Chapter VI

On the contrary, in economics there is a growing tendency to put· an equals-sign between value and the result of exchange-relationships. 3 This is to give the· theory of value a much more modest role than it had before. Reference to price says even less than reference to "supply and demand". Attention is confined to the result of the conflict between the various factors, disregarding the forces behind them, such as (for supply): number of hours worked and scarcity of resources, and (for demand): subjective valuations of needs and sacrifices-to name the most important ones. One is admittedly on safer ground, when sticking to prices with their special faculty of synthesizing a variety of heterogeneous factors. In a barter market prices can be expressed by their relative barter-relationship to other goods, but the machinery of exchange is simplified and improved by the use of a generally accepted means of exchange which makes it possible to express every price in terms of one good or one unit of value. Thus monetary economy represents a perfection of barter economy.

1A monetary economy was also assumed in the previous chapter, so. its implica tions might have been discussed there, but it was natural to concentrate in that chapter on the consequences of the communist-marxist· principles. 2As we shall see later (cf., inter alia, Chapter XIV), an attempt is made in this discussion to extend the concept of price and to let prices represent something more and something else than an expression of exchange-relationships. 3Cf. L. M. Fraser: Economic Thought and Language, London, 1937, p. 351. 65 66 ECONOM:IC CALCULATION IN THE SOCIALIST SOCIETY In Chapter II it was said that we had no right to assume the existence of either value~relationships, markets or prices in a socialist society. However, further examination 1 has shown that a monetary economy is a necessity, since a moneyless one does not provide that common denominator which is essential for calculation; and that claims to consumption (warrants) made out for an amount are better than ones made out for a specific quantity of goods in natura and give their possessor a better opportunity of choosing what needs he will satisfy and to what extent (within the scope of the goods offered and of the purchasing power at his disposal). Finally, the conclusion has been reached that if the central authority is to avoid producing goods in which no one has any interest, it will have to take account of the utility of its products. If not, the goods produced might be said to have a value according to the yardstick used, for example, hours of work, but not necessarily to have one in the eyes of the consumer.

Does acceptance of these conclusions imply that our socialist society is to have a monetary economy similar to that in a capitalist society? Not necessarily. There are essential differences. Seeing that the State is the one and only employer and the sole owner of all goods which may be sold and bought, the socialist society can manage with a giro-or clearing arrangement. In theory it has particularly favourable facilities for carrying this through. Theoretically all payments (whether for raw materials, power, transport or semi~finished products at every stage of production, for wages and consumer goods), can be made. by means of cross book-entries. The common unit of reckoning used may be anything one likes, from hours of work to motor-headlights or tillers, as long as it is used consistently. No sort of cover is necessary and one can imagine that the socialist community can manage with money ofan entirely different kind to that used in the capitalist society.

In this connection it may be mentioned that money-of-account, of the kind possible in a socialist community, has come to occupy a dominant. place in certain central treatises on monetary theory. Professor J. M. Keynes begins his treatise on money as follows: "Money-of-Account, namely, that in which Debts and Prices and General Purchasing Power are expressed, is the primary concept of a Theory of Money." 2 In his "Currency and Credit"3, R. G. Hawtrey takes as his starting point a civilized society which makes use of credit (money-of-account), but not of circulating money. The object of this abstract experiment is to determine the specific function of money, and Hawtrey arrives ICf. Chapters IV and V. 2A Treatise on Money, London, 1930, I, pp. 3-4. 3Curren~y and Credit, 3rd Edn., London, 1928, Chapter I.

MONEY AND FORMATION OF PRICES OF·CONSUMER GOODS 67 at the conclusion that it is that of settling debts (which may be performed by "money-of-account"). Even if the facilities for using money-of-account and clearing arrangements are especially favourable in a socialist society and even if it is theoretically possible to use them, there exist great practical difficulties. It is, of course, out of the question that the buyer and seller should meet in person to adjust debits and credits. Drafts or cheques would have to be used. With a cheque-or a clearing arrangement misuse would be possible by overdrawing accounts. Even if one assumes that fraud is unthinkable in a socialist society, it will nevertheless be possible for accounts to be overdrawn through forgetfulness, failure to make note of expenditure, or faulty book keeping. If the person receiving the cheque was obliged to see that there was cover, the practical usefulness of such a clearing system would, of course, be further curtailed. The chief objection, however, is that such an arrangement would be most unwieldy. One has only ·to think how troublesome it would be to write out a cheque every time one wanted to disburse .some small account, such as a tip or bus fare. The amount of bookkeeping necessary to write up all the milliards of transactions would be enormous. Sooner or later people would start demanding circulation-money as well. l Whether this circulation-money is to be called money or not, is a question of definition and of subordinate interest to this discussion.

