The Purchasing Power of Money

CHAPTER XI
STATISTICAL VERIFICATION. GENERAL HISTORICAL REVIEW

§ 1

SINCE both the level of prices and the quantity of money in circulation cannot in practice be perfectly measured, and since the level of prices depends upon other factors besides the quantity of money,—viz. the quantity of circulating credit, the velocities of circulation of that credit and of money, and the volume of business,—it would be absurd to expect any exact correspondence between variations in the quantity of circulating money and variations in the price level; and it is likewise absurd to state, as some have stated, that the absence of exact statistical correspondence proves the absence of any influence of quantity of money on price level. Nevertheless, when the volume of money changes greatly and quickly, the effect on prices from this cause is usually so great as to make itself manifest.

The general trend of prices has usually been upward, as Figure 10*53 shows. According to the diagram prices are now about five times as high as a thousand years ago and are from two to three times as high as in the period between 1200 and 1500 A.D.*54 Beginning with the last-named date, or shortly after the discovery of America, prices have almost steadily risen.

The discovery of America was followed in 1519 by the invasion of Mexico under Cortez and, twenty years later, by Pizarro's conquest of Peru. From these conquests and the consequent development of New World mining of precious metals, dates the tremendous production of gold, and especially of silver, during the sixteenth century. From the discovery of America until the after effects of its discovery began to be felt, or, to be exact, through the year 1544, the average annual output of gold was less than five million dollars, and of silver about the same.*55 The rich mines of Potosi in Bolivia were discovered in 1546. From 1545 to 1560 the annual production of silver averaged eighteen millions, which was over fourfold the previous rate. The product of gold also increased, though slightly. The product of gold also increased, though slightly. The rates of production for both metals rose steadily (with slight interruption, 1811-1840) up to the present time.

These new world mines began to pour their product into Europe: first into Spain, the chief owner of the mines, then, by trade, into the Netherlands and other parts of Europe, and then into the Orient—that great "sink of silver." Accordingly, as Cliffe Leslie*56 has shown, prices rose first in Spain, then in the Netherlands, and then in other regions.

But, though the new supplies of the precious metals distributed themselves very gradually through Europe, and the rise of prices was consequently in some regions delayed, there can be no doubt that they rose or that the rise was great. The rise between the discovery of America and the beginning of the nineteenth century was several hundred per cent. This rise was simultaneous with an increase of the stock of the precious metals, because production outran consumption.

Although the total production of the precious metals continued to increase until 1810, the ratio of the yearly production to the existing stock became gradually less. Corresponding to this slackening of production, and presumably because of it, prices did not continue to rise at the same rapid rate as at first. Furthermore, with the development of trade with the East, more and more of the new supplies found their way thither. The most rapid rise occurred during the sixteenth century.

§ 2

The stock of money metals at any time in any country is evidently the difference between the total product and the sum of the consumption and the net export. Jacob*57 has estimated roughly the stock in Europe at various dates. The following table compares the estimated metallic stocks in Europe with the estimated price levels:—

With the enormous increase in the quantity of the precious metals, small wonder if prices have risen!

We see that there has been a general increase (1) in the stock of money metals, and (2) in the price level, and that the greatest increase of each was in the sixteenth century. We find also that the prices did not increase as fast as the quantity of money. This relative slowness on the part of prices was to be expected, because of the increased volume of business. This, we know, must have come with increased population and with progress in the arts—especially the arts of trade—and with development in transportation. As to changes in the velocity of circulation of money we know absolutely nothing.

§ 3

During the last century the price movements have been more carefully recorded and show many ups and downs. The most complete statistics (those of Sauerbeck) are for England. They are represented in Figure 11.*59

As is well known, English prices were inflated by the issue of irredeemable paper during the Napoleonic wars. This period of the paper standard extended from 1801 to 1820. But prices in paper were only slightly higher than prices in gold, and the chief price movements (except in a few years) were but slightly affected by the existence of a paper standard. The main periods of price movements in England since 1789 may be stated as follows:—

Prices rose 1789-1809, stock increasing.
Prices fell 1809-1849, stock stationary.
Prices rose 1849-1873, stock increasing.
Prices fell 1873-1896, stock increasing slightly.
Prices rose 1896-present, stock increasing.

In each case is cited the movement in the stock of money metals in Europe as given in the table of Del Mar.*60

The only period which does not, at first glance, agree with what we might expect if our theory of price levels in relation to money is correct, is the period 1873-1896. Of the other four periods, three are periods of rising prices and increasing stocks. The fourth is a period of a stationary stock; and since the volume of trade undoubtedly increased, a fall of prices was naturally to be expected.

The exceptional period 1873-1896—a period of falling prices—is probably to be accounted for by the increasing volume of trade and the successive demonetization of silver by various countries.

The foregoing parallelism between monetary stocks and prices is somewhat remarkable in view of the in completeness of the data.*61 In the table there are lacking, not only exact statistics as to the volume of trade and all statistics whatever of velocity of circulation, but also statistics of the volume of bank notes, government notes, and deposit currency. We know, however, that modern banking, which had scarcely developed at all before the French Revolution, developed rapidly throughout the nineteenth century. It is also known that banking and deposit currency developed more rapidly during the third period in the table (1849-1873) than during the fourth (1873-1896),*62 which fact contributes somewhat to explain the contrast between the price movements of these two periods.

