping with the original philosophical conception of Marx and, in the long term, more viable. That is the response we favor, which obliges us to explain what we would advocate as a model socialist commonwealth.
In the following pages we reconstruct a concept of socialism that is at once consistent with the conceptual framework of Marx’s writings on economics and influenced by the lessons of hitherto-existing socialism.1 We differ from most people writing on socialist economics in the years since the collapse of Communism in that we continue to advocate a planned economy. In doing so we are in the company of David Laibman (1992), and before setting out our own conception of socialism in may be useful to register our points of agreement and disagreement with his work, which is distinguished by a good knowledge of how planning systems actually worked.
Laibman distinguishes three phases of development of socialist planning:
We distinguish between project planning, in which the plan focuses resources on strategic sectors/and or projects; and systemic planning, in which all sectors of the economy are brought within a single framework. The latter, in turn, may be subdivided into a command phase and a comprehensive phase. ¼ We have then three overall stages in the development of planning: project; systemic-command; systemic-comprehensive. To each of these corresponds an evolutionary stage in the development of markets, or commodity relations. Our thinking about markets, then, is placed in a historical and social framework; this is essential if actual market processes are to be analysed in a way that avoids entrapment in the ahistorical dogma of `the’ market, or the `free’ market. (Laibman, 1992, p. 62)
Examples of these phases or stages of planning are, for the project stage the USSR in the 1930s, for the systemic command phase the postwar Soviet model (i.e., the stable 5-year plan system in the USSR, say, from the 1950s), and for the systemic-comprehensive phase the Soviet system between the late 1960s and the late 1980s (Laibman, 1992, p. 84).
One of Laibman’s concerns is to explain why, despite the fact that the `historical model of pure socialism ¼ rested on the total absence of commodity relations’ (p. 63), commodity relations persisted in all stages of the actual planned economy. He has a difficult task. His main aim is to criticize market socialism, but he must somehow rationalize the fact that market relations seem to have expanded in the transition between the `command’ and `comprehensive’ phases of planning. He has little difficulty in explaining the persistence of commodity relations during the period of project planning, attributing this to the persistence of non-socialist property relations in agriculture as in the classic Soviet account. This argument, however, suggests that as agriculture was transformed into a system of state farms, and hence brought within the planning system, commodity relations should have tended to disappear.
That market relations did not in fact disappear in the systemic-command phase is explained by Laibman as a function of a relatively low level of socialist consciousness. While, in this phase, the individual enterprise does not have property rights in the means of production it uses, and `in principle acts on behalf of society as a whole,’ nonetheless
consciousness at the enterprise ¼ is limited; people in practice adopt the standpoint of the production unit at which their concrete experience is located. They therefore cannot rely on the political evaluation of their work in relation to the plan; they require `secondary confirmation’of the social utility of their work by means of market exchange. (p. 65)
We find this derivation of the persistence of the commodity form unconvincing. It is more plausible to see people’s consciousness as being shaped by their production relations than vice versa. Only in a society in which goods take on the commodity form will sale seem relevant or meaningful. The problem is to explain why the commodity form itself should have persisted.
This is not difficult to understand if we take into account the familiar litany of problems suffered by the `command’ economy. Citing Nove (1969), Laibman summarizes these problems as follows.
With higher levels of industrialisation and increasing complexity, the strain becomes intolerable as planning bodies try to master and use increasing amounts of information; this results in ever more serious errors and inefficiency. At the level of the micro-unit (enterprise), plan targets set in physical output terms result in distortion; since enterprises are rewarded in terms of plan fulfillment, they have perverse incentives to bias the assortment of output towards either heavier components (when targets are set in terms of weight), or lighter ones (when targets are set in terms of number). (Laibman, 1992, p. 64)
Against this background the maintenance of the commodity form is no mystery. If the sales of intermediate goods-sales in which both buyers and sellers were state enterprises-were `secondary confirmation’ of anything, it was that the direct socialization of labour had not been achieved. The primary confirmation was, of course, the well known deficiencies of the planning system described above. These deficiencies-which became progressively more apparent as actual socialism moved beyond the `project’ phase-are all expressions of an inability of the planning authorities to master the complexity of the structure of expanded reproduction. From this flow both micro-structural imbalances and the aggregated plans that generated the perverse incentives.
Were the planning authorities able to plan down to the detail, say, of the universal product bar codes used in the West, then the tales about producing massive nails to meet plan targets by weight would not arise. In this view of things, the problem was the excessive autonomy of the enterprises. The ability of enterprises to choose the composition of output meant that the system had too many degrees of freedom. No level of `socialist conciousness’ could, without more information from the plan, ensure that the right mix of enterprise output was achieved. Resort to sale, as confirmation of the correctness of the mix, merely reconfirms that labour is still only indirectly social.
In this light the increased reliance on market relations within the system of plan formation, under the later `systemic-comprehensive’ phase, should be seen not as an advance, as Laibman argues, but rather as a precipitate. Given the limited problem-solving capacity of the planners, the surplus coordination precipitates out as market relations. What Laibman fails to ask-and he is by no means alone in this-is what limits the solubility of planning problems?
Why did the regulatory capacity not keep pace with the growing complexity of the economy? Why could the increased mass of economic interrelations not be dissolved in the growth of the planning solution?
Our hypothesis is that it was a case of inadequate development of the forces of production. The forces of production in question were those of information technology. Planning is a material labour process that operates on information (as, for that matter, is market coordination). It does this with determinate means of production. At one stage these may be the letters, ledger forms and mathematical tables operated on by the human computors of the early GOSPLAN. At another level the means of coordination could be supercomputers and the Internet. These different modes of material production have very different productivities: the historic productivity gap between manufacture and automated industry pales when compared to the productivity differentials in information processing brought about by computer technology.
Our argument is that for any given complexity of productive structure there is a minimum threshold of planning sophistication required. Should the planners lack the information processing ability to reach this sophistication, the excess co-ordination must condense or precipitate out as market relations. While for any but the simplest of economies this threshold cannot be achieved by bureaucratic labour, it can be achieved by modern information technology provided that intelligent use is made of the theory of value.
Among economists hostile to central planning, there is a belief that a system of distributed, decentralised decision making is bound to be superior to a centralised one. It is interesting to contrast this approach with that taken by computer scientists and complexity theorists studying parallelism and optimisation problems.
In computing the presumption is the opposite: one tries first to solve problems using a single processor and only if this becomes impossible, does one embark on the risky path of parallel or distributed processing. It has been learned from experience that solving problems in parallel is a great deal harder than it at first seems.
A distributed system of decision making – like the market – might be superior to a centralised one on two distinct axes. It might be faster, and it might come up with better solutions to problems.
It is often true that if more people work in parallel on something, they will finish the job faster. Often, but famously not always, nine women don’t produce a baby in one month. More generally, whether parallelism is worth while depends upon the ratio of work done to communications. Dividing a decision making task up, whether on computers or among humans, soon results in a situation where most machines or people are waiting idly for input from their collaborators. Communications becomes the bottleneck to the whole process. 2Generally, the further the messages have to go, and the more human intervention is needed to send a message the slower the whole process becomes. In a market system, the sending of messages can be very slow, since one of the ways firms `communicate’ is by altering the level of their production, which may change the price level. Messages of this sort take months or years to send.
In a computerised system the same principles apply. Designers of fast machines try to bring all of their components as close together as possible. This limits the time that messages passing at the speed of light will take to get from one part of the machine to another.
A distinct issue is whether the solution arrived at by parallel processes will be as optimal as that arrived at by a single process. The question arises because in parallel decision making, each process makes decisions on the basis of out of date information about what other processes are doing. This is an inevitable consequence of communications delays.
In practical applications where parallelism can not be avoided, such as distributed databases for airline bookings, considerable trouble has to be gone through to make sure that the decision making processes have interlocks which make them, in principle, serialisable. That is to say, to be correct, one has to show that the parallel decision making process is formally equivalent to a single serial process. In the absence of such interlocks, one could get the situation where two agents in distinct towns check if a seat is free. They may both be informed that it is. Each then books it, and issues a ticket, and updates the database with the name of the passenger who has booked it. The end result, is that two passengers are issued with tickets to the same seat, but only the name of one of them is entered in the database.
