Transcriber’s Note:
The cover image was created by the transcriber and is placed in the public domain.
PAGE | |
---|---|
Preface | vii |
Chapter I. | |
Rules for the Definition and Application of Terms in Political Economy | 1 |
Chapter II. | |
On the Definition of Wealth by the French Economists | 8 |
Chapter III. | |
On the Definition and Application of Terms by Adam Smith | 10 |
Chapter IV. | |
Application of the term Utility by M. Say | 19 |
Chapter V. | |
On the Definition and Application of Terms by Mr. Ricardo | 23 |
Chapter VI. | |
On the Definition and Application of Terms by Mr. Mill, in his “Elements of Political Economy.” | 37 |
viChapter VII. | |
On the Definition and Application of Terms, by Mr. Macculloch, in his “Principles of Political Economy.” | 69 |
Chapter VIII. | |
On the Definition and Use of Terms by the Author of “A Critical Dissertation on the Nature, Measure, and Causes of Value.” | 125 |
Chapter IX. | |
Summary of the Reasons for Adopting the subjoined Definition of the Measure of Value | 203 |
Chapter X. | |
Definitions in Political Economy | 234 |
Chapter XI. | |
Remarks on the Definitions | 249 |
The differences of opinion among political economists have of late been a frequent subject of complaint; and it must be allowed, that one of the principal causes of them may be traced to the different meanings in which the same terms have been used by different writers.
The object of the present publication is, to draw attention to an obstacle in the study of political economy, which has now increased to no inconsiderable magnitude. But this could not be done merely by laying down rules for the definition and application of terms, and defining conformably viiito them. It was necessary to show the difficulties which had resulted from an inattention to this subject, in some of the most popular works on political economy; and this has naturally led to the discussion of certain important principles and questions of classification, which it would be most desirable to settle previously, as the only foundation for a correct definition and application of terms.
These are the reasons for the arrangement and mode of treating the subject which has been adopted.
In a mathematical definition, although the words in which it is expressed may vary, the meaning which it is intended to convey is always the same. Whether a straight line be defined to be a line which lies evenly between its extreme points, or the shortest line which can be drawn between two points, there never can be a difference of opinion as to the lines which are comprehended, and those which are not comprehended, in the definition.
The case is not the same with the definitions in the less strict sciences. The classifications in natural history, notwithstanding all the pains which have been taken with them, 2are still such, that it is sometimes difficult to say to which of two adjoining classes the individuals on the confines of each ought to belong. It is still more difficult, in the sciences of morals and politics, to use terms which may not be understood differently by different persons, according to their different habits and opinions. The terms virtue, morality, equity, charity, are in every-day use; yet it is by no means universally agreed what are the particular acts which ought to be classed under these different heads.
The terms liberty, civil liberty, political liberty, constitutional government, &c. &c., are frequently understood in a different sense by different persons.
It has sometimes been said of political economy, that it approaches to the strict science of mathematics. But I fear it must be acknowledged, particularly since the great deviations which have lately taken place from the definitions and doctrines of Adam Smith, that it approaches more nearly to the sciences of morals and politics.
It does not seem yet to be agreed what 3ought to be considered as the best definition of wealth, of capital, of productive labour, or of value;—what is meant by real wages;—what is meant by labour;—what is meant by profits;—in what sense the term ‘demand’ is to be understood,[1] &c. &c.
As a remedy for such differences, it has been suggested, that a new and more perfect nomenclature should be introduced. But though the inconveniences of a new nomenclature are much more than counterbalanced by its obvious utility in such sciences as chemistry, botany, and some others, where a great variety of objects, not in general use, must be arranged and described so as best to enable us to remember their characteristic distinctions; yet in such sciences as morals, politics, and political economy, where the terms are comparatively few, and of constant application in the daily concerns of life, it is impossible to suppose 4that an entirely new nomenclature would be submitted to; and if it were, it would not render the same service to these sciences, in promoting their advancement, as the nomenclatures of Linnæus, Lavoisier, and Cuvier, to the sciences to which they were respectively applied.
Under these circumstances, it may be desirable to consider what seem to be the most obvious and natural rules for our guidance in defining and applying the terms used in the science of political economy. The object to be kept in view should evidently be such a definition and application of these terms, as will enable us most clearly and conveniently to explain the nature and causes of the wealth of nations; and the rules chiefly to be attended to may, perhaps, be nearly included in the four following:—
First. When we employ terms which are of daily occurrence in the common conversation of educated persons, we should define and apply them, so as to agree with the sense in which they are understood in this ordinary use of them. This is the best and 5more desirable authority for the meaning of words.
Secondly. When the sanction of this authority is not attainable, on account of further distinctions being required, the next best authority is that of some of the most celebrated writers in the science, particularly if any one of them has, by common consent, been considered as the principal founder of it. In this case, whether the term be a new one, born with the science, or an old one used in a new sense, it will not be strange to the generality of readers, nor liable to be often misunderstood.
But it may be observed, that we shall not be able to improve the science if we are thus to be bound down by past authority. This is unquestionably true; and I should be by no means inclined to propose to political economists “jurare in verba magistri,” whenever it can be clearly made out that a change would be beneficial, and decidedly contribute to the advancement of the science. But it must be allowed, that in the less strict sciences there are few definitions to which some plausible, nay, even real, objections are not to be 6made; and, if we determine to have a new one in every case where the old one is not quite complete, the chances are, that we shall subject the science to all the very serious disadvantages of a frequent change of terms, without finally accomplishing our object.
It is acknowledged, however, that a change may sometimes be necessary; and when it is, the natural rules to be attended to seem to be,
Thirdly. That the alteration proposed should not only remove the immediate objections which may have been made to the terms as before applied, but should be shown to be free from other equal or greater objections, and on the whole be obviously more useful in facilitating the explanation and improvement of the science. A change which is always itself an evil, can alone be warranted by superior utility taken in the most enlarged sense.
Fourthly. That any new definitions adopted should be consistent with those which are allowed to remain, and that the same terms should always be applied in the same sense, except where inveterate custom has established 7different meanings of the same word; in which case the sense in which the word is used, if not marked by the context, which it generally is, should be particularly specified.
I cannot help thinking that these rules for the definitions in political economy must be allowed to be obviously natural and proper, and that if changes are made without attention to them, we must necessarily run a great risk of impeding, instead of promoting, the progress of the science.
Yet, although these rules appear to be so obvious and natural, as to make one think it almost impossible that they should escape attention, it must be acknowledged that they have been too often overlooked by political economists; and it may tend to illustrate their use and importance; and possibly excite a little more attention to them in future; to notice some of the most striking deviations from them in the works of writers of the highest reputation.
It will not be worth while to advert to the misnomers of the mercantile system; but the system of the French Economists was a scientific one, and aimed at precision. Yet it must be acknowledged that their definition of wealth violated the first and most obvious rule which ought to guide men of science, as well as others, in the use of terms. Wealth and riches are words in the commonest use; and though all persons might not be able at once to describe with accuracy what they mean when they speak of the wealth of a country, yet all, we believe, who intend to use the term in its ordinary sense, would agree in saying that they do not confine the term either to the gross raw produce, or the neat raw produce of such country. And it is quite certain that two countries, with both the same gross raw 9produce, and the same neat raw produce, might differ most essentially from each other in a great number of the most universally acknowledged characteristics of wealth, such as good houses, good furniture, good clothes, good carriages, which, in the one case, might be possessed only by a few great landlords, and a small number of manufacturers and merchants; and in the other case, by an equal, or greater proportion of landlords, and a much greater number of manufacturers and merchants. This difference might take place without any difference in the amount of the raw produce, the neat produce, or the population, merely by the conversion of idle retainers and menial servants into active artisans and traders. The result, therefore, of comparing together the wealth of different countries, according to the sense of that term adopted by the Economists, and according to the sense in which it is generally understood in society, would be totally different. And this circumstance detracts in a very great degree from the practical utility of the works of the Economists.
In adverting to the terms and definitions of Adam Smith, in his “Wealth of Nations,” I think it will be found that he has less frequently and less strikingly deviated from the rules above laid down, and that he has more constantly and uniformly kept in view the paramount object of explaining in the most intelligible manner the causes of the wealth of nations, according to the ordinary acceptation of the expression, than any of the subsequent writers in the science, who have essentially differed from him. His faults in this respect are not so much that he has often fallen into the common error, of using terms in a different sense from that in which they are ordinarily applied in society, but 11that he is sometimes deficient in the precision of his definitions; and does not always, when adopted, adhere to them with sufficient strictness.
His definition of wealth, for instance, is not sufficiently accurate; nor does he adhere to it with sufficient uniformity: yet it cannot be doubted that he means by the term generally the material products which are necessary, useful, and agreeable to man, and are not furnished by nature in unlimited abundance; and I own I feel quite convinced that it is in this sense in which it is most generally understood in society, and in which it may be most usefully applied, in explaining the causes of the wealth of nations.
In adopting the labour which a commodity will command as the measure of its value, he has not, as it appears to me, given the most conclusive reasons for it, nor has he in all cases made it quite clear whether he means the labour which a commodity will command, or the labour worked up in it. He has more frequently failed in not adhering 12practically to the measure he had proposed, and in substituting as an equivalent the quantity of corn a commodity will command, which, as a measure of value, has properties essentially distinct from labour. Yet, with all this, it must be acknowledged that he has generally used the terms labour and value in the sense in which they are ordinarily understood in society, and has, with few exceptions, applied labour as the measure of value in the way in which it may be made most extensively useful in the explanation of the science.
It has been sometimes objected to Adam Smith, that he has applied the term productive in a new and not very appropriate sense. But if we examine the manner in which this term is applied in ordinary conversation and writing, it must be allowed that, whatever meaning may be thought to attach to it, from its derivation, it is practically used as implying causation in regard to almost any effect whatever. Thus we say that such and such things are productive of the best effects, 13others of the very worst effects, and others are unproductive of, or do not produce, any perceptible effects; meaning by these expressions, that some things cause the best effects, others the worst effects, others, again, cause no perceptible effects; and these effects may, of course, apply according to the context, and the subject under discussion, to the health of the body, the improvement of the mind, the structure of society, or the wealth of a nation.
Now, Adam Smith was inquiring into the nature and causes of the wealth of nations; and having confined the term wealth to material objects, and described human labour as the main source of wealth, he clearly saw the necessity of making some distinction between those different kinds of labour which, without reference to their utility, he could not but observe had the most essentially distinct effects, in directly causing that wealth, the nature of which he was investigating. He called one of these kinds of labour productive, or productive of wealth, and the other unproductive, or not productive of wealth; and knowing that it 14would occasion interminable confusion, and break down all the barriers between production and consumption, to attempt to estimate the circumstances which might indirectly contribute to the production of wealth, he described productive labour in such a way, as to leave no doubt that he meant the labour which was so directly productive of wealth, as to be estimated in the quantity or value of the material object produced.
In his application of the terms productive and unproductive, therefore, he does not seem to have violated the usage of common conversation and writing; and it appears to me, that, if we fully and impartially consider the consequences of making no distinction between different kinds of labour, we must feel the conviction that the terms which he has adopted are pre-eminently useful for the purpose to which they are applied—that is, to enable him to explain, intelligibly and satisfactorily, the causes of the wealth of different nations, according to the ordinary meaning which men attach to the term wealth, whatever may be their theories on the subject.
15Where Adam Smith has most failed in the use of his terms, is in the application of the word real. The real value of a commodity he distinctly and repeatedly states to be the quantity of labour which it will command, in contradistinction to its nominal value, that is, its value in money, or any other specific commodity named. But while he is thus using the word real, in this sense, he applies it to wages in a totally different sense, and says, that the real wages of labour are the necessaries and conveniencies of life which the money received by the labourer will enable him to command. Now, it must be allowed that both these modes of applying the word real, cannot be correct, or consistent with each other. If the value of labour varies continually with the varying quantity of the necessaries and conveniencies of life which it will command, it is completely inconsistent to bring it forward as a measure of real value. And if it can, with propriety, be brought forward as a measure of the real value of commodities, it follows necessarily that the average value of a given quantity of labour, of a given description, can 16never be considered as in the slightest degree affected by the varying quantity of commodities for which it will exchange. Of this Adam Smith seemed to be fully aware in the fifth chapter of his first book, where he says distinctly, that when more or less goods are given in exchange for labour, it is the goods that vary, not the labour.
It is evident, therefore, that to get right, we must cease to use the term real, in one or other of the meanings in which it has been applied by Adam Smith.
If the term had never been applied in political economy in a different sense from that in which it was first used by Adam Smith, there could be no doubt that it might be advantageously continued, and the expression real value might answer its purpose very well, and save any question respecting the substitution of some other term, such as intrinsic, positive, absolute, or natural. But as the term real has been very generally applied, by most writers, to wages, implying the real quantity of the means of subsistence and comfort which the labourer is enabled to 17command, in contradistinction to his nominal or money wages, the matter cannot be so easily settled, and we must come to some determination as to which of the two meanings it would be most advisable to reject.
Adhering to the rules which have been laid down, it will probably be acknowledged that the term real, when applied to the means of obtaining something in exchange, seems more naturally to imply the power of commanding the necessaries, conveniencies, and luxuries of life, than the power of commanding labour. A certain quantity of wealth is something more real, if the word real be used in its most ordinary sense, than a certain quantity of labour; and if, on this account, we continue to apply the term real to wages, we must express by positive, absolute, intrinsic, or natural, what Adam Smith has expressed by the word real, as applied to value: or it would be still better if political economists would agree in assigning a distinct meaning to the term value, as contradistinguished from price, whenever the value of a commodity is mentioned without mentioning any specific article in which 18it is proposed to estimate it, in the same manner as the price of a commodity is universally understood to mean price in money, whenever the term is used without referring specifically to some other article.
If, however, it should be found that the term real, in the sense in which it is first and most frequently applied by Adam Smith, has by usage got such fast hold of this meaning, that it cannot easily be displaced; and, further, if it be thought that an adjunct of this kind to the term value will sometimes be wanted in explanations, and that to express what Adam Smith means, the term real is preferable to either of the terms intrinsic, positive, absolute, or natural, there would be little objection to letting it retain its first meaning, provided we took care not to use it in application to the wages of labour, as implying the necessaries, conveniencies, and amusements of life. Instead of real wages, we must then say corn wages, commodity wages, wages in the means of subsistence, or something of the kind. But the other change is obviously more simple, and therefore in my opinion preferable.
It would lead me too far and into too many repetitions, if I were to go through the principal definitions of the continental political economists, and examine the manner in which they have used their terms in reference to the obvious rules above laid down; but I cannot resist noticing one very signal deviation from them in the justly distinguished work of M. Say. It relates to the term utility.
It must be allowed by those who are acquainted with M. Say’s work, first, that he has used the term utility in a sense totally different from that in which it is used in common conversation, and in the language of those who are considered as the best authorities in political economy. Proceeding upon the principle, that nothing can be valuable which is not useful to some person or other, he has strangely identified utility and value, and 20made the utility of a commodity proportionate to its value, although the custom is universal of distinguishing between that which is useful and that which is merely high-priced, of that which is calculated to satisfy the acknowledged and general wants of mankind, and that which may be only calculated to satisfy the capricious tastes of a few. He has thus violated the first and most obvious rule for the use of terms.
Secondly, he has gone directly against the usage of the best writers in political economy, and particularly against the authority of Adam Smith, whom he himself considers as the main founder of the science. Adam Smith has declared his opinion in the most decided manner on this subject, by contrasting value in use, and value in exchange, and illustrating the distinction between them by adducing the marked instances of a diamond and water. M. Say, therefore, in the manner in which he has applied the term utility, has violated the second obvious rule for the use of terms, as well as the first.
Thirdly, the objections to the old terms in 21use, wealth and value, if there were any, do not by any means seem to have been such as to warrant the introduction of a new term. The object of M. Say seems to have been to show, that production does not mean production of new matter in the universe, but I cannot believe that even the Economists had this idea; and it is quite certain that Adam Smith’s definition of production completely excludes it. “There is one sort of labour,” he says, “which adds to the value of the subject on which it is bestowed * * * and as it produces a value may be called productive.”[2] There is, certainly, no question here about the creation of new matter. And as M. Say observes, that when things are in their ordinary and natural state their value is the measure of their utility, while he had before affirmed that riches were in proportion to value,[3] it is difficult to conceive what beneficial purpose he could have in view in introducing the term utility thus made synonymous with value or riches.
22Fourthly, as the terms useful and utility are in such very common use, when applied in their accustomed sense, and cannot easily be supplied by others, it is extremely difficult to confine their application to the new sense proposed by M. Say. It is scarcely possible not to use them sometimes, as M. Say himself has done, according to their ordinary acceptation; but this necessarily introduces uncertainty and obscurity into the language of political economy.
M. Say had before made little or no distinction between riches and value, two terms which Mr. Ricardo justly considers as essentially different. He then introduces another term, utility, which, as he applies it, can hardly be distinguished from either of the others. The new term, therefore, could not have been called for; and it must be allowed that the use of it in the sense proposed, violates all the most obvious rules for the introduction of a new term into any science.
Although it must be allowed that the criterion of value which Mr. Ricardo has endeavoured to establish is an incomplete one, yet I cannot but think that he has conferred an important benefit on the science of political economy, by drawing a marked line of distinction between riches and value. A difference had perhaps been felt by most writers, but none before him had so strongly marked it, and attached so much importance to it. He agrees entirely with Adam Smith in the following definition of riches: “Every man is rich or poor according to the degree in which he can afford to enjoy the necessaries, conveniencies, and amusements of human life.”[4] And adds an observation in which I think he is quite right. “Value, then, essentially differs from riches; for value depends not on 24abundance, but on the difficulty or facility of production.”[5] He subsequently says, “although Adam Smith has given the correct description of riches which I have more than once noticed, he afterwards explains them differently, and says that a man must be rich or poor, according to the quantity of labour which he can afford to purchase. Now this description differs essentially from the other, and is certainly incorrect; for suppose the mines were to become more productive, so that gold and silver fell in value, from the greater facility of production; or that velvets were to be manufactured by so much less labour than before, that they fell to half their former value; the riches of all those who purchased these commodities would be increased; one man might increase the quantity of his plate, another might buy double the quantity of velvet; but with the possession of this additional plate and velvet, they could employ no more labour than before, because, as the exchangeable value of velvet and of plate would be lowered, they must part with proportionably 25more of these species of riches to purchase a day’s labour. Riches then cannot be estimated by the quantity of labour which they will purchase.”[6]
In these remarks I entirely agree with Mr. Ricardo. If riches consist of the necessaries, conveniencies, and luxuries of life, and the same quantity of labour will at different times, and under different circumstances, produce a very different quantity of the necessaries, conveniencies, and luxuries of life, then it is quite clear that the power of commanding labour, and the power of commanding the necessaries, conveniencies and luxuries of life are essentially distinct. One, in fact, is a description of value, and the other of wealth.
But though Mr. Ricardo has fully succeeded in showing that Adam Smith was incorrect in 26confounding wealth and value, even according to his own descriptions of them; yet he has nowhere succeeded in making out the propriety of that peculiar view of value which forms the most prominent feature of his work.
He has not confined himself to the assertion, that what he calls the value of a commodity is determined by the quantity of labour worked up in it; but he states, in substance, the following proposition, that commodities exchange with each other according to the quantity of manual labour worked up in them, including the labour worked up in the materials and tools consumed in their production, as well as that which is more immediately employed.[7]
Now this proposition is contradicted by universal experience. The slightest observation will serve to convince us, that after making all the required allowances for temporary deviations from the natural and ordinary course of things, the class of commodities subject to this law of exchange is most extremely confined, while the classes, not subject to it, embrace the great mass of commodities. 27Mr. Ricardo, indeed, himself admits of considerable exceptions to his rule; but if we examine the classes which come under his exceptions, that is, where the quantities of fixed capital employed are different and of different degrees of duration, and where the periods of the returns of the circulating capital employed are not the same, we shall find that they are so numerous, that the rule may be considered as the exception, and the exceptions the rule.
Yet, notwithstanding these admissions, he proceeds with his rule as if there had been few or no exceptions to it: he especially estimates the value of wages by the quantity of human labour worked up in them; and as it is quite true, that if we look only to this element of value, the value of wages has a tendency to rise in the progress of cultivation and improvement, he has attributed the fall of profits which usually takes place in rich countries to the rise in the value of wages; and, in fact, has founded his whole theory of profits, which has been considered as the crowning achievement in the science, upon the rise and fall in the 28value of wages. “It has been my endeavour,” he says, “to show throughout this work, that the rate of profits can never be increased but by a fall of wages.”[8] Again he observes, “Profits—it cannot be too often repeated—depend on wages; not on nominal but real wages; not on the number of pounds which may be annually paid to the labourer, but on the number of days’ work necessary to obtain these pounds.”[9]
Real wages, then, according to Mr. Ricardo’s definition, are determined by the quantity of labour worked up in the articles, which the labourer receives as a remuneration for his labour, whether food and clothing, or money.
Now the meaning here attached to the term real wages, on which Mr. Ricardo’s theory of profits is made to depend, is quite unusual, and decidedly contradicts all the most obvious rules which suggest themselves for the application of terms in any science.
In the first place, no one we believe ever heard, before the time of Mr. Ricardo, this 29term used in conversation in such a manner, that an increase of real wages would generally imply a diminution in the means of subsistence and comfort among the labouring classes and their families. Yet this would be the case, according to the sense in which Mr. Ricardo uses the term. Speaking of the different situations of the landlord and the labourer, in the progress of society, after describing the increasing wealth of the landlord, he says, “The fate of the labourer will be less happy; he will receive more money-wages it is true, (and the money of Mr. Ricardo is here used as measuring what he calls real wages;) but his corn wages will be reduced; and not only his command of corn, but his general condition will be deteriorated.” With a continued increase of real wages, “the condition of the labourer will generally decline, while the condition of the landlord will always be improved.”[10]
Secondly, No writer that I have met with, anterior to Mr. Ricardo, ever used the term wages, or real wages, as implying proportions. 30Profits, indeed, imply proportions; and the rate of profits had always justly been estimated by a per centage upon the value of the advances. But wages had uniformly been considered as rising or falling, not according to any proportion which they might bear to the whole produce obtained by a certain quantity of labour, but by the greater or smaller quantity of any particular produce received by the labourer, or by the greater or smaller power which such produce would convey, of commanding the necessaries and conveniencies of life. Adam Smith in particular had often used the term real wages, and always in the most natural sense possible, as implying the necessaries and conveniencies of life, which, according to the common language and feelings of men, might justly be considered as more real than money, or any other particular article in which the labourer might be paid. And the use of the term, in this sense, by Adam Smith, and most other political economists, necessarily made the new interpretation given to it more strange, and more unwarranted.
31Thirdly, There were no objections to the sense in which the term was before applied. It was both natural and useful. Nor was a new interpretation of it wanted for the purpose of explanation. All the effects of the wages of labour upon profits might have been clearly described, by stating, that profits are determined by the proportion of the whole produce which goes to pay the wages of labour, without calling this proportion, whether small or great in quantity, the real wages of labour, and without asserting that, as the value of wages rises, profits must proportionably fall. That profits are determined by the proportion of the whole produce which goes to pay the wages of labour, is a proposition, which, when correctly explained, will be found to be true, and to be confirmed by universal experience; while the proposition, that as the value of wages rises profits proportionably fall, cannot be true, except on the assumption that commodities, which have the same quantity of labour worked up in them, are always of the same value, an assumption which probably will not be found to be true in one case 32out of five hundred; and this, not from accidental or temporary causes, but from that natural and necessary state of things, which, in the progress of civilisation and improvement, tends continually to increase the quantity of fixed capital employed, and to render more various and unequal the times of the returns of the circulating capital. The introduction, therefore, of a new meaning of the term real wages, has not certainly the recommendation of being more useful.
