THE PURCHASING POWER OF MONEY THE MACMILLAN COMPANY NEW YORK • BOSTON • CHICAGO SAN FRANCISCO MACMILLAN & CO., Limited LONDON • BOMBAY • CALCUTTA MELBOURNE THE MACMILLAN CO. OF CANADA, LTD. TORONTO THE PURCHASING POWER OF MONEY ITS DETERMINATION AND RELATION TO CREDIT INTEREST AND CRISES BY mVING FISHER morassoR of political economt in tale UNIVKBSITT ASSISTED BY HARRY G. BROWN nrSTKUCTOB IN POLITICAL ECONOMY IN TALB UNIVERSITT HEW AMU JIFVI3II' EOiTION 466 66 Nein gork THE MACMILLAN COMPANY 1920 AU rights reserved CiOPTSraHT, Iffll, By the MACMILLAN COMPAlTr. Set up and electrotyped. Published March, igit. Nodvaoli 9rni« J. S. Gashing Co. — Berwick dc Smith 0*. Norwood, Mass., U.S.A. 2-2-3 ^S5p dc-p ■ > Co THE MEMORY OF SIMON NEWCOMB 6BEAT SCIENTIST, INSPIRING FRIEND, PIONEER IN THE STUDY OF ** SOCIETARY circulation" PREFACE TO THE FIRST EDITION The purpose of this book is to set forth the principles determining the purchasing power of money and to apply those principles to the study of historical changes in that purchasing power, including in particular the recent change in " the cost of living," which has aroused world-wide discussion. If the principles here advocated are correct, the pur- chasing power of money — or its reciprocal, the level of prices — depends exclusively on five definite factors : (1) the volume of money in circulation ; (2) its velocity of circulation ; (3) the volume of bank deposits subject to check ; (4) its velocity ; and (5) the volume of trade. Each of these five magnitudes is extremely definite, and their relation to the purchasing power of money is defi- nitely expressed by an "equation of exchange." In my opinion, the branch of economics which treats of these five regulators of purchasing power ought to be recog- nized and ultimately will be recognized as an exact science, capable of precise formulation, demonstration, I and statistical verification. The main contentions of this book are at bottom simply a restatement and amplification of the old "quantity theory " of money. With certain corrections in the usual statements of that theory, it may still be called funda- mentally sound. What has long been needed is a candid reexamination and revision of that venerable theory rather than its repudiation. Yet in the voluminous literature on money, there seems to be very little that approaches accurate formulation and VIU PREFACE TO THE FIRST EDITION rigorous demonstration, — whether theoretical or statis- tical. In making this attempt at reconstruction, I have the satisfaction of finding myself for once a conservative rather than a radical in economic theory. It has seemed to me a scandal that academic economists have, through outside clamor, been led into disagreements over the fundamental propositions concerning money. This is due to the confusion in which the subject has been thrown by reason of the political controversies with which it has become entangled. As some one has said, it would seem that even the theorems of Euclid would be challenged and doubted if they should be appealed to by one political party as against another. At any rate, since the "quantity theory " has become the subject of political dispute, it has lost prestige and has even come to be regarded by many as an exploded fallacy. The attempts by pro- moters of unsound money to make an improper use of the quantity theory — as in the first Bryan campaign — led many sound money men to the utter repudiation of the quantity theory. The consequence has been that, espe- cially in America, the quantity theory needs to be rein- ' troduced into general knowledge. Besides aiming to set forth the principles affecting the purchasing power of money, this book aims to illustrate and verify those principles by historical facts and statis- tics. In particular, the recent rise in prices is examined in detail and traced to its several causes. The study of the principles and facts concerning the purchasing power of money is of far more than academic interest. Such questions affect the welfare of every in- habitant of the civilized world. At each turn of the tide of prices, millions of persons are benefited and other millions are injured. PREFACE TO THE FIRST EDITION IX For a hundred years the world has been suffering from periodic changes in the level of prices, producing alter- nate crises and depressions of trade. Only by knowledge, both of the principles and of the facts involved, can such fluctuations in future be prevented or mitigated, and only by such knowledge can the losses which they entail be avoided or reduced. It is not too much to say that the evils of a variable monetary standard are among the most serious economic evils with which civilization has to deal ; and the practical problem of finding a solution of the dif- ficulty is of international extent and importance. I have proposed, very tentatively, a remedy for the evils of mon- etary instability. But the time is not yet ripe for the acceptance of any working plan. What is at present most needed is a clear and general public understanding of principles and facts. Toward such an end this book aims to contribute : — 1. A reconstruction of the quantity theory. 2. A discussion of the best form of index number. 3. Some mechanical methods of representing visually the determination of the level of prices. 4. A practical method of estimating the velocity of circulation of money. 5. The ascertainment statistically of the bank deposits in the United States which are subject to check, as distinct from " individual deposits," as usually published. 6. An improved statistical evaluation of the volume of trade, as well as of the remaining elements in the equation of exchange. 7. A thorough statistical verification of the (recon- structed) quantity theory of money. As it is quite impossible to do justice to some of these subjects without the use of mathematics, these have been freely introduced, but have been relegated, so far as X PREFACE TO THE FIRST EDITION possible, to Appendices. This plan, which is in accord- ance with that previously adopted in The Nature of Cap- ital and Income and The Rate of Interest, leaves the text almost wholly nonmathematical. Most of the statistical results review and confirm the conclusions of Professor Kemmerer in his valuable Money and Credit Instruments in their Relation to General Prices, which appeared while the present book was in course of construction. I am greatly indebted to Professor Kemmerer for reading the entire manuscript and for much valuable criticism throughout. My thanks are due to Professor F. Y. Edgeworth of All Souls' College, Oxford, and to Professor A. W. Flux of Manchester for kindly looking through the manuscript of the Appendix on index numbers and for suggestions and criticisms. To Dr. A. Piatt Andrew, now Assistant Secretary of the Treasury, my thanks are due for his kindness, as Special Assistant to the National Monetary Commission, in putting the resources of that Commission at my dis- posal, and in working out, from the records of the office of the Comptroller of the Currency, the volume of deposits subject to check at various dates in the past. For cooper- ation in carrying out these same calculations, I am like- wise indebted to Mr. Lawrence O. Murray, Comptroller of the Currency. These valuable figures are the first of their kind. To Mr. Gilpin of the New York Clearing House, my thanks are due for his kindness in furnishing various figures asked for and cited specifically in the text. To Mr. Richard M. Hurd, President of the Lawyers Mortgage Co., I am indebted for reading parts of the manuscript and for valuable criticism. To Mr. John O. Perrin, President of the American National Bank of Indianapolis, I am i^idebted for statistics PREFACE TO THE FIRST EDITION XI of the "activity" of bank accounts in liis bank, and for similar figures I am indebted to the officers of the National New Haven Bank and the City Bank of New Haven. My thanks are due to the Economic Journal for permis- sion to use unaltered some parts of my article on "The Mechanics of Bimetallism," which first appeared in that journal in 1894. My thanks are due to the Journal of the Royal Statistical Society for similar permission with reference to my article on "A Practical Method for estimating the Velocity of Circulation of Money," which appeared in December, 1909. A number of my students have rendered valuable service in gathering and coordinating statistics. I would especially mention Mr. Seimin Inaoka, Mr. Morgan Porter, Mr. N. S. Fineberg, Mr. W. E. Lagerquist, now in- structor at Cornell Universit}'-, Messrs. G. S. and L. A. Dole, Dr. John Bauer, now assistant professor at Cornell University, Dr. John Kerr Towles, now instructor at the University of Illinois, Dr. A. S. Field, now instructor at Dartmouth College, Mr. A. G. Boesel, Mr. W. F. Hicker- nell, Mr. Yasuyiro Hayakawa, Mr. Chester A. Phillips, and Mr. R. N. Griswold. Mr. Griswold performed the lengthy calculations involved in ascertaining an index of the volume of trade. There are two persons to whom I am more indebted than to any others. These are my brother, Mr. Her- bert W. Fisher, and my colleague. Dr. Harry G. Brown. To my brother my thanks are due for a most searching criticism of the whole book from the standpoint of peda- gogical exposition, and to Mr. Brown for general criticism and suggestions as well as for detailed work throughout. In recognition of Mr. Brown's assistance, I have placed his name on the title-page. IRVING FISHER. Yale University, February, 1911. PREFACE TO THE SECOND EDITION The second edition is a reprint of the first with the fol- lowing changes : — 1. Correction of occasional misprints. 2. Addition of data for 1910, 1911, and 1912, in the tables on pages 301, 317, and the diagram between pages 306 and 307. 3. A change in Figure 1 (page 13) to make it conform to the facts for 1912. 4. Changes in the table on page 147 with accompany- ing text to make the data correspond to the facts for 1912. 5. The insertion of an addendum on pages 492-493, giving the revised figures for deposits subject to check as calculated by Professor Wesley Clair Mitchell. 6. An appendix to the second edition (page 494 ff.) on "standardizing the dollar." For corrections of misprints and various helpful criti- cisms of the first edition I am under great obligations to a number of friends and correspondents and particularly to Major W. E. McKechnie, of the Indian Medical Service, Etawah, United Provinces, India ; Professor Warren M. Persons, Colorado College, Colorado Springs, Colo. ; Mr. J. M. Keynes, Editor, Economic Journal, Kings College, Cambridge ; Carl Snyder, author, New York City ; James Bonar, Deputy Master of the Royal Mint, Ottawa, Canada ; Professor Allyn A. Young, Washington University, St. Louis, Mo.; Professor Stephen Bauer, Director, Interna- tional Office of Labor Legislation, Basle, Switzerland ; Professor Wesley Clair Mitchell, New York City ; Pro- fessor O. M. W. Sprague, Harvard University. PREFACE TO THE SECOND EDITION XUl I have endeavored to avoid disturbing the plates of the first edition more than was absolutely necessary. Other- wise I should have been glad to incorporate some changes to make use of some valuable but general criticisms. In particular I should have liked to modify somewhat the statement of the theory of crises in Chapter IV and in Chapter XI to make use of the helpful criticism of Miss Minnie Throop England, of the University of Nebraska, in The Quarterly Journal of Economics^ November, 1912 ; also to meet a criticism of Mr. Keynes' to the effect that, while my book shows that the changes in the quantity of money do affect the price level, it does not show how they do so. To those who feel the need of a more defi- nite picture of how the price level is affected by a change in the quantity of money I refer the reader to my Ele- mentary Principles of Economics^ pages 242-247, and to other writers on this subject, particularly Cairns. IRVING FISHER. SUGGESTIONS TO READERS 1. The general reader will be chiefly interested in Chapters I-VIII. 2. The cursory reader will find the gist of the book in Chapter II. 3. Objectors to the quantity theory will find their theo- retical and statistical objections discussed in Chapters VIII and XII respectively. 4. Studeyits of financial history should read Chapter XII. 5. Currency reformers should read Chapter XIII. 6. The appendices are addressed mainly (though not exclusively) to mathematical economists^ for whom the chief interest will probably lie with the Appendix to Chapter X, on Index Numbers, (which should be read as a whole,) and § 6 of the Appendix to Chapter XII, on the Method of Determining Velocity of Circulation. 7. The remainder of the Appendix to Chapter XII is supplied chiefly in order that statistical critics may be en- abled to verify the processes described in the text. 8. Chapter X and its Appendix are of chief interest to students of index numbers^ a subject as fascinating to some as it is dr}'- to others. 9. The analytical table of contents, the index, and the running page headings have been constructed with espe- cial reference to the varying needs of different classes of readers. The book is, however, designed to constitute a complete whole, and it is hoped that as many as possible of those who approach it from special viewpoints may, in the end, read it aU. SUMMARY OF CONTENTS OHAPTBB PAGB I. Primary Definitions 1 » II. Purchasing Power of Monet as belated to the "Equa- tion OF Exchange " 8 ^ in. Influence of Deposit Currenct on the Equation and therefore on Purchasing Power 33 ~. IV. Disturbance of Equation and of Purchasing Power DURING Transition Periods 55 -«. v. Indirect Influences on Purchasing Power . . .74" VI. Indirect Influences (Continued) 90 Vn. Influence of Monetary Systems on Purchasing Power 112 VIII. Influence of Quantity of Money and Other Factors ON Purchasing Power and on Each Other . . 149 IX. The Dispersion of Prices makes necessary an Index OF Purchasing Power 184 X. The Best Index Numbers of Purchasing Power . . 198 XL Statistical Verification. General Historical Review 234 - Xll. Statistical Verification. Recent Years . , . 276 '' Xlll. The Problem of making Purchasing Power more Stable 319 y Appendix to Chapter II 349 Appendix to Chapter III . . ...... 367 Appendix to Chapter V ........ 370 Appendix to Chapter VI. .••••.. 372 Appendix to Chapter VII .••.•.«. 376 Appendix to Chapter VIII . ..•••.. 379 Appendix to Chapter X .••••••. 885 Appendix to Chapter XII ..••■••. 430 XV ANALYTICAL TABLE OF CONTENTS CHAPTER I FrIMART DEFmiTIOMS PAOK § 1. Wealth and exchange • . 1 § 2. Exchangeable goods 4 § 3. Circulation of money against goods 6 CHAPTER II Pdrchasing Power of Monet as related to thb Equation OF Exchange § 1. The various circulating media .... § 2. The equation of exchange arithmetically expressed § 3. The equation of exchange mechanically expressed § 4. The equation of exchange algebraically expressed § 5. Couclusiou and illustrations .... 8 14 21 24 28 CHAPTER m Influence of Deposit Currency on the Equation AND therefore ON PURCHASING POWEB § 1. The mystery of circulating credit 33 § 2. The basis of circulating credit 40 §3. Banking limitations . . . . . . . , .42 § 4, The revised equation of exchange . . . ... 47 § 5. Deposit currency normally proportioned to money ... 49 § 6. Summary 63 CHAPTER IV Disturbance of Equation and of Purchasing Power during Transition Periods § 1. Tardiness of interest adjustment to price movements . . 55 § 2. How a rise of prices generates a further rise . . . . 58 § 3. Extent of disturbances in equation . . .... 61 § 4. How a rise of prices culminates in a crisis .... 64 XVIU ANALYTICAL TABLE OP CONTENTS PAOB § 5. Completion of the credit cycle 67 § 6. Summary 72 CHAPTER V Indirect Influences on Purchasing Power § 1. Influence of conditions of production and consumption on trade and therefore on prices . . ... 74 § 2. Influence of conditions connecting producers and consumers on trade and therefore on prices 77 § 3. Influence of individual habits on velocities of circulation and therefore on prices 79 § 4. Influence of systems of payment on velocities of circulation and therefore on prices . . . .... 83 § 5. Influence of general causes on velocities of circulation and therefore on prices ........ 87 § 6. Influences on the volume of deposits subject to check and therefore on prices 88 CHAPTER VI Indirect Influences (Continued) § 1. Influence of foreign trade on the quantity of money and there- fore on prices 90 § 2. Influence of melting and minting on the quantity of money and therefore on prices 96 § 3. Influence of the production and consumption of money metals on the quantity of money and therefore on prices . . 99 § 4. Mechanical illustration of these influences .... 104 CHAPTER Vn Influence of Monetary Systems on Purchasing Powbb § 1. Gresham's Law 112 § 2. Cases when bimetallism fails immediately .... 116 § 3. Cases when bimetallism fails after production overtakes con- sumption 121 § 4. The limping standard ; the gold-exchange standard . . 127 § 5. Bimetallism in France 132 ANALYTICAL TABLE OF CONTENTS XIX PAGE § 6. Lessons of French experiment 135 § 7. The limping standard in India 138 § 8. The limping standard in the United States .... 140 § 9. General description of system in the United States . . . 143 CHAPTER VIII IwPLtTENCE OF Quantity of Money and Other Factors on Purchasing Power and on Each Other § 1. The equation of exchange implies no causal sequence . . 149 § 2, Effects of a change in money (M). Quantity theory in causal sense 151 § 3. Quantity theory not strictly true during transition periods . 159 § 4. Effects of a change in deposits (M') relatively to money (M) 162 § 5. Effects of changes in velocities of circulation ( V and F') . 164 § 6. Effects of changes in volume of trade (the §'s) . . . 165 § 7. Can the price level be regarded as cause as well as effect ? . 169 § 8. Distinction between causation of individual prices and the price level 174 § 9. Summary 181 CHAPTEE IX The Dispersion of Prices makes necessary an Index OP Purchasing Power § 1. Some prices cannot respond readily to price movements . . 184 § 2. Consequently other prices must over-respond .... 190 § 3. Transformation of the right side of the equation of exchange iTom'EpQ to PT 194 § 4. Summary 196 CHAPTER X The Best Index Numbers op Purchasino Power § 1. Forms of index numbers 198 § 2. Various purposes of index numbers 204 § 3. An index number as a standard of deferred payments . . 208 § 4. Deferred payments based on total exchanges .... 217 § 5. Practical restrictions 225 § 6. Summary 231 XX ANALYTICAL TABLE OF CONTENTS CHAPTER XI Statistical Verification. General Histobicai. Review §1- §2. §3. §4. §5. §6. §7. §8. §9. §10. §11. § 12. § 13. § 14. §15. §16. §17. §18. §1- §2. §3. §4. §5. §6. §7. §8. §9. §10, §11. The last thousand years . The last four centuries The nineteenth century . Its five price movements . Retrospect . . . , Outlook Paper money . . . . Paper money in France . Paper money in England'. Paper money in Austria . Early American paper money . The " greenbacks " . Confidence in the greenback . Confederate paper money Deposit currency and crises Particular crises Velocity of deposits and crises Summary CHAPTER Xn Statistical Verification. Recent Yeaes Professor Kemmerer's statistics, 1879-1908 New estimates for M and M', 1890-1909 New estimates for M'V and T"', 1896-1909 New estimates for 3IV and T^ 1896-1909 Estimates for T and P, 1896-1909 , P directly and indirectly calculated Correcting discrepancies .... The final results ..... The comparative importance of price-raising causes Influence of antecedent causes such as tariffs, etc. Results and by-products of Chapter XII . PAQX 234 237 238 240 246 248 250 252 253 255 256 258 261 263 265 267 270 274 276 280 282 285 290 292 298 304 307 311 315 CHAPTER XIII The Problem of making Purchasing Power more Stable § 1. The problem of monetary reform 319 § 2. Bimetallism as a solution 323 § 3. Other proposed solutions 328 § 4. The tabular standard 332 § 5. The writer's proposal ........ 337 § 6. Summary and conclusion 348 APPENDICES Appendix to Chapter II FAec § 1 (to Ch. II, § 3). The concept of an average .... 349 § 2 (to Ch. 11, § 6). The concept of velocity of circulation . . 362 § 3 (to Ch. II, § 6). "Arrays" ofp's, Q's, a,ndpQ's . . .355 § 4 (to Ch. II, § 5). " Arrays " of e's, m's, and F's . . .358 § 5 (to Ch. II, § 5). The coin-transfer concept of velocity and the concept of time of turnovei 362 § 6 (to Ch. II, § 6). Algebraic demonstration of equation of ex- change 364 § 7 (to Ch. II, § 6). P must be a specific form of average in order to vary directly as M and V and inversely as the §'s, . . 364 Appendix to Chapter III § 1 (to Ch. in, § 2). " Arrays " of A;'s and r's . . . .367 § 2 (to Ch. in, § 4). Algebraic demonstration of equation of ex- change including deposit currency 368 Appendix to Chapter V § 1 (to Ch. V, § 5). Effect of time credit on equation of exchange 370 Appendix to Chapter VI § 1 (to Ch. VI, § 1) . Modification of equation of exchange required by international trade 372 Appendix to Chapter Vn § 1 (to Ch. VII, § 2) . Money substitutes unlike other substitutes . 376 § 2 (to Ch. VII, § 2). Limits for ratios within which bimetallism is possible 378 Appendix to Chapter VHI § 1 (to Ch. VIII, § 6). Statistics of turnover at Yale University . 379 § 2 (to Ch. VIII, § 8). Four types of commodities contrasted . 382 xxi XXll APPENDICES Appendix to Chapter X § 1. Each form of index number for prices implies a correlative index number for quantities 385 § 2. Index numbers for prices occur in arithmetical pairs as also do index numbers for quantities .... § 3. General meanings of p's and §'s § 4. Review of 44 formulae, heading table columns § 5. Review of 8 tests, heading table rows § 6. The interior of the table ; column 11 in particular § 7. The 44 formulae compared .... § 8. Reasons for preferring the median practically § 9. Summary 390 392 393 400 408 418 425 428 Appendix to Chapter XII § 1 (to Ch. Xn, § 1). Professor Kemmerer's calculations . . 430 § 2 (to Ch. XII, § 2) . Method of calculating M . . . .432 § 3 (to Ch. XII, § 2). Method of calculating J/' . . . .434 § 4 (to Ch. XII, § 3). Method of calculating M'V for 1896 and 1909 441 § 5 (to Ch. XII, § 3). Method of calculating MV for 1897-1908 446 § 6 (to Ch. XII, § 4). General practical formula for calculating V 448 § 7 (to Ch. XII, § 4). Application of formula to calculations of F for 1896 and 1909 460 § 8 (to Ch. XII, § 4). Interpolating Values of Ffor 1897-1908 . 477 § 9 (to Ch. XII, § 5). Method of calculating T . . . .478 § 10 (to Ch. xn, § 5). Method of calculating P . . . .486 § 11 (to Ch. XII, § 7). Mutual adjustments of calculated values oi M, M, V, W, P, T 488 § 12 (to Ch. XII, § 8). Credit and cash transactions. Comparison with Kinley's estimates 491 § 13 (to whole Ch.). Addendum to second edition . . . 492 Appendix on " Standardizing the Dollar." 