If one accepts the extreme definition that everything which is accepted as a means of payment is money, then su.ch circulation-money will be money. Also if the central authority only allows communal goods to be distributed against these warrants and forbids any private barter, the members of the community would be forced to accept and use them as money. It is, however, necessary to point out that this circulation money in a socialist state differs in various respects from money in the capitalist society. One difference is that in a socialist society its circulation will be considerably ·restricted in that the members of the community, qua consumers, can only use it to make purchases in the State's shops and will not be in a position to employ it for buying means of pro duction, raw materials or semi-manufactured goods, or for barter transactions among themselves. This means that the process of circulation will be much shorter than in the capitalist society. In the socialist. community the money will go from the State (or one of its IDr. Tisch has gone thoroughly into the question of how far the consumption warrants, on which her proposed solution of the problem of calculation is based, are money or not. In this connection she gives a full and interesting summary of the rather mixed discussion of money in marxist literature. After a thorough discussion Dr. Tisch comes to the conclusion that money-of-account and clearing alone are impracticable.

68 ECONOMIC CALCULATION IN THE SOCIALIST SOCIETY organs) to the consumer and from him back to the State (or one of its organs). Thus circulation-money is really rather a misleading term and it might be better to call them claims-to-consumption (warrants), keeping in mind, however, that they are made out for amounts and not for naturalia. The fact that the process of circulation is short in distance does not necessarily mean that it will be so in time, for even these· claims to consumption can serve as a means of storing value, as a means of hoarding. It seems doubtful that hoarding is such a decisive factor for trade and employment in capitalist societies as Professor Keynes makes out (in his "General Theory of Employment, Interest, and Money"), but, in any case, the consequences of hoarding are not necessarily the same in a. socialist community, considering that the volume of pro duction of capital goods as well as of consumers' goods is entirely in the hands of the central authority. Whether or not this is compatible with the aim of maximum production for satisfying needs will be discussed in Chapter XVI. For the time being we shall assume that it is so and that the central authority has sovereign powers for fixing the kind and volume of production, and that it keeps labour employed, even to the extent of starting an undertaking solelY for the sake of employment. In this case the only result of hoarding 1 in a socialist community would be that a proportion of the consumer goods put at the disposal of the members during the period in question would remain unsold.

If the aim is continuous turnover and accumulation of goods is considered undesirable, there are various solutions imaginable. One would be to reduce prices so far that it would be more tempting to buy than to hoard. (We shall return shortly to this alternative and the question of fixing prices for consumer goods.) Another way would be to give the money a restricted term of validity, so that the members of the community would either have to use them within a certain period or get nothing for them. Seeing that money in a socialist community has a very restricted use, such limitation of the period of validity is more natural than the "stamping" proposed, e.g., by Silvio Gesell. On the other hand the use of money with a restricted period of validity may have unfortunate secondary effects. (See Chapter VIII.) A third solution would be to make an extra distribution of pur chasing power in the shape of higher wages or a social dividend. 2 Such an extra distribution may create certain difficulties. If the unsold 1That the concept of hoarding has been variously interpreted in modern literature, is not relevant in this connection. (See "The Concept of Hoarding", by Joan Robinson, in Economicjournal,June, 1938.) 2For the form of distribution of "social dividends", see the discussion between Dr. Lerner and Dr. Lange in ReviewofEconomicStudies,October, 1936, and February, 1937· MONEY AND FORMATION OF PRICES OF CONSUMER GOODS 69 goods were of such a kind or quality that nobody wanted them, then the consumers will merely hoard the extra money as well in the hope of being able later to buy goods more suitable to their desires and needs.

If such goods did come on the market later the central authority would run the risk of more money being suddenly used (or more· warrants presented) than there were goods to meet. Such a situation would arise from other causes, too, for instance, from fear of a·rise in prices, or simply because the desire to buy had increased owing to a change of season, which can call forth both unconscious physiological reactions and a desire to provide oneself with clothes and other articles suitable to the time of year. To eliminate the factor of uncertainty inherent in the hoarding of warrants, it has been suggested that the central authority should tempt the people to invest in bonds, so that it at any time has full control over all outstanding purchasing power. To achieve this, the bonds must be long-term or, preferably, unredeemable. To admit of con version, redemption or sale would give the savers the opportunity of making that sudden use of their money which it is desired to avoid.

