Principles of Economics

Marshall, Alfred
Display paragraphs in this book containing:
First Pub. Date
London: Macmillan and Co., Ltd.
Pub. Date
8th edition
67 of 71




§ 1. When Ricardo was addressing a general audience, he drew largely upon his wide and intimate knowledge of the facts of life, using them "for illustration, verification, or the premises of argument." But in his Principles of Political Economy "the same questions are treated with a singular exclusion of all reference to the actual world around him*88." And he wrote to Malthus in May, 1820 (the same year in which Malthus published his Principles of Political Economy considered with a view to their practical application), "Our differences may in some respects, I think, be ascribed to your considering my book as more practical than I intended it to be. My object was to elucidate principles, and to do this I imagined strong cases, that I might show the operation of those principles." His book makes no pretence to be systematic. He was with difficulty induced to publish it; and if in writing it he had in view any readers at all, they were chiefly those statesmen and business men with whom he associated. So he purposely omitted many things which were necessary for the logical completeness of his argument, but which they would regard as obvious. And further, as he told Malthus in the following October, he was "but a poor master of language." His exposition is as confused as his thought is profound; he uses words in artificial senses which he does not explain, and to which he does not adhere; and he changes from one hypothesis to another without giving notice.


If then we seek to understand him rightly, we must interpret him generously, more generously than he himself interpreted Adam Smith. When his words are ambiguous, we must give them that interpretation which other passages in his writings indicate that he would have wished us to give them. If we do this with the desire to ascertain what he really meant, his doctrines, though very far from complete, are free from many of the errors that are commonly attributed to them.


He considers, for instance (Principles, Ch. I. § 1), that utility is "absolutely essential" to (normal) value though not its measure; while the value of things "of which there is a very limited quantity...varies with the wealth and inclinations of those who are desirous to possess them." And elsewhere (Ib. Ch. IV.) he insists on the way in which the market fluctuations of prices are determined by the amount available for sale on the one hand, and "the wants and wishes of mankind" on the other.


Again, in a profound, though very incomplete, discussion of the difference between "Value and Riches" he seems to be feeling his way towards the distinction between marginal and total utility. For by Riches he means total utility, and he seems to be always on the point of stating that value corresponds to the increment of riches which results from that part of the commodity which it is only just worth the while of purchasers to buy; and that when the supply runs short, whether temporarily in consequence of a passing accident, or permanently in consequence of an increase in cost of production, there is a rise in that marginal increment of riches which is measured by value, at the same time that there is a diminution in the aggregate riches, the total utility, derived from the commodity. Throughout the whole discussion he is trying to say, though (being ignorant of the terse language of the differential calculus) he did not get hold of the right words in which to say it neatly, that marginal utility is raised and total utility is lessened by any check to supply.


§ 2. But while not thinking that he had much to say that was of great importance on the subject of utility, he believed that the connection between cost of production and value was imperfectly understood; and that erroneous views on this subject were likely to lead the country astray in practical problems of taxation and finance; and so he addressed himself specially to this subject. But here also he made short cuts.


For, though he was aware that commodities fall into three classes according as they obey the law of diminishing, of constant, or of increasing return; yet he thought it best to ignore this distinction in a theory of value applicable to all kinds of commodities. A commodity chosen at random was just as likely to obey one as the other of the two laws of diminishing and of increasing return; and therefore he thought himself justified in assuming provisionally that they all obeyed the law of constant return. In this perhaps he was justified, but he made a mistake in not stating explicitly what he was doing.


He argued in the first Section of his first Chapter that "in the early stages of society" where there is scarcely any use of capital, and where any one man's labour has nearly the same price as any other man's, it is, broadly speaking, true that "the value of a commodity, or the quantity of a commodity for which it will exchange, depends on the relative quantity of labour which is necessary for its production." That is, if two things are made by twelve and four men's labour for a year, all the men being of the same grade, the normal value of the former will be three times that of the latter. For if ten per cent. has to be added for profits on the capital invested in the one case, ten per cent. will need to be added in the other also. [If w be a year's wages of a worker of this class, the costs of production will be 4w.110/100, and 12w.110/100: and the ratio of these is 4 : 12, or 1 : 3.]