Fourthly, the new sense in which the term real wages is used, is not maintained with consistency, or applied to old facts and opinions, with a proper allowance for the change that has been made. This is almost unavoidable, when old terms, which are quite familiar in one sense, are applied in another and different sense. It is particularly remarkable in Mr. Ricardo’s use of his artificial money, which is meant to be the measure of real wages. Thus, he says, “It may be proper to observe, that Adam Smith, and all the writers who have followed him, have, without one exception that I know of, maintained, that a rise in 33the price of labour would be uniformly followed by a rise in the price of all commodities. I hope I have succeeded in showing that there are no grounds for such an opinion, and that only those commodities would rise which had less fixed capital employed upon them than the medium in which price was estimated, and that all those which had more would positively fall in price when wages rose. On the contrary, if wages fell, those commodities only would fall which had a less proportion of fixed capital employed upon them than the medium in which price was estimated; all those which had more would positively rise in price.”[11]
Now all these effects of a rise or fall in the wages of labour, depend entirely upon wages being estimated in Mr. Ricardo’s imaginary money. Estimated in this way, and in this way alone, Mr. Ricardo’s statement would be correct. But neither Adam Smith, nor any of his followers, down to the time of Mr. Ricardo, ever thought of estimating the price of wages in this way. And estimating them in the way to which they were always accustomed, that 34is in money, as they found it, they are quite justified in what they have said. According to Adam Smith, at least, who estimates the value of commodities by the quantity of labour which they will command, if the money wages of labour universally rise, the value of money proportionably falls; and when the value of money falls, Mr. Ricardo himself says, that the prices of goods always rise.
The difference, therefore, between Mr. Ricardo and Adam Smith in this case, arises from Mr. Ricardo’s forgetting that he was using the term price of labour in a different sense from that in which it was used in the proposition objected to.
In the same manner, Mr. Ricardo’s very startling proposition respecting the effects of foreign trade, namely, that “no extension of foreign trade will immediately increase the amount of value in a country,” arises entirely from his using the term value in a different sense from that in which it had been used by his predecessors.
If the value of foreign commodities imported is to be estimated by the quantity of labour 35worked up in the commodities sent out to purchase them, then it is quite true that, whatever may be the returns, their value is unsusceptible of increase. But if the value of foreign commodities imported be estimated in the way in which they had ever been estimated before, that is, either in the money, in the labour, or in the mass of commodities which they would command when brought home, then there cannot be the least doubt that the immediate effect of a prosperous venture which gives great profits to the merchants concerned would be to increase the amount of value in the country. The value of the returns compared with the value of the outgoings would, in this particular trade, be greater than usual; and it is quite certain, that this increase of value in one quarter would not necessarily be counterbalanced by a decrease of value in any other. Practically, indeed, nothing is more usual than a simultaneous rise in the value of the great mass of commodities from a prosperous trade, whether this value be estimated in money or in labour.
It must be allowed, then, that Mr. Ricardo 36has been very far from cautious in the definition and application of his terms, in treating of some of the most fundamental principles of political economy; and I have very little doubt, as I have stated elsewhere, that this is one of the reasons why many of the readers of his work have found great difficulty in understanding it. When old and very familiar terms are used in a new sense, it is scarcely possible for the writer to be always consistent in their application, and extremely difficult to the reader always to be aware of the sense meant to be affixed to them.
Mr. Mill, in his Elements of Political Economy, professedly lays no claim to discovery. His main object seems to have been to give the substance of Mr. Ricardo’s work in a more concentrated form, and with a better arrangement; and this object he has accomplished. In the definition and application of his terms he nearly follows Mr. Ricardo; but it may be useful to notice a few cases, where he has either made the errors of Mr. Ricardo’s definitions more prominent, or has altered without improving them.
On his first approach to the question of value, he describes the causes which determine it much more inaccurately than Mr. Ricardo. He says, that “the value of commodities is determined by the quantity of capital 38and labour necessary to produce them.”[12] But this is obviously untrue and quite inconsistent with what he says afterwards respecting the regulator of value. It may be correct, and I fully believe it is, to estimate the value of labour by its quantity; but how can we estimate the value of different kinds of machinery, or different kinds of raw materials by their quantity? The quantity of raw material contained in a coarse and thick piece of calico, as compared with a very fine and thin piece of muslin, worked up by the same quantity of labour, may be four or five times greater, while the value of it, and the degree in which it affects the value of the commodity, may be actually less. We cannot, in short, measure the value of any product of labour by its bulk or quantity; and it must therefore be essentially incorrect to say, that the value of commodities is determined by the quantity of capital and labour necessary to produce them.
Proceeding afterwards to investigate more minutely what it is, which in the last resort 39determines the proportion in which commodities exchange for one another, he observes, that “as all capital consists in commodities, it follows, of course, that the first capital must have been the result of pure labour. The first commodities could not be made by any commodities existing before them. But if the first commodities, and of course the first capital, were the result of pure labour, the value of this capital, the quantity of other commodities for which it would exchange, must have been estimated by labour. This is an immediate consequence of the proposition which we have just established, that where labour was the sole instrument of production, exchangeable value was determined by the quantity of labour which the production of the commodity required. If this be established, it is a necessary consequence that the exchangeable value of all commodities is determined by quantity of labour.”[13]
Now this necessary consequence, which is here so confidently announced, does not appear to me to follow either from this statement, or 40from any thing which is said subsequently. Allowing that the first commodities, if completed and brought into use immediately, might be the result of pure labour, and that their value would therefore be determined by the quantity of that labour; yet it is quite impossible that such commodities should be employed as capital to assist in the production of other commodities, without the capitalist being deprived of the use of his advances for a certain period, and requiring a remuneration in the shape of profits.
In the early periods of society, on account of the comparative scarcity of these advances of labour, this remuneration would be high, and would affect the value of such commodities to a considerable degree, owing to the high rate of profits. In the more advanced stages of society, the value of capital and commodities is largely affected by profits, on account of the greatly increased quantity of fixed capital employed, and the greater length of time for which much of the circulating capital is advanced before the capitalist is repaid by the returns. In both cases, the rate at which 41commodities exchange with each other, is essentially affected by the varying amount of profits. It is impossible, therefore, to agree with Mr. Mill, when he says, “It appears by the clearest evidence, that quantity of labour in the last resort determines the proportion in which commodities exchange for one another.”[14]
On the same grounds Mr. Mill is quite incorrect, in calling capital hoarded labour. It may, perhaps, be called hoarded labour and profits; but certainly not hoarded labour alone, unless we determine to call profits labour. This Mr. Mill himself could not but see; and consequently, in his second edition, he has deserted Mr. Ricardo, and boldly ventured to say, that “profits are in reality the measure of quantity of labour.”[15] But as this very peculiar and most unwarranted abuse of terms belongs, I believe, originally to Mr. Macculloch, it may be best to defer the more particular examination of it, till I come to consider the definitions and application of terms adopted by Mr. Macculloch.
42In a work like that of Mr. Mill, which has so much the air of logical precision, one should have hoped and expected to find superior accuracy in the definitions, and great uniformity in the application of his terms, in whatever sense he might determine to use them; but in this the reader will be disappointed. It is difficult, for instance, to infer from the language of Mr. Mill, whether a commodity is to be considered as altering in its value in proportion to its costs of production, or in proportion to its power of commanding other commodities, and they are certainly not the same.
At the commencement of his seventh section, of chap. iii., entitled, “What regulates the Value of Money,” he says,
“By the value of money is here to be understood the proportion in which it exchanges for other commodities, or the quantity of it which exchanges for a certain quantity of other things.”
This is, to be sure, a very lax description of the value of money, very inferior in point of accuracy, even to what would be understood by the general power of purchasing. What 43are the things a certain quantity of which is here alluded to? and if these things change in the costs of their production, will money be proportionally affected?
But we have a different and better description of value in the next section. It is there said, that “gold and silver are, in reality, commodities. They are commodities for the attainment of which labour and capital must be employed. It is cost of production which determines the value of these as of other ordinary productions.”[16]
Now, if cost of production determines the value of money, it follows that, while the cost of producing a given quantity of money remains the same, its value remains the same. But it is obvious that the value of money may remain the same in this sense of the term, while, owing to the alterations which may be taking place in the costs of producing the commodities alluded to, the quantity of other things for which it will exchange may be essentially different. Which of the two, then, is the true criterion of the value of money? It is surely most desirable 44that the student in political economy should not be left in the dark on this subject; yet Mr. Mill gives him no assistance; and he is left to decide between two very different meanings as well as he can.
But, perhaps, the most culpable confusion of terms which Mr. Mill has fallen into, is in relation to demand and supply; and as he has a more original and appropriate claim to this error than any other English writer, and its belief leads to very important consequences, the notice of it is particularly called for.
In the first place, no person can have turned his attention, in the slightest degree, to the language of political economy, either in conversation or books, without being fully aware that the term demand is used in two very distinct senses; one implying the quantity of the commodity consumed, and the other the amount of sacrifice which the purchasers are willing to make in order to obtain a given portion of it. In the former sense, an increase of demand is but very uncertainly connected with an increase of value, or a further encouragement to production, as in general the 45greatest increase of such kind of demand takes place in consequence of a very abundant supply and a great fall in value. It is the other sense alone to which we refer, when we speak of the demand compared with the supply as determining the values and prices of commodities; and in this latter sense of the term demand, which, perhaps, is in the most frequent use, an increase of supply is so far from increasing demand that it diminishes it, while a diminution of demand increases it.
Secondly, it has been generally agreed, that when the quantity of a commodity brought to market is neither more nor less than sufficient to supply all those who are able and willing to give the natural and necessary price for it, the demand may then, and then only, be said to be equal to the supply; because, if the quantity wanted by those who are able and willing to give the natural price exceed the supply, the demand is said to be greater than the supply, and the price rises above the ordinary costs of production; and if the quantity wanted by those who are able and willing to give the natural price fall short of 46the supply, the demand is said to be less than the supply, and the price falls below the ordinary costs of production. This is the language of Adam Smith, and of almost all writers on political economy, as well as the language of common conversation when such subjects are discussed. Indeed it is difficult to conceive in what other sense it could, with any propriety, be said, that the supply was equal to the demand, because in any other sense than this, the supply of a commodity might be said to be equal to the demand, whether it were selling at double or the half of its cost.
Thirdly, it must be allowed, that according to the best authorities in books and conversation, what is meant by the glut of a particular commodity is such an abundant supply of it compared with the demand as to make its price fall below the costs of production; and what is meant by a general glut, is such an abundance of a large mass of commodities of different kinds, as to make them all fall below the natural price, or the ordinary costs of production, without any proportionate rise of price in any other equally large mass of commodities.
47With these preliminary definitions, we may proceed to examine some of the arguments by which Mr. Mill endeavours to show that demand and supply are always equal in the aggregate; that an over supply of some commodities must always be balanced by a proportionate under supply of others; and that, therefore, a general glut is impossible.
If Mr. Mill had always strictly adhered to that meaning of the term demand for a commodity which signifies the quantity consumed, he might have maintained the position with which he heads the third section of his fourth chapter, namely, that consumption is co-extensive with production. This, however, is, in reality, no more than saying, that if commodities were produced in such abundance as to be sold at half their cost of production, they would still be somehow or other consumed—a truism equally obvious and futile. But Mr. Mill has used the term demand in such a way, that he cannot shelter himself under this truism. He observes, “It is evident that whatever a man has produced, and does not wish to keep for his own consumption, is a 48stock which he may give in exchange for other commodities. His will, therefore, to purchase, and his means of purchasing, in other words, his demand, is exactly equal to the amount of what he has produced, and does not mean to consume.”[17]
Here it is evident that Mr. Mill uses the term demand in the sense of the amount of sacrifice which the purchaser is able to make, in order to obtain the commodity to be sold, or, as Mr. Mill correctly expresses it, his means of purchasing. But it is quite obvious that his means of purchasing other commodities are not proportioned to the quantity of his own commodity which he has produced, and wishes to part with; but to its value in exchange; and unless the value of a commodity in exchange be proportioned 49to its quantity, it cannot be true that the demand and supply of every individual are always equal to one another. According to the acknowledged laws of demand and supply, an increased quantity will often lower the value of the whole, and actually diminish the means of purchasing other commodities.
Mr. Mill asks, “What is it that is necessarily meant, when we say that the supply and the demand are accommodated to one another? It is this (he says) that goods which have been produced by a certain quantity of labour, exchange for goods which have been produced by an equal quantity of labour. Let this proposition be attended to, and all the rest is clear. Thus, if a pair of shoes is produced by an equal quantity of labour as a hat, so long as a hat exchanges for a pair of shoes, so long the supply and demand are accommodated to one another. If it should so happen that shoes fell in value, as compared with hats, which is the same thing as hats rising in value, as compared with shoes, this would imply that more shoes had been brought to 50market, as compared with hats. Shoes would then be in more than due abundance. Why? Because in them the produce of a certain quantity of labour would not exchange for the produce of an equal quantity. But for the very same reason, hats would be in less than due abundance, because the produce of a certain quantity of labour in them would exchange for the produce of more than an equal quantity in shoes.”[18]
Now, I have duly attended, according to Mr. Mill’s instructions, to the proposition which is to make all the rest clear; and yet the conclusions at which he wishes to arrive, appear to me as much enveloped in darkness as ever. This, indeed, was to be expected from the proposition itself, which obviously involves a most unwarranted definition of what is meant, when we say that the supply and the demand are accommodated to one another. It has already been stated that what has hitherto been meant, both in conversation and in the writings of the highest authority on political economy, by the supply being accommodated 51to, or equal to the demand, is, that the supply is just sufficient to accommodate all those who are able and willing to pay the natural and necessary price for it, in which case, of course, it will always sell at what Adam Smith calls its natural price.
Now, unless Mr. Mill is ready to maintain that people would still say that the supply of a commodity was accommodated to the demand for it, whether it were selling at three times the cost of its production, or only one-third of that cost, he cannot maintain his definition. He cannot, for instance, deny that hats and shoes may be both selling below the costs of production, although they may exchange for each other in such proportions, that the hats produced by a certain quantity of labour may exchange for the shoes produced by the same quantity of labour. But can it be said on this account, that the supply of hats is suited to the demand for hats, or the supply of shoes suited to the demand for shoes, when they are both so abundant that neither of them will exchange for what will fulfil the conditions of their continued supply? 52And supposing that, while both are selling below the costs of production, shoes should fall still lower than hats, what would be the consequence? According to Mr. Mill, “shoes would then be in more than due abundance. Why? Because in them the produce of a certain quantity of labour would not exchange for the produce of an equal quantity. But for the very same reason, hats would be in less than due abundance, because the produce of a certain quantity of labour in them would exchange for the produce of more than an equal quantity in shoes.”[19]
It will be most readily allowed that, in the case supposed, shoes will be in more than due abundance, though not for the reason given by Mr. Mill. But how can it be stated, with the least semblance of truth, that hats would be in less than due abundance, when, by the very supposition, they are selling at a price which will not re-purchase the quantity of labour employed in producing them.
Nothing can show more distinctly than the very case here produced by Mr. Mill, that his 53proposition or definition, which is to clear up everything, is wholly inapplicable to the question; and that to represent the abundance or deficiency of the supply of one commodity, as determined by the deficiency or abundance of another, is to give a view of the subject totally different from the reality, and calculated to lead to the most absurd conclusions. There is hardly any stage of society subsequent to the division of labour, where the state of the supply compared with the demand of shoes is essentially affected by the state of the supply compared with the demand for hats; and in the present state of society in this country, where the question of a general glut has arisen, it is still more irrelevant to advert to any other objects as efficient causes of demand for a particular commodity, except those which relate to the costs of producing it.
The hop-planter who takes a hundred bags of hops to Weyhill fair, thinks little more about the supply of hats and shoes than he does about the spots in the sun. What does he think about, then? and what does he want to exchange his hops for? Mr. Mill seems to 54be of opinion that it would show great ignorance of political economy, to say that what he wants is money; yet, notwithstanding the probable imputation of this great ignorance, I have no hesitation in distinctly asserting, that it really is money which he wants, and that this money he must obtain, in the present state of society, in exchange for the great mass of what he has brought to market, or he will be unable to carry on his business as a hop-planter; and for these specific reasons; first, that he must pay the rent of his hop grounds in money; secondly, that he must pay for his poles, his bags, his implements, &c., &c., in money; thirdly, that he must pay the numerous labourers which he employs on his grounds, during the course of the next year, in money; and fourthly, that it is in money, and in money alone of all the articles brought to the fair, that he can calculate his profits.
It is perfectly true, that both the landlords and the labourers who are paid in money will finally exchange it for something else, as no one enjoys money in kind, except the miser; but the landlord who may spend perhaps a 55good deal in post-horses, dinners at inns, and menial servants, would be little likely to accept from the hop-planter the articles which he could get at the fair in exchange for his hops; and though the expenditure of the labourer is much more simple, and may be said to consist almost entirely in food and clothing, yet it is quite certain that the power of commanding a given quantity of labour can never be represented, with any approach towards correctness, by a given quantity of corn and clothing. As a matter of fact, the labourer in this country is paid in money; and while it often happens that for many years together the money-price of labour remains the same, the money-price of corn is continually altering, and the labourer may, perhaps, receive the value of twice as much corn in one year as he does in another.
What an entirely false view, then, does it give of the real state of things, what a complete obscuration instead of illustration of the subject is it, to represent the demand for shoes as determined by the supply of hats, or the demand for hops by the supply of cloth, cheese, 56or even corn. In fact, the doctrine that one half of the commodities of a country necessarily constitute an adequate market or effectual demand for the other half, is utterly without foundation. The great producers who are the great sellers, before they can venture to think about the supplies of hats, shoes, and cloth, on which they may perhaps expend a tenth part of a tenth part of what they have brought to market, must first direct their whole attention to the replacing of their capital, and to the question whether, after replacing it, they will have realized fair profits. Whatever may be the number of intermediate acts of barter which may take place in regard to commodities—whether the producers send them to China,[20] or sell them in the place where they are produced: the question as to 57an adequate market for them, depends exclusively upon whether the producers can replace their capitals with ordinary profits, so as to enable them successfully to go on with their business.
But what are their capitals? They are, as Adam Smith states, the tools to work with, the materials to work upon, and the means of commanding the necessary quantity of labour. Colonel Torrens, therefore, is quite right, when he says, “that an increased production of those articles which do not form component parts of capital, cannot create an increased effectual demand, either for such articles themselves, or for those other articles which do form component parts of capital.”[21] And, perhaps, he may be considered as making some approaches towards the truth, when he says, that “effectual demand consists in the 58power and inclination, on the part of consumers, to give for commodities, either by immediate or circuitous barter, some greater proportion of all the ingredients of capital than their production costs.”[22] But in this latter position, he is still very far from representing what actually takes place. When we consider how much labour is directly employed in the production of the great mass of commodities, and recollect further, that raw materials and machinery, the other two branches of capital, are mainly produced by labour, it is obvious that the power of replacing capitals will mainly depend on the power of commanding labour: but a given quantity of what Colonel Torrens calls the ingredients of capital, can never represent a given quantity of labour; and consequently, if a given quantity of labour be necessary in any production, a very different quantity of the ingredients of capital would be required at different times, to occasion the same effectual demand for it. It is far, therefore, from being true, that if the 59ingredients of capital, represented by a hundred and ten quarters of corn, and a hundred and ten suits of clothing, were increased to “two hundred and twenty quarters of corn, and two hundred and twenty suits of clothing, the effectual demand for the article would be doubled.”[23]
It is still further from the truth, “that increased supply is the one and only cause of increased effectual demand;”[24] and most happy is it for mankind that this is not true. If it were, how difficult would it be for a society to recover itself, under a temporary diminution of food and clothing! But by a kind provision of nature, this diminution, within certain limits, instead of diminishing, will increase effectual demand. The theory of demand and supply, shows that the food and clothing thus diminished in quantity, will rise in value; and universal experience tells us, that, as a matter of fact, the money-price of the remaining food and clothing will for a time rise in a greater degree than in proportion to the diminution 60of its quantity, while the money-price of labour may remain the same. The necessary consequence will be, the power of setting in motion a greater quantity of productive industry than before.[25]
There is no assumption so entirely fatal to a just explanation of what is really taking place in society, as the assumption, that the natural wages of labour in food and clothing are always nearly the same, and just about sufficient to maintain a stationary population. All the most common causes of an acceleration or retardation in the movements of the great machine of human society, involve variations, 61and often great variations, in the real wages of labour. Commodities in general, and corn most particularly, are continually rising or falling in money-price, from the state of the supply as compared with the demand, while the money-price of labour remains much more nearly the same. In the case of a rise of corn and commodities, the real wages of common day-labour are necessarily diminished: the labourer obtains a smaller proportion of what he produces; profits necessarily rise; the capitalists have a greater power of commanding labour; more persons are called into full work, and the increased produce which follows, is the natural remedy for that state of the demand and supply, from whatever cause arising, which had occasioned the temporary rise in the money-price of commodities. On the other hand, if corn and other commodities fall in money-price, as compared with the money-price of labour, it is obvious that the day-labourer, who gets employment, will be able to buy more corn with the money which he receives; he obtains a larger proportion of what he produces; profits necessarily fall; the 62capitalists have a diminished power of commanding labour; fewer persons are fully employed, and the diminished production which follows, is the natural remedy for that state of the demand and supply, from whatever cause arising, which occasioned the temporary fall in the money-price of commodities. The operation of these remedial processes to prevent the continuance of excess or defect, is so much what one should naturally expect, and is so obviously confirmed by general experience, that it is inconceivable that a proposition should have obtained any currency which is founded on a supposed law of demand and supply diametrically opposed to these remedial processes.
It will be recollected, that the question of a glut is exclusively whether it may be general, as well as particular, and not whether it may be permanent as well as temporary. The causes above mentioned act powerfully to prevent the permanence either of glut or scarcity, and to regulate the supply of commodities so as to make them sell at their natural prices. But this tendency, in the natural course of 63things, to cure a glut or a scarcity, is no more a proof that such evils have never existed, than the tendency of the healing processes of nature to cure some disorders without assistance from man, is a proof that such disorders have not existed.
But to return more particularly to Mr. Mill. After asserting that the supply is the demand, and the demand is the supply, so frequently, that the unwary reader must feel quite at a loss to know which is which, he comes to a distinct conclusion, which is so directly contradicted both by theory and experience, as to show either that his premises must have been false, or that what he calls his indissoluble train of reasoning consists of mere unconnected links. He says, “It is therefore universally true, that as the aggregate demand and aggregate supply of a nation never can be unequal to one another, so there never can be a superabundant supply in particular instances, and hence a fall in exchangeable value below the cost of production, without a corresponding deficiency of supply, and hence a rise in exchangeable value beyond 64cost of production in other instances. The doctrine of the glut, therefore, seems to be disproved by a chain of reasoning perfectly indissoluble.”[26]
While commodities are merely compared with each other, it is unquestionably true that they cannot all fall together, or all rise together. But when they are compared with the costs of production, as they are in the above passage, it is evident that, consistently with the justest theory, they may all fall or rise at the same time. For what are the costs of production? They are either the quantity of money necessary to pay the labour worked up in the commodity, and in the tools and materials consumed in its production, with the ordinary profits upon the advances for the time that they have been advanced; or they are the quantity of labour in kind required to be worked up in the commodity, and in the tools and materials consumed in its production with such an additional quantity as is equivalent to the ordinary profits upon the advances for the time that they have been advanced.
65Now it surely cannot be denied theoretically, that all commodities produced in this country may fall in comparison with a commodity produced in Mexico. As little can it be denied theoretically that all commodities produced by British labour may fall as compared with that labour, either from an unusually increased supply of such commodities, or a diminution of demand for them. And when, from these theoretical concessions, required by the universally acknowledged laws of demand and supply, we turn to the facts, we see with our own eyes, and learn from authority which there is no reason whatever for doubting, that a very large mass of commodities does at times fall below the costs of production, whether those costs be estimated in money or labour, without the slightest shadow of pretence for saying that any other equally large mass is raised proportionally above the costs of production.
Even within the very last year, it is a matter of the most public notoriety that the cotton manufactures, the woollen manufactures, the linen manufactures, the silk manufactures, 66have all fallen below the costs of production, including ordinary profits. To go no further, the amount of these manufactures, taken together, must, on a rough estimate, exceed seventy millions of pounds sterling. And if this mass of commodities, partly from over production and over trading, and partly from their necessary consequences, the shock to confidence and credit and the diminution of bills of exchange and currency, have fallen below the ordinary costs of production, what man is there credulous enough to believe that there must have been, according to the language of Mr. Mill, “a corresponding deficiency of supply, and hence a rise of exchangeable value beyond cost of production in other instances”? I doubt, indeed, much, whether satisfactory evidence could be brought to show that a single million’s worth of goods has risen above the cost of production, while seventy millions’ worth have fallen below it.
Consequently, if the definition of a general glut be a fall in a great mass of commodities below the costs of production, not counterbalanced by a proportionate rise of some 67other equally large mass of commodities above the costs of production, Mr. Mill’s conclusion against the existence of a general glut, founded on “a chain of reasoning perfectly indissoluble,” seems to be utterly without foundation.