494 ADDENDUM (In this revised edition figures for 1910-1912 are given on pages 304, 317. See also page 492.) Data have just become available by which to bring down through 1910 the statistics of Chapter XII. The results are as follows : Magnitudes in the Equation of Exchange for 1910* M M' V V P T MV+M'V PT As first calculated 1.64 7.24 21 52.8 103.7 397 416 412 As finally adjusted 1.64 7.23 21 52.7 104.0 399 415 415 The table shows that the figures, as first calculated, conform admirably to the equation of exchange. The adjustment needed, to produce perfect conformity, in only one case reaches the half of one per cent ! From the adjusted figures we may calculate the per- centages of cash and check transactions (^MV-i- MV ■\- M' V and M'V -i-MV^M'V'~). These are 8% and 92%, which may be added to the table on page 317. The ratio of deposits to money (^M'/M^ is 4.4, which shows * The above figures may be inserted by the reader in the tables on pages 280, 281, 284, 285, 290, 292, 293, 304. The methods of deriving the figures are in general the same as those explained in the Appendix to Chapter XII. The antecedent figures on which the above table depends may be inserted by the reader as foUovys : M. On page 432 add (to the bottom of columns 1-8 incl. ) in the table the following : 1910, 3.42, 3.42, .32, 1.41, 3.3%, 1.46, 1.G4. M'. It is not necessary to complete the table on page 435, as the Comptroller's Report for 1910 (p. 64) gives for the first time deposits subject to check (7.82 billions). To this 7.82, however, three corrections are needed : (1) subtract .29 for " savings accounts " improperly included (estimated for me by the Comptroller's Office at half of the figure in note a, lower table, p. 64, Comptr. Rpt.) ; (2) subtract .54 as "exchanges for clearing house" (= f times those for national banks) ; (3) add .25 xxiii XXIV ADDENDUM a great increase over 1909. The disproportionate growth of deposits relatively to money and the excessive velocity of circulation (52.7) of deposits, substantially equal to the unprecedented figure for 1909, are disquieting symptoms and serve only to confirm the forebodings in the text. For aid in working out the figures in this addendum I am indebted to three of my students, Mr. H. A. W. Duckert, Mr. J. M. Shortliffe, and Mr. M. G. Hastings. as the Comptroller's Office estimate, for me, of unreported deposits sub- ject to check. By applying these corrections we obtain 7.24. V. I have simply taken 21 as a safe approximate estimate on the basis of the previous statistics of V (p. 478) and its assumed relation to V'. M' V and V. Add to columns 1-7 of table on page 448 the following : 1910, 97.3, 66.4, 429.3, .89 (by extrapolation, an unsafe guide), 382, 52.8. P. This is obtained (on the principles of the table on page 487) from the index number 131.6 of wholesale prices for 1910 (kindly supplied in advance of publication by the Bureau of Labor) and the average price 96.2 of stocks as given by the Commercial and Financial Chronicle, both being compared with the respective figures for 1909, viz. 126.5 and 97.5. They are combined by " weighting " the wholesale prices 10 and the stock prices 1 and reducing the results so that the average for 1909 shall be 100. T. This is obtained : (a) by continuing columns 1-5 of the table on page 479 by inserting : 1910, 160, 113, 162, 154 ; (the extension of column 2 for 1910 is made by means of somewhat more complete data than those enumerated on pages 480-482) ; (ft) by combining the result, 154, obtained for column 5 with the figures for railway cars handled. These were 19.8 millions for 1909 and 22.3 for 1910. Column 5 being weighted 10 and the car figures 1, we get as indices of trade : for 1909, 1718, and for 1910, 1763, showing an increase of 2.6%, which, applied to the (corrected) estimate of the absolute trade of 1909, viz. 387 billions, gives 397 as the absolute trade in 1910. (The opportunity is here taken to correct an inadvertence on pp. 480 ff. It should have been there stated that, of the 44 categories mentioned, some are alternative and not independent items, viz. those having the same names and differing only in the number of cities ; also that the dates given do not imply that the items opposite are in all cases used for all the intervening time, but only for such periods as the items were actually available.) It is noticeable that the changes in business in 1910 as compared with 1909 are somewhat irregular ; the sales of stocks have declined ; exports and imports (both of them) have declined about 10 %. THE PURCHASING POWER OF MONEY THE PUECHASING POWER OF MONEY CHAPTER II PRIMARY DEFINITIONS In order to make clear the relation which the topic treated in this book bears to the general subject of economics, some primary definitions are necessary. In the first place, economics itself may be defined as the science of wealth, and wealth may be defined as material objects owned by human beings. Of wealth, therefore, there are two essential attributes : material- ity and appropriation. For it is not all material things that are included under wealth, but only such as have been appropriated. Wealth does not include the sun, moon, and other heavenly bodies, nor even all parts of the surface of this planet, but only such parts as have been appropriated to the use of mankind. It is, then, appropriated parts of the earth's surface and the appropriated objects upon it which constitute wealth. For convenience, wealth may be classified under three heads : real estate, commodities, and human beings. Real estate includes the surface of the earth and the other wealth attached thereto — improvements such as buildings, fences, drains, railways, street im- ' This chapter is mainly a condensation of Chapters I and II of the author's Nature of Capital and Income, New York (Macmillan), 1906. B 1 2 THE PURCHASING POWER OF MONEY [Chap. I provements, and so on. Commodities include all mova- ble wealth (except man himself), whether raw materials or finished products. There is one particular variety of commodity — a certain finished product — ■ which is of especial importance in the subject of which this book treats ; namely, money. Any commodity to be called " money " must be generally acceptable in exchange, and any commodity generally acceptable in exchange should be called money. The best example of a money commodity is found to-day in gold coins. Of all wealth, man himself is a species. Like his horses or his cattle, he is himself a material object, and like them, he is owned ; for if slave, he is owned by another, and if free, by himself.^ But though human beings may be considered as wealth, human qualities, such as skill, intelligence, and inventiveness, are not wealth. Just as the hardness of steel is not wealth, but merely a quality of one par- ticular kind of wealth, — hard steel, — so the skill of a workman is not wealth, but merely a quality of another particular kind of wealth — skilled workman. Similarly, intelligence is not wealth, but an intelhgent man is wealth. Since materiality is one of the two essential attributes of wealth, any article of wealth may be measured in physical units. Land is measured in acres ; coal, in tons ; milk, in quarts ; and wheat, in bushels. There- fore, for estimating the quantities of different articles of wealth, all the various physical units of measurement * If we wish to include only slaves as wealth and not free men, we shall have to amend our definition of wealth so as to read : Wealth consists of material objects owned by human beings and external to the owner. For the purpose of this book it makes no practical difference whether this narrower meaning or the broader one be employed. Sec. 1] PRIMARY DEFINITIONS 3 may be employed : linear measure, square measure, cubic measure, and measure by weight. Whenever any species of wealth is measured in its physical units, a first step is taken toward the measure- ment of that mysterious magnitude called ''value." Sometimes value is looked upon as a psychical and some- times as a physical phenomenon. But, although the determination of value always involves a psychical process — judgment — yet the terms in which the re- sults are expressed and measured are physical. It is desirable, for the sake of clearness, to lead up to the concept of value by means of three preliminary concepts ; namely, transfer, exchange, and price. A transfer of wealth is a change in its ownership. An exchange consists of two mutual and voluntary transfers, each in consideration of the other. When a certain quantity of one kind of wealth is exchanged for a certain quantity of another kind, we may divide one of the two quantities by the other, and obtain the price of the latter. For instance, if two dollars of gold are exchanged for three bushels of wheat, the price of the wheat in gold is two thirds of a dollar per bushel ; and the price of the gold in wheat is one and a half bushels per dollar. It is to be noticed that these are ratios of two physical quantities, the units for measuring which are quite different from each other. One commodity is measured in bushels, or units of volume of wheat, the other in dollars, or units of weight of gold. In general, a price of any species of wealth is merely the ratio of two physical quantities, in whatever way each may originally be measured. This brings us, at last, to the concept of value. The value of any item of wealth is its price multiplied by its quantity. Thus, if half a dollar per bushel is the price 4 THE PURCHASING POWER OF MONEY [Chap. 1 of wheat, the value of a hundred bushels of wheat is fifty dollars. §2 Hitherto we have confined our discussion to some of the consequences of the first prerequisite of wealth — that it must be material. We turn now to the second prerequisite, namely, that it must be owned. To own wealth is simply to have the right to benefit by it that is, the right to enjoy its services or benefits. Thus the owner of a loaf of bread has the right to benefit bj it by eating it, by selling it, or by otherwise disposing of it. The man who owns a house has the right to benefit by enjoying its shelter, by selling it, or by rent- ing it. This right, the right to or in the benefits of wealth — or more briefly, the right to or in the wealth it- self — • is called a " property right " or simply "property." If things were always owned in fee simple, i.e. if there were no division of ownership, — no partnership rights, no shares, and no stock companies, — there would be little practical need to distinguish property from wealth ; and as a matter of fact, in the rough popular usage, any article of wealth, and especially real estate, is often inaccurately called a "piece of property." But the ownership of wealth is frequently divided; and this fact necessitates a careful distinction between the thing owned and the rights of the owners. Thus, a railroad is wealth. Its shares and bonded debt are rights to this wealth. Each owner of shares or bonds has the right to a fractional part of the benefits from the railway. The total of these rights comprises the complete owner- ship of, or property in, the railway. Like wealth, property rights also may be meas- ured ; but in units of a different character. The units Sec. 2] PRIMARY DEFINITIONS 5 of property are not physical, but consist of abstract rights to the benefits of wealth. If a man has twenty- five shares in a certain railway company, the measure- ment of his property is twenty-five units just as truly as though he had twenty-five bushels of wheat. What he has is twenty-five rights of a specific sort. There exist various units of property for measuring property, as there are various units of wealth for meas- uring wealth ; and to property may be applied pre- cisely the same concepts of transfer, exchange, price, and value which are applied to wealth. Besides the distinction between wealth and property rights, another distinction should here be noted. This is the distinction between property rights and certificates of those rights. The former are the rights to use wealth, the latter are merely the written evidence of those rights. Thus, the right to receive dividends from a railroad is property, but the writjten paper evidencing that right is a stock certificate. TT!&^%jh.t to a railway trip is a property right, the ticket evidencing that right is a certificate of property. The promise of a bank is a property right ; the bank note on which that promise is engraved is a certificate of property. Any property right which is generally acceptable in exchange may be called "money." Its printed evi- dence is also called money. Hence there arise three meanings of the term money, viz. its meaning in the sense of wealth ; its meaning in the sense of property ; ^ and its meaning in the sense of written evidence. From the standpoint of economic analysis the prop- erty sense is the most important. ^ Cf. Menger, Handworterbuch der Siaatswissenschaften, Jena (Fischer), Vol. IV, 1900, Article, "Geld," pp. 69-71. 6 THE PURCHASING POWER OP MONEY [Chap. I What we have been speaking of as property is the right to the services, uses, or benefits of wealth. By benefits of wealth is meant the desirable events which occur by means of wealth. Like wealth and property, benefits also may be measured, but in units of a still different character. Benefits are reckoned either ' ' by time, " — as the services of a gardener or of a dwelling house; or "by the piece," — as the use of a plow or a telephone. And just as the concepts of transfer, exchange, price, and value apply to wealth and property, so do they apply to benefits. The uses (benefits) of wealth, with which we have been \4ealing, should be distinguished from the utility of wealth. The one means desirable events, the other, the desirability of those events. The one is usually outside of the mind, the other always inside. Whenever we speak of rights to benefits, the benefits referred to are future benefits. The owner of a house owns the right to use it from the present instant on- ward. Its past use has perished and is no longer sub- ject to ownership. The term "goods" will be used in this book simply as a convenient collective term to include wealth, property, and benefits. The transfer, exchange, price, and value of goods take on innumerable forms. Under price alone, as thus fully applied to goods, fall rent, wages, rates of interest, prices in terms of money, and prices in terms of other goqds. But we shall be chiefly con- cerned in this book with prices of goods in terms of money. §3 Little has yet been said as to the relation of wealth, property and benefits to time. A certain quantity of goods may be either a quantity existing at a partic- Sec. 3] PRIMARY DEFINITIONS 7 ular instant of time or a quantity produced, consumed, transported, or exchanged during a period of time. The first quantity is a stock, or fund, of goods ; the second is a flow, or stream, of goods. The amount of wheat in a flour mill on any definite date is a stock of wheat, while the monthly or weekly amounts which come in or go out constitute a flow of wheat. The amount of mined coal existing in the United States at any given moment is a stock of mined coal ; the weekly amount mined is a flow of coal. There are many applications of this distinction ; for instance, to capital and income. A stock of goods, whether wealth or property, existing at an instant of time is called capital. A flow of benefits from such capital during a period of time is called " income." In- come, therefore, is one important kind of economic flow. Besides income, economic flows are of three chief classes, representing respectively changes of condition (such as production or consumption), changes of position (such as transportation, exportation, and importation), and changes of ownership, which we have already called " transfers." Trade is a flow of transfers. Whether for- eign or domestic, it is simply the exchange of a stream of transferred rights in goods for an equivalent stream of transferred money or money substitutes. The second of these two streams is called the "circulation" of money. The equation between the two is called the "equation of exchange " ; and it is this equation that constitutes the subject matter of the present book. CHAPTER II PURCHASING POWER OF MONEY AS RELATED TO THE EQUATION OF EXCHANGE § 1 We define money as what is generally acceptable in exchange for goods. ^ The facility with which it may thus be exchanged, or its general acceptability, is its distinguishing characteristic. The general acceptability may be reenforced by law, the money thus becoming what is known as "legal tender"; but such reenforce- ment is not essential. All that is necessary in order that any good may be money is that general acceptability attach to it. On the frontier, without any legal sanc- tion, money is sometimes gold dust or gold nuggets. In the Colony of Virginia it was tobacco. Among the Indians in New England it was wampum. "In German New Guinea the bent tusks of a boar are used as money. In Cahfornia red birds' heads have been used in the same way." ^ Stone money and shell money are so used in Melanesia.^ "In Burmah Chinese gambling counters are used as money. Guttapercha tokens issued by 1 For discussions on the definition of money, see A. Piatt An- drew, "What ought to be called Money" in Quarterly Journal of Economics, Vol. XIII ; Jevons, Money and the Mechanism of Ex- change, London (Kegan Paul) and New York (Appleton), 1896; Palgrave, Dictionary of Political Economy; Walker, Money, and other treatises and textbooks. 2 Sumner, Folkways, Boston (Ginn), 1907, p. 147. 3 Ibid., p. 150. 8 Sec. 1] THE EQUATION OF EXCHANGE 9 street car companies in South America are said to be used in the same way." ^ Not many years ago in a town in New York state, similar tokens got into local circulation until their issue was forbidden by the United States government. In Mexico large cacao beans of relatively poor quahty were used as money, and on the west coast of Africa little mats were used.^ The list could be extended indefinitely. But whatever the substance of such a commodity, it is general exchange- ability which makes it money. On the other hand, even what is made legal tender may, by general usage, be deprived of its practical char- acter as money. During the Civil War the govern- ment attempted to circulate fifty-dollar notes, bearing interest at 7.3 per cent, so that the interest amounted to the very easily computed amount of a cent a day. The notes, however, failed to circulate. In spite of the attempt to make their exchange easy, people preferred to keep them for the sake of the interest.^ Money never bears interest except in the sense of creating con- venience in the process of exchange. This convenience is the special service of money and offsets the apparent loss of interest involved in keeping it in one's pocket instead of investing. There are various degrees of exchangeability which must be transcended before we arrive at real money. Of all kinds of goods, perhaps the least exchangeable is real estate. Only in case some person happens to be found who wants it, can a piece of real estate be traded. A mortgage on real estate is one degree more exchange- able. Yet even a mortgage is less exchangeable than a well-known and safe corporation security ; and a cor- 1 Sumner, Folkways, p. 148. '^ Ibid. ' See Jevons, Money and the Mechanism of Exchange, p. 245. 10 THE PURCHASING POWER OF MONEY [Chap. II poration security is less exchangeable than a government bond. In fact persons not infrequently buy govern- ment bonds as merely temporary investments, intending to sell them again as soon as permanent investments yielding better interest are obtainable. One degree more exchangeable than a government bond is a bill of exchange ; one degree more exchangeable than a bill of exchange is a sight draft ; while a check is almost as exchangeable as money itself. Yet no one of these is really money for none of them is ^^ generally accept- able." If we confine our attention to present and normal conditions, and to those means of exchange which either are money or most nearly approximate it, we shall find that money itself belongs to a general class of property rights which we may call " currency " or '^ circulating media." Currency includes any type of property right which, whether generally acceptable or not, does actually, for its chief purpose and use, serve as a means of exchange. Circulating media are of two chief classes : (1) money ; (2) bank deposits, which will be treated fully in the next chapter. By means of checks, bank deposits serve as a means of payment in exchange for other goods. A check is the "certificate" or evidence of the transfer of bank deposits. It is acceptable to the payee only by his consent. It would not be generally accepted by strangers. Yet by checks, bank deposits even more than money do actually serve as a medium of exchange. Practically speaking, money and bank deposits subject to check are the only circulating media. If post-office orders and telegraphic transfer are to be included, they may be regarded as certificates of transfer of special deposits, the post oflfice or telegraph company serving Sec. 1] THE EQUATION OF EXCHANGE 11 the purpose, for these special transactions, of a bank of deposit. But while a bank deposit transferable by check is included as circulating media, it is not money. A bank note, on the other hand, is both circulating medium and money. Between these two lies the final line of distinction between what is money and what is not. True, the line is delicately drawn, especially when we come to such checks as cashier's checks or certified checks, for the latter are almost identical with bank notes. Each is a demand liability on a bank, and each confers on the holder the right to draw money. Yet while a note is generally acceptable in exchange, a check is specially acceptable only, i.e. only by the con- sent of the payee. Real money rights are what a payee accepts without question, because he is induced to do so either by ''legal tender" laws or by a well-established custom.^ Of real money there are two kinds : primary and fiduciary. Money is called " primary" if it is a commod- ity which has just as much value in some use other than money as it has in monetary use. Primary money has its full value independently of any other wealth. Fidu- ciary money, on the other hand, is money the value of which depends partly or wholly on the confidence that the owner can exchange it for other goods, e.g. for primary money at a bank or government office, or at any rate for discharge of debts or purchase of goods of mer- chants. The chief example of primary money is gold coin ; the chief example of fiduciary money is bank notes. The qualities of primary money which make for exchangeability are numerous. The most important * See Francis Walker, Money, Trade, and Industry, New York (Holt), 1879, Chapter I. 12 THE PURCHASING POWER OF MONEY [Chap. II are portability, durability, and divisibility.