§ 4

We may, therefore, summarize the course of price movements during the nineteenth century by the following probable statements:—

1. Between 1789 and 1809 prices rose rapidly, the index numbers of Jevons moving from 85 to 157 when prices are expressed in the gold standard, or 161 when expressed in paper.*63 That is, prices practically doubled in twenty years. This rise was due to the increased stock of gold and silver, which in turn was due to their large production during this period as compared with the periods before and after. The production of silver was especially great.*64 The Napoleonic wars with their destruction of wealth and interference with trade probably exercised some influence in the same direction.*65

2. Between 1809 and 1849 prices fell. The fall was measured by Jevons as a fall from 157, gold (or 161, paper), to 64. That is, in forty years prices were reduced to less than half, or, to be more exact, to two fifths. This fall in prices was presumably due to the lull in the production of the precious metals, which prevented the aggregate stock from keeping pace with the volume of business. Indeed, the aggregate stock remained stationary while the volume of business increased. Even the development of bank currency was insufficient to offset the continued increase in the volume of business. It is interesting to observe that this period of falling prices was interrupted by a temporary rise after 1833, which Jevons was at a loss to account for, but which was apparently due to the inflow of Russian gold after the discoveries of gold in Siberia in 1830.*66

3. Between 1849 and 1873 (although with two notable interruptions) prices rose. They rose, according to Jevons's figures supplemented by Sauerbeck's,*67 from 64 to 86, and according to Sauerbeck's alone, from 74 to 111. That is, in 24 years prices increased, according to one calculation, by one third; according to another, by one half. This rise was presumably in consequence of the gold inflation following the famous California gold discoveries in 1849 and Australian discoveries in 1851 and 1852. The simultaneous rapid development of banking contributed to the same result in spite of the continued increase in trade.

4. Between 1873 and 1896 prices fell. This fall was presumably due to the slackening in the production of gold; to the adoption of the gold standard by nations previously on a silver basis, and the consequent withdrawal of gold by these new users from the old; to the arrest of the expansion of silver money consequent on the closure of mints to silver; to the slackening in the growth of banking; and to the ever present growth of trade.*68

During the long fall of prices from 1873 to 1896, country after country adopted the gold standard. We have already seen that Germany adopted the gold standard in 1871-1873, thus helping to render impossible the maintenance of bimetallism by the Latin Union. The Scandinavian monetary union adopted the gold standard in 1873. Between that date and 1878 the countries of the Latin Union suspended the free coinage of silver and came practically to a gold basis. In the United States the legislation of 1873 signified that with resumption (which took place in 1879), the country would come to a gold basis, although no considerable amount of silver, except for small change, had been coined here for several decades previously. The Netherlands virtually adopted the gold standard in 1875-1876, Egypt in 1885, Austria in 1892, India in 1893, Chili in 1895, Venezuela and Costa Rica in 1896, Russia, Japan, and Peru in 1897, Ecuador in 1899, and Mexico in 1905. In fact, most countries of importance have now definitely adopted the gold standard.

The preceding figures apply only to gold countries. But in 1873 gold and silver countries, as it were, fell asunder. It is interesting, therefore, to inquire whether the movement of prices in gold countries was parallel or antithetical to that in silver countries. As might be expected, we find it antithetical. The demonetization of silver in gold countries made a greater amount of that metal available for silver countries. Accordingly, we find that prices rose in India from 107 in 1873 to 140 in 1896,*69 in Japan from 104 in 1873 to 133 in 1896,*70 and in China from 100 in 1874 to 109 in 1893.*71 These figures, although not as reliable and representative as the figures for gold countries, agree in indicating a rise of prices. The amount of rise is differently indicated, ranging roughly from 10 per cent to 35 per cent. The following table shows the contrast between the gold and silver countries as between 1873-1876 and 1890-1893, the last year being that of the closure of the Indian mint to silver.*72

PRICES IN GOLD AND SILVER COUNTRIES*73
	GOLD	SILVER
1873-1876	100	100
1890-1893	78	117

We see that gold prices fell a little more than 20 per cent and that silver prices rose a little less than 20 per cent.*74

If some way had been contrived by which gold and silver could have been kept together (say by world-wide bimetallism), prices would not have fallen so much in gold countries, or risen so much (if at all) in silver countries, but would probably have fallen in gold countries slightly - probably about 10 per cent up to 1890-1893 and more up to 1896. This is because the stocks of specie in silver countries were less than half those in gold countries*75 (including those with the "limping standard" from left-over silver); so that had there been a transfer of a given amount of silver from the silver Orient to the gold Occident, this would have affected Oriental prices about twice as much as Occidental.

The transition of India from the silver to the gold side has left about nine tenths*76 of the specie (gold and overvalued silver) in the gold column. In other words, the world is now practically on a gold basis. The result has been to make Indian prices move in sympathy with European prices, *77 instead of in opposition.

5. From 1896 to the present, prices have been rising because of the extraordinary rise in gold production and the consequent increase in money media of all kinds. The gold of South Africa combined with the gold from the rich mines of Cripple Creek and other parts of the Rocky Mountain Plateau, and reënforced by gold from the Klondike, caused, and is still causing, a repetition of the phenomenon of half a century ago.

That there has been a distinct rise in prices is evident from the figures of all index numbers. Those of The Economist, Sauerbeck, Dun, the Labor Bureau Reports, and Bradstreet are given on the following page.

ENGLISH			AMERICAN
Close of Dec.	Economist	Sauerbeck	Dun	Labor Bureau	Bradstreet
1896	1950	61	74	90	59
1897	1890	62	72	90	61
1898	1918	64	77	93	66
1899	2145	68	85	102	72
1900	2126	75	91	111	79
1901	1948	70	91	109	76
1902	2003	69	102	113	79
1903	2197	69	99	114	79
1904	2136	70	97	113	79
1905	2342	72	98	116	81
1906	2499	77	105	123	84
1907	2310	80		130	89
1908	2197	73		123	80
1909	2373	74		127	85

The high points of 1900 and 1907, as compared with the low level of 1896, must be regarded as at least partly due to expansion of credit. The fairest comparison (to eliminate the effects of undue changes in credit) is perhaps that of the years 1896, 1903, and 1909. That the rise of prices has been world wide is evidenced not only by index numbers, which are only available for a limited number of countries, but by general impressions of consumers and by special reports and investigations.*78

The period 1896-1909 for the United States will be studied in more detail in the following chapter.