If the hoarding should be the result of a desire to keep the money for later purchases of goods, investment in ·long-term bonds would not appear a very attractive alternative. In order to attract purchases the rate of interest would have to be high. If people were allowed to invest in presumably safe State bonds, bearing a high rate of interest, it would be possible for the parsimonious eventually to live a render existence, and this many socialists consider incompatible with the socialist society. Between money in a capitalist society and money in a socialist one (money being used in the widest sense of the word, including means of-accounting and claims-to-consumption) there is also the large and essential difference, that money-of-account does not necessarily need to have the same denomination as claims-to-consumption or bear any relation to them in respect of value. This is due to the fact that the State is the sole owner of the means of production and the one and only employer. Production and distribution do not take place uno acto, as they do in a capitalist society in which distribution occurs directly out of the production of the economic goods and their exchange, and as its ultimate result. The socialist state on the other hand can allow distribution to be quite independent of the process of production (within the scope of the goods available for distribution).

As long as the central authority has the only sai in the distribution of communal goods, it can employ one unit of reckoning for production and another for distribution. The matter, however, takes on a different complexion if it is admitted 70 ECONOMIC CALCULATION IN THE SOCIALIST SOCIETY that the Central authority should take into account the utility of the goods produced. This introduces a new factor and necessitates a yardstick to which the central authority must refer. Now, what is useful will depend on the end. That which is useful in a community aiming at rapidly building up its war-machine, is not necessarily so in one which looks to satisfy the needs of its members. Taking maximum satisfaction of needs as a reasonable working hypothesis, an important question arises: who is to decide what can best serve as means of satisfying the needs of the members of the community, and what is useful for them? Here there are two possi bilities: either that the members of the community themselves determine the nature and intensity of their needs, or that others, e.g.

the central authority, do it for them. To those that believe that people differ not least in their tastes and needs, it will appear unreasonable that the central authority should determine how the individual is to satisfy his needs, or that production should be based on scales of needs fixed by that authority. There are, however, some who think that the State, whether in a socialist or capitalist community, best knows what is good for the members of the community to consume.! Both alternatives have been put forward in the discussion of economic calculation in socialist communities. Most of the participants assume, as a matter of course, that there must be free choice of goods and occupation. Some have .even made this a precondition in their proposed solution of the problem. There are, however, others who think that the problem of calculation can only be solved if these facilities are abolished. Whether one should accept the one or the other alternative is not for the theoretical economist to decide. Both will be discussed here. First, let us assume that the members of the community have a free choice of goods and occupation. 2 IThe different individuals' ideas about this matter seem to depend on their attitude to their fellow men and surroundings in general. People seem-irrespective of "race" and in every phase of civilization-to be divided into two different types: the one demands, as a matter of course, respect for the individual's personal character and tastes, while the other prefers common rules for all, feels himself attracted by the mass and thinks that others, too, should merge and incorporate themselves in the mass. It would take too long to discuss here what lies at the back of these two attitudes, or to analyse the attempts to rationalize the state of mind that seems to be derived from instinct. Analogous and equally divergent views are to be found among the Realists and Nominalists of the Middle Ages.

2As has been mentioned the majority of those who have taken part in the dis cussion of the possibility of calculation in the socialist community think that this solution is the natural one. Dr. A. P. Lerner wrote in "Economic Theory" in the Review of Economic Studies, 1934-5, that scales of consumption worked out by the bureaucrats of the central authority would be undemocratic and incompatible with socialist ideals. Dr. Oskar Lange concurred in this in his article "On the Economic Theory of Socialism" in the same journal for 1936-7. Dr. Lerner, for his part, makes the reservation that he speaks as a "human being and a sympathizer with socialis tends" • MONEY AND FORMATION OF PRICES OF CONSUMER GOODS 71 Free choice of goods may mean two things. Either that the con sumers can freely choose between the goods available, but that they have no chance of directing the course of production. In that case the preference scale chosen by the central authority is the only deciding factor. Or it may mean that the consumers are able to influence production, which is changed according to their preferences as shown by their effective demand. This latter is the usual interpretation given to free consumer choice and the one which will be used here.

It has been mentioned that a socialist community might under certain conditions have different denominations for money-of-account and consumption-warrants. If we assume that the members of the community shall themselves determine their needs and the trend of production, this contention becomes untenable. In that case it would be more natural to have the same. denomination for the prices of necessaries and of means of production (that is to say, to have the same monetary unit for money-of-account and claims-to-consumption), so that the prices of consumer goods can be used for direct calculation and exercise their influence on the prices of means of production and vice versa. This does not imply that one common denomination for money-of-account and claims-to-consumption makes prices in a socialist community serviceable as data for calculation; that point will be gone into later. First, we must quite briefly discuss certain consequences of the assumption that there is a free choice of occupation. .

