I. The Economic Basis of Class Conflict1
THE TASK which has been assigned to me in this symposium is to analyze the economic conditions which may give rise to conflict between different classes. Accepting the view that such conflicts arise from a disharmony of interests, I have to set out in what senses and on what occasions the disharmony may be said to be economic in origin.
2. Subjective and Objective Interest
Now it would be possible to spend much time making precise exactly what we mean by an "economic basis," an "economic interest" and so on: and in view of the great ambiguity of language here, it is probable that such inquiry would not be unprofitable. But for purposes of this paper I shall assume that we know roughly what is meant by such phrases, leaving any ambiguities to be disentangled in the course of the spoken discussion. There is one matter, however, on which it is necessary to be more explicit: the distinction between what I shall call subjective and objective communities of interest.
I can best explain shortly what I have in mind by means of an example. The members of a group of peasant proprietors believe that a certain line of common action will increase the value of their sales: such belief constitutes, I would say, a subjective community of interest. If at the same time it is true that, in the absence of disturbing circumstances, such action will actually have this effect, then, I would say, they have an objective community of interest. I should even say that there was an objective community of interest if such conditions existed unbeknown to the producers in question: for the real position would be such that, if the conditions were perceived, then common action would present certain attractions.
At first sight, this distinction may be thought to be very trivial. But I do not think that this is so. Of course it is possible that, to the majority of cases of subjective communities of interest, there may correspond the real conditions which are believed to exist. But there is nothing in the nature of things which would lead us to suppose that people are never mistaken. And certainly popular discussion of such matters suffers greatly from failure to keep the distinction in mind. How often do we hear it said that, whether or not the workers have actually anything to gain from class conflict, they believe that they have, and that therefore the conflict exists? And then, if it is objected that, in such and such a situation, no feeling of this sort has been present, it is replied that, in any case, the underlying conditions are such that, whether it is realized or not, there is an actual class conflict.
3. The Market Theory of Group Conflict
Now my task, as I understand it, is to investigate the basis of clashes of objective communities of interest. No doubt it is always belief which is the immediate cause of action. It is important, therefore, to trace the causes of all beliefs, whether they be correct or erroneous. But that is not my object here. However much we are interested in the causes of error, be they in any sense economic or otherwise, what we want to discover in these discussions is whether there are inherent in certain social structures, not merely optical illusions, but objective disharmonies which exist whether perceived or not.
Having said this, I ought perhaps to say at once that I believe in the possibility of such disharmonies. I do not believe that, under any conceivable conditions, social life is entirely free from potential conflict arising from such causes: and I certainly do not think that, in the absence of suitable institutional restraints and remedies, there is any inherent tendency to harmony which will prevent such conflicts emerging. If there be any "invisible hand" in a noncollectivist order, it only operates in a framework of deliberately contrived law and order. What is more to the point, I believe that simple economic analysis casts considerable light on these matters.
But — and perhaps this is where one who approaches these matters from the economic point of view may differ from his fellow sociologists — I do not think we are likely to secure results which are very helpful if we start, in advance of our analysis, by laying down hard-and-fast definitions of classes and then asking if anything can be said about them from this or that point of view. For a class, after all, is not something which is necessarily given independent of the forces which are operative in society; and I think we shall be much more likely to get a proper sense of proportion if we look first at the conditions conducive to conflict and then arrange our classifications accordingly than if we proceed in the reverse direction. For this reason I cannot believe that it is the most satisfactory starting point to assume that the only classification worth discussing is a vertical stratification. I do indeed think that there is something to say about vertical stratification; but I am inclined to think that we get a false sense of its proportionate importance if we ignore other groupings, which are certainly not essentially vertical.
The clue to a successful approach to the problem, as I see it, lies in the analysis of markets. In the exchange society the market is, as it were, the reflection of the whole network of economic relationships. The market phenomena, the prices of products and factors, provide the immediate stimuli to action, the indices according to which men arrange their consumption and production. There may be matters lying beyond immediate market analysis to which we should also give attention. But it is surely to the market that we should look first if we wish to understand the economic basis of conflict.
Let us therefore start by considering the interests of groups, homogeneous in their market relations to particular goods or services, either as buyers or sellers. We can at once discern two main types of situation in which there may exist conflict between the interest of such groups and the interests of others: firstly when, as individuals, they are confronted by monopolistic groups: secondly when, as groups, their position is contrasted with others.
