Chapter V

Joint Demand and Supply

§1. Marginal Cost under Joint Supply. Several references have been made above to joint products, a relation which it will be convenient now to describe as that of Joint Supply. Our sense of symmetry should make us look for a parallel relation on the side of demand; and it is not far to seek. There is a "joint demand" for carriages and horses, for golf clubs and golf balls, for pens and ink, for the many groups of things which we use together in ordinary life. But the most important instances of Joint Demand are to be found when we pass from consumers' to producers' goods. There, indeed, Joint Demand is the universal rule. Iron ore, coal and the services of many grades of operatives are all jointly demanded for the production of steel; wool, textile machinery and again the services of many operatives are jointly demanded for the production of woollen goods (to mention in each case only a few things out of a very extensive list). Now we have already noted that, when commodities are jointly supplied, there is an obvious difficulty in allocating to each of them its proper share of the joint cost of production. There is a similar difficulty in estimating the utility of a commodity which is demanded jointly with others. Thus, the utility of wool is derived from that of the woollen goods which it helps to make. But the utility of the factories, the machinery and the operatives employed in the woollen and worsted industries is derived from precisely the same source. How much, then, of the utility of woollen goods should be attributed to the wool and how much to the textile machinery? Can we make any sense of the notion of utility as applying to one of these things, taken by itself? And, if not, how can we explain the price of a thing like wool in terms of utility and cost, since we cannot disentangle its cost from that of mutton, nor its utility from that of a great variety of other things?

Here the conception of the margin enables us to grapple with a problem which would otherwise be insoluble. For, while it is impossible to separate out the total utility and cost of wool, it is not impossible to disentangle its marginal utility and its marginal cost. The proportion in which wool and mutton are supplied cannot be radically transformed; but it can be varied within certain limits, by rearing, for instance, a different breed of sheep. Variations of this kind have been an important feature of the economic history of Australasia, where sheep farming is the leading industry. Before the days of cold storage, Australia and New Zealand could not export their mutton to European markets, though they could export their wool. Wool was accordingly much the most valuable product; the mutton was sold in the home markets, where, the supply being very plentiful, the price was very low. In the circumstances, the Australasian farmers naturally concentrated on breeding a variety of sheep whose wool-yielding were superior to their mutton-yielding qualities. The development of the arts of refrigeration led in the eighties to an important change. It became possible to obtain relatively high prices for frozen mutton in overseas markets. There was, therefore, a marked tendency, especially in New Zealand, to substitute, for the merino, the crossbred sheep which yields a larger quantity of mutton and a smaller quantity of wool of poorer quality. Now if we calculate the cost of maintaining the number of merino sheep which will yield a given quantity of wool, and calculate the cost of maintaining the larger number of crossbred sheep which will be required to yield the same quantity of wool (allowing for differences of quality) the extra cost which would be incurred in the latter case must be attributed entirely to the extra mutton that would be obtained. This extra cost we can regard as constituting the marginal cost of mutton. So long as this marginal cost falls short of the price of mutton, it will be profitable to extend further the substitution of crossbred for merino sheep. The process of substitution will in fact be continued until we reach the point at which the marginal cost is about equal to the price. Similarly by starting with the numbers of merino and crossbred sheep which would yield the same quantity of mutton, we can calculate the marginal cost of wool; and again the tendency will be for this marginal cost to be equal to the price.[1]