Free choice of occupation can mean both free choice of profession and of the positions for which one may be· qualified. Where there is no free choice of occupation the members of the community-either on the basis of intelligence and physio-psychological tests or arbitrarily-will have assigned to them the profession they must choose, and they will have to take the situation assigned to them by the central authority or its agents. Here again, the economist as such is not able to say which alternative should be chosen, even though it is probable that the amount performed will be greater where the work is voluntary,. than if compulsory. Should this assumption be correct, as it is generally supposed to be, there will be economic consequences attendant on the choice, which will thus influence the possibility of attaining the end of maximum production to satisfy needs. Here, too, we shall discuss both suppositions, but, for the time being, we assume that there is a free choice of both profession Seeing that Dr. Lerner and Dr. Lange have embarked on a value-judgment in using such a yardstick as "democratic or undemocratic" one would seem entitled to ask whethet it were not undemocratic that the socialist state should (I) forbid the members of the community to produce goods they and their fellow-men need, (2) forbid them to engage people for this end, and (3) forbid them to have other employers than the State.

72 ECONOMIC CALCULATION IN THE SOCIALIST SOCIETY and occupation, keeping in mind, however, that the choice will necessarily be restricted, when there is ex-definitioneonly one employer. Even if the choice of occupation is free in theory, those looking for work may, in practice, have to take what is offered them. The assumption of free choice of occupation does· not greatly affect the question of the possibility of calculation, but it will put certain difficulties in the way of achieving optimal distribution and optimal use of the community's manpower. As this is of a certain theoretical interest, it will be discussed briefly here, even though it does lie somewhat outside the scope of a discussion on calculation. The natural result of a free choice of occupation will be a rush for the professions and situations which are considered the most attractive. If the State is not able to order people to take up the various types of work, it will have to employ a graded wage-scale if it is to obtain optimal distribution of manpower.' This creates a difficulty for those who think that a· graduated scale of wages is incompatible with the principle of just and equal distribution. Dr.

Oskar Langel has taken up the question and found certain difficulties, for he is not one of those who assume that egoistic-individualistic behaviour will disappear under socialism. Dr. Lange realizes that differentiation of incomes is required in order to obtain optimal distribution of labour, but he tries to show that no real differentiation is necessary. It is important for him to prove this, as he wants to have bothofree choice of occupation and goods and a distribution of income which will maximize "the total welfare of the community". In his opinion this· can only be achieved if the following two con ditions are satisfied: (I) that the distribution is such that the same demand-price offered by the different consumers represents an equal urgency of needs, as is obtained if the marginal utility of income is the same for all consumers, and (2) that the distribution leads to such apportionment of the services of labour between the different occupa tions as to make the difference of the value of the marginal product of labour in the different occupations equal to the differences in the marginal disutility involved in their pursuit.

In Dr. Lange's opinion the first condition is satisfied, when all consumers have the same income, "assuming the marginal utility curves of income to be the same for all individuals". This assumption is wholly unrealistic. It is inadmissible to assume that similar incomes give the same marginal utility curves (see Chapter IV). As Mrs. Joan Robinson puts it, "It is not really justifiable to talk about maximum satisfaction' to a whole .population", for the reason that "to a strictly l"On the Economic Theory of Socialism, II", in the Review of Economic Studies, February, 1937.

MONEY AND FORMATION OF PRICES OF CONSUMER GOODS 73 logical mind any discussion of utility to more than one individual is repugnant" .1 Dr. Lange is aware that the assumption does not correspond to reality, but holds that "such differences (of 'sensitiveness') would disappear in the relatively homogeneous social stratification of a socialist society and all differences as to 'sensitiveness' would be of a purely individual character". This new assumption, that the citizens of the socialist state will grow into unsensitive automata is doubtful, but very revealing, and so is his disregard of individual characteristics. Dr. Lange admits (a) that optimal distribution of labour necessitates differentiated income, and (b) that this may seem to be in contradiction with condition (I). Now he says that this is not really so. By putting leisure, safety, agreeableness of work, etc., into the utility-scales of the individuals, "the disutility of any occupation can be represented as opportunity-cost".