^ The chief quality of fiduciary money which makes it exchangeable is its redeemability in primary money, or else its im- posed character of legal tender. Bank notes and all other fiduciary money, as well as bank deposits, circulate by certificates often called ''tokens." ''Token coins" are included in this de- scription. The value of these tokens, apart from the rights they convey, is small. Thus the value of a silver dollar, as wealth, is only about forty cents ; that is all that the actual silver in it is worth. Its value as prop- erty, however, is one hundred cents ; for its holder has a legal right to use it in paying a debt to that amount, and a customary right to so use it in payment for goods. Likewise, the property value of a fifty-cent piece, a quarter, a ten-cent piece, a five-cent piece, or a one-cent piece is considerably greater than its value as wealth. The value of a paper dollar as wealth — for instance, a silver certificate — is almost nothing. It is worth just its value as paper, and no more. But its value as property is a hundred cents, that is, the equiva- lent of one gold dollar. It represents to that extent a claim of the holder on the wealth of the community. Figure 1 indicates the classification of all circulating media in the United States. It shows that the total amount of circulating media is about 8| billions, of which about 7 billions are bank deposits subject to check, and 1^ billions, money ; and that of this 1^ billions of money, 1 billion is fiduciary money and only about I a billion, primary money. In the present chapter we shall exclude the consider- ation of bank deposits or check circulation and confine our attention to the circulation of money, primary 1 See Jevons, Money and the Mechanism of Exchange, Chapter V. Sec. 1] THE EQUATION OF EXCHANGE 13 and fiduciary. In the United States, the only primary money is gold coin. The fiduciary money includes (1) token coins, viz. silver dollars, fractional silver, and minor coins ("nickels" and cents); (2) paper money, viz. (a) certifi- Deposits subject TO CHECK Fiduciary MONE Y MOWEY BlLUION Billions Fig. 1. cates for gold and sil- ver, and (6) promissory notes, whether of the United States govern- ment ("greenbacks"), or of the National banks. Checks aside, we may classify exchanges into three groups : the /'exchange of goods against goods, or bar- ter; the exchange of . money against money, or changing money; and the \5' exchange of money against goods, or purchase and sale. Only the last-named species of exchange makes up what we call the "circulation" of money. The circulation of money signifies, therefore, the aggregate amount of its transfers against goods. All money held for circula- tion, i.e. all money, except what is in the banks and United States government's vaults, is called "money in circulation." The chief object of this book is to explain the causes determining the purchasing power of money. The purchasing power of money is indicated by the quan- tities of other goods which a given quantity of money will buy. The lower we find the prices of goods, the larger the quantities that can be bought by a given amount of money, and therefore the higher the purchas- 14 THE PURCHASING POWER OF MONEY [Chap. II ing power of money. The higher we find the prices of goods, the smaller the quantities that can be bought by a given amount of money, and therefore the lower the purchasing power of money. In short, the purchas- ing power of money is the reciprocal of the level of prices ; so that the study of the purchasing power of money is identical with the study of price levels. §2 Overlooking the influence of deposit currency, or checks, the price level may be said to depend on only three sets of causes : (1) the quantity of money in circu- lation ; (2) its "efficiency" or velocity of circulation (or the average number of times a year money is exchanged for goods) ; and (3) the volume of trade (or amount of goods bought by money). The so-called "quantity theory," ^ i.e. that prices vary proportionately to money, has often been incorrectly formulated, but (overlooking checks) the theory is correct in the sense that the level of prices varies directly with the quantity of money in circulation, provided the velocity of circulation of that money and the volume of trade which it is obliged to perform are not changed. The quantity theory has been one of the most bitterly contested theories in economics, largely because the recognition of its truth or falsity affected powerful ^ This theory, though often crudely formulated, has been accepted by Locke, Hume, Adam Smith, Ricardo, Mill, Walker, IMarshall, Hadley, Fetter, Kemmerer and most writers on the subject. The Roman Julius Paulus, about 200 a.d., stated his belief that the value of money depends on its quantity. See Zuckerkandl, Theorie des Preises; Kemmerer, Money and Credit Instruments in their Relation to General Prices, New York (Holt), 1909. It is true that many writers still oppose the quantity theory. See especially, Laughlin, Principles of Money, New York (Scribner), 1903. Sec. 2] THE EQUATION OF EXCHANGE 15 interests in commerce and politics. It has been main- tained — and the assertion is scarcely an exaggeration — that the theorems of Euclid would be bitterly con- troverted if financial or political interests were in- volved. The quantity theory has, unfortunately, been made the basis of arguments for unsound currency schemes. It has been invoked in behalf of irredeemable paper money and of national free coinage of silver at the ratio of 16 to 1. As a consequence, not a few ''sound money men," believing that a theory used to support such vagaries must be wrong, and fearing the political effects of its propagation, have drifted into the position of opposing, not only the unsound propaganda, but also the sound principles by which its advocates sought to bolster it up.^ These attacks upon the quantity theory have been rendered easy by the imperfect com- prehension of it on the part of those who have thus invoked it in a bad cause. Personally, I believe that few mental attitudes are more pernicious, and in the end more disastrous, than those which would uphold sound practice by denying sound principles because some thinkers make unsound application of those principles. At any rate, in scien- tific study there is no choice but to find and state the liji varnished truth. The quantity theory will be made more clear by the equation of exchange, which is now to be explained. The equation of exchange is a statement, in math- * See Scott, "It has been a most fruitful source of false doctrines regarding monetary matters, and is constantly and successfully employed in defense of harmful legislation and as a means of pre- venting needed monetary reforms." Money and Banking, New York, 1903, p. 68. 16 THE PURCHASING POWER OF MONEY [Chap. II ematical form, of the total transactions effected in a certain period in a given community. It is obtained simply by adding together the equations of exchange for all individual transactions. Suppose, for instance, that a person buys 10 pounds of sugar at 7 cents per pound. This is an exchange transaction, in which 10 pounds of sugar have been regarded as equal to 70 cents, and this fact may be expressed thus : 70 cents = 10 pounds of sugar multiplied by 7 cents a pound. Every other sale and purchase may be expressed similarly, and by adding them all together we get the equation of exchange for a certain -period in a given community. During this same period, however, the same money may serve, and usually does serve, for several transactions. For that reason the money side of the equation is of course greater than the total amount of money in cir- culation. The equation of exchange relates to all the purchases made by money in a certain community during a cer- tain time. We shall continue to ignore checks or any circulating medium not money. We shall also ignore foreign trade and thus restrict ourselves to trade within a hypothetical community. Later we shall reinclude these factors, proceeding by a series of approximations through successive hypothetical conditions to the actual conditions which prevail to-day. We must, of course, not forget that the conclusions expressed in each suc- cessive approximation are true solely on the particular hypothesis assumed. The equation of exchange is simply the sum of the equations involved in all individual exchanges in a year. In each sale and purchase, the money and goods ex- changed are ipso facto equivalent ; for instance, the money paid for sugar is equivalent to the sugar bought. Sec. 2] THE EQUATION OF EXCHANGE 17 And in the grand total of all exchanges for a year, the total money paid is equal in value to the total value of the goods bought. The equation thus has a money side and a goods side. The money side is the total money paid, and may be considered as the product of y the quantity of money multiplied by its rapidity of circulation. The goods side is made up of the products of quantities of goods exchanged multiplied by their respective prices. The important magnitude, called the velocity of cir- culation, or rapidity of turnover, is simply the quotient y^ obtained by dividing the total money payments for • goods in the course of a year by the average amount of money in circulation by which those payments are effected. This velocity of circulation for an entire com- munity is a sort of average of the rates of turnover of money for different persons. Each person has his own rate of turnover which he can readily calculate by di- viding the amount of money he expends per year by the average amount he carries. • Let us begin with the money side. If the number of dollars in a country is 5,000,000, and their velocity of circulation is twenty times per year, then the total amount of money changing hands (for goods) per year is 5,000,000 times twenty, or $100,000,000. This is the money side of the equation of exchange. Since the money side of the equation is $100,000,000, the goods side must be the same. For if $100,000,000 has been spent for goods in the course of the year, then $100,000,000 worth of goods must have been sold in that year. In order to avoid the necessity of writing out the quantities and prices of the innumerable va- rieties of goods which are actually exchanged, let us assume for the present that there are only three kinds 18 THE PURCHASING POWER OF MONEY [Chap. U of goods, — bread, coal, and cloth ; and that the sales are: — 200,000,000 loaves of bread at $ .10 a loaf, 10,000,000 tons of coal at 5.00 a ton, and 30,000,000 yards of cloth at 1 .00 a yard. The value of these transactions is evidently $100,000,- 000, i.e. $20,000,000 worth of bread plus $50,000,000 worth of coal plus $30,000,000 worth of cloth. The equation of exchange therefore (remember that the money side consisted of $5,000,000 exchanged 20 times) is as follows : — S5,000,000 X 20 times a year = 200,000,000 loaves X $ .10 a loaf + 10,000,000 tons X 5.00 a ton + 30,000,000 yards X 1.00 a yard. This equation contains on the money side two magni- tudes, viz. (1) the quantity of money and (2) its velocity of circulation ; and on the goods side two groups of magnitudes in two columns, viz. (1) the quantities of goods exchanged (loaves, tons, yards), and (2) the prices of these goods. The equation shows that these four sets of magnitudes are mutually related. Because this equation must be fulfilled, the prices must bear a relation to the three other sets of magnitudes, — quantity of money, rapidity of circulation, and quan- tities of goods exchanged. Consequently, these prices must, as a whole, vary proportionally with the quantity of money and with its velocity of circulation, and in- versely with the quantities of goods exchanged. Suppose, for instance, that the quantity of money were doubled, while its velocity of circulation and the quantities of goods exchanged remained the same. Then it would be quite impossible for prices to remain unchanged. The money side would now be Sec. 2] THE EQUATION OF EXCHANGE 19 $10,000,000 X 20 times a year or $200,000,000 ; whereas, if prices should not change, the goods would remain $100,000,000, and the equation would be violated. Since exchanges, individually and collectively, always involve an equivalent quid pro quo, the two sides must be equal. Not only must purchases and sales be equal in amount — since every article bought by one person is necessarily sold by another — but the total value of goods sold must equal the total amount of money exchanged. Therefore, under the given conditions, prices must change in such a way as to raise the goods side from $100,000,000 to $200,000,000. This doubling may be accomplished by an even or uneven rise in prices, but some sort of a rise of prices there must he. If the prices rise evenly, they will evidently all be exactly doubled, so that the equation will read : — $10,000,000 X 20 times a year = 200,000,000 loaves X $ .20 per loaf + 10,000,000 tons X 10.00 per ton + 30,000,000 yards X 2.00 per yard. If the prices rise unevenly, the doubling must evidently be brought about by compensation ; if some prices rise by less than double, others must rise by enough more than double to exactly compensate. But whether all prices increase uniformly, each being exactly doubled, or some prices increase more and some less (so as still to double the total money value of the goods purchased), the prices are doubled on the average} This proposition is usually expressed by saying that the "general level of prices" is raised twofold. From the mere fact, therefore, that the money spent for goods 1 This does not mean, of course, that their simple arithmetical average is exactly doubled. For definition of an average or "mean" in general, see § 1 of Appendix to (this) Chapter II. 20 THE PURCHASING POWER OF MONEY [Chap. II must equal the quantities of those goods multipHed by their prices, it follows that the level of prices must rise or fall according to changes in the quantity of money, unless there are changes in its velocity of circulation or in the quantities of goods exchanged. If changes in the quantity of money affect prices, so will changes in the other factors — quantities of goods and velocity of circulation — affect prices, and in a very similar manner. Thus a doubling in the velocity of circulation of money will double the lev«l of prices, provided the quantity of money in circulation and the quantities of goods exchanged for money remain as be- fore. The equation will become : — $5,000,000 X 40 times a year = 200,000,000 loaves X $ .20 a loaf + 10,000,000 tons X 10.00 a ton + 30,000,000 yards X 2.00 a yard, or else the equation will assume a form in which some of the prices will more than double, and others less than double by enough to preserve the same total value of the sales. Again, a doubling in the quantities of goods exchanged will not double, but halve, the height of the price level, provided the quantity of money and its velocity of circulation remain the same. jJnder these circum- stances the equation will become : — ),000,000 X 20 times a year = 400,000,000 loaves X S .05 a loaf + 20,000,000 tons X 2.50 a ton + 60,000,000 yards X .50 a yard, or else it will assume a form in which some of the prices are more than halved, and others less than halved, BO as to preserve the equation. Sec. 3] THE EQUATION OF EXCHANGE 21 Finally, if there is a simultaneous change in two or all of the three influences, i.e. quantity of money, velocity of circulation, and quantities of goods ex- changed, the price level will be a compound or resultant of these various influences. If, for example, the quan- tity of money is doubled, and its velocity of circulation is halved, while the quantity of goods exchanged remains constant, the price level will be undisturbed. Like- wise, it will be undisturbed if the quantity of money is doubled an^ the quantity of goods is doubled, while the velocity of circulation remains the same. To double the quantity of money, therefore, is not always to double prices. We must distinctly recognize that the quantity of money is only one of three factors, all equally impor- tant in determining the price level. §3 The equation of exchange has now been expressed by an arithmetical illustration. It may be also represented visually, by a mechanical illustration. Such a repre- sentation is embodied in Figure 2. This represents a Fig. 2. mechanical balance in equilibrium, the two sides of which symbolize respectively the money side and the goods side of the equation of exchange. The weight at the left, symbolized by a purse, represents the money in circulation; the ''arm" or distance from the fulcrum at which this weight (purse) is hung represents the 22 THE PURCHASING POWER OF MONEY [Chap. II efficiency of this money, or its velocity of circulation. On the right side are three weights, — bread, coal, and cloth, symbolized respectively by a loaf, a coal scuttle, and a roll of cloth. The arm, or distance of each from the fulcrum, represents its price. In order that the lever arms at the right may not be inordinately long, we have found it convenient to reduce the unit of measure of coal from tons to hundredweights, and that of cloth from yards to feet, and consequently to enlarge corre- spondingly the numbers of units (the measure of coal changing from 10,000,000 tons to 200,000,000 hundred- weights, and that of the cloth from 30,000,000 yards to 90,000,000 feet). The price of coal in the new unit per hundredweight becomes 25 cents per hundredweight, and that of cloth in feet becomes SS-g- cents per foot. We all know that, when a balance is in equilibrium, the tendency to turn in one direction equals the tend- ency to turn in the other. Each weight produces on its side a tendency to turn, measured by the product of the weight by its arm. The weight on the left produces, on that side, a tendency measured by 5,000,000 x 20; while the weights on the right make a combined opposite tendency measured by 200,000,000 x .10 + 200,000,000 x .25-^90,000,000 X ..33i The equality of these opposite tendencies represents the equation of exchange. An increase in the weights or arms on one side re- quires, in order to preserve equilibrium, a proportional increase in the weights or arms on the other side. This simple and familiar principle, applied to the symbolism here adopted, means that if, for instance, the velocity of circulation (left arm) remains the same, and if the trade (weights at the right) remains the same, then any increase of the purse at the left will require a lengthen- ing of one or more of the arms at the right, represent- Sec. 3] THE EQUATION OF EXCHANGE 23 ing prices. If these prices increase uniformly, they will increase in the same ratio as the increase in money ; if they do not increase uniformly, some will increase more and some less than this ratio, maintaining an average. Likewise it is evident that if the arm at the left lengthens, and if the purse and the various weights on the right remain the same, there must be an increase in the arms at the right. Again, if there is an increase in weights at the right, and if the left arm and the purse remain the same, there must be a shortening of right arms. In general, a change in one of the four sets of mag- nitudes must be accompanied by such a change or changes in one or more of the other three as shall main- tain equilibrium. As we are interested in the average change in prices rather than in the prices individually, we may simplify this mechanical representation by hanging all the right- hand weights at one average point, so that the arm shall represent the average prices. This arm is a ''weighted average" of the three original arms, the weights being literally the weights hanging at the right. This averaging of prices is represented in Figure 3, which visualizes the fact that the average price of goods X__Mi Pro (right arm) varies directly with the quantity of money (left weight), and directly with its velocity of circulation 24 THE PURCHASING POWER OF MONEY [Chap. II (left arm), and inversely with the volume of trade (right weight). §4 We now come to the strict algebraic statement of the equation of exchange. An algebraic statement is usually a good safeguard against loose reasoning ; and loose reasoning is chiefly responsible for the sus- picion under which economic theories have frequently fallen. If it is worth while in geometry to demonstrate carefully, at the start, propositions which are almost self-evident, it is a hundredfold more worth while to demonstrate with care the propositions relating to price levels, which are less self-evident ; which, indeed, while confidently assumed by many, are contemptuously rejected by others. Let us denote the total circulation of money, i.e. the amount of money expended for goods in a given com- munity during a given year, by E (expenditure) ; and the average amount of money in circulation in the com- munity during the year by M (money). M will be the simple arithmetical average of the amounts of money existing at successive instants separated from each other by equal intervals of time indefinitely small. If we divide the year's expenditures, E, by the average amount of money, M, we shall obtain what is called the average rate of turnover of money in its exchange E for goods, — , that is, the velocity of circulation of money. ^ This velocity may be denoted by V, so that E / ijj = ^J then E may be expressed as MV. In words : the total circulation of money in the sense of ^ For discussion of the concept of velocity of circulation, see §§ 2, 4, 5 of Appendix to (this) Chapter II. Sec. 4] THE EQUATION OF EXCHANGE 25 money expended is equal to the total money in circu- lation multiplied by its velocity of circulation or turn- over. E ovMV, therefore, expresses the money side of the equation of exchange. Turning to the goods side of the equation, we have to deal with the prices of goods exchanged and quantities of goods exchanged. The average^ price of sale of any particular good, such as bread, purchased in the given community during the given year, may be represented by p (price) ; and the total quantity of it purchased, by Q (quantity) ; hkewise the average price of another good (say coal) may be represented by p' and the total quantity of it exchanged, by Q' ; the average price and the total quantity of a third good (say cloth) may be represented by p" and Q" respectively ; and so on, for all other goods exchanged, however numerous. The equation of exchange may evidently be expressed as follows : ^ — MV = pQ + v'Q' + P"Q" + etc. ^ This is an average weighted according to the quantities pur- chased on various occasions throughout the period and country considered. See § 3 of Appendix to (this) Chapter II. 2 An algebraic statement of the equation of exchange was made by Simon Newcomb in his able but little appreciated Principles of Political Economy, New York (Harper), 1885, p. 346. It is also expressed by Edgeworth, " Report on Monetary Standard." Report of the British Association for the Advancement of Science, 1887, p. 293, and by President Hadley, Economics, New York (Putnam), 1896, p. 197. See also Irving Fisher, "The Role of Capital in Eco- nomic Theory," Economic Journal, December, 1899, pp. 515-521, and E. W. Kemmerer, Money and Credit Instruments in their Rela- tion to General Prices, New York (Holt), 1907, p. 13. While thus only recently given mathematical expression, the quantity theory has long been understood as a relationship among the several fac- 26 THE PURCHASING POWER OF MONEY [Chap. II The right-hand side of this equation is the sum of terms of the form pQ — a price multipHed by a quantity bought. It is customary in mathematics to abbreviate such a sum of terms (all of which are of the same form) by using ''2 " as a symbol of summation. This symbol does not signify a magnitude as do the symbols Af , V, p, Q, etc. It signifies merely the operation of addition and should be read ''the sum of terms of the following type." The equation of exchange may therefore be written : — MV = SpQ. That is, the magnitudes E, M, V, the p's and the Q's relate to the entire community and an entire year ; but they are based on and related to corresponding magni- tudes for the individual persons of which the community is composed and for the individual moments of time of which the year is composed.^ The algebraic derivation of this equation is, of course, essentially the same as the arithmetical derivation previously given. It consists simply in adding together the equations for all individual purchases within the community during the year.- By means of this equation, MV = ^pQ, the three theorems set forth earlier in this chapter may be now expressed as follows : — (1) If V and the Q's remain invariable while M varies in any ratio, the money side of the equation will vary tors : amount of money, rapidity of circulation, and amount of trade. See Mill, Principles of Political Economy, Book III, Chapter VIII, § 3. Ricardo probably deserves chief credit for launching the theory. " For the relations subsisting between these magnitudes (as relating to the whole community and the whole year), and the corresponding elementary magnitudes relating to each individual and each moment, see § 4 of the Appendix to (this) Chapter II. ^ See § 6 of Appendix to (this) Chapter II. Sec. 4] THE EQUATION OF EXCHANGE 27 in the same ratio and therefore its equal, the goods side, must vary in that same ratio also ; consequently, either the p's will all vary in that ratio or else some p's will vary more than in that ratio and others enough less to compensate and maintain the same average.