If facts so notorious as these to which I have adverted are either boldly denied, or considered as undeserving attention, in founding the theories of political economy, there is an end at once to the utility of the science.
On the subject of the wages of labour, Mr. Mill has added his authority to the peculiar views and language of Mr. Ricardo. He says, “Whatever the share of the labourer, such is the rate of wages; and, vice versâ, whatever the rate of wages, such is the share of the commodity or commodities’ worth which the labourer receives.”[27] Perhaps the term rate of wages used by Mr. Mill to express the proportion of the produce which falls to the share of the labourer is in some respects preferable to the term real wages, used by Mr. Ricardo for the same purpose; but still it is highly 68objectionable, because it is an old and familiar term used in an entirely new sense. When the expressions high or low rates of wages were used, before the time of Mr. Ricardo and Mr. Mill, no one understood them to mean the proportion of the produce awarded to the labourer. In fact, this meaning had not been before conveyed by any appropriate terms in the language of political economy; yet it is a meaning the expression of which was much wanted in explaining the theory of profits. To express it, therefore, a new term should certainly have been chosen, and not an old one, which was familiar in a different sense. There seems to be no objection to the term proportionate wages, which has been used by Mr. Macculloch.
On the whole, it must be allowed, that Mr. Mill in his Elements of Political Economy has but little attended to the most obvious rules which ought to guide political economists in the definition and application of their terms. They are often unsanctioned by the proper authorities, and rarely maintained with consistency.
However incautious Mr. Ricardo and Mr. Mill may have been in the definition and application of their terms, I fear it will be found that Mr. Macculloch has been still more so; and that, instead of growing more careful, the longer he considers the subject, he seems to be growing more rash and inconsiderate.
The expositors of any science are in general desirous of calling into their service definite and appropriate terms; and for this purpose their main object is to look for characteristic differences, not partial resemblances. Mr. Macculloch, on the other hand, seems to be only looking out for resemblances: and proceeding upon this principle, he is led to confound material with immaterial objects; productive with unproductive labour; capital with revenue; the food of the labourer 70with the labourer himself; production with consumption; and labour with profits.
That this is not an exaggerated view of what has been stated by Mr. Macculloch, in his Principles of Political Economy, any person who reads the work with attention may satisfy himself.
Mr. Macculloch’s definition of wealth, which he considers as quite unexceptionable, is, “those articles or products which possess exchangeable value, and are either necessary, useful, or agreeable.”[28]
It is not, perhaps, quite unexceptionable to use the term value in a definition of wealth. It is something like explaining ignotum per ignotius. But independently of this objection, the definition is so worded, that it is left in doubt whether immaterial gratifications are meant to be included in it. They are not in general designated by the terms articles or products; and it is only made clear that it is intended to include them by a collateral remark on my definition of wealth, which I confine specifically to material objects, and by a subsequent 71definition of productive labour, which is made to include every gratification derived from human exertion.
Mr. Macculloch, in the article on Political Economy which he published in the Supplement to the Encyclopædia Britannica, had excluded these kinds of gratification from his definition of wealth, and had given such reasons for this exclusion, as would fully have convinced me of its propriety, if I had not been convinced before. He observes that, “if political economy were to embrace a discussion of the production and distribution of all that is useful and agreeable, it would include within itself every other science; and the best Encyclopædia would really be the best treatise on political economy. Good health is useful and delightful, and therefore, on this hypothesis, the science of wealth ought to comprehend the science of medicine: civil and religious liberty are highly useful, and therefore the science of wealth must comprehend the science of politics: good acting is agreeable, and therefore, to be complete, the science of wealth must embrace a discussion of the principles of the histrionic art, and 72so on. Such definitions are obviously worse than useless. They can have no effect but to generate confused and perplexed notions respecting the objects and limits of the science, and to prevent the student ever acquiring a clear and distinct idea of the inquiries in which he is engaged.”[29]
On these grounds he confined wealth to material products; but, in the same treatise, he included, in his definition of productive labour, all those sources of gratification which he had, with such good reason, excluded from his definition of wealth. When he had done this, however, he could not but be struck with 73the inconsistency of saying that wealth consisted exclusively of material products, and yet that all labour was equally productive of wealth, whether it produced material products or not. To get rid of this inconsistency, he has now altered his definition, by leaving out the term material products; and it remains to be seen, whether in so doing he has not essentially deviated from the most obvious rules which should direct us in defining our terms.
His definition of wealth, as explained by what he subsequently says of productive labour, now includes all the gratifications derived from menial service and followers, whatever may be their number.
Now let us suppose two fertile countries with the same population and produce, in one of which it was the pride and pleasure of the landlords to employ their rents chiefly in maintaining menial servants and followers, and in the other, chiefly in the purchase of manufactures and the products of foreign commerce. It is evident that the different results would be nearly what I described in 74speaking of the consequences of the definition of the Economists. In the country, where the tastes and habits of the landlords led them to prefer material conveniencies and luxuries, there would, in the first place, be in all probability a much better division of landed property; secondly, supposing the same agricultural capital, there would be a very much greater quantity of manufacturing and mercantile capital; and thirdly, the structure of society would be totally different. In the one country, there would be a large body of persons living upon the profits of capital; in the other, comparatively a very small one: in the one, there would be a large middle class of society; in the other, the society would be divided almost entirely between a few great landlords and their menials and dependents: in the one country, good houses, good furniture, good clothes, and good carriages, would be in comparative abundance; while in the other, these conveniencies would be confined to a very few.
Now, I would ask, whether it would not be the grossest violation of all common language, 75and all common feelings and apprehensions, to say that the two countries were equally rich?
Mr. Macculloch, however, has discovered that there is a resemblance between the end accomplished by the menial servant or dependent, and by the manufacturer or agriculturist. He says, “The end of all human exertion is the same; that is, to increase the sum of necessaries, comforts, and enjoyments; and it must be left to the judgment of every one to determine what proportion of these comforts he will have in the shape of menial services, and what in the shape of material products.”[30]
It will, indeed, be readily allowed, that even the third footman who stands behind a coach, and seems only to add to the fatigue of the horses and the wear and tear of the carriage, is still employed to gratify some want or wish of man, in the same manner as the riband maker or the lace maker. It will further be readily allowed, that it is by no means politic to interfere with individuals in the modes of spending their incomes. But 76does it at all follow from this, that if these different kinds of labour have very different effects on society in regard to wealth, as the term is understood by the great mass of mankind, that they should not be distinguished by different appellations, in order to facilitate the explanation of these different effects? Mr. Macculloch might unquestionably discover some resemblance between the salt and the meat which it seasons: they both contribute, when used in proper proportions, to compose a palatable and nutritive meal, and in general we may leave it to the taste and discretion of the individual to determine these proportions; but are we on that account to confound the two substances together, and to affirm that they are equally nutritive? Are we to define and apply our terms in such a way as to make it follow from our statements, that, if the individual were to compound his repast of half salt and half meat, it would equally conduce to his health and strength?
But Mr. Macculloch states, that a taste for the gratifications derived from the unproductive labourers of Adam Smith “has exactly 77the same effect upon national wealth as a taste for tobacco, champagne, or any other luxury.”[31] This may be directly denied, unless we define wealth in such a manner as will entitle us to say that the enjoyments derived by a few great landlords, from the parade of menial servants and followers, will tell as effectually in an estimate of wealth as a large mass of manufacturers and foreign commodities. But when M. Chaptal endeavoured to estimate the wealth of France, and Mr. Colquhoun that of England, we do not find the value of these enjoyments computed in any of their tables. And certainly, if wealth means what it is understood to mean in common conversation and in the language of the highest authorities in the science of Political Economy, no effects on national wealth can or will be more distinct than those which result from a taste for material conveniencies and luxuries, and a taste for menial servants and followers. The exchange of the ordinary products of land for manufactures, tobacco, and champagne necessarily generates capital; 78and the more such exchanges prevail the more do those advantages prevail which result from the growth of capital and a better structure of society; while an exchange of necessaries for menial services, beyond a certain limited amount, obviously tends to check the growth of capital, and, if pushed to a considerable extent, to prevent accumulation entirely, and to keep a country permanently in a semi-barbarous state.
Mr. Macculloch, when not under the influence of his definition, justly observes, that “the great practical problem, involved in that part of the science of political economy which treats of the production of wealth, must resolve itself into a discussion of the means whereby the greatest amount of necessary, useful, and desirable products may be obtained with the least possible quantity of labour.”[32] But among the unproductive labourers of Adam Smith there is no room for such saving of labour. The pre-eminent advantages 79to be derived from capital, machinery, and the division of labour, are here almost entirely lost; and in most instances the saving of labour would defeat the very end in view, namely, the parade of attendance, and the pride of commanding a numerous body of followers.
Now, if the employment of the labour required to produce material conveniencies and luxuries necessarily occasions the creation and distribution of capital, and, further, affords room for all the advantages resulting from the saving of labour and the most extended use of machinery; while the employment of the labour, called by Adam Smith unproductive, is necessarily cut off from all these benefits, I would ask whether these two circumstances alone do not form a sufficiently marked line of distinction amply to justify the classification of Adam Smith; and the utility of such a classification, in explaining the causes of the wealth of nations, is most obvious and striking.
So difficult is it, consistently, to maintain a definition which contradicts the common usage of language, and the common feelings 80of mankind, that I have not the least doubt, if Mr. Macculloch himself were to travel through two countries of the kind before described, that is, one flourishing in manufactures and commerce, and the other, though with the same population and food, furnishing little more to the great mass of its people than panem et Circenses, he would call the latter poor, and the former comparatively rich.
Now, what must have been the cause of this difference? Adam Smith would give a simple, sufficient, and most intelligible reason for it. He would say, that the number and powers of those whom he had called productive labourers, had been much greater in one country than in the other. This seems to be a clear and satisfactory explanation. How Mr. Macculloch could explain the matter according to doctrines which make no difference between the different kinds of labour, I am utterly at a loss to conjecture[33].
81Perhaps, however, he would say, upon recollection, that his definition of wealth did not oblige him to allow that there would really be any difference in the wealth of these two countries. In that case, I think it may be very safely said that his definition of wealth violates all the most obvious rules for the definition and application of terms. It is opposed to the meaning of the term wealth as used in common conversation; it is opposed to the meaning of the term wealth as applied by the writers of the highest authority in political economy; it is so far from removing the little difficulties which had attended former definitions of wealth and productive labour, that it very greatly aggravates them; it so contradicts our common habits and feelings, that it is scarcely possible to maintain it with consistency.
Mr. Macculloch’s definition of capital has exactly the same kind of character as his definition of wealth, namely, that of being so extended as to destroy all precision, and to confound objects which had before been most usefully separated, with a view to the explanation 82of the causes of the wealth of nations. The alteration of a definition seems with Mr. Macculloch to be a matter of very slight consequence. The following passage is certainly a most extraordinary one. “The capital of a country may be defined to be that portion of the produce of industry existing in it, which can be made directly available, either to the support of human existence, or to the facilitating of production. This definition differs from that given by Dr. Smith, which has been adopted by most other economists. The whole produce of industry belonging to a country, is said to form its stock; and its capital is supposed to consist of that portion only of its stock, which is employed in the view of producing some species of commodities. The other portion of the stock of a country, or that which is employed to maintain its inhabitants, without any immediate view to production, has been denominated its revenue, and is not supposed to contribute anything to the increase of its wealth.”
“These distinctions seem to rest on no good foundations. Portions of stock employed 83without any immediate view to production, are often by far the most productive. The stock, for example, that Arkwright and Watt employed in their own consumption, and without which they could not have subsisted, was laid out as revenue; and yet it is quite certain that it contributed infinitely more to increase their own wealth, as well as that of the country, than any equal quantity of stock expended on the artisans in their service. It is always extremely difficult to say whether any portion of stock is, or is not, productively employed; and any definition of capital which involves the determination of this point, can only serve to embarrass and obscure a subject that is otherwise abundantly simple. In our view of the matter, it is enough to constitute an article capital, if it can either directly contribute to the support of man, or assist him in appropriating or producing commodities; but the question respecting the mode of employing an article ought certainly to be held to be, what it obviously is, perfectly distinct from the question whether that article is capital. For any thing that we can à priori know to 84the contrary, a horse yoked to a gentleman’s coach may be just as productively employed as if he were yoked to a brewer’s dray, though it is quite plain, that whatever difference may really obtain in the two cases, the identity of the horse is not affected; he is equally possessed, in the one case as well as the other, of the capacity to assist in production, and so long as he possesses that capacity, he ought to be viewed, independently of all other considerations, as a portion of the capital of the country.”[34]
If these doctrines were admitted, there would be an end, at once, of all classifications, and of all those appropriate designations which so essentially assist us, in explaining what is going forward in society. If the distinction between the whole mass of the products of a country, and those parts of it which are applied to perform particular functions, rests on no solid foundation, it may be asked, on what better foundation does the distinction between the mass of the male population of a country, and the classes of lawyers, physicians, 85manufacturers, and agriculturists rest? They all equally come under the general denomination of men; but particular classes are most usefully distinguished by particular appellations founded on the particular functions which they generally perform.
The bread which I consume myself, or give to a menial servant, is a part of the general produce of the country, and may not be different from that which is advanced to a manufacturer or agriculturist. When I or my servant consume the bread, it performs a most necessary and important service, no less than the maintenance of life and health; but in obtaining this service my wealth is pro tanto diminished. On the other hand, if I give the same kind of bread as wages to a manufacturer or agricultural labourer, it will not, with regard to me, perform so necessary an office as before, but it will perform an essentially different one with regard to my wealth, it will increase my wealth instead of diminishing it. In an inquiry into the causes of the wealth of nations, does not this difference 86in the functions which the same advances perform require to be marked by a particular appellation?
Accordingly, both in the language of common conversation and of the best writers, revenue and capital have always been distinguished; by revenue being understood, that which is expended with a view to immediate support and enjoyment, and by capital, that which is expended with a view to profit. But in the language of Mr. Macculloch, in the passage above quoted, it is the capacity to perform particular functions, and not the habitual performance of them, that justifies particular designations. A coach-horse, drawing a chariot in the Park, has the capacity of being employed in a brewer’s dray or a farmer’s waggon: “whatever difference may really obtain in the two cases, the identity of the horse is not affected; he is equally possessed, in the one case as in the other, of the capacity to assist in production; and so long as he possesses that capacity, he ought to be viewed, independently of all other 87considerations, as a portion of the capital of the country.”
This appears to me to be very little different from saying that a man who is capable of being made to perform the functions of a judge ought to be denominated a judge; because, whether he sits on the bench or in the court below, the identity of the man is the same; he is equally possessed, in the one case as well as the other, of the capacity to assist in the decision of causes, and so long as he possesses that capacity he ought to be viewed, independently of all other considerations, as one of the judges of the country. It is said, that the French are astonished at the small number of judges in England. If this kind of comprehensive nomenclature were adopted, their wonder would soon cease.
The whole of the incomes of every person in a society, in whatever way they may be actually employed, might be employed, as far as they would go, directly in the support of man. Consequently, according to the definitions 88of Mr. Macculloch, all incomes are capital. But he is not satisfied even with this very unusually-extended meaning of the term. He can trace a resemblance between a working man and a working horse, and is, consequently, led to say, “However extended the sense previously attached to the term capital may at first sight appear, I am satisfied that it ought to be interpreted still more comprehensively. Instead of understanding by capital all that portion of the produce of industry extrinsic to man, which may be applicable to his support, and to the facilitating of production, there does not seem to be any good reason why man himself should not, and very many why he should, be considered as forming a part of the national capital. Man is as much the produce of labour as any of the machines constructed by his agency; and it appears to us, that in all economical investigations he ought to be considered in precisely the same point of view.”[35]
That there is some resemblance between a 89working man and a working horse cannot for a moment be doubted; but is that sufficient reason why they should be confounded together under the name of capital? The question is not whether there is a partial resemblance between these two objects, but whether there is a characteristic difference; and surely there is a sufficient distinction in all economical investigations between a free man, and the horse, the machine, or the food which he uses, to warrant a different designation, especially when one of the greatest objects of all economical investigations, and certainly the most worthy, has been how to secure at all times a full sufficiency of the produce of industry extrinsic to man as compared with man himself.
It has been hitherto usual to say, that the happiness of the labouring classes of society depends chiefly upon the rate at which the capital of the country increases, compared with its population; but if the capital of the country includes its population, there is no meaning in the statement. Yet hardly any writer that I know of has more frequently 90made this statement than Mr. Macculloch himself. Nothing, indeed, can show more strikingly the extreme difficulty of maintaining consistently new and unusual definitions, than the frequency with which he seems to be compelled to use terms in their old and accustomed sense, notwithstanding the different definitions which he has given of them.
Thus, in his very peculiar and most untenable argument on the effects of absenteeism in Ireland, one of the reasons which he gives, why the absence of the landlords does not diminish the wealth of that country is, that they do not remove any capital from it, but merely what they would spend on their own gratifications. If, however, the definition of the capital of a country, as stated by Mr. Macculloch, be “that portion of the produce of industry existing in it which can be made directly available either to the support of human existence or to the facilitating of production,” it follows necessarily that they remove a considerable quantity of capital, as it will hardly be denied that the corn, cattle, and butter produced from their estates (which, 91after all the mystery about bills of exchange is done away, are practically the main articles exported to England for the payment of their rents) may be made directly available to the support of human existence.
Mr. Macculloch is also disposed to recommend emigration as one of the best means of relieving the distress of Ireland, by altering the proportion between capital and labour; but if, according to him, in all economical discussions, man is to be considered as capital, precisely like the machine which he uses or the food which he consumes, the emigration of a portion of the population will be to deprive the country of a portion of its capital, which has always been considered as most pernicious. Whatever, therefore, may be the merits or demerits of Mr. Macculloch’s reasoning on these subjects, independently of his definitions, it is obvious that the application of his definitions at once destroys it.
It need hardly be repeated, that in all the less strict sciences, definitions and classifications are seldom perfect and complete; but no reasonable man will refuse to take advantage 92of an imperfect instrument which is essentially useful, if no other more perfect one can be obtained. If it be found useful, with a view to an explanation of the causes of the wealth of nations, to make a distinction between the labours of agriculturists and manufactures, as compared with menial servants, followers, and buffoons, the utility of this distinction is not destroyed, though its perfect accuracy may be impeached, because, in a few instances, the labour of the menial servant is very similar to that of the productive labourer. The classification is formed upon the general character and general effects of one sort of industry as compared with another; and if, in these respects, the line of distinction is sufficiently marked, it is mere useless cavilling to dwell upon particular instances.[36]
93But even in the very case on which Mr. Macculloch lays his principal stress, the difference is such as fully to warrant a different classification. It is, no doubt, true that, to have a fire in an attic in London, it is equally necessary that the coals should be brought up stairs from the cellar, as that they should be brought up from the bottom of the coal-mine to the surface: it is equally true that there is some resemblance between carrying coals from the bottom of a house to the top, and carrying them from the bottom of a mine to the top; but there is still a most decided and characteristic difference in the two cases.
The miner is paid by the owner or worker of the mine, for the express purpose of increasing 94his wealth; the value of the miner’s labour is, therefore, charged with a profit upon the price of the coals; and the result of it would regularly enter into any estimate of national wealth. But when the same owner or worker of coal-mines pays a menial servant for bringing coals up from the yard to the drawing-room, he pays him for the express purpose of facilitating and rendering more convenient and agreeable, the consumption of that wealth which he had obtained through the instrumentality of the miner. The two instruments are used for purposes distinctly different, one to assist in obtaining wealth, the other to assist in consuming it. In an inquiry into the causes of the wealth of nations, I cannot easily conceive a more distinct and useful line of demarcation.
On the same principle, if it be found useful with a view to explanations, to distinguish, by a different name, the stock destined for immediate consumption, from the stock employed or kept, with a view to profit, surely we must not wait to investigate the peculiar talents of each individual, before we venture 95to characterise the nature of his expenditure; and if we find such men as Arkwright and Watt[37] most naturally and properly reserving, for their immediate consumption, the means of keeping up a handsome or splendid establishment for the gratification of themselves, their family, and their friends, make an exception in their favour, and call such an expenditure an outlay of capital, instead of a consumption of revenue, as we should call it in the case of all ordinary persons. Such an inquiry would impose a duty upon the writers in political economy, which it would be perfectly absurd to attempt to fulfil, as it would quite defeat the end of the proposed classifications; and with regard to the distinguished characters adverted to, it would surely be most unnecessary. In an estimate of national wealth, the genius of a Newton or a Milton is necessarily underrated, which only shows that there are other sources of admiration and delight besides wealth. But such men as Arkwright and Watt are quite safe in the hands of the political economist. The result 96of their genius and labour is exactly of that description which is estimated in the very great addition which it makes to the capital and revenue of the country, in the most natural and ordinary acceptation of these terms. And when the effects of their genius have been estimated in this way, it would not only lead to inextricable difficulty, but it would be obviously a double entry, to estimate, in addition, the value of the men as extraordinary machines. It would be like estimating the value of a commodity produced by a skilful artificer, and then adding his high wages, and putting both into an estimate of national wealth.
But it is difficult to say what may not be called wealth, or what labour may not be called productive, in Mr. Macculloch’s nomenclature. According to his view of the subject, any sort of exertion, or any sort of consumption which tends, however indirectly, to encourage production, ought to be denominated productive; and before we venture to call the most trivial sort of exercise or amusement, such as blowing bubbles, or building houses of cards 97unproductive, we must wait to see whether the person so employed does not work the harder for it afterwards.[38] But, not to mention the impossibility of any, the most useful classification, if such doctrines were admitted, and we were required to wait the result in each particular case, and make exceptions accordingly, I will venture to affirm, that if we once break down the distinction between the labour which is so directly productive of wealth as to be estimated in the value of the object produced, and the labour or exertion, which is so indirectly a cause of wealth, that its effect is incapable of definite estimation, we must necessarily introduce the greatest confusion into the science of political economy, and render the causes of the wealth of nations inexplicable. There is no kind of exertion or amusement which may not, upon this principle, be called productive. Walking, riding, driving, card-playing, billiard-playing, &c. &c. may all be, indirectly, causes of production; and according to Mr. Macculloch, “it is very like a truism, 98to say, that what is a cause of production must be productive.”[39]
But of all the indirect causes of production, the most powerful, beyond all question, is consumption.
If man were not to consume, how scanty, comparatively, would be the produce of the earth. Consumption, therefore, is the main fundamental cause of production; and if we are to put indirect causation on a footing with direct causation, as suggested by Mr. Macculloch, we must rank in the same class, the manufacturer and the billiard player, the producer and the consumer.
It is impossible that the science of political economy should not most essentially suffer from such a confusion of terms. Nothing can be clearer, than that, with a view to any thing like precision, and the means of intelligible explanation, it is absolutely necessary to designate by a different name the labour which is directly productive of wealth, from that which merely encourages it.
Another most extraordinary and inconceivable 99misnomer of Mr. Macculloch is, the extension of the term labour to all the operations of nature, and every variety of profits.
Adam Smith, and all other writers, who have happened to fall in my way, have meant, by the term labour, when unaccompanied by any specific adjunct, the exertions of human beings; and by the term wages of labour, the remuneration, whether in produce or money, paid to those human beings for their exertions. When Mr. Ricardo stated, that commodities exchanged with each other according to the quantity of labour worked up in them, there cannot be the least doubt that he meant the quantity of human labour immediately employed in their production, together with that portion of human labour worked up in the fixed and circulating capitals consumed in aiding such production. And it is undoubtedly true, referring merely to the relation of one commodity to another, and supposing all other things the same; that is, supposing profits to be the same, the proportion of fixed and circulating capitals to be the same, and the duration of the fixed capitals 100and the times of the returns of the circulating capitals the same, that then the relative values of the commodities will be determined by the quantity of human labour worked up in each.
But Mr. Macculloch could not but see that it was scarcely possible to take up two commodities, of different kinds, in which all these things would be the same, and, consequently, that such a supposition would be so inapplicable to the mass of commodities, as to be perfectly useless; and yet, without such a supposition, the proposition would be obviously false.