Thus, in Dr. Lange's opinion, the choice of an occupation offering a lower money income, but also a smaller disutility, may be interpreted as purchaseof leisure, safety, agreeableness ofwotk, etc. He also hints at the possibility of the central authority paying the same money income to any citizen, and charging a price for the pursuit of each occupation. He goes on to draw the conclusion that not only is there no contradiction between the two conditions, but that condition (2) is necessary in order to satisfy condition (I). There are several reasons for not accepting Dr. Lange's subtle argument (which seems, by the way, to be based on arguments advanced by Adam Smith). In the first place, he is arguing in a circle, or, rather, is contradicting himself: he starts by requiring differentiated income, but arrives at the conclusion that the incomes are in reality the same, seeing that from the higher incomes must be subtracted the disutility, which makes it carry a higher salary, and to the lower must be added the advantages (in the shape of leisure, safety, etc.) which make them low. One cannot have at one time differentiated and equal wages. One must. either stick to nominal wages, expressed in amounts, or to real wages, where advantages and disadvantages are taken into account.

In the second place, optimal distribution of labour does not merely mean geographical distribution (as .to places and industries) and a distribution of manual labour, but optimal exploitation of efficiency, professional knowledge, capabilities and readiness to assume responsi bility. An employer-in this case the central authority-cannot, when fixing wages, merely take into consideration whether the position in question represents greater or less amenity in his own or the opinion of the applicant.· Such criteria cannot be used for grading wages. lEconomics of Imperfect Competition, London, 1933,. p. 3I~. F 74 ECONOMIC CALCULATION IN THE SOCIALIST SOCIETY The chief thing and the real criterion must be the importance of the performance for the production-result. Unless one believes that all individuals are possessed of the same efficiency, professional knowledge, capabilities and readiness to assume responsibility, the application of this criterion will create differentiated wages in conflict with Dr. Lange's conditions.

We now revert to the question of how the central authority is to procure reliable data for regulating production. Incorrect data regard ing preferences will make itself felt throughout the whole system and distort production in all stages. Wrong production resulting from incorrect data means a loss of utility for the community. The question of how correct data about consumers' preferences is to be obtained in the socialist society has not been gone into very deeply in the discussion on the possibilities of calculation, even by those who postulate production of needs and free choice of goods (and occupation). Dr. Herbert Zassenhaus has expressed the opinion l that the sceptics have been too much occuped with distribution in the socialist community. This is a surprising assertion. To the sceptics belongs the honour of having started the discussion, but to the advocates belongs the honour of having most discussed how.satis faction of needs should be determined, presumably from a feeling that certain difficulties were involved. But even after their treatment, there exist greater problems than have hitherto been pointed out.

Ballots (e.g. together with mannequin parades) have been suggested for obtaining data for determining consumers' preferences. Such· an expedient does not solve the difficulties. It is quite impossible for consumers to give a scale of preferences for all consumer goods through such ballots and still less so to indicate the nuances in their desires or the relative strength of their different needs. Further, the final calculation would be enormously difficult, quite apart from the fact that the consumers would have to keep continuously balloting, as the number of consumers, their needs and the relative strength of their needs, are constantly changing. Besides, such an artificial fOfm of voting is entirely superfluous if claims-to-consumption, made out for amounts (money) are used. In his purchases. the consumer then has an opportunity, as simple as it is effective, of showing his prefer: ences for the goods available. He is (within the range of goods offered) able to maximize the utility of the purchasing power at his disposal, ,so that the marginal utility of each unity of income (unit of reckoning) becomes the same for all goods.

It might appear that this solves the problem: the demand will show lSee "NeuerePlanwirtschaftsliteratur und die Theorie der Planwirtschaft", in Zeitschriftfiir Nationalok.onomie,1936.

MONEY AND FORMATION OF PRICES OF CONSUMER GOODS 75 whether an article is preferred or not; if at the end of the period of reckoning there is a shortage of one article, this would show that more ought to be produced. Unsold goods would be an indication that less should be produced (at the prices fixed). However, it is not as easy as that. Only under certain definite conditions will a surplus or shortage of goods be an indication of preference on the part of the consumer. In the first place, an oppor tunity of choice must really exist, not only from among different kinds of goods, but from among goods of the same kind, at different prices and of different quality . .Another and equally important condition is that there shall not be issued more claims-to-consumption than there are consumer goods. If that should be done (or if the same kind of money is used for consumer goods as for the payment of raw materials and semi manufactured goods, and a part of this latter is used to buy consumer goods), there will be a shortage of these goods. In such a case there will be no indication of how much shall be produced of the different goods and qualities. Just as in tennis a score of 6-0, 6-0 gives no indication of how much better the winner is, so stocks of unsold goods do not reveal how strongly the different goods are desired.