§ 5

It will be seen that the history of prices has in substance been the history of a race between the increase in media of exchange (M and M') and the increase in trade (T), while (we assume) the velocities of circulation were changing in a much less degree. Knowing little of the variations in the development of trade, we may tentatively assume a steady growth, and pay chief attention to the variations of circulating media. Sometimes the circulating media shot ahead of trade and then prices rose. This was undoubtedly the case in the periods numbered 1, 3, and 5 of the five periods just considered. Sometimes, on the other hand, circulating media lagged behind trade and then prices fell. This must have been the case in the periods numbered 2 and 4.

It is important to emphasize at this point a fact mentioned in a previous chapter; namely, that the breakdown of bimetallism and the consequent division of the world into a gold section and a silver section have made each section more sensitive than before to fluctuations in the production of the precious metals. The present flood of gold can spread itself only over the gold section of the world, and not over the whole world as was virtually the case with the Californian gold immediately after 1849. At that time gold displaced silver in bimetallic France and sent it to the Orient. In this way the Orient afforded relief for bimetallic countries by draining off silver and making room for gold; and the bimetallic countries thereby afforded relief to gold countries also.

Since 1873, therefore, the gold reservoir of Europe and America has been separated from the silver reservoir of the East, with the consequence that the European and American reservoir level has been made more sensitive to either a scarcity or a superabundance of gold. The result has been to aggravate both the fall of prices from 1873 to 1896 and the present rise, although the later effect is mitigated by the previous extension of the gold standard.

§ 6

The outlook for the future is apparently toward a continued rise of prices due to a continued increase in the gold supply. To-day almost as much gold is produced every year as was produced in the whole of the 16th century.

The most careful review of present gold-mining conditions shows that we may expect a continuance of gold inflation for a generation or more. "For at least thirty years we may count on an output of gold higher than, or at least comparable to, that of the last few years."*79 This gold will come from the United States, Alaska, Mexico, the Transvaal, and other parts of Africa and Australia, and later from Colombia, Bolivia, Chili, the Ural Province, Siberia, and Korea. It must be remembered that it is the stock of gold and not the annual production which influences the price level; and that the stock will probably continue to increase for many years after the production has begun to decline,—as long, in fact, as production keeps above consumption.

A lake continues to rise long after the freshet which feeds it has reached its maximum. So the stock of gold will continue to increase long after the annual production of gold has stopped increasing. Whether or not prices will continue to rise depends on whether the increase in gold and the circulating media based on gold continues to exceed the growth of trade. It is the relation of gold to trade that chiefly affects prices. Even if the stock of gold should increase for many years, prices may not rise; for trade may increase still faster. If the annual additions of gold to the total stock remain constant and consequently the stock continually increases, the ratio between the constant annual addition and the increasing stock will evidently decrease, and the increase in stock will count for less and less in raising prices.*80

It is difficult to predict the future growth of trade and therefore impossible to say for how long gold expansion will keep ahead of trade expansion. That for many years, however, gold will outrun trade seems probable, for the reason that there is no immediate prospect of a reduction in the percentage growth of the gold stock, nor of an increase in the percentage growth of trade. Not only do mining engineers report untold workable deposits in outlying regions (for instance a full billion of dollars in one region of Colombia alone), but any long look ahead must reckon with possible and probable cheapening of gold extraction. The cyanide process has made low grade ores pay. If we let imagination run a little ahead of our times, we may expect similar improvements in the future whereby still lower grades may be worked or possibly the sea compelled to give up its gold. Like the surface of the continents, the waters of the sea contain many thousand times as much gold as all the gold thus far extracted in the whole history of the world. It is to be hoped that the knowledge of how to get this hidden treasure may not be secured. To whatever extent inventors and gold miners might be enriched thereby, scarcely a worse economic calamity can be imagined than the resulting depreciation. It may be, however, that only by such a calamity can the nations of the world be aroused to the necessity of getting rid of metallic standards altogether.

§ 7

We have briefly summarized the history of price movements since the discovery of America and shown their relation to the stock of the precious metals. But, as we have emphasized in previous chapters, the precious metals do not include all forms of circulating media. Paper money and bank deposits have come during the nineteenth century to occupy very important places in currency systems.

We shall not attempt any complete review of the effects of paper money on prices. The best that can be done is to mention briefly the most striking cases of paper money inflation and contraction. These are all cases of irredeemable paper money. When paper money is redeemable, its possible increase is restricted by that fact and, what is more important, the effects of the increase are dissipated over so large an area as to have little perceptible effect on prices. This dissipation takes place through the export of specie from the country in which the paper issues occur. Though the paper cannot be itself exported, it can displace gold or silver, which amounts to the same thing so far as spreading out the effect on prices is concerned.

But when the paper is irredeemable, after specie has been expelled from circulation (whether by export, melting, or hoarding in anticipation of disaster) there is no such spreading-out effect. The effects on prices are then entirely local and therefore greatly magnified.*81

The consequence is that the most striking examples of price inflation are cases of irredeemable paper money. The rise of prices is often still further aggravated by the gradual substitution of other and better money or resort to barter, which further restricts the sphere in which the paper is used and within that sphere makes it the more redundant. Where the paper money is looked upon with disfavor, for whatever reason—whether because its promised redemption has been indefinitely postponed, or simply because of the bare fact that it is depreciating, or because of any other consideration—its sphere of use is restricted.*82 Creditors and tradesmen avoid taking it if they can, by "contracting out" in advance; by barter; by fixing a double set of prices, one in paper and the other in some other money; and by outright refusal. In the end it may happen that the paper ceases to be used at all. In that case its value depreciates indefinitely and therefore prices (so far as still expressed in terms of paper) rise indefinitely.