Instead, however, of correcting Mr. Ricardo’s proposition, as he was naturally called upon to do, by adding to the human labour worked up in the commodity, any other element which was found ordinarily to affect its value, and calling it by its ordinary name, he chose to retain Mr. Ricardo’s language, but entirely to alter its meaning. There is nothing that may not be proved by a new definition. A composition of flour, milk, suet, and stones is a plum-pudding; if by stones be meant plums. Upon this principle, Mr. 101Macculloch undertakes to show, that commodities do really exchange with each other according to the quantity of labour employed upon them; and it must be acknowledged, that in the instances which he has chosen he has not been deterred by apparent difficulties. He has taken the bull by the horns. The cases are nearly as strong as that of the plum-pudding.[40]
They are the two following—namely, that the increase of value which a cask of wine acquires, by being kept a certain number of years untouched in a cellar, is occasioned by the increased quantity of labour employed upon it; and that an oak tree of a hundred years’ growth, worth 25l., which may not have been touched by man, beast, or machine for a century, derives its whole value from labour.
Mr. Macculloch acknowledges that Mr. Ricardo was inclined to modify his grand principle, that the exchangeable value of commodities depended on the quantity of labour required for their production, so far as to allow that the additional exchangeable value that 102is sometimes given to commodities, by keeping them after they have been purchased or produced until they become fit to be used, was not to be considered as an effect of labour, but as an equivalent for the profits which the capital laid out on the commodities would have yielded had it been actually employed.[41] This was looking at the subject in the true point of view, and showing that he would not get out of the difficulty by changing the meaning of the term labour; but Mr. Macculloch says—
“I confess, however, notwithstanding the hesitation I cannot but feel in differing from so great an authority, that I see no good reason for making this exception. Suppose, to illustrate the principle, that a cask of new wine, which cost 50l., is put into a cellar, and that at the end of twelve months it is worth 55l., the question is, whether ought the 5l. of additional value given to the wine to be considered as a compensation for the time the 50l. worth of capital has been locked up, or ought it to be considered as the value of 103additional labour actually laid out on the wine. I think that it ought to be considered in the latter point of view, and for this, as it appears to me a most satisfactory and conclusive reason, that if we keep a commodity, as a cask of wine which has not arrived at maturity, and on which therefore a change or effect is to be produced, it will be possessed of additional value at the year’s end; whereas, had we kept a cask of wine which had already arrived at maturity, and on which no beneficial or desirable effect could be produced for a hundred or a thousand years, it would not have been worth a single additional farthing. This seems to prove incontrovertibly that the additional value acquired by the wine during the period it has been kept in the cellar is not a compensation or return for time, but for the effect or change that has been produced on it. Time cannot of itself produce any effect, it merely affords space for really efficient causes to operate; and it is therefore clear, that it can have nothing to do with the value.”[42]
104On this passage it should be remarked, in the first place, that the question stated in it is not the main question in reference to the new meaning which Mr. Macculloch must give to the term labour, in order to make out his proposition. He acknowledges that the increased value acquired by the wine is either owing to the operation of nature during the year in improving its quality, or to the profits acquired by the capitalist for being deprived for a year from using his capital of 50l. in any other way. But in either case Mr. Macculloch’s language is quite unwarranted. When he uses the expression, “additional labour actually laid out upon the wine,” who could possibly imagine that, instead of meaning human labour, he meant the processes carried on by nature in a cask of wine during the time that it is kept. This is at once giving an entirely new meaning to the term labour.
But, further, it is most justly stated by Mr. Ricardo, that when the powers of nature can be called into action in unlimited abundance, she always works gratis; and her processes never add to the value, though they 105may add very greatly to the utility of the objects to which they are applied.
This truth is also fully adopted and strongly stated by Mr. Macculloch himself. “All the rude products (he says) and all the productive powers and capacities of nature are gratuitously offered to man. Nature is not niggardly or parsimonious; she neither demands nor receives an equivalent for her favours. An object which it does not require any portion of labour to appropriate or to adapt to our use may be of the very highest utility, but as it is the free gift of nature, it is utterly impossible it can be possessed of the smallest value.”[43] Consequently, as the processes which are carrying on in the cask of wine, while it is kept, are unquestionably the free gift of nature, and are at the service of all who want them, it is utterly impossible, even if their effects were ten times greater than they are, that they should add in the smallest degree to the price of the wine. It is, no doubt, perfectly true, as stated by Mr. Macculloch, that if wine were not improved 106by keeping, it would not be worth a single additional farthing after being kept a hundred or even a thousand years. But this proves nothing but that, in that case, no one would ever think of keeping wine longer than was absolutely necessary for its convenient sale or convenient consumption.
The improvement which wine derives from keeping is unquestionably the cause of its being kept; but when on this account the wine-merchant has kept his wine, the additional price which he is enabled to put upon it is regulated upon principles totally distinct from the average degree of improvement which the wine acquires. It is regulated exclusively, as stated by Mr. Ricardo, by the average profits which the capital engaged in keeping the wine would have yielded if it had been actively employed; and that this is the regulating principle of the additional price, and not the degree of improvement, is quite certain: because it would be universally allowed that if, in the case supposed by Mr. Macculloch, the ordinary rate of profits had been 20 per cent., instead of 10 per cent., a 107cask of new wine, worth 50l., after it had been kept a year, would have been increased in value 10l. instead of 5l., although the processes of nature and the improvement of the wine were precisely the same in the two cases; and there cannot be the least doubt, as I said before, that if the quality of wine, by a year’s keeping, were ordinarily improved in a degree ten times as great as at present, the prices of wines would not be raised; because, if they were so raised, all wine-merchants who sold kept wines would be making greater profits than other dealers.
Nothing then can be clearer than that the additional value of the kept wine is derived from the additional amount of profits of which it is composed, determined by the time for which the capital was advanced and the ordinary rate of profits.
The value of the oak tree of a hundred years’ growth is derived, in a very considerable degree, from the same cause; though, in rich and cultivated countries, where alone it could be worth 25l., rent would necessarily form a part of this value.
108If the number of acorns necessary on an average to rear one good oak were planted by the hand of man, they would be planted on appropriated land; and as land is limited in quantity, the powers of vegetation in the land cannot be called into action by every one who is in possession of acorns, in the same way as the improving operations of nature may be called into action by every person who possesses a cask of wine. But setting this part of the value aside, and supposing the acorns to be planted at a certain expense, it is quite clear, that almost the whole of the remaining value would be derived from the compound interest or profits upon the advances of the labour required for the first planting of the acorns, and the subsequent protection of the young trees. A much larger part, therefore, of the final value of the tree than of the final value of the wine would be owing to profits.
Now, if we were to compare an oak tree, worth 25l., with a quantity of hardware worth the same sum, the value of which was chiefly made up of human labour; and as the reason 109why these two objects were of the same value, were to state that the same quantity of labour had been worked up in them—we should obviously state a direct falsity, according to the common usage of language; and nothing could make the statement true, but the magical influence of a new meaning given to the term labour. But to make labour mean profits, or fermentation, or vegetation, or rent, appears to me quite as unwarrantable as to make stones mean plums.
To measure profits by labour is totally a different thing. Adam Smith always keeps wages, profits, and rent quite distinct; and when he mentions one of them, never thinks of including in the same term any other. But he observes, that “labour measures the value not only of that part of the price of a commodity which resolves itself into labour, but of that which resolves itself into rent, and of that which resolves itself into profit.”[44] This is perfectly just; and, in particular, nothing can be more natural and obvious than to measure by labour the increase of value which commodities derive from profits; because 110profits are a per centage upon the advances, and the main original advances in the great mass of commodities are the necessary quantity of labour.[45]
Thus, if a hundred days’ labour be advanced for a year,[45] in order to produce a commodity, and the rate of profits be 10 per cent., it is impossible in any way to represent so correctly the increase of value which the commodity derives from profits as by adding 10 per cent., or whatever may be the rate of profits, to the quantity of labour actually employed, and saying, that the completed commodity when sold would be worth ten days’ labour more than the quantity of labour worked up in it. On the other hand, if we were ignorant of the rate of profits, but found that a hundred days’ labour advanced for a year would produce a commodity which would ordinarily sell for the value of one hundred and ten days, we might safely conclude that ordinary profits were 10 per cent.
111Now, if we were to compare two commodities, on each of which a hundred days’ labour had been employed, and one of them could be brought to market immediately, the other in not less time than a year, it is quite obvious, that we could not say that they would exchange with each other according to the quantity of labour worked up in them; but we evidently could say, that they would exchange with each other according to the quantity of labour and of profits worked up in them, and that one of them would be 10 per cent. more valuable than the other, because profits had added the value of ten days’ labour to the labour actually employed upon the one; while there being no profits in the other, its value was only in proportion to the labour actually employed upon it.
And in general, while the slightest examination of what is passing around us must convince us that commodities, under deduction of rent and taxes, do not ordinarily exchange with each according to the quantity of human labour worked up in them, the same examination will convince us that, under 112the same deduction, they do ordinarily exchange with each other, according to the quantity of human labour and of profits worked up in them; and further, that the quantity of human labour worked up in them, with the profits upon the advances for the time that they have been advanced, is correctly measured by the quantity of human labour of the same kind which the commodity so composed will ordinarily command.
We must carefully, therefore, distinguish between measuring profits by labour, and meaning profits by labour; and while the first is obviously justifiable, and may be in the highest degree useful, it must be allowed, that the latter contradicts all the most obvious rules for the use of terms: it contradicts the usage of common conversation: it contradicts the highest authorities in the science of political economy: it embarrasses all explanations; and it cannot be maintained with consistency.
Though Mr. Macculloch’s work affords other instances of a want of attention, on a point so important in all philosophical discussions, 113as appropriate and consistent definitions, I will only notice further, his use of the term real. He applies it to wages, in two senses entirely different.
In part iii. p. 294, he says, “But if the variation in the rate of wages be real, and not nominal, that is, if the labourer be getting either a greater or less proportion of the produce of his industry, or a greater or less quantity of money of invariable value, this will not happen.” Here, it is evident that Mr. Macculloch applies the term real to wages, in the sense of proportional wages, that is, as Mr. Ricardo applied it.
In part iii. p. 365, Mr. Macculloch says, “If the productiveness of industry were to diminish, proportional wages might rise, notwithstanding that real wages, or the absolute amount of the produce of industry falling to the share of the labourer, might be diminished. Here, the term real wages is used as synonymous with the absolute amount of produce falling to the share of the labourer, that is, in the sense in which Adam Smith has applied it.”
114I have already observed, that Adam Smith’s application of the term real wages, to the absolute quantity of the produce earned by the labourer, seems to be a most natural one; and Mr. Ricardo’s application of the same term to the proportion of the produce earned by the labourer, a most unnatural one. Mr. Macculloch, therefore, was quite right, in introducing the term proportionate wages, to express Mr. Ricardo’s meaning; but why not adhere to it? Why should he, in some places, mean, by real wages, proportionate wages, and, in other places, something totally different?
In the application of the term real to value, Mr. Macculloch adopts the meaning of Mr. Ricardo. He says, indeed, “that it is to Mr. Ricardo’s sagacity, in distinguishing between the quantity of labour required to produce commodities, and the quantity of labour for which they will exchange, and in showing, that while the first is undeniably correct as a measure of their real, and generally speaking, of their exchangeable values, the second, instead of being an equivalent proposition, is frequently opposed to 115the first, and consequently, quite inaccurate, that the science is indebted for one of its greatest improvements.”[46]
I should be sorry to think that Mr. Ricardo’s services to the science of political economy should rest principally upon the frail foundation, on which they are here placed; a foundation, which, as we have seen, Mr. Macculloch himself cannot defend, without totally altering the meaning of Mr. Ricardo’s words.
This is evident, in various passages of Mr. Macculloch’s work. In his section on value, part ii. p. 216, he thus expresses himself: “assuming the toil and trouble of acquiring any thing to be the measure of its 116real value, or of the esteem in which it is held by its possessor.” Again, he says, p. 219, “the real value of a commodity, or the estimation in which it is held by its possessor, is measured or determined by the quantity of labour required to produce or obtain it.”
In these two passages, he obviously identifies the real value of a commodity with the estimation in which it is held. But, surely, in this case, the term real must be applied as Adam Smith applies it, and not as Mr. Ricardo applies it? Can it be contended for a moment, that a commodity, which, on account of the necessary remuneration for profits, sells for ten per cent. above the value of the human labour worked up in it, is not held in higher estimation, than a commodity which sells for ten per cent. less, on account of the value of the labour employed upon it not having been increased by profits? Would it not be absolutely certain, that if the latter could be obtained by the sacrifice of a hundred days’ labour, it would be necessary to make the sacrifice of a hundred 117and ten days’ labour, or some equivalent for it, in order to obtain the former? Consequently it follows necessarily, that if the real value of a commodity be considered as synonymous with the estimation in which it is held, such value must be measured by the quantity of labour which it will command, and not the quantity worked up in it.
Mr. Macculloch thus states Mr. Ricardo’s main proposition:[47] “a commodity, produced by a certain quantity of labour, will, in the state of the market now supposed, (that is, when the market is not affected by either real or artificial monopolies, and when the supply of commodities is equal to the effectual demand,) uniformly exchange for, or buy any other commodity, produced by the same quantity of labour.”
Now, if the term labour be taken in the sense in which it is used by Mr. Ricardo, the proposition is contradicted by universal experience. If, on the other hand, the term labour be considered as including profits, the proposition is true; but only because it is 118a totally different one from that of Mr. Ricardo, owing to a most unwarrantable perversion of terms.
It appears, then, on the whole, that although Mr. Macculloch has at different times compared Adam Smith to Newton and to Locke, he has, in the definition and application of his terms, differed from him on almost all the most important subjects of Political Economy,—in the definition of wealth, the definition of capital, the definition of productive and unproductive labour, the definition of profits, the definition of labour simply, and the definition of real value, though, in the last instance, it is rather professedly than substantially.[48]
However highly I may respect the authority of Adam Smith, and however inconvenient at first a great change of terms and meanings must necessarily be, yet if it could 119be made out that such changes would essentially facilitate the explanation and improvement of the science of political economy, I should have been the last to oppose them. But after considering them with much attention, I own I feel the strongest conviction that they are eminently the reverse of being useful, with a view to an explanation of the nature and causes of the wealth of nations; or, in more modern, though not more appropriate phrase, the production, distribution, and consumption of wealth.
I have too much respect for Mr. Macculloch to suppose that he has differed from Adam Smith on so many points with the intention of giving to his work a greater air of originality. This is, no doubt, a feeling which not unfrequently operates in favour of changes; but I do not think it did on the present occasion. I should rather suppose that he adopted them in consequence of seeing some objections to Adam Smith’s definitions, without being sufficiently aware that, in the less strict sciences, nothing is so easy as to find some objection to a definition, 120and nothing so difficult as to substitute an unobjectionable one in its place.
Whether the definitions substituted for those of Adam Smith on the present occasion have removed the objections to them which Mr. Macculloch may have felt, I cannot be a competent judge; but even supposing them to have done this, I think I can confidently affirm that they have left other objections, beyond all comparison greater and more embarrassing. And on this point I would beg those of my readers who are inclined to pay attention to these subjects, seriously and candidly to trace the consequences to the science of political economy, in regard to its explanation and practical application, of adopting Mr. Macculloch’s definitions. They are not, indeed, all his own; but the very extraordinary extension which he has given to the term capital, the making of no distinction between directly productive consumption and consumption that is only indirectly productive; and the extension of the term labour, without any adjunct, to mean profits, fermentation, 121and vegetation, belong, I believe, exclusively to Mr. Macculloch; and, I think, it will be found that they are beyond the rest strikingly calculated to introduce uncertainty and confusion into the science.
The tendency of some of our most popular writers to innovate without improving, and their marked inattention to facts, leading necessarily to differences of opinion and uncertainty of conclusion, have been the main causes which have of late thrown some discredit on the science of political economy. Nor can this be a matter of much surprise, though it may be of regret.
At a period, when all the merchants of our own country, and many in others, find the utmost difficulty in employing their capitals so as to obtain ordinary profits, they are repeatedly told that, according to the principles of political economy, no difficulty can ever be found in employing capital, if it be laid out in the production of the proper articles; and that any distress which they may have suffered is exclusively owing to a wrong application of their capital, such as “the production 122of cottons, which were not wanted, instead of broad cloths, which were wanted.”[49] They are, further, gravely assured, that if they find any difficulty in exchanging what they have produced, for what they wish to obtain for it, “they have an obvious resource at hand; they can abandon the production of the commodities which they do not want, and apply themselves directly to the production of those that they do want, or of substitutes for them;”[50] and this consolatory recommendation is perhaps addressed to a merchant who is desirous of obtaining, by the employment of his capital at the ordinary rate of profits, such an income as will enable him to get a governess for his daughters, and to send his boys to school and college.
At such times, assertions like these, and the proposal of such a remedy, appear to me little different from an assertion, on supposed philosophical principles, that it cannot rain, when crowds of people are getting wet 123through, and the proposal to go without clothes in order to prevent the inconvenience arising from a wet coat. If assertions so contrary to the most glaring facts, and remedies so preposterously ridiculous, in a civilized country,[51] are said to be dictated by the principles of political economy, it cannot be matter of wonder that many have little faith in them. And till the theories of popular writers on political economy cease to be in 124direct opposition to general experience; and till some steadiness is given to the science by a greater degree of care among its professors, not to alter without improving,—it cannot be expected that it should attain that general influence in society which (its principles being just) would be of the highest practical utility.
It might be thought that I was not called upon to notice the deviations from the most obvious rules for the use of terms in a Critical Dissertation on the Nature, Measure, and Causes of Value, by an anonymous writer. But the great importance of the subject itself at the present moment, when it may be said to be sub judice, the tone of scientific precision in which the dissertation is written, notwithstanding its fundamental errors, and the impression which it is understood to have made among some considerable political economists, seem to call for and justify attention to it.
The author, in his preface, observes, that “Writers on political economy have generally contented themselves with a short definition 126of the term value, and a distinction of the property denoted by it into several kinds, and have then proceeded to employ the word with various degrees of laxity. Not one of them has brought into distinct view the nature of the idea represented by this term, or the inferences which a full perception of its meaning immediately suggests; and the neglect of this preliminary has created differences of opinion and perplexities of thought which otherwise could never have existed.”[52]
Now it appears to me, that the author, at his first setting out, has in an eminent degree fallen into the very errors which he has here animadverted upon.
He begins by stating, very justly, that “value, in its ultimate sense, appears to mean the esteem in which any object is held;” and then proceeds to state, in the most lax and inconsequent manner, that “It is only when objects are considered together as subjects of preference or exchange that the specific feeling of value can arise. When they are so considered, our esteem for one 127object, or our wish to possess it, may be equal to, or greater, or less than our esteem for another; it may, for instance, be doubly as great, or, in other words, we would give one of the former for two of the latter. So long as we regarded objects singly, we might feel a great degree of admiration or fondness for them, but we could not express our emotions in any definite manner. When, however, we regard two objects, as subjects of choice or exchange, we appear to acquire the power of expressing our feelings with precision; we say, for instance, that one a is, in our estimation, equal to two b.... The value of a is expressed by the quantity b, for which it will exchange, and the value of b is, in the same way, expressed by the quantity of a.”[53]
So, then, it appears, as a consequence of value, meaning the esteem in which an object is held, that if there were two sorts of fruit in a country, called a and b, both very plentiful in the summer, and both very scarce in the winter; and if in both seasons they were to bear the same relation to each other, the 128feelings of the inhabitants with regard to the fruit a would be expressed with precision, by saying that, as it would always command the same quantity of the fruit b, it would continue to be of the same value—that is, would be held in the same estimation in summer as in winter.
It appears, further, that in a country where there were only deer, and no beavers or other products to compare them with, the specific feeling of value for deer could not arise among the inhabitants; although, on account of the high esteem in which they were held, any man would willingly walk fifty miles in order to get one!! These are, to be sure, very strange conclusions, but they follow directly from the previous statements.
The author, however, nothing daunted, goes on to say, that “If from any consideration, or number of considerations, men esteem one a as highly as two b, and are willing to exchange the two commodities in that ratio, it may be correctly said that a has the power of commanding two b, or that b has the power of commanding half of a.”
129“The definition of Adam Smith, therefore, that the value of an object expresses the power of purchasing other goods which the possession of that object conveys, is substantially correct; and as it is plain and intelligible, it may be taken as the basis of our subsequent reasonings without any further metaphysical investigation.”[54]
In a Critical Dissertation on Value, which is introduced with a heavy complaint against all preceding political economists for neglecting the preliminary labour necessary to give a full perception of its meaning, it might naturally have been expected, that previous to the final adoption of the meaning in which it was intended to use the term, throughout the dissertation, the consideration, or number of considerations, which induce men to prefer one object to another, or to give two b for one a, should be carefully investigated. But nothing of this kind is done. A definition of the value of an object by Adam Smith, which, as he afterwards clearly shows, requires explanation and modification, is arbitrarily 130adopted, or, in the language of the author, is “taken as the basis of his subsequent reasonings, without any further metaphysical investigation.”
That this first general description of value in exchange by Adam Smith does not, without further explanation, convey to the reader the prevailing meaning which he himself attaches to the term, is obvious in many passages of his work, and particularly in his elaborate inquiry into the value of silver during the four last centuries. He there shows, in the most satisfactory manner, that, in the progress of cultivation and improvement, there is a class of commodities, such as cattle, wood, pigs, poultry, &c., which, on account of their becoming comparatively more scarce and difficult of attainment, necessarily rise in value; yet he particularly states, that this rise in their value is not connected with any degradation in the value of silver,[55] although it is obvious that, other things being the same, a pound of silver would have a smaller power of purchasing other goods.
131Nothing, indeed, can be clearer than that this general description of value requires further explanation. There is the greatest difference imaginable between an increased power in any object of purchasing other goods, arising from its scarcity and the increased difficulty of procuring it; and the increase of its power to purchase other goods arising from the increased plenty of such goods and the increased facility of procuring them. Nor is it easy to conceive any distinction more vital to the subject of value, as the term is generally understood, or more necessary to “a full perception of its meaning.”
I cannot but think, therefore, that the author, under all the circumstances of the case, was not justified in adopting this definition of Adam Smith without further investigation.
But the adoption of this definition by the author in so unceremonious a manner, though quite inconsistent with the declarations in the preface, and most unpromising in regard to any improvement of the science which might have been expected from the dissertation, is 132by no means the gravest offence which he has committed in the opening of his subject.
Adam Smith’s definition, taken as it stands, however imperfect it may be, would still serve as a rough but useful standard of value in those cases where, in using the most ordinary forms of expression, some kind of standard is tacitly referred to, and no other more accurate one had been adopted.
But how is this definition of Adam Smith to be interpreted? If we understand it in the sense usually conveyed by the terms employed, it is impossible to doubt that by the power of purchasing other goods is meant the power of purchasing other goods generally. Who, then, could have conceived before-hand that the author would have inferred from this definition that he was justified in representing the power of purchasing other goods by the power of purchasing any one sort of goods which might first come to hand?—so that, considering the value of money in this country to be proportioned to its general power of purchasing, it would be correct to say that 133the value of an ounce of silver was proportioned to the quantity of apples which it would command; and that when it commanded more apples, the value of silver rose—when it commanded fewer apples, the value of silver fell.
It is, no doubt, quite allowable to compare any two commodities whatever together in regard to their value in exchange, and, among others, silver and apples. It is also allowable to say, though it would in general sound very strange, that the value of an ounce of silver, estimated in apples, is the quantity of apples it will command, provided that, by thus using the qualifying expression estimated in apples, immediately after the word value, we distinctly give notice to the reader that we are not going to speak of the exchangeable value of silver generally, according to the definition of Adam Smith, but merely in the very confined sense of its relation to one particular article. But if, without this distinct notice to the reader, we simply say that the value of an ounce of silver is expressed by the quantity of apples for which it will exchange, 134or, in the words of the author, that “the value of a is expressed by the quantity of b, for which it will exchange,” nothing can be more clear than that we use the term value in a manner totally unwarranted by the previous definition, that is, in a sense quite distinct from that in which Adam Smith uses it in the description of value adopted by the author.
Putting the corn and the circulating medium of a country out of the question, the relations of which to labour and the costs of producing various commodities are tolerably well known, I think no one, in ordinary conversation, has ever been heard to express the general power of purchasing by the power of purchasing some one particular commodity. I certainly, at least, myself never recollect to have heard these two very distinct meanings confounded. It would, indeed, sound very strange, if a person returning from India, on being asked what was the value of money in that country, were to mention the quantity of English broad cloth which a given quantity 135of money would exchange for, and to infer, in consequence, that the value of money was lower in India than in England.
