[99] Of course, we do not mean to question the immense light which history throws upon the nature of existing social forces.

[100] Theory of Political Economy, 4th ed., p. 34.

[101] Art. "Geld," in Handwörterbuch der Staatswissenschaften.

[102] Cf. Helfferich, Das Geld, Leipzig, 1903, for the same terminology, pp. 485-486.

[103] Exchange creates values. It does not necessarily create utilities. Wheat going from a famine-stricken part of India to a place where it will sell for higher prices does not gain in utility thereby.

[104] A possible exception to this general statement might be made for Professor H. J. Davenport, who would insist that his version of the utility theory is based on "relative marginal utility," rather than on marginal utility in Böhm-Bawerk's fashion. No critic has been more merciless than he in the criticism of the Austrian confusions of demand-curves with utility-curves, etc. But it is not clear to me that Professor Davenport has freed himself from the general doctrine that he criticises. I am not sure that he would accept Schumpeter's version of the Austrian theory as correct. It may be possible to read Schumpeter's doctrine into chapter 7 of Davenport's admirable Economics of Enterprise, but it is not clear that one could read it in the chapter! That individual price-offer depends on the marginal utilities of alternative goods, in comparison with the marginal utility of the good in question, Davenport does emphasize. But the complication that not merely the utilities of alternative goods, but also their prices, have to be taken into account, and that this involves circular reasoning when an effort is made to give a summary of the whole system of prices by means of individual utility calculations, he does not, so far as I can see, grapple with. He summarizes the thing on p. 104: "The steps, then, are from (1) utility to (2) marginal utility, thence to (3) the comparison of marginal utilities, and finally to (4) price-offer." He takes no account here of the complication that the third step is in large degree a comparison, not of marginal utilities proper, but rather, of "subjective values in exchange." Yet just in this lies a vital difficulty of utility theory, in so far as it attempts to explain causation. Moreover, Professor Davenport is seeking to explain the causal relation of utility to demand, the old Austrian problem. The explanation of demand is, indeed, the problem with which all theories of value must come to terms, if they are to be of any use. As we have seen, Schumpeter's schema has no bearing whatever on the explanation of demand, or on causation of any sort. Schumpeter's scheme leaves money out, and demand-curves run in money terms. Davenport's scheme assumes money—and "purchasing power." (Loc. cit., 91.) We have seen in the chapter on "Supply and Demand" that the notion of demand and supply involves money and a fixed absolute value of money. Professor Davenport is thus doubly assuming value, the thing to be explained! Laws of "relative marginal utility" developed on the assumption of money, and in abstraction from changes in the value of money, are not likely to be of service when the problem of the value of money itself is taken up. On pp. 95-96, Davenport comes closest to Schumpeter's doctrine, saying that "the total situation is directive of each individual in it," and that there are "mutual reactions," such that particular facts are both effects and causes, illustrated by the last person who jumps on a crowded raft—does he sink the others, or do they sink him? This recognizes the complexity of the problem, but it is not clear that it even purports to do more than that. What is called for is a definition of the essential elements in that "total situation," with precise statement as to what is assumed constant and what is allowed to vary, and an analysis of the "mutual reactions," with a starting point and a terminus ad quem,—an equilibrium in which "mutual reactions" cease to trouble with their endless circle! Schumpeter's schema, though meeting criticism on other scores, does meet this logical test, but Davenport's does not appear to do so.

It is interesting to note that Professor Alvin S. Johnson, in his review of the Economics of Enterprise, concludes that Professor Davenport, instead of meaning by "relative marginal utility" anything of the sort that Schumpeter has in mind in his equilibrium picture of all utilities to all individuals, really has an absolute value in mind. (Quarterly Journal of Economics, May, 1914, pp. 433-436.) There is much in Professor Davenport's book to justify this interpretation.

Professor Davenport's application of "utility" to the problem of the value of money will be found on pp. 267-275 of the Economics of Enterprise. The general discussion of money and credit in the Economics of Enterprise has been exceedingly illuminating to me, and my indebtedness to it will appear in the present book.

