The Purchasing Power of Money/chapter 7

by Irving Fisher

CHAPTER VII INFLUENCE OF MONETARY SYSTEMS ON PURCHASING POWER § 1 VII.1 THUS far we have considered the influences that determine the purchasing power of money when the money in circulation is all of one kind. The illustration given in the previous chapter shows how the money mechanism operates when a single metal is used. We have now to consider the monetary systems in which more than one kind of money is used.

VII.2 One of the first difficulties in the early history of money was that of keeping two (or more) metals in circulation. One of the two would become cheaper than the other, and the cheaper would drive out the dearer. This tendency was observed by Nicolas Oresme, afterwards Count Bishop of Lisieux, in a report to Charles V of France, about 1366, and by Copernicus about 1526 in a report or treatise written for Sigismund I, King of Poland.*66 Macleod in his Elements of Political Economy, published in 1857,*67 before he had become aware of the earlier formulations of Oresme and Copernicus,*68 gave the name "Gresham's Law" to this tendency, in honor of Sir Thomas Gresham, who stated the principle in the middle of the sixteenth century. The tendency seems in fact, to have been recognized even among the ancient Greeks, being mentioned in the "Frogs" of Aristophanes:*69—

"For your old and standard pieces valued and approved and tried, Here among the Grecian nations and in all the world beside, Recognized in every realm for trusty stamp and pure assay, Are rejected and abandoned for the trash of yesterday, For a vile, adulterate issue, drossy, counterfeit, and base Which the traffic of the City passes current in their place." VII.3 Gresham's or Oresme's Law is ordinarily stated in the form, "Bad money drives out good money," for it was usually observed that the badly worn, defaced, light-weight, "clipped," "sweated," and otherwise deteriorated money tended to drive out the full-weight, freshly minted coins. This formulation, however, is not accurate. It is not true that "bad" coins, e.g. worn, bent, defaced, or even clipped coins, will drive out other money just because of their worn, bent, defaced, or clipped condition. Accurately stated, the Law is simply this: Cheap money will drive out dear money. The reason the cheaper of two moneys always prevails is that the choice of the use of money rests chiefly with the man who gives it in exchange, not with the man who receives it. When any one has the choice of paying his debts in either of two moneys, motives of economy will prompt him to use the cheaper. If the initiative and choice lay principally with the person who receives, instead of the person who pays the money, the opposite would hold true. The dearer or "good" money would then drive out the cheaper or "bad" money.

VII.4 What then becomes of the dearer money? It may be hoarded, or go into the melting pot, or go abroad,—hoarded and melted from motives of economy, and sent abroad because, where foreign trade is involved, it is the foreigner who receives the money, rather than ourselves who give it, who dictates what kind of money shall be accepted. He will take only the best, because our legal-tender laws do not bind him.

VII.5 The better money might conceivably be used in exchange at a premium, i.e. at its bullion value; but the difficulties of arranging payments in it, which would be satisfactory to both parties, are such that in practice it is never so used in large quantities. In fact, the force of Gresham's Law is so great that it will even sacrifice the convenience of a whole nation. For instance, in Italy fifteen years ago the overissue of paper money drove not only gold across the Alps, but also silver and copper. These could circulate in Southern France at a par with corresponding coins there because France and Italy belonged to the Latin Union. Consequently, for a time there was very little small change left, below the denomination of 5 lire notes. Customers at retail stores often found it impossible to make their purchases because they lacked the small denominations necessary, and because the storekeeper lacked the same small denominations, and could not make change. To meet the difficulty, 30,000,000 of 1 lire notes were issued, and these were so much in demand that dealers paid a premium for them.

VII.6 Gresham's Law applies not only to two rival moneys of the same metal; it applies to all moneys that circulate concurrently. Until "milling" the edges of coins was invented and a "limit of tolerance" of the mint (deviation from the standard weight) was adopted, much embarrassment was felt in commerce from the fact that the clipping and debasing of coin was a common practice. Nowadays, however, any coin which has been so "sweated" or clipped as to reduce its weight appreciably ceases to be legal tender, and being commonly rejected by those to whom it is offered ceases to be money. Within the customary*70 or legal limits of tolerance, however—that is, as long as the cheaper money retains the "money" power—it will drive out the dearer.

