Do We Need the Market?

Dear Editors,

Thank you for publishing the thoughtful and fair-minded review of my book, From Marx to Mises, in the June 1993 issue of the Socialist Standard. Within the space permitted, I will here reply only to Robin Cox's most important point: his approach to the economic calculation problem.

Decision makers in any industrial system using advanced technology are faced with choices involving comparisons of specific quantities of different factors. These comparisons can never be made in kind.

If you can get the same result with one gallon of oil as you can get with two gallons of oil, it is preferable to use one gallon of oil and save the second gallon for other uses. That purely technical calculation can be done in kind. But faced with a choice where one gallon of oil can be saved by using more of some other resource, no decision can be made in kind. Is the saving of one gallon of oil greater or less than the increased use of some quantity of an other resource (say, five minutes of maintenance labor, or a few days' shorter life of a machine through increased wear and tear, or one ounce of some mineral employed in a chemical reaction)? If such comparisons have to be made, then we need to be able to reduce different factors to common units of cost.

Your review seems to imply that no such comparisons would any longer have to be made under socialism. In From Marx to Mises I give several examples, and many more can be found in the economic literature, or simply by consulting one's own knowledge of economic history or of places where one has worked. If we really do not need to make such comparisons, then the whole Misesian case collapses, and your remarks about stock control are unnecessary. But I maintain that such comparisons have to be made day by day in any unit of industrial production, no matter what the social institutions.

Robin Cox asks us to suppose that the demand for good X increases, and that the current output of resource A is "insufficient to meet this increased demand". He says that "the sensible solution would be to search for some more abundant resource (B) that could substitute for A. At the same time the falling stock of A would constrain the multifarious users of A to economise on it".

I have three comments:

1. Most commonly, every one of the types of resources used in producing X would be simultaneously used in the production of thousands of other things (let us collectively call theseY). So if the output of X is increased, that of Y must be reduced. The current output of A could itself be increased (as well as being withdrawn from production of Y), but any increase in output of A would involve costs (reductions in output elsewhere). Typically, output of A should be increased somewhat, as well as reducing output of Y. Both these developments reduce output of some goods.

2. In the market, people will automatically switch to using more substitutes for A, but the exact degree of this substitution will be different in all the different lines of production which use A. This can be left to the judgement of the people in those different lines of production because there is a public and objective specification of how scarce A is: the price of A. Robin Cox's statement that searching for B, a substitute for A, is "the sensible solution" over simplifies the problem. Aside from the fact that reallocation of resources has to occur immediately, while this "searching" is still going on, such searching occurs all the time, with respect to all factors, and B, like A, will most likely be a factor used in producing many other goods, with a consequent reduction in output of those other goods if more of B is used to make X. The only practicable way to tell how "abundant" B is, by comparison with A, is to look at their relative prices. And B is almost certainly not a perfect substitute for A, so it will still be up to thousands of separate, local decision-makers to determine whether any technical superiority of A is outweighed, in each particular case, by the comparative cheapness of B. This again requires prices (or some other cost indexes) of A and B.

3. Robin Cox states that the falling stocks of A would constrain users of A to economize on A. Are all users simply told that total stocks of A have gone down, and relied upon to reduce their use of it? The stocks might have fallen for many different reasons, so the users all have to act on their theories as to the reasons for the fall; the users have to estimate by precisely how much to cut their rate of use, bearing in mind that such adjustments are always costly (is the use of A to be cut so much, for instance, that any machine which continuously requires inputs of A should be immediately junked?). How does each user of some quantity of A, out of, say, fifty thousand users, knowing that each of the other 49,999 users are making a similar decision, which perhaps ought to be different in every case, determine his response to the news that stocks of A have fallen by ten per cent? If all the users reduce their use of A by the amount they judge best, there is no guarantee that the outcome is the appropriate total reduction in use. Again, a price for A can do things which news about its total stock cannot do. We cannot do without some general unit of cost, to compare the costs of aggregates of different factors, and we do not know of any feasible unit other than prices denominated in money.

DAVID RAMSAY STEELE, Chicago.

Reply

You may not be able to conceive of production without money and prices, but we can. The definitive answer to your supposed "economic calculation problem" is a (largely) self-regulating system of stock control in which calculations are made in kind rather than in terms of a common unit like money. A self-regulating system of stock control will permit producers in a socialist society (workplace councils, industry councils etc) to ascertain more or less immediately the availability of stocks of any particular item throughout the system; the communications technology to enable this to happen is already in place. Given this, your assertion that the "only practicable way to tell how 'abundant' B is, by comparison with A, is to look at the relative prices" is absurd.