^ (2) If M and the Q's remain invariable while V varies in any ratio, the money side of the equation will vary in the same ratio, and therefore its equal, the goods side, must vary in that ratio also; consequently, the p's will all vary in the same ratio or else some will vary more and others enough less to compensate. (3) If Af and V remain invariable, the money side and the goods side will remain invariable ; consequently, if the Q's all vary in a given ratio, either the p's must all vary in the inverse ratio or else some of them will vary more and others enough less to compensate. We may, if we wish, further simplify the right side by writing it in the form PT where P is a weighted average of all the p's, and T is the sum of all the Q's. P then represents in one magnitude the level of prices, and T represents in one magnitude the volume of trade. This simplification is the algebraic interpretation of the mechanical illustration given in Figure 3, where all the goods, instead of being hung separately, as in Figure 2, were combined and hung at an average point representing their average price. We have derived the equation of exchange, MV = ^vQ, by adding together, for the right side, the sums expended by different persons. But the same reasoning would have derived an equation of exchange by taking the sums received by different persons. The results of ' For the nature of the average here involved and for the aver- ages involved in the other two following cases, see § 7 of Appendix to (this) Chapter II. 1/ 28 THE PURCHASING POWER OF MONEY [Chap. II the two methods will harmonize if the community has no foreign trade ; for, apart from foreign trade, what is expended by one person in the community is necessarily received by some other person in that community. If we wish to extend the reasoning so as to apply to foreign trade, we shall have two equations of ex- change, one based on money expended and the other on money received or accepted by members of the com- munity. These will always be approximately equal and may or may not be exactly equal within a country ac- cording to the "balance of trade" between that coun- try and others. The right side of the equation based on expenditures will include, in addition to the domes- tic quantities already represented there, the quantities of goods imporlsd and their prices, but not those ex- ported ; whilo the reverse will be true of the equation based on receipts. §5 This completes our statement of the equation of exchange, except for the element of check payments, which is reserved for the next chapter. We have seen that the equation of exchange has as its ultimate basis the elementary equations of exchange pertaining to given persons and given moments, in other words, the equa- tions pertaining to individual transactions. Such ele- mentary equations mean that the money paid in any transaction is the equivalent of the goods bought at the price of sale. From this secure and obvious premise is derived the equation of exchange MV = '^pQ, each element in which is a sum or an average of the like elementary elements for different individuals and differ- ent moments, thus comprising all the purchases in the community during the year. Finally, from this equa- Sec. 5] THE EQUATIOJS OF EXCHANGE 29 tion we see that prices vary directly as M and F, and inversely as the Q's, provided in each case only one of these three sets of magnitudes varies, and the other two remain unchanged. Whether to change one of the three necessarily disturbs the others is a question re- served for a later chapter. Those who object to the equation of exchange as a mere truism are asked to defer judgment until they have read Chapter VIII. To recapitulate, we find then that, under the con- ditions assumed, the price level varies (1) directly as the quantity of money in circulation (M), (2) directly as the velocity; of its circulation (F), (3) inversely as the volume of trade done by it (T). The first of these three relations is worth emphasis. It constitutes the " quantity theory of money." So important is this principle, and so bitterly con- tested has it been, that we shall illustrate it further. As already indicated, by "the quantity of money" is meant the number of dollars (or other given monetary units) in circulation. This number may be changed in several ways, of which the following three are most important. Their statement will serve to bring home to us the conclusions we have reached and to reveal the fundamental peculiarity of money on which they rest. As a first illustration, let us suppose the government to double the denominations of all money ; that is, let us suppose that what has been hitherto a half dollar is henceforth called a dollar, and that what has hitherto been a dollar is henceforth called two dollars. Evi- dently the number of "dollars" in circulation will then be doubled ; and the price level, measured in terms of the new "dollars," will be double what it would otherwise be. Every one will pay out the same coins as though no such law were passed. But he will, in 30 THE PURCHASING POWER OF MONEY [Chap. II each case, be paying twice as many ''dollars." Fol example, if $3 formerly had to be paid for a pair of shoes, the price of this same pair of shoes will now become $6. Thus we see how the nominal quantity of money affects price levels. A second illustration is found in a debased currency. Suppose a government cuts each dollar in two, coining the halves into new ''dollars" ; and, recalling all paper notes, replaces them with double the original number — • two new notes for each old one of the same denomination. In short, suppose money not only to be renamed, as in the first illustration, but also reissued ; prices in the de- based coinage will again be doubled just as in the first illustration. The subdivision and recoinage is an im- material circumstance, unless it be carried so far as to make counting difficult and thus to interfere with the convenience of money. Wherever a dollar had been paid before debasement, two dollars — i.e. two of the old halves coined into two of the new dollars — will now be paid instead. In the first illustration, the increase in quantity was simply nominal, being brought about by renaming coins. In the second illustration, besides renaming, the further fact of recoining is introduced. In the first case the number of actual pieces of money of each kind was unchanged, but their denominations were doubled. In the second case, the number of pieces is also doubled by splitting each coin and reminting it into two coins, each of the same nominal denomination as the original whole of which it is the half, and by similarly redoubUng the paper money. For a third illustration, suppose that, instead of doubhng the number of dollars by splitting them in two and recoining the halves, the government duplicates Sec. 5J THE EQUATION OF EXCHANGE 31 each piece of money in existence and presents the du- phcate to the possessor of the original.^ (We must in this case suppose, further, that there is some effectual bar to prevent the melting or exporting of money. Otherwise the quantity of money in circulation will not be doubled : much of the increase will escape.) If the quantity of money is thus doubled, prices will also be doubled just as truly as in the second illustration, in which there were exactly the same denominations. The only difference between the second and the third illustrations will be in the size and weight of the coins. The weights of the individual coins, instead of being reduced, will remain unchanged ; but their number will be doubled. This doubling of coins must have the same effect as the 50 per cent debasement, i.e. it must have the effect of doubling prices. The force of the third illustration becomes even more evident if, in accordance with Ricardo's presentation,^ we pass back by means of a seigniorage from the third illustration to the second. That is, after duplicating all money, let the government abstract half of each coin, thereby reducing the weight to that of the debased coinage in the second illustration, and removing the only point of distinction between the two. This ''seigniorage" abstracted will not affect the value of the coins, so long as their number remains unchanged. In short, the quantity theory asserts that (provided velocity of circulation and volume of trade are un- changed) if we increase the number of dollars, whether ^ Cf. J. S. Mill, Principles of Political Economy, Book III, Chap- ter VIII, § 2. Ricardo in his reply to Bosanquet uses an illus- tration similar in principle though slightly different in form. See Works, 2d ed., London (Murray), 1852, p. 346. » Works, 2d ed., London (Murray), 1852, pp. 346 and 347 (reply to Bosanquet, Chapter VI) ; see also pp. 213 and 214. 32 THE PURCHASING POWER OF MONEY [Chap. II by renaming coins, or by debasing coins, or by increasing coinage, or by any other means, prices will be increased in the same proportion. It is the number, and not the weight, that is essential. This fact needs great emphasis. It is a fact which differentiates money from all other goods and explains the peculiar manner in which its purchasing power is related to other goods. Sugar, for instance, has a specific desirability dependent on its quantity in pounds. Money has no such quality. The value of sugar depends on its actual quantity. If the quantity of sugar is changed from 1,000,000 pounds to 1,000,000 hundredweight, it does not follow that a hundredweight will have the value previously possessed by a pound. But if money in circulation is changed from 1,000,000 units of one weight to 1,000,000 units of another weight, the value of each unit will remain unchanged. The quantity theory of money thus rests, ultimately, upon the fundamental pecuUarity which money alone of all goods possesses, — the fact that it has no power to satisfy human wants except a power to purchase things which do have such power. ^ 1 Cf . G. F. Knapp, Staatliche Theorie des Geldes, Leipzig, 1905 ; L. von Bortkiewiez, " Die geldtheoretischen und die wahrungspoli- tisehen Consequenzen des ' Nominalismus,' " Jahrbuch fiir Gesetz- gebung, Verwaltung und Volkswirtschaft, October, 1906 ; Bertrand Nogaro, " L'experience bimetalliste du XIX siecle et la theorie generale de la monnaie," Revue d'Economie politique, 1908. CHAPTER III INFLUENCE OF DEPOSIT CURKENCY ON THE EQUATION AND THEREFORE ON PURCHASING POWER § 1 We are now ready to explain the nature of bank deposit currency, or circulating credit. Credit, in general, is the claim of a creditor against a debtor. Bank deposits subject to check are the claims of the creditors of a bank against the bank, by virtue of which they may, on demand, draw by check specified sums of money from the bank. Since no other kind of bank deposits will be considered by us, we shall usually refer to "bank deposits subject to check" simply as ''bank deposits." They are also called "circulating credit." Bank checks, as we have seen, are merely certificates of rights to draw, i.e. to transfer bank de- posits. The checks themselves are not the currency; the bank deposits which they represent are the currency. It is in the connection with the transfer of bank deposits that there arises that so-called "mystery of banking" called "circulating credit." Many persons, including some economists, have supposed that credit is a special form of wealth which may be created out of whole cloth, as it were, by a bank. Others have maintained that credit has no foundation in actual wealth at all, but is a kind of unreal and inflated bubble with a precarious, if not wholly illegitimate, existence. As a matter of fact, bank deposits are as easy to under- stand as bank notes, and what is said in this chapter D 33 34 THE PURCHASING POWER OF MONEY [Chap. Ill of bank deposits may in substance be taken as true also of bank notes. The chief difference is a formal one, the notes circulating from hand to hand, while the deposit currency circulates only by means of special orders called "checks." To understand the real nature of bank deposits, let us imagine a hypothetical institution, — a kind of primitive bank existing mainly for the sake of deposits and the safe keeping of actual money. The original bank of Amsterdam was somewhat like the bank we are now imagining. In such a bank a number of people deposit $100,000 in gold, each accepting a receipt for the amount of his deposit. If this bank should issue a ''capital account," or statement, it would show $100,000 in its vaults and $100,000 owed to depositors, as fol- lows : — Assets Liabilities Gold $100,000 I Due depositors . . $100,000 The right-hand side of the statement is, of course, made up of smaller amounts owed to individual de- positors. Assuming that there is owed to A, $10,000, to B, $10,000, and to all others $80,000, we may write the bank statement as follows : — Assets Liabilities Gold $100,000 $100,000 Due depositor A . . $10,000 Due depositor B . . 10,000 Due other depositors 80,000 $100,000 Now assume that A wishes to pay B $1000. A could go to the bank with B, present certificates or checks for $1000, obtain the gold, and hand it over to B, who might then redeposit it in the same bank, merely hand- Sec. 1] DEPOSIT CURRENCY 35 ing it back through the cashier's window and taking a new certificate in his own name. Instead, however, of both A and B visiting the bank and handhng the money, A might simply give B a check for $1000. The transfer in either case would mean that A's holding in the bank was reduced from S 10,000 to $9000, and that B's was increased from $10,000 to $11,000. The state- ment would then read : — Assets Liabilities Gold $100,000 $100,000 Due depositor A . . $ 9,000 Due depositor B . . 11,000 Due other depositors 80,000 $100,000 Thus the certificates, or checks, would circulate in place of cash among the various depositors in the bank. What really changes ownership, or ''circulates," in such cases is the rigM to draw money. The check is merely the evidence of this right and of the transfer of ohis right from one person to another. In the case under consideration, the bank would be con- ducted at a loss. It would be giving the time and labor of its clerical force for the accommodation of its deposi- tors, without getting anything in return. But such a hypothetical bank would soon find — much as did the bank of Amsterdam^ — that it could ''make money" by lending at interest some of the gold on deposit. This could not offend the depositors; for they do not expect or desire to get back the identical gold they de- posited. What they want is simply to be able at any time to obtain the same amount of gold. Since, then, 1 See Dunbar's Theory and History of Banking, 2d ed., edited by O. M. W. Sprague, New York and London (G. P. Putnam's Sons), 1901, pp. 113-116. 36 THE PURCHASING POWER OF MONEY [Chap. Ill their arrangement with the bank calls for the payment^ not of any particular gold, but merely of a definite amount, and that but occasionally, the bank finds it- self free to lend out part of the gold that otherwise would lie idle in its vaults. To keep it idle would be a great and needless waste of opportunity. Let us suppose, then, that the bank decides to loan out half its cash. This is usually done in exchange for promissory notes of the borrowers. Now a loan is really an exchange of money for a promissory note which the lender — in this case the bank — receives in place of the gold. Let us suppose that so-called borrowers actually draw out S50,000 of gold. The bank thereby exchanges money for promises, and its books will then read : — Assets Gold reserve . . . Promissory notes . $50,000 50,000 $100,000 Liabilities Due depositor A . , Due depositor B . , Due other depositors $ 9,000 11,000 80,000 $100,000 It will be noted that now the gold in the bank is only $50,000, while the total deposits are still S100,000. In other words, the depositors now have more "'money on deposit" than the bank has in its vaults ! But, as will be shown, this form of expression involves a popular fallacy in the word ''money." Something good is behind each loan, but not necessarily money. Next, suppose that the borrowers become, in a sense, depositors also, by redepositing the $50,000 of cash which they borrowed, in return for the right to draw out the same sum on demand. In other words, suppose Sec. 1] DEPOSIT CURRENCY 37 that after borrowing $50,000 from the bank, they lend it back to the bank. The bank's assets will thus be enlarged by $50,000, and its obligations (or credit extended) will be equally enlarged; and the balance sheet will become : — Assets Gold reserve . . Promissory notes . iX Ct^ $100,000 50,000 $150,000 Liabilities Due depositor A . . Due depositor B . . Due old depositors . Due new depositors, i.e. the borrowers . 5 9,000 11,000 80,000 50,000 $150,000 What happened in this case was the following : Gold was borrowed in exchange for a promissory note and then handed back in exchange for a right to draw. Thus the gold really did not budge; but the bank received a promissory note and the depositor a right to draw. Evidently, therefore, the same result would have followed if each borrower had merely handed in his prom- issory note and received, in exchange, a right to draw. As this operation most frequently puzzles the beginner in the study of banking, we repeat the tables repre- senting the conditions before and after these ''loans," i.e. these exchanges of promissory notes for present rights to draw.^ 1 In the ultimate analysis, and outside of its function of insuring credit, a bank is really an intermediary between borrowers and lenders. It is by virtue of bringing borrowers and ultimate lenders together and providing the former with a supply of loans which would not otherwise exist, that a bank simultaneously tends to lower the rate of interest and increase the supply of credit cur- rency. See paper by Harry G. Brown, in the Quarterly Journal of Economics, August, 1910, on "Commercial Banking and the Rate of Interest." 4 p f* I* r* 38 THE PURCHASING POWER OF MONEY [Chap. Ill Before the Loans Assets Liabilities Gold reserve . . . $100,000 1 Due depositors . . After the Loans $100,000 Gold reserve . . . Promissory notes . . $100,000 50,000 Due depositors . . $150,000 Clearly, therefore, the intermediation of the money in this case is a needless complication, though it may help to a theoretical miderstanding of the resultant shifting of rights and liabilities. Thus the bank may receive deposits of gold or deposits of promises. In exchange for the promises it may give, or lend, either a right to draw, or gold, — the same that was deposited by another customer. Even when the borrower has only a promise, by fiction he is still held to have de- posited money ; and like the original cash depositors, he is given the right to make out checks. The total value of rights to draw, in whichever way arising, is termed ''deposits." Banks more often lend rights to draw (or deposit rights) than actual cash, partly because of the greater convenience to borrowers, and partly because the banks wish to keep their cash reserves large, in order to meet large or unexpected demands. It is true that if a bank loans money, part of the money so loaned will be redeposited by the persons to whom the borrowers pay it in* the course of business ; but it will not neces- sarily be redeposited in the same bank. Hence the average banker prefers that the borrower should not withdraw actual cash. Besides lending deposit rights, banks may also lend their own no^es, called ''bank notes." And the principle governing bank notes is the same as the prin- Sec. 1] DEPOSIT CURRENCY 39 ciple governing deposit rights. The holder simply gets a pocketful of bank notes instead of a bank account. In either case the bank must be always ready to pay the holder — to ''redeem its notes" — as well as pay its depositors, on demand, and in either case the bank exchanges a promise for a promise. In the case of the note, the bank has exchanged its bank note for a cus- tomer's promissory note. The bank note carries no interest, but is payable on demand. The customer's note bears interest, but is payable only at a definite date. Assuming that the bank issues $50,000 of notes, the balance sheet will now become : — ., 'L, Assets Gold reserve . . . . $100,000 Loans 100,000 Liabilities $200,000 Due depositors , Due note holders % $150,000 50,000 $200,000 We repeat that by means of credit the deposits (and notes) of a bank may exceed its cash. There would be nothing mysterious or obscure about this fact, nor about credit in general, if people could be induced not to think of banking operations as money operations. To so represent them is metaphorical and misleading. They are no more money operations than they are real estate transactions. A bank depositor, A, has not ordinarily ''deposited money"; and whether he has or not, he certainly cannot properly say that he "has money in the bank." What he does have is the bank's promise to pay money on demand. The bank owes him money. When a private person owes money, the creditor never thinks of saying that he has it on deposit in the debtor's pocket. 40 THE PURCHASING POWER OF MONEY [Chap. Ill §2 It cannot be too strongly emphasized that, in any balance sheet, the value of the liabilities rests on that of the assets. The deposits of a bank are no excep- tion. We must not be misled by the fact that the cash assets may be less than the deposits. When the uninitiated first learn that the number of dol- lars which note holders and depositors have the right to draw out of a bank exceeds the number of dollars in the bank, they are apt to jump to the conclusion that there is nothing behind the notes or deposit liabil- ities. Yet behind all these obligations there is always, in the case of a solvent bank, full value ; if not actual dollars, at any rate dollars' worth of property. By no jugglery can the liabilities exceed the assets except in insolvency, and even in that case only nominally, for the true value of the liabiUties (''bad debts") will only equal the true value of the assets behind them. These assets, as already indicated, are largely the notes of merchants, although, so far as the theory of banking is concerned, they might be any property whatever. If they consisted in the ownership of real estate or other wealth in ''fee simple," so that the tangible wealth which property always represents were clearly evident, all mystery would disappear. But the effect would not be different. Instead of taking grain, machines, or steel ingots on deposit, in exchange for the sums lent, banks prefer to take interest- bearing notes of corporations and individuals who own, directly or indirectly, grain, machines, and steel ingots : and by the banking laws the banks are even compelled to take the notes instead of the ingots. The bank finds itself with liabilities which exceed its cash assets ; Sec. 2] DEPOSIT CURRENCY 41 but in either case the excess of habilities is balanced by the possession of other assets than cash. These other assets of the bank are usually liabilities of business men. These liabilities are in turn supported by the assets of the business men. If we continue to follow up the ultimate basis of the bank's liabilities we shall find it in the visible tangible wealth of the world. This ultimate basis of the entire credit structure is kept out of sight, but the basis exists. Indeed, we may say that banking, in a sense, causes this visible, tangible wealth to circulate. If the acres of a landowner or the iron stoves of a stove dealer cannot circulate in [y literally the same way that gold dollars circulate, yet the landowner or stove dealer may give to the bank a note on which the banker may base bank notes or de- posits ; and these bank notes and deposits will circulate like gold dollars. Through banking, he who possesses wealth difficult to exchange can create a circulating medium. He has only to give to a bank his note — for which, of course, his property is liable — get in return the right to draw, and lo ! his comparatively unexchange- able wealth becomes liquid currency. To put it crudely, banking is a device for coining into dollars land, stoves, and other wealth not otherwise generally exchangeable. It is interesting to observe that the formation of the great modern ''trusts" has given a considerable impetus to deposit currency ; for the securities of large corporations are more easily used as ''collateral se- curity" for bank loans than the stocks and bonds of small corporations or than partnership rights. We began by regarding a bank as substantially a cooperative enterprise, run for the convenience and at the expense of its depositors. But, as soon as it reaches the point of lending money to X, Y , and Z, on time, 42 THE PURCHASING POWER OF MONEY [Chap. Ill while itself owing money on demand, it assumes toward X, Y, and Z and its cash depositors risks which the depositors would be unwilling to assume. To meet this situation, the responsibility and expense of running the bank are taken by a third class of people — stockholders — who are willing to assume the aug- mented risk for the sake of the chance of profit. Stock- holders, in order to guarantee the depositors against loss, put in some cash of their own. Their contract is, in effect, to make good any loss to depositors. Let us sup- pose that the stockholders put in $50,000, viz. S40,000 in cash and $10,000 in the purchase of a bank building. The accounts now stand : — Assets Liabilities Cash $140,000 Loans 100,000 Building 10,000 $250,000 Due depositors . . $150,000 Due note holders . . 