Much of what has been said of Davenport's "relative utility" theory may also be said of Wicksteed's. (Common Sense of Political Economy, London, 1910.) This is in many ways a remarkable book, characterized by excellencies of many different sorts. But it fails to present the utility theory in such a way as to avoid circular reasoning. Wicksteed sees the confusion of utility-curves with demand-curves, and protests vigorously and at length against it. (E. g., pp. 147-150.) He starts out by assuming money and a set of market prices. His earlier chapters are given to showing how the individual adjusts himself to the market, bringing his "marginal utilities" of various goods into harmony with the market prices. He recognizes that he has made these assumptions (pp. 130-131), and that he cannot use the results thus achieved as an explanation of the market prices. They are "our goal, not our starting point." But by pp. 161-162 he finds himself with the "suspicion" that nothing special or peculiar is to be found in the laws of "market or current prices—a phenomenon which it is obviously impossible to regard as ultimate, which demands explanation, and which we have not yet explained.... Much remains to be done, but we can already see that the preferences of each individual help to determine the terms or conditions under which the choice of other members of the community must be exercised. If you take the individuals of the community two and two it is clear that the marginal preferences of each determine the limits within which direct exchanges with the other can be entertained, and we must already have at least a presentiment that the collective scale is the register of the final and precise 'resultant' of all these mutually determining conditions and forces."

This seems to forecast Schumpeter's doctrine, but in the development which follows, we do not find it. The heart of his analysis of the causation of prices is in ch. vi, on "Markets." The "summary" which precedes that chapter again suggests Schumpeter's analysis—the notion of an all-embracing equilibrium. But when we get into the detailed analyses of the chapter we find nothing more than an exceedingly good account of the process by which supply and demand of particular goods, considered separately, become equated, through two-sided competition, and under conditions of monopoly. Instead of "relative marginal utilities," we see customers coming into the market with various money-prices in mind, and sellers trying out various money-prices—not marginal utilities, nor yet two or more marginal utilities in comparison with one another, but rather, money-prices, which, in the minds of the buyers may be supposed to represent "subjective values in exchange," based on both marginal utilities and objective prices of other things that enter into the budget, and which, in the minds of sellers, represent estimates of the prices which buyers may be induced to pay. Wicksteed does not transcend the circle. Finally, despite his caution to avoid the more glaring forms of the circle, and the confounding of demand-curves with utility-curves, and of utility with value, he does lapse into it in its completest form in expounding the Austrian doctrine of cost of production. "The only sense, then, in which cost of production can affect the value of one thing is the sense in which it is itself the value of another thing. Thus what has been variously termed utility, ophelemity, or desiredness, is the sole and ultimate determinant of all exchange values." (P. 391.) Here is the illicit leap from marginal demand price to marginal utility which all utility theorists make, sooner or later! It is true that costs in one place are reflections of demand elsewhere. But it is not true that costs in one place have any definite quantitative relation to utilities in another place!

When Wicksteed comes to discuss the value of money, he makes slight use of the notion of abstract ratios among relative utilities, and employs a concept which he has nowhere vindicated or explained: the value of money, as distinct from the reciprocal of the price-level, treating the value of money as something which can be directly influenced by sinister rumors affecting the credit of the Government, and which can be an independent cause affecting velocity of circulation, and the amount of trade done by means of money. Loc. cit., p. 623. See infra, our chapter on "Velocity of Circulation."

The only writers I know at first hand who have really thought the thing through, and avoided the circle in form, are Schumpeter and Irving Fisher. (Mathematical Investigations in the Theory of Value and Prices, Trans. Conn. Acad. of Arts and Sciences, 1892. See bibliographical note, supra, in this chapter.) I have given an exposition of Schumpeter, rather than Fisher, because the former has put the doctrine in non-mathematical form. In the text I have indicated the limitations of their doctrine. Fisher definitely avows the impossibility of applying the doctrine to the problem of the value of money. Purchasing Power of Money, p. 174. Schumpeter doesn't apply it to money, and when he tries to work out a utility doctrine of money, he lapses into the Austrian circle in a very obvious form. In later writings, Fisher also seems to forget the limitations imposed on utility theory in his earlier essay. In his Elementary Principles, ed. 1912, Fisher lists (pp. 408-409) a great multitude of factors that might affect the price of pig iron, and then says: "Back of these causes lie other causes, multiplying endlessly as we proceed backward. But if we trace back all these causes to their utmost limits, they will all resolve themselves into changes in the marginal desirability or undesirability of satisfactions and of efforts, respectively, at different points of time, and in the marginal rate of impatience as between any one year and the next." Here these marginal psychic magnitudes, which in the earlier essay appeared merely as surface phenomena, resultants of a total situation, proportional to prices, causes of nothing, merely symptoms of a completed equilibrium, are erected into atomic veræ causæ, the ultimate ultimates!