Let us consider each of these cases separately.
(i) The case where individual buyers and sellers are confronted with monopoly need not detain us long. It is well known that, if supply or demand is in the hands of a monopoly, the prices paid or the prices realized by the individuals with whom the monopoly deals may be less favorable than if the market were competitive. If, for instance, peasant producers are confronted with a single buyer or small number of buyers acting in concert, they may feel that their interests are damaged by such conditions; and very often they will be right. Similarly, if they buy their implements from a monopoly or if they borrow money from a closed group of moneylenders, there will exist a like community of interest. Situations of this sort will often be ephemeral, scarcely worthy to be dignified by the name of group or class conflict. But in certain conditions, where mobility is low, they may persist or tend to repeat themselves: and in such cases far-reaching divisions may appear. I have mentioned already the agrarian case. I am inclined to think that some, though certainly not the larger part, of the history of trade unionism is to be explained in similar terms. The only sense which the economist can attach to the term exploitation of labor is as a description of what happens where a group of competing workers is confronted by a monopolistic buyer. The frequency of such a situation is capable of great exaggeration. But when it does occur, and when for long periods no competing demand arises on the employers' side, sharp conflict is likely to be engendered.
This is all very simple. I ought perhaps to add at this point that by "monopoly" I mean monopoly in the sense of control of the sources of supply or demand by one seller or buyer or, at most, by a small number. I do not mean conditions such that, under competition, the individuals on one side of the market get higher incomes than their opposite numbers. If, for instance, Lancashire were the sole source of supply of cotton goods and if all the manufacturing units there were controlled by one group of sellers, I should say that there was a Lancashire monopoly. I should not say that there was a monopoly, however, if the different mills in Lancashire were in competition with each other, even if the proceeds of sale gave incomes there higher than elsewhere. The word "monopoly," if applied to such a case, can only mean scarcity, and although it has been so applied by many economic historians and by a few competent economists, I believe that it is a usage which can only cause confusion and which conceals differences which it is important to keep clearly in view.2
(ii) Let us now turn to the second case in which conflict is incipient — the case where the interest of the group as group is in disharmony with the interests of others. In a purely formal sense this is simply the opposite side of the picture we have just been examining. But it has an interest of its own quite sufficient to warrant separate examination.
Speaking broadly, we may say that the interest of such a group acting as a group is to maximize its net takings: in the first approximation we may neglect the other advantages and disadvantages which will actually influence action. Now this is an aim which may well be in conflict with the interests of other members of society. If, for instance, the conditions of demand are such that a smaller quantity sells for a higher aggregate value, then usually the group interest calls for restriction; whereas the interests of society demand the contrary. It is said that a century or so ago, when high transport costs still limited supplies of foreign grain to English markets, it was an acceptable toast for English farmers to drink to "a bloody war and a wet harvest." Here clearly was a sharp conflict of interest.
It would be possible to dilate at length on the exact nature of the market conditions which lead to different degrees of such disharmony. But these are matters of technique well discussed in standard works, and for our purposes they are less interesting than certain broader considerations.3
Let us note in the first place that, so far as the cash incentive is concerned, the conflict we have discovered is between society (or other groups) and the members of the group acting as a group. It does not exist between society and the members of the group acting individually. For each member of the group, with any given degree of restriction, the more he himself can sell the better. When the individual farmer raised his glass to the dreadful toast we have just been considering, he must have murmured, "But, Lord, don't let the rain fall on my crop!" It is only if individuals are able to act as a closed group that the latent conflict becomes actual. When it is said that the infringers of restriction agreements act in an "antisocial" manner, what is meant is that they are acting in an antigroup manner. "Social" conduct in the sense of group loyalty may quite easily lead to social conflict. The qualities which lead a man to be loyal to his profession or to his industry in a maneuver to raise prices, or to prevent prices from adjusting themselves to a state of greater abundance, may, according to some systems of ethics, be thought to be valuable in themselves. But, from the point of view of the rest of society, it is the interloper who acts in the social interest.
But let us notice, in the second place, that so long as the services of the producers are not specialized and so long as neither they nor the members of other groups enjoy permanent monopolistic status, the conflict of interest we have discovered may itself not be permanent in character.