Even in the event of the central authority issuing the exact number of claims-to-consumption to absorb the necessaries produced-a postulate often put forward in the discussion of socialism-there will still be no serviceable data for determining what the consumers really want. If periodic shortages occur, the central authority will not only lack data for these periods, but the whole foundation for determining consumers' preferences may be destroyed, for the consumers will soon learn from experience and use their claims-to-consumption to buy arty goods obtainable, so as to get somethingin exchange. The same thing will happen, if the· consumers should for one reason or another come to mistrust the value of the money. Such a mistrust would also be a strong stimulant to substitute real values for money, even though the need for the articles in question were not great at the time. Fear of possible inflation would, therefore, likewise contribute to destroy the value of sales-statistics as indications for production.

Whether the demand for consumer goods will give reliable data, will also depend on other factors such as the length of the period of accounting and observation and the facilities for buying on credit and by instalments. Of still greater importance is the question whether the central authority uses fixed prices or prices varying with the demand for the different articles (of different qualities) during the period of accounting. The central authority has supreme power in fixing prices, inasmuch as it has a complete monopoly both as producer and distributor. Here we come up against a problem of calculation 76 ECONOMIC CALCULATION IN THE SOCIALIST SOCIETY of a special kind. Even if we assume for the moment that the problem of calculating production is solved and that there are sufficient goods of every kind to give full facilities for choice, there still remains the problem of how selling prices are to be determined. If fixed prices are maintained and there is not enough of a certain article at the price fixed, the result will be that the central authority receives no indication of how strong the demand for the article in question has been, as there will be queues of would-be consumers, and, presumably, forced rationing. On the other hand, if prices are graded for the different goods according to demand, the central authority gets a stimulated sale of goods for which there has been little demand and a retarded sale of those which have been in greater demand. This last procedure also represents a form of "rationing", but a rationing more in harmony with the aim we have assumed in this chapter, namely, that the members of the community shall themselves be able to determine the scale of preference of their needs and to what extent they wish each of them to be satisfied. If the prices of articles are increased as demands increase and stocks diminish, the number of consumers asking for them will be restricted and the result will be that the goods go to those who feel the need for them so strongly that they are willing to make the corresponding higher sacrifice, that is, to pay the higher price (and can pay it).

It is obvious from what has just been said that varying prices will give the best indication of the demand and of the strength of the demand, but the adoption of this solution raises the question of whether the central authority is to increase and lower prices equally along the whole line, or take into account the fact that not only does the elasticity of demand vary for the different articles and qualities, but it also varies as between different villages and districts, and also at different periods and seasons. In other words, is the central authority to regard its retail shops as monopolistic selling branches (what they really are) or as independent units possessing full freedom in their price policies? It is obvious that this latter alternative will give the members of the community the best opportunity to grade the satis faction of their needs. It also represents the greatest degree of free competition in the last link in the process of production. Whether such competition and such an independent price policy is compatible with socialist ideals, is a question to which we shall revert later. (See Chapter XIII.) If competition and independent price policy are adopted, it is evident that one of the principles, from which the champions of socialism expect so much, is broken, viz. centralized planning and centralized initiative-taking.

Even if this objection were passed over, the alternative of varying prices will not solve all the difficulties for the central authority: the MONEY AND FORMATION OF PRICES OF CONSUMER GOODS 77 demand for an individual article is not only a function of its price, but of the prices of all articles offered for sale. In addition, it depends on the existence of substitute goods, complementary goods and their prices. (See Chapter IV.) The data the central authority can hope to get from the different shops will scarcely prove a sound basis for planning production. If it is possible at aU. to work out curves of demand on the basis of such data (as has been asserted), these will vary not only because of possible differences in th~ demand in the town or country district concerned, but also because the shops in the district in question have altered the prices of other goods. In theory the problem is not insoluble, but it will put enormous practical difficulties in the way of calculation.

In addition it must be pointed out that the way the central authority fixes prices will have a strong influence on the consumers' purchases. The height of a price will influence not only the consumers' demand for these individual articles, but also the use made of their total purchasing power. One must also remember that the retail shops can obviously not be left to fix their prices without regard to what they themselves have to pay for the articles. Their power of fixing their prices according to the demand has its natural limits in the production costs (including selling costs). How and to what extent the central authority should take into account the cost of production (which depends on the value of the means of production for which there are no markets) and to what extent it shall take into account ·data of demand, and how these data are to be collated, is merely one of the problems of calculation the socialist community has to face. If the aim "maximum production for satisfaction of needs" is accepted, it is self-evident that the consumers' readiness to pay rising prices for an article should in the last resort lead to more of it being produced, which would mean that more labour power and more capital (means of production and land) should be employed. The task' before the central authority is to assure an automatic increase in supply.when demand rises.