§ 2 VII.7 The obvious effect of Gresham's Law is to decrease the purchasing power of money at every opportunity. The history of the world's currencies is largely a record of money debasements, often at the behest of the sovereign. Our chief purpose now, in considering Gresham's Law, is to formulate more fully the causes determining the purchasing power of money under monetary systems subject to the operation of Gresham's Law. The first application is to bimetallism.

VII.8 In order to understand fully the influence of any monetary system on the purchasing power of money, we must first understand how the system works.*71 It has been denied that bimetallism ever did work or can be made to work, because the cheaper metal will drive out the dearer. Our first task is to show, quite irrespective of its desirability, that bimetallism can and does "work" under certain circumstances, but not under others. To make clear when it will work and when it will not work, we shall continue to employ the mechanical illustration*72 of the last chapter, in which the amount of gold bullion is represented by the contents of reservoir Gb (Figs. 6, 7). Here, as before, we represent the purchasing power or value of gold by the distance of the water level below the zero level, OO. In the last chapter, our figure represented only one metal, gold, and represented that metal in two reservoirs,—the bullion reservoir and the coin reservoir. We shall now, one step at a time, elaborate that figure. First, as in Figure 6 a, we add a reservoir for silver bullion (Sb), a reservoir of somewhat different shape and size from Gb.

This reservoir may be used to show the relation between the value or purchasing power of silver and its quantity in the arts and as bullion. Here, then, are three reservoirs. At first the silver one is entirely isolated; but after a while we shall connect it with the middle one. For the present, let us suppose that the middle one, which contains money, is entirely filled with gold money only (Fig. 6 a), no silver being yet used as money. In other words, the monetary system is the same as that discussed in the last chapter. The only change we have introduced is to add to the picture another reservoir (Sb), entirely detached, showing the quantity and value of silver bullion.

VII.9 We next suppose a pipe opened at the right, connecting Sb with the money reservoir; that is, we introduce bimetallism. Under bimetallism, governments open their mints to the free coinage of both metals at a fixed ratio, i.e. a fixed ratio between the said metals. For instance, if a silver dollar contains 16 grains of silver for every grain of gold in a gold dollar, the ratio is said to be 16 to 1. Under this system, the debtor has the option, unless otherwise bound by his contract, of making payment either in gold or in silver money. These, in fact, are the two requisites of complete bimetallism, viz. (1) the free and unlimited coinage of both metals at a fixed ratio, and (2) the unlimited legal tender of each metal at that ratio.*73 These new conditions are represented in Figure 6 b (and later, Fig. 7 b), where a pipe gives silver an entrance into the money or central reservoir.*74

VII.10 What we are about to represent is not the relations between mines, bullion, and arts, but the relations between bullion (two kinds) and coins. We may, therefore, disregard for the present all inlets and outlets except the connections between the bullion reservoirs and coin reservoir.

VII.11 Now in these reservoirs the surface distances below OO represent, as we have said, purchasing power of gold and silver. But each unit of silver (say each drop of silver water, whether as money or as bullion) contains sixteen times as many grains as each unit of gold (say each drop of gold water, whether as money or as bullion). That is, a unit of water represents a dollar of gold or a dollar of silver. All we wish to represent is the relative purchasing power of corresponding units.

VII.12 The waters representing gold and silver money are separated by a movable film f. In Figure 6 a this film is at the extreme right; in Figure 6 b, at the extreme left; in Figure 7 a, again at the right; and in Figure 7 b, midway.