In regard to the opinions and practice of other writers on political economy, most of them have considered the general power of purchasing, and the power of purchasing a particular commodity as so essentially distinct, that they have given them different names. The only authority quoted with approbation by the author, is Colonel Torrens, whose views, as to the nature of value, appear to him, he says, to be sounder than those of any other writer. Yet, what does Colonel Torrens say on this subject?—“The term exchangeable value expresses the power of purchasing with respect to commodities in general. The term price denotes the same power with respect to some particular commodity, the quantity of which is given. Thus, when I speak of the exchangeable value of cotton as rising or falling, I imply, that it will purchase a greater or less quantity of corn, and wine, and labour, and other marketable commodities; but when I 136talk of the price of cotton as rising or falling, I mean, that it will purchase a greater or less quantity of some one particular commodity, such as corn, or wine, or labour, or money, which is either expressed or understood. Exchangeable value may rise, while price falls, or fall while price rises. For example; if cotton were, from any cause, to acquire twice its former power of purchasing, with respect to goods in general, while gold, the particular commodity in which the price of cotton is expressed, rose in a still higher ratio, and acquired four times its former power in the market, then, though the exchangeable value of cotton would be doubled, its price would fall one half. Again; if cotton would purchase only half the former quantity of commodities, while it purchased twice the quantity of some particular commodity, such as corn, or wine, or labour, or money, then its exchangeable value would have sunk one half, while its price, as expressed in corn, or wine, or labour, or money, became double. And again; if cotton, and the particular commodity in which price is expressed, should rise or fall in the same 137proportion with each other, then the exchangeable value of cotton, or its general power of purchasing, would fluctuate, while its price remained stationary.”[56]
It appears then, that, whether Colonel Torrens’s view of value be quite correct or not, he draws the most marked line of distinction possible between the power of purchasing generally, and the power of purchasing a particular commodity, and is decidedly of opinion, that the latter, which is the sense in which the author uses the term value, should not be called value, but price. The authority of Colonel Torrens, therefore, whose views on the subject of value the author considers as so sound, is directly against him.
But not only does Colonel Torrens attach a very different meaning to the term value, from that in which it is used by the author throughout the greatest part of his work, but the author himself, in his notes and illustrations,[57] has given extracts from almost all the distinguished writers in political economy, 138expressly for the purpose of showing the universality of an opinion respecting the nature and measure of value directly opposed to his own. The writers to whom he refers, are Adam Smith, Sir James Stuart, Lord Lauderdale, M. Storch, M. Say, Mr. Ricardo, myself, Colonel Torrens, Mrs. Marcet, Mr. Mill, the Templar’s Dialogues, and Mr. Blake.
In the case of a proposition the nature of which admits of a logical proof, authority is of no consequence; but in a question which relates to the meaning to be attached to a particular term, it is quite incredible that any person should thus have ventured to disregard it.
Much, however, of inconsistency, of illogical inference, and disregard of authority, might have been forgiven, if the proposed change in the meaning of the term value would introduce a much greater degree of clearness and precision into the language of political economy, and, in that way, be eminently useful to the progress of the science.
But, what would be the consequence of adopting the meaning which the author attaches 139to the term value, and of allowing, according to his own words, that “the value of a is expressed by the quantity of b for which it will exchange, and the value of b is, in the same way, expressed by the quantity of a?”[58] One of these consequences is strikingly described in the following passage of the author’s chapter on Real and Nominal Value, a distinction which he is pleased to call unmeaning. “The value of a commodity denoting its relation in exchange to some other commodity, we may speak of it as money-value, corn-value, cloth-value, according to the commodity with which it is compared: and hence there are a thousand different kinds of value, as many kinds of value as there are commodities in existence, and all are equally real and equally nominal.”[59]
This is precision with a vengeance. Now, though I am very far from intending to say that the writers on political economy have been sufficiently agreed as to the precise 140meaning which they attach to the terms value of a commodity, when no express reference is made to the object with which it is to be compared, yet, by drawing a marked line of distinction between what has been called the real value of commodities and their nominal value, or, more correctly, between their value and their price, they have avoided the prodigious confusion which would arise from a commodity having a thousand or ten thousand different values at the same time. Whenever they use the term value of a commodity alone, and speak of its rising or falling, if they do not mean money-price, they refer either to its power of purchasing generally, or to something expressive of its elementary cost of production.
In either case, some general and very important information is communicated; but the value of a commodity, in the sense understood by the author, might be expressed a hundred different ways, without conveying a rational answer to any person who had inquired about it.
Further; the use of the term value, in the 141sense understood by the author, is entirely superfluous. It has exactly the same meaning as the term price, except that the term price has this very decided advantage over it, namely, that when the price of a commodity is mentioned, without an express reference to any other object in which it is to be estimated, political economists have universally agreed to understand it as referring to money. This is a prodigious advantage in favour of the term price, and tends greatly to promote both facility and precision in the language of political economy. When I ask, what is the price of wheat in Poland? no one has the least doubt about my meaning, and I should, without fail, get the kind of answer I intended. But if I asked, what was the value of wheat in Poland? I might, according to the author, be answered in a thousand different ways, all equally proper, and yet not one of the answers be of the kind I wanted. Of course, whether I use the term value or price, if I always expressly subjoin the object to which I mean to refer, it will be quite indifferent to which term I resort. But it is vain to suppose that the public will 142submit to such constant and unnecessary circumlocution. It would quite alter the language of political economy; and the kind of abbreviation which has taken place in application to the term price could not take place in regard to value, according to the doctrines of the author; because, when the value of a commodity is used alone, like the price of a commodity, no one object rather than another is entitled to a preference for the expression of that value. The author says distinctly in a note,[60] that money-value has no greater claim to the general term value than any other kind of value. It is quite clear, therefore, that if the term value is only to be applied in the sense in which it is applied by the author, it would be much better to exclude it at once from the vocabulary of political economy as utterly useless, and only calculated to produce confusion.
It may be further observed, that the sense in which the author proposes to apply the term value, is so different from the sense in which it is understood in ordinary conversation, 143and among the best writers, that it would be quite impossible to maintain it with consistency. The author himself, however obstinately, at times, he seems to persevere in the peculiar meaning which he has given to the term value, frequently uses it by itself, without reference to any particular article in which he proposes to express it. Even in the titles of some of his chapters he does this; and when in Chapter XI. he discusses the distinction between value and riches, and in Chapter XI. the causes of value, we are entitled to complain, that he has not acted according to the instructions which he has given to others, and told us, either expressly, or by implication, in what article the value here mentioned is to be expressed.
Again; when he mentions the value of that corn which is produced on lands paying rent, and when he speaks, as he frequently does, of the value of capital,[61] he does not tell 144us in what he means to express the value of corn, or of capital, although he thinks that such a reference, either expressed or implied, is always necessary, and particularly says, “In the preceding pages it has been shown, that we can express the value of a commodity only by the quantity of some other commodity for which it will exchange.”[62]
The meaning, therefore, which he gives to the term value is such, that he cannot and does not maintain it consistently himself, much less can he expect that others should so maintain it.
It appears, then, that the author has arbitrarily adopted a meaning of the term value quite unwarranted by the usage of ordinary conversation, directly opposed to the 145authority of the best writers on political economy, pre-eminently and conspicuously useless; and of such a nature that it cannot be maintained with consistency.
And what does he do with his definition after so adopting it?
He applies it to try the truth of a number of propositions advanced by different writers, who, according to his own showing, have used the term in a very different sense.
This, I own, appears to me much the same kind of proceeding as if a person were to define a straight line to be something essentially different from a line lying evenly between its two extremes, and then were gravely to apply it to one proposition after another of Euclid, and show, as might easily be done, granting the definition, that the conclusions of the Grecian geometer were all wrong.
The perseverance with which the author proceeds gravely to apply his peculiar definition of value to other writers, who have defined it differently, is truly curious, and must be allowed to be a great waste of time and labour. If, as he says he has repeatedly 146stated, “to know the value of an article at any period is merely to know its relation in exchange to some other commodity;”[63] and if, as I believe, no previous writer, in referring to the value of an article at any period ever thought or said that it could be expressed by its relation in exchange to any other contemporary commodity indifferently, it might at once be presumed, without further trouble, that almost all former propositions involving the term value would turn out to be either false or futile. It was quite unnecessary for him, therefore, to go into the detail; but as he has done so, it may be useful to follow him in some of his conclusions, as it may assist in drawing attention to a subject which lies at the bottom of many of the difficulties in political economy, and has not been sufficiently considered.
One of the first effects of the author’s definition is to destroy the distinction between what many writers of great authority have called real value, and nominal value. I have already had occasion to observe, that Adam 147Smith, by applying the term real wages to express the necessaries and conveniencies of life earned by the labourer, had precluded himself from the power of applying it consistently to the value of a commodity, in order to express its power of commanding labour; because it is well known that the same quantity of labour will both produce and command, at different times and under different circumstances, a very different quantity of the necessaries and conveniencies of life. But putting aside for the present this acknowledged inconsistency of Adam Smith, and taking real value as distinguished from nominal in the sense in which the writers who have so applied it intended, the author’s observations on these writers are not a little extraordinary.
After noticing the doctrines of Adam Smith, Mr. Ricardo, and myself, on the subject of real and nominal value, he says, “After the disquisition on the nature of value in the preceding chapter, the distinction of it in this way must appear to be merely arbitrary and incapable of being turned to any use. What information is conveyed or what advance in 148argument is effected, by telling us that value estimated in one way is real, but in another, is nominal?”[64] He afterwards goes on to say, in reference to a passage in the Templar’s Dialogues, “It would not, however, probably have been written, had the author attended to the simple fact, that value must always imply value in something, and unless that something is indicated, the word conveys no information. Now, as the terms nominal and real do not denote anything in this way, they convey no precise information, and are liable to engender continual disputes, because their meaning is arbitrarily assumed.”[65]
These appear to me, I confess, to be very extraordinary observations. It must surely be allowed, that to compare a commodity either with the mass of other commodities, or with the elementary costs of production, is most essentially distinct from comparing it with some particular commodity named. And if so, writers are bound so to express themselves as to convey to their readers, which of 149the two they intend to refer to. Whether these writers have chosen the very best terms to express these ideas is another question; but that the ideas themselves are quite different, and that it is essential to the language of political economy that they should be distinguished by different terms, cannot admit of a doubt. It appears to me, therefore, almost inconceivable that the author should say, “What information is conveyed, or what advance in argument is effected, by telling us, that value estimated in one way is real, but in another, is nominal?” It might as well be said, that, in speaking of our planetary system, no information is conveyed by using different adjuncts to the term distance, in order to distinguish between the distances of the planets from the sun, and the relations of their distances to each other. And supposing it had been the habit of most writers to call the first distances real and the second relative, would it not be most strange to say that the distinction in this way of distance into two kinds is incapable of being turned to any use, as all distance is relative?
150The author is repeatedly dwelling upon the relative nature of value, as if he alone had considered it in this light; but no other writer that I have met with has ever appeared to me to use the term value without an intelligible reference expressed or implied to something else; and when the author says, in the passage above quoted, that value must always imply value in something which ought to be indicated, and that the terms nominal and real do not denote anything in this way, he appears to me, I own, to assert what is entirely without foundation. M. Say, for instance, in a passage quoted by the author in his notes,[66] observes, “There is this difference between a real and a relative variation of price; that the former is a change of value arising from an alteration of the changes of production; the latter a change arising from an alteration in the ratio of value of one particular commodity to other commodities.” Now is it possible to say with truth, that the real and relative values here described do not both refer to other objects, and that these 151objects are not so different as to require to be distinguished?
The author may, perhaps, say, that if both expressions are meant to be relative, why use the terms real, positive, or absolute? The answer is, that the usage of our language allows it, and that nothing is more common than the use of the terms real, positive, and absolute, in contradistinction to relative, when the former terms have relation to some more general object, particularly to anything which is considered as a standard, whether accurate or inaccurate.
Thus, in the illustration before adverted to, although all distances are relative, it would be quite justifiable to say, that if the earth was moving towards the farthest part of her orbit, her positive, absolute, or real distance from the sun was increasing, although her distance relatively to that of some other planet or comet, moving from the sun with greater velocity, was diminishing. Tall and short, rich and poor, are relative terms: yet surely we should be warranted in saying, that Peter was not only taller than his three brothers, but, 152really or positively, a tall man. In the first case he is said to be tall in relation to three individuals; but a stranger, knowing nothing of the height of these individuals, would obtain very little information from the statement. He would not know whether Peter was four feet, five feet, or six feet high: in the latter case, Peter is said to be tall in relation to the average or standard height of the race of men spoken of; and though the stranger might not have in his mind a perfectly accurate notion of this standard, yet he would immediately have before him the height of Peter within a few inches, instead of a few feet.
On the same principle, would it not be most ridiculous for any person gravely to propose that as rich and poor are relative terms, no one should ever call a man rich without mentioning at the same time the individual in relation to whom he was rich? It is perfectly well known, that when, in any particular place or country, a man is said to be a rich man, the term refers to a sort of loose standard, expressing either a certain command over the goods of this life, or a certain superiority in this 153respect over the mass of the society, which superiority it had been the custom to mark by this expression. In either case, it would be allowable to call the man really or positively rich. But if the proposed change were adopted, and instead of saying that Mr. John Doe was a rich man, we could only say that he was rich in relation to Mr. Richard Roe, as poor Richard might be little better than a pauper, Mr. Doe might, after all, be in very narrow circumstances.
It is clear, therefore, not only that the terms real and positive may be legitimately applied in contradistinction to relative, when a relation to some more general object or standard is intended; but that the difference between the two sorts of relations is of the utmost importance, and ought to be carefully distinguished. It is not easy to conceive, therefore, how any writer could suppose that the language of political economy would be improved by a definition which would destroy this distinction, and make as many kinds of value as there are commodities, all equally real and equally nominal. In reference to 154all other political economists, whenever they have used the term value of a commodity, without specifically mentioning the object in which they intended to estimate it, I have always felt myself authorised, consistently with their general language, to consider them as referring tacitly either to the mass of commodities, to the state of the supply compared with the demand, or to the elementary costs of production. But when the author of the Critical Dissertation uses the term value, which he does frequently without specific application, his general doctrine must leave the reader quite at a loss to conjecture what he means.
Proceeding on the same strange misapprehension or perversion of the language of other writers, the author says of the writer of the Templar’s Dialogues, “Following Mr. Ricardo, he appears entirely to lose sight of the relative nature of value, and, as I have remarked in the preceding chapter, to consider it as something positive and absolute; so that if there were only two commodities in the world, and they should both, by some circumstances 155or other, come to be produced by double the usual quantity of labour, they would both rise in real value, although their relation to each other would be undisturbed. According to this doctrine every thing might at once become more valuable by requiring at once more labour for its production; a position utterly at variance with the truth, that value denotes the relation in which commodities stand to each other as articles of exchange. Real value, in a word, is on this theory considered as the independent result of labour; and, consequently, if under any circumstances the quantity of labour is increased, the real value is increased. Hence the paradox, that it is impossible for a continually to increase in value—in real value observe, and yet command a continually decreasing quantity of b, and this although they were the only two commodities in existence. For it must not be supposed that the author means that a might increase in value in relation to a third commodity c, while it commanded a decreasing quantity of b; a 156proposition which is too self-evident to be insisted on; but he means that a might increase in a kind of value called real, which has no reference to any other commodity whatever.” Apply to the position of this author the rule recommended in the last chapter; inquire, when he speaks of value, value in what? and all the possible truth on the subject appears in its naked simplicity. He adds afterwards again, “value must be value in something, or in relation to something.”[67]
Now let the reader recollect that this passage was written by a person who sets out with saying that value in its ultimate sense appears to mean the esteem in which any object is held, and it will appear most remarkable.
In the first place, what can the author possibly mean by speaking of the kind of value here called real, as if it had no relation to any thing else? The Templar, it must surely be allowed, has explained himself with sufficient 157clearness that by real value he means value in relation to the producing labour.
Secondly, I would ask the writer, who says that the value of a commodity means the esteem in which it is held, whether the labour required to produce a commodity does not, beyond all comparison, express more nearly the esteem in which the commodity is held, than a reference to some other commodity the producing labour of which is utterly unknown, and may therefore be one day or one thousand days?
I have already stated that I decidedly differ from Mr. Ricardo, and it follows of course that I differ equally from the Templar, in thinking that the value of a commodity may be correctly expressed by referring to the producing labour alone; but compared with the expression of value proposed to be substituted by the author of the Critical Dissertation, it has a prodigious superiority. Let us try both, for instance, by the touch of the talisman recommended by the author himself. Let the question be the value of silver before the discovery of the American mines; and 158let us ask, as directed, value in relation to what? The Templar would answer, value in relation to the producing labour; and though in this answer a material ingredient of elementary value is omitted, yet I should collect from it some tolerable notion of the esteem in which silver was held at that time; and if I found, on comparison, that the producing labour was now three or four times less, I should be able, with tolerable certainty, to infer, that silver had grown more plentiful; and that four centuries ago a given quantity of silver was held in much greater esteem, that is, people would make a much greater sacrifice in order to obtain it, than at present.
On the other hand, if the author of the Critical Dissertation should speak of the value of silver before the discovery of the American mines, and we should ask, value in relation to what? the answer would be, “I have repeatedly stated that to know the value of an article at any period is merely to know its relation in exchange to some other commodity;” consequently, we should know the 159value of silver in the fifteenth century, or the esteem in which it was held, by comparing it with calicoes, although we might know nothing at all about the difficulty or facility of obtaining calicoes at that time. And if we were to proceed, as in the former case, and, with a view to ascertain the esteem in which silver was held in the fifteenth century, as compared with the esteem in which it is held in the nineteenth, were to mark the relation of silver to calicoes in the two periods, it would appear, that as, owing to the improvements in the cotton machinery, a given quantity of silver would command more calicoes now than formerly, silver should be considered as being held in higher estimation now than four centuries ago. Yet no person, I believe, not even the author himself, would agree to this conclusion. He would probably say that the comparison was merely between silver and calicoes, and had nothing to do with anything else. If this be all he means, why does he confuse his readers by stating that value means the esteem in which a commodity is held? and why does he say 160that to know the value of an article at any period is merely to know its relation in exchange to some other commodity? If all he means by the value of a commodity is its relation to some other, why did he not at once say, without ever talking about esteem, that the value of one commodity in relation to any other was expressed by the quantity of that other for which the first would exchange; and that, when the first rose in relation to the other, the other would always fall proportionably in relation to the first? If he had so expressed himself, his proposition would have obtained universal consent; it would have been a truism which had never been denied. But as long as he continues to talk of the esteem in which commodities are held, his readers must consider him as peculiarly inconsistent, if, on the supposition of there being only two commodities in existence, he prefers measuring the esteem in which one of them is held by its relation to the other, rather than by its relation to the producing labour. And they must further think, that while he continues to state that “to know the 161value of an article at any period is merely to know its relation in exchange to some other commodity,” he is stating a proposition which, according to the usual sense in which the word value is understood when so placed, is totally unfounded. No man, I believe, but the author would venture to say that he should know the value of silver four hundred years ago by knowing the quantity of calicoes which an ounce of silver would then command.
The sixth chapter of the author is entitled “On Measures of Value;” and the discussion of this subject leads him to such strange conclusions, that one cannot but feel the greatest surprise at his not seeing that he must have been proceeding in a wrong course. He ridicules the notion of its being necessary that a commodity should possess invariable value, in order to form a perfect measure of value. Such a notion, which he says in a note has been entertained by all the most distinguished writers in political economy, he civilly calls an utter absurdity. According to the doctrines and language of the author, no relation exists between the value of a commodity at one 162time and the value of the same sort of commodity at another; and “the only use of a measure of value, in the sense of a medium of comparison, is between commodities existing at the same time.”[68]
“If this be so, it is, no doubt, quite absurd in political economists to look for anything approaching towards an invariable measure of value, or even to talk of one commodity or object being more steady or constant in its value than another. At the same moment, bags of hops are as good a measure of the relative value of commodities as labour or money. With regard to money, indeed, the author particularly observes, that from the relations between corn and money, at two different periods, no other relation can be deduced; we do not advance a step beyond the information given. * * We cannot deduce the relation of value between corn at the first, and corn at the second period, because no such relation exists, nor, consequently, can we ascertain their comparative power over other commodities. If we made the 163attempt, it would be, in fact, endeavouring to infer the quantities of corn which exchanged for each other at two different periods of time, a thing obviously absurd. And further, money would not be here discharging a particular function any more than the other commodity. We should have the value of corn in money and the value of money in corn, but one would be no more a measure or medium of comparison than the other.”[69]
From all this it follows necessarily that we must on no account say, that butter has been rising during the last month; if we do, we shall be convicted of the absurdity of proposing to exchange the butter which was consumed three weeks ago with the butter now on our table, in order to ascertain that a pound of the former will command less than a pound of the latter. For the same reason, we must not on any account say, that the value of wheat fell very greatly from 1818 to 1822, and rose considerably from 1822 to 1826. We must not venture to compare the value of the advances of a master manufacturer 164with the value of his returns; or, in estimating the rate of his profits, presume to prefer money, which generally changes slowly and inconsiderably in its power of setting labour to work, to hops, which change so rapidly and greatly, &c. &c. In short, the whole of the language and inferences of the business of buying and selling, and making money, must be altered and adapted to the new definitions and doctrines.
It is quite astonishing that these consequences should not have startled the author, and made him turn back. If he had but adhered to his first description of value, namely, the esteem in which an object is held; or even if he had interpreted his second definition of value, namely, “the power of purchasing other goods,” according to the ordinary and natural meaning of the expression, he could never have been led into the strange mistake of supposing, that when people have talked of the value of a commodity at one period, compared with the value of the same kind of commodity at another, they could only refer to the rate at which they would 165actually exchange with each other, which, as no exchange could in such a case take place, would be absurd. What then did they mean? They obviously meant either to compare the esteem in which a commodity was held at one period with the esteem in which it was held at another, founded on the state of its supply compared with the demand, and ordinarily on its costs of production; or to compare the general power of purchasing which a commodity possessed at one period with its general power of purchasing at another period. And will the author venture to assert, that there are not some objects better calculated than others to measure this esteem, or measure this general power of purchasing at different periods? Will the author maintain, that if, in reference to two periods in the same country, a commodity of a given kind will in the second period command double the quantity of labour that it did in the first, we could not with much more certainty infer that the esteem for it had greatly increased, than if we had taken calicoes or currants as the medium of comparison? Or would the author, 166upon a little reflection, repeat again what he says in the passage last quoted, that from the relations between corn and money in two successive seasons, we can deduce no other relation, * * “money would not be here discharging a particular function any more than the other commodity. We should have the value of corn in money and the value of money in corn, but one would be no more a measure or medium of comparison than the other.”[70]
To me, at least, these statements appear utterly unfounded. If the money-price of corn has risen this year to double what it was in the last, I can infer, with almost absolute certainty, that corn is held in much higher estimation than it was. I can be quite sure that the relation of corn to other articles, besides money, has most essentially changed, and that a quarter of corn will now command a much greater quantity of labour, a much greater quantity of cloth, a much greater quantity of hardware, a much greater quantity of hats and shoes, than it did the year before: 167in short, that it will command nearly double the quantity of all other commodities which are in their natural and ordinary state, and have not been essentially affected by the causes which have operated upon the price of corn.
Where then is the truth of saying, that from the altered relation between corn and money we deduce no other relation? It is perfectly obvious that we can deduce and do deduce a great number of other most important relations; and, in fact, do ascertain, though not with perfect accuracy, yet with a most desirable and useful approach to it, the degree of increase in the power of corn to command in exchange the mass of other commodities.
On the other hand, from the diminished power of money in relation to corn, we cannot infer that money has fallen nearly in the same proportion in relation to other commodities. If an ounce of silver will now command only half a bushel of wheat, instead of a whole bushel, we can by no means infer that an ounce of silver will therefore command only about half the quantity of labour, half the 168quantity of cloth, half the quantity of hardware, half the quantity of hats and shoes, and of all those commodities which are in their natural and ordinary state. To all these objects money will probably bear nearly the same relation as before.
Where, then, is the truth of saying, that money would not be here discharging a particular function more than the other commodity? Broad, glaring, and incontrovertible facts show, that for short periods money does perform the function of measuring the variations in the general power of purchasing possessed by the corn; but that the corn does not measure the variations in the general power of purchasing possessed by the money. This is one of the instances of that extraordinary inattention to facts which, most unfortunately for the science of political economy, the professors of it have lately indulged themselves in.
The author has said a great deal in good set phrase about the false analogy involved in the application of the term measure to the value of commodities at different periods; and 169gravely states the difference between measuring length at different periods and measuring value.