But he went on to show that these assumptions cannot be properly made in later stages of civilization, and that the relation of value to cost of production is more complex than that with which he started; and his next step was to introduce in Section II. the consideration that "labour of different qualities is differently rewarded." If the wages of a jeweller are twice as great as those of a working labourer, an hour's work of the one must count for two hours' work of the other. Should there be a change in their relative wages, there will of course be a corresponding change in the relative values of things made by them. But instead of analysing, as economists of this generation do, the causes which make (say) jewellers' wages change from one generation to another relatively to those of ordinary labourers, he contented himself with stating that such variations cannot be great.


Next in Section III. he urged that in reckoning the cost of production of a commodity, account must be taken not only of the labour applied immediately to it, but also of that which is bestowed on the implements, tools and buildings with which such labour is assisted; and here the element of time, which he had carefully kept in the background at starting, was necessarily introduced.


Accordingly in Section IV. he discusses more fully the different influences exerted on the value of "a set of commodities" [he uses this simple method sometimes to evade the difficulties of the distinctions between prime cost and total cost]: and especially he takes account of the different effects of the application of circulating capital which is consumed in a single use, and of fixed capital; and again of the time for which labour must be invested in making machinery to make commodities. If that be long, they will have a greater cost of production and be "more valuable to compensate for the greater length of time, which must elapse before they can be brought to market."


And lastly in Section V. he sums up the influence which different lengths of investment, whether direct or indirect, will have upon relative values; arguing correctly that if wages all rise and fall together the change will have no permanent effect on the relative values of different commodities. But he argues if the rate of profits falls it will lower the relative values of those commodities the production of which requires capital to be invested a long while before they can be brought to market. For if in one case the average investment is for a year and requires ten per cent. to be added to the wages bill for profits; and in another is for two years and requires twenty per cent. to be added; then a fall of profits by one-fifth will reduce the addition in the latter case from 20 to 16, and in the former only from 10 to 8. [If their direct labour cost is equal the ratio of their values before the change will be 120/110 or 1.091; and after the change 116/108 or 1.074; a fall of nearly two per cent.] His argument is avowedly only provisional; in later chapters he takes account of other causes of differences in profits in different industries, besides the period of investment. But it seems difficult to imagine how he could more strongly have emphasized the fact that Time or Waiting as well as Labour is an element of cost of production than by occupying his first chapter with this discussion. Unfortunately however he delighted in short phrases, and he thought that his readers would always supply for themselves the explanations of which he had given them a hint.


Once indeed, in a note at the end of the sixth Section of his first Chapter, he says:—"Mr Malthus appears to think that it is a part of my doctrine that the cost and value of a thing should be the same; it is, if he means by cost, 'cost of production' including profits. In the above passage, this is what he does not mean, and therefore he has not clearly understood me." And yet Rodbertus and Karl Marx claim Ricardo's authority for the statement that the natural value of things consists solely of the labour spent on them; and even those German economists who most strenuously combat the conclusions of these writers, are often found to admit that they have interpreted Ricardo rightly, and that their conclusions follow logically from his.


This and other facts of a similar kind show that Ricardo's reticence was an error of judgment. It would have been better if he had occasionally repeated the statement that the values of two commodities are to be regarded as in the long run proportionate to the amount of labour required for making them, only on the condition that other things are equal: i.e., that the labour employed in the two cases is equally skilled, and therefore equally highly paid; that it is assisted by proportionate amounts of capital, account being taken of the period of its investment; and that the rates of profits are equal. He does not state clearly, and in some cases he perhaps did not fully and clearly perceive how, in the problem of normal value, the various elements govern one another mutually, and not successively in a long chain of causation. And he was more guilty than almost anyone else of the bad habit of endeavouring to express great economic doctrines in short sentences*89.