Whatever the situation, the equation of exchange continues to hold true though its significance becomes of less importance, because T, instead of comprising practically all trade, comes to mean only that disappearing portion of trade still transacted by means of paper money.

The value of irredeemable paper money is, therefore, extremely precarious. If once it starts depreciating—from whatever cause—it is likely to depreciate further, not simply because of the ever present temptation to further issue, but also because of a growing public sentiment against it which sooner or later restricts its use.*83 In many cases the irredeemable paper money continues to be used with sufficient acceptability to give it a virtual monopoly as a medium of exchange.

Although theoretically irredeemable paper money may be the cheapest and most easily regulated form of currency, and although, in some cases, it has remained a stable currency for a considerable period, the lesson of history is emphatically that irredeemable paper money results in monetary manipulation, business distrust, a speculative condition of trade, and all the evils which flow from these conditions.

§ 8

One of the early paper-money schemes was that of John Law, who established a bank of issue in France in 1716. Two years later (December 4, 1718), the bank was taken over by the Crown. Soon shrewd traders were acquiring specie for notes and exporting the specie secretly, although exportation of specie was illegal. May 27, 1720, only four years after its establishment, the bank stopped payment of specie. By November of the very same year the paper had fallen to one tenth of its par value, and after this it became utterly worthless.

The case of the assignats of the French Revolution is classic.*84 It was in December, 1789, that the first issue, four hundred million francs, was ordered, based ostensibly on the landed property of the nation. The notes were issued in April, 1790, and bore 3 per cent interest. According to the original plan, all of the assignats received in payment for land were to be burned. But original plans seem never to be carried out with respect to paper money. Instead, a hundred millions were reissued in the form of small notes. Prices began to rise. In June of the year 1791, six hundred millions more were issued. Depreciation to the extent of 8 to 10 per cent immediately followed. Specie was rapidly disappearing. Another three hundred millions of francs were ordered in December, 1791. By February, the assignats were over 30 per cent below par. In April, 1792, came a decree for the issue of three hundred millions more, and in July for the same amount additional. Most prices were very high, but wages seem to have still remained at the level of 1788. By December 14 of 1792, thirty-four hundred million francs had been issued in assignats, of which six hundred millions had been burned, leaving twenty-eight hundred millions in circulation. Laws were enacted to fix maximum prices, but were evaded. By 1796, forty-five billion francs had been issued, of which thirty-six billions were in circulation. In February of that year the gold louis, of 25 francs, was worth 7200 francs in assignats; and the assignats were worth 1/288 of par. A new kind of paper money, the mandats, was next issued, but soon fell to 5 per cent of its nominal value. In the end the twenty-five hundred million mandats and the thirty-six billion assignats were repudiated and became entirely worthless.

§ 9

England's experience with irredeemable paper money was more temperate. Under the stress of the Napoleonic wars, the Bank of England suspended cash payments in 1797. This nullified the force which automatically limited overissue. The bank resumed cash payments in 1821. During much of the intervening period of paper money, prices in paper were very high. The following table of Jevons shows the relative prices in notes and specie from 1801 to 1820:*85—

YEAR	GOLD STANDARD	PAPER STANDARD
1801	140	153
1802	110	119
1803	125	128
1804	119	122
1805	132	136
1806	130	133
1807	129	132
1808	145	149
1809	157	161
1810	142	164
1811	136	147
1812	121	148
1813	115	149
1814	114	153
1815	109	132
1816	91	109
1817	117	120
1818	132	135
1819	112	117
1820	103	106

The causes of the rise of prices were discussed in the famous Bullion Report. The general conclusion reached was that a "rise of the market price of gold above its mint price will take place," if the local currency of any particular country, "being no longer convertible into gold, should at any time be issued to excess. That excess cannot be exported to other countries, and, not being convertible into specie, it is not necessarily returned upon those who issued it; it remains in the channel of circulation, and is gradually absorbed by increasing the prices of all commodities. An increase in the quantity of the local currency of a particular country will raise prices in that country exactly in the same manner as an increase in the general supply of precious metals raises prices all over the world. By means of the increase of quantity, the value of a given portion of that circulating medium, in exchange for other commodities, is lowered. In other words, the money prices of all other commodities are raised—that of bullion with the rest." This is an excellent statement of the philosophy of irredeemable paper money when that money is sufficiently within bounds to remain in general use. No mention is made of partial or complete abandonment of use because of worthlessness. The reason is doubtless that in England the paper money never reached this pass, as it undoubtedly did in many instances in France, Austria, America, and elsewhere.

§ 10

Austrian experience with paper money is instructive.*86 Like so many of the European banks, that of Austria was used by the government as an instrumentality for obtaining loans. This was done by allowing the bank to issue large sums in notes. The wars with Napoleon demanded supplies, and during these wars the issue was largely increased. In 1796 the note issue was 47,000,000 gulden; in 1800 it was 200,000,000; in 1806, it was 449,000,000. The notes were much below par. In 1810 the bank notes fell successively to 1/5, 1/8 and about 1/11 of par. In 1811 a proclamation openly valued them at one fifth of their nominal value and decreed their exchange at this rate for redemption notes, called the Viennese legal tender, which became the Austrian legal-tender currency. But even these new issues soon fell to 1/216 of their face value (May, 1812) and 1/338 of their face value (June, 1812), while the bank notes were at 1690 to 100 in silver. New issues were added under a different name until, in 1816, the amount of paper money was over 638,000,000, with prices, of course, tremendously inflated. In 1816 was founded the Austrian national bank, which was intended to draw in the paper money. From time to time thereafter the amount of paper circulation was reduced, but not without occasional relapses. At the present time Austria has no paper money which is not at par.