We can, perhaps, best see what is involved here if we take an example rather more dynamic in character than the case of the farmers and the wet harvest. Let us suppose that there occurs an invention which facilitates the production of wheat. If the conditions of production are competitive, there will be an incentive to individual producers to utilize this invention; it will diminish costs of production. But if the elasticity of demand for wheat is low, then, if the supply is increased, the fall in price necessary to clear the market may be such as to lower receipts even more than the lowering of costs of production. From the point of view of the group therefore, for the time being at least, there is a community of interest in conflict with the interests of consumers; and if group action can be organized through the instrumentality of the state or by spontaneous agreement to limit competition, there is an incentive to restrict the exploitation of the invention.
But suppose that this does not happen, and suppose that the producers of wheat are free to turn their activities and resources to the production of other commodities. What happens? If we assume monetary conditions conducive to full employment, we may expect transfers to other lines of production to take place, until the incomes of transferable factors in wheat production have risen once more so as to be as favorable as the incomes obtainable elsewhere for similar services. The mobile producers, as producers, are no worse off than before. As consumers they share with others the benefits of the cheapening invention.
But notice that this is only possible if there is mobility. There must be no obstacles to the entry of new producers to other branches of production: that is to say, there must not be group monopoly elsewhere. Furthermore, the resources concerned must be capable of employment elsewhere: they must not be specialized on the line of production at first affected. Now this is not always the case. Young laborers may be able to turn to other things. Older men may find it more difficult. Moreover, material resources may be so specialized that changes of the sort we are discussing may involve a permanent lowering of their yield: in the case we have been considering it is probable that, whatever the eventual position of farmers and laborers, the owners of land will suffer some tendency to a permanent lowering of their rents. In such cases, for factors which are more or less specific to particular branches of production, there may be a long-period as well as a short-period conflict of their interest and the interest of other groups. It is of course arguable that, even here, the balance of long-run advantage is against restriction: if restriction becomes general, there is likely to be loss all around. But important as are such considerations — and on a broad view they are very important — they are on a plane of analysis different from that on which we are moving at present. If there is no mobility, then conflicts of interest of the kind we have been discussing are clearly possible.
4. The Significance of Market Homogeneity
Such, in very general and loose terms, is an outline of what I would venture to call the market theory of group conflict. When the conditions of supply and demand are such as either to confront buyers and sellers with monopolistic organizations or to permit buyers or sellers themselves to act as groups, then the objective conditions of conflict are present. It is not certain that consciousness of these conditions will always develop, or that, if it does, it will necessarily be so important and permanent as to give rise to anything which may usefully be described as class conflict. Clearly this depends partly upon the degree of the disharmony, partly upon the apparatus of communication, partly upon propaganda and leadership. In a society in which many different changes are taking place at the same time and in which the general institutional framework is conducive to mobility, many minor conflicts of this sort may escape notice or cancel each other out. But obviously this may not happen. If there are extensive institutional influences conducive to an absence of mobility and the creation of positions of privilege, conflicts may develop which are certainly worthy of the term class conflicts. In societies in which producers' organizations are given powers of exclusion based on statute or coercive custom, such conflicts must be permanently present. It is not difficult to think of historical examples: the guild societies of the European Middle Ages, the caste societies of the East. If we try to get the social phenomena of our own time into proper perspective, it is arguable that the European agrarian reaction, which is based essentially upon a clash of interests of this nature, is quantitatively more important than any other conflict of the day. And certainly the stream of tendencies which, when they appear on the right, are known as corporativism and, when they appear on the left, as self-government in industry, works strongly to create such conflicts. Speaking generally, I would be prepared to argue that the whole theory of social classes has tended to suffer from a serious lack of proportion because of a neglect of conflicts of this kind. Preconceived ideas of social stratification have prevented us perceiving other classifications which are historically and actually at least of comparable importance.