§2. Marginal Utility under Joint Demand. On the side of demand there exist as a rule similar possibilities of variation. Some machinery, some labor, some materials of various kinds, are all indispensable in the production of any manufactured commodity. But the proportions in which these factors are combined together can be varied, and are frequently varied in practice as the result of the ceaseless pursuit of economy by business men. To produce pig-iron, you need both coal and iron ore; but, if coal becomes more costly, it is possible to economize its use. Machinery and labor must be used together, in some cases in proportions which are absolutely fixed. But there is in nearly every industry a debated question as to whether the introduction of some further labor-saving machine would be worth while, or some improved machine which would represent the substitution of more capital plus less labor for less capital plus more labor. A farmer can cultivate his land, to use a common expression, more intensively or less intensively; in other words, he can apply larger or smaller quantities of capital and labor (the proportion between which he can also vary) to the same amount of land. The problem is essentially the same as that of the substitution of the crossbred for the merino. We can take the various possible combinations of the factors of production, and contrast two cases in which different quantities of one factor are employed, together with equal quantities of the others. The extra product which will be yielded in the case in which the larger quantity of the varying factor is employed can then be regarded as the marginal product (or marginal utility) of the extra quantity of that factor; and we can say that the employment of this factor will be pushed forward to the point where this marginal product will be roughly equal to the price that must be paid for it. We can thus lay down the most important proposition that the relation between marginal utility and price holds good generally of the ultimate agents of production; that the rent of land, the wages of labor, and, we can even add, the profits of capital tend to equal their (derived) marginal utilities, or, as it is sometimes expressed, their marginal net products.

Whenever, therefore, the proportions in which two or more things are produced or used together can be varied, the relations of joint supply and joint demand are perfectly consistent with a specific marginal cost and marginal utility for each commodity.

§3. A contrast between Cotton and Cotton-seed, and Wool and Mutton. But it sometimes happens that such variations cannot be made. Thus, it has not been found possible (so far as I am aware) to alter the proportions in which cotton lint and cotton-seed are yielded by the cotton plant. Roughly speaking, you get about 2 pounds of cotton-seed for every 1 pound of cotton lint (or raw cotton), and though this proportion may vary somewhat from plantation to plantation, it is upon the knees of the gods, and not upon the will of the planter that the variation depends. We cannot, therefore, speak with accuracy of the separate marginal costs of raw cotton and cotton-seed. It is true that some plantations are so far distant from any seed-crushing mill that it is not worth while to sell the seed as a commercial product; and it might seem, therefore, as though we might regard the entire costs of cotton growing on such plantations as constituting the marginal costs of raw cotton. But planters, so situated, derive a considerable value from their cotton-seed by using it as fodder for their live stock or as a manure. You can, of course, argue that proper allowance is automatically made for this factor, as a deduction from the costs of raw cotton, when you add up the expenses of the plantation. In the same way you can deduct the price which a planter who sells his cotton-seed obtains for it, from the total costs of the plantation, and call the remainder the costs of the raw cotton. But this is really to reason in a circle. For in either case the magnitude of the deduction depends on the marginal utility of the cotton-seed. And the notion of the cost of anything becomes blurred and blunted if we so use it that it must be deduced from the utility of something else, which is not an agent in the production of the thing in question.

This point is not merely an academic one. It means that we cannot explain the relative prices of cotton lint and cotton-seed in terms of cost at all, whether marginal or otherwise. The influence of cost will be confined to the sum of the prices of the two things. Upon this sum it will exert precisely the same influence as it exerts upon price in general, by affecting the total quantities of the two things that will be supplied. But upon the distribution of this sum between lint and seed, cost will exert no influence whatever, because it cannot affect the proportions in which they are supplied. It may assist some readers if I state the matter in more concrete terms. Cost of production will be one of the factors which will result in the production of an annual cotton crop in the United States of, let us say, 10 million tons of seed cotton. This crop will yield roughly 6-2/3 million tons of cotton-seed, and 3-1/3 million tons (or rather more than 13 million bales) of lint. The combined price received by the planter of (let us say) 14.4 cents for 1 pound of lint plus 2 pounds of seed should correspond roughly to the marginal joint costs of production. But the factor of cost has no influence at all in determining that this combined price is made up of a price of 12 cents per pound for lint, and only 1.2 cents per pound (or $24 per ton) for cotton-seed. To account for this we must rely entirely upon demand. We can say, shortly, that the respective prices must be such as will enable the demand to carry off 6-2/3 million tons of seed, and 3-1/3 million tons of raw cotton. Or we can go further and say that the marginal utility of a pound of raw cotton, when 3-1/3 million tons are supplied, is ten times as great as that of a pound of seed when 6-2/3 million tons are supplied.