§ 3. There are few writers of modern times who have approached as near to the brilliant originality of Ricardo as Jevons has done. But he appears to have judged both Ricardo and Mill harshly, and to have attributed to them doctrines narrower and less scientific than those which they really held. And his desire to emphasize an aspect of value to which they had given insufficient prominence, was probably in some measure accountable for his saying, "Repeated reflection and inquiry have led me to the somewhat novel opinion that value depends entirely upon utility" (Theory, p. 1). This statement seems to be no less one-sided and fragmentary, and much more misleading, than that into which Ricardo often glided with careless brevity, as to the dependence of value on cost of production; but which he never regarded as more than a part of a larger doctrine, the rest of which he had tried to explain.


Jevons continues:—"We have only to trace out carefully the natural laws of variation of utility as depending upon the quantity of commodity in our possession, in order to arrive at a satisfactory theory of exchange, of which the ordinary laws of supply and demand are a necessary consequence...Labour is found often to determine value, but only in an indirect manner by varying the degree of utility of the commodity through an increase or limitation of the supply." As we shall presently see, the latter of these two statements had been made before in almost the same form, loose and inaccurate as it is, by Ricardo and Mill; but they would not have accepted the former statement. For while they regarded the natural laws of variation of utility as too obvious to require detailed explanation, and while they admitted that cost of production could have no effect upon exchange value if it could have none upon the amount which producers brought forward for sale; their doctrines imply that what is true of supply, is true mutatis mutandis of demand, and that the utility of a commodity could have no effect upon its exchange value if it could have none on the amount which purchasers took off the market. Let us then turn to examine the chain of causation in which Jevons' central position is formulated in his Second Edition, and compare it with the position taken up by Ricardo and Mill. He says (p. 179):—

"Cost of production determines supply.
Supply determines final degree of utility.
Final degree of utility determines value."

Now if this series of causations really existed, there could be no great harm in omitting the intermediate stages and saying that cost of production determines value. For if A is the cause of B, which is the cause of C, which is the cause of D; then A is the cause of D. But in fact there is no such series.


A preliminary objection might be taken to the ambiguity of the terms "cost of production" and "supply"; which Jevons ought to have avoided, by the aid of that technical apparatus of semi-mathematical phrases, which was at his disposal, but not at Ricardo's. A graver objection lies against his third statement. For the price which the various purchasers in a market will pay for a thing, is determined not solely by the final degrees of its utility to them, but by these in conjunction with the amounts of purchasing power severally at their disposal. The exchange value of a thing is the same all over a market; but the final degrees of utility to which it corresponds are not equal at any two parts. Jevons supposed himself to be getting nearer the foundations of exchange value when in his account of the causes which determine it, he substituted the phrase "final degree of utility," for "the price which consumers are only just willing to pay,"—the phrase which in the present treatise is condensed into "marginal demand price." When for instance describing (Second Edition, p. 105) the settlement of exchange between "one trading body possessing only corn, and another possessing only beef," he makes his diagram represent "a person" as gaining a "utility" measured along one line and losing a "utility" measured along another. But that is not what really happens; a trading body is not "a person," it gives up things which represent equal purchasing power to all of its members, but very different utilities. It is true that Jevons was himself aware of this; and that his account can be made consistent with the facts of life by a series of interpretations, which in effect substitute "demand-price" and "supply-price" for "utility" and "disutility": but, when so amended, they lose much of their aggressive force against the older doctrines, and if both are to be held severely to a strictly literal interpretation, then the older method of speaking, though not perfectly accurate, appears to be nearer the truth than that which Jevons and some of his followers have endeavoured to substitute for it.


But the greatest objection of all to his formal statement of his central doctrine is that it does not represent supply price, demand price and amount produced as mutually determining one another (subject to certain other conditions), but as determined one by another in a series. It is as though when three balls A, B, and C rest against one another in a bowl, instead of saying that the position of the three mutually determines one another under the action of gravity, he had said that A determines B, and B determines C. Someone else however with equal justice might say that C determines B and B determines A. And in reply to Jevons a catena rather less untrue than his can be made by inverting his order and saying:—

Utility determines the amount that has to be supplied,
The amount that has to be supplied determines cost of production,
Cost of production determines value, because it determines the supply price which is required to make the producers keep to their work.