Now there is one feature of this theory which has already been mentioned in passing but which is so fundamental that it deserves a separate emphasis: the groups whose communities of interest are thus analyzed are homogeneous in relation to the market. Either they sell or they buy commodities which the market ranks as similar. This does not mean, of course, that the commodities concerned need be exactly the same. Commodities which are closely related may be the subject of a community of interest. Insofar as they are competing substitutes, one grade of wheat may be the subject of a community of interest as against other grades: but there may be a wider community of interest relating to wheat in general. Obviously it is a matter of degree. We must not draw hard-and-fast lines. But this should not blind us to the fact that, as market homogeneity weakens, so the objective community of interest weakens too: and there comes a point at which it disappears altogether. Because the markets for their various products move together there may be a potential community of interest between producers of various types of cereals. There is no community of interest of this sort between the producers of, say, cereals and coffee.
5. The Theory of the Class War
All this becomes highly relevant when we proceed to apply our apparatus to the examination of what some think is the most important conflict of class interest, the alleged conflict of interest between the propertied and the propertyless — the class war of Marxian theory.
I do not think we shall be doing great injustice to the orthodox theory of this subject if we take as our starting point its international character. In The Communist Manifesto it is the workers of the world who are exhorted to unite. And since it was necessary to exhort them to unite, and since it is clear that even today there is no very marked international proletarian solidarity, we are surely on fairly firm ground if we assume that it is an objective community of interest between the propertyless, an objective clash between their interest and the rest of society which is in question. Whatever the apologia on the part of what I believe would be called subjectivist epigoni, the main Marxian theory asserts an objective community of interest among the proletariat the world over.
But, judged by our tests, this assertion does not seem immediately plausible. For the different groups of laborers in the different national areas do not supply services which the market treats as homogeneous. They do not sell identical commodities. In the present organization of the world into different national groups, with its almost total prohibition of international migration, the interests of laborers in different parts of the world are often violently opposed to each other. Measures which perpetuate the poverty of Japanese laborers are beneficial to the Lancashire cotton operative. Indeed, it can easily be shown that not infrequently there is more solidarity between the interests of groups of laborers in one place and groups of property owners in another than between the laborers of the two places in question. The restriction of migration of Prussian peasants into the United States of America, for instance, which benefits certain classes of American labor, benefits also certain classes of Prussian landlord: in the one case the relative price of certain kinds of labor is maintained; in the other, the relative price of certain kinds of land. There is a sense in which it may be said that the long-run interests of the majority of the human race are not in fundamental disharmony. Our discussion of the significance of mobility and the general interest in restricting restriction, perhaps, shows the way in which such a proposition could be developed. But an objective community of interest of the various individuals composing the international proletariat is not disclosed by the methods of analysis which we have so far been using.
I stress this point because, despite the overwhelming realistic and quantitative significance of the international problem, it has become the fashion to neglect it. For reasons which are not easy to follow, attempts to argue the matter on the international plane are looked upon as bad form or as pedantry. But it is surely not bad form to discuss the merits of a theory on the plane chosen by its proponents: and it is hardly pedantry to discuss the world as it is rather than a hypothetical closed society. I think that people have got into far too facile a habit of saying simply that "of course" national divisions may "overlay" the divisions of social classes, and then going on to assume that the creation of a classless society within each national area would automatically eliminate these divisions between national areas. There is, in fact, as I have argued elsewhere, strong reason to suppose that it would do exactly the contrary.4
Nevertheless, for the sake of argument, let us abstract from the complications of national grouping. Let us inquire concerning the community of interest of the propertyless in a closed society, which, however, in other respects has the extent and the complexity of the world as we actually know it. Even so, I suggest the classification does not conform to our criteria. It is true that the classical economists were in the habit of generalizing widely about a class of factors of production which they labeled "labor." But I do not think that this is a part of their work which has shown itself particularly valuable for purposes of realistic analysis. There is not one, there are many separate types of labor services: and their several relations in production and in the market do not give rise to a general community of interest of the type we have been discussing. On the contrary, the interests of the different groups are often sharply opposed: and if exclusive group organization is possible, violent conflict may develop. We have only to think of the relations between male and female labor, skilled and unskilled, white and colored5 — to adopt classifications only one degree less crude than that which we are discussing — to convince ourselves of this. Measures which benefit one class may injure another and injure the other far more than any injury inflicted on property owners.