If accordingly the demand for cotton-seed were to expand considerably owing, say, to the discovery of some new use for the oil, which is its most valuable constituent; the effect would be first a rise in the price of cotton-seed, and, subsequently, by stimulating cotton growing, a more plentiful supply and a lower price for raw cotton. And so far at least as the increased supply is concerned, this must necessarily be the effect, "other things being equal"; though, to be sure, it might be outweighed and obscured by other influences such as the boll-weevil. But it is not the case that an increased demand for mutton must necessarily increase the supply or lower the price of wool; and it is most unlikely to do so in any similar degree. For, here, the separate marginal costs of the two things exert their influence. An increased demand for mutton will stimulate sheep farming, but it will also stimulate the substitution of crossbred for merino breeds; and the resultant of these two opposite tendencies upon the supply of wool is logically indeterminate. As a matter of history we know that the development of cold storage in the eighties (which we may regard for the present purpose as equivalent to an increased demand for Australian mutton) caused considerable perturbation in the woollen and worsted industries of Yorkshire. They were faced with a dwindling supply and a soaring price of merino wool; and the adaptability with which they met the situation, and won prestige for the crossbred tops, and yarns and fabrics, to which they largely turned is a matter of just pride in the trade to-day. The fact, however, that this alteration in the supply of wool was a matter not only of quantity but of quality, while it takes nothing from the substance of the preceding argument, makes it difficult to draw a clear moral, bearing on the present issue, from this incursion into history.

§4. The Importance of being Unimportant. The above contrast between cases in which variation is possible, and those in which it is not possible, is reproduced with a heightened significance when we turn back to joint demand. The cases are perhaps less common in which it is impossible to alter the proportions in which different commodities are jointly demanded, but there are many cases in which it is not nearly worth while to do so (and this amounts to very much the same thing). Cases of this sort are especially likely to occur when we are dealing with a commodity which accounts for only a tiny fraction of the costs of the industry which is its chief consumer. Sewing cotton, for example, is jointly demanded, with many other things, by the tailoring and other clothing trades; but the money which these trades spend on sewing cotton is so small a part of their total expenditure, that no ordinary variation in its price is likely to make it worth while to study the ways and means of using it in smaller quantities. When sewing cotton is bought by the domestic consumer, considerations which are fundamentally the same, though somewhat different in form, point to a similar conclusion. It is thus very difficult to assign to sewing cotton a specific marginal utility. This difficulty is of great importance in connection with the possibilities of monopolistic exploitation. For it means that the demand blade of the scissors upon which we rely to cut off excrescences of price is blunted, and if accordingly the producers constitute a strong enough combination to control the supply blade, they will possess an unusual power of advancing their selling prices as they choose. I am far from suggesting that Messrs. J. & P. Coats are to be condemned as an extortionate monopoly. On the contrary, during 1919, when the profits in highly competitive industries like the main branches of the cotton and woollen trades, soared exuberantly, the record of this concern seems to me one of distinct moderation. But the present point is that they possess an exceptional power to fix the price of sewing cotton as they choose, and that this is attributable in no small degree to the fact that sewing cotton constitutes an essential but relatively trifling item in the expenses of the processes in which it is employed.