§ 11

Many of the American colonies had experience with paper money. In fact, one of the grievances against England was the parliamentary prohibition of paper money issues! In practically all cases*87 there was overissue and depreciation. This was true, for example, in Massachusetts, where paper money was issued to pay the expenses of the expeditions against Canada,*88 and in Rhode Island,*89 which suffered more, perhaps, from paper money than any of the others. Following are figures for Rhode Island taken from the account book of Thomas Hazard (the entries and memoranda extending from 1750 to 1785), which show the height and variability of prices.*90

1755. HAY £20 PER LOAD
CORN PER BUSHEL		BUTTER PER POUND
1751	25 s.	1751	7 s.
1758	50 s.	1760	16 s.
1762	100 s.
WOOL PER POUND		POTATOES PER BUSHEL
1752	8 s.	1750	10 s.
1756	12 s.	1753	20 s.
1759	28 s.	1774	35 s.
1768	32 s.

We had also during the Revolution a national experience with Continental paper money which gave rise to the derogatory phrase, still current, "not worth a continental." Depreciation began almost from the moment of issue (1775), and finally the money was recognized by Congress itself to have reached 1/40 of the nominal value.*91 All prices, of course, were tremendously high. Even the new tenor paper given for the old emissions at the rate of a dollar for forty*92 declined rapidly in value. A bushel of wheat was worth, at one time, seventy-five dollars, coffee four dollars a pound, and sugar three dollars a pound.*93 It is interesting to observe that, in this case, the depreciation seems to have been accentuated far beyond what mere overissue tended to produce, by a distrust of the money and a refusal to receive it in trade. Several classes were disinclined to receive it to begin with, and as confidence waned, the number who were unwilling to receive it increased. Barter frequently took the place of trade with money.*94

The depreciation was doubtless much greater because the paper money of various colonies helped to overflow the circulation, competing with the congressional money and limiting its sphere of circulation.

§ 12

The effects were so disastrous that, in the Constitution of the United States, a provision was incorporated prohibiting any state from issuing "bills of credit." But, during the Civil War, the temptation again came to resort to this easy way of securing means of payment; and the federal government itself issued United States notes or "greenbacks." The banks had already suspended specie payments so that gold was at a slight premium in bank paper.*95

These greenbacks were issued from time to time during the war with resulting depreciation as their quantity increased,—a depreciation greater or less also according as failure or success of the Union armies affected confidence in the paper money. The amounts issued were: $150,000,000 by the act of February 25, 1862; $150,000,000 by act of July 11, 1862; $150,000,000 authorized by acts of January 17 and March 3, 1863. Besides the greenbacks (issued in denominations in no case under a dollar), there was some issue of fractional currency and of interest-bearing notes running for a brief period, both of which were also made legal tender.*96 The rise in prices is shown by the following table:*97—

INDEX NUMBERS OF PRICES DURING GREENBACK DEPRECIATION
YEAR	PRICE OF GOLD*98 IN GREENBACKS	INDEX NOS. OF NORTHERN PRICES (1860=100)
		Falkner*99		Dun*102 in Paper	Mitchell*102 Median in Paper
		In Gold*100	In Paper*101
1861	100	94	94	89	96
1863	144	91	132	150	134
1865	163	107	232	169	158
1867	138	123	166	164	150
1869	136	112	152	143	158
1871	112	123	136	132	130
1873	114	115	129	124	130
1875	115	115	129	117	121
1877	105	107	114	95	100
1879	100	95	95	85	85

It has been asserted that the rise of prices during the greenback depreciation was not due to the quantity of the greenbacks, but to the public distrust of greenbacks. The truth is probably that it was due to both. Distrust was evident and restricted the sphere of greenbacks very materially. California and, in fact, all the region west of the Rocky Mountains, made strenuous efforts to prevent the circulation of greenbacks,—efforts which were largely successful. And naturally the greenbacks could not circulate in the South. These restrictions alone would confine their circulation to a population of about 20 millions out of a total population in 1860 of 31 millions, that is, to less than two thirds of the entire population. Therefore the volume of trade for which the greenbacks were used must have been greatly reduced. The total circulating currency during the war is not known with certainty; but the best estimates of the various forms of circulating media are those compiled by Mitchell.*103 Though he modestly warns the reader against any attempt to cast up sums, his results may be considered as at least of some value. The totals, omitting money in the Treasury and interest-bearing forms, which were known to have only a very sluggish circulation, we find to be as follows:—

YEAR	ROUGHLY ESTIMATED CIRCULATION IN LOYAL STATES*104	PRICES OF ALL ARTICLES AVERAGED*105 ACCORDING TO IMPORTANCE, COMPRISING 68.60% OF TOTAL EXPENDITURE
1860	433	100
1861	490	94
1862	360	104
1863	677	132
1864	708	172
1865	774	232
1866	759	188

Considering the unreliability of the figures for currency*106and the lack of data as to the other magnitudes in the equation of exchange, there is here a rough correspondence between the volume of the currency and the level of prices.