But let us once more waive such realistic objections. Let us imagine that we are dealing with that remote world in which all labor is homogeneous in efficiency and disposition and in which the phrase "the labor market" is not a simplification but a reality. Are we here confronted with an objective clash of interest in either of the senses of our market theory? I doubt it. For, in the first place, it does not seem plausible to argue as if such a group were confronted by a general monopoly on the part of the employers. I do not deny the existence of buyers' monopoly in particular branches of industry: indeed it has been part of my contention that, where such conditions exist, there group conflicts are latent. But a universal monopoly of employers is something very different from this: and I do not think there is any empirical ground for regarding it as a plausible assumption. A universal employers' monopoly is a chimera of the imagination. Nor, in the second place, do I think the situation is to be saved by assuming that the laborers have anything to gain from restriction. If the elasticity of demand for labor were less than unity, then certainly there would exist disharmony, at any rate in the short run; a restriction of the supply of labor would cause the rate of wages to rise more than proportionately, so that the total amount of wages would increase. But I have yet to see a demonstration that this is likely to be the case. The economists known to me who have thought it worthwhile to speculate about so highly unrealistic an instance, have come to a different conclusion. I will not press this point, for it seems to me rather futile to attempt an empirical solution of a problem which is not given empirically. But I would suggest that, if the theory of objective clashes of interest here is to be anything more than mere dogmatic assertion, some other approach is needed.
6. Class Conflict and Inequality of Opportunity
But cannot such an approach be provided? It is certainly worthwhile making the attempt.
Let us revert to our former examples. It will be remembered that we argued that if producers were confined to particular lines of production by inability to secure opportunities elsewhere, then objective conflicts of interest were present. Now, in the purely individualistic society there are opportunities for the sale of some kinds of labor which are only accessible to those who have command of capital, either for expenditure on training or for security against the risks of enterprise. It is not necessary to quote examples: they are the commonplaces of the elementary textbooks. Is it not the case that those who lack such advantages are confined to work at margins lower than otherwise would be necessary? And if this is so, is this not a sense in which there may be said to be an objective conflict of interest between the propertied and the propertyless?
We seem to be on firmer ground now: and I do not doubt that there is a core of truth here which is important. But let us be quite sure what exactly is involved.
In the first place we should observe that the degree of conflict here depends essentially on the inequality of income: it is only insofar as inequality of income is due to inequality of property holding that the category of property is involved.6 It follows, therefore, that the degree of conflict will in part depend on the distribution of property. In any society in which there exists inequality of property holding there are bound to exist some inequalities of economic opportunity. But the extent to which this becomes a serious social problem will depend essentially on the degree of inequality. In a society in which differences of property holding were narrow, the problem would be much less formidable than in a society in which differences of property holding were great. Before, therefore, it can be claimed that such conflicts are inevitable in a society in which private property is permitted, it must be shown that there is inevitability about the present framework of law relating to property. And, in spite of the enthusiastic support given to the view by Marxians, I am yet to be persuaded that it is true.7
Secondly, I do not think that the situation which we are now contemplating is one which can be most conveniently described as one of monopolistic exclusion. If we are properly to grasp the rather complex position here, it is necessary to make certain distinctions. If entry to a profession, for instance, were only to be secured by the payment of a heavy fee quite unrelated to the cost of training, I should agree that the use of the term monopoly was appropriate. If it were limited to members of certain race or caste, again I should hold the term to be justified. The caste society is the monopolistic society par excellence: it is a society in which economic conflict of interest must be permanent. But if the position of advantage were one to which anyone with property of a certain amount might have access who incurred the necessary expenses, and if there were no legal obstacles to accumulation, then I should not use this term. Membership of a certain income group is not the same as membership of an exclusive caste or estate. There are points of similarity. But there are also points of difference. The position may be the same as regards advantage to exploit a given opportunity: it would be silly to pretend that the fact that a man may accumulate property in the future gives him the same advantage at any moment as the actual possession of property. But, whereas the advantage of a closed caste or estate is based on a right of exclusion that serves no economic function whatever, the advantage that comes from property comes from an institution that, in some forms at any rate, can be said to discharge a function that is necessary and that, in some way or other, would have to be discharged whatever the structure of society. If the institution of private property is necessary to secure an organization of production adequate to the demands of society, the inequalities that it may engender cannot be described as restrictionist. There will be more to be said about this later on. But I submit that it constitutes a difference of substance.