The a figures represent conditions before the mints are opened to silver. The b figures represent conditions after they have been opened and Gresham's Law has operated. If, just previous to the introduction of bimetallism, the silver level in Sb is below the gold level in Gb the statute introducing bimetallism will be inoperative, i.e. the silver bullion will not flow uphill, as it were, into the money reservoir; but if, as in Figure 6 a or in 7 a, the silver level is higher, then as soon as the mints are open to silver, it will flow into circulation. Being at first cheaper than gold, it will push out the gold money through the left tube (i.e. by melting) into the bullion market. This expulsion of gold may be complete, as shown in Figure 6 b, or only partial, as shown in Figure 7 b. The expulsion will continue just as long as there is a premium on gold; that is, as long as the silver level in the bullion reservoir is above the gold level in the money reservoir; i.e. as long as silver bullion is cheaper than gold money.

VII.13 Let mm, as shown in Figure 6 a, be the mean level; that is, a level such that the volume x above it equals the combined vacant volumes y and z below it. This line, mm, remains the mean level, whatever may be the distribution of the contents among the three reservoirs. As soon as the connecting pipe is inserted, silver will flow into the money reservoir and, in accordance with Gresham's Law, will displace gold.

VII.14 Here we have to distinguish two cases: (1) when the silver x above the mean line, mm, exceeds the total contents of the money reservoir below this line; (2) when x is less than said lower contents. In the first case, it is evident that silver will sweep gold wholly out of circulation, as shown in Figure 6 b, where the film has moved from the extreme right to the extreme left. The contents of silver in the bullion reservoir are less than before, and the contents of gold in the bullion reservoir greater than before.

VII.15 But this redistribution is only the first effect of opening the mints to silver. The balance between production and consumption has been upset both for gold and for silver. The increased value of silver (lowered level in Sb) has stimulated production, bringing into operation silver mines (uncovered inlets at right); and, on the other hand, the decreased value of gold (raised level in Gb) has discouraged gold production, shutting off gold mines (covered inlets at left). Like alterations are effected in the outflows, i.e. the consumption, waste, and absorption of each metal.

VII.16 The result is that the levels resulting from the first redistribution will not necessarily be permanent. They may recede toward their original respective levels, and under all ordinary conditions will do so. But in any case,—and this is the point to be emphasized,—they cannot return entirely to those levels. Such a supposition would be untenable, as the following reasoning shows. Suppose, for the moment, that silver should return to its original level. Then the silver inflow (production) would also return to its original rate dependent on that level, but the silver outflow (consumption, waste, etc.) would be greater than originally. The consumption in the arts would be the same; but the waste and absorption of silver money constitute an additional drain. Therefore, consumption (equal to production before) will now exceed production, and the high original level cannot be maintained. The conclusion follows that, whatever the new level of permanent equilibrium, it lies below the old. The same argument, mutatis mutandis, proves that for gold the new level of permanent equilibrium lies above the old. The gap between the two original levels has therefore been reduced. Even though bimetallism has failed to bring about a concurrent circulation of both metals and a parity of values at the given coinage ratio, it has resulted in reducing the value of the dearer metal (gold) and increasing that of the cheaper (silver). This effect of mutual approach will be referred to in discussing the second case which follows.

§ 3 VII.17 So much for the first case, where x is larger than the contents of the money reservoir below mm. In the second case, x is supposed to be smaller than the contents of the money reservoir below the line mm; that is, there is not enough silver to push all the gold out of circulation. Under these circumstances, disregarding for the moment any change in production or consumption, the opening of the pipe—the opening of the mints to silver—will bring the whole system of liquids to the common level mm. In other words, the premium on gold bullion will disappear (Fig. 7b), and its purchasing power and the purchasing power of silver bullion will be a mean between their original purchasing powers, this mean being the distance of the mean line, mm, below OO. In other words, bimetallism in this case succeeds; that is, it will establish and maintain an equality for a time between the gold and silver dollars in the money reservoir.

VII.18 But the equilibrium which we have just found is a mere equalization of levels produced by a redistribution of the existing stocks of gold and silver among the various reservoirs. It will be disturbed as soon as these stocks are disturbed. A permanent equilibrium requires that the stocks shall remain the same,—requires, in other words, an equality between production and consumption for each metal. After the inrush of silver from the silver bullion to the money reservoir, it is evident that the production and consumption of gold need no longer be equal to each other, nor need the production and consumption of silver be equal to each other. The same stimulation of silver production and discouragement of gold production will occur that occurred in the case considered in the last section. The result may be that silver will, in the end, entirely displace gold; or again it may fail to do so.