§ 13

It is necessary to remember that the confidence with which we have to deal is not primarily confidence in redemption, but confidence in the paper money,—its purchasing power. This confidence may rest on expectation of redemption or on other conditions, particularly the expectation of further inflation or contraction. The explanation of the value of the greenbacks appears to me to be in brief as follows:—

The Redemption Act of 1875 announced the intention of our government to redeem the greenbacks on and after January 1, 1879. Each greenback being thus kept equal to the discounted value of a full dollar due January 1, 1879, they rose steadily toward par as that date approached. Some of them were withdrawn from circulation to be held for the rise. The value of a greenback dollar could not be much less than this discounted value of the gold dollar promised in 1879; otherwise speculators might withdraw the greenbacks wholly. This would pay them well provided they were certain that the government's promise would be fulfilled. On the other hand, the value of greenbacks could not, with paper money redundant for trade, be greater than said discounted value, because in that case speculators would return it all to the circulation, the prospective rise being "too small to repay the interest" lost in carrying it. Thus speculation acted as a regulator of the quantity of money.

Thus the rise in value of the greenbacks, like other coming events, cast its shadow before. It was "discounted in advance." It is quite true that confidence in redemption was here the ultimate cause of the appreciation of the paper money; but the readjustments caused by this confidence include a reduction in the quantity of the money in circulation. Without such readjustment the appreciation would be impossible, as the equation of exchange plainly shows. We should note, however, that if the price of the currency were already sufficiently high, the prospect of future redemption would not further raise it. It might happen that the value of the currency was already above the discounted par value promised at the time set for redemption. In such a case there need not be any speculation or any immediate rise in value until the date of redemption drew near enough to make itself felt. On the other hand, when, during the war, the government announced a further issue of a paper currency already depreciated, the public anticipated its further depreciation by releasing such hoards and stocks as were available; in other words, by accelerating the circulation of money. Each man hastened to spend his money before an expected rise of price, and his very action hastened that rise.

Announcements of federal defeats in the war acted in the same way, being signals that further issues of greenbacks might be required; while announcement of victories acted in the opposite way, being like signals of probable redemption.

When appreciation is anticipated, there is a tendency among owners of money to hoard or hold it back, and, among owners of goods, to sell them speedily; the result being to decrease prices by reducing the velocity of circulation and increasing the volume of trade. When, on the contrary, depreciation is anticipated, there is a tendency among owners of money to spend it speedily and among owners of goods to hold them for a rise, the result being to raise prices by increasing the velocity of circulation and decreasing the volume of trade. In other words, the expectation of a future rise or fall of prices causes an immediate rise or fall of prices.

These anticipations respond so promptly to every sign or rumor that superficial observers have regarded the rise and fall of the greenbacks as related directly and solely to expected redemption and as having no relation to quantity. These observers overlook the real mechanism at work; they fail to see that these effects, though quick, are slight and limited. They are the simple adjustments of transition periods described in Chapter IV. It would be a grave mistake to reason, because the losses at Chickamauga caused greenbacks to fall 4 per cent in a single day, that their value had no relation to their volume. This fall indicated a slight acceleration in the velocity of circulation, and a slight retardation in the volume of trade; but, under ordinary circumstances, it is only slightly that the velocity of circulation can thus be accelerated; while to make trade stagnate long or completely would require a cataclysm.

§ 14

In the South it is "impossible to state even approximately how many Confederate treasury notes were outstanding at any time."*107 Professor Schwab has, however, given the value of gold in Confederate currency and index numbers of prices in the South. He concludes:*108—

"This movement of the gold premium corresponds roughly with the amount of government notes outstanding in each period. The relatively rapid increase in the issue of notes after August, 1862, during the last months of 1863, and again during the last months of the war, is reflected in the rapid increase of the gold premium at those three times. When the amount of outstanding notes remained stationary at the beginning of 1863, there was a somewhat slower advance of the gold premium during those months; while the shrinking of the outstanding notes during the first half of 1864 is distinctly reflected in a temporary decline of the premium.

"In the North during the Civil War the course of gold premium only remotely suggested the amount of notes outstanding at any time. The premium rose most rapidly, or, in other words, the notes sank in value most rapidly, at the beginning of 1863, recovering again during the second quarter of that year, declining after August, 1863, to their lowest point in the summer of 1864, and rising again during the last months of the war.*109 The value of the 'greenback' was much more a barometer of popular feeling as to the eventual outcome of the war than a gauge of their amount in circulation, for the latter did not materially increase after July, 1863, and certainly not after July, 1864. In fact, the gold value of the federal 'greenback' ran closely parallel with the gold value of the federal bonds during the war. This is also true of the confederate bonds and treasury notes. These two sets of parallel fluctuations were evidently caused by the changing credit of the two governments concerned.

"A general index number for either section, based both on a simple and a weighted average, can be constructed. The lines plotted to indicate these two sets of figures do not run parallel, but converge and diverge during different periods of the war, converging at those times when events in the military, the political, or the financial field discouraged the South, and correspondingly encouraged the North in the general belief that the war was approaching an end; diverging at those times when federal reverses, or similar events in other than the military field, raised the hopes of the South, and led to belief on both sides that the war would be protracted."*110

We thus see that redundancy of issue produces a rise of prices, not only because of increased quantity, but because of decreased confidence,*111 which affects the sphere of use of the money and therefore the volume of trade performed by the money, and accelerates its velocity.

§ 15

We have given historical instances of the effects on prices of changes in the precious metals and in paper money.

There remain to be considered historical instances of the effects on prices of changes in deposit currency. The price movements due to changes in deposit currency usually include those culminations called crises and depressions.