In the absence of such interlocks, the optimality of decision making can degrade massively. In a recent paper3 Macready and co-workers have shown that as one adds processors to an optimisation problem, one initially finds that the solutions arrived at are marginally better than would be achieved using a single process. However, for any given complexity of problem one hits a distinct phase transition, beyond which the answers arrived at become very much worse than those produced by a single processor. The particular system for which they have demonstrated this property, is that of finding the minimum energy configuration for what is called a `spin glass’. A spin glass is a simplified model of interacting magnetic domains in the presence of an ambient field. The model almost certainly generalises to economic problems. This is both because spin glasses are a simple and well established paradigmatic model for optimsation in interconnected systems, and also because such phase transitions now seem to be a general phenomenon in massive decision or constraint satisfaction problems.4
The labor theory of value is the conceptual foundation of all Marx’s writing on economics. Our claim is that this also provides both a moral philosophy for, and a set of economic policies for, socialism. The irony is that both the philosophy and the policies have been effectively ignored by orthodox `Marxism-Leninism.’ The theory of value has been relegated to the analysis of capitalism, and dismissed as of no relevance to socialism.
One of Marx’s fundamental philosophical axioms was that appearances can be deceptive.5 In astronomy the appearance is of a fixed earth and an orbiting sun, the opposite of the reality. The task of science to uncover the underlying causal mechanisms that give rise to these appearances.
In economics, the appearance is that things are valuable because people are willing to pay money for them. The foundation of scientific political economy with Adam Smith starts out from the recognition that this is an illusion, that there is an underlying cause of prices that is independent of the subjective wishes of purchasers.
The value of a thing is the number of hours of society’s labour needed to make it. (Marx, 1970, pp. 38-39)
It is important to recognize that value exists prior to, and independently of, sale, and thus its appearance as exchange value. At any given time, the labor needed to produce any given object is an objective fact. If a car requires 1000 hours for its production, that is what it has cost society. In a market economy, this social cost will determine its exchange value. But even if the product is never brought to market-as with a Ford engine used within the Ford Motor Company-a portion of society’s labor went into its production and thus it has value. Generalizing, the concept of value, or social labor time, also applies to non-market economies.6
Exchange value is the specific form assumed by value in market economies. In a market economy, the relative prices or exchange values of commodities are determined by their value, their labor content.7 For value to appear as exchange value a large number of independent producers and/or consumers must exist. The labor of these independent producers is unplanned. Only when their products are exchanged does the necessary social labor content of the products get measured.
The measurement performed by the market assesses the necessity of the labor in two senses-whether the commodity was produced efficiently, and whether it was produced in the `right’ quantity. If a firm uses out of date production techniques, it will have wasted social labor; this is reflected in its commodities selling for less than their actual labor content. If too much of a commodity comes onto the market, again social labor has been wasted and the price will be below its actual labor content. In this way the market validates the social necessity of privately performed labor.
Bourgeois authors make much of the information-processing power of the market. This, it is alleged, thanks to the large number of simultaneous processors, can lead to `incredibly fast and accurate decision making.’8 In fact the market is a very inefficient information processor. This inefficiency is inherent in the nature of exchange value.
What information does a price give you about a thing? Consider a TV set: its price gives you one number-$200, say-which at best gives a measure of the relative amount of labor required to produce the commodity. But this labor has actually been of many different types. Some of it was performed directly in the TV factory, some was embodied in the value of the chips, resistors, etc. used to make the TV. Hundreds of parts go into a TV, each of which is itself the product of a complex production process. In addition to labor its production has other costs. The chip factories release volatile solvents into the atmosphere, contributing to pollution. The phosphors used in the screen use rare-earth metals whose stocks are finite. Work on the production lines bonding chips destroys the eyesight of Asian women.
Nearly all this information is lost, condensed down into a price, a single monetary figure. When you buy a TV the only information that the market transmits to you is its price. The complexity of the implications of producing more TVs-or more cars or more foreign holidays-goes far beyond what can be encoded in the price. An implicit recognition of this lies at the heart of the widespread belief that you cannot reduce everything to money.
Moreover, prices only tell you about present conditions. Suppose that in 1994 a computer costs $1000. Thinking this good value, 30 million customers around the world decide to buy one. Manufacturers, in response to demand, place orders for memory and microprocessor chips. It is soon discovered that the chip factories do not have the capacity to meet the demand. The prices quoted for chips were not sufficient to cover the costs of opening new production lines. Delivery times on chips rise to three months, four months, six months. Seeing this, Fujitsu, Oki and Hitachi put up their chip prices and build new factories. In the meantime, the price of computers is put up to $1300. By the time new plants are in production, demand for computers has fallen off, prices collapse, and newly opened factories have to close.
This sort of instability, which occurs on a cyclical basis in most industries, is an inevitable consequence of the limited bandwidth of the channel that passes information in a market economy: price.
The standard bourgeois claim is that socialist central planning entails an enormous suffocating bureacratic apparattus, and that it must eventual choke under information overload9. This myth has been repeated so often, that it is believed even when the facts so evidently contradict it.
Take the word bureaucracy literally, it refers to those working in bureaus or offices. The observable fact about the socialist economies is that they employed far fewer people in bureaus or offices than capitalist economies at a comparable stage of development. Capitalist cities are high-rise, their skylines dominated by office tower blocks. Socialist cites were low rise, dominated by the long sheds of industry. Material production not information processing dominated their economies. In fact it is capitalist economies that are dominated by, choked by a constantly rising overhead of unproductive bureacratic work, for what else is the banking, insurance, sales and marketing that fills the tower blocks?
The bureau was the manufactury of information processing, the locus of a formal subordination of information processing labour. With its subdivision of mental labour it stands to information processing in the same relation as Adam Smith’s pin manufactury to material processing. The next stage in development, the real subordination of labour to machinery is achieved first with the Hollerith tabulator and then with the computer center. What the self acting Mule was to material production, the Hollerith machine, and the IBM was to information processing.
Necessity is not the mother of invention, but of application. All of the essential ingredients of information technology were invented in the early 19th century. In its conceptual sophistication, computational power and engineering elegance, Babbage’s machines of the 1830s and 1840s put ENIAC and Colossus to shame. But despite generous state funding for their R&D there was no commercial need for them, at the then current level of capitalist development. The technology was stillborn, forgotten until this century.
The initial impetus for rebith came again from les raisons d’ êtat: the census for Hollerith and the tabulator, army ballistics for Mauchley and the ENIAC, intelligence gathering for Turing and the Colossus. But the impetus for their general application came from the information processing crisis of mid 20th century capitalism, the deluge of commercial correspondance, the truckloads of cheques cleared daily.
By analysing (in a very conservative way) the information processing costs implicit in a market system in contrast to a centrally a planned system, and examining the laws which govern how the respective costs grow as a function of the scale of the economy, one can both explain why the gathering and communication of information poses such a problem for market economies, and show why it was less of a problem for socialist ones. We explain this in the Appendix rather than in the main body of the text, as the argument is pretty technical.
If labor is taken as the substance of value, then the value of labor itself is always unity. An hour of labor is worth an hour of labor. Since hourly wages vary, and since the goods that can be bought with an hour’s wages invariably cost less than an hour to make, it is clear that either
- workers are systematically cheated by being paid less than the value of their labor. This was the conclusion of Rodbertus;10 or
- wages are not actually the price of labor, but the cost of hire of the ability to work. This was the view adopted by Marx.
Marx’s view has perhaps more logical elegance. Whichever view one adopts, the conclusion is much the same: capitalism leads to workers’ being exploited.
Exploitation is being forced to do unpaid work for others. In some cases unpaid labor appears as such: the work of the slave for the master or a wife for her husband. It may be disguised in terms of love or duty, but that it is unpaid is beyond question. In the case of wage labor, it is only possible to detect exploitation if you know how many minutes’ work are required to produce the goods that can be bought with an hour’s wages. Since this is difficult to work out, workers are unlikely to realize just how much they are exploited.
Although the labor contract between employer and employee is in theory a voluntary one, it is in practice entered into under duress. The employer is in a much stronger position and in practice dictates the terms of employment. The worker often faces the alternative of unemployment. Someone who has been unemployed for a while, or who fears unemployment, will be glad of any job and won’t be too particular about the conditions. The employer does not face the same constraints. There are usually many applicants for each job, so that if some people try to negotiate their rate of wages there will be others ready to undercut them.