VII.19 There may be, then, two possibilities. One possibility is obvious, namely, that gold may be completely driven out, the result being the same as already represented in the lower part of Figure 6. In the second possibility, gold will not be pushed out.

VII.20 The reality of this second possibility will be clear if we attempt first to deny it. Suppose, therefore, that the film f be at the extreme left, and permanent equilibrium finally established. In the illustrative mechanism the gold level will be lower than before, and the silver level higher. How much lower and higher depends evidently on technical conditions of the production and consumption corresponding to the situation. It is of course not inconceivable that the gold level may be so much lower and the silver level so much higher as to make their relative positions reversed, i.e. to make the gold level higher than the silver level. But in this event it is quite impossible that the film f should be at the left. Gold, now being the cheaper, would flow into circulation and displace silver. Under the conditions we are now imagining, the film cannot stay at either extreme. If it is at the right, silver will be cheaper than gold and will move it leftward; if it is at the left, gold will be cheaper than silver and will move it rightward. Under these circumstances, evidently, equilibrium must lie between these extremes, as in Figure 7 b. The conditions of production and consumption under which bimetallism can succeed are therefore (1) that under silver monometallism a gold dollar would in equilibrium be cheaper than a silver dollar, and (2) that under gold monometallism silver would be cheaper than gold. A bimetallic level, therefore, when bimetallism is feasible, must always lie between the levels which the two metals would have assumed under gold monometallism, gold being currency and silver not, and for the same reasons it must lie between the levels which the two metals would have under silver monometallism, silver being currency and gold not.*75 In all our reasoning we have supposed a given legal ratio between the two metals. But bimetallism, impossible at one ratio, is always possible at another. There will always be two limiting ratios between which bimetallism is possible.*76

VII.21 It is easy to show that the two limiting ratios for a single nation are narrower than for a combination of nations, since the currency reservoir is, for one nation, smaller than for many, while the arts reservoirs are virtually larger by the amount of the monometallic currencies of the remaining nations. When bimetallism has broken down at one ratio, it can always be set in operation again at another, but the transition requires a depreciation of the currency. The only way of reintroducing the metal which has passed out of the currency reservoir is by lowering the amount of it in the monetary unit,—unless the still more drastic measure is adopted of raising the coinage weight of money already in circulation.

VII.22 It should also be pointed out that two nations cannot both maintain bimetallism at two different ratios unless the difference is less than the cost of shipment. One of the two nations would lose the metal which it undervalued and find itself on a monometallic basis.

VII.23 A few additional observations may now be stated. The temporary and normal equilibriums which have been considered separately are in fact quite distinctly separated in time. The time of redistributing existing stocks of metal, according to a newly enacted law, depends on the rapidity of transportation, melting, and minting, and would be measured in months or weeks. Normal equilibrium, however, depends on the slow working of changes in the rates of production and consumption, and would be measured in years. The normal equilibrium, if once established, is permanent so long as the conditions of production and consumption do not change. Slight alterations of these conditions—the exhaustion of mines, the discovery of new leads, etc.—will cause slight variations in the proportions of gold and silver money, that is, in the position of the film f. The oscillations of this film (and not of the price ratio as in the case of two unconnected commodities) reflect these changing conditions. But, in all probability, this film will sooner or later reach one of its limits. The probable time for such an event is, however, very long. The gold currency of the world is, roughly speaking, perhaps $5,000,000,000; the annual production of silver, reckoning at its present market price, is roughly about $100,000,000. Supposing a system of international bimetallism at, say, 36 to 1 to be initially in normal equilibrium, consider the effect of an enormous increase in the silver production, say a half, or $50,000,000. Then a hundred years would be required to push out gold without taking into account the fact that, as the pushing proceeds, the excess of production over consumption steadily declines. If this excess dwindles uniformly from $50,000,000 to zero, the period would be double, or two hundred years. When we add to these considerations the fact that, while the stimulus to the production of one metal acts quickly, the ensuing check to the production of the other acts more slowly, owing to the fixity of the "sunk" capital, and that, therefore, the volume of the currency is greater at the end than at the beginning; also the fact that the currency reservoir is itself constantly expanding; and finally the fact that fluctuations of production are likely to be in either direction, and for either metal, we may be tolerably confident that, if initially successful with the film near the middle position, international bimetallism would continue successful for many generations. The initial success depends, as has been seen, upon the ratio enacted.