It is interesting to contrast this with a yet more recent statement by a philosopher who has undertaken a defence of the utility theory of economic value, Professor R. B. Perry, in the Quarterly Journal of Economics, for May, 1916. Considering the contentions of the present writer that many general social causes, in addition to the individual utilities concerned with consumption, are needed to explain changes in the values of goods, such as changes in fashion, mode, in general business confidence, in moral attitude toward different sorts of consumption, in the distribution of wealth, in taxes and other laws, Professor Perry says: "If the Austrian School has neglected this, then it needs to be corrected. But the essential contention of that school remains, so far as I can see, unaltered; in that these changes work through individuals and have their point of application in a more or less rational comparison of needs made by the individual buyer or seller. Whatever affects these individual schedules on a sufficiently large scale will affect prices. But to ignore the individual channels through which these forces pass, is elliptical." (Pp. 469-470. Italics mine.) Now I call attention to several points in the foregoing. First, I would contrast it with the doctrine quoted from Professor Fisher's Elementary Principles. Where Fisher puts the utilities far back in the realm of ultimate causation, making them the source from which spring all the proximate social causes which might affect the price of pig iron (such as "a trade war," "a change in fashion," a "change in incomes," "decreasing foresight," etc., loc. cit., p. 409), Professor Perry would make individual utility schedules the final focal point, toward which converge, and through which pass, all the causal forces, however richly explained by antecedent social factors, which affect prices. The utility theory of value means all things to all men!

But a second point with reference to Professor Perry's doctrine. It is perfectly true that all social activities are the work of individuals. Society is nothing apart from the individuals who make it up. To think of society and the individual as separate and antithetical is a fallacy which I have criticised in detail in Part III of Social Value. The social value theory does not mean that there are social forces which do not run through individual channels. This is not to accept the notion that individuals are really, in their psychical nature, isolated monads, however. There is a functional unity of individual minds, and no individual can be understood in abstraction from society. But this view is as old as Aristotle. I have not contended that prices can change apart from the mental activities of individual men, working upon one another. So far there may be no issue with Professor Perry.

But there is a big issue when he contends that all the causation is focussed in individual utility schedules, and in a more or less rational comparison of needs made by the individual buyer and seller. This is demonstrably erroneous. Let us assume, for example, that utility schedules of every individual New Yorker remain unchanged, but that, through a change in the law (the work of individual men, under the influence of their own individual emotions and ideas, of, say, ethical character), incomes in New York City are equalized. Hold rigidly to the assumption that there are no changes in utility schedules. Will there not be, none the less, a radical readjustment of prices? Will not the prices of Riverside palaces and steam yachts sink and the prices of things which the poor esteem rise? The utility-curves of the erstwhile rich, assumed to remain unchanged, no longer count for so much as before in the market. The rich cannot go so far down their curves in the consumption process as before. The poor, or those who had been poorest, now count for more in the market. They can lower their margins. In other words, the forces affecting the distribution of wealth, in so far as they are legal and moral in character, at least, may affect the price-situation, without altering utility schedules. Some social factors, as changes in mode and fashion, will work through the utility schedules, but others will not. One big variable affecting prices which need not, in idea, at least, affect utility schedules at all, and whose main influence is anyhow not directed through them, is the volume of business confidence. This factor we shall analyze in our discussion of credit, infra. Professor Perry thus escapes only part of the criticism which we have made (Social Value, pp. 45 and 56) of the Austrian theory: (1) that it abstracts the individual from his vital contacts with other individuals, and (2) that, within the individual mind thus abstracted, the Austrians make a further abstraction, taking as relevant only the interests concerned with consumption of economic goods, summed up in the utility schedules. The second criticism applies to Professor Perry as well. Men's total interests are not summed up in utility schedules, and do not affect prices exclusively via utility schedules.

It may be noticed, also, with reference to Professor Perry's discussion that he has misconstrued the Austrian theory in conceiving it as an analysis of an historical process, with a beginning and an end, instead of a static picture, in which preëxisting individual factors come into equilibrium. (Loc. cit., 475.) He seeks thus to avoid the Austrian circle, but as we have shown in the discussion of von Mises in the text, this way is not open to the Austrians.