These factors conspire to force employees to sell themselves at exploitatively low wage rates. Just how low can be seen by looking at the breakdown of the US working day shown in Table 1. The table records, at five-year intervals, how the value added through labor has been distributed between the workers and their bosses. We can see that as time went on, the number of minutes per hour that workers put in for their bosses-the surplus labor time, in Marxian language-tended to rise. By the end of the period, productive workers were working only about 18 minutes per hour to pay their own wages. The remaining time went to the benefit of the employing class.
|Year||Mins. per hour||Mins. per hour|
|worked for self||worked for|
|Source : Derived from Moseley (1991).|
Not only is capitalist exploitation unjust, it is also conducive to economic inefficiency in the form of the wasting of labor time, human society’s key resource. Consider the drive on the part of capitalist entrepreneurs to minimize their costs of production. They can do this in two main ways: They can try to make their workers work longer hours for lower wages, and/or they can adopt a more efficient technology. To the capitalist it is a matter of indifference how she cuts her costs. If sweated labor is cheaper than adopting new technology, then sweated labor it will have to be. The capitalist buys her labor by the hour and is reluctant to waste it. She employs time and motion study to check that she is making good use of what she has bought. But still, she buys labor cheap: if she did not, there would be no profit in it. Here is the contradiction: what is bought cheap is never truly valued.
The lower wages are, the greater the profit, but when wages are low employers can afford to squander labor. The capitalist is one step above the slaveholder in rationality, but that step can be a small one. For instance, the `navigators’ who built Britain’s railway system in the Victorian era worked with the same tools as Hadrian’s slaves building roads and aqueducts-muscle power, pick and shovel. The one great technical advance in two millennia was the wheel barrow, a Chinese invention. The navvies had it, the slaves did not. The railway was the product of the machine age: it was not beyond the wit of Stevenson or Brunel to design steam-powered mechanical excavators. They did not bother because wage slaves could be had cheap.
Right up through the 1950s in British docks, dockers labored to unload ships with techniques that had not changed since the middle ages. Hired by the day, they did the work of slaves without even the security that went with slavery. It took full employment, strong labor unions and better wages to persuade the capitalist class that it was worth investing in mechanical excavators and containerization. A similar story can be told of any number of sweated trades-garment making, toy making, etc.-where wages are low. Here again, production technology is stagnant, the incentive to innovate low.
One of the criticisms that economic reformers leveled at the old price and wage structure in the USSR was that the low level of wages there led to this same sort of waste of labor. In the USSR wages were kept low and a significant part of people’s incomes came in the form of heavily subsidized housing and public services. In opposition, reformers advocated a change in the price and wage system so that services would cost more while wages would be raised to compensate. They claimed that the higher price of labor would then act as an incentive for innovation.11 The argument is valid, but it does not go far enough. The problem arises because the wage-that is, the price paid for labor, rather than labor time itself-is used in costings. Given this, the results of all comparisons of the costs of different production techniques will be influenced in part by the distribution of income. To avoid this we need an objective measure of the amount of labor used to produce things: their true value.
Marx pointed out the inherent limitations involved in cost calculations under the capitalist system, and how these held back potential developments of technology.
Suppose, then, a machine cost as much as the wages for a year of the 150 men that it displaces, say £ 3,000; this £ 3,000 is by no means the expression in money of the labour added to the object produced by these 150 men before the introduction of the machine, but only of the portion of their years labour which was expended for themselves and represented by their wages. On the other hand, the £ 3,000, the money value of the machine, expresses all the labour expended on its production, no matter in what proportion this labour constitutes wages for the workman, and surplus value for the capitalist. Therefore, though a machine costs as much as the labour power displaced by it costs, yet the labour materialised in it is even then much less than the living labour it replaces.12 (Marx, 1970, p392)
Marx then goes on to define what a rational criterion for the employment of machinery would be.
The use of machinery for the exclusive purpose of cheapening the product, is limited in this way, that less labour must be expended in producing the machinery than is displaced by the employment of the machinery.(Marx, 1970, p392)
This limit would act as an upper bound in any mode of production, but this dgree if rationality can not be achieved in a capitalist economy. To each mode of production there exists its own proper form of economic calculation, its own proper form of economising.
For the capitalist, however, this use is still more limited. Instead of paying for the labour, he pays only the value of the labour power employed; therefore the limit to his using a machine is fixed by the difference between value of the machine and the value of the labour power replaced by it. (Marx, 1970, p392)
As so often in Capital Marx is able to make these critical observations capitalist production, because, at least in his minds eye, he is looking back at it from the standpoint of communism. One can only criticise a system once one can visualise an alternative. Just as Smith’s critique of the unproductive expenditures of the landowning class was an anticipation of a fully bourgeois form of economy, Marx’s criticism here is an anticipation of communism. It is an anticipation in which the fetishistic historically partial form of economic calculation brought allowed by the monetary rationality of the market, is replaced by a direct calculation of social rather than private costs. Here is the esoteric implication of the whole critique of commodity fetishism: this calculation can only be in terms of labour time, of value not price.
It follows from the definition given above that value exists in any society with a social division of labor, which must include socialism. The question is how value appears in a socialist economy. Must it appear indirectly through the mediation of exchange value or can it appear directly as a relation between the labor times?
Marx certainly seemed to think that in a socialist commonwealth value relations would appear transparently as relations of labor time:
Let us now picture to ourselves ¼ a community of free individuals, carrying on their work with the means of production in common, in which the labour-power of all the different individuals is consciously applied as the combined labour-power of the community. All of the characteristics of Robinson[ Crusoe]’s labour are here repeated, but with this difference, that they are social instead of individual. Everything produced by him was exclusively the result of his own personal labour, and therefore only an object of use for himself. The total product of our community is a social product. One portion serves as fresh means of production and remains social. But another portion is consumed by the members as means of subsistence. A distribution of this portion among them is consequently necessary. The mode of this distribution will vary with the productive organization of the community, and the degree of historical development attained by the producers. We will assume, but merely for the sake of a parallel with the production of commodities, that the share of each individual producer in the means of subsistence is determined by his labour-time. Labour-time would, in that case, play a double part. Its apportionment in accordance with a definite social plan maintains the proper proportion between the different kinds of work to be done and the various wants of the community. On the other hand, it also serves as a measure of the portion of the common labour borne by each individual, and of his share in the part of the total product destined for individual consumption. The social relations of the individual producers, with regard both to their labour and to its products, are in this case perfectly simple and intelligible, and that with regard not only to production but also to distribution. (Marx, 1970, pp. 78-79)
This quotation from Marx contains two basic ideas that we develop further in the following pages. First, there is the conception of a socialist system of payment in the form of labor tokens, an egalitarian replacement for the capitalist wages system. Second, we have the conception of a socialist planning system using social labor time as its unit of account and basic principle of costing. It is noteworthy that no actual socialist society of the twentieth century has done as Marx envisaged, namely eliminated money and substituted a labor-time accounting system, yet we shall argue that this is both feasible and highly desirable on grounds of both equity and efficiency. We begin by expanding on the idea of payment in labor tokens.
It was a common assumption of nineteenth-century socialism that people should be paid in labor tokens. We encounter the idea in various forms in Owen, Marx,13 Lassalle, Rodbertus and Proudhon. Debate centred on whether or not this implied a fully planned economy. With the enthusiasm of a pioneer, Owen tried to introduce the principle into England via voluntary co-operatives. Later socialists concluded that Owen’s goal would be attainable only with the complete replacement of the capitalist economy.
Whilst Marx was very complimentary about Owen, he was critical of the schemes of Proudhon and Rodbertus. It is worth considering the Marxian critique of ‘labour money’ schemes; for there may appear to be a tension between the latter critique and Marx’s own proposals. Indeed, the ‘critique of labour money’ is open to a (mis)reading which takes it as critical of any attempt to depart from the market system, towards a direct calculus of labour time. This reading has been made by writers as far apart as Karl Kautsky and Terence Hutchison.
The basic object of Marx and Engels’s critique might be described as a naive socialist’ appropriation of the Ricardian theory of value. If only, the reformers argue, we could impose the condition that all commodities really exchange according to the labour embodied in them, then surely exploitation would be ruled out. Hence the schemes, from John Gray in England, through a long list of English ‘Ricardian socialists’, to Proudhon in France, to Rodbertus in Germany, for enforcing exchange in accordance with labour values. 14 From the standpoint of Marx and Engels, such schemes, however, honourable the intentions of their propagators, represent a Utopian and indeed reactionary attempt to turn back the clock to a word of ,simple commodity production’ and exchange between independent producers owning their own means of production. The labour-money utopians failed to recognize two vital points. First, capitalist exploitation occurs through the exchange of commodities in accordance with their labour values (with the value of the special commodity labour-power determined by the labour content of the workers’ means of subsistence). Secondly, although labour content governs the long-run equilibrium exchange ratios of commodities under capitalism, the mechanism whereby production is continually adjusted in line with changing demand, and in the light of changing technologies, under the market system, relies on the divergence of market prices from their long-run equilibrium values. Such divergences generate differential rates of profit, which in turn guide capital into branches of production where supply is inadequate, and push capital out of branches where supply is excessive, in the classic Smith/Ricardo manner. If such divergence is ruled out by fiat, and the signalling mechanism of market prices is hence disabled, there will be chaos, with shortages and surpluses of specific commodities arising everywhere.