VII.24 It is to be observed that bimetallism can never avoid a slight premium. On the contrary, it is this difference of level which supplies the force which compells change from one point of equilibrium to another.*77

VII.25 In a series of years, the bimetallic level remains intermediate between the changing levels which the two metals would separately follow. Bimetallism spreads the effect of any single fluctuation over the combined gold and silver markets.*78 The steadying power of bimetallism depends on the breadths of the reservoirs, and not on the position of the film f. It remains in full force, no matter what may be the proportions of gold and silver money, and is as great when only one nation is bimetallic as when the whole world adopts the system. Even if only Switzerland had a system of bimetallism in successful operation, it would, until its breakdown, keep together and equalize the currencies of the whole world, wherever either gold or silver was standard. In fact, the world would have all the benefits of bimetallism enjoyed by Switzerland without its evils and dangers. This international function would cease abruptly as soon as the system of bimetallism should fail.

VII.26 It should be pointed out that the equalizing effect maintained is relative only. It is conceivable that one metal would be steadier alone than when joined to the other. In a later chapter we shall consider the extent to which this equalization is an advantage. Here we confine ourselves to showing merely the mechanical operation of the bimetallic system.*79

§ 4 VII.27 Bimetallism is to-day a subject of historical interest only. It is no longer practiced; but its former prevalence has left behind it in many countries, including France and the United States, a monetary system which is sometimes called the "limping" standard. Such a system comes about when, in a system of bimetallism, before either metal can wholly expel the other, the mint is closed to the cheaper of them, but the coinage that has been accomplished up to date is not recalled. Suppose silver to be the metal thus excluded, as in France and the United States. Any money already coined in that metal and in circulation is kept in circulation at par with gold. This parity may continue even if limited additional amounts of silver be coined from time to time. There will then result a difference in value between silver bullion and silver coin, the silver coin being overvalued. This situation is represented in Figure 8.

Here the pipe connection between the money reservoir and the silver-bullion reservoir has been, as it were, cut off, or, let us say, stopped by a valve which refuses passage of silver from the bullion reservoir to the money reservoir but not the reverse (for no law ever can prevent the melting down of silver coins into bullion). Newly mined silver cannot now become money, and thus lower the purchasing power of the money.

VII.28 On the other hand, new supplies of gold continue to affect the value of currency, as before,—the value, not only of the gold, but also of the concurrently circulating overvalued silver. If more gold should flow into the money reservoir, it would raise the currency level. Should this level ever become higher than the level of the silver bullion reservoir, silver would flow from the money reservoir into the bullion reservoir; for the passage in that direction (i.e. melting) is still free. So long, however, as the currency level is below the silver level, i.e. so long as the coined silver is worth more than the uncoined, there will be no flow of silver in either direction. The legal prohibition prevents the flow in one direction, and the laws of relative levels prevent its flow in the other.

VII.29 In the case just discussed, the value of the coined silver will be equal to the value of gold at the legal ratio. Precisely the same principle applies in the case of any money, the coined value of which is greater than the value of its constituent material. Take the case, for instance, of paper money. So long as it has the distinctive characteristic of money,—general acceptability at its legal value,—and is limited in quantity, its value will ordinarily be equal to that of its legal equivalent in gold. If its quantity increases indefinitely, it will gradually push out all the gold and entirely fill the money reservoir, just as silver would do under bimetallism if produced in sufficiently large amounts. Likewise, credit money and credit in the form of bank deposits would have this effect. To the extent that they are used, they lessen the demand for gold, decrease its value as money, and cause more of it to go into the arts or to other countries.