50,000 Due stockholders . 50,000 $250,000 The accounts as they now stand include the chief fea- tures of an ordinary modern bank, — a so-called "bank of deposit, issue, and discount." We have seen that the assets must be adequate to meet the liabilities. We now wish to point out that the form of the assets must be such as will insure meeting the lia- bilities promptly. Since the business of a bank is to fur- nish quickly available property (cash or credit) in place of the "slower" property of its depositors, it fails of its purpose when it is caught with insufficient cash. Yet it "makes money " partly by tying up its quick property, i.e. lending it out where it is less accessible. Its prob- Sec. 3] DEPOSIT CURRENCY 43 lem in policy is to tie up enough to increase its prop- erty, but not to tie up so much as to get tied up itself. So far as anything has yet been said to the contrary, a bank might increase indefinitely its loans in relation to its cash or in relation to its capital. If this were so, deposit currency could be indefinitely inflated. There are limits, however, imposed by prudence and sound economic policy, on both these processes. In- solvency and insufficiency of cash must both be avoided. Insolvency is that condition which threatens when loans are extended with insufficient capital. Insufficiency of cash is that condition which threatens when loans are extended unduly relatively to cash. Insolvency is reached when assets no longer cover liabilities (to others than stockholders), so that the bank is unable to pay its debts. Insufficienc}'^ of cash is reached when, although the bank's total assets are fully equal to its liabilities, the actual cash on hand is insufficient to meet the needs of the instant, and the bank is unable to pay its debts on demand. The less the ratio of the value of the stockholders' interests to the value of liabilities to others, the greater is the risk of insolvency; the risk of insufficiency of cash is the greater, the less the ratio of the cash to the demand liabilities. In other words, the leading safe- guard against insolvency lies in a large capital and sur- plus, but the leading safeguard against insufficiency of cash lies in a large cash reserve. Insolvency proper may befall any business enterprise ; insufficiency of cash relates especially to banks in their function of redeeming notes and deposits. Let us illustrate insufficiency of cash. In our bank's accounts as we left them, there was a reserve of $140,000 of cash, and $200,000 of demand liabilities (deposits 44 THE PURCHASING POWER OF MONEY [Chap. Ill and notes). The managers of the bank may think this reserve of $140,000 unnecessarily large or the loans unnecessarily small. They may then extend their loans (extended to customers in the form of cash, notes or deposit accounts) until the cash reserve is reduced, say to $40,000, and the liabihties due depositors and note holders increased to $300,000. If, under these cir- cumstances, some depositor or note holder demands $50,000 cash, immediate payment will be impossible. It is true that the assets still equal the liabilities. There is full value behind the $50,000 demanded ; but the understanding was that depositors and note holders should be paid in money and on demand. Were this not a stipulation of the deposit contract, the bank might pay the claims thus made upon it by transfer- ring to its creditors the promissory notes due it from its debtors ; or it might ask the customers to wait until it could turn these securities into cash.f. Since a bank cannot follow either of these plans, it tries, where insufficiency of cash impends, to forestall this condition by "calling in" some of its loans, or if none can be called in, by selling some of its securities or other property for cash. But it happens unfortunately that there is a limit to the amount of cash which a bank can suddenly realize. No bank could escape failure if a large percentage of its note holders and depositors should simultaneously demand cash payment.- The paradox of a panic is well expressed by the case of the man who inquired of his bank whether it had cash available for paying the amount of his deposit, saying, "If you can pay me, I don't want it ; but if you can't, ^ See Irving Fisher, The Nature of Capital and' Income, Chapter V. 2Cf. Ricardo, Works, 2d ed., London (Murray), 1852, p. 217 {Principles of Political Economy and Taxation, Chapter XXVII). Sec. 3] DEPOSIT CURRENCY 45 I do." Such was the situation in 1907 in Wall Street. All the depositors at one time wanted to be sure their money ' ' was there. ' ' Yet it never is there all at one time. Since, then, insufficiency of cash is so troublesome a condition, — so difficult to escape when it has arrived, and so difficult to forestall when it begins to ap- proach, — a bank must so regulate its loans and note issues as to keep on hand a sufficient cash reserve, and thus prevent insufficiency of cash from even threaten- ing. It can regulate the reserve by alternately selling securities for cash and loaning cash on securities. The more the loans in proportion to the cash on hand, the greater the profits, but the greater the danger also. In the long run a bank maintains its necessary reserve by means of adjusting the interest rate charged for loans. If it has few loans and a reserve large enough to support loans of much greater volume, it will endeavor to extend its loans by lowering the rate of interest. If its loans are large and it fears too great demands on the reserve, it will restrict the loans by a high interest charge. Thus, by alternately raising and lowering in- terest, a bank keeps its loans within the sum which the reserve can support, but endeavors to keep them (for the sake of profit) as high as the reserve will support. If the sums owed to individual depositors are large, relatively to the total liabilities, the reserve should be proportionately large, since the action of a small num- ber of depositors can deplete it rapidly.^ Similarly, the reserves should be larger against fluctuating deposits (as of stock brokers) or those known to be temporary.^ ' Victor Morawetz, The Banking and Currency Problem in the United States, New York (The North American Review Publish- ing Co.), 1909, pp. 36 and 37. Also Kem merer, Money and Prices, 1909, p. 80. 2 Ibid. 46 THE PURCHASING POWER OF MONEY [Chap. Ill The reserve in a large city of great bank activity needs to be greater in proportion to its demand liabilities than in a small town with infrequent banking transactions. Experience dictates differently the average size of deposit accounts for different banks according to the general character and amount of their business. For every bank there is a normal ratio, and hence for a whole community there is also a normal ratio — an average of the ratios for the different banks. No absolute numerical rule can be given. Arbitrary rules are often imposed by law. National banks in the United States, for instance, are required to keep a re- serve for their deposits, varying according as they are or are not situated in certain cities designated by law as ''reserve" cities, i.e. cities where national banks hold deposits of banks elsewhere. These reserves are all in defense of deposits. In defense of notes, on the other hand, no cash reserve is required, — that is, of national banks. True, the same economic princi- ples apply to both bank notes and deposits, but the law treats them differently. The government itself chooses to undertake to redeem the national bank notes on demand. The state banks are subject to varying restrictions.^ Thus the requirement as to the ratio of reserve to de- posits varies from 12| per cent to 22^ per cent, being usually between 15 per cent and 20 per cent. Of the reserve, the part which must be cash varies from 10 per cent (of the reserve) to 50 per cent, being usually 40 per cent. Such legal regulation of banking reserves, however, * "Digest of State Banking Statutes," in Reports of the National Monetary Commission, 61st Congress, 2d Session Senate Document, No. 353. Sec. 4] DEPOSIT CURRENCY 47 is not a necessary development of banking. In Can- ada, the law makes the notes practically coordinate with the deposits. Indeed, banking may exist with- out government regulations at all. ''George Smith's money " furnishes an illustration. George Smith, Alex- ander Mitchell and others established in 1839 an Insurance Company which, though forbidden to exer- cise "banking privileges," issued certificates of deposit payable to bearer, and these certificates were actually circulated like bank notes. ^ §4 The study of banking operations, then, discloses two species of currency : one, bank notes, belonging to the category of money; and the other, deposits, be- longing outside of that category, but constituting an excellent substitute. Referring these to the larger category of goods, we have a threefold classification of goods : first, money ; second, deposit currency, or simply deposits ; and third, all other goods. And by the use of these, there are six possible types of ex- change : — (1) Money against money, (2) Deposits against deposits, (3) Goods against goods, (4) Money against deposits, (5) Money against goods, (6) Deposits against goods. For our purpose, only the last two types of exchange are important, for these constitute the circulation of currency. As regards the other four, the first and third have been previously explained as "money changing" and "barter" respectively. The second and fourth * See Horace White, Money and Banking 48 THE PURCHASING POWER OF MONEY [Chap. Ill are banking transactions : the second being such as the selling of drafts for checks, or the mutual cancel- lation of bank clearings; and the fourth being such operations as the depositing or the withdrawing of money, by depositing cash or cashing checks. The analysis of the balance sheets of banks has pre- pared us for the inclusion of bank deposits or circulating credit in the equation of exchange. We shall still use M to express the quantity of actual money, and V to express the velocity of its circulation. Similarly, we shall now use M' to express the total deposits subject to transfer by check ; and 7' to express the average velocity of circulation. The total value of purchases in a year is therefore no longer to be measured by MV, but by MV + M'V. The equation of exchange, there- fore, becomes : — MV + M'V = 2pQ = PT} Let us again represent the equation of exchange by means of a mechanical picture. In Figure 4, trade, ^Tr^H.M.^„.,|,„,j,.,n_„in,..,j_.„,|,„,||i,.jiii,jii,ij.|.,j,M,j^ Fig. 4. as before, is represented on the right by the weight of a miscellaneous assortment of goods ; and their average price by the distance to the right from the fulcrum, or 1 The equation of exchange is also stated by Kemmerer, Money and Credit Instruments in their Relation to General Prices, so as to inelude bank credit, although in a somewhat different way. That credit acts on prices in the same manner as money is by no means a newly established principle. See, for example, MiU, Principles oj Political Economy, Book III, Chapter XII, §§ 1, 2. Sec. 5] DEPOSIT CURRENCY 49 the length of the arm on which this weight hangs. Again at the left, money (ikf ) is represented by a weight in the form of a purse, and its velocity of circulation ( V) by its arm ; but now we have a new weight at the left, in the form of a bank book, to represent the bank de- posits {M'). The velocity of circulation (V) of these bank deposits is represented by its distance from the fulcrum or the arm at which the book hangs. This mechanism makes clear the fact that the average price (right arm) increases with the increase of money or bank deposits and with the velocities of their circulation, and decreases with the increase in the volume of trade. Recurring to the left side of the equation of exchange, OT MV + M'V, we see that in a community without bank deposits the left side of the equation reduces simply to MV, the formula used in Chapter II ; for in such a community the term " M'V " vanishes. The introduction of M' tends to raise prices. That is, the hanging of the bank book on the left requires a lengthen- ing of the arm at the right. Just as E was used to denote the total circulation of money, MV, so we may now use E' to denote the total circulation of deposits, M'V. Like E, M, and V, so also E', M', and V are sums and averages of corresponding magnitudes pertaining to different parts of the year, or different persons.^ §5 With the extension of the equation of monetary circulation to include deposit cii'culation, the influence 1 The mathematical analysis of E', M', and V in terms of "arrays" of e"s, m"s, and t;"s, etc., is precisely parallel to that of E, M, and V, given in the Appendix to Chapter II. See also §§1 and 2 of Appendix to (this) Chapter III. 50 THE PURCHASING POWER OF MONEY [Chap. IH exerted by the quantity of money on general prices becomes less direct ; and the process of tracing this influence becomes more difficult and complicated. It has even been argued that this interposition of circulating credit breaks whatever connection there may be be- tween prices and the quantity of money. ^ This would be true if circulating credit were independent of money. But the fact is that the quantity of circulating credit, M' , tends to hold a definite relation to M, the quantity of money in circulation; that is, deposits are normally a more or less definite multiple of money. Two facts normally give deposits a more or less definite ratio to money. The first has been already explained, viz. that bank reserves are kept in a more or less definite ratio to bank deposits. The second is that individuals, firms, and corporations preserve more or less definite ratios between their cash trans- actions and their check transactions, and also between their money and deposit balances.^ These ratios are determined by motives of individual convenience and habit. In general, business firms use money for wage payments, and for small miscellaneous transactions included under the term ''petty cash"; while for settlements with each other they usually prefer checks. These preferences are so strong that we could not imagine them overridden except tempora- 1 An almost opposite view is that of Laughlin that normal credit cannot affect prices because it is not an offer of standard money and cannot affect the value of the standard which alone determines general prices. See the Principles of Money, New York (Seribner), 1903, p. 97. Both views are inconsistent with that upheld in this book. 2 This fact is apparently overlooked by Laughlin when he argues that there is not "any reason for limiting the amount of the deposit currency, or the assumption of an absolute scarcity of specie re- serves." See Principles of Money, p. 127. Sec. 5] DEPOSIT CURRENCY 51 rily and to a small degree. A business firm would hardly pay car fares with checks and liquidate its large liabilities with cash. Each person strikes an equilibrium between his use of the two methods of payment, and does not greatly disturb it except for short periods of time. He keeps his stock of money or his bank balance in constant adjustment to the payments he makes in money or by check. Whenever his stock of money becomes relatively small and his bank balance relatively large, he cashes a check. In the opposite event, he deposits cash. In this way he is constantly converting one of the two media of exchange into the other. A private individual usually feeds his purse from his bank account ; a retail com- mercial firm usually feeds its bank account from its till. The bank acts as intermediary for both. In a given community the quantitative relation of deposit currency^ to money is determined by several considerations of convenience. In the first place, the more highly developed the business of a community, the more prevalent the use of checks. Where business is conducted on a large scale, merchants habitually transact their larger operations with each other by means of checks, and their smaller ones by means of cash. Again, the more concentrated the population, the more prevalent the use of checks. In cities it is more convenient both for the payer and the payee to make large payments by check ; whereas, in the coun- try, trips to a bank are too expensive in time and effort to be convenient, and therefore more money is used in proportion to the amount of business done.^ Again, 1 The convenient expression "deposit currency" is used by Laughlin, The Principles of Money, p. 118. * See Kinley's "Credit Instruments," Report of the National 52 THE PURCHASING POWER OF MONEY [Chap. Ill the wealthier the members of the community, the more largely will they use checks. Laborers seldom use them; but capitalists, professional and salaried men use them habitually, for personal as well as business transactions. There is, then, a relation of convenience and custom between check and cash circulation, and a more or less stable ratio between the deposit balance of the average man or corporation and the stock of money kept in pocket or till. This fact, as applied to the country as a whole, means that by convenience a rough ratio is fixed between M and M'. If that ratio is disturbed temporarily, there will come into play a tendency to restore it. Individuals will deposit surplus cash, or they will cash surplus deposits. Hence, both money in circulation (as shown above) and money in reserve (as shown previously) tend to keep in a fixed ratio to deposits. It follows that the two must be in a fixed ratio to each other. It further follows that any change in M, the quantity of money in circulation, requiring as it normally does a proportional change in M', the volume of bank de- posits subject to check, will result in an exactly pro- portional change in the general level of prices except, of course, so far as this effect be interfered with by concomitant changes in the F's or the Q's. The truth of this proposition is evident from the equation MV + M'V =^pQ; for if, say, M and M' are doubled, while V and F' remain the same, the left side of the equation is doubled and therefore the right side must be doubled also. But if the Q's remain Monetary Commission, Senate Document, 399, 61st Congress, 2d Session, 1910, p. 188. Sec. 6] DEPOSIT CURRENCY 53 unchanged, then evidently all the p's must be doubled, or else if some are less than doubled, others must be enough more than doubled to compensate. §6 The contents of this chapter may be formulated in a few simple propositions : — (1) Banks supply two kinds of currency, viz. bank notes — which are money ; and bank deposits (or rights to draw) — which are not money. (2) A bank check is merely a certificate of a right to draw. (3) Behind the claims of depositors and note holders stand, not simply the cash reserve, but all the assets of the bank. (4) Deposit banking is a device by which wealth, incapable of direct circulation, may be made the basis of the circulation of rights to draw. (5) The basis of such circulating rights to draw or deposits must consist in part of actual money, and it should consist in part also of quick assets readily ex- changeable for money. (6) Six sorts of exchange exist among the three classes of goods, money, deposits, and other goods. Of these six sorts of exchange, the most important for our pres- ent purposes are the exchanges of money and deposits against goods. (7) The equation of money circulation extended so as to include bank deposits reads thus : — MV + M'V = ^pQ or PT. (8) There tends to be a normal ratio of bank deposits (ilf' ) to the quantity of money {M) ; because business C^ 54 THE PURCHASING POWER OF MONEY [Chap. Ill convenience dictates that the available currency shall be apportioned between deposits and money in a cer- tain more or less definite, even though elastic, ratio. (9) The inclusion of deposit currency does not nor- mally disturb the quantitative relation between money and prices. CHAPTER IV DISTURBANCE OF EQUATION AND OF PURCHASING POWER DURING TRANSITION PERIODS In the last chapter it was shown that the quantity of bank deposits normally maintains a definite ratio to the quantity of money in circulation and to the amount of bank reserves. As long as this normal relation holds, the existence of bank deposits merely magnifies the effect on the level of prices produced by the quantity of money in circulation and does not in the least distort that effect. Moreover, changes in velocity or trade will have the same effect on prices, whether bank deposits are included or not. But during periods of transition this relation between money (M) and deposits (M') is by no means rigid. We are now ready to study these periods of transition. The change which constitutes a transition may be a change in the quantity of money, or in any other factor of the equation of exchange, or in all. Usually all are involved, but the chief factor which we shall select for study (together with its effect on the other factors) is quantity of money. If the quantity of money were suddenly doubled, the effect of the change would not be the same at first as later. The ultimate effect is, as we have seen, to double prices ; but before this happens, the prices oscillate up and down. In this chapter we shall consider the temporary effects during 55 56 THE PURCHASING POWER OF MONEY [Chap. IV the period of transition separately from the permanent or ultimate effects which were considered in the last chapter. These permanent or ultimate effects follow after a new equilibrium is established, — if, indeed, such a condition as equilibrium may be said ever to be established. What we are concerned with in this chapter is the temporary effects, i.e. those in the tran- sition period. The transition periods may be characterized either by rising prices or by falling prices. Rising prices must be clearly distinguished from high prices, Suiid falling from low. With stationary levels, high or low, we have in this chapter nothing to do. Our concern is with ris- ing or falling prices. Rising prices mark the transition between a low and a high level of prices, just as a hill marks the transition between flat lowlands and flat highlands. Since the study of these acclivities and declivities is bound up with that of the