'Abundance' is a relationship between supply and demand, where the former exceeds the latter. In socialism a buffer of surplus stock for any particular item, whether a consumer or a producer good, can be produced, to allow for future fluctuations in the demand for that item, and to provide an adequate response time for any necessary adjustments.

Achieving 'abundance' can be understood as the maintenance of an adequate buffer of stock in the light of extrapolated trends in demand. The relative abundance or scarcity of a good would be indicated by how easy or difficult it was to maintain such an adequate buffer stock in the face of a demand trend (upward, static, downward). It will thus be possible to choose how to combine different factors for production, and whether to use one rather than another, on the basis of their relative abundance/scarcity. By following the rule of using the minimum necessary amounts of the least abundant factors it will be possible to ensure their efficient allocation. Money as a "general unit of cost" would not come into it.

In further asserting that "if the output of X is increased, output of Y must be reduced" you are begging the question at issue, which is precisely whether or not resources are and always will be scarce. It is to assume that society's resources are fully streiihed and that there are no reserves to draw upon. But given the productivity of modem technology and the elimination of capitalist waste, there are likely to be substantial untapped reserves. In addition, socialist society can, as just explained, deliberately plan to produce surpluses of various items just to meet the eventuality you have in mind.

With regard to human resources in particular, even today under capitalism tens of millions of people are unemployed. Though of course in socialism no one will be "employed" as such, the average workload for individuals is likely to be much less (thus resulting in a sizeable reservoir of labour) and the opportunities for individuals to move flexibly from one kind of work to another much greater. This will make much less likely the occurrence of the bottlenecks you foresee in the produc¬tion of any particular good following an unexpected increase in demand for it.
But scarcity is not simply a function of supply; it is also a function of demand. It is in this area that the anarcho-capitalist critique of socialism, based on its premise of infinite demand, is particularly weak and unrealistic. For it takes little, if any, account of the effect of the social environment on the likely structure and size of demand in socialism.

In a system of capitalist competition, there is a built-in tendency to stimulate demand to a maximum extent. Firms, for example, need to persuade customers to buy their products or they go out of business. They would not otherwise spend the vast amounts they do spend on advertising those products. At the same time, there is in capitalist society a tendency for individuals to seek to validate their sense of worth through the accumulation of possessions. This is not surprising for if, as Marx contended, the prevailing ideas of society are those of its ruling class then we can understand why, when the wealth of that class so preoccupies the minds of its members, such a notion of status should be so deep-rooted. It is this which helps to underpin the myth of infinite demand. In socialism, status based upon the material wealth at one's command, would be a meaningless concept. Why take more than you need when you can freely take what you need? In socialism the only way in which individuals can command the esteem of others is through their contribution to society, and the more the movement for socialism grows the more will it subvert the prevailing capitalist ethos, in general, and its anachronistic notion of status, in particular. Nor do we accept your premise that prices arise out of conditions of scarcity. They arise out of conditions of private property. So even if genuine shortages occur in the conditions of common ownership that will exist in socialism - it is likely that some shortages (e.g. decent housing) will persist (if only as a receding problem) into the early stages of socialism - this will not undermine the new society by leading to the re-emergence of money and prices.

For socialism to be established, there are two fundamental preconditions that must be met. Firstly, the productive potential of society must have been developed to the point where, generally speaking, we can produce enough for all. This is not now a problem as we have long since reached this point. However, this does require that we appreciate what is meant by "enough" and that we do not project on to socialism the insatiable consumerism of capitalism.

Secondly, the establishment of socialism presupposes the existence of a mass socialist movement and a profound change in social outlook. It is simply not reasonable to suppose that the desire for socialism on such a large scale, and the conscious understanding of what it entails on the part of all concerned, would not influence the way people behaved in socialism and towards each other. Would they want to jeopardise the new society they had helped create? Of course not.

One must therefore assume that whatever shortages may persist can be tackled by some system of direct rationing and will be borne with forbearance - even, one might say, with a sense of altruistic restraint. For whatever the problems that socialism may have to contend with, and there will still be many, if the alternative has to be the re-instatement of capitalism then there would not be a real alternative
Editors.