Able and penetrating though Professor Perry's discussion is, on the psychological side, it fails, I think, to take adequate account of the complexities with which the economist and sociologist must deal.

In general, I find no version of the utility theory of value which is defensible, and, above all, no effort to apply it to the value of money which has met with success.

[105] Vide Taussig, Principles, I, 432.

[106] "Der Bankzins als Regulator der Waarenpreise," Conrad's Jahrbücher, 1897.

[107] Loc. cit., ch. 8.

[108] Cf. ch. on "Economic Value."

[109] Nicholson, J. S., Money and Monetary Problems, pp. 64-66; 71-73.

[110] Works, McCulloch ed. 1852, p. 213.

[111] Cf. the criticism of Nicholson by W. A. Scott, Money and Banking, 1903 ed., ch. 4.

[112] Cf. Mill, Principles, Bk. III, ch. xiii, par. 1. "Nothing more is needful to make a person accept anything as money, and even at any arbitrary value, than the persuasion that it will be taken from him on the same terms by others." It is not quite fair to identify Mill's doctrine with the circle stated above, however, since Mill couples it with a reference to convention, resting on the influence of government—a mention, without analysis, of some of the factors to be discussed shortly.

[113] Cf. Knies, Das Geld, I, p. 140.

[114] Cf. Social Value, ch. 2. Infra, our chapter on "The Functions of Money."

[115] Das Geld, Leipzig, 1903, p. 477.

[116] Laughlin, rejoinder to Clow, "The Quantity Theory and its Critics," in Jour. of Pol. Econ., 1902.

[117] Principles of Money, passim.

[118] Cf. Social Value, pp. 132-136, and supra, ch. on "Marginal Utility and Value of Money."

[119] Strictly speaking, there is no marginal utility, but only a "subjective value in exchange," for money of the sort here discussed. See supra, the chapter on "Marginal Utility."

[120] The psychological reactions of the people in times of stress and uncertainty toward different kinds of money cannot be predicted with any certainty, and there seems to be absolutely no definite or universal law governing the matter. The present writer collected a lot of newspaper clippings at the outbreak of the present World War. From these it appears that in both Paris and Berlin there was a very great distrust of bank-notes, and an insistence by retailers, restaurants, landladies, etc., on coin. But silver, which was not standard money, seems to have been accepted without question. When hoarding is referred to in these clippings, it is invariably gold that is mentioned. A similar hoarding of gold took place during the Balkan crisis at the time of the outbreak of the war between the Balkan Allies and Turkey. Professor E. E. Agger informs me, however, that he has found some evidence that bank-notes as well as gold were hoarded in Austria, at this time.

Sometimes we have a suspension of Gresham's law, and an acceptance of all kinds of money at varying ratios. The following clipping from the Boston Herald of March 17, 1914, illustrates this: "Douglas, Ariz., March 16.—Four kinds of money are now circulating in the Mexican territory controlled by the Constitutionalists. These are United States currency, the first issues of the Constitutionalist government and of Sonora state, and 'Villa money,' or that issued by Chihuahua at the instance of the rebel military commander. United States takes precedence. Merchants in Sonora, in order to protect themselves and at the same time observe the laws requiring acceptance of the rebel currency issues, have established a sliding scale of prices. This was discovered when five merchants were arrested at Cananea by Constitutionalist secret service men, who found that for American money they could buy goods for less than half the amount exacted when payment was offered in Mexican currency. The uncertainty of the rebel campaign against Torreon is reflected in the money market. To-day Constitutionalist sold for 22 and 28 cents American on the peso. Mexican federal currency commanded from 30 to 32 cents." In the experience of travellers who have discussed the matter with the writer, there was little of this flexibility of relation between paper money and coin in Berlin, or Paris at the outbreak of the present War. Where paper was refused, it was absolutely refused, and where it was accepted, it seems to have been accepted without discount. No doubt, a fuller investigation would reveal all manner of variation in the behavior of different people in different centres, and at the same centres, at the outbreak of the War.

[121] Money and Banking, 1903 ed., pp. 58-60; 101-104.

[122] Principles of Money, p. 530.