If it is agreed that a policy of autonomous pricing for the individual shop is compatible with socialism, it is natural to ask why the demand for consumer goods in a socialist society should be less suitable a guide to production than in a private-capitalist one. The explanation is that the basis is entirely different. In a private capitalist society there usually exist alongside the competing autonomous retailers several autonomous, more or less, competing wholesalers, all of whom have established data of costs on which to work. They aim first and fore most at obtaining a maximum profit and do not consciously endeavour to contribute to the maximum satisfaction of the needs of the buying public, which, in a socialist society, is the declared aim. Further, there 78 ECONOMIC CALCULA'flON IN T'Hn SOCIALIS'I'SOCIETY is only one monopolistic seller and no markets which automatically provide data for calculation. In a private-'capitalistic society even the monopolies have certain cost-data at their disposal. If nothing else, there will usually exist prices of goods which may substitute the monopoly's raw materials as well as finished products. Even assuming that the monopoly was absolute and there was no basis of comparison or calculation for either raw materials or the finished product, the task would still be infinitely much easier for a monopolistic concern in a private capitalist society than for the central authority, becaus~ the monopoly only concerns itself \vith obtaining maximum profit.

It is natural to ask whether the central authority could not likewise aim at maximum profit! and thus facilitate its task. It might, of course, but having absolute monopolistic power both as producer and as distributor, there is no limit to its ability to maximize profits. Our discussion is based on the assumption that it is maximum production for needs whjch is the end. In that case the task is entirely different, viz. in the first place to obtain data to determine the changing need$ of the members of the community, and, secondly, to incorporate these data in the system of calculation so that they give an indication of how the limited resources can best be employed in alternative uses to achieve this end. One can imagine various intermediate forms between fixed and variable prices. The most obvious is to maintain fixed low prices (or to have special shop's with fixed low prices) for necessaries and a free market with variable prices for luxuries and other articles. This, however, would give the central authority the least reliable guide to production. Whether one assumes. that there is a certain system in the distribution of claims-to-consumption (a certain connection between the claims distributed and the total price of the goods offered) or mere arbitrariness, the result will be that there will be a "run" on the goods with particularly low prices (or to the shops which sell them). They will always be sold out, from which there will be no other conclusion to be drawn than that at these low prices a larger quantity might have been sold. In the free market, on the other hand, prices will be forced up, because only a small part of the claims can be used in the shops with the cheaper goods. There will be more left for the purchase of free articles, and the central authority will have no indication whatever of the extent to which needs have been s~tisfied in respect of the cheap and the free articles.

The problems and difficulties connected with fixing prices for lAs we shall see later (Appendix C) the concept of profit will not be foreign to the socialist community.

MONEY AND FORMATION OF PRICES OF CONSUMER GOODS 79 necessaries in a socialist society have, as mentioned, scarcely been touched on by those taking part in this discussion, although many of them are aware that to have correct data regarding consumers' preferences is a necessary condition for obtaining that valuation o( the means of production which is indispensable to achieve the given end. The question has been taken up by the first Englishman who took part in the discussion, H. D. Dickinson.! On the subject of calculation of necessaries, he writes: "The selling agency will sell on the basis of what the market will bear, raising the price when stocks fall short and lowering it when they accumulate." Mr. Dickinson realizes that "construction of demand schedules" is a very difficult task, since the demand for an article is not merely a function of its price, but of the prices of all other articles. (On this point he refers to Cassels' Theory of Social Economy, new edition, pp. 139-147.) There are many objections to Mr. Dickinson's solution, or rather, different suggestions for a solution (see also Chapter X), but here we shall consider only his proposal to solve the problem of calculating necessaries. Dr. A. P. Lerner, who likewise believes in the possibility of calculation in the socialist society, says of Mr. Dickinson's solution. 2 " ... where the concepts have not been sufficiently refined, where a complex thing appears simple, there will be transferred from the capitalist to the socialist society institutions and organi zations which find themselves in conflict with their environment.

This kind of error is seen in Mr. Dickinson's use of statistically determined demand curves. In the capitalist society it is perhaps possible to get some estimate of a demand curve upon the assump tion in each case that other prices are constant. Mr. Dickinson transfers curves drawn up in this way to be added together, thus contradicting the assumption in each case that prices are constant." On one point Dr. Lerner' criticism seems to be built on faulty premises, and that is where he upbraids Dickinson for assuming that price is a function of only one variable. Dickinson specifically draws attention to the theoretical difficulties arising out of the demand for an article being a function of the price of all other articles. However, there are objections to others of Dickinson's proposals, inter alia to that that the prices of necessities shall be put as high as "the market will bear". If the state trebles the price of foodstuffs from one week to the other, ISee "Price Formation in a Socialist Community" in EconomicJournal, June, 1933.