adjustment of interest rates, our first task is to present a brief statement regarding the effects of rising and falling prices^ on the rate of interest. Indeed, the chief object of this chapter is to show that the peculiar behavior of the rate of interest during transition periods is largely responsible for the crises and depressions in which price movements end. It must be borne in mind that although business loans are made in the form of money, yet whenever a man borrows money, he does not do this in order to hoard the money, but to purchase goods with it. To all intents and purposes, therefore, when A borrows one hundred dollars from B in order to purchase, say, one ' For a fuller statement, see Irving Fisher, The Rate of Interest, New York (Macmillan), 1907, Chapters V, XIV. Sec. 1] TRANSITION PERIODS 57 hundred units of a given commodity at one dollar per unit, it may be said that B is virtually lending A one hundred units of that commodity. And if at the end of a year A returns one hundred dollars to B, but the price of the commodity has meanwhile advanced, then B has lost a fraction of the purchasing power originally loaned to A. For even though A should happen to return to B the identical coins in which the loan was made, these coins represent somewhat less than the original quantity of purchasable commodities. Bear- ing this in mind in our investigation of interest rates, let us suppose that prices are rising at the rate of 3 per cent each year. It is plain that the man who lends $100 at the beginning of the year must, in order to get 5 per cent interest in purchasing power, receive back both $103 (then the equivalent of the $100 lent) plus 5 per cent of this, or a total of $108.15. That is, in order to get 5 per cent interest in actual purchasing power, he must receive a little more than 8 per cent interest in money. The 3 per cent rise of prices thus ought to add approximately 3 per cent to the rate of interest. Rising prices, therefore, in order that the relations between creditor and debtor shall be the same during the rise as before and after, require higher money interest than stationary prices require. Not only will lenders require, but borrowers can afford to pay higher interest in terms of money ; and to some extent competition will gradually force them to do so.' Yet we are so accustomed in our business dealings to consider money as the one thing stable, — to think of a "dollar as a dollar" regardless of the passage of time, that we reluctantly yield to this process of readjustment, thus rendering it very slow and imperfect. When prices ^ Rate o/ Interest, Chapter XIV. 58 THE PURCHASING POWER OF MONEY [Chap. IV are rising at the rate of 3 per cent a year, and the nor- mal rate of interest — i.e. the rate which would exist were prices stationary — is 5 per cent, the actual rate, though it ought (in order to make up for the rising prices) to be 8.15 per cent, will not ordinarily reach that figure; but it may reach, say, 6 per cent, and later, 7 per cent. This inadequacy and tardiness of adjustment are fostered, moreover, by law and custom, which arbitrarily tend to keep down the rate of interest. A similar inadequacy of adjustment is observed when prices are falling. Suppose that, by the end of a year, $97 will buy as much as $100 at the beginning. In that case the lender, in order to get back a purchasing power equivalent to his principal and 5 per cent interest, should get, not $105, but only $97 + 5 per cent of $97 or $101.85. Thus the rate of interest in money should in this case be 1.85 per cent, or less than 2 per cent, instead of the original 5 per cent. In other words, the 3 per cent fall of prices should reduce the rate of interest by approximately 3 per cent. But as a matter of fact, such a perfect adjustment is seldom reached, and money interest keeps far above 2 per cent for a considerable time.^ §2 We are now ready to study temporary or transitional changes in the factors of our equation of exchange. Let us begin by assuming a slight initial disturbance, such as would be produced, for instance, by an increase in the quantity of gold. This, through the equation of ex- change, will cause a rise in prices. As prices rise, profits of business men, measured in money, will rise also, even if the costs of business were to rise in the same * Rate of Interest, loc. cit. Sec. 2] TRANSITION PERIODS 59 proportion. Thus, if a man who sold $10,000 of goods at a cost of $6000, thus clearing $4000, could get double prices at double cost, his profit would be double also, being $20,000 - $12,000, which is $8000. Of course such a rise of prices would be purely nominal, as it would merely keep pace with the rise in price level. The business man would gain no advantage, for his larger money profits would buy no more than his former smaller money profits bought before. But, as a matter of fact, the business man's profits will rise more than this because the rate of interest he has to pay will not ad- just itself immediately. Among his costs is interest, and this cost will not, at first, rise. Thus the profits will rise faster than prices. Consequently, he will find him- self making greater profits than usual, and be en- couraged to expand his business by increasing his bor- rowings. These borrowings are mostly in the form of short-time loans from banks; and, as we have seen, short-time loans engender deposits. As is well known, the correspondence between loans and deposits is re- markably exact. ^ Therefore, deposit currency (M') will increase, but this extension of deposit currency tends further to raise the general level of prices, just as the in- crease of gold raised it in the first place.^ Hence prices, which were already outstripping the rate of interest, tend to outstrip it still further, enabling borrowers, ^ See J. Pease Norton, Statistical Studies in the New York Money Market (Macmillan), 1902, chart at end. * See article by Knut Wicksell in the Jahrbucher fur National- ohoiomie, 1897 (Band 68), pp. 228-243, entitled "Der Bankzins als Rej^ulator der Warenpreise." This article, while not dealing di- rectly with credit cycles as related to panics, points out the con- nection between the rate of interest on bank loans and changes ii Ithe level of prices due to the resulting expansion and contraction of^uch loans. 60 THE PURCHASING POWER OF MONEY [Chap. IV who were already increasing their profits, to increase them still further. More loans are demanded, and although nominal interest may be forced up some- what, still it keeps lagging below the normal level. Yet nominally the rate of interest has increased; and hence the lenders, too, including banks, are led to become more enterprising. Beguiled by the higher nominal rates into the belief that fairly high interest is being realized, they extend their loans, and with the resulting expansion of bank loans, deposit currency (M'), already expanded, expands still more. Also, if prices are rising, the money value of collateral may be greater, making it easier for borrowers to get large credit.^ Hence prices rise still further.^ This sequence of events may be briefly stated as follows : — 1. Prices rise (whatever the first cause may be; but we have chosen for illustration an increase in the amount of gold). 2. The rate of interest rises, but not sufficiently. 3. Enterprisers (to use Professor Fetter's term), en- couraged by large profits, expand their loans. 4. Deposit currency (M') expands relatively to money (M). 5. Prices continue to rise, that is, phenomenoin No. 1 is repeated. Then No. 2 is repeated, and so on. In other words, a slight initial rise of prices sets in motion a train of events which tends to repeat itself. Rise of prices generates rise of prices, and continues to do so as long as the interest rate lags behind its normal figure. 1 See Kinley, Money, New York (Maomillan), 1904, p. 223. * See Wicksell, op. cit. I 'i Sec. 3J TRANSITION PERIODS 61 §3 The expansion of deposit currency indicated in this cumulative movement abnormally increases the ratio of M' to M. This is evident if the rise of prices begins in a change in some element or elements in the equation other than the quantity of money ; for if M remains constant and M' increases, the ratio M' to M must in- crease also. If M increases in any ratio, M' will increase in a greater ratio. If it increased only in the same ratio, prices would increase in that ratio (assuming velocities and quantities unchanged); and if prices increased in that ratio, loans (which being made to buy goods must be adjusted to the prices of goods) would have to be increased in that ratio in order to secure merely the same goods as before. But enterprisers, wishing to profit by the lag in interest, would extend the loans beyond this old or original point. Therefore, deposits based on loans would increase in a greater ratio. That is, the ratio M' to M would increase. In other words, during the period while M is increasing, M' increases still faster, thus disturbing the normal ratio between these two forms of currency. This, however, is not the only disturbance caused by the increase in M. There are disturbances in the Q's (or in other words T) in V, and in V. These will be taken up in order. Trade (the Q's) will be stimulated by the easy terms for loans. This effect is always observed during rising prices, and people note approvingl}^ that ''business is good" and "times are booming." Such statements represent the point of view of the ordinary business man who is an "enterpriser-borrower." They do not represent the sentiments of the creditor, the salaried man, or the laborer, most of whom are silent but 62 THE PURCHASING POWER OF MONEY [Chap. IV long-suffering, — paying higher prices, but not getting proportionally higher incomes. , The first cause of the unhealthy increase in trade lies / K in the fact that prices, like interest, lag behind their ' full adjustment and have to be pushed up, so to speak, by increased purchases. This is especially true in cases where the original impetus came from an increase in money. The surplus money is first expended at nearly the old price level, but its continued expenditure grad- ually raises prices. In the meantime the volume of purchases will be somewhat greater than it would have been had prices risen more promptly. In fact, from the point of view of those who are selling goods, it is the possibility of a greater volume of sales at the old prices which gives encouragement to an increase of prices. Seeing that they can find purchasers for more goods than before at the previously prevailing prices, or for as many goods as before at higher prices, they will charge these higher prices. But the amount of trade is dependent, almost en- tirely, on other things than the quantity of currency, so that an increase of currency cannot, even temporarily, very greatly increase trade. In ordinarily good times practically the whole community is engaged in labor, pro- ducing, transporting, and exchanging goods. The in- crease of currency of a "boom" period cannot, of itself, increase the population, extend invention, or increase the efficiency of labor. These factors pretty definitely limit the amount of trade which can be reasonably carried on. So, although the gains of the enterpriser- borrower may exert a psychological stimulus on trade, though a few unemployed may be employed, and some others in a few lines induced to work overtime, and al- though there may be some additional buying and selling Sec. 3] TEANSITION PERIODS 63 which is speculative, yet almost the entire effect of an increase of deposits must be seen in a change of prices. Normally the entire effect would so express itself, but transitionally there will be also some increase in the Q's. We next observe that the rise in prices — fall in the purchasing power of money — will accelerate the y^ circulation of money. We all hasten to get rid of any commodity which, like ripe fruit, is spoiling on our hands. ^ Money is no exception; when it is depreciat- ing, holders will get rid of it as fast as possible. As they view it, their motive is to buy goods which appre- ciate in terms of money in order to profit by the rise in their value. The inevitable result is that these goods «x rise in price still further. The series of changes, then, initiated by rising prices, expressed more fully than before, is as follows : — "^ 1. Prices rise. 2. Velocities of circulation (F and V) increase; the rate of interest rises, but not sufficiently. 3. Profits increase, loans expand, and the Q's in- crease. 4. Deposit currency {M') expands relatively to money (M). 5. Prices continue to rise; that is, phenomenon No. 1 is repeated. Then No. 2 is repeated, and so on. ^ It will be noticed that these changes now involve all * For statistical proof, see Pierre des Essars, Journal de la SociitS de Statistique de Paris, April, 1895, p. 143. The figures relate only to velocity of bank deposits. No corresponding figures for velocity of circulation of money exist. Pierre des Essars has shown that in European banks V reaches a maximum in crisis years almost without fail. The same I find true in this country as shown by the ratio of clearings to deposits in New York, Boston, and Phila- delphia. 64 THE PURCHASING POWER OF MONEY [Chap. IV the magnitudes in the equation of exchange. They are temporary changes, pertaining only to the transition period. They are Hke temporary increases in power and readjustments in an automobile climbing a hill. § 4 Evidently the expansion coming from this cycle of / causes cannot proceed forever. It must ultimately spend itself. The check upon its continued operation lies in the rate of interest. It was the tardiness of the rise in interest that was responsible for the abnormal condition. But the rise in interest, though belated, is progressive, and, as soon as it overtakes the rate of rise in prices, the whole situation is changed. If prices are rising at the rate of 2 per cent per annum, the boom will continue only until interest becomes 2 per cent higher. It then offsets the rate of rise in prices. The banks are forced in self-defense to raise interest be- cause they cannot stand so abnormal an expansion of loans relatively to reserves. As soon as the interest rate becomes adjusted, borrowers can no longer hope to make great profits, and the demand for loans ceases to expand. There are also other forces placing a limitation on i\ further expansion of deposit currency and introducing a tendency to contraction. Not only is the amount of deposit currency limited both by law and by prudence to a certain maximum multiple of the amount of bank reserves ; but bank reserves are themselves limited by the amount of money available for use as reserves. Further, with the rise of interest, the value of certain collateral securities, such as bonds, on the basis of which loans are made, begins to fall. Such securities, being worth the discounted value ofj fixed sums, fall Sec. 4] TRANSITION PERIODS 65 as interest rises ; and therefore they cannot be used as collateral for loans as large as before. This check to loans is, as previously explained, a check to deposits also. With the rise of interest, those who have counted on renewing their loans at the former rates and for the former amounts are unable to do so. It follows that some of them are destined to fail. The failure (or prospect of failure) of firms that have borrowed heavily from banks induces fear on the part of many depositors that the banks will not be able to realize on these loans. Hence the banks themselves fall under suspicion, and for this reason depositors demand cash. Then occur ''runs on the banks," which deplete the bank reserves at the very moment they are most needed.^ Being short of reserves, the banks have to curtail their loans. It is then that the rate of interest rises to a panic figure. Those enterprisers who are caught must have currency ^ to liquidate their obligations, and to get it are willing to pay high interest. Some of them are destined to become bankrupt, and, with their failure, the demand for loans is correspondingly reduced. This culmination of an upward price movement is what is called a crisis,^ — ' a condition characterized by bankruptcies, and the bank- * A part of the theory of crises here presented is similarly ex- plained in a paper by Harry G. Brown, Yale Review, August, 1910, entitled "Typical Commeroial Crises versus a Money Panic." 2 Irving Fisher, Rate of Interest, pp. 325, 326. ^ This is the definition of a crisis given by Juglar and the his- tory of crises which he gives in detail corresponds to the descrip- tion. See Juglar, Des Crises Commerciales et de leur retour periodique en France en Angleterre et aux Etats-Unis. 2d ed., Paris (Guillaumin), 1889, pp. 4 and 5. See also translation of part dealing with the United States, by De Courcey W. Thom, A Brie} History of Panics in the United States, New York (Putnam), 1893, pp. 7-10. ¥ 66 THE PURCHASING POWER OF MONEY [Chap. IV ruptcies being due to a lack of cash when it is most needed. It is generally recognized that the collapse of bank credit brought about by loss of confidence is the essential fact of every crisis, be the cause of the loss of confidence what it may. What is not generally recognized, and what it is desired in this chapter to emphasize, is that this loss of confidence (in the typical commercial crisis here described) is a consequence of a belated adjustment in the interest rate. It is not our purpose here to discuss nonmonetary causes of crises, further than to say that the monetary / causes are the most important when taken in connection with the maladjustments in the rate of interest. The other factors often emphasized are merely effects of this maladjustment. " Overconsumption " and ''over- investment ' ' are cases in point . The reason many people spend more than they can afford is that they are relying on the dollar as a stable unit when as a matter of fact its purchasing power is rapidly falling. The bond- holder, for instance, is beguiled into trenching on his capital. He never dreams that he ought to lay by a sinking fund because the decrease in purchasing power of money is reducing the real value of his principal. Again, the stockholder and enterpriser generally are beguiled by a vain reliance on the stability of the rate of interest, and so they overinvest. It is true that for a time they are gaining what the bondholder is losing and are therefore justified in both spending and in- vesting more than if prices were not rising ; and at first they prosper. But sooner or later the rate of interest rises above what they had reckoned on, and they awake to the fact that they have embarked on enterprises which cannot pay these high rates. 3ec. 5] TRANSITION PERIODS 67 Then a curious thing happens : borrowers, unable to get easy loans, blame the high rate of interest for conditions which were really due to the fact that the previous rate of interest was not high enough. Had the previous rate been high enough, the borrowers never would have overinvested. §5 The contraction of loans and deposits is accompanied by a decrease in velocities, and these conspire to pre- vent a further rise of prices and tend toward a fall. The crest of the wave is reached and a reaction sets in. Since prices have stopped rising, the rate of interest, which has risen to compensate the rise of prices, should fall again. But just as at first it was slow to rise, so now it is slow to fall. In fact, it tends for a time to rise still further. The mistakes of the past of overborrowing compel the unfortunate victims of these mistakes to borrow still further to protect their solvency. It is this special abnormality which marks the period as a ''crisis." Loans are wanted to continue old debts or to pay these debts by creating new ones. They are not wanted because of new investments but because of obligations connected with old (and ill-fated) investments. The problem is how to get extricated from the meshes of past commitments. It is the problem of liquidation. Even when interest begins to fall, it falls slowly, and failures continue to occur. Borrowers now find that interest, though nominally low, is still hard to meet. Especially do they find this true in the case of contracts made just before prices ceased rising or just before they began to fall. The rate of interest in these cases is agreed upon before the change in conditions takes place. 68 THE PURCHASING POWER OF MONEY [Chap. IV There will, in consequence, be little if any adjustment in lowering nominal interest. Because interest is hard to pay, failures continue to occur. There comes to be a greater hesitation in lending on any but the best security, and a hesitation to borrow save when the prospects of success are the greatest. Bank loans tend to be low, and consequently deposits {M') are reduced. The contraction of deposit currency makes prices fall still more. Those who have borrowed for the purpose of buying stocks of goods now find they cannot sell them for enough even to pay back what they have borrowed. Owing to this tardiness of the interest rate in falling to a lower and a normal level, the sequence of events is now the opposite of what it was before : — 1. Prices fall. 2. The rate of interest falls, but not sufficiently. 3 . Enterpriser-borrowers, discouraged by small profits, contract their borrowings. 4. Deposit currency {M') contracts relatively to money (M). 5. Prices continue to fall; that is, phenomenon No. 1 is repeated. Then No. 2 is repeated, and so on. Thus a fall of prices generates a further fall of prices. The cycle evidently repeats itself as long as the rate of interest lags behind. The man who loses most is the business man in debt. He is the typical business man, and he now complains that ''business is bad." There is a "depression of trade." During this depression, velocities (V and V) are abnormally low. People are less hasty to spend money or checks when the dollars they represent are rising in purchasing power. The Q's (or quantities in trade) decline because (1) the initiators of trade — the enter- priser-borrowers — are discouraged ; (2) the inertia of Sec. 5] TRANSITION PERIODS 69 high prices can be overcome only by a falling off of expenditures ; (3) trade against money which alone the Q's represent gives way somewhat to barter. For a time there is not enough money to do the business which has to be done at existing prices, for these prices are still high and will not immediately adjust them- selves to the sudden contraction. When such a ' ' money famine" exists, there is no way of doing all the business except by eking out money transactions with barter. But while recourse to barter eases the first fall of prices, the inconvenience of barter immediately begins to operate as an additional force tending to reduce prices by inducing sellers to sell at a sacrifice if only money can be secured and barter avoided ; although this ef- fect is partly neutralized for a time by a decrease in the amount of business which people will attempt under such adverse conditions. A statement includ- ing these factors is : — 1. Prices fall. 2. Velocities of circulation (F and V) fall; the rate of interest falls, but not sufficiently. 3. Profits decrease; loans and the Q's decrease. 