VII.30 So long as the quantity of silver or other token money, e.g. paper money, is too small to displace gold completely, gold will continue in circulation. The value of the other money in this case cannot fall below that of gold. For if it should, it would, by Gresham's Law, displace gold, which we have supposed it is not of sufficient quantity to do. The parity between silver coin and gold under the "limping" standard is, therefore, not necessarily dependent on any redeemability in gold, but may result merely from limitation in the amount of silver coin. Such limitation is usually sufficient to maintain parity despite irredeemability. This is not always true, however; for if the people should lose confidence in some form of irredeemable paper or token money, even though it were not overissued, it would depreciate and be nearly as cheap in money form as it is in the raw state. A man is willing to accept money at its face value so long as he has confidence that every one else is ready to do the same. But it is possible, for instance, for a mere fear of overissue to destroy this confidence. The payee, who, under ordinary circumstances, submits patiently to whatever money is customary or legal tender, may then take a hand and insist on "contracting out" of the offending standard.*80 That is, he may insist on making all his future contracts in terms of the better metal,—gold, for instance,—and thus contribute to the further downfall of the depreciated paper.

VII.31 Irredeemable paper money, then, like our irredeemable silver dollars, may circulate at par with other money, if limited in quantity and not too unpopular. If it is gradually increased in amount, such irredeemable money may expel all metallic money and be left in undisputed possession of the field.

VII.32 But though such a result—a condition of irredeemable paper money as the sole currency—is possible, it has seldom if ever proved desirable. Unless safeguarded, irredeemability is a constant temptation to abuse, and this fact alone causes business distrust and discourages long-time contracts and enterprises. Irredeemable paper money has almost invariably proved a curse to the country employing it. While, therefore, redeemability is not absolutely essential to produce parity of value with the primary money, practically it is a wise precaution. The lack of redeemability of silver dollars in the United States is one of the chief defects in our unsatisfactory monetary system, and a continuing danger.

VII.33 It is possible to have various degrees of redeemability. One of the most interesting systems of partial redeemability is the system now known as the gold-exchange standard, by which countries, not themselves on a strict gold basis, nevertheless maintain substantial parity with gold through the foreign exchanges. By this system the government or its agent, while not redeeming its currency in gold, redeems it in orders on gold abroad. That is, the government sells bills of exchange on London or New York at a stated price. The currency which it thus receives, and in a sense redeems, it keeps out of circulation until the price of foreign exchange falls (i.e. until the demand for redemption ceases).

VII.34 The gold-exchange standard may be regarded as a kind of limping standard with the added feature of partial redemption.

VII.35 This added feature, however, greatly modifies the nature of the limping standard. The limping standard without the gold-exchange attachment may at any time break down, if the silver (or whatever else the overvalued money may be) should become so redundant, relatively to trade, as completely to displace gold. As soon as all gold is driven abroad, parity with gold ceases. But with the gold-exchange system this catastrophe is avoided. In fact, with this system it is not necessary to have gold in circulation at any time. The willingness of the government to sell foreign exchange at a fixed price, and to lock up the silver it receives thereby, takes that much currency out of circulation just as effectively as though the equivalent of gold had been exported. So long as the government is willing and able to maintain the price of bills of exchange with a gold country, it, ipso facto, maintains approximate parity with gold.*81

§ 5 VII.36 We have now to illustrate, by historical examples, the principles just explained. The first and most important case is that of France. The ratio of 15½ to 1 was adopted by France in 1785 and continued by the law of 1803. The history of France and the Latin Union during the period from 1785, and especially from 1803, to 1873 is instructive. It affords a practical illustration of the theory that when conditions are favorable, gold and silver can be kept tied together for a considerable period by means of bimetallism. During this period the public was ordinarily unconscious of any disparity of value, and only observed the changes from the relative predominance of gold to the relative predominance of silver in the currency and vice versa. In the wholesale bullion market, it is true, there were slight variations from the ratio of 15½ to 1. But such variations simply supplied the force to restore equilibrium.