2See "Economic Theory and Socialist Economy'" in Review of Economic ftudies.• 1934-5,P· 52.

80 ECONOMIC CALCULATION IN THE SOCIALIST SOCIETY the market will "bear" it, for people must have foodstuffs and these, in accordance with the assumptions, can only be obtained from the state's shops. Thus a price-policy which aimed at fully exploiting the state's monopolistic position, would not throw much light on con sumers' preferences. The people who ought to be particularly interested in the reliability of the data expressing consumers' preferences, are those who propose "the trial and error method" . We shall return to these solutions (see Chapter XI), and here discuss only the fixation of prices for necessaries. What the advocates of the "trial and error method" really mean is, that it is of no consequence what prices are originally fixed. In their opinion the fixing of prices-and the problem of economic calculation in the socialist community in general-can be solved if the central authority takes any surplus or shortage of necessaries as an indication of whether more or less of the article in question is to be produced.

The first to propose this trial and error method was Enrico Barone, 1 but its most ardent champion has been Professor Fred. M. Taylor. (See his paper The Guidance of Productionin a SocialistState, with which he opened the meeting of the American Economic Association in Chicago on 27th December, 1928.2) For Professor Taylor the problem is simple. He wants the central authority to fix temporary prices for all goods, including means of production, and goes on to say: "If, in regulating productive processes, the authorities were actually using for any particular f3'ctor a valuation which was too high or too low, that fact would soon disclose itself in unmistak able ways ... too high a valuation of any factor would cause the stock of that factor to show a surplus at the end of the pro ductive period."3 In the same way too Iowa valuation would lead to a shortage of the article or factor of production in question. Thus, after various attempts, one would be able to fix the correct accountancy prices for the factors of production. In his treatise On the Economic Theor:y of Socialism4 Dr.

Oskar Lange strongly supports Professor Taylor's solution. He mentions that it is the historically given prices which will be the basis for· "the process of successive trials", but says that one can just as well start with a set of fortuitous prices, e.g. by drawing numbers out of anurn. 5 As we have tried to demonstrate in this chapter, the question is far l"Uber die Okonomische Theorie der Planwirtschaft" in Zeitschrift fur Nationalok,onomie, Vol. V, Vienna, 1934. 2See American Economic Review, 192.9, p. 1 (also referred to on p. 4). 3Dp. cit., p. 7. 4See Review of Economic Studies, October, 1936. 5Dr. Lange is referring to Walras: Elementd'economiepolitiquepure.

MONEY AND FORMATION OF PRICES OF CONSUMER GOODS 81 from simple. The extent to which a surplus or shortage of goods will serve as data for consumers' preferences will depend on many factors, inter alia on the stability of the value of money, on the greater or smaller selection of goods, on whether the prices are variable or fixed, and, in the latter case, on whether the central authority's valuation coincides with that of the consumers. It must further be mentioned that the advocates of the "trial and error method" obviously assume that the community will be static, and expect that conditions will remain unaltered, while the "trial" is being made. On this assumption it is possible that the method suggested would one day lead to a satisfactory result, but the problem before us is to get a satisfactory result in a community where there is continuous variation. We shall come back to the other aspects of this attempted solution in Chapter XI.

The "trial and errormeth6d" has also been proposed by W. Crosby Roper Jr.l who thinks that a shortage of necessaries will be a "clear indication" 2 of what the central authority has to do. At the same time he is, however, fully aware of the difficulties and is strongly sceptical of being able to obtain reliable data: "This description of the process makes it seem rather simple and easily accomplished. It is a question, apparently, of adjusting a few mistakes at the beginning and then sitting down to watch the system work. But again we ignore the almost incredible com plication of the economic process. . . . At the establishment of a price system with perhaps only one or two considerable errors (an almost unbelievable assumption), those one or two errors would involve changes extending through the whole structure. If the number of serious mistakes were greater, it would take a con siderable time and a great deal of careful calculation to reach a position of equilibrium, where factors would be priced exactly according to marginal productivity, where these prices would be equal for factors of equal efficiency, and where the whole theoretical system of stable equilibrium was realised. As a matter of fact, this equilibrium could be reached only in a static economy which can never exist."3 Mr. Roper well expresses the fact that the indications are after all not so clear and that the consequences of incorrect data will be serious and far-reaching.

lSee The Problemof Pricingin a SocialistState, Cambridge, Mass., 1931. 20p. cit., p. 58. 30p. cit., pp. 58-59.