4. Deposit currency (M') contracts relatively to money (M). Prices continue to fall ; that is, phenomenon No. 1 is repeated. Then No. 2 is repeated, and so on. The contraction brought about by this cycle of causes becomes self-Hmiting as soon as the rate of interest overtakes the rate of fall in prices. After a time, normal conditions begin to return. The weakest producers have been forced out, or have at least been prevented from expanding their business by increased loans. The strongest firms are left to build up a new credit structure. The continuous fall of prices has 70 THE PURCHASING POWER OP .VIONEY [Chap. IV made it impossible for most borrowers to pay the old high rates of interest ; the demand for loans diminishes, and interest falls to a point such that borrowers can at last pay it. Borrowers again become willing to take ventures ; failures decrease in number ; bank loans cease to decrease; prices cease to fall; borrowing and carrying on business become profitable ; loans are again demanded ; prices again begin to rise, and there occurs a repetition of the upward movement already described. We have considered the rise, culmination, fall, and recovery of prices. These changes are abnormal oscillations, due to some initial disturbance. The up- ward and downward movements taken together con- stitute a complete credit cycle, which resembles the forward and backward movements of a pendulum.^ In most cases the time occupied by the swing of the commercial pendulum to and fro is about ten years. While the pendulum is continually seeking a stable position, practically there is almost always some oc- currence to prevent perfect equilibrium. Oscillations are set up which, though tending to be self-corrective, are continually perpetuated by fresh disturbances. \ Any cause which disturbs equilibrium will suffice to I set up oscillations. One of the most common of such causes is an increase in the quantity of money. ^ An- other is a shock to business confidence (affecting enter- prise, loans, and deposits). A third is short crops, affecting the Q's. A fourth is invention. The factors in the equation of exchange are there- ' For a mathematical treatment of this analogy, see Pareto, Cours d'Sconomie politique, Lausanne, 1897, pp. 282-284. * Such would seem to be the explanation of the panic of 1907. Cf. Irving Fisher, Rate of Interest, p. 336. Sec. 5] TRANSITION PERIODS 71 fore continually seeking normal adjustment. A ship in a calm sea will ''pitch" only a few times before coming to rest, but in a high sea the pitching never ceases. While continually seeking equiUbrium, the ship continually encounters causes which accentuate the oscillation. The factors seeking mutual adjustment are money in circulation, deposits, their velocities, the Q's and the p's. These magnitudes must always be linked together by the equation MV + M'V = 2pQ. This represents the mechanism of exchange. But in order to conform to such a relation the displacement of any one part of the mechanism spreads its effects during the transition period over all parts. Since periods of transition are the rule and those of equi- librium the exception, the mechanism of exchange is almost always in a dynamic rather than a static con- dition. It must not be assumed that every credit cycle is so marked as to produce artificially excessive business ac- tivity at one time and "hard times" at another. The rhythm may be more or less extreme in the width of its fluctuations. If banks are conservative in making loans during the periods of rising prices, and the expansion of credit currency is therefore limited, the rise of prices is likewise limited, and the succeeding fall is apt to be less and to take place more gradually. If there were a bet- ter appreciation of the meaning of changes in the price level and an endeavor to balance these changes by ad- justment in the rate of interest, the oscillations might be very greatly mitigated. It is the lagging behind of the rate of interest which allows the oscillations to reach so great proportions. On this point Marshall well says : " The cause of alternating periods of in- flation and depression of commercial activity ... is 72 THE PURCHASING POWER OF MONEY [Chap. IV intimately connected with those variations in the real rate of interest which are caused by changes in the purchasing power of money. For when prices are likely to rise, people rush to borrow money and buy goods, and thus help prices to rise ; business is inflated, and is managed recklessly and wastefully ; those working on borrowed capital pay back less real value than they borrowed, and enrich themselves at the expense of the community. When afterwards credit is shaken and prices begin to fall, every one wants to get rid of commodities which are falling in value and to get hold of money which is rapidly rising; this makes prices fall all the faster, and the further fall makes credit shrink even more, and thus for a long time prices fall because prices have fallen." ^ A somewhat different sort of cycle is the seasonal fluctuation which occurs annually. Such fluctuations, for the most part, are due, not to the departure from a state of equilibrium, but rather to a continuous adjust- ment to conditions, which, though changing, are normal and expected. As the autumn periods of harvesting and crop moving approach, there is a tendency toward a lower level of prices, followed after the passing of this period and the approach of winter by a rise of prices. § 6 In the present chapter we have analyzed the phe- nomena characteristic of periods of transition. We have found that one such "boom" period leads to a reaction, and that the action and reaction complete a cycle of "prosperity" and "depression." It has been seen that rising prices tend towards a ^ Marshall, Principles of Economics, Sth ed., London (Macmillan), 1907, Vol. I, p. 594. Sec. 6] TRANSITION PERIODS 73 higher nominal interest, and f alUng prices tend towards a lower, but that in general the adjustment is incomplete. With any initial rise of prices comes an expansion of loans, owing to the fact that interest does not at once adjust itself. This produces profits for the enterpriser-bor- rower, and his demand for loans further extends de- posit currency. This extension still further raises prices, a result accentuated by a rise in velocities though somewhat mitigated by an increase in trade. When interest has become adjusted to rising prices, and loans and deposits have reached the limit set for them by the bank reserves and other conditions, the fact that prices no longer are rising necessitates a new adjustment. Those whose business has been unduly extended now find the high rates of interest oppressive. Failures result, constituting a commercial crisis. A reaction sets in ; a reverse movement is initiated. A fall of prices, once begun, tends to be accelerated for reasons exactly corresponding to those which operate in the opposite situation. CHAPTER V INDIRECT INFLUENCES ON PURCHASING POWER § 1 Thus far we have considered the level of prices as affected by the volume of trade, by the velocities of circulation of money and of deposits, and by the quan- tities of money and of deposits. These are the only influences which can directly affect the level of prices. Any other influences on prices must act through these five. There are myriads of such influences (outside of the equation of exchange) that affect prices through these five. It is our purpose in this chapter to note the chief among them, excepting those that affect the volume of money {M) ; the latter will be examined in the two following chapters. We shall first consider the outside influences that affect the volume of trade and, through it, the price level. The conditions which determine the extent of trade are numerous and technical. The most important may be classified as follows : — 1. Conditions affecting producers. (a) Geographical differences in natural resources. (6) The division of labor. (c) Knowledge of the technique of production. (d) The accumulation of capital. 2. Conditions affecting consumers. (a) The extent and variety of human wants. 74 Sec. 1] INDIRECT INFLUENCES 75 3. Conditions connecting producers and consumers. (a) Facilities for transportation. (b) Relative freedom of trade. (c) Character of monetary and banking systems. (d) Business confidence. 1 (a). It is evident that if all localities were exactly alike in their natural resources and in their compara- tive costs of production little or no trade would be set up between them. It is equally true that the greater the difference in the costs of production of different arti- cles in different localities, the more likely is there to be trade between them and the greater the amount of that trade. Primitive trade had its raison d'etre in the fact that the regions of this earth are unlike in their prod- ucts. The traders were travelers between distant coun- tries. Changes in commercial geography still produce changes in the distribution and volume of trade. The exhaustion of gold and silver mines in Nevada and of lumber in Michigan have tended to reduce the volume of trade of these regions, both external and internal. Contrariwise, cattle raising in Texas, the production of coal in Pennsylvania, of oranges in Florida, and of ap- ples in Oregon have increased the volume of trade for these communities respectively. 1 (6). Equally obvious is the influence of the division of labor. Division of labor is based in part on differences in comparative costs or efforts as between men, — cor- responding to geographic differences as between coun- tries. These two, combined, lead to local differentia- tion of labor, making, for example, the town of Sheffield famous for cutlery, Dresden for china, Venice for glass, Paterson for silks, and Pittsburg for steel. 1 (c). Besides local and personal differentiation, the state of knowledge of production will affect trade. 76 THE PURCHASING POWER OF MONEY [Chap. V The mines of Africa and Australia were left unworked for centuries by ignorant natives but were opened by white men possessing a knowledge of metallurgy. Vast coal fields in China await development, largely for lack of knowledge of how to extract and market the coal. Egypt awaits the advent of scientific agriculture, to usher in trade expansion. Nowadays, trade schools in Germany, England, and the United States are in- creasing and diffusing knowledge of productive tech- nique. 1 (d). But knowledge, to be of use, must be applied ; and its application usually requires the aid of capital. The greater and the more productive the stock or capital in any community, the more goods it can put into the currents of trade. A mill will make a town a center of trade. Docks, elevators, warehouses and railway terminals help to transform a harbor into a port of commerce. Since increase in trade tends to decrease the general level of prices, anything which tends to increase trade likewise tends to decrease the general level of prices. We conclude, therefore, that among the causes tending to decrease prices are increasing geographical or personal specialization, improved productive technique, and the accumulation of capital. The history of commerce shows that all these causes have been increasingly operative during a long period including the last century. Consequently, there has been a constant tendency, from these sources at least, for prices to fall. 2 (a). Turning to the consumers' side, it is evident that their wants change from time to time. This is true even of so-called natural wants, but more con- spicuously true of acquired or artificial wants. Wants are, as it were, the mainsprings of economic Sec. 2] INDIRECT INFLUENCES 77 activity which in the last analysis keep the economic world in motion. The desire to have clothes as fine as the clothes of others, or finer, or different, leads to the multiplicity of silks, satins, laces, etc. ; and the same principle apphes to furniture, amusements, books, works of art, and every other means of gratification. The increase of wants, by leading to an increase in trade, tends to lower the price level. Historically, during recent times through invention, education, and the emulation coming from increased contact in centers of population, there has been a great intensification and diversification of human wants and therefore increased trade. Consequently, there has been from these causes a tendency of prices to fall. §2 3 (a). Anything which facilitates intercourse tends to increase trade. Anything that interferes with inter- course tends to decrease trade. First of all, there are the mechanical facilities for transport. As Macaulay said, with the exception of the alphabet and the printing press, no set of inventions has tended to alter civilization so much as those which abridge distance, — such as the railway, the steamship, the telephone, the tele- graph, and that conveyer of information and advertise- ments, the newspaper. These all tend, therefore, to decrease prices. 3 (b). Trade barriers are not only physical but legal. A tariff between countries has the same influence in decreasing trade as a chain of mountains. The freer the trade, the more of it there will be. In France, many communities have a local tariff (octroi) which tends to interfere with local trade. In the United States trade is free within the country itself, but between the 78 THE PURCHASING POWER OF MONEY [Chap. V United States and other countries there is a high pro- tective tariff. The very fact of increasing faciUties for transportation, lowering or removing physical barriers, has stimulated nations and communities to erect legal barriers in their place. Tariffs not only tend to decrease the frequency of exchanges, but to the extent that they prevent international or interlocal division of labor and make countries more alike as well as less productive, they also tend to decrease the amounts of goods which can be exchanged. The ulti- mate effect is thus to raise prices. 3 (c). The development of efficient monetary and banldng systems tends to increase trade. There have been times in the history of the world when money was in so uncertain a state that people hesitated to make many trade contracts because of the lack of knowledge of what would be required of them when the contract should be fulfilled. In the same way, when people cannot depend on the good faith or stability of banks, they will hesitate to use deposits and checks. 3 (d). Confidence, not only in banks in particular, but in business in general, is truly said to be "the soul of trade." Without this confidence there can- not be a great volume of contracts. Anything that tends to increase this confidence tends to increase trade. In South America there are many places wait- ing to be developed simply because capitalists do not feel any security in contracts there. They are fearful that by hook or by crook the fruit of any investments they may make will be taken from them. We see, then, that prices will tend to fall through increase in trade, which may in turn be brought about by improved transportation, by increased freedom of trade, by improved monetary and banking systems, Sec. 3] INDIRECT INFLUENCES 79 and by business confidence. Historically, during recent years, all of these causes have tended to grow in power, except freedom of trade. Tariff barriers, however, have only partly offset the removal of physical barriers. The net effect has been a progressive lowering of trade re- strictions, and therefore the tendency, so far as this group of causes goes, has been for prices to fall. §3 Having examined those causes outside the equation which affect the volume of trade, our next task is to consider the outside causes that affect the velocities of circulation of money and of deposits. For the most part, the causes affecting one of these velocities affect the other also. These causes may be classified as follows : — 1. Habits of the individual. (a) As to thrift and hoarding. (6) As to book credit. (c) As to the use of checks. 2. Systems of payments in the community. (a) As to frequency of receipts and of disburse- ments. (6) As to regularity of receipts and disbursements. (c) As to correspondence between times and amounts of receipts and disbursements. 3. General Causes. (a) Density of population. (6) Rapidity of transportation. 1 (a). Taking these up in order, we may first con- sider what influence thrift has on the velocity of cir- culation. Velocity of circulation of money is the same thing as its rate of turnover It is found by dividing the total payments effected by money in a year by the 80 THE PURCHASING POWER OF MONEY [Chap. V amount of money in circulation in that year. It de- pends upon the rates of turnover of the individuals who compose the society. This velocity of circula- tion or rapidity of turnover of money is the greater for each individual the more he spends, with a given average amount of cash on hand; or the less average cash he keeps, with a given yearly expenditure. The velocity of circulation of a spendthrift may be presumed to be greater than the average.^ He is al- ways apt to be ''short" of funds, — to have a small average balance on hand. But his thrifty neighbor takes care to provide himself with cash enough to meet all contingencies. The latter tends to hoard and lay by his money, and will, therefore, have a slower velocity of circulation. When, as used to be the custom in France, people put money away in stockings and kept it there for months, the velocity of circulation must have been extremely slow. The same principle applies to deposits. In a certain university town the banks often refuse to take deposits from students of spend- ing habits because the average balances of the latter are so low ; or insist on a special stipulation that the balance shall never fall below $100. Hoarded money is sometimes said to be withdrawn from circulation. But this is only another way of say- ing that hoarding tends to decrease the velocity of circulation. A man who is thrifty is usually, to some extent, a hoarder either of money ^ or of bank deposits. Laborers who save usually keep their savings in the form of ^ Cf . Jevons, Money and the Mechanism of Exchange, New York (Appleton), 1896, p. 336. ^ Cf. Harrison H. Brace, Gold Production and Future Prices, New York (Bankers' Publishing Co.), 1910, p. 122. Sec. 3] INDIRECT INFLUENCES 81 money until enough is accumulated to be deposited in a savings bank. Those who have bank accounts will likewise accumulate considerable deposits when preparing to make an investment. Banks whose de- positors are "rapidly making money" and periodi- cally investing the same, have, it is said, less active accounts than banks whose depositors ''live up to their incomes." 1 (6). The habit of "charging," i.e. using book credit, tends to increase the velocity of circulation of money, because the man who gets things "charged" does not need to keep on hand as much money as he would if he made all payments in cash. A man who pays cash daily needs to keep cash for daily contin- gencies. The system of cash payments, unlike the system of book credit, requires that money shall be kept on hand in advance of purchases. Evidently, if money must be provided in advance, it must be pro- vided in larger quantities than when merely required to liquidate past debts. This is true for two reasons : First, in advance of purchases, there is always uncer- tainty as to when money will be needed and how much, while after bills are incurred, the exact sum needed is known. Secondly, and as a consequence of the first circumstance, money held in advance must be held a longer time than money received after a use for it has been contracted for. In short, to keep money in advance requires (a) a larger margin for unforeseen contingencies and (6) a longer period before being dis- bursed during which the money is idle. In the system of cash payments, a man must keep money idle in advance lest he be caught in the embarrassing position of lacking it when he most needs it. With book credit, he knows that even if he should be caught without a 82 THE PURCHASING POWER OP MONEY [Chap. V cent in his pocket, he can still get supplies on credit. These he can pay for when money comes to hand. Moreover, this money need not lie long in his pocket. Immediately it is received, there is a use awaiting it to pay debts accumulated. Now, to shorten the period of waiting evidently decreases the average balance carried, even if in the end the same sums are received and disbursed. For instance, a laborer receiving and spending $7 a week, if he cannot ''charge," must make his week's wages last through the week. If he spends $1 a day, his weekly cycle must show on successive days at least as much as $7, $6, $5, $4, $3, $2, and $1, at which time another $7 comes in. This makes an average of at least $4. But if he can charge every- thing and then wait until pay day to meet the resulting obligations, he need keep nothing through the week, paying out his $7 when it comes in. His weekly cycle need show no higher balances than $7, $0, $0, $0, $0, $0, $0, the average of which is only SI. Through book credit, therefore, the average amount of money or bank deposits which each person must keep at hand to meet a given expenditure is made less. This means that the rate of turnover is increased ; for if people spend the same amounts as before, but keep smaller amounts on hand, the quotient of the amount spent divided by the amount on hand must increase. But we have seen that to increase the rate of turn- over will tend to increase the price level. Therefore, book credit tends to increase the price ^ level. More- over, a community can to some extent cover the relative scarcity of money of a period when business is large ' This indirect effect on the price level must not be confused with the direct effect sometimes claimed. See § 1 of Appendix to (this) Chapter V. Sec. 4] INDIRECT INFLUENCES 83 with the relative surplus of a period when fewer de- mands are made on its supply of money. Otherwise, to maintain the same general level of prices, there would have to be considerably more money when business was large ; and this money, unless it were some form of elastic bank currency which could be canceled and retired, would lie idle during those seasons when busi- ness was slack. In short, book credit economizes money (M) even though it may not economize money payments {E) and therefore increases the velocity of circulation of money {E/M). 1 (c). The habit of using checks rather than money will also affect the velocity of circulation; because a depositor's surplus money will immediately be put into the bank in return for a right to draw by check. Banks thus offer an outlet for any surplus pocket money or surplus till money, and tend to prevent the existence of idle hoards. In like manner surplus deposits may be converted into cash — that is, ex- changed for cash — as desired. In short, those who make use both of cash and deposits have the opportunity, by adjusting the two, to prevent either from being idle. We see, then, that three habits — spendthrift habits, the habit of charging, and the habit of using checks — • all tend to raise the level of prices through their effects on the velocity of circulation of money, or of deposits. It is believed that these habits (except probably the first) have been increasing rapidly during modern times. §4 2 (a). The more frequently money or checks are received and disbursed, the shorter is the average in- terval between the receipt and the expenditure of money 84 THE PURCHASING POWER OF MONEY [Chap. V or checks and the more rapid is the velocity of circula- tion. This may best be seen from an example. A change from monthly to weekly wage payments tends to increase the velocity of circulation of money. If a laborer is paid weekly $7 and reduces this evenly each day, ending each week empty-handed, his average cash, as we have seen, would be a little over half of $7 or about $4. This makes his turnover nearly twice a week. Under monthly payments the laborer who receives and spends an average of $1 a day will have to spread the $30 more or less evenly over the follow- ing 30 days. If, at the next pay day, he comes out empty-handed, his average money during the month has been about $15. This makes his turnover about twice a month. Thus the rate of turnover is more rapid under weekly than under monthly payments. The same result would hold if we assumed that, instead of ending the cycle empty-handed, he ended it with a given fraction — say half — of his wages unspent. Under weekly payments, he would begin with $10.50, and end with $3.50, averaging about $7. Under monthly payments he would thus begin with an average of $45, and end with $15, averaging about $30. In the former case his average velocity of circulation would be once a week and in the latter once a month. The turnover will thus still be about four times as rapid under weekly as under monthly payment. Thus if the distribution of expenditure over the two cycles should have exactly the same ''time shape "^ (distribution in time), weekly payments would ac- ^ Compare Adolphe Landry, "La Rapidite de la Circulation Monetaire," Extrait de La Revue d' Economie 'politique, Fevrier, 1905. Sec. 4] INDIRECT INFLUENCES 85 celerate the velocity of circulation in the same ratio which a month bears to a week. As a matter of his- tory, however, it is not likely that the substitution of weekly payments for monthly payments has in- creased the rapidity of circulation of money among workingmen fourfold, because the change in another element, book credit, would be likely to cause a some- what compensatory decrease. Book credit is less likely to be used under weekly than under monthly payments. Where this book-credit habit or habit of '^ charging" is prevalent, the great bulk of money is spent on pay day. It is probable that the substitu- tion of weekly for monthly payments, when it has taken place, has enabled many workingmen, who formerly found it necessary to trade on credit, to make their own payments in cash, thus tendmg to decrease the velocity of turnover of money. Frequency of disbursements evidently has an effect similar to the effect of frequency of receipts ; i.e. it tends to accelerate the velocity of turnover, or cir- culation. 