Socialist Standard April 1994

Beyond Capitalism

In 1968 an article written by a member of the Socialist Party, entitled "Smash Cash", appeared in the magazine OZ. Some years later the author of that article, David Ramsay Steele, was converted to the free market ideas of Ludwig von Mises which he expounds in a book just published, From Marx to Mises: Post Capitalist Society and the challange of Economic Calculation.

Mises (1881-1973) was an Austrian economist whose disciples included Hayek and Lionel Robbins. Early this century, against a background of ascendant Bolshevism, he developed what Steele calls "the most powerful objection that has been made to Marxian socialism". In the 1920s and 1930s, in the wake of the collapse of the Bolshevik experiment of "war communism" (1918-1921), Mises's ideas were widely discussed but from the 1940s began to fade as the fad for Keynesian interventionism grew. But for the failure of, and recent retreat from, Keynesianism itself, Mises might have remained an obscure nonentity. However, the "new groundswell of anti-socialist pro-market opinion" that emerged in the 1980s has prompted a reappraisal of the man — particularly since the collapse of the Soviet empire.

So what is the Misesian objection to socialism? Some of the evidence presented can be summarily dismissed. For example, Steele refers to the failure of "war communism" under the Bolsheviks but the pre¬conditions for socialist revolution did not exist in Russia at the time. The means of production were insufficiently developed to permit a socialist system to function, (even assuming you could have "socialism in one country"). Nor was there the necessary level of popular support to bring it about. Reference to the fatuous claims of "Bolshevik leaders" that they were "abolishing commodity production and money" is irrelevant; you cannot impose socialism on an unwilling (and still largely peasant) population, unaware of what it entails.
Such specious evidence aside, there remains the theoretical argument that there is something in the nature of socialism that makes it inherently untenable. This has two basic components which, though linked, can be separately analysed.

Central Planning

Steele points out that planning as such is by no means incompatible with the market. Within the market there are numerous plans but the interconnections between them are unplanned or "anarchic". The proposal to go beyond planning the parts of production to planning the whole of it, thus requiring a "single vast plan" dispensing with the numerous plans of the market, is what has generally been called "central planning", (though "total planning" might convey the meaning better).

But could a total planning of worldwide production be achieved? Production today is integrated through a complex division of labour. The production of even a simple item is inextricably bound up with the production of numerous other items. When we consider that these in turn require inputs of various sorts for their production, we have an inkling of just how difficult it would be to centrally plan in advance output in every conceivable line of production.

Theoretically, this could be done by constructing a vast "input-output" table. But, for logistical reasons, the best this could achieve is a drastically simplified picture of the input-output linkages that make up a production system; at best the number of items it could handle would probably amount to a few hundred. While each might represent a broad category of goods lumped together for convenience, in reality there are hundreds of thousands of different goods so that any decision made on the basis of such crude aggregated data is likely to result in gross misallocation of resources.

But the problem would not end there for once the Plan has been formulated, assuming it could be, it would require considerable coercion to implement it: how else would it be possible to ensure that targets specified by it were met? This would put it at odds with the democratic nature of a socialist society. Democratic participation would be precluded since this requires informed decision-making whereas no individual can possibly absorb or utilize more than a tiny fraction of the information to operate the whole production system. Inevitably, the power to make decisions would gravitate into the hands of a central administration (or re-emergent state).

According to Steele, "central planning" lies at the very heart of the Marxist vision. Whether Marx and Engels did themselves support Steele's concept of central planning is unclear. Some of their remarks suggest strongly that they did; others suggest the opposite, as Steele himself concedes (p. 316). This ambiguity is well captured by the phrase "anarchy of production". Steele interprets this as pejorative reference to the fact that the total pattern of production is unplanned but it could equally allude to the ungovernable laws of a capitalist economy which manifest themselves through the trade cycle.

A recession occurs not simply because there is unbalanced growth between different sectors of the economy (which might imply the need for centralized planning) but because of the knock-on consequences of such growth within the context of a market economy: an economy-wide contraction of production because the monetary flows are interrupted. As Marx pointed out, it is because of "the very connection between the mutual claims and obligations, between purchases and sales", that disproportionate growth leads to recession (Theories of Surplus Value, Chapter 11, 4c). Once you remove the market you break that connection, then any overproduction of some good no longer has a knock-on effect on the rest of production and can be remedied simply by consciously adjusting output to the level required. In this sense "conscious social control" replaces the "anarchy of (market) production".