I was not aware that people were ignorant of this difference. As I said before, whenever mention is made of the value of a commodity at different periods, I have always thought that a reference has been intended either to its general power of purchasing, or to something calculated to express the estimation in which it was held at these different periods, founded on the state of its supply compared with the demand, or the elementary costs of its production.
But if the term has been generally understood in this way, people must have been fully aware that value was essentially different from length: they would know perfectly well that a piece of cloth of a yard long would continue to be a yard long when it was sent to China; but that its value, that is, its general power of purchasing in China, or the estimation in which it was held there, would probably be essentially altered. But allowing this most marked distinction, and that the 170value of a commodity cannot be so well defined, and its variations so accurately measured, as the length of a commodity—where is the false analogy of endeavouring to measure these variations as well as we can? We cannot certainly describe the wealth of a merchant, nor measure the increase of his wealth during the last four years, with the same exactness as we can describe the height of a boy, and measure the amount of his growth during the same period. We can perform the latter operation with the most perfect precision by means of a foot-rule. The nature of wealth, and the best instruments used to measure its increase, are such, that the same precision is unattainable; but there is no false analogy involved in the process of measuring the wealth of a merchant at one time with his wealth four years before, by the number of pounds sterling which he possesses now, as compared with the number of pounds sterling he possessed at the former period. What false analogy is involved in applying money to measure the value of the advances of a manufacturer, as compared with the value of his 171returns, in order to estimate his profits? and what can the author mean by saying, that no relation of value can exist between commodities at different periods;[71] and that it is a case where money has no function to perform?
Notwithstanding such assertions, we see every day the most perfect conviction prevailing among all agriculturists, merchants, manufacturers, and shopkeepers, and among all writers on political economy, except the author, that to estimate the relation of commodities, at different periods, in regard to their general power of purchasing, and particularly the power of purchasing labour, the main instrument of production, is a most important function, which it is peculiarly desirable to have performed; and that, for moderately short periods, money does perform this function with very tolerable accuracy. And for this specific reason; that, for moderately short periods, a given quantity of money will represent, more nearly than any other commodity, the general power of purchasing, and particularly the power of setting 172labour in motion, so vital to the capitalist. It will approach, in short, more nearly than any other commodity, to that invariability which the author thinks so utterly useless in a measure of value, and the very mention of which seems to excite his indignation.[72]
It is, in fact, by means of this same steadiness of value in the precious metals, which they derive from their great durability, and the consequent uniformity of their supply in the market, that they are enabled to perform their most important functions. Hops, or corn, as before stated, will measure the relative values of commodities at the same time and place; but let the author or reader attempt to estimate the profits of a capitalist in hops or corn, by the excess of the value of his advances above the value of his returns so estimated, and he will soon be bewildered. If a very plentiful year of corn were to succeed to a comparatively scarce one, the farmer, estimating both his outgoings and incomings in the corn of each year, might appear to gain above fifty per cent., 173while, in reality, he might have lost, and might not be able, without trenching on his capital, to employ as many men on his farm as the year before. On the other hand, if a comparatively scarce year were to succeed to a plentiful one, his profits, estimated in corn, might appear to be less than nothing, and yet he might have been an unusual gainer, in reference to his general power of purchasing labour and other commodities, except corn. If the hop-planter were to estimate his advances and returns in hops, it is obvious that the results would be of the same kind, but aggravated in degree.
It must be allowed, then, that the commercial world have acted most wisely in selecting, for their practical measure of value, a commodity which is not only peculiarly convenient in its form, but is, in general, subject only to slow changes of value; and possesses, therefore, that steadiness in its power of purchasing labour and commodities, without which, all confidence in carrying on mercantile enterprises, of any duration, would be at an end.
174But though the precious metals are a very useful and excellent measure of value for those periods, within which almost all mercantile transactions are begun and completed; yet, as Adam Smith very justly observes, they are not so for very long periods; not because there is no function for them to perform, but because, in the course of four hundred years, they are found to lose that uniformity of value, which, in general, they retain so well during four years.
I can by no means, therefore, agree with the author, when he says, speaking of the precious metals, that, “in regard to measuring or comparing value, there is no operation that can be intelligibly described, or consistently imagined, but may be performed by the media of which we are in possession.”[73] Surely, to measure the relative power of a commodity over labour and the mass of other commodities, at different and distant times, is an operation which may be both consistently imagined, and intelligibly described; yet it is quite certain, that, in regard to distant 175periods, the precious metals will not perform this well. Would the author himself venture to say, that the general power of purchasing possessed by an ounce of silver in the time of Edward the Third, was not very much greater than the general power of purchasing possessed by an ounce of silver in the time of George the Fourth; or, that the same quantity of agricultural labour, at these two periods, would not much more nearly have represented the same general power of purchasing? The author seems equally unfortunate when he launches out in praise of the precious metals as a measure of value, as when he says that they do not perform this function better than corn.
It will be observed that, in speaking of the values of commodities, at different periods, as meaning their different powers of purchasing at those periods, the kind of value referred to is, exclusively, value in exchange. And, in reference to value in exchange, exclusively, it appears to be of the utmost importance to the language of political economy, 176to distinguish between the power of purchasing generally, and the power of purchasing any one commodity.
But it must not be imagined that when the estimation in which a commodity is held at different periods is referred to, as determined at the time by the state of the supply compared with the demand, and ordinarily by the natural and necessary conditions of its supply, or by the elementary costs of its production, which are equivalent expressions, that value in exchange is lost sight of. Yet the author is continually falling into this kind of misapprehension, and into a total forgetfulness of his first account of the meaning of value, in his examination of Mr. Ricardo’s views, as to the uses of a measure of value, in which, he says, a singular confusion of thought is to be discovered.[74]
“Suppose, he observes, that we had such a commodity as Mr. Ricardo requires for a standard: suppose, for instance, all commodities to be produced by labour alone, and silver to be produced by an invariable quantity 177of labour. In this case, silver would be, according to Mr. Ricardo, a perfect measure of value. But in what sense? What is the function performed? Silver, even if invariable in its producing labour, will tell us nothing of the value of other commodities. Their relations in value to silver, or their prices, must be ascertained in the usual way; and, when ascertained, we shall certainly know the values of commodities in relation to each other; but in all this, there is no assistance derived from the producing labour of silver being a constant quantity.”[75]
I have already described the function which silver would have to perform in this case, namely, either to measure the different powers of purchasing possessed by commodities at different periods, or to measure the different degrees of estimation in which they were held at these different periods.
Now, in the first place, with regard to the general power of purchasing, can it be denied for a moment, that, granting all the premises, as the author does hypothetically, silver, so 178produced, would be, beyond comparison, a better measure of the power of purchasing generally, than silver as it has been actually produced? It would be secured from that greatest source of variation in the general power of purchasing occasioned by the variation in its own producing labour; and an ounce of such silver would command much more nearly the same quantity of labour and commodities, for four or five hundred years together, than an ounce of silver derived from mines of greatly varying fertility.
Secondly, with regard to the estimation in which a commodity is held, it is not easy to conceive a more complete measure. If all commodities were produced by labour alone, and exchanged with each other according to the producing labour; and if silver were produced by an invariable quantity of labour, the quantity of silver given for a commodity in the market at different periods, would express almost accurately the relative estimation in which it was held at these periods; because it would express at once the relative sacrifice which people were willing to make, 179in order to obtain such a commodity at these different periods; the relative conditions of the supply, or elementary costs of production, of such commodity at these periods; and the proportion of the produce to the producer, or the relative state of the demand, as compared with the supply of such commodity at these different periods. And if the value of a commodity means, as the author has told us in the first sentence of his book, the esteem in which it is held, Mr. Ricardo’s measure would certainly do all which he proposed it should do; and this specifically on account of its invariability in relation to the estimation in which it was held.
It would not merely indicate, as the author states, in which of two commodities varying in relation to each other, at different periods, the variation had taken place;[76] but it would express the precise amount of the variation; that is, if it appeared by documents that the price of a yard of cloth of a certain quality four hundred years ago was twenty shillings, and its price at present was only ten shillings, 180it would follow, that the estimation in which it was held, or its value, had fallen one-half; because, as all commodities are, by the supposition, produced by labour alone, the sacrifice with which it could be obtained, the necessary conditions of its supply, or the elementary costs of its production, had diminished one-half.
The variations of a commodity, in relation to this kind of standard, would further show, with great exactness, the variations in its power of commanding all those commodities which had not altered in the conditions of their supply, or the elementary costs of production. If a commodity rose or fell in this standard price, at different periods, it would necessarily rise or fall exactly in the same proportion in its power of commanding, in exchange, all those commodities which had not altered in the conditions of their supply, or their elementary costs of production.
But still, it will be readily acknowledged, that, even granting all that the author has granted hypothetically to Mr. Ricardo, it is 181not true that such silver would be an accurate measure of the general power of purchasing. Although the circumstance of its invariability, in regard to its producing labour, would give it a prodigious superiority over all other commodities even in this respect, yet, as the producing labour of many commodities may vary in the progress of society, it is quite impossible that the same quantity of any one object can, through successive periods, represent the same general power of purchasing. This is universally allowed; and as it would be clearly desirable to have one rather than two definitions of value, the question is, whether, both on this account, and on account of the universal language and practice of society, for short periods, it would not be decidedly better to confine the term value of a commodity, when used generally, to the estimation in which it is held, determined by the state of the supply compared with the demand, and ordinarily by the elementary costs of production, rather than to its general power of purchasing. There is very nearly an accurate measure of 182the former; it is universally acknowledged that there cannot be an accurate measure of the latter; and further, it is most important to remark that, in adopting the former, our language would much more nearly coincide with the ordinary language of society in referring to variations of value, than if we adopted the latter.
As a matter of fact, when a rise in the value of hops or of corn is spoken of, who ever thinks about the changes which may have taken place in the values of iron, flax, or cabbages? For short periods, we consider money as nearly a correct measure of the values of commodities, as well as of their prices; and if hops and corn have risen in this measure, we do not hesitate to say that their values have risen, without the least reference to cloths, calicoes, or cambrics. This is a clear proof that, in general, when we speak of the variations in the values of commodities, we do not measure them by the variations in their general power of purchasing, but by some sort of standard which we think better represents the varying estimation 183in which they are held, determined at all times by the state of the supply compared with the demand, and, on an average, by the elementary costs of production.
The only variations in the general power of a commodity to purchase, which are susceptible of a distinct and definite measure, are those which arise from causes which affect the commodity itself, and not from the causes which affect the innumerable articles against which it is capable of being exchanged. In speaking, therefore, of the variations in the value of particular commodities, it is not only more accordant with the accustomed meaning attached to the expression, but absolutely necessary with a view to precision, to consider them as exclusively proportioned to, and measured by, the amount of the causes of value operating upon themselves.
Mr. Ricardo, therefore, quite consistently with his own hypothesis, considers a commodity, the producing labour of which has doubled, as having increased to double its former value. It has increased in relation to 184a standard which, according to him, is the sole cause of value; it will command just double the quantity of all those commodities which have not altered in their producing value; and if it will not command just double the quantity of other commodities, it is not because it will not command just double the value which it did before, but because, on account of the changes in the producing labour of the other commodities, double the quantity of them has become more or less than double the value.
On the same principle, Adam Smith considers the value of cattle as rising in the progress of cultivation and improvement, although the value of land, the value of wood, the value of poultry, &c., might rise still higher, and, consequently, a given quantity of cattle might, with regard to some commodities or sets of commodities, have its power of purchasing diminished. But in saying that the value of cattle rises in the progress of cultivation, he means to say, that it rises in relation to a standard, namely, the labour a commodity will command, which represents 185at different periods the state of the supply of cattle compared with the demand, and, on an average, the elementary costs of their production; and, consequently, much better represents the estimation in which they are held than any commodity or set of commodities. “Labour,” he observes, “it must always be remembered, and not any particular commodity, or set of commodities, is the real measure of the value both of silver and of all other commodities.”[77]
Even the author himself has a chapter on the causes of value; and here he finds it absolutely necessary to estimate the causes affecting one commodity as distinct from the causes affecting another; although, according to his previous doctrine, the value of one commodity might be just as powerfully affected by causes operating upon another commodity as by causes operating upon itself: If a and b be compared, the value of a will be equally doubled, whether the elementary cost of a be doubled or the elementary cost of b be diminished one half; and so no doubt 186it would, if the relation of a to b were alone considered. But what does this prove? not that the value of a is not very differently affected in the two cases, according to the most ordinary, the most useful, and the most correct acceptation of the term value; but that to confine the term value, as the author does, to the mere relation of any one commodity to any other, is to render it pre-eminently futile and useless.
In first separating value in exchange from value in use, it may be allowable to distinguish it by the title of the power of purchasing other goods, as Adam Smith has done, though never to interpret this power as the power of purchasing any one sort of goods, as the author has done. But the moment we come to inquire into the variations of the values of commodities at different periods, we must, with any view to precision and utility, draw a marked line of distinction between a variation in the power of purchasing derived from causes affecting the particular purchasing commodities, and the variations in the power of purchasing which 187may arise from causes operating upon the purchased commodities. We must confine our attention exclusively to the former; and for this purpose refer to some standard which will best enable us to estimate the variations in the elementary costs of production, and in the state of the demand and supply of these commodities, as the best criterion of their varying value, or the varying estimation in which they are held at different periods.
On these grounds, Mr. Ricardo, consistently with his peculiar theory, measures the varying values of commodities at different periods by their producing labour.
And Adam Smith, consistently with his more just and applicable theory, measures the values of commodities at different periods by the labour which they will command.
Among the author’s chapters is one (the seventh) entitled “On the Measure of Value proposed by Mr. Malthus.”
In order to prepare himself for the refutations intended, he sums up his principal doctrines respecting value; and as they are here brought into a small compass, I cannot resist 188the temptation of quoting them in his own words.
He says, “It has been shown that the value of labour, like that of any other exchangeable article, is denoted by the quantity of some other commodity for which a definite portion of it will exchange, and must rise or fall as that quantity becomes greater or smaller, these phrases being only different expressions of the same event. Hence, unless labour always exchanges for the same quantity of other things, its value cannot be invariable, and, consequently, the very supposition of its being, at one and the same time, invariable, and capable of measuring the variations of other commodities, involves a contradiction.”
“It has also been shown, that to term anything immutable in value, amidst the fluctuations of other things, implies that its value at one time may be compared with its value at another time, without reference to any other commodity, which is absurd, value denoting a relation between two things at the same time; and it has likewise been shown, 189that in no sense could an object of invariable value be of any peculiar service in the capacity of a measure.
“These considerations,” he says, “are quite sufficient to overturn the claims of the proposed measure, as maintained by its advocate.”[78]
I am most ready to acknowledge that they are amply sufficient for the purpose, if they are true. But is it possible that doctrines can be true, which, having no other foundation than a most arbitrary and unwarranted interpretation of a definition of Adam Smith, lead directly to the subjoined conclusions?
First; That the value of labour rises or falls as a given portion of it will exchange for a greater or less quantity of silk or any other commodity, however unconnected with the labourer’s wants; so that if silks were to fall to one-half their price, the value of labour would be doubled.
Secondly; That the value of corn in one year cannot be compared with the value of corn in another, because value denotes 190only a relation between two things at the same time.
And thirdly, That the comparative steadiness in the value of the precious metals, for short periods, is of no service to them in the capacity of a measure of value.
The decision of the question, as to the truth of doctrines necessarily leading to such conclusions, may be safely left to the reader. But to return to the main subject of the chapter, namely, the measure of value proposed by me.
In a publication entitled “The Measure of Value stated and illustrated,” I had given reasons, which appeared to me convincing, for adopting labour, in the sense in which it is generally understood and applied by Adam Smith, as the measure of value; and further to illustrate the subject, and bring into one view the results of different suppositions respecting the varying fertility of the soil and the varying quantity of corn paid to the labourer, I added a table in which different suppositions of this kind are made.
In reference to this table the author observes, 191that “In the same way any article might be proved to be of invariable value, for instance, ten yards of cloth. For whether we gave 5l. or 10l. for the ten yards, the sum given would always be equal in value to the cloth for which it was paid, or, in other words, of invariable value in relation to cloth. But that which is given for a thing of invariable value must itself be invariable, whence the ten yards of cloth must be of invariable value.”[79]
This comparison shows either a most singular want of discrimination, or a purposed disregard of the premises on which the table is founded. These premises are, that the natural and necessary conditions of the supply of the great mass of commodities, or, in other words, their elementary costs of production, are, the accumulated and immediate labour necessary to produce them, with the addition of the ordinary profits upon the whole advances for the time they have been advanced; and that the ordinary values of commodities at different periods, according 192to the most customary application of the term, are determined by the elementary costs of production at those periods, that is, by the labour and profits worked up in them.
If these premises be just, the table correctly illustrates all that it was intended to illustrate. If the premises be false, the whole falls to the ground.
Now, I would ask the author, what sort of resemblance there is between ten yards of cloth and ten days’ labour? Is cloth the universal and the main instrument of production? Is the advance of an adequate quantity of cloth the natural and necessary condition of the supply of all commodities? Has any one ever thought of calling cloth and profits the elementary costs of production? or has it ever been proposed to estimate the values of commodities at different periods by the different quantities of cloth and profits worked up in them?
If these questions cannot be answered in the affirmative, it is obvious that what may be true and important with regard to labour, may 193be perfectly false or futile in regard to any product of labour.[80] The whole depends upon the mode of estimating the values of commodities.
It would, no doubt, be an absurd tautological truism merely to state, that the varying wages of a given quantity of labour will always command the same quantity of labour; but if it were previously shown that the quantity of labour which a commodity commands represents exactly the quantity of labour 194worked up in it, with the profits upon the advances, and does therefore really represent and measure those natural and necessary conditions of the supply, those elementary costs of production which determine value; then the truism that the varying wages of a given quantity of labour always command the same quantity of labour, must necessarily involve the important truth, that the elementary costs of producing the varying wages of a given quantity of labour must always be the same.
It is obvious to any person inspecting the table, that the uniform numbers in the seventh column, illustrating the invariable value of the wages of a given number of men, might, with perfect certainty, have been stated without the intermediate steps; but if they had been so stated, no conclusion respecting the constancy of the value of such wages could have been drawn. The intermediate steps, which show that the value of the wages of ten men is there estimated by the causes which had been previously shown to determine the values of all commodities, can alone warrant 195the conclusion that the uniform numbers in the seventh column imply uniformity of value in the wages.
Mr. Ricardo had stated repeatedly, that the value of the wages of labour must necessarily rise in the progress of society. He builds, indeed, the whole foundation of his theory of profits on the rise and fall of the value of labour. The table shows that, if we estimate the value of wages by the labour worked up in them, that is, by one element of value, Mr. Ricardo is right, and the value of wages will really rise as poorer land is taken into cultivation; but that, if we estimate the value of wages by the labour and profits worked up in them, that is, by the two elementary ingredients of value, the value of wages will remain the same.
The author says that, from the remarks he has made, the reader will perceive that Mr. Malthus’s “Table illustrating the invariable value of labour,” absolutely proves nothing;[81] and he concludes his chapter with observing, that his “cursory review evinces that the 196formidable array of figures in the table yields not a single new or important truth.”[82]
I was not aware that it was ever expected from a tabular arrangement, that it should afford logical proofs of new propositions; but, if the author means that, taking the whole publication together, it contains nothing new or important, though I may be bound to believe it in relation to his own reading and his own views, I cannot help doubting it a little in regard to the reading and views of many others; and I am quite certain that, with regard to myself, the view I there took of the subject of value, and of the reasons for adopting labour as its measure, was, in many of its parts, quite new to me a year before the publication.
In the first place; I had nowhere seen it stated, that the ordinary quantity of labour which a commodity will command must represent and measure the quantity of labour worked up in it, with the addition of profits. But, as soon as my attention was strongly drawn to this truth, the labour which a commodity 197would ordinarily command appeared to me in a new light. I had before considered labour as the most general and the most important of all the objects given in exchange, and, therefore, by far the best measure of the general power of purchasing of any one object; but after I became aware that, by representing the labour worked up in a commodity, with the profits, it represented the natural and necessary conditions of its supply, or the elementary costs of its production, its importance, as a measure, appeared to me very greatly increased.
Secondly; I had nowhere seen it stated that, however the fertility of the soil might vary, the elementary costs of producing the wages of a given quantity of labour must always necessarily be the same. Colonel Torrens, in adverting to a measure of value, says, “In the first place, exchangeable value is determined by the cost of production; and there is no commodity, the cost of producing which is not liable to perpetual fluctuation. In the second place, even if a commodity could be found which always required the 198same expenditure for its production, it would not, therefore, be of invariable exchangeable value, so as to serve as a standard for measuring the value of other things. Exchangeable value is determined, not by the absolute, but by the relative, cost of production.”[83]
I had been convinced, however, that, with a view to superior accuracy and utility, and a more complete conformity to the language and practice of society, in estimating the varying values of commodities for short periods, it was necessary to separate the variations in the power of a commodity to purchase, into two parts; the first, derived from causes operating upon the commodity itself; the second, from causes operating upon other commodities; and, in speaking of the variations in the exchangeable value of a commodity, to refer only to the former. In this case it is obvious that, according to Colonel Torrens, we should possess a measure of value if we could find an object the cost of producing which was always the same.
199Now it is shown, in the “Measure of value stated and illustrated,” that the conditions of the supply of labour, or the elementary costs of producing the corn wages of a given number of men, estimated just in the same way as we should estimate the elementary costs of producing cloth, linens, hardware, or any other commodity, must of necessity always remain the same.
I own that these two necessary qualities of the labour, which commodities will ordinarily command, were practically new to me; and, when forced on my attention, and accompanied by the conviction above described, as to the most correct and useful definition of value, made me view labour as a measure of value, so far approaching towards accuracy, considering the nature of the subject, that it might fairly be called a standard.
The publication was also marked by another peculiarity, which I cannot but consider as of some importance: namely, the constant use of the term labour and profits, instead of the customary one, labour and capital.
It must be allowed that the expression 200labour and capital is essentially tautological. In every definition of capital I have met with, the means of commanding labour are included; and there can be no doubt that machinery and raw materials require labour for their production of the same general description, and usually in as large a proportion, as the labour advanced by the last capitalist. Speaking loosely, we may indeed use the expression labour and capital, meaning by capital, when so used, all that part of the general description of capital which does not consist of the means of commanding the immediate labour required. But when we are engaged in an inquiry into the elements of value, nothing can be more unphilosophical than to talk of labour and capital. Excluding rent and taxes, the only elements concerned in regulating the value of commodities are labour and profits, including, of course, in such labour, the labour worked up in the raw materials, and that portion of the machinery worn out in the production; and including in the profits, the profits of the producers of the raw materials and machinery. 201To say that the values of commodities are regulated or determined by the quantity of capital and labour necessary to produce them is essentially false. To say that the values of commodities are regulated by the quantities of labour and profits necessary to produce them is, I believe, essentially true. And if so, it was a point of some importance to substitute the expression labour and profits for the customary one of labour and capital.
I have been detained longer than I intended by the Critical Dissertation on the Nature, Measures, and Causes of Value. There is still matter of animadversion remaining; but were I to go on I should tire my readers, if I have not done it already.
The author, when not under the influence of his peculiar definitions, makes some very just observations; and the work is exceedingly well written; which makes it a matter of greater surprise that its main proposition should be so strikingly adverse to the principle of utility, and so peculiarly calculated 202to retard the progress of that science which it must have been intended to promote.
I do not think it necessary to the object I have in view, to proceed further with these remarks on the definition and use of terms among political economists. What I have already said, if just, will be sufficient to show that much uncertainty has arisen from our[84] negligence on this important point, and much improvement might be expected from greater attention to it. I shall now, therefore, proceed to define some of the principal terms in political economy, as nearly as I can, according to the rules laid down. But before I begin, I think it may be useful to give a summary of the reasons for adopting the subjoined definition of the measure of value.
As a preliminary, it may be proper to state, that it seems absolutely essential to the language of political economy, that the expression value of a commodity, like the expression price of a commodity, should have some fixed and determined sense attached to it. Every person who has either written or talked on the subject of political economy, has been constantly in the habit of using the term without specifically expressing the object of comparison intended: and if it were true, that we might with equal propriety suppose any one of a thousand different objects referred to, it might easily be shown, that all past writers who had used the term value had talked the greatest nonsense; and all future writers must abound in the most tedious circumlocutions and the most futile propositions.