Let us then turn to Ricardo's doctrine which, though unsystematic and open to many objections, seems to be more philosophic in principle and closer to the actual facts of life. He says, in the letter to Malthus already quoted:—"M. Say has not a correct notion of what is meant by value when he contends that a commodity is valuable in proportion to its utility. This would be true if buyers only regulated the value of commodities; then indeed we might expect that all men would be willing to give a price for things in proportion to the estimation in which they held them; but the fact appears to me to be that the buyers have the least in the world to do in regulating price; it is all done by the competition of the sellers, and, however really willing the buyers might be to give more for iron than for gold, they could not, because the supply would be regulated by cost of production.... You say demand and supply regulates value [sic]; this I think is saying nothing, and for the reason I have given in the beginning of this letter: it is supply which regulates value, and supply is itself controlled by comparative cost of production. Cost of production, in money, means the value of labour as well as of profits." (See pp. 173-6 of Dr Bonar's excellent edition of these letters.) And again in his next letter, "I do not dispute either the influence of demand on the price of corn or on the price of all other things: but supply follows close at its heels and soon takes the power of regulating price in his [sic] own hands, and in regulating it he is determined by cost of production."


These letters were not indeed published when Jevons wrote, but there are very similar statements in Ricardo's Principles. Mill also, when discussing the value of money (Book III. ch. IX. § 3), speaks of "the law of demand and supply which is acknowledged to be applicable to all commodities, and which in the case of money as of most other things, is controlled but not set aside by the law of cost of production, since cost of production would have no effect on value if it could have none on supply." And again, when summing up his theory of value (Book III. ch. XVI. § 1), he says:—"From this it appears that demand and supply govern the fluctuations of prices in all cases, and the permanent values of all things of which the supply is determined by any agency other than that of free competition: but that, under the régime of free competition, things are, on the average, exchanged for each other at such values and sold for such prices as afford equal expectation of advantage to all classes of producers; which can only be when things exchange for one another in the ratio of their cost of production." And, on the next page, speaking of commodities which have a joint cost of production, he says, "since cost of production here fails us we must resort to a law of value anterior to cost of production and more fundamental, the law of demand and supply."


Jevons (p. 215), referring to this last passage, speaks of "the fallacy involved in Mill's idea that he is reverting to an anterior law of value, the law of supply and demand, the fact being that in introducing the cost of production principle, he has never quitted the law of supply and demand at all. The cost of production is only one circumstance which governs supply and thus indirectly influences values."


This criticism seems to contain an important truth; though the wording of the last part is open to objection. If it had been made in Mill's time he would probably have accepted it; and would have withdrawn the word "anterior" as not expressing his real meaning. The "cost of production principle" and the "final utility" principle are undoubtedly component parts of the one all-ruling law of supply and demand; each may be compared to one blade of a pair of scissors. When one blade is held still, and the cutting is effected by moving the other, we may say with careless brevity that the cutting is done by the second; but the statement is not one to be made formally, and defended deliberately*90.


Perhaps Jevons' antagonism to Ricardo and Mill would have been less if he had not himself fallen into the habit of speaking of relations which really exist only between demand price and value as though they held between utility and value; and if he had emphasized as Cournot had done, and as the use of mathematical forms might have been expected to lead him to do, that fundamental symmetry of the general relations in which demand and supply stand to value, which coexists with striking differences in the details of those relations. We must not indeed forget that, at the time at which he wrote, the demand side of the theory of value had been much neglected; and that he did excellent service by calling attention to it and developing it. There are few thinkers whose claims on our gratitude are as high and as various as those of Jevons: but that must not lead us to accept hastily his criticisms on his great predecessors*91.