The economic history of the last century has been characterized by a succession of crises. Juglar in his description of the conditions preceding crises mentions the signs of great prosperity, the enterprise and the speculation of all kinds, the rising prices, the demand for labor, the rising wages, the ambition to become at once rich, the increasing luxury, and the excessive expenditure.*112

A crisis is, as Juglar in fact defines it, an arrest of the rise of prices. At higher prices than those already reached purchasers cannot be found. Those who had purchased, hoping to sell again for profit, cannot dispose of their goods.*113

Our previous analysis has shown us that, before a crisis, while prices are ascending, there is a great increase in bank deposits; and that these, being a circulating medium, accelerate the rise.

It has been pointed out that, with trade international, the rise of prices, resulting from expansion of deposits, is also international. Even if, in some of the countries, deposits should not expand, a rise of the price level would nevertheless occur. The expansion of deposits even in one country of considerable size would, by tending to raise prices there, cause the export of gold. Thus, in other countries the supply of money would increase and prices rise also. This would tend to stimulate expansion of deposits in these other countries and bring about a further rise. Even, therefore, if credit expansion did not begin at the same time in all the principal commercial countries, the beginning of it in one country would be quickly communicated to others. For the same reason the arrest of rising prices and the beginning of falling prices would occur at about the same time in most of the principal countries. As a matter of fact this is what we find to be the case. Juglar has made out a table showing the crises in England, France, and the United States from 1800 to 1882.*114 With the addition of the dates of later crises the table is as follows:—

FRANCE	ENGLAND	UNITED STATES
1804	1803
1810	1810
1813-1814	1815	1814
1818	1818	1818
1825	1825	1826
1830	1830
1836-1839	1836-1839	1837-1839
1847	1847	1848
1857	1857	1857
1864	1864-1866	1864
1873	1873	1873
1882	1882	1884
1889-1890	1890-1891	1890-1891
		1893
1907	1907	1907

§ 16

A study of Juglar's or Thom's tables will show that, in general, bank note circulation and bank deposit circulation increase before a crisis and reach a maximum at the time of the crisis. Index numbers of prices show the same general trend.

Thus,*115 for the United States, the crisis of 1837-1839 shows that circulation of state banks increased each year from 61 millions in 1830 to 149 in 1837 and fell to 116 in the next year; that individual deposits rose each year from 55 millions in 1830 to 127 in 1837 and fell to 84 the next year; that from 1844 to 1848, the date of the next crisis, circulation rose from 75 millions to 128, falling back to 114 the next year, and that the deposits rose from 84 millions to 103, falling back to 91; that from 1851 to 1857, the date of the next crisis, circulation rose from 155 millions to 214, falling the next year to 155, and that the deposits rose from 128 millions to 230, falling the next year to 185. These facts—that prices and deposits rose, culminated, and fell together in reference to the crises of 1837, 1846, and 1857—are confirmed by figures for per capita circulation and deposits given by Sumner.*116 These show the characteristic sharp check to expansion in the crisis years, mild in the mild crisis of 1846 and pronounced in the more pronounced crises of 1837 and 1857. Corresponding phenomena occurred at the next crisis, 1863-1864. After this time, the chief statistics are for national banks, and these show similar results. Thus, from 1868 to 1873, national bank circulation rose from 295 millions to 341 and then fell, while in the same period deposits rose from 532 millions to 656 and then fell. Similar, though less marked, movements occurred in the milder crises of 1884 and 1890, which is the last included in Thom's tables. The crisis of 1893 was exceptional and largely confined to the United States, being chiefly due to the fear as to the stability of the gold standard without much reference to currency and deposit expansion.*117 Whereas in the typical speculative cycle the ratio of deposits to reserves gradually increases until it reaches a maximum just before the crisis, as it did in 1873, 1884, and 1907, this did not happen in 1893. It is true that the deposits of national banks were larger in 1892 than in 1890 or 1891, but they were no larger relatively to reserves, though possibly this fact is to be accounted for by an increase of reserves following the slight crisis of 1890-1891. It is true, also, that the ratio of the deposits of national banks to reserves was high in 1893, but this was due, not to an expansion of deposits, for deposits decreased during that year, but to the runs on the banks and consequent depletion of their reserves.*118 The crisis of 1907, on the other hand, was, like that of 1857, typically a crisis of currency expansion. The facts in reference to this crisis will be discussed more fully in the following chapter.

In France the same tendency of circulation and deposits to reach a maximum at or about a crisis and recede immediately afterward is illustrated fairly well,*119 especially for deposits.

For the Bank of England we find the same general correspondence between crises, circulation, and private deposits.*120

§ 17

Not only do money and deposit currency (M and M') rise regularly to a maximum at the time of a crisis, but their velocity of circulation, so far as statistics indicate, goes through the same cycle. Pierre des Essars has demonstrated this beyond peradventure, so far as velocity of circulation of deposits is concerned.*121

For the United States we have scarcely any statistics of velocity of circulation of deposits, but those for two New Haven banks and for an Indianapolis bank, which I have secured for the last few years, show a maximum in the crisis year 1907.

After a crisis, a decrease occurs in M, M', V, and V'. Bank reserves are increased, and this causes a decrease in M.