Perhaps the point will be made clearer if we turn from the selling prices of commercial products, in regard to which there is a strong and not ineffective public sentiment against "profiteering," to the remuneration of different classes of labor. With an instinctive disposition towards megalomania, it is often claimed in Great Britain that the miners, being a very numerous and well-organized body of workpeople, were in a stronger strategic position than most workpeople for exacting the remuneration they desire. It is quite true that a stoppage of work in the coal industry causes us a high degree of inconvenience, and temporary concessions may thereby be obtained which might otherwise have been refused. But this is a dubious advantage, and we grossly exaggerate its real importance. The truth is that the strategic position of the miners in regard to wages questions is by no means strong. For their wages constitute a very large percentage of the cost of coal; and the price of coal in its turn is a most important element in the costs of many of the industries which are its principal consumers. Great Britain, moreover, is far from possessing a monopoly of coal. If, accordingly, the wages of the miners are temporarily pushed up to a high point, the result will certainly be a diminished demand for British coal, which will lead before long to their fighting a losing battle to maintain the concessions they have won. Contrast their position with that of the steel smelters, whose wages (high though the wage rates are) constitute a very small percentage of the costs of steel production, and we must agree I think that we have in this distinction the main reason why the steel smelters, though they hardly ever go on strike, have as a rule been able to do so much better for themselves than the miners.

When a commodity or service is such that an appreciable alteration in its price has only a slight effect upon the quantity demanded, the demand is said to be inelastic. Conversely, when a small change in price greatly alters the quantity demanded, we call the demand elastic. In the former case, it is worth nothing, a larger aggregate sum of money will be spent upon the thing when its price is high than when it is low, while the opposite is true in the latter case. This distinction is of considerable importance in connection with many problems (e.g. of taxation); and the terms, elastic demand and inelastic demand, are worth remembering. We may thus express the above conclusions by saying that the demand for sewing-cotton is highly inelastic, and that the demand for coal miners is more elastic than that for steel smelters.

§5. Capital and Labor. Cases in which it is impracticable to make any variation in the proportions in which different things are used together are, however, the exception rather than the rule. Where variation is possible, we are confronted with an uncertainty as to the way in which an increased supply of one thing will react on the demand for another, similar to our uncertainty as to whether an increased demand for mutton would augment or diminish the supply of wool. It is, for instance, of the highest importance to give a clear answer, if we can, to the question whether an increased supply of capital will increase the demand for labor. The chief effect of an increased supply of capital is to facilitate the extended use of expensive machines: to some extent these machines will increase the demand for labor; to some extent they will be substituted for it. Which of these two tendencies will outweigh the other we cannot be absolutely sure. But fortunately we can be far more nearly sure than was possible in the analogous case of wool and mutton. An increase in the supply of capital increases the demand for the commodities, from which the demand for labor is derived, in both the senses discussed in Chapter II. First it makes them cheaper to buy, and thus increases the quantity that will be bought. It is this that is parallel to the effect of an increased demand for mutton in making it more profitable to breed sheep. But it also serves to increase the purchasing power with which to buy commodities, because it increases the aggregate real wealth of the community, and it thus serves to raise the whole demand curve. This last consideration is so important as to make it overwhelmingly probable, apart from the evidence of history, that an increase in the supply of capital (and the same may be said of an increase in the supply of the other agents of production) will on balance increase the demand for labor. The evidence of history points to the same conclusion. The history of the last hundred years displays an unprecedented accumulation of capital, and an unprecedented extension of machinery, associated with an unprecedented improvement in the standard of living throughout the whole community. This is powerful testimony in favor of the view that an increase in the supply of capital and the use of machinery will usually enhance on balance the demand for labor. Moreover, though this is not conclusive, there is little room for doubt that an obstructive attitude towards the extension of machinery in a particular country, or a particular district, is misguided. For its effect must be to make production more costly there than it is elsewhere, and to lead, slowly perhaps, but very surely, to the transference of the industry to other regions.