One point which emerges repeatedly in the Marxian critique is this: according to the labour theory of value, it is socially necessary labour time which governs equilibrium prices, and not just ‘raw’ labour content (Marx, 1963: 20-21, 66, 204-205). But in commodity-producing society, what is socially necessary labour emerges only through market competition. Labour is first of all ‘private’ (carried out in independent workshops and enterprises), and it is validated or constituted as social only through commodity exchange. The social necessity of labour has two dimensions. First of all, we are referred to the technical conditions of production and the physical productivity of labour. Inefficient or lazy producers, or those using outmoded technology, will fail to realize a market price in line with their actual labour input, but only with the lesser amount which is defined as ‘necessary’. Secondly, there is a sense in which the social necessity of labour is relative to the prevailing structure of demand. If a certain commodity is overproduced relative to demand, it will fail to realize a price in line with its labour value – even if it is produced with average or better technical efficiency. The proponents of labour money want to shortcircuit this process, to act as if all labour were immediately social. The effects within commodity-producing society cannot but be disastrous.
Now the lesson which Marx and Engels read to the labour-money socialists, concerning the beauties of the supply/demand mechanism under capitalism and the foolishness of the arbitrary fixing of prices in line with actual labour content, are obviously rather pleasing to the critics of socialism. It appears that Kautsky also read the critique of labour money as casting doubt on the Marxian objective of direct calculation in terms of labour content, so that by the 1920s the figure widely regarded as the authoritative guardian of the Marxian legacy in the west had effectively abandoned this central tenet of classical Marxism. 15 From the account of the critique of labour money we have given, the limits of that critique should be apparent. What Marx and Engels are rejecting is the notion of fixing prices according to actual labour content in the context of a commodity-producing economy where production is private. In an economy where the means of production are under communal control, on the other hand, labour does become ‘directly social’, in the sense that it is subordinated to a preestablished central plan. Here the calculation of the labour content of goods is an important element in the planning process. And here the reshuffling of resources in line with changing social needs and priorities does not proceed via the response of profit-seeking firms to divergences between market prices and long-run equilibrium values, so the critique of labour money is simply irrelevant. This is the context for Marx’s suggestion for the distribution of consumer goods through labour tokens.
The significance of labor tokens is that they establish the obligation on all to work by abolishing unearned incomes; they make the economic relations between people transparently obvious; and they are egalitarian, ensuring that all labor is counted as equal. It is the last point that ensured that they were never adopted under the bureaucratic state socialisms of the twentieth century. What ruler or manager was willing to see his work as equal to that of a mere laborer?
The difference between a labor-token system and the hire of labor-power can be shown via some contemporary illustrations.
- Suppose you engage a self-employed plumber to fix the toilet. The plumber will judge how long it will take and quote on that basis. On completion of the job you pay the plumber for parts and labor. You do not purchase his ability to work for a day, you pay for the actual work done. If he does not finish the job he does not get paid-it was up to him to judge how long it would take. Self-employed, he has an incentive to get his estimates right.
- Suppose, on the other hand, you call out a repairman employed by a service company to fix the heating. You are likely to be charged for time actually taken. The service company need have no control over how hard or efficiently the repairman works, as the system of charging means that it can never lose. The company purchases his labor-power at $10 per hour and sells it on to you at $40. In this case you are being re-sold labor-power, not the labor actually performed.
- Finally, suppose that you took out a maintainance contract for $80 per annum. The service company is now selling you the promise of work actually done, labor, and has the responsibility and incentive to ensure that the work is done efficiently and to time.
Payment in labor tokens implies payment for work actually done as in cases 1 and 3. When Owen proposed such payment for artisans, this was unproblematic. Proof of work done was provided by the product delivered to the `labor exchange.’ In a modern economy it implies either a system of piecework, or detailed work study to arrive at estimates of time required under conditions of average skill to perform a task.
How much better off would the average person be under the socialist system of payment? How much would one hour’s labor produce? We estimate that in the US in 1990 an hour’s labor produced goods worth about $25 (see Table 2). This means that payment in terms of labor money would be equivalent to an hourly rate of $25 per hour in 1990 money, or about $875 for a 34.5-hour week. This is of course before tax.
|GDP ($ billion)||5522|
|less Capital consumption||602|
|equals Net value product ($ billion)||4920|
|Number of employees:|
|equals Total productive workers||112.5|
|Average weekly hours||34.5|
|times Weeks per year||50|
|equals Hours per worker-year||1725|
|hence Net value product (billion hours)||194|
|hence Rate of value-creation ($ per hour)||25.4|
|Note : Data from Annual Abstract of US Statistics, and the Bureau of Labor Statistics’ Employment and Earnings Survey .|
According to the Bureau of Labor Statistics figures, the median weekly income for women workers in 1990 was $348. Given an egalitarian pay scale of $875 a week, the majority of the female workforce would have seen their incomes more than doubled. Although men are generally paid about a third more than women, the great majority of men also would benefit from the socialist principle of payment according to labor. The median male wage was $485, so the majority of male workers would see wage increases of more than 75 per cent. What this shows is that the great majority of employees are exploited. The gains that they would make if they were no longer exploited would more than offset any erosion of differentials that they might suffer under an egalitarian pay scheme. The very substantial increase that almost all employees would experience is possible because property income is abolished under a socialist system of payment. Socialism involves employees as a group benefiting at the expense of shareholders and other property owners.
The obvious objection to an egalitarian payment system is that it makes no provision for different types of labor. In capitalist economies relatively skilled or educated labor power is generally better paid. Why?
One explanation for this salary premium is that it compensates for both the expenses of education or training and earnings foregone. The extent to which people in capitalist economies are responsible for the financing their own training varies, but in all cases there is an element of earnings foregone, in that people could earn more-at first, that is-by going straight into employment after completing the basic education than they will receive during the years of additional education. In order to generate a sufficient supply of educated labor power, therefore, the more highly educated workers must be paid a premium once they move into employment.
Just how realistic is this? Is it really a `sacrifice’ to be a student compared, for instance, to leaving school and working on a building site? Compared to many working-class youngsters, students have an easy time. The work is clean. It is not too demanding. There are good social facilities and a rich cultural life. Is this an experience that demands financial compensation in later life?
Even if the compensation argument is an accurate reflection of reality in capitalist countries this does not mean that professional workers should obtain the same sort of differentials in a socialist commonwealth. The costs of education and training then would be borne fully by the state. Not only would the education itself be free, as it has been in Britain, but in addition students could receive a regular wage during their period of study. Study is a valid and socially necessary form of work. It produces skilled labor as its `output.’ It should therefore be rewarded accordingly. So there need be no individual expense or earnings-loss on the part of a student that has to be compensated for. The class system prevents a large part of the population from reaching their full potential. Children grow up in deprived areas without ever realising the opportunities that education presents. Their aspirations stunted from infancy, the majority, with some realism, assume that only menial work is open to them; and who needs an education for that?
This attitude reflects the jobs that children see their parents doing, and these jobs would not immediately change if a revolution in society instituted equal pay. Equality would not raise the educational and cultural levels overnight; but in time the democratic presumption behind it would. Equal pay is a moral statement. It says that one person is worth as much as any other. It says, `Citizens, you are all equal in the eyes of society: you may do different things but you are no longer divided into upper and lower classes.’
Talk of equality of educational opportunity is just hollow cant so long as hard economic reality reminds you that society considers you inferior. Beyond what it buys, pay is a symbol of social status, of esteem and self-esteem. Level pay and you will produce a revolution in self-esteem. Along with the comfort and security it would bring to the mass of people would come a rise in their expectations for themselves and their children.
When we discussed who would gain from socialism, we said that in 1990 the average worker would have had an income of $25 per hour were they paid the full value of their labor, we were not claiming that everybody would be free to spend all of this each week. There would still be taxes. In a socialist commonwealth the level of personal taxation to support education, health services, public investment, scientific research and so on is likely to be higher than at present. As against this, less tax would have to be needed to fund social security in a fully employed socialist economy. But an allocation of national income through the taxation system is fundamentally different from exploitation because the tax system can be subject to democratic control.