Steele tells us that Marx set much store by the supposed tendency towards centralization in capitalism — and thus, following Steele's logic, a steady diminution in capitalist anarchy — since it would lead to the "breakdown of the rationale of capitalist production according to the law of value" (p. 72). Yet, curiously, we find Engels in Socialism: Utopian and Scientific saying the exact opposite, that "anarchy" in capitalism grows to a "greater and greater height". This is hardly consistent with Steele's understanding of "anarchy of production" though it does square with the view that capitalist crises get progressively worse.

Perhaps Marx and Engels were at times inconsistent or muddled on this question. But why should it concern us? We are not dogmatists; if they were advocates of central planning in the sense of total planning, we would dissociate ourselves from this aspect of Marxism, and for the best of reasons: a centrally-planned economy is neither feasible nor compatible with the nature of socialism.

What really matters is not what Marx and Engels may have said, but what a rejection of central planning entails. In a model of socialism in which total production is not centrally planned there would be, as in capitalism (and in Steele's sense of the word) an "anarchic" or spontaneously ordered system of production. But there the similarities would end.

Local production

Steele does acknowledge the existence of models that reject both central planning and the market, one being Kropotkin's "anarchocommunism". According to this, "groups of producers (would) govern themselves and federate for occasional common purposes" (p. 217). Such groups would effectively organize production on self-sufficient lines but, in Steele's view, the imposition of local autarky would precipitate a "violent reduction in living standards" (p. 322); it would entail the disintegration of the spatial division of labour and the comparative advantages of regional specialization. Whatever the merits or otherwise of this argument, we do not have to accept that the only alternative to central planning in socialism is localized autarky. In fact the Socialist Party has long argued that socialist production would be carried on at several levels — local, regional and global — though Steele implies incorrectly that this is a relatively recent development and that until the 1980s the Party supported central planning (p. 417).

There may of course be a tendency to produce more things locally in socialism than at present. But given that there will remain a considerable degree of interaction between communities (however defined) in the form of material flows, on what basis will these interactions occur? To suggest that only the market can integrate worldwide production and that "inter-regional movements of goods" cannot be "regulated without trade or prices" is mistaken; it is to fail to distinguish between market-exchange and what anthropologists call "generalized reciprocity" under which goods are transferred between and within communities on the basis of giving and receiving rather than buying and selling.

Generalized reciprocity is a mode of transaction in which neither the value of what is given is calculated (it has no price) nor the time of its "repayment" specified. It thus denotes a kind of social obligation — that we do not simply take from society without giving back something in return — as summed up in the social rule "from each according to their ability to each according
to their needs". In this sense, socialism can be characterized as a system of generalized reciprocity.

Economic calculation

Since total planning is an unworkable proposition, it would suffice to show that it was essential to socialism to prove that socialism itself could not work. Steele does not exactly pursue this line of argument; he seems to regard the matter as secondary to his main argument that, in the absence of market prices, economic inefficiency will result, leading to a significant decline in output. This is the so-called "economic calculation argument".

According to this argument scarcity is an unavoidable fact of life; it "applies to any goods where the decision to use a unit of that good entails giving up some other potential use". In other words, whatever one decides to do has an "opportunity cost" — that is the opportunity to do something else which one thereby forgoes; economics is concerned with the "allocation of scarce resources"; the most efficient allocation is one that uses up the least resources in the production of any item and thus leaves the most resources over for other uses, so mimimizing the opportunity costs of producing that item.

Steele provides a hypothetical examples of a "widget" which can be produced by three methods (p. 6): Method A (5 lbs of rubber and 5 lbs of wood), Method B (5 lbs of rubber and 4 lbs of wood) and method C (4 lbs of rubber and 5 lbs of wood). Clearly, B and C are more efficient in terms of resource use than A but how do we know whether B is better than C or vice versa?

According to the economic calculation argument, this requires being able to compare different factors of production, like rubber and wood, and that means reducing them to a common unit. By adding up the total costs involved (expenditure) and the benefits gained (income) in terms of this common unit, we can determine which particular method is most economically efficient by the magnitude of net profit it yields. Of course, there may not be any net profit, in which case it would be uneconomic to produce widgets; the value of resources used up in making them would exceed the value of the widgets themselves and there would be less resources left over for other uses.

According to Steele, such calculations are made possible by virtue of the existence of money; socialism would rapidly regress into chaotic inefficiency unless it can replace money with some alternative that enables society to continue to make such calculations. However, he does concede that capitalist entrepreneurs who make these calculations may often get their sums wrong; given the "anarchic" nature of capitalist production they must rely on shrewd estimates. But what matters is that there is some "objective test of the accuracy of (these) estimates" which profit and loss provides by rewarding those whose estimates are correct and weeding out those whose estimates are incorrect.