VII.37 From 1803 until about 1850 the tendency was for silver to displace gold. In our mechanical terms there was, for the most part, an inflow on the right-hand side of the money reservoir, and the film was gradually pressed leftward. The statistics for the movements of gold and silver are not given separately and continuously before 1830. But from 1830 to 1847, inclusive, there was a net export of gold of 73,000,000 francs, although five of the years showed an import, making an average export of over 4,000,000 francs a year.*82 From 1830 until 1851 there was a net importation of silver in every year, amounting to a total for the period of 2,297,000,000 francs or an average of over 104,000,000 francs a year.*83 The statistics for silver are taken to 1851 because after that year the movement for silver was reversed, while for gold the inward flow began with 1848. Silver was displacing gold and filling up the currency reservoir. Nevertheless, the reservoir was expanding so fast, that is, trade was increasing, that there was no increase of prices, but rather a decrease. By 1850, the film had practically reached its limit. Bimetallism would have broken down and resulted in silver monometallism then and there, except for the fact that, as though to save the day, gold had just been discovered in California. The consequence of the new and increased gold production was a reverse movement, an inflow of gold into the French currency and an outflow of silver. From 1848 to 1870, inclusive, the net importation of gold amounted to 5,153,000,000 francs or over 224,000,000 francs a year, while the net exportation of silver from 1852 to 1864, inclusive, amounted to 1,726,000,000 francs or nearly 133,000,000 francs a year.*84 Gold was displacing silver and filling the currency. It seemed probable that France would be entirely drained of her silver currency and come to a gold basis. France formed with Belgium, Italy, and Switzerland in 1865, and Greece in 1868, the Latin Monetary Union. The amount of silver in the subsidiary coins was reduced, but the standard silver coins were kept at the old ratio with gold. But the new gold mines were gradually exhausted, while silver production increased, with the consequence that there was again a reversal of the movement. From 1871 to 1873, inclusive, the exportation of gold netted 375,000,000 francs, or an average of 125,000,000 francs a year, while from 1865 to 1873, inclusive, the net importation of silver was 860,000,000 francs, or over 94,000,000 francs a year. Thus, even before the gold began to flow out, in 1871, silver had begun to flow in, i.e. in 1865. Silver gradually pushed gold out of circulation and, had not France and the other countries of the Latin Union successively suspended the free coinage of silver in 1873-1878, they would have found themselves on a silver, instead of a gold, basis. It has been claimed by bimetallists that this action in demonetizing silver was itself the cause of the breakdown. The truth is, that the breakdown was the cause of demonetization, although demonetization, by keeping back silver from circulation and keeping gold in circulation, did operate to widen the breach already made.

VII.38 The film, in other words, was close to the left limit, and the currency reservoir was filled for the most part with silver. The Latin Union might conceivably have maintained bimetallism longer if other countries had joined with them. But it had to absorb, not only much of the silver provided by the mines, but also a considerable amount which had previously formed part of the monetary stock of Germany and which, at the adoption of the gold standard by that country following the Franco-Prussian War, was thrown on the market. That is, not only silver mines, but countries demonetizing silver, dumped silver on the Latin Union. Add to this the movement toward the gold standard in Scandinavia and the United States, and it becomes evident that the obstacles were many for a union comprising so few, and mostly unimportant, states.

VII.39 The parity with gold of the silver remaining in circulation in the Latin Union is now preserved on the principles explained earlier in the chapter, viz. by limiting its quantity, as well as by making it full legal tender and receivable for public dues.