In a democracy where the citizens can influence the level of taxation, taxes represent resources that people have consented to allocate to public purposes. By contrast, the distribution of income brought about by the market economy is not, and can never be, the result of democratic decisions. But nor can it be said that our present tax system is fundamentally democratic. The only democratic tax change is one approved by plebiscite.
How can complicated questions like taxes be reduced to simple yes/no questions that the people can vote on? Here again Marx’s economic writings point a way forward. Marx describes how the working day of wage laborers is divided into two parts, one of which they spend working for themselves (to pay their wages) and another during which they work for their employer.
We have seen that the labourer, during one portion of the labour-process, produces ¼ the value of his means of subsistence. ¼ If the value of those necessaries represents on an average the expenditure of six hours’ labour, the workman must on average work for six hours to produce that value. (Marx, 1970, p. 216)
During the second period of the labour-process, that in which his labour is no longer necessary labour, the workman ¼ creates surplus-value which, for the capitalist, has all the charms of creation out of nothing. This portion of the working-day I name surplus labour-time, and to the labour expended during that time, I give the name of surplus-labour. (Marx, 1970, p. 217)
In a socialist society we can still speak of surplus labor. No longer would labor be devoted to producing luxuries for the wealthy, but any net investment by society must come under the heading of surplus. Work to support the non-producers-children, the retired and the invalids-can, on the other hand, only be considered surplus from the most individualistic standpoint. From the point of view of society as a whole this labor is necessary, and, considering a person’s whole life, the hours that she puts in during her working life to support the young, the sick and the old would be be balanced by the times that she herself is supported by others. Nevertheless, while working, we as individuals are not free to consume that portion of the national product that goes to pay pensions or run the hospitals and schools. Taxes must be deducted from our incomes to pay for it.
Issues of taxation could be presented to people in this sort of way: `How many hours are you willing to work per week to support the public health system? It is presently three hours per week; do you want to (1) reduce this by 10 minutes, (2) leave it the same, or (3) increase this by 10 minutes?’ Then people would have something that they could understand. Such questions could readily be put in a referendum so that each year the annual budget was approved by the whole people.
What would be the best form of taxes in a socialist commonwealth? The USSR traditionally raised most of its state revenue from a sales or turnover tax on the produce of nationalized industry. This gave the superficial illusion that people were not heavily taxed, since income tax could be kept down to negligible levels, and it is taxes on income that are most visible. The effect, however, was to depress the price of labor relative to other commodities, with the deleterious effects that we described above. There is a better alternative.
Ironically, in view of the strenous efforts made by Marxists to oppose the `Community Charge’ (better known as the poll tax) under Mrs Thatcher’s Tory goverment, a poll tax would actually be a good tax in a socialist commonwealth. The obvious objection to such a tax under capitalism is that it is regressive. A poll tax of £ 300 a year was 10 percent for a girl earning £ 60 a week, but only 1 percent for a professional man earning £ 2500 a month. If everybody earned the same hourly rate then this objection would fall, provided that those who had retired or were unable to work were exempted.
Socialists argue for progressive income taxes in capitalist countries in order to redistribute income from the rich. Meanwhile right-wing economists argue against income tax on the grounds that it is a disincentive to work. It is interesting how the advocates of social inequality think that the wealthy respond to quite different incentives from the poor. If the rich are to be persuaded to work, it seems that the only thing that they respond to is still greater wealth: hence the paramount importance of reducing taxes on high incomes. When dealing with the poor, in contrast, it is held that there is nothing like the lash of still greater poverty as a work incentive: hence the paramount importance of strictly limiting the benefits to which they are entitled. In a society of huge inequalities, arguments about incentives inevitably have this odor of hypocrisy. But again, stripped from their capitalist context they gain in rationality. If there is no inequality arising from exploitation, then someone who chooses to work only four hours a day would, under a proportional income tax make only half the contribution to the health and education system of someone who works an eight-hour day. Would this be fair?
The advantage of a poll tax in a commonwealth is that it establishes that all have the same obligation to work for the common good before they work for themselves. Once you had done your three hours a day to pay your tax, each additional hour worked would be your own. For every additional hour, you would get goods that had cost society one hour to produce.
The capacity of any society to produce is finite. So too, is the demand for any one particular good. Some goods, e.g. water in a rainy country, can be produced to satisfy our needs with a minimal expenditure of labor. But by definition these goods account for only a small part of the value of a nation’s output. The more valuable part is hard won by labor, our ultimate scarce resource.16
Technology may reduce the labor required for some things, or even abolish whole branches of the division of labor. But as fast as it does this it creates new trades and specialisms, and, by opening up new vistas of the possible, engenders new and more sophisticated tastes. By the standards of the nineteenth-century founders of the socialist movement, the workers of Eastern Europe in 1989 lived a life of plenty: Owen and Lassalle had never heard of CDs and videos.
It is a fact, not of economics but of geology, that 1,200,000,000 Chinese and 700,000,000 Indians are not all going to be able to drive BMWs, or even Chevvies. Since scarcity cannot be imagined away, socialism must have a practical and fair way of dealing with it. Basically, there are two options, the rationing of scarce goods, or a price system of some kind.
Rationing makes good sense for services such as health care, where needs can be determined objectively rather than subjectively. In countries with socialized medicine, decisions about the medical procedures needed by a patient are made by doctors, not patients. The assumption is that doctors are better placed to arrive at an objective assessment of what is wrong with the patient, and thus the treatment needed, than the patients themselves.
Where needs are best judged by the individual, on the other hand, the wisdom of rationing depends on the distribution of income. Rationing is the best way of ensuring that scarce goods are fairly distributed if incomes are unequal, since it prevents the rich cornering the market. In case of food in an emergency, formal rationing will ensure that everyone can get enough to survive. Given the sort of egalitarian income distribution discussed earlier, however, a price system is more efficient than rationing.
Advocates of the market compare it to a system of voting which makes the consumer `sovereign.’ This it does, but as Rodbertus long ago pointed out,17 the consumers and the people are two different groups. Consumers are those with money. Only those who already possess something can have their wants satisfied. The unemployed, with only their unwanted labor to offer, have no votes in this system.
If, however, we first assume a highly egalitarian income distribution this objection to the market would not apply. So long as the market is restricted to consumer goods, there is no reason why it should be incompatible with socialism.
The basic principle of a socialist market in consumer goods can be stated quite simply. All consumer goods are marked with their labor values, i.e. the total amount of social labor which is required to produce them.18 But aside from this, the actual prices (in labor tokens) of consumer goods will be set, so far as possible, at market-clearing levels. Market-clearing prices are prices which balance the supply of goods (previously decided upon when the plan is formulated) and the demand. By definition, these prices avoid manifest shortages and surpluses. The appearance of a shortage (excess demand) will result in a rise in price which will cause consumers to reduce their consumption of the good in question. The available supply will then go to those who are willing to pay the most. The appearance of a surplus will result in a fall in price, encouraging consumers to increase their demands for the item.
Suppose a radio requires 10 hours of labor. It will then be marked with a labor value of 10 hours, but if an excess demand emerges, the price will be raised so as to eliminate the excess demand. Suppose this price happens to be 12 labor tokens. The radio then has a price to labor-value ratio of 12/10, or 1.2. Planners (or their computers) record this ratio for each consumer good. The ratio will vary from product to product, sometimes around 1.0, sometimes above (if the product is in strong demand), and sometimes below (if the product is relatively unpopular). The planners then follow this rule: Increase the target output of goods with a ratio in excess of 1.0, and reduce it for those with a ratio less than 1.0.
The point is that these ratios provide a measure of the effectiveness of social labor in meeting consumers’ needs (production of `use-value,’ in Marx’s terminology) across the different industries. If a product has a ratio of market-clearing price to labor-value above 1.0, this indicates that people are willing to spend more labor tokens on the item (i.e. work more hours to acquire it) than the labor time required to produce it. But this in turn indicates that the labor devoted to producing this product is of above-average `social effectiveness.’ Conversely, if the market-clearing price falls below the labor-value, that tells us that consumers do not `value’ the product at its full value: labor devoted to this good is of below-average effectiveness. Parity, or a ratio of 1.0, is an equilibrium condition: in this case consumers `value’ the product, in terms of their own labor time, at just what it costs society to produce it.19
There are therefore two mechanisms whereby the citizens of a socialist commonwealth can determine the allocation of their combined labor time. At one level, they vote periodically on the allocation of their labor between broadly-defined uses such as consumer goods, investment in means of production, and the health service. At another level, they `vote’ on the allocation of labor within the consumer goods sector via the spending of their labor tokens.