Thus, the market process is "self-correcting" while socialism supposedly lacks such a mechanism and has no way of discovering least-cost factor combinations since it has no common unit by which to make cost comparisons. One proposal to get round this problem is that of labour-time accounting with labour time serving as a common unit of cost. But this is beset with numerous difficulties, not the least of which is how do you weight different kinds of labour.

Stock control

There is, however, another proposal which Steele considers which, in fact, turns out to be the definitive answer to the economic calculation argument — namely "calculation in kind". By definition, this assumes that you do not need a common unit of calculation as such a need exists "only where there is commodity exchange". Steele concedes that "some calculations, even within the market, can be done in kind" (p. 86), but believes the scope for this is limited. He also contends that it fails to address, the "fundamental question of how to compare the costs of alternative aggregates of factors" (p. 123), but this only assumes what it needs to prove: that such comparisons are necessary.

Given that socialism will still need to concern itself with the efficient allocation of resources (among other things), how will this be achieved through calculation in kind? The answer rests crucially upon a recognition that production in socialism cannot be totally planned in advance. Steele stumbles over parts of that answer but the long shadow of central planning that pervades his book prevents him from seeing it. This problem is compounded by the fact that the major proponent of calculation in kind to whom he refers, Otto Neurath, was himself an advocate of central planning.

Decentralized production entails a self-regulating system of stock control. Stocks of goods held at distribution points would be monitored, their rate of depletion providing vital information about the future demand for such goods, information which will be conveyed to the units producing these goods. The units would in turn draw upon the relevant factors of production and the depletion of these would activate yet other production units further back along the production chain. There would thus be a marked degree of automacity in the way the system operated.
The maintenance of surplus stocks, as Marx pointed out, would provide a buffer against unforeseen fluctuations in demand (Capital Vol.2) but it would also be relevant to the task of efficient allocation. This is so because of the inverse relationship between the supply of any good and the need to economize or use less of it: the scarcer a good the greater the need for economization.

An analogy might help here. According to the 19th century agricultural chemist, Justus von Liebig, there is a "law of the minimum" whereby plant growth is limited by the availability of whatever nutrient is scarcest — usually fixed nitrogen. Some working principle analogous to the "law of the minimum" would apply to the allocation of factor inputs in socialism.

Suppose that the demand for a good X as registered through the system of stock control rose permanetly but that the production of resource A was insufficient to meet this increased demand. In the face of an already committed pattern of factor allocation, the sensible solution would be to search for some more abundant resource B that could substitute for A. At the same time the falling stocks of A would constrain the multifarious users of A to economize on it.

This feedback permits a continual process of discrimination in the use of resources according to their comparative availability. In short, it proves precisely the kind of self-correcting mechanism which Steele claims is the exclusive property of a market system.

Market tyranny

With this claim rebutted, most of what remains of the theoretical argument against socialism falls away. For example, Steele contends that much of the economic activities carried on in capitalism, which socialists regard as wasteful, is in fact "productive". But this is to be expected: if you cannot accept the possibility of an alternative to the market then anything that is functional to the operation of a market economy must, by definition, be "productive". Conversely, once that possibility is accepted the appalling, ever-mounting burden of capitalist waste becomes glaringly evident.

Capitalism, with its obsessive preoccupation with reducing costs is, in many respects the very antithesis of efficient production. We do not refer here only to its obvious structural costs; there is also the important category of externalized costs to which Steele makes barely a passing mention. These arise out of the competitive pressure on capitalist enterprises to pass on — or "externalize" — some of their costs, the repercussions of which, as many documented cases of environmental disasters testify, far exceed any savings that might have been achieved for the sake of profit.

Despite its basic anti-socialist position, this is a book that is both highly informative and lucidly written; it has much to recommend it. But perhaps, ironically, its most enduring value will be as powerful stimulus to fresh thinking about the kind of society we want, a society free of the tyranny of the market.

ROBIN COX. Socialist Standard June 1993

further reading
The "Economic Calculation" Controversy: unravelling of a myth by Robin Cox
Socialism and Calculation (.pdf)
Socialism as a Practical Alternative (SPGB pamphlet 1994 - .pdf)
Online book by (late SPGB member) Pieter Lawrence