204But the author of the Critical Dissertation on Value has certainly done injustice to the writers who have gone before him, in supposing that when they have used the term value of a commodity, no reference was implied, if it was not expressed. As I stated before, they must be considered as referring, in some form or other, either to its general power of purchasing, or, to the estimation in which it was held, determined by the state of its supply compared with the demand, and, on an average, by the elementary costs of production; and as it would be perfectly ridiculous to suppose, that when the values of commodities, at different periods, are spoken of generally, by respectable writers, they could mean to refer to individual commodities not intended to represent, more or less accurately, the above objects of reference; it is obvious, that the ultimate reference implied must be confined to one of these, or their equivalents.
I have already given my reasons for thinking it more correct and useful to refer to the estimation in which a commodity is held, 205determined as above described, rather than to its general power of purchasing; but, as others may be of a different opinion, it may be useful to include among the reasons for adopting labour as a measure of value, its qualities as a measure of the general power of purchasing.
Supposing, then, that the exchangeable value of a commodity were defined to be its general power of purchasing, this must refer to the power of purchasing the mass of commodities; but this mass is quite unmanageable, and the power of purchasing it can never be ascertained. With a view, therefore, to its practical application, it would unquestionably be our endeavour to fix upon some object, or set of objects, which would best represent an average of the general mass. Now, of any one object, it cannot for a moment be denied that labour best represents an average of the general mass of productions. There is no commodity considered by society as wealth, for which labour is not, in the first instance, exchanged; there are very few for which it is not exchanged in great quantities: 206and this can be said of no other object, except labour, and the circulating medium which represents it. It is, at once, the first, the universal, and the most important object given in exchange for all commodities; and if to this we add, that while there is one large class of commodities, such as raw products, which in the progress of society tends to rise as compared with labour, there is another large class of commodities, such as manufactured articles, which at the same time tends to fall; it may not be far from the truth to say, that the portion of the average mass of commodities which a given quantity of labour will command in the same country, during the course of some centuries, may not very essentially vary.
Allowing, however, that it would vary, and that labour is an imperfect measure of the general power of purchasing; yet, if some sort of standard more applicable than the mass of commodities be required, and labour appears to be beyond comparison the best representative of this mass, there will be a very powerful reason for adopting labour 207as the practical measure of value, even among those who may persevere in thinking that the best definition of value in exchange is the general power of purchasing.
To those, however, who hold the opinion that the variations in the exchangeable value of a commodity and the variations in its power of purchasing are not identical, and that a commodity increases in exchangeable value only when it will command a greater value in exchange, while its power of purchasing may increase merely because it will command a greater quantity of commodities which have confessedly fallen in value, the reasons for adopting labour as the measure of value will be found to increase tenfold in force.
There are various ways of describing value in the sense here understood; and the slightest examination of them will show that the labour which a commodity will command can alone be the measure of such value.
First; The author of the Critical Dissertation on Value has commenced his work by a description of it, in which I entirely agree with him. He says, as I have before stated, that “value, in its ultimate sense, appears to 208mean the esteem in which any object is held. But it is obvious that the degree of this esteem cannot be measured by comparing it with another commodity about which we know as little as of the first. The comparison with money would leave us as much in the dark as ever, if we did not previously know the estimation in which money was held.”[85] Even the mere relative values of two commodities cannot be inferred by putting them side by side, and looking at them for any length of time. Before we can attain even this partial conclusion, we must refer each of them to the desires of man, and the means of production; that is, we must make a previous comparison, in order to ascertain the value of each before we can venture to say what relation one bears to the other. It is this primary comparison which, independently 209of any secondary comparison, determines the estimation in which the commodity is held. And as this primary comparison can only be represented by the exchange with labour, it is certain that, if we define the value of a commodity to be the estimation in which it is held, the quantity of labour which it will command can alone measure this estimation.
Secondly: Locke, most justly looking to the foundation of all value, considers the value of commodities as determined by the proportion of their quantity to their vent, or of the supply to the demand; but the varying vent or demand for one commodity cannot possibly be represented by the varying quantity of another commodity for which it is exchanged, unless the second commodity remain steady in regard to labour. If at one time I give two pounds of hops for a yard of cloth, and at another time only one, it does not at all follow that the demand for cloth has diminished; on the contrary, it may be increased, and in giving the value of one pound of hops, I may have enabled the cloth manufacturer to set more men to work, and to 210obtain higher profits than when I gave the value of two pounds. But the demand for a commodity, though not proportioned to the quantity of any other commodity which the purchaser is willing and able to give for it, is really proportioned to the quantity of labour which he will give for it; and for this reason: the quantity of labour which a commodity will ordinarily command, represents exactly the effectual demand for it; because it represents exactly that of labour and profits united necessary to effect its supply;[86] while the actual quantity of labour which a commodity will command when it differs from the ordinary quantity, represents the excess or defect of demand arising from temporary causes. If then looking to the foundation of all value, namely; the limitation 211of the supply as compared with the wants of mankind, we consider the value of commodities at any time or place as proportioned to the state of their supply compared with the demand at that time and place, it is evident that the quantity of labour of the same time and place which any commodity, or parcels of commodities, will command, can alone represent and measure the state of the supply of them as compared with the demand,[87] and their values as founded on this relation.
Thirdly: It has often been stated that the value of a commodity is determined by the sacrifice which people are willing to make in order to obtain it; and this seems to be perfectly true. But the question recurs, how are we to measure this sacrifice? It is obvious that we cannot measure it by the quantity of another commodity which we are willing to give in exchange for it. When I give more calicoes, or more potatoes, than I did before, for a certain quantity of hardware, it does not 212at all follow that I make a greater sacrifice in order to obtain what I want. On the contrary, if calicoes and potatoes had both fallen in price, the one from improved machinery and the other from the abundance of the season, my sacrifice might even have been less rather than greater. Even the quantity of money which is given for a commodity is no measure of the sacrifice made to obtain it. Though it is an excellent measure of the variations in the sacrifice made, at the same time and place; yet without further information, it will tell us nothing either about the amount, or the variations at different places and times. The giving of an ounce of silver was a very different sacrifice in the time of Edward I. from what it is at present. It is obvious, therefore, that the sacrifice which we are willing to make, in order to obtain a particular commodity, is not proportioned to the quantity of any other commodity for which it will exchange, but to the difficulty with which such quantity, whether more or less, is attained. Now labour can measure this difficulty, but nothing else can. If, then, the 213value of a commodity be determined by the sacrifice which people are willing to make in order to obtain it, it is the labour given for a commodity, and labour alone, which can measure this sacrifice.
Fourthly: In the Measure of Value Stated and Illustrated, I considered the value of commodities as, on an average, determined by the natural and necessary conditions of their supply. These conditions I stated to be the accumulated and immediate labour worked up in commodities with the ordinary profits upon the whole advances for the time that they were advanced. And it appeared, both in the early part of the discussion, and in the Table, that the quantity of labour which a commodity would ordinarily command must represent and measure the quantity of labour worked up in it with the addition of profits. It was certainly a very remarkable fact, that when Mr. Ricardo chose the labour worked up in commodities “as, under many circumstances, an invariable standard,” and rejected the labour which they would ordinarily purchase as subject to as many fluctuations as 214the commodities compared with it,[88] he should not have seen that the labour which a commodity will ordinarily command, necessarily involves his own proposition, with that addition to it merely which can alone make it correct; and that it is precisely because the labour which a commodity will ordinarily command measures the labour actually worked up in it with the addition of profits, that it is justifiable to consider it as a measure of value. If then the ordinary value of a commodity be considered as determined by the natural and necessary conditions of its supply, it is certain that the labour which it will ordinarily command is alone the measure of these conditions.
Fifthly: The values of commodities are often said to be determined by the costs of production. When the costs of production do not refer to money, but to those simple elements of production, without an adequate quantity of which, whatever may be their price in money, the commodity cannot be produced, they are precisely the same as the 215natural and necessary conditions of the supply. The elementary costs of production, excluding rents and taxes, are the labour and profits required to produce a commodity. Of these it has been already shown, that the labour which the commodity will ordinarily command is alone the measure; and allowing that we could obtain with tolerable exactness the average price of common agricultural labour at different times and in different countries, and that when the prices of all other sorts of labour were once established, they would (as assumed by Adam Smith and Mr. Ricardo) continue to bear nearly the same relation to each other in the further progress of cultivation and improvement, it is certain that the quantity of common agricultural labour which a commodity would ordinarily command at any place and time would measure, with a near approach to accuracy, the elementary costs of production at that place and time; so that commodities, which at two different periods in the same country would ordinarily command the same quantity of agricultural labour, might fairly 216be said to be equal to each other in their elementary costs of production, and, of course, in their values, if their values be determined by their elementary costs of production.
Sixthly: It may be said that the value of a commodity must be proportioned to its supply compared with the number of its producers. This appears, indeed, to be strikingly the case in the early periods of society when many commodities are obtained, almost exclusively, by labour. If fruits are to be procured, or game killed or caught, by labour alone, or assisted only by capital of very little value, the quantity obtained, on an average, by a day’s labour must represent, with a great approach to accuracy, the degree of scarcity in which commodities exist compared with the producers of them working for a certain time. But the degree in which the supply of a commodity is limited, as compared with the numbers, powers, and wants of those who wish to obtain it, is the foundation of all value. Here the producers are both the effectual demanders and the consumers; and the produce obtained on an average by a single 217producer must represent the supply compared with the numbers, powers, and wants of the demanders. If a large quantity of produce be obtained by a producer, the commodity will be in abundance, and will be considered as of comparatively little value; if a small quantity be obtained by a producer, the commodity will be scarce, and will be considered as of comparatively great value. If it be the custom of the country for the producers to work only four hours a-day instead of ten or twelve, the commodities produced will bear a comparatively small proportion to the numbers of the producers and effectual demanders, and will consequently be of much higher value, than in those countries where it is the custom to work for the greater number of hours; and, on the other hand, if the producers, besides working ten or twelve hours a-day, are aided by ingenious instruments and great skill in the use of them, the commodities produced will be in unusual plenty compared with the producers, and will be considered as proportionally of low value. In all these cases the value of the commodity is 218evidently determined by the relation between its quantity and the number of its producers.
Now though, in the more advanced stages of society, the producer is not always at the same time the demander and consumer; yet the effectual demand for commodities must, on an average, be proportioned to the productive services set in motion to obtain them;[89] and when the different kinds of producers are reduced to a common denominator, such as common agricultural day-labour, and profits are deducted as the remuneration of the capitalist, and rent as the remuneration of the landowner, the proportion which the remaining produce bears to the number of such producers must represent, exactly in the same manner as in the early periods of society, the degree of scarcity in which the commodity exists compared with the producers; and therefore the value of the commodity is measured by the quantity of it which will command a day’s common labour. In fact, if it 219be once allowed that when labour is exclusively concerned, the number of days’ labour necessary to produce a commodity at any place and time represents the natural value of the commodity at that place and time[90], then, as it is quite certain that the value in exchange of any other commodity compared with the first, will be accurately in proportion to the respective quantities of the same kind of labour which they will command, it follows necessarily, that the value of the second commodity must always be in proportion to the quantity of labour it will command, however its value may have been affected by profits, rents, taxes, monopolies, or the accidental state of its supply compared with the demand.
Seventhly: It has been stated that the values of commodities must be proportioned to the causes of value operating upon them. The author of the Critical Dissertation has a chapter on the causes of value, and, at the conclusion of it, adverting to the variety of considerations operating upon the human 220mind, which he thinks have been overlooked by political economists, he observes, “these considerations are the causes of value; and the attempt to proportion the quantities in which commodities are exchanged for each other to the degree in which one of these considerations exists, must be vain and ineffectual. All, in reality, that can be accomplished on this subject, is to ascertain the various causes of value; and, when this is done, we may always infer, from an increase or diminution of any of them, an increase or diminution of the effect.”[91]
These remarks, it must be allowed, are justly applicable to those who propose to measure the values of commodities by the quantity of labour actually bestowed upon them; but in no degree to those who propose to measure them by the quantity of labour which they will command. We have already shown that the labour which commodities will command measures that paramount cause of value which includes every other; namely, the state of the supply as compared 221with the demand. Whatever may be the number and variety of considerations operating on the mind in the interchange of commodities, whether merely the common elementary costs of production, or whether these costs have been variously modified by taxes, by portions of rent, by monopolies strict or partial, and by temporary scarcity or abundance, the result of the whole must appear in the state of the supply compared with the demand; and in the case of an individual article, the supply of which may be considered as given, the demand, must be proportioned to the sacrifice which the purchasers are able and, under all the circumstances, willing to make in order to attain it.
But it has already been shown that it is the command of labour which the purchasers are able and willing to transfer to the sellers, and not any particular commodity, except in proportion as it will command labour, that can alone represent the sacrifice of the purchasers. The labour, therefore, which a commodity will command, or which the purchasers are willing to give for it, measures 222the result of all the causes of value acting upon it,—of all the various considerations operating upon the mind in the interchange of commodities.
Whether then we consider the value of a commodity at any place and time as expressed by the estimation in which it is held; whether we consider it as founded entirely on the state of the supply as compared with the demand; whether we consider it as determined by the sacrifice which people are willing to make in order to obtain it; or by the natural and necessary conditions of its supply; or by the elementary costs of its production; or by the number of its producers; or by the result of all the causes of value operating upon it, it is plain that the labour which it will ordinarily command in any place will measure its natural and ordinary value; and the labour which it will actually command will measure its market value.
It must always be recollected, however, that in any sense in which we can use the term value of a commodity, there must be a reference, either expressed or implied, to some 223place and time, in the same manner as when we use the term price of a commodity. We all well know that the price of the same kind of commodity of the same quality, weight, and dimensions, is very different in different places and at different times; and this must be equally true in regard to the value of a commodity. It follows that, from the very nature of the thing, the value of a commodity cannot be expressed or measured independently of place and time. It is this quality which so essentially distinguishes the value of a commodity from its length or weight; but it does not necessarily destroy its capability of being measured.
It is true, however, that a very general opinion has prevailed among political economists, even since the publication of Adam Smith’s work, that from the very nature of value, so essentially different from length or weight, it cannot admit of a regular and definite measure.[92] This opinion seems to 224me to have arisen principally from two causes.
First—a proper distinction has seldom been made between the definitions of wealth and value. Though the meanings of these two terms have by no means always been considered as the same, yet the characteristics of one of them have been continually allowed 225to mix themselves with the characteristics of the other. This appears even in Adam Smith himself. When he says, that a man is rich or poor according to the quantity of the necessaries, conveniencies, and luxuries of life which he can command, he gives a most correct definition of wealth; but when he afterwards says, that he is rich or poor according to the quantity of labour which he can command, he evidently confounds wealth with value. The former is a definition of wealth; and of this, or of the general power of purchasing, which too much resembles it, there is no measure. The latter is his own definition or expression of real value; and of this the very terms which he uses show that there is a measure. The measure is distinctly expressed in the terms.
The second principal cause which has prevented labour from being received, according to the language of Adam Smith, as “alone the ultimate and real standard by which the value of all commodities can at all times and places be estimated and compared,”[93] is, 226that in different periods, and in different countries, it is not really true, as stated by him, that the labourer in working “lays down the same portion of his ease, his liberty, and his happiness.”[94] There is the best reason to believe that the labourer in India, and in many other countries, neither exerts himself so much while he is working, nor works for so many hours a day as an English labourer. A day’s labour, therefore, is not invariable either in regard to intensity or time. But still it appears to me that, for the reasons before stated, that is, because the labour of each place and time measures at that place and time the estimation in which a commodity is held, the state of its supply compared with the demand, the elementary costs of its production, the natural and necessary conditions of the supply, the proportion of the produce to the producers, &c. it must be considered as measuring, with a fair approach towards accuracy, the values of commodities at these places and times, so as to answer the question,—what was the value 227of broad-cloth of a certain description in the time of Edward III. in England? or, what is the value of money at present in China? The nature of the measure, and the reason why the varying intensity of the labour and the different number of hours employed in the day, do not disqualify it from performing its functions, may perhaps be illustrated by the following comparison:—
Let us suppose that the heights of men in different countries were extremely different, varying from six feet to six inches, and that the trees, shrubs, houses, utensils, and every other product or article were all in proportion, and that the foot-rule in each country bore the same relation to the race of human beings which inhabited it as the English foot-rule does to Englishmen: then, though it is obvious that the length of ten feet in one nation might extend over a much larger portion of space than ten feet in another nation; yet the foot of each nation would measure with accuracy the relative estimation in which men and things were held in regard to height, length, breadth, &c. It would 228determine whether a man was tall or short in the estimation of his fellow-citizens; whether his shoulders were broad or narrow; whether his circumference was great or small; and not only whether Mr. Pike’s nose was longer than Mr. Chub’s, but whether it was not, in the accustomed language of the country, absolutely a long nose, although perhaps it might not extend to a quarter of an English inch. On the other hand, if, instead of referring to the measure of each country, we were to refer always to an English foot, though we should be able to ascertain the relative portions of space which all the men to whom we applied our measure occupied, we should make sad havoc with the estimates which they and their countrymen had formed of their own heights, and many certainly would be considered as very short who had before always been considered as very tall. Now it must be allowed that the value of a commodity, as it changes with place and time, and depends upon the wants and caprices of man and the means of satisfying them, resembles 229more the estimate of tall or short, broad or narrow, than a portion of space capable of being ascertained by a measure unchangeable by time and place.
When we speak of the value of silver in China, we cannot possibly mean the value of an ounce of Chinese silver brought to London, where, if it were pure, it would be precisely of the same value as an ounce of pure silver which had been in London from time immemorial. What alone we can correctly mean is, the estimation in which the ounce of silver is held in China, determined, at the time, by the state of the supply compared with the demand, and ordinarily by the quantity of Chinese labour and profits necessary to produce it; and if this be what we mean by the value of an ounce of silver in China, there can be no doubt that Chinese labour, and Chinese labour alone, can measure it. Even, however, if we mean the relation of an ounce of silver to all the commodities in China in succession, it would be impossible practically to form any approximation towards a just notion of the result, 230except by referring the silver to Chinese labour.
It might be allowed, perhaps, that labour would be a still more satisfactory measure of value, in all countries and at all times, if the physical force exerted in a day’s labour were always the same; and probably this is sometimes not far from being the case in a few countries as compared with each other, and more frequently in the same country at different periods. The English agricultural labourers in the time of Edward III., though probably less skilful, worked, I should conceive, for nearly the same number of hours, and exerted nearly the same physical force, as our labourers at present. Under such circumstances, and in the same country, agricultural labour seems to be a measure of value from century to century calculated to satisfy the scruples of the most fastidious. But even when it is acknowledged, that the labourer at different times and in different countries does not always lay down the same portion of his ease, his liberty, and happiness, the quality of labour, 231as a measure of value, is not essentially impaired; and it appears to me always true, that when commodities in different countries and at different times have been found to command the same quantity of the agricultural labour of each country and time, they may with propriety be said to have been held in the same estimation, and considered as of the same value.
We may now proceed to the definition of some of the most important terms in common use among political economists, particularly those which have been most controverted. Whenever it has been thought necessary either to deviate from the general rule of employing terms according to their ordinary meaning, or to determine between two meanings both of which have some authorities in their favour, I have always been guided in my choice by what appeared to me the superior practical utility of the meaning selected in explaining the causes of the wealth of nations.[95]
232The reader will be aware, from the manner in which I have treated the subject, and the discussions into which I have allowed myself to enter, that what I consider as the main obstacle to a more general agreement among political economists, is rather the differences of opinion which have prevailed as to the classes of objects which are to be separated from each other by appropriate names, than as to the names which these classes should receive. It seems indeed to be pretty generally and most properly agreed, that the principal names which have been so long in use should remain. It would certainly be an Herculean task to change them, nor would any change which could be adopted in the present state of things remove the real difficulties. It has been most justly observed by Bacon, that “to say, where notions cannot be fitly reconciled, that there wanteth a 233term or nomenclature for it, is but a shift of ignorance.” When some people think that every sort of gratification, whether arising from immaterial or material objects, from spiritual comfort or comfortable clothing, should be designated by the same appropriate term; while others think it of great use and importance that they should be distinguished, it is obvious that such different notions cannot be reconciled by a new nomenclature. The grand preliminary required is that the notions should be fitly reconciled; and till this is done, a change of names would be perfectly futile. Preserving therefore, generally, the old names, the great practical question is, what they are to include and what they are to exclude?
1. The material objects necessary, useful or agreeable to man, which have required some portion of human exertion to appropriate or produce.
2. The quality of being serviceable or beneficial to mankind. The utility of an object has generally been considered as proportioned to the necessity and real importance of these services and benefits.
All wealth is necessarily useful; but all that is useful is not necessarily wealth.
3. Has two meanings—value in use, and value in exchange.
4. Is synonymous with Utility. It rarely 235occurs in political economy, and is never implied by the word value when used alone.
5. The relation of one object to some other, or others in exchange, resulting from the estimation in which each is held. When no second object is specified, the value of a commodity naturally refers to the causes which determine this estimation, and the object which measures it.
Value is distinguished from wealth in that it is not confined to material objects, and is much more dependent upon scarcity and difficulty of production.
6. The creation of objects which constitute wealth.
7. The portion of wealth created by production.
8. Land, labour, and capital. The two original sources, are land and labour; but the aid which labour receives from capital is 236applied so very early, and is so very necessary in the production of wealth, that it may be considered as a third source.
9. The soil, mines, waters, and fisheries of the habitable globe. It is the main source of raw materials and food.
10. The exertions of human beings employed with a view to remuneration. If the term be applied to other exertions, they must be particularly specified.
11. The labour which is so directly productive of wealth as to be capable of estimation in the quantity or value of the products obtained.
12. All labour which is not directly productive of wealth. The terms productive and unproductive are always used by political economists in a restricted and technical sense exclusively applicable to the direct production or non-production of wealth.
13. The exertion of the human faculties and powers to accomplish some desirable end. No very marked line is drawn in common language, or by political economists, between industry and labour; but the term industry generally implies more superintendence and less bodily exertion than labour.
14. Accumulated wealth, either reserved by the consumer for his consumption, or kept, or employed with a view to profit.
15. That portion of the stock of a country which is kept or employed with a view to profit in the production and distribution of wealth.
16. That portion of stock employed with a view to profit which yields such profit while it remains in the possession of the owner.
17. That portion of stock employed with a 238view to profit which does not yield such profit till it is parted with.
18. That portion of stock or wealth which the possessor may annually consume without injury to his permanent resources. It consists of the rents of land, the wages of labour, and the profits of stock.
19. The employment of a portion of revenue as capital. Capital may therefore increase without an increase of stock or wealth.
20. In modern times, implies the accumulation of capital, as few people now lock up their money in a box.
21. That portion of the produce of land which remains to the owner after all the outgoings belonging to its cultivation are paid, including the ordinary profits of the capital employed.
22. The average rent of land as before defined, estimated in money.
23. That portion of the produce of land which is not actually consumed by the cultivators.
24. The remuneration paid to the labourer for his exertions.
25. The wages which the labourer receives in the current money of the country.
26. The necessaries, conveniences, and luxuries of life which the wages of the labourer enable him to command.
27. The ordinary wages paid to the labourer by the day, week, month, or year, according to the custom of the place where he is employed. They are generally estimated in money.
28. Has generally been understood to mean the average money-price of common day-labour, and is not therefore different from the rate of wages, except that it more specifically refers to money.
29. The whole earnings of the labourer in a given time, which may be much more or much less than the average rate of wages, or the price of common day-labour.
30. The price in money of a given quantity of human exertion of a given strength and character, which may be essentially different from the common price of day-labour, or the whole money-earnings of the labourer in a given time.
31. The labour worked up in the raw materials and tools applied to the production of other commodities.
32. When stock is employed as capital in 241the production and distribution of wealth, its profits consist of the difference between the value of the capital advanced, and the value of the commodity when sold or used.
33. The per centage proportion which the value of the profits upon any capital bears to the value of such capital.
34. The net profits of a capital in money separated from the risk and trouble of employing it.
35. That portion of the gross profits of capital, independent of monopoly, which remains after deducting the net profits, or the interest of money.
36. The profits which arise from the employment of capital where the competition is not free.
37. The advance of the quantity of accumulated and immediate labour necessary to their production, with such a per centage upon the whole of the advances for the time they have been employed as is equivalent to ordinary profits. If there be any other necessary conditions of the supply arising from monopolies of any description, or from taxes, they must be added.
38. An expression exactly equivalent to the conditions of the supply.
39. The quantity of labour for which the commodity will exchange, when it is in its natural and ordinary state.
40. The estimation in which it is held at that place and time, determined in all cases 243by the state of the supply compared with the demand, and ordinarily by the elementary costs of production which regulate that state.