A possible criticism of using labor values in economic calculation is that they cannot take into account environmental costs. This is valid, but the same applies to using money. Endangered species have no money to pay for their own protection. If environmental factors do enter into the accounts of profit and loss it is only indirectly-for example, by having the government levy a carbon tax on fuel to deter atmospheric pollution.
A planned economy could produce the same effect by more direct means. It could set physical targets for carbon consumption at so many million tons a year. These limits can be fed in as constraints to the planning computers, which on the basis of their knowledge of current technology can compute how production must be adjusted to deal with the new conditions. A consequence will be that planned production of carbon-intensive consumer goods is restricted. Coal for domestic heating will rise in price relative to natural gas. This kind of premium would be permanent, not a transitory phenomenon of adjustment, and would, like a carbon tax, provide a source of public revenue.
A few decades ago there was little doubt in the minds of socialists that planning was the wave of the future. This was borne out by the rapid advance of the planned economies, which with sputnik and Gagarin seemed to outpace the muddled inefficiency of the capitalist economies. Today, of course, the picture looks different.
Ever since the 1920s bourgeois economists had been claiming that the problems of economic calculation involved with planing an economy were so complex that they could not be solved. It was claimed that without the feedback mechanisms of the market decision-making would be arbitrary and inefficient. So long as the Soviet economy had a rate of growth in excess of the West these ideas did not seem very plausible. But when that economy became more complex, and growth slowed, these bourgeois criticisms seemed to gain relevance. It does seem plausible that a central planning agency can no longer cope with the myriad detail of a modern economy.
In his book The Economics of Feasible Socialism (1983), Alec Nove argued persuasively that if a planning agency, with all its access to national statistical data, cannot plan effectively, then it is even more hopeless for a collection of decentralized workers’ committees to do it. So are we left with no alternative to the market? Must we content ourselves with the advocacy of worker shareholdings?
We believe not. One of us is a computer scientist, and for the last few of years has been researching the possibilities of using modern computers to solve planning problems. We believe that it can now be conclusively demonstrated that the bourgeois arguments against socialist planning are outdated.20 The problems of calculation that seemed daunting in the past can now be readily handled by powerful computers.
Many people are now familiar with the spreadsheet programs like Lotus-123 that are used on personal computers to prepare company plans. The problem of drawing up a plan for an economy can be thought of as a giant spreadsheet or matrix M. The rows of the the spreadsheet represent the different economic activities, while the columns represent the products used by these activities. If the first row represented electricity production and the second represented oil production then M 1,2 [row 1, col 2] would be the amount of oil used to produce electricity and M 2,1 [row 2, col 1] the amount of electricity used to produce oil. The last column of the spreadsheet will hold the total amount produced by each process-so many tera-kilowatt hours of electricity and so many hundred million barrels of oil, etc. The bottom row of the spreadsheet shows the total inputs of each product used in all the production processes. The problem is to ensure that the total planned output of each product is at least equal to the total planned use of that product. What we know to start off with are the technical properties of the processes: one barrel of oil produces so many kilowatt hours. We also know the stocks of capital goods and means of production at the start of the year. What we must do is allocate these to different production processes in such a way as to meet the above constraint.
This is the problem of balancing the plan. In mathematical terms, it involves the solution of a very large number of simultaneous equations, perhaps 10 million equations if the plan is expressed in full detail. This task was quite beyond Gosplan, the Soviet central planning agency, but we have argued that it is now feasible in a matter of minutes or at worst hours-modest compared to the jobs run by physicists and weather forecasters. The two key requirements are (a) efficient algorithms and data-structures, and (b) current-generation high performance computers. The details of our case are spelled out in Cockshott and Cottrell (1993), Cottrell and Cockshott (1993a, 1993b). Gosplan, by contrast, was able to balance the Soviet plans only in highly aggregated terms; and as Nove (1983, 1986) has argued, this sort of exercise is never sufficient to ensure coherence where it matters, at the very specific level of particular goods.
In addition to producing a balanced plan, the optimization of an economy-wide plan is another highly computation-intensive task. The standard approach to this is to treat it as a linear programming problem. The planners specify a desired set of proportions for the outputs and, via the simplex method,21 compute the allocation of means of production that will maximize the output, in those proportions. The problem with this approach is that the running time of an algorithm based on the simplex method can easily turn out to grow as the cube of the number of industries, or worse. Suppose the plan to be optimized deals with 100,000 distinct products. Then we are talking of some 1,000,000,000,000,000 computer instructions to solve the problem. On a 1960s computer the program would have taken some 30 years to run.22 Soviet economic planners resorted to running smaller linear programs, handling only a few hundred key products. At this size, the equations could be solved. This explains one of the strengths of the Russian economy. It did well on certain key projects, such as the space program, which could be given high priority in the planning process. But even today in the West there just is not the computer power available to apply the same techniques more widely.
One of the authors has developed an alternative plan-optimization algorithm, based on concepts from the field of articial intelligence, whose computational demands are very much less than those of linear programming. The running time of this algorithm, described in detail in Cockshott (1990), is less than proportional to the square of the number of products, making it feasible for very large systems. It has been used to optimize the plan for a model economy of about 4000 industries on a desktop computer (a Sun 3) in about five minutes. Unlike simplex, the alternative algorithm is not guaranteed to find the global maximum-but then, of course, neither is any market mechanism! There is no technical reason why the US could not have a completely planned economy. Each work place would have PCs linked to a network of computers within the enterprise which would in turn be linked to a continent-wide network of supercomputers. The workplace would build up a local spreadsheet of its production capabilities and raw materials requirements. These would be transmitted through the hierarchy of machines which would balance up supplies and demands and draw up plans accordingly.
Using computer networks, rival teams of economists could present to the public several alternative continental plans each of which would provide full employment but directed towards different ends: improving our public transport, investing more in industrial equipment, implementing energy-saving measures, improving housing conditions, etc. People could then vote at a referendum for which of these development plans they wanted, knowing that the various alternatives had been thoroughly costed and proved feasible.
Marx’s critique of capitalism has two main aspects: capitalism is unjust, being based on the exploitation of labor, and-although it was a tremendously progressive system relative to previous modes of production-it is also inefficient, this inefficiency taking the forms of both the `anarchy’ of the market and the systematic squandering of labor. We have attempted to outline a vision of a socialist system that answers both of these issues, using the Marxian labor theory of value as a basis. Our suggestions have, we trust, been bold and controversial. We believe it is better for socialists to be bold than to be timid. In this way socialist ideas present themselves with greater clarity and distinctness; and it should be noted that a bold redistribution of income, of the sort we have proposed via the labor-token system, stands to benefit far more people than a timid redistribution. We hope we have made the point that the labor theory of value, understood in a broad sense, provides a definite moral principle with which to oppose the market order. Lacking such a principle, socialists are reduced to tinkering at the margins, which will never fire the imagination of those who are oppressed and exploited in the present system.
Appendix: information processing under market and plan
In this appendix we examine the laws governing the communications costs of capitalist and socialist modes of production, and show why they are lower in the latter. Communications cost is a measure of work done to centralise or disseminate economic information: we will use the conceptual apparatus of algorithmic information theory (Chaitin, 1982) to measure this cost.
Consider the problem of how fast, and with what communications overhead, an economy can converge on a balanced structure of production. It can be demonstrated that this can be done faster and at less communications cost by a centrally planned system.
It is worth considering initially the dynamics of convergence on a fixed target, since, however limited this example ist, a control system with a faster impulse response will also be faster at tracking a moving target.
Consider an economy E = [A,c,r,w] with n producers each producing distinct products using a vector of production functions A, a well defined vector of final consumption expenditure c that is independent of the prices of the n products, an exogenously given wage rate w and a compatible rate of profit r. Then there exists a possible balanced structure e = [U,p] where U is the commodity flow matrix and p a price vector. The first stage of our argument is to use information theory define a metric in terms of which the proximity of an economy to balance can be specified.
We will assume, as is the case in commercial arithmetic, that all quantities are expressed to some finite precision rather than being real numbers. How much information is required to specify this equilibrium point?