It seemed right to select Jevons' attack for reply, because, in England at all events, it has attracted more attention than any other. But somewhat similar attacks on Ricardo's theory of value had been made by many other writers. Among them may specially be mentioned Mr Macleod, whose writings before 1870 anticipated much both of the form and substance of recent criticisms on the classical doctrines of value in relation to cost, by Profs. Walras and Carl Menger, who were contemporary with Jevons, and Profs. v. Böhm-Bawerk and Wieser, who were later.


The carelessness of Ricardo with regard to the element of Time has been imitated by his critics, and has thus been a source of twofold misunderstanding. For they attempt to disprove doctrines as to the ultimate tendencies, the causes of causes, the causœ causantes, of the relations between cost of production and value, by means of arguments based on the causes of temporary changes, and short-period fluctuations of value. Doubtless nearly everything they say when expressing their own opinions is true in the sense in which they mean it; some of it is new and much of it is improved in form. But they do not appear to make any progress towards establishing their claim to have discovered a new doctrine of value which is in sharp contrast to the old; or which calls for any considerable demolition, as distinguished from development and extension, of the old doctrine.


Ricardo's first chapter has been discussed here with sole reference to the causes which govern the relative exchange values of different things; because its chief influence on subsequent thought has been in this direction. But it was originally associated with a controversy as to the extent to which the price of labour affords a good standard for measuring the general purchasing power of money. In this connection its interest is mainly historical: but reference may be made to an illuminating article on it by Prof. Hollander in the Quarterly Journal of Economics, 1904.

Notes for this chapter

Compare the concluding remarks of Book V.: and Appendix B, 5.
See an admirable article on Ricardo's Use of Facts in the first volume of the Harvard Quarterly Journal of Economics, by the late Professor Dunbar.
Prof. Ashley in a suggestive criticism of this Note, as part of an attempted "Rehabilitation of Ricardo" (Economic Journal, Vol. I.), insists that it has been commonly believed that Ricardo did in fact habitually think of mere quantities of labour as constituting cost of production, and governing value, subject only to "slight modifications"; and that this interpretation of him is the most consistent with his writings as a whole. It is not disputed that this interpretation has been accepted by many able writers: otherwise there would have been little need for rehabilitating, i.e. clothing more fully his somewhat too naked doctrines. But the question whether Ricardo is to be supposed to have meant nothing by the first chapter of his book, merely because he did not constantly repeat the interpretation clauses contained in it, is one which each reader must decide for himself according to his temperament: it does not lend itself to be solved by argument. It is here claimed not that his doctrines contained a complete theory of value: but only that they were in the main true as far as they went. Rodbertus and Marx interpreted Ricardo's doctrine, to mean that interest does not enter into that cost of production which governs (or rather takes part in governing) value: and as regards this Prof. Ashley appears to concede all that is claimed here when (p. 480) he takes it as beyond question that Ricardo "regarded the payment of interest, that is, of something more than the mere replacement of capital, as a matter of course."
See V. III. 7.
See an article on Jevons' Theory by the present writer in the Academy for April 1, 1872. The edition of his Theory brought out by his son in 1911, contains an Appendix on his account of interest, with special reference to that article (see also above VI. I. 8). He contends that his father's theory is "true as far as it goes" though he "followed the unfortunate practice of the Ricardian school by abstracting for treatment certain ideas, and assuming that his readers are familiar with their relations and taking his point of view." The son may be accepted as the true interpreter of the father: and the debts of economics to the father are no doubt so great as to be comparable with its transcendent obligations to Ricardo. But Jevons' Theory had a combative side, as well as a constructive. In great part it was an attack on what he called in his Preface, "that able but wrong-headed man, David Ricardo" who "shunted the car of Economic science on to a wrong line." His criticisms on Ricardo achieved some apparently unfair dialectical triumphs, by assuming that Ricardo thought of value as governed by cost of production without reference to demand. This misconception of Ricardo was doing great harm in 1872: and it seemed necessary to show that Jevons' Theory of Interest, if interpreted as he interpreted Ricardo, is untenable.

End of Notes

67 of 71

Return to top