[123] Written in December, 1914.

[124] Cf. Clow, F. R., "The Quantity Theory and its Critics," Jour. of Pol. Econ., 1902, p. 602.

[125] Cf. Emery, Speculation, pp. 90-91.

[126] Cf. Böhm-Bawerk's criticisms of the "use" theory of interest. (Capital and Interest, passim.) Both use theories and productivity theories are probably suggested, in part, by peculiarities which money possesses in pre-eminent degree. See infra, the chapter on the "Functions of Money."

[127] A more precise analysis of all these points will be given in the chapter on "The Functions of Money."

[128] Cf. Professor Taussig's account of expansions and contractions of the silver currency in his Silver Situation, passim.

[129] For bibliography, see Am. Econ. Rev., Dec., 1914, pp. 838-839.

[130] New York, 1911. All references to this book in the present volume are to the 1913 edition, which contains some new matter.

[131] Standard of Value, London, 1912, p. 48, n.

[132] Papers and Proceedings, Supplement to March, 1913, number of American Econ. Review, p. 131.

[133] American Econ. Rev., Supplement to March, 1916, number, p. 138.

[134] Loc. cit., pp. 31-32.

[135] Loc. cit., pp. 175ff.

[136] "The Passiveness of Prices," infra.

[137] Particularly in view of the elaborate statistics, to be considered below, with which it is sought to make the equation realistic.

[138] Loc. cit., p. 16ff.

[139] Loc. cit. p. 25.

[140] Ibid., p. 26.

[141] Ibid., p. 27.

[142] Where it is not meaningless, as at various points in the theory of mechanics, the product is always of a different denomination from either factor.

[143] Vide our ch. on "Supply and Demand," supra, for a discussion of Mill's doctrine as to the "demand" for money.

[144] What is here said of Fisher's equation of exchange applies, for the most part, to all versions of it.

[145] Loc. cit., p. 298. Cf. our chapter, infra, on "Statistical Demonstrations of the Quantity Theory."

[146] Purchasing Power of Money, p. 290.

[147] The amplified equation is MV + M´V´ = PT, which takes account of bank-credit. This is explained, infra.

[148] Loc. cit., p. 487. I recur to this point in discussing the statistics of the "equation of exchange" in ch. 19.

[149] Infra, ch. on "Quantity Theory and World Prices."

[150] Loc. cit., p. 48.

[151] Loc. cit., p. 370. The same position is taken by Kemmerer, Money and Credit Instruments, pp. 68 et seq. Mill denies the validity of these distinctions. See Principles, Bk. III, ch. 12, Par. 8.

[152] The above was written before the discussion in the Annalist (Feb. 7, Feb. 21, March 6, March 13, March 20, 1916) in which the present writer urged that Professor Fisher had greatly exaggerated the volume of trade in the United States by taking banking transactions as representative of trade. In reply (see especially the number for Feb. 21, pp. 245 et seq.) Professor Fisher maintains that the overcounting to which I call attention is offset by undercounting, and considers offsetting book-credits, which actually dispense with the use of money and checks, an important element in the undercounting. I am unable to reconcile this position with the reasons given for excluding book-credits from the "equation of exchange." A detailed discussion of the points at issue appears in later chapters, particularly in the chapter on "Statistical Demonstrations of the Quantity Theory."

[153] Quarterly Journal of Economics, vols. 8 and 9; Political Economy, pp. 169-175; Money, chs. 3-8.

[154] In our analysis of bank-loans, infra, we shall find reason to hold that Walker, though false to the logic of the quantity theory, comes nearer to a tenable doctrine than do Kemmerer, Fisher, Andrew, and most other quantity theorists.

[155] Principles, Bk. III, chs. 11 and 12.

[156] Purchasing Power of Money.

[157] Loc. cit., pp. 50-51.

[158] Loc. cit., p. 280.

[159] A. W. Atwood, "Hoarded Gold," Saturday Evening Post, Dec. 12, 1914, p. 26.

[160] Cf. Kinley, D., The Use of Credit Instruments, Senate Document 399, 1910, pp. 192-194.

[161] Ibid., pp. 102-103. In the same volume, on p. 200, the figures are given incorrectly, as 70% checks and 30% cash. C. A. Phillips, Readings in Money and Banking, 1916, p. 151, repeats this erroneous statement.