Since, then, currency and velocity both increase before a crisis, reach a maximum at the crisis, and fall after the crisis, it is small wonder that prices follow the same course. That they do is the real meaning of a crisis. In fact, as we have seen, Juglar defines a crisis as an arrest of a rise of prices. The index numbers of prices show the rise, maximum, and slump for almost every crisis year for which price statistics exist.*122

The following figures are designed to present a picture of the crisis of 1907 in the United States as illustrating the culmination of a typical credit cycle:—

Year	Deposits*123 of National Banks (millions)	Reserves*123 of National Banks (millions)	Ratio*123 of National Banks to Reserves	Clearings*124 (billions)	M'V'*124 (billions)	Index*125 Number of Prices, P. (Jan.)	Per Cent*125 Rise of Prices during Year.	Money Interest*126 New York Price, m, Two-name 60-day Paper	Virtual*126 Interest
1904	3.31	658	5.0	113	228	113.2 .7	4.2	3.5
1905	3.78	649	5.8	144	279	114.0	5.3	4.3	-1.0
1906	4.06	651	6.2	160	315	120.0	6.6	5.7	-0.9
1907	4.32	692	6.2	145	323	127.9	-1.7	6.4	8.1
1908	4.38	849	5.1	132	294	125.7	—	4.4	—

We notice, in the first column, a steady and rapid increase in the deposits of national banks up to, and including, the crisis year. Though deposits for 1908 do not decrease, yet they remain almost stationary as compared with those of the previous year. The second column, that for reserves, shows, as we should expect, a large increase in the year after the crisis, the banks having fortified themselves against the decrease of business confidence. We find, then (third column), an increase in the ratio of deposits to reserves, the highest ratio being reached in 1906 and 1907, not because reserves were depleted,—on the contrary, they were expanding,—but because deposits were expanding still more rapidly. If the theory presented in Chapter IV is correct, it is precisely this high ratio of deposits to reserves, brought about by failure of interest to rise with rise of prices, which forced the banks to raise their rates of discount and so check further expansion of credit. Then came the crisis and the short succeeding depression. The next column, headed "clearings," is indicative of the volume of check transactions, the circulation of deposit currency. As a fairly constant proportion of checks is settled through the various clearing houses of the country, clearings may fairly be regarded as somewhat of a criterion of M'V'. The fifth column is derived from the fourth and from other data, and is intended as an estimate of M'V'. These two columns increase through 1906, but (since they relate to the whole year and not to a point in the middle of the year) begin to show the effects of the credit slump in the fall of 1907, so that their growth is arrested somewhat in that year, and still more in the year after. We should expect to find, then, a rise of prices reaching a maximum with 1907 and falling in 1908, and this we do find in column six. Column seven shows the per cent rise during each year. Thus, for January, 1904, the index number or P is 113.2, and for January, 1905, it is 114.0. The rise, therefore, is a little less than 1 percent. The minus sign signifies a fall. The eighth column is for rates of interest and indicates, as we should expect, a rise, culminating in 1907. Virtual interest—that is, the interest in terms of commodities—was exceedingly low during the years immediately preceding 1907, because prices were rising so fast. This is shown in column nine, where the nominal interest (measured in money) is corrected by the rise or fall of prices to give interest as measured in actual purchasing power. With the culmination of the cycle in 1907 and the resultant fall of prices, we find virtual interest suddenly becomes very high. No wonder that borrowing enterprisers often found it hard to make both ends meet.

The facts as to credit cycles, then, completely confirm the analysis already given in previous chapters and indicate that prices rise and fall with cycles of currency and velocity. For the benefit of those who doubt whether the expansion of deposit currency raises prices, or whether the rise of prices creates deposit currency, it may be added that facts, as well as theory, show that the former relationship is the true one (although temporarily, as during 1904-1907, there exists a reaction of prices on deposits). Miss England has shown, for instance, that loans and deposits expand before prices rise, and that, though prices often fall before loans and deposits shrink, this anomalous order of events is explainable by the revival of trade following a crisis.*127

No attempt has been made in this chapter to review all the phenomena or even all the typical phenomena of crises. We are not here concerned with crises except in relation to currency. Our concern is with the magnitudes entering the equation of exchange, especially M, M', and V', for these are the immediate elements the variations in which affect the price level and cause it to rise and fall.

§ 18

This chapter has been devoted to an historical study of changes in the quantity of currency and of the effects of these changes on prices. We have seen that, on the whole, increases in the amount of money have tended to raise prices from century to century during the last thousand years, and especially since the discovery of America. The changes in the last century, or more exactly, from 1789 to 1909, have been considered in somewhat more detail, covering five periods of alternately rising and falling prices. We have seen evidence to connect these price movements with changes in the quantity of money and in the volume of business. The periods 1789-1809, 1849-1873, and 1896-1909 were periods of rising prices and large increases of the money supply. In the period 1809-1849 prices fell presumably because of a falling off in gold and silver production and a continuing increase of business; while between 1873 and 1896, although the world's stock of precious metals was increasing slowly, prices in gold countries fell, because in addition to the increasing volume of business there was a stampede of nations to adopt the gold standard and demonetize or limit the coinage of silver.

We have observed the recent continual increase of gold production and found reasons for the tentative prediction that the gold production of the future would continue excessive and probably cause the present rise of prices to continue for some time in the future.

We have described some of the chief examples of paper money inflation and shown that the records for circulation and price changes bear out in a general way the principles set forth in previous chapters. The paper money experiences of France during the French Revolution, of England during the Napoleonic wars, of Austria, the American colonies, the United States, and the Confederacy have been briefly reviewed. We have noted that in these cases, as in others, prices depended on the quantity of money, its velocity, and the volume of business. We have seen that the apparent exceptions due to lack of confidence in paper money are not really exceptions, because lack of confidence works itself out through the magnitudes in the equation of exchange. Distrust increases the velocity of circulation, and decreases the trade performed by the money. We have shown that the general effect of irredeemable paper money issues, which are almost always in large quantities, despite pledges to the contrary, has been to raise prices.

Finally, our study of deposit circulation and crises has afforded further illustration. Preceding a typical crisis, there is, in general, a tendency for deposits to increase and also for their velocity of circulation to increase, while prices tend to rise. Following the crisis comes a decrease in bank deposits and their velocity of circulation, an increase in bank reserves, with a corresponding tendency to diminish money in circulation, and a fall of prices. In the years of the principal crises these took place simultaneously in different countries.