Nevertheless, when account is taken of all these points, there can be no doubt that inequality of opportunity due to inequality of income does create clashes of interest in a significant and objective sense. And although I hold that some inequality of income rests upon institutions that discharge a necessary social function and a function whose benefits are not confined to the property holder, I hold too that, up to the point at which attempts to remedy such inequality take a form that definitely reacts unfavorably on the long-run interest of the inferior income groups, these clashes continue to exist. I would add that, in my judgment, in rich countries there is probably a considerable margin before this point is reached. Often as it is misrepresented for propagandist purposes, belief in a system of international liberalism does not exclude belief in energetic measures designed to reduce inequality of opportunity.
7. The Class War and the System of Free Enterprise
But is this all that can be said about the matter? The Marxian case is certainly not limited to inequalities of work income due to inequalities of opportunity: it concerns the whole range of inequalities of income due to the existence of private property of any kind. Is there no sense in which it can be said that the propertyless would be better off if the whole institution of private property were abolished?
Now it is possible that, in the very short run, a general expropriation of property and sharing out of the proceeds might raise the wealth immediately available to the average proletarian.8 This obviously would apply to any sort of confiscatory transfer between other "classes" — Jews and Aryans, whites and blacks, men and women, etc. In the same way, if we consider forms of taxation and subsidy which run far short of complete expropriation, similar gains and losses are discernible: if this paper were a comprehensive treatment of its subject, much space would have to be devoted to conflicts in the sphere of public finance. But it is not in this short-run sense of mere "divisionism" that the Marxian asserts a conflict of interest. The Marxian equally with the liberal knows that a mere process of redistribution settles nothing. He asserts rather that in the long run, too, there is conflict inasmuch as the productive functions now discharged by private ownership could be better discharged another way.
To investigate the legitimacy of such a claim obviously lies outside the scope of this paper. But I think it is worthwhile insisting that a mere assertion of this sort does not itself establish an objective conflict of interest: the assertion must be proved. And in order that it may be proved, it is necessary to show that a society in which private ownership of the means of production has been abolished will satisfy the demands of the citizens of the world better than before. It is necessary to show that this is possible, not merely on a provincial or a national, but also on an international scale. And, while I am aware that there are some who think that such proof can be given, I venture to suggest that nothing of the kind has as yet been generally accepted, and that, indeed, the expectations of such a society which have hitherto been prevalent have rested on assertions and theorems which even the most enthusiastic socialist of today would admit to have been mistaken.9 Certainly the actual experience of those states which have approached nearest to the ideal of total collectivism affords little support to optimism in this respect.
Hitherto, then, such community of interest of the international proletariat as has rested on this claim has been subjective rather than objective in nature. The objective clashes of interest which can be actually demonstrated to operate in the world of reality suggest a classification of social groups more related to the phenomena of the market and to possible limitations on industrial and international mobility than any which rest on a general division between the propertied and the propertyless. In such a classification horizontal clashes of interest are at least as important as vertical.
1. This paper was first presented as part of a symposium on "Class Conflict and Social Stratification," held under the auspices of the Institute of Sociology in the autumn of 1937.
2. For a further discussion of this point see pp. 49–50 below.
3. Such matters are further discussed below in the paper on The Economics of Restrictionism.
4. Economic Planning and International Order, pp. 60–68.
5. On the conflict between labor groups of various racial origins see A. Plant, "The Economics of the Native Question," Voorslag: A Magazine of South African Life and Art (Durban), 1, no. 2, pp. 21–38; and W.H. Hutt, "The Economic Position of the Bantu in South Africa," in Western Civilization and the Natives of South Africa (Routledge). It is greatly to be regretted that Professor Plant's highly important paper is not available in a more accessible form.
6. There would be inequality of opportunity in a propertyless society if incomes were markedly unequal and if education and training were financed only by the family unit.
7. I have suggested elsewhere (Economic Planning and International Order, pp. 264–265) that the present distribution of property could be substantially modified in the direction of greater equality by the substitution of the succession duty principle for the principle on which the estate duties are now levied.
8. On a world scale the amount available for redistribution per head would be small. The more important sources for leveling up the lower incomes would be the high work incomes of the West!
9. It is probable that Marx himself thought that computation in terms of labor costs would be an adequate basis for socialist planning.