Assuming that we have some efficient binary encoding method and that I(s) is a measure in bits of the information content of the data structure s using this method, then the balanced state can be specified by I(e), or, since the balance is in a sense given in the starting conditions, it can be specified by I(E)+I(ps) where ps is a program to solve an arbitrary system of production equations. In general we have I(e) £ I(E)+I(ps). In the following we will assume that I(e) is specified by I(E)+I(ps).
Let I(x|y) be the conditional or relative information (Chaitin, 1982) of x given y. The conditional information associated with any arbitrary configuration of the economy, k = [Uk, pk], may then be expressed relative to the balanced state, e, as I(k|e). If k is in the neighbourhood of e we should expect that I(k|e) £ I(k). For instance, suppose that we can derive Uk from A and an intensity vector uk which specifies the rate at which each industry operates then
where pu is a program to compute Uk from some A and some uk. Since Uk is a matrix and uk a vector, each of scale n, we can assume that I( Uk) > I(uk).
As the economy nears balance the conditional information required to specify it will shrink, since uk starts to approximate to ue.23 Intuitively we only have to supply the difference vector between the two, and this will require less and less information to encode, the smaller the distance between uk and ue. A similar argument applies to the two price vectors pk and pe. If we assume that the system follows a dynamic law that causes it to converge on balance then we should have the relation I(kt+1|e) < I(kt|e).
We now construct a model of the amount of information that has to be transmitted between the producers of a market economy in order to move it towards balance. We make the simplifying assumptions that all production process take one timestep to operate, and that the whole process evolves synchronously. We assume the process starts just after production has finished, with the economy in some random non-equilibrium state. We further assume that each firm starts out with a given selling price for its product. We also, out of generosity to the capitalist case, ignore the overheads of cheque clearing and payment. We specify the economy as a set of communication sequential processes24, with each firm constituting a distinct process. Each firm i carries out the following procedure.
- It writes to all its suppliers asking them their current prices.
- It replies to all price requests that it gets, quoting its current price pi.
- It opens and reads all price quotes from its suppliers.
- It estimates its current per-unit cost of production.
- It calculates the anticipated profitability of production.
- If this is above r it increases its target production rate ui by some fraction. If profitability is below r a proportionate reduction is made.
- It now calculates how much of each input j is required to sustain that production.
- It sends off to each of its suppliers j, an order for amount Uij of their product.
- It opens all orders that it has received and
- totals them up.
- If the total is greater than the available product it scales down each order proportionately to ensure that what it can supply is fairly distributed among its customers.
- It dispatches the (partially) filled orders to its customers.
- If it has no remaining stocks it increases its selling price by some increasing function of the level of excess orders, while if it has stocks left over it reduces its price by some increasing function of the remaining stock.
- It receives all deliveries of inputs and determines at what scale it can actually proceed with production.
- It commences production for the next period.
Experience with computer models of this type of system indicates that if the readiness of producers to change prices is too great, the system could be grossly unstable. We will assume that the price changes are sufficiently small to ensure that only damped oscillations occur. The condition for movement towards equilibrium is then that over a sufficiently large ensemble of points k in phase space, the mean effect of an iteration of the above procedure is to decrease the mean error for each economic variable by some factor 0 £ g < 1. Under such circumstances, while the convergence time in vector space will clearly follow a logarithmic law-to converge by a factor of D in in vector space will take time of order log1/g(D)-in information space the convergence time will be linear. Thus if at time t the distance from equilibrium is I(kt|e), convergence to within a distance e will take a take a time of order
where d is a constant related to the number of economic variables that alter by a mean factor of g each step. The convergence time in information space, for small e, will thus approximate to a linear function of I(k|e) which we can write as DI(k|e).
We are now in a position to express the communications costs of reducing the conditional entropy of the economy to some level e. Communication takes place at steps 1, 2, 8 and 9c of the procedure. How many messages does each supplier have to send, and how much information must they contain?
Letters through the mail contain much redundant pro forma information: we will assume that this is eliminated and the messages reduced to their bare essentials. The whole of the pro forma will be treated as a single symbol in a limited alphabet of message types. A request for a quote would thus be the pair [R,H] where R is a symbol indicating that the message is a quotation request, and H the home address of the requestor. A quote would be the pair [Q,P] with Q indicating the message is a quote and P being the price. An order would similarly be represented by [O,Uij], and with each delivery would go a dispatch note [N,Uij] indicating the actual amount delivered, where Uij £ Uij.
If we assume that each of n firms has on average m suppliers, the number of messages of each type per iteration of the procedure will be nm. Since we have an alphabet of message types (R, Q, O, N) with cardinality 4, these symbols can be encoded in 2 bits each. We will further assume that (H, P, Uij, Uij) can each be encoded in binary numbers of b bits. We thus obtain an expression for the communications cost of an iteration of 4nm(b+2). Taking into account the number of iterations, the cost of approaching the equilibrium will be 4nm(b+2) DI(k|e).
Let us now contrast this with what would be required in a planned economy. Here the procedure involves two distinct procedures, that followed by the (state-owned) firm and that followed by the planning bureau. The firms do the following:
- In the first time period:
- They send to the planners a message listing their address, their technical input coefficients and their current output stocks.
- They receive instructions from the planners about how much of each of their output is to be sent to each of their users.
- They send the goods with appropriate dispatch notes to their users.
- They receive goods inward, read the dispatch notes and calculate their new production level.
- They commence production.
- They then repeatedly perform the same sequence replacing step 1a with:
- They send to the planners a message giving their current output stocks.
The planning bureau performs the complementary procedure:
- In the first period:
- They read the details of stocks and technical coefficients from all of their producers.
- They compute the equilibrium point e from technical coeffients and the final demand.
- They compute a turnpike path (Dorfman, Samuelson and Solow, 1958) from the current output structure to the equilibrium output structure.
- They send out for firms to make deliveries consistent with moving along that path.
- In the second and subsequent periods:
- They read messages giving the extent to which output targets have been met.
- They compute a turnpike path from the current output structure to the equilibrium output structure.
- They send out for firms to make deliveries consistent with moving along that path.
We have previously shown that with computer technology the steps (b) and (c) can be undertaken in a time that is small relative to the production period (Cockshott 1990, Cockshott and Cottrell 1993).
Comparing the repsective information flows, it is clear that the number of orders and dispatch notes sent per iteration is invariant between the two modes of organisation of production. The only difference is that in the planned case the orders come from the center whereas in the market they come from the customers. These messages will again account for a communications load of 2nm(b+2). The difference is that in the planned system there is no exchange of price information. Instead, on the first iteration there is a transmission of information about stocks and technical coefficients. Since any coefficient takes two numbers to specify, the communications load per firm will be: (1+2m)b. For n firms this approximates to the nm(b+2) that was required to communicate the price data.
The difference comes on subsequent iterations, where, assuming no technical change, there is no need to update the planners’ record of the technology matrix. On i-1 subsequent iterations, the planning system has therefore to exchange only about half as much information as the market system. Furthermore, since the planned economy moves on a turnpike path to equilibrium, its convergence time will be less than that of the market economy. The consequent communications cost is 2nm(b+2)(2 + (i-1)) where i < DI(k|e).
The consequence is that, contrary to Hayek’s claims, the amount of information that would have to be transmitted in a planned system is substantially lower than for a market system. The centralised gathering of information is less onerous than the commercial correspondence required by the market. In addition, the convergence time of the market system is slower. The implication of faster convergence for adaptation to changing rather than stable conditions of production and consumption are obvious.
In addition, it should be noted that in our model for the market, we have ignored any information that has to be sent around the system in order to make payments. In practice, with the sending of invoices, cheques, receipts, clearing of cheques etc., the information flow in the market system is likely to be twice as high as our estimates
- Bland, R., 1981,
- `The allocation of resources by linear programming,’ Scientific American , June.
- Chaitin, G. J. (1982).
- `Algorithmic information theory’. In Encyclopedia of Statistical Sciences, vol. 1. New York: Wiley, pp. 38-41.
- Cockshott, W. P., 1990,
- `Application of artificial intelligence techniques to economic planning,’ Future Computer Systems , vol. 2, no. 4, pp. 429-43.
- Cockshott, W. P.
- and Cottrell, A., 1989, `Labour value and socialist economic calculation,’ Economy and Society , vol. 18, February, pp. 71-99.
- Cockshott, W. P.
- and Cottrell, A., 1993, Towards a New Socialism , Nottingham, England: Spokesman.
- Cottrell, A.
- and Cockshott, W. P., 1993a, `Calculation, complexity and planning: the socialist calculation debate once again,’ Review of Political Economy , vol. 5, no. 2, pp. 73-112.
- Cottrell, A.