§6. Conclusions as to Joint Supply and Joint Demand. Here, however, we are beginning to digress. Let us sum up in a general form our conclusions as to the way in which changes in the supply or demand of a commodity react upon the demand or supply of the other things with which it is jointly demanded or supplied. Everything turns, as we have seen, on the possibility of variation in the proportions in which the things are used or produced together; and this, it is also clear, is a matter of degree. Our conclusions, therefore, had best take the following form:—

§7. Composite Supply and Composite Demand. Joint Demand and Joint Supply do not complete the list of relations between the demand and supply of different things. Between tea and coffee, or beef and mutton there is a relation of a different kind. These things are in large measure what we call "substitutes" for one another. An increased supply, and a lower price of mutton, will probably induce us to consume less beef. This relation it is convenient to describe as Composite Supply. Beef and mutton make up a composite supply of meat; tea and coffee a composite supply of a certain type of beverage. For any group of things, between which the relation of Composite Supply exists, we can say, with complete generality, that an increased supply of one of them will tend to diminish the demand for the others. Parallel to the relation of Composite Supply is that of Composite Demand. There are frequently several alternative uses in which a commodity or service can be employed; and these alternative uses make up a composite demand for the thing in question. Thus railways, gasworks, private households and a great variety of industries contribute to a Composite Demand for coal. It is worth noting that there is frequently an association in practice between Joint Demand and Composite Supply on the one hand; and between Joint Supply and Composite Demand on the other. Wool and mutton, for instance, we have described as an instance of Joint Supply; but, in so far as the proportions of wool and mutton can be varied, we can regard these things as constituting a Composite Demand for sheep. And this conception may help us to retain a clearer and more orderly picture of the problems we have discussed above. We can regard the fact that wool and mutton are produced together as their Joint Supply aspect, and the fact that these proportions can be varied as their Composite Demand aspect; and the question as to whether an increased demand for mutton will increase the supply of wool turns upon whether the former aspect is more important than the latter. Similarly labor and machinery, employed together for the same purpose, form an instance of Joint Demand; but in so far as they can be substituted for one another, they constitute a Composite Supply of alternative agents of production.

These four relations of Joint Demand, Joint Supply, Composite Demand and Composite Supply are well worth remembering and distinguishing from one another. They are of immense importance in every branch of economic affairs. There are hardly any economic problems upon which we are fitted to express an opinion, unless we have a lively sense of the far-reaching ramifications of cause and consequence, of the subtle and often unexpected interconnections between different industries and different markets. To gape at these complexities in a confused stupor is as foolish as it is to ignore them. But confusion and stupor are only too likely to represent our final state of mind, if we attempt to deal with these complications, one by one as they occur to us, in a piecemeal and haphazard fashion. We need a clear method, a systematic plan by which we may search them out, and fit them into place. The four relations which we have enumerated supply us with such a plan and method. For they represent something more than a series of pompous names for familiar notions. They constitute a classification of the various ways in which the demand and supply of one thing can affect the demand and supply of others; a classification which is exhaustive when we add the relation of derived demand, and an analogous relation on the supply side which we must now notice.

§8. Ultimate Real Costs. Just as the utility of "producers' goods" is derived from that of the "consumers' goods" which they help to make; so the cost of any commodity is derived from the cost of the things which help to make it. Moreover, just as we recognize that the utility of "consumers' goods" lies at the back of all demand, and constitutes the ultimate end of all production; so we cannot but feel, however obscurely, that behind the phenomena of money costs, there must lie certain ultimate costs, of which all money costs are but the measure. But when we try to explain what the nature of these real costs may be, we are plunged in difficulty. Wages, it may indeed seem at first sight, present no trouble. There is the effort and the fatigue, the unpleasantness of human labor, to represent real costs. But can we suppose that these things are measured with any approach to accuracy by the wages which are paid in actual fact? Is it true, even as a broad general rule, that the services which are most arduous and most disagreeable command the highest price? And wages are not the only ingredient of money costs. There are profits: to what real costs do profits correspond? More difficult still, to what does rent correspond? These plainly are not questions upon which he who runs may read. It will be necessary to devote the next four chapters to their elucidation.