[162] Cf. Sprague, Crises under the National Banking System, Nat. Monetary Commission Report, pp. 71-75; 200, 202.

[163] Cf. also p. 280 of Fisher's Purchasing Power of Money.

[164] Kemmerer (Money and Credit Instruments, p. 80) maintains that, "under perfectly static conditions," money in circulation and money in bank reserves will keep a fixed relation to one another. He offers no argument to support this view. Of course, "under perfectly static conditions," everything keeps in fixed relation to everything else. The volume of credit will keep a fixed relation to the number of laborers and to the supply of clocks. But this would hardly establish causal connections! Fisher multiplies "fixed relations" of various kinds, without, so far as very diligent search can tell, offering any argument to support them. Thus, we have on p. 105 the statement, "We have seen that normally the quantities of other currency are proportional to the quantity of primary money, which we are supposing to be gold." Where this thesis has been demonstrated, he does not indicate. In view of the fact that gold has been the one really flexible element in our money supply, the thesis is hardly credible. On pp. 146-147, facing this difficulty, Fisher says: "Since, however, almost all the money can be used as bank reserves, even national bank-notes being so used by state banks and trust companies, the proportionate relations between money in circulation, money in reserves, and bank-deposits will hold approximately true as the normal condition of affairs. The legal requirements as to reserves strengthen the tendency." Here is a very substantial growth in the doctrine, with only one new argument, namely, that concerning legal reserve requirements—which gives minimal ratios, not fixed ratios. In what way the fact that most kinds of money can serve as legal reserves gives reason for the doctrine of fixed proportions is not made clear. For Professor Fisher, however, it seems quite enough, for on p. 162, in the heart of his causal theory, he boldly announces: "There must be some relation between the amount of money in circulation, the amount of reserves, and the amount of deposits. Normally we have seen that the three remain in given ratios to each other." (Italics mine.) It is doubtless somewhat dangerous to make a confident negative statement concerning a book which has no index. But careful reading of all that has preceded this statement reveals no references to this topic except those quoted above. "We have seen" is not a legitimate premise when so important an issue is involved. In our discussion of reserves in the section on credit, as well as in the discussion of the volume of trade, it will appear that no "normal" or "static" relations of this kind are possible.

[165] "The price-level outside of New York City, for instance, affects the price-level in New York City only via changes in the money in New York City. Within New York City it is the money which influences the price-level, and not the price-level which influences the money. The price-level is effect and not cause." (Loc. cit., p. 172.)

[166] Loc. cit., p. 50.

[167] W. C. Mitchell, Business Cycles, p. 306.

[168] Ibid., p. 325.

[169] J. P. Norton, Statistical Studies in the New York Money Market, p. 71, and chart opposite p. 72.

[170] Ibid., chart facing p. 72.

[171] Cf. Mitchell, loc. cit., chart, p. 298, and text, p. 295. As the ratio of reserves to money in circulation was greater in 1911 than in 1894, and as the ratio of deposits to reserves was also higher, we have a still wider variation in the ratio of money in circulation to deposits—M:M´

[172] See the striking figures collected by A. P. Andrew for 1907. Quart. Jour. of Econ., Feb. 1908, p. 297.

[173] Infra, our discussions of the relations of volume of money and credit to volume of trade, and our discussion of credit in the constructive part of the book. The theory of money and credit must be a dynamic theory.

[174] Senate Document, No. 405, 1910. For the Bank of England, see p. 25; for the Crédit Lyonnais, pp. 224-226; for the Deutsche Bank, pp. 374-375.

[175] Statist, 1912, p. 577.

[176] "The Prospects of Money," British Economic Journal, Dec. 1914.

[177] Cf. Ashley, W. J., Gold and Prices, N. Y., 1912, pp. 21 et seq.

[178] Cf. von Mises, "The Foreign Exchange Policy of the Austro-Hungarian Bank," British Econ. Jour., 1909, vol. 19. Cf. Keynes, Indian Currency and Finance.

[179] Conant, Principles of Money and Banking, vol. II, p. 50. In 1899, the reserve of the Bank of Belgium consisted of 107 millions (francs) in specie, and 108 millions in foreign bills.

[180] Principles of Economics, vol. I, pp. 432 et seq.

[181] In the chapter on "Quantity Theory and International Gold Movements," infra.