- and Cockshott, W. P., 1993b, `Socialist planning after the collapse of the Soviet Union,’ Revue Européene des Sciences Sociales , tome XXXI, pp. 167-85.
- Delphy, C., 1984,
- Close to Home: A Materialist Analysis of Women’s Oppression , Hutchinson.
- Farjoun, E.
- and Machover, M., 1983, Laws of Chaos , London: Verso.
- Hayek, F. A. (1945).
- `The use of knowledge in society’. American Economic Review , vol. 35, pp. 519-30.
- Hayek, F. A. (1955).
- The Counter-Revolution of Science . New York: The Free Press.
- Hoare, C. A. R. 1978
- `Communicating Sequential Processes’, Comm. ACM, Vol 21, pp 666-677.
- Hutchison, T. W. 1981
- The politics and philosophy of economics: Marxians, Keynesians and Austrians. New York: New York University Press.
- Kautsky, K. 1902
- The social revolution. Chicago: Charles Kerr.
- Kautsky, K. 1925
- The labour revolution. London: George Allen and Unwin.
- Laibman, D., 1992,
- `Market and plan: the evolution of socialist social structures in history and theory,’ Science & Society , vol. 56, no. 1, pp. 60-91.
- Macready, W. 1996
- and Siapas, A.G., Kauffman, S. A. `Criticality and parallelism in combinatorial optimisation’, Science Vol 271, 5 Jan 1996, pp56-58.
- Marx, K. 1963
- The poverty of philosophy, New York: International Publishers.
- Marx, K., 1970,
- Capital, A Critique of Political Economy , Volume I, Moscow: Progress Publishers (first published 1864).
- Marx, K., 1971,
- A Contribution to the Critique of Political Economy , Moscow: Progress Publishers ( first published 1859).
- Marx, K., 1971a
- Capital, A Critique of Political Economy , Volume III, Moscow: Progress Publishers (first published 1894).
- Marx, K., 1974,
- `Critique of the Gotha Programme,’ in The First International and After (Political Writings, Volume 3), ed. D. Fernbach, Harmondsworth: Penguin (written 1875).
- Moseley, F., 1991,
- The Falling Rate of Profit in the Postwar United States Economy , New York: Macmillan.
- Nove, A., 1983,
- The Economics of Feasible Socialism , London: George Allen and Unwin.
- Nove, A., 1986,
- The Soviet Economic System , third edition, Boston: Unwin Hyman.
- Prosser, P., 1996
- `An empirical study of phase transitions in binary constraint satisfaction models’, Artificial Intelligence, Vol 81, Nos 1..2, March 1996, pp81-109.
- Rodbertus, K., 1904,
- Das Kapital , Paris: Griard and Brière.
- Smith, A., 1937,
- The Wealth of Nations , New York: The Modern Library (first published 1776).
- Sraffa, P., 1960,
- Production of Commodities by Means of Commodities , Cambridge: Cambridge University Press.
- Steedman, I., 1977,
- Marx after Sraffa , London: New Left Books.
- Stone, H., 1987
- High Performance Computer Architecture, Addison Wesley.
1 A fuller exposition of many of the ideas presented here may be found in our recent book, Towards a New Socialism (Cockshott and Cottrell, 1993).
2 A good discussion of this is given in Stone 1987 Chap 6
3 Macready 96.
4 For another example see Prosser 96.
5 `That in their appearance things often represent themselves in inverted form is pretty well known in every science except political economy’ (Marx, 1970, p. 537).
6 `Under the rural patriarchal system of production, when spinner and weaver lived under the same roof-the women of the family spinning and the men weaving, say for the requirements of the family-yarn and linen were social products and spinning and weaving social labor within the framework of the family. But their social character did not appear in the form of yarn becoming a universal equivalent for linen’ (Marx, 1971, p. 33). For a good analysis of this case see also Delphy (1984).
7 In the posthumously published volume III of Capital (Marx, 1971a), Marx qualified this with the idea that in a capitalist economy what he called `prices of production’ operate. These prices involve goods whose production is capital intensive selling above their values. The theory of prices of production was further developed by the neo-ricardian economist Sraffa (1960) and has been used by critics of Marx to try to invalidate his analysis of exploitation (Steedman, 1977). The theory of prices of production is premissed on equality of rates of profit across branches of production. It has been demonstrated by Farjoun and Machover (1983) that this premise does not hold empirically, and that the predictions of the labor theory of value as described in Volume I of Capital offer a better approximation to reality.
8 Senator A. Gore, Scientific American , September 1991.
9 For instance see Hayek 1945 and 1955.
10 Karl Rodbertus, an influential socialist theorist of the 19th century, and founder along with Ferdinand Lassalle of `state socialism.’
11 Once in power, we find the `reformers’ follow the more familiar policy of raising prices, but not wages to compensate.
12 “These mute agents (machines) are always the produce of much less labour than that which they displace, even when they are of the same money value” (Ricardo, 1. C., p. 40)
13 Besides the quotation from Capital given above, the Critique of the Gotha Programme contains a particularly clear account of the idea: `[T]he individual producer gets back from society-after the deductions-exactly what he has given to it. What he has given it is his individual quantum of labour. For instance, the social working day consists of the sum of the individual hours of work. The individual labour time of the individual producer thus constitutes his contribution to the social working day, his share of it. Society gives him a certificate stating that he has done such and such an amount of work (after the labour done for the comunal fund has been deducted), and with this certificate he can withdraw from the social supply of means of consumption as much as costs an equivalent amount of labour. The same amount of labour he has given to society in one form, he receives back in another’ (Marx, 1974, p. 346).
14 Marx criticizes Proudhon’s scheme in his Poverty of philosophy (  1963), and deals with John Gray in his 1859 Contribution to the critique of political economy (the relevant section of which is reprinted as an Appendix to Marx, 1963), while Engels tackles Rodbertus’s variant in his 1884 Preface to the first German edition of The povertv of philosophy (again, in Marx, 1963). Between Marx in 1847 and Engels in 1884 we find a consistent line of attack on such proposals.
15 In his book The social revolution (1902: 129-33), Kautsky offers a brief and rather ambiguous discussion of the ‘law of value’ and socialism, which combines state ments of the classical Marxian theses with strangely incongruous comments on the ‘indispensability’ of money. In his later work, The labour revolution (1925: 261-70) the formulations of Marx and Engels are dropped in favour of a general argument for the necessity of money and prices. This argument appears to owe something to the ‘critique of labour money’ already discussed; it also draws on the idea that the measurement of labour content is impracticable – it ‘could not be achieved by the most complicated State machinery imaginable’ (p. 267).
16 `The annual labor of every nation is the fund which originally supplies it with all the necessaries and conveniences of life which it annually consumes, and which consists always either in the immediate produce of that labor, or in what is purchased with that produce from other nations.’-Adam Smith.
17 `In a social state of this description’ [i.e., a capitalist one] `people produce not with a view to satisfying the needs of labour, but the needs of possession; in other words, they produce for those who possess.’ (Rodbertus, 1904, p. 161)
18 In normal commercial practice, entrepreneurs going into business decide on the selling price for their product by adding up the costs of wages plus raw materials and adding `normal’ profits, all in money terms. They can determine the prices of raw materials from trade catalogs, and the wages they will have to pay by scanning the job adverts of the local paper. The equivalent calculations in terms of labor time are in principle even easier. If catalogs are published stating how much labor went into each raw material, then all that has to be done is add these costs in terms of indirect labor to the number of hours of direct labor used. None of this requires any computational technology unavailable in Marx’s day-ledgers, printed catalogues and clerks with training in arithmetic would have been enough. Today, of course it could be done using computer networks, which could put the catalogues on-line and update them daily. For an analysis of the computer resources needed see Cockshott and Cottrell (1989).
19 This means that the objective of socialist retail markets should be to run at break even level, making neither a profit nor a loss; the goods being sold off cheap compensate for those sold at a premium.
20 For a longer presentation of the argument see Cockshott (1990), Cockshott and Cottrell (1989, 1993).
21 See Bland (1981).
22 We mention a 1960s computer here because the mainframes available to the Soviet planners were mostly based on Western designs of this period.
23 Note that this information measure of the distance from balance, based on a sum of logarithms, differs from a simple Euclidean measure, based on a sum of squares. The information measure is more sensitive to a multiplicity of small errors than to one large error. Because of the equivalence between information and entropy it also measures the conditional entropy of the system.
24 See Hoare 1978.
File translated from TEX by TTH, version 1.30.