2 (6). Regularity oi payment also facilitates the turn- over. "When the workingman can be fairly certain of both his receipts and expenditures, he can, by close calculation, adjust them so precisely as safely to end each payment cycle with an empty pocket. This habit is extremely common among certain classes of city laborers. On the other hand, if the receipts and expenditures are irregular, either in amount or in time, prudence requires the worker to keep a larger sum on hand, to insure against mishaps.^ Even when fore- known with certainty, irregular receipts require a larger average sum to be kept on hand. This state- * Compare Landry, ibid. 86 THE PURCHASING POWER OF MONEY [Chap. V merit holds, at least, if we assume that the frequency of payments per year is the same as in the case of regular payments, and that the "tune shape" of ex- penditures between receipts is also the same. Thus, suppose that a workman spends at the rate of $1 a day and receives at the average rate of $1 a day. The average amount that he will require to keep on hand will be less if his receipts occur once every fortnight than if they occur at intervals of three weeks and one week respectively in alternation. For, supposing he tries to come out empty-handed just before each pay- ment, in the former case he will evidently need an aver- age sum each fortnight of $7 ; but in the latter case, he will need for the first period of three weeks, or twenty- one days, $10.50, and in the second period $3.50, the average of which — remembering that the $10.50 applies for three weeks and the $3.50 for one week — will be $8.75. We may, therefore, conclude that reg- ularity, both of receipts and of payments, tends to increase velocity of circulation. 2 (c). Next, consider the synchronizing of receipts and disbursements, i.e. making payments at the same intervals as obtaining receipts. Where payments such as rent, interest, insurance and taxes occur at periods irrespective of the times of receipts of money, it is often necessary to accumulate money or deposits in advance, thus increasing the average on hand, with- drawing money from use for a time, and decreasing the velocity of circulation. This result may, however, be obviated if the individual is willing and able to borrow in order to meet his tax or other special ex- pense, repaying the loan later at his convenience. This is one of the ways in which banking, as already explained, through loans and deposits, serves the con- Sec. 5] INDIRECT INFLUENCES 87 venience of the public and increases the velocity of circulation of money and deposits. Similarly book credit may obviate the inconveniences arising from the disharmony between the times of receipt and dis- bursement; for we have already seen that it is a great convenience to the spender of money or of deposits, if dealers to whom he is in debt will allow him to postpone payment until he has received his money or his bank deposit. This arrangement obviates the necessity of keeping much money or deposits on hand, and therefore increases their velocity of circulation. We conclude, then, that synchronizing and regu- larity of payment, no less than frequency of pa3nnent, have tended to increase prices by increasing velocity of circulation. §5 3 (a). The more densely populated a locality, the more rapid will be the velocity of circulation. ^ There is definite evidence that this is true of bank deposits. The following figures ^ give the velocities of circulation of deposits in ten cities, arranged in order of size: — Paris 116 Lisbon 29 Berlin 161 Indianapolis 30 Brussels 123 New Haven 16 Madrid 14 Athens 4 Rome 43 Santa Barbara 1 Madrid is the only city seriously out of its order in respect to velocity of circulation. 1 This is pointed out by Kinley, Money, New York (Macmillan), 1904, p. 156. ^ These figures are the medians of those of Pierre des Essars for European banks (Journal de la Societe de Statistique de Paris, April, 1895) supplemented by data secured by me from a few American banks. 88 THE PURCHASING POWER OF MONEY [Chap. V 3 (b). Again, the more extensive and the speedier the transportation in general, the more rapid the circulation of money.* Anything which makes it easier to pass money from one person to another will tend to increase the velocity of circulation. Railways have this effect. The telegraph has increased the velocity of circulation of deposits, since these can now be transferred thousands of miles in a few minutes. Mail and express, by facilitat- ing the transmission of bank deposits and money, have likewise tended to increase their velocity of circulation. We conclude, then, that density of population and rapidity of transportation have tended to increase prices by increasing velocities. Historically this con- centration of population in cities has been an im- portant factor in raising prices in the United States. Ordinarily, the velocity of circulation of money and the velocity of circulation of deposits will be similarly influenced by similar causes. In time of panics, however, if the confidence of depositors is shaken, the tendency is for deposits to be withdrawn while money is hoarded. Hence, for a time, the two velocities may change in opposite directions, although there are no good statistics for verifying this supposition. §6 Lastly, the chief specific outside influences on the volume of deposits subject to check are: — (1) The system of banking and the habits of the people in utilizing that system. (2) The habit of charging. 1. It goes without saying that a banking system must 1 Cf. Jevons, Money and the Mechanism of Exchange, New York (Appleton), 1896, p. 336; also Kinley, Money, New York (Mao- miUan), 1904, pp. 156 i nd 157. Sec. 6] INDIRECT INFLUENCES 89 be devised and developed before it can be used. The invention of banking has made deposit currency pos- sible, and its adoption has undoubtedly led to a great increase in deposits and consequent rise of prices. Even in the last decade the extension in the United States of deposit banking has been an exceedingly powerful influence in that direction. In Europe de- posit banking is still in its infancy. 2. ''Charging" is often a preliminary to payment by check, rather than by cash. If a customer did not have his obligations ''charged," he would pay in money and not by check. ^ The ultimate effect of this practice, therefore, is to increase the ratio of check payments to cash payments {E' to E) and the ratio of deposits to money carried {M' to M), and therefore to increase the amount of credit currency which a given quantity of money can sustain. This effect, the substitution of checks for cash pay- ments, is probably by far the most important effect of "charging," and exerts a powerful influence toward raising prices. ^ Andrew, " Credit and the Value of Money." Reprint from Papen and Proceedings of the Seventeenth Annual Meeting American Economic Association, December, 1904, p. 10. CHAPTER VI INDIRECT INFLUENCES (continued) We have now considered those influences outside the equation of exchange which affect the volume of trade (the Q's), the velocities of circulation of money and deposits (V and V), and the amount of deposits (M'). We have reserved for separate treatment in this chapter and the following the outside influences that affect the quantity of money (M). The chief of these may be classified as follows : — 1. Influences operating through the exportation and importation of money. 2. Influences operating through the melting or mint- ing of money. 3. Influences operating through the production and consumption of money metals. 4. Influences of monetary and banking systems, to be treated in the next chapter. The first to be considered is the influence of foreign trade. Hitherto we have confined our studies of price levels to an isolated community, having no trade re- lation with other communities. In the modern world, however, no such community exists, and it is important to observe that international trade gives present-day problems of money and of the price level an interna- tional character. If all countries had their irredeem- able paper money, and had no money acceptable elsewhere, there could be no international adjustment 90 Sec. 1] INDIRECT INFLUENCES CONTINUED 91 of monetary matters. Price levels in different countries would have no intimate connection. Indeed, to some extent the connection is actually broken between exist- ing countries which have different metallic standards, — for example, between a gold-basis and a silver-basis country, — although through their nonmonetary uses the two metals are still somewhat bound together. But where two or more nations trading with each other use the same standard, there is a tendency for the price levels of each to influence profoundly the price levels of the other. The price level in a small country like Switzerland depends largely upon the price level in other countries. Gold, which is the primary or full weight money in most civihzed nations, is constantly travelling from one country or community to another. When a single small country is under consideration, it is therefore preferable to say that the quantity of money in that country is determined by the universal price level, rather than to say that its level of prices is determined by the quantity of money within its borders. An individual country bears the same relation to the world that a lagoon bears to the ocean. The level of the ocean depends, of course, upon the quantity of water in it. But when we speak of the lagoon, we reverse the statement, and say that the quantity of water in it depends upon the level of the ocean. As the tide in the outside ocean rises and falls, the quantity of water in the lagoon will adjust itself accordingly. To simplify the problem of the distribution of money among different communities, we shall, for the time being, ignore the fact that money consists ordinarily of a material capable of nonmonetary uses and may be melted or minted. 92 THE PURCHASING POWER OF MONEY [Chap. VI Let US, then, consider the causes that determine the quantity of money in a state hke Connecticut. If the level of prices in Connecticut temporarily falls below that of the surrounding states, Rhode Island, Massa- chusetts, and New York, the effect is to cause an export of money from these states to Connecticut, because people will buy goods wherever they are cheapest and sell them wherever they are dearest. With its low prices, Connecticut becomes a good place to buy from, but a poor place to sell in. But if outsiders buy of Connecticut, they will have to bring money to buy with. There will, therefore, be a tendency for money to flow to Connecticut until the level of prices there rises to a level which will arrest the influx. If, on the other hand, prices in Connecticut are higher than in surrounding states, it becomes a good place to sell to and a poor place to buy from. But if outsiders sell to Con- necticut, they will receive money in exchange. There is then a tendency for money to flow out of Connecticut until the level of prices in Connecticut is lower. But it must not be inferred that the prices of various articles or even the general level of prices will become precisely the same in different countries. Distance, ignorance as to where the best markets are to be found, tariffs, and costs of transportation help to maintain price differences. The native products of each region tend to be cheaper in that region. They are exported as long as the excess of prices abroad is enough to more than cover the cost of transportation. Practically, a commodity will not be exported at a price which would not at least be equal to the price in the country of origin, plus the freight. Many commodities are shipped only one way. Thus, wheat is shipped from the United States to England, but not from England to the United Sec. 1] INDIRECi" INFLUENCES CONTINUED 93 States. It tends to be cheaper in the United States. Large exportations raise its price in America toward the price in England, but it will usually keep below that price by the cost of transportation. Other commodities that are cheap to transport will be sent in either direc- tion, according to market conditions. But, although international and interlocal trade will never bring about exact uniformity of price levels, it will, to the extent that it exists, produce an adjustment of these levels toward uniformity by regulating in the manner already described the distribution of money. If one commodity enters into international trade, it alone will suffice, though slowly, to act as a regulator of money distribution ; for in return for that com- modity, money may flow and, as the price level rises or falls, the quantity of that commodity sold may be correspondingly adjusted. In ordinary intercourse between nations, even when a deliberate attempt is made to interfere with it by protective tariffs, there will always be a large number of commodities thus acting as outlets and inlets. And since the quantity of money itself affects prices for all sorts of commodi- ties, the regulative effect of international trade applies, not simply to the commodities which enter into that trade, but to all others as well. It follows that now- adays international and interlocal trade is constantly regulating price levels throughout the world. We must not leave this subject without emphasizing the effects of a tariff on the purchasing power of money. When a country adopts a tariff, the tendency is for the level of prices to rise. A tariff obviously raises the prices of the ''protected" goods. But it does more than that, — it tends also to raise the prices of goods in general. Thus, the tariff first causes a decrease 94 THE PURCHASING POWER OF MONEY [Chap. VI in imports. Though in the long run this decrease in imports will lead to a corresponding decrease in exports, yet at first there will be no such adjustment. The foreigner will, for a time, continue to buy from the pro- tected country almost as much as before. This will result temporarily in an excess of that country's exports over its imports, or a so-called ''favorable" balance of trade, and a consequent inflow of money. This inflow will eventually raise the prices, not alone of protected goods, but of other goods as well. The rise will con- tinue till it reaches a point high enough to put a stop to the ''favorable" balance of trade. Although the "favorable balance" of trade created by a tariff is temporary, it leaves behind a permanent increase of money and of prices. The tariff wall is a sort of dam, causing an elevation in the prices of the goods impounded behind it. This fact is sometimes overlooked in the theory of international trade as commonly set forth. Emphasis is laid instead on the fact that in the last analysis the trade is of goods for goods, not of money for goods, and that a tariff on imports reduces, not only imports, but exports also, — that it merely interrupts temporarily the virtual barter between nations. The effect of a tax on imports is likened to that of a tax on exports. But in respect to effects on price levels a tax on imports and a tax on exports are diametrically opposed. If we place our tax on exports, we first interfere with exports. The imports are not checked until money has flowed out and has reduced the general price level enough to de- stroy the "unfavorable" balance of trade first created. We conclude that the general purchasing power of money is reduced by a tariff and that it would be in- creased by a tax on exports. Sec. 1] INDIRECT INFLUENCES CONTINUED 95 This is, perhaps, the chief reason why a protective tariff seems to many a cause of prosperity. It furnishes a temporary stimulus, not only to protected industries, but to trade in general, which is really simply the stimu- lus of money inflation. Our present interest in international trade, however, is mainly directed to its effects on international price levels. Except for the export or import of money to adjust the price levels, international trade is at bottom merely an interchange of goods. Where the price level is not concerned, the money value of the goods sold by a country will exactly equal the value of those bought. Only when there is a difference in these values, or a "balance of trade," will there be any flow of money and consequently any tendency to modify the price level.^ We have shown how the international and interlocal equilibrium of prices may be disturbed by differential changes in the quantity of money alone. It may also be disturbed by differential changes in the volume of bank deposits ; or in the velocity of circulation of money ; or in the velocity of circulation of bank de- posits ; or in the volume of trade. But whatever may be the source of the difference in price levels, equilibrium will eventually be restored through an international or interlocal redistribution of money and goods brought about by international and interlocal trade. Other elements in the equation of exchange than money and commodities cannot be transported from one place to another. Except for transitional effects, then, international differences of price levels produce changes only in one 1 For mathematical statement, see § 1 of Appendix to (this) Chapter VI. 96 THE PURCHASING POWER OF MONEY [Chap. VI of the elements in the equation of exchange, — the volume of money. Practically, of course, transition periods may be incessant or chronic. It seldom happens that a nation has no balance of trade. For decades Ori- ental nations took silver from Occidental nations even when silver was, under the bimetallic regime, at a stable ratio with gold. In Europe there was a consequent long-continued tendency for prices to fall, and in Asia a tendency to rise, with all the other transitional effects involved. §2 We have seen how M in the equation of exchange is affected by the import or export of money. Con- sidered with reference to the M in any one of the coun- tries concerned, the M's, in all the others are ''outside influences." Proceeding now one step farther, we must consider those influences on M that are not only outside of the equation of exchange for a particular country, but out- side those for the whole world. Besides the monetary inflow and outflow through import and export, there is an inflow and outflow through minting and melting. In other words, not only do the stocks of money in the world connect with each other like interconnecting bodies of water, but they connect in the same way with the outside stock of bullion. In the modern world one of the precious metals, such as gold, usually plays the part of primary money, and this metal has two uses, — a monetary use and a commodity use. That is to say, gold is not only a money material, but a commodity as well. In their character of commodities, the precious metals are raw materials for jewelry, works of art, and other products into which they may be wrought. It is Sec. 2] INDIRECT INFLUENCES CONTINUED 97 in this unmanufactured or raw state that they are called bullion. Gold money may be changed into gold bullion, and vice versa. In fact, both changes are going on con- stantly, for if the value of gold as compared with other commodities is greater in the one use than in the other, gold will immediately flow toward whichever use is more profitable, and the market price of gold bullion will determine the direction of the flow. Since 100 ounces of gold, y^ fine, can be transformed into $1860, the market value of so much gold bullion, ^^ fine, must tend to be $1860. If it costs nothing to have bullion coined into money, and nothing to melt money into bullion, there will be an automatic flux and reflux from money to bullion and from bullion to money that will prevent the price of bullion from varying greatly. On the one hand, if the price of gold bullion is greater than the money which could be minted from it, no matter how slight the difference may be, the users of gold who require bullion — notably jewelers — will save this difference by melting gold coin into bullion. Con- trariwise, if the price of bullion is less than the value of gold coin, the owners of bullion will save the differ- ence by taking bullion to the mint and having it coined into gold dollars, instead of selling it in the bullion market. The effect of melting coin, on the one hand, is to decrease the amount of gold money and increase the amount of gold bullion, thereby lowering the value of gold as bullion and raising the value of gold as money; thereby lowering the price level and restoring the equal- ity between bullion and money. The effect of minting bullion into coin is, by the opposite process, to bring the value of gold as coin and the value of gold as bullion again into equilibrium. In practice, the balance is 98 THE PURCHASING POWER OF MONEY [Chap. VI probably ^ maintained chiefly by turning newly mined gold into the one or the other use according to the mar- ket. By thus feeding the two reservoirs according to their respective needs there is saved the necessity of any great amount of interflow between money and the arts. Where a charge — called " seigniorage " — is made for changing bulhon into coin, or where the process in- volves expense or delay, the flow of bullion into currency will be to that extent impeded. But under a modern system of free coinage and with modern methods of metallurgy, both melting and minting may be performed so inexpensively and so quickly that there is practically no cost or delay involved. In fact, there are few in- stances of more exact price adjustment than the ad- justment between gold bullion and gold coin. It follows that the quantity of money, and therefore its purchasing power, is directly dependent on that of gold bullion. The stability of the price of gold bullion expressed in gold coin causes confusion in the minds of many people, giving them the erroneous impression that there is no change in the value of money. Indeed, this stability has often been cited to show that gold is a stable standard of value. Dealers in objects made of gold seem to misunderstand the significance of the fact that an ounce of gold always costs about $18.60 in the United States or £3 17s. 10-2