41. The estimation in which it is held when it is in its natural and ordinary state, determined by the elementary costs of its production, or the conditions of its supply.
42. The quantity of labour which it will command or exchange for at that place and time.
43. The quantity of labour for which it will exchange at that place and time, when it is in its natural and ordinary state.
44. The quantity of money for which it 244exchanges at that place and time, the money referring to the precious metals.
45. The price in money which will pay the elementary costs of its production, or the money conditions of its supply.
46. The quantity offered, or ready to be immediately offered, for sale.
47. Has two distinct meanings: one, in regard to its extent, or the quantity of commodities purchased; and the other, in regard to its intensity, or the sacrifice which the demanders are able and willing to make in order to satisfy their wants.
48. The quantity of the commodity purchased, which generally increases with the increase of the supply, and diminishes with the diminution of it. It is often the greatest when commodities are selling below the costs of production.
49. The sacrifice which the demanders are able and willing to make in order to satisfy their wants. It is this species of demand alone which, compared with the supply, determines prices and values.
50. The quantity of a commodity wanted by those who are able and willing to pay the costs of its production.
51. The sacrifice which the demanders must make, in order to effectuate the continued supply of a commodity.
52. The quantity of labour for which the commodity will exchange, when in its natural and ordinary state.
53. The demand for a commodity is said 246to be in excess above the supply, when, either from the diminution of the supply, or the increase of the effectual demand, the quantity in the market is not sufficient to supply all the effectual demanders. In this case the intensity of the demand increases, and the commodity rises, in proportion to the competition of the demanders, and the sacrifice they are able and willing to make in order to satisfy their wants.
54. The supply of a commodity is said to be in excess above the demand, or there is a partial glut, when, either from the superabundance of supply, or the diminution of demand, the quantity in the market exceeds the quantity wanted by those who are able and willing to pay the elementary costs of production. It then falls below these costs in proportion to the eagerness of the sellers to sell; and the glut is trifling, or great, accordingly.
55. A glut is said to be general, when, either from superabundance of supply or diminution of demand, a considerable mass of commodities falls below the elementary costs of production.
56. A given demand, in regard to price, is a given quantity of money intended to be laid out in the purchase of certain commodities in a market; and a given demand, in regard to value, is the command of a given quantity of labour intended to be employed in the same way.
57. Prices and values vary as the demand directly and the supply inversely. When the demand is given, prices and values vary inversely as the supply; when the supply is given, directly as the demand.
58. The destruction wholly or in part of any portions of wealth.
59. The consumption or employment of wealth by the capitalist, with a view to future production.
60. The consumption of wealth, as revenue, with a view to the final purpose of all production—subsistence and enjoyment; but not with a view to profit.
Def. 1. The reader will be aware that, in almost all definitions, the same meaning may be conveyed in different language, and that it is the meaning rather than the mode of expressing it that should be the main object of our consideration. The essential question in the definition of wealth is, whether or not it should be confined to material objects, and the reader is already apprised of my reasons for thinking that it should. Even M. Say, who admits “les produits immatériels,” allows, as I have before stated (p. 93), that the multiplication of them “ne fait rien pour la richesse;” and M. Storch, in his able “Cours d’Economie Politique,” though he justly lays great stress on what he calls les biens internes, with a view to civilization and the indirect production of wealth, confines the term richesses to biens externes, or material objects; 250and according to this meaning treats of the Théorie de la Richesse Nationale, in the first, and far the largest, part of his work. Altogether, I can feel no doubt that some classification of this kind, or some separation of material from immaterial objects is, in the highest degree, useful in a definition of wealth.
The latter part of the definition is of minor importance. It is intended to exclude such material objects as air, light, rain, &c.—which, however necessary and useful to man, are seldom considered as wealth; and, perhaps, it is more objectionable to exclude them, by the introduction of the term exchangeable value into a definition of wealth, than in the mode which has been adopted. If the latter clause were not added, the only consequence would be, that, in comparing different countries together, such objects as air, light, &c., would be neglected as common quantities.
Def. 2. I have already alluded to the manner in which M. Say has applied the 251term Utility. His language cannot be considered as consistent, when he says that the price of an article is the measure of its utility, although it might be, according to his own expression, la chose la plus inutile.[96] It is much better for the science of political economy that the term should retain its natural and ordinary meaning. All wealth is no doubt useful, but there are so very many immaterial, and some material objects which are highly useful, and yet not wealth, that there can be no excuse for confounding them. M. Storch has not escaped the same kind of error.
Def. 5. Two articles are never exchanged with each other without a previous estimation being formed of the value of each, by a reference to the wants of mankind and the means of production. This general and most important relation to the means of production, and the labour which represents these means, seems to be quite forgotten by those who imagine that there is no relation implied 252when the value of a commodity is mentioned without specific reference to some other commodity.
M. Say, under the head Valeur des Choses, observes, “c’est la quantité d’autres choses évaluables qu’on peut obtenir en échange d’elle.”[97] This is a most vague and uncertain definition, and much less satisfactory than the general power of purchasing.
M. Storch says, that “la valeur des choses, c’est leur utilité relative;” but this certainly cannot be said unless we completely change the natural and ordinary meaning either of utility or value.
Neither M. Say nor M. Storch has sufficiently distinguished utility, wealth, and value.
Def. 6. The term creation is not here meant to apply to the creation of matter, but to the creation and production of the objects which have been defined to be wealth.
Defs. 11 and 12. If wealth be confined to material objects, it must be allowed to be 253peculiarly convenient and useful, in explaining the causes of the wealth of nations, to have some appropriate term for that species of labour which directly produces wealth; and as the principal founder of the science of political economy has used the terms productive labour in the restricted sense necessary for this special purpose, perhaps few objections would have been made to it, if it had not involved all other kinds of labour, however useful and important, under the apparently disparaging designation of unproductive. This is a consequence, no doubt, to be regretted: yet, when it has been repeatedly stated that the term unproductive, as applied by Adam Smith, in no degree impeaches the utility and importance of such labour, but merely implies that it does not directly produce gross wealth, the mere name ought not to decide against a classification for which it appears from experience that it is very difficult to find a satisfactory substitute.
In M. Storch’s “Considérations sur la Nature du Revenu National,” he does not appear to me to give a correct view of what 254Adam Smith means by productive labour.[98] The difficulty of classification above alluded to appears strikingly in this treatise. There is some plausibility in the system, and it is explained with ingenuity and ability; but I think that the adoption of it would destroy all precision in the science of political economy.
Defs. 19 and 20. I have never been able to understand how the accumulation of capital and the difference between saving and spending can be distinctly explained, if we call all labour equally productive.
Def. 23. It is this gross surplus of the land which furnishes the means of subsistence to the inhabitants of towns and cities. Besides the rents of land, which are powerfully effective in this respect, a large part of what, in the division of the produce of land, would fall to the shares of the farmers and labourers, is exchanged by them for other objects of convenience and gratification, thus giving the 255main necessaries of life to a great mass of persons not immediately connected with the soil. The proportion which this mass of persons may bear to the cultivators will depend upon the natural fertility of the soil, and the skill with which it has been improved, and continues to be worked.
Defs. 28 and 30. In a valuable publication on the Price of Corn, and Wages of Labour, by Sir Edward West, which has just fallen into my hands, he proposes that the price of labour should mean the sum paid for a given quantity of labour of a given character. I quite agree with him in thinking that it would be useful to have some appropriate term to express this meaning; but, as the price of labour has certainly not hitherto been used in this sense, and as it would be, in almost all cases, extremely difficult to give an answer to a question respecting the price of labour so understood, it would certainly be proper to vary the expression in some degree, in order to prepare people for a new meaning. In 256Definition 30, therefore, I have given this meaning to The price of effective labour.
Def. 31. It would save time and circumlocution, which is one of the great objects of appropriate terms, if, in speaking of the labour worked up in commodities, the labour worked up in the capital necessary to their production were designated by the term accumulated labour, as contradistinguished from the immediate labour employed by the last capitalist. We must always recollect, however, that labour is not the only element worked up in capital.
Def. 38. I have used the word elementary, in order to show that money-costs are not meant. On account of the doubt which may arise in this respect when the term costs of production is used alone, and the further doubt, whether ordinary profits are always included, I am decidedly of opinion that the conditions of the supply is a more expressive and less uncertain term for the same meaning. I do not find, however, that generally 257it is so well understood. I have defined, therefore, the costs of production with the addition of the word elementary, and including profits, as having precisely the same meaning as the conditions of the supply. I once thought it might be better not to include profits in costs of production; but as Adam Smith has included them, and more particularly as the profits worked up in the capital necessary to any production must form a part of the advances or costs in any sense in which the word costs can be used, I think it best, on the whole, to include necessary profits in the elementary costs of production. They are obviously included in the necessary conditions of the supply.
Defs. 39 and 40. In speaking, of the quantity of labour for which a commodity will exchange, as a measure either of the conditions of its supply or of its value, it must always be understood, that the different kinds of labour which may have been employed to produce it, must be reduced to labour of one description and of the lowest denomination, 258namely, common agricultural day-labour, estimated on an average throughout the year. This is the kind of labour which is always referred to when labour is spoken of as a measure.
Def. 57. It is not true, as stated by M. Say, that prices rise in the direct ratio of the quantity demanded, and the inverse ratio of the quantity supplied.[99] They only vary in this way, when the demand is understood to mean the sacrifice which the demanders are able and willing to make, in order to supply themselves with what they want; which may be represented in regard to price by the quantity of money ready to be employed in purchases in a market. When the demand for labour is spoken of, it can only relate to extent; and a greater demand can only signify a power of commanding a greater quantity of labour.
Def. 59. The only productive consumption, properly so called, is the consumption 259or destruction of wealth by capitalists with a view to reproduction. This is the only marked line of distinction which can be drawn between productive and unproductive consumption. The workman whom the capitalist employs certainly consumes that part of his wages which he does not save, as revenue, with a view to subsistence and enjoyment; and not as capital, with a view to production. He is a productive consumer to the person who employs him, and to the state, but not, strictly speaking, to himself. Consumption is the great purpose and end of all production. The consumption of wealth, as revenue, with a view to support and enjoyment, is even more necessary and important than the consumption of wealth as capital; but their effects are essentially different in regard to the direct production of wealth, and they ought therefore to be distinguished.
260I am far from meaning to present the foregoing definitions to the notice of the reader as in any degree complete; either in regard to extent, or correctness. In extent, they have been purposely limited, and in regard to correctness, I am too well aware of the difficulty of the subject to think that I have succeeded in making my definitions embrace all I wish, and exclude all I wish. I am strongly, indeed, disposed to believe, that in the sciences of morals, politics, and political economy, which will not admit of a change in the principal terms already in use, the full attainment of this object is impossible; yet a nearer approach to it is always something gained. I should not indeed have been justified in offering these definitions to the public, if I had not thought that they were, on the whole, less objectionable, and would be more useful in explaining the causes of the wealth of nations than any which I had seen. But I am conscious of some anomalies, and probably there are some more of which I am not conscious. Knowing, however, that 261the attempt to remove them might destroy useful classifications, I shall not consider a few individual cases, of little importance, as valid objections.
It is known that Adam Smith gave few regular definitions; but the meanings in which he used his terms may be collected from the context, and to these I have, in a considerable degree, adhered. For some I have been indebted to M. Say; others are my own; and in all, I have endeavoured to follow the rules for the definition and use of terms laid down at the beginning of this treatise. I shall consider my object as fully answered, if what I have done, should succeed in drawing that degree of attention to the subject which may lead to the production of something of the same kind, more correct and more useful, and so convincing as to be generally adopted.
1. It may seem strange to the reader, but it is nevertheless true, that the meanings of all these terms, which had been settled long ago, and in my opinion with a great approach towards correctness, by Adam Smith, have of late been called in question, and altered.
2. Wealth of Nations, b. ii. c. iii. p. i. vol. ii. 6th ed.
3. Traité d’ Economie Politique, liv. i. c. i. pp. 2, 4, 4th ed.
4. Wealth of Nations, b. i. c. v. p. 43. 6th edit.
5. Polit. Econ. c. xx. p. 320. 3rd Edit.
6. Polit. Econ. c. xx. p. 326. 3rd edit.—It may be remarked, by the way, that Mr. Ricardo here uses labour as a measure of value in the sense in which I think it ought always to be used, and not according to his own theory. He measures the exchangeable value of the plate and velvet, not by the quantity of labour worked up in them, but by the quantity they will command or employ.
7. Polit. Econ. c. i. sec. iii. pp. 16, 18, 3rd edit.
8. Polit. Econ. c. vii. p. 137, 3rd edit.
9. Id. p. 152.
10. Polit. Econ. c. v. p. 98, 3rd edit.
11. Polit. Econ. c. i. sec. vi. p. 45, 3rd edit.
12. Elements of Polit. Econ. c. ii. sec. iii. p. 75, 2nd edit.
13. Elements of Polit. Econ. c. iii. sec. ii. p. 92.
14. Elements of Polit. Econ. c. iii. sec. ii. p. 94.
15. Id. c. iii. sec. ii. p. 95.
16. Sec. viii. p. 188.
17. Elements of Polit. Econ. c. iv. s. iii. p. 225. If the demand of every individual were equal to his supply, in the correct sense of the expression, it would be a proof that he could always sell his commodity for the costs of production, including fair profits; and then even a partial glut would be impossible. The argument proves too much. It is very strange that Mr. Mill should not have seen what appears to be so very obvious,—that supply must always be proportioned to quantity, and demand to value.
18. Elem. of Polit. Econ. c. iv. s. iii. p. 233.
19. Elem. of Polit. Econ. c. iv. s. iii. p. 234.
20. Foreign trade is, no doubt, mainly a trade of barter; but the question whether British woollens find an adequate market in the United States, does not depend upon their purchasing the same quantity of tobacco as usual, but upon whether the tobacco, or whatever the returns may be, will purchase the British money or the British labour necessary to enable the woollen manufacturer to carry on his business successfully. If both woollen manufactures and tobacco are below the costs of production in money or labour, both parties may be carrying on a losing trade, at the time when the rate at which the two articles exchange with each other is the same as usual. This is the answer to the pamphlet, which M. Say addressed to me some years ago.
21. On the Production of Wealth, c. vi. s. vi. p. 349.
22. On the Production of Wealth, c. vi. s. vi. p. 349.
23. On the Production of Wealth, c. vi. s. vi. p. 345.
24. Id. p. 348.
25. It is quite astonishing that political economists of reputation should be inclined to resort to any kind of illustration, however clumsy and inapplicable, rather than refer to money. I suppose they are afraid of the imputation of thinking that wealth consists in money. But though it is certainly true that wealth does not consist in money, it is equally true that money is a most powerful agent in the distribution of wealth; and those who, in a country where all exchanges are practically effected by money, continue the attempt to explain the principles of demand and supply, and the variations of wages and profits, by referring chiefly to hats, shoes, corn, suits of clothing, &c., must of necessity fail.
26. Elem. of Polit. Econ. c. iv. s. iii. p. 234.
27. Elements of Polit. Econ. c. ii. sec. ii. p. 41.
28. Principles of Political Economy, part i. p. 5.
29. These remarks were principally directed against Lord Lauderdale’s definition of wealth—all that man desires as useful and delightful to him; but they apply with nearly equal force to Mr. Macculloch’s present definition, which is limited to those objects which possess exchangeable value. According to Mr. Macculloch’s own statement, health is purchased from the physician, and the gratification derived from acting from the actor; and it must be allowed that it is impossible to enjoy the benefits of civil and religious liberty without paying those who administer a good government. It has been said by Mr. Hallam, with some truth, that the liberties of England were chiefly obtained by successive purchases from the crown.
30. Principles of Polit. Econ., part iv. p. 406.
31. Principles of Polit. Econ., part iv. p. 410.
32. Principles of Polit. Econ., part ii. p. 71. This language has absolutely no meaning, if all labour be equally productive in regard to national wealth.
33. Mr. Macculloch dwells very much upon the extreme importance of accumulation to the increase of national wealth. But how are the gratifications afforded by menial servants to be accumulated?
34. Principles of Polit. Econ., part ii. p. 92.
35. Principles of Polit. Econ., part ii. p. 114.
36. This is very justly stated in Mr. Mill’s “Elements of Political Economy,” ch. iv. sec. i. p. 219, 2d edit.: both Mr. Ricardo and Mr. Mill, indeed, fully allow the distinction between productive and unproductive labour. M. Say, though he calls the labour of the menial servant productive, makes a distinction between the labour which is productive of material products and the labour which is productive of immaterial products. Of the latter products he says, “En favorisant leur multiplication, on ne fait rien pour la richesse, on ne fait que pour la consommation.”—Table Analytique, liv. i. ch. 13. This is a most characteristic difference; and though I prefer the classification of Adam Smith, as more simple, I should allow that, on these principles, the causes of the wealth of nations may be clearly explained. But I own myself utterly at a loss to conceive how they can be explained, if all labour be considered as equally productive.
37. Elem. of Polit. Econ. part ii. p. 93.
38. Princip. of Polit. Econ., part iv. p. 409.
39. Princip. of Polit. Econ., part iv. p. 411.
40. Principles of Polit. Econ., part iii., pp. 313, 317.
41. Principles of Polit. Econ., part iii. p. 313.
42. Principles of Polit. Econ., part iii. p. 313.
43. Principles of Polit. Econ. part ii. p. 69.
44. Wealth of Nations, b. i. c. vi.
45. It must always be recollected, that the advance of a certain number of days’ labour necessarily involves the wages paid for them, however these wages may vary in quantity. But the essential advance is the quantity of labour, not the quantity of money or corn.
46. Principles of Polit. Econ., part iii. p. 223. This is a most remarkable passage to come from Mr. Macculloch, who, though he agrees with Mr. Ricardo in words, has, in reality, deserted him, and agrees in substance with Adam Smith. According to the new meaning, which Mr. Macculloch has given to the term profits—the quantity of labour required to produce a commodity, is precisely equal to the quantity of labour for which it will ordinarily exchange, and certainly not equal to what Mr. Ricardo meant by the quantity of labour bestowed upon it.
47. Principles of Polit. Econ., part iii, s. 1. p. 221.
48. A person who uses a term in a particular sense practically defines it in that sense. Mr. Macculloch sometimes makes what have hitherto always been considered as profits mean labour; and sometimes makes labour, when used simply without any adjunct, mean fermentation, vegetation, or profits.
49. Macculloch’s Principles of Polit. Econ., part ii. p. 189.
50. Id. p. 190.
51. I own I want words to express the astonishment I feel at the proposal of such a remedy. A man, under the intoxication of what he conceives to be a new and important discovery, may be excused for occasionally making a rash statement; but that a proposal directly involving the discontinuance of the division of labour should, in a civilized country, be repeated over and over again by succeeding writers, and considered as an obvious resource in a sudden fall of profits, absolutely passes my comprehension. What a strange and most inapt illustration too, is it to talk about the possessors of broad cloths wanting to change them for silks! Who ever heard of a great producer of any commodity wishing to obtain an equivalent for it in some one other sort of completed commodity? If he is to produce what he wants, it must not be silks, but raw materials, tools, corn, meat, coats, hats, shoes and stockings, &c. &c.; and this is the obvious resource which is at hand in a glut!!!
52. Preface, p. 5.
53. Dissertation on Value, c. 1. p. 3.
54. Dissertation on Value, c. 1. p. 4.
55. Wealth of Nations, b. i. c. xi.
56. Production of Wealth, c. i. p. 49.
57. P. 242.
58. Dissertation on Value, c. i. p. 3.
59. C. ii. p. 39.
60. Dissertation on Value, c. iii. p. 58.
61. Dissertation on Value, c. xi. p. 194, 224. In the question between Colonel Torrens and Mr. Mill, “Whether the value of commodities depends upon capital as the final standard,” the author decides against Mr. Mill, but surely without reason. Mr. Mill cannot be wrong in thinking, that no progress whatever is made towards tracing the value of a commodity to its elements, by saying, that its value is determined by the value of the capital employed to produce it. The question still remains, how is the value of the capital determined? As to what the author says, p. 202, about the amount of capital, unless this amount be estimated in money, which quite alters the question, it is entirely inapplicable as a standard.
62. C. viii. p. 160.
63. C. vi. p. 135.
64. Dissertation on Value, c. ii. p. 58.
65. Id. p. 39.
66. P. 240.
67. Dissertation on Value, c. ii. p. 40.
68. Dissertation on Value, c. vi. p. 117.
69. Dissertation on Value, c. vi. p. 117.
70. Dissertation on Value, c. vi. p. 117.
71. Dissertation on Value, c. vi. p. 113, et seq.
72. Dissertation on Value, c. vi. p. 110.
73. Dissertation on Value, c. vi. p. 102.
74. Dissertation on Value, c. vi. p. 120.
75. Dissertation on Value, c. vi. p. 122.
76. Dissertation on Value, c. vi. p. 121.
77. Wealth of Nations, b. i. c. xi. p. 291, 6th edit.
78. Dissertation on Value, c. vii. p. 140.
79. Dissertation on Value, c. vi. p. 145.
80. It has always been a matter of great surprise to me that I should have been accused of arbitrarily adopting labour as the measure of value. If there be not a most marked and characteristic distinction between labour and any product of labour, I do not know where a characteristic distinction between two objects is to be found; and surely I have stated this distinction often enough, and brought forward the peculiar qualities of labour as my reasons for thinking that it may be taken as a measure of value. Opinions may differ as to the sufficiency of these reasons, or as to the degree of accuracy with which it will serve the purpose of a measure. But how it can be said that I have adopted it arbitrarily, is quite unintelligible to me. If I had merely stated, that I had adopted it because it was the main element in the natural costs of production, there could have been no ground for such a charge.
81. Dissertation on Value, c. vii. p. 148.
82. Dissertation on Value, c. vii. p. 150.
83. On the Production of Wealth, c. i. p. 56.
84. I am very ready to include myself among those political economists who have not been sufficiently attentive to this subject.
85. If in a foreign country, in which the relation of money to men and labour was unknown to us, we were told that a quarter of corn was selling for four ounces of silver, we should not know whether there was a famine, and corn was held in the highest estimation, or whether there was a glut of corn, and it was held in the lowest estimation. The very term estimation, as applied to commodities, must of necessity refer to man and labour.
86. It is a truth fruitful in important consequences, that the labour which commodities will command when in their natural state, by representing accurately the quantity of labour and profits necessary to produce them, must represent accurately the effectual demand for them. And this holds good at different places and times, referring of course to the labour of the same description at each place and time.
87. What could give us any information respecting the scarcity of a commodity in China, or the state of its supply as compared with the demand, but a reference to Chinese labour?
88. Principles of Polit. Econ., c. i. s. i. p. 5. 3d edit.
89. M. Say’s comprehensive expression, “Services productifs,” includes profits and rents as well as labour; but it is certain that labour will measure accurately the value of the whole amount of these services.
90. If this concession be once made, the whole question respecting labour as a measure of value is at once decided.
91. C. xi. p. 232.
92. I own that I was myself for a very long time of this opinion; but I am now perfectly convinced that I was wrong, and that Adam Smith was quite right in the prevailing view which he took of value, though he did not always strictly adhere to it. I am also convinced that it would be a great improvement to the language of political economy, if, whenever the term value, or value in exchange, is mentioned without specific reference, it should always be understood to mean value in exchange for labour,—the great instrument of production, and primary object given in exchange for every thing that is wealth; in the same manner as, when the price of a commodity is mentioned without specific reference, it is always understood to mean price in money—the universal medium of exchange, and practical measure of relative value. I am further convinced that the view of value here taken throws considerable light on the nature of demand and the means of expressing and measuring it, and that just view of value is absolutely necessary to a correct explanation of rents, profits, and wages. These convictions on my mind, which have acquired increase of strength the longer I have considered the subject, must be my apology to the reader for dwelling on it longer than, in considering it cursorily, he may think it deserves.
93. Wealth of Nations, b. i. c. v.
94. Wealth of Nations, b. i. c. v.
95. It is specifically on this ground that I think the meaning of the term Wealth should be confined to material objects; that productive labour should be confined to that labour alone which is directly productive of wealth; and that value, or value in exchange, when no specific object is referred to, should mean value in exchange for the means of production, of which labour, the great instrument of production, is alone the representative.
96. Traité d’Economie Politique, Epitome, vol. ii. p. 506, 4th edit.
97. Epitome, vol. ii. p. 507.
98. C. iv. p. 83.
99. Vol. ii. p. 17. 4th edition.