§ 6 VII.40 It is strange that the lessons of the French and other experiments do not seem to be generally understood either by monometallists or bimetallists. For instance, uncompromising monometallists have pointed to the variation in the value of gold and silver during the three quarters of a century as disproving the possibility of maintaining a legal ratio. They might as well point to the ripples on a pond or the slight gradient of a river, as disproving the fact that water seeks a level. These ripples are really evidence of the process of seeking a level and are trifling as compared with those which in all probability would have taken place had there not been a legal ratio. The diagram and tables used in Shaw's History of the Currency and the similar diagram here given show that during the period of inflowing silver, 1803 to 1850, in spite of the great increase in the quantity of silver, the ratio was changed from 15½ to 1 by at most only .75 points or slightly over 4.8 per cent in any year, and the average departure was only .29 points or 1.9 per cent. Moreover, the greater part of the deviation is explainable by the seigniorage charge then in force in France.*85 During the succeeding period from 1851 to 1870, characterized largely by an inflow of gold, the maximum departure (in the opposite direction) was .31 points or 2 per cent, with an average departure of .14 points or .9 per cent, while during the succeeding period of inflowing silver and outflowing gold, from 1871 to 1873, the ratio rose above 15½ to 1 by a maximum of .42 points or 2.7 per cent and an average of .21 points or 1.4 per cent. Contrast these figures with those since 1873.*86 The maximum departure from the ratio of 15½ to 1 since 1873 is 23.65 points, or 152.6 per cent, and the average departure 10.4 points, or 67.1 per cent.*87 The history of the ratio is shown in Figure 9.

VII.41 On the other hand, bimetallists have often failed to see that this experiment illustrates the limits as well as the possibilities of bimetallism. In 1850 bimetallism had almost broken down in France and would have been succeeded by silver monometallism had not the increased production of gold reversed the flow. In 1865, gold had largely driven out silver. By 1873, gold had again largely disappeared, and it seems evident that it would have disappeared entirely had not the suspension of the free coinage of silver followed. A continuance of bimetallism at a ratio of 15½ to 1 by France and the Latin Union alone would doubtless have been impossible. Yet the attempt, though a failure, would have kept the ratio nearer 15½ to 1 than it actually has been; for the Union would have furnished a large market for silver. Possibly bimetallism could have been maintained longer, despite increased silver production, had not several other countries adopted the gold standard in these critical years. This fact helped to flood the countries of the Latin Union with silver and drain them of their gold. These countries were suffering all the expense and trouble of maintaining the ratio between gold and silver, while other countries were reaping most of the benefits. Herein lies one of the weaknesses of bimetallism as a practical political proposition,—each country prefers that some other country or countries should be the ones to adopt it. There is little prospect, therefore, in the future, of any single country taking the initiative, and still less of any international agreement.

§ 7 VII.42 The system now in use in France is also employed in many other countries which, like France, have been forced to adopt it or else become silver-standard countries. After the rupture of the bimetallic tie, which until 1873 linked all gold and silver countries together, the commercial world broke into two parts, gold-standard countries and silver-standard countries; and many desiring to join the ranks of the former, but in danger of being thrust among the latter, saved themselves by closing their mints to silver and thereby adopting the limping standard. One of these countries was British India.

VII.43 The case of India is interesting because it never was a bimetallic country, and at the time of the adoption of the present system, in which gold is the standard, no gold was in circulation. The mints were closed to silver in June, 1893, and the legal ratio put the rupee at 16d. At first, to the great discomfiture of those who had advocated the new system, this value was not maintained. But failure at first was to be expected because no gold was in circulation, and unsuspected coined stores of silver existed to swell the circulation in spite of the closure of the mints. Moreover, the government accepted from banks and others considerable amounts of silver which had been shipped to India before the closing of the mints, and coined it, and a considerable amount was withdrawn from the government reserve and put into circulation. The value of the rupee fell by 1895 to as low as 13d. But even from the first the value of the rupee kept above the value of its contained silver. If it fell as compared with the then appreciating gold, it rose as compared with the value of silver bullion. Surely this may be held to show that the value of money has some relation to its quantity, apart altogether from the quantity and the value of the constituent material. Furthermore, the value of the rupee rose gradually, even in relation to the gold standard, from 13d. in 1895 to 15 1/3; in 1898 and to 16d., the legal par, by 1899, where it has since remained. As the Indian government has, during the last decade, paid out rupees for gold on demand, at this rate, the value of the rupee cannot go appreciably higher. Should it do so, gold would be presented for rupees, more rupees would have to be issued, and this would continue until their value had fallen to 16d. per rupee.*88

VII.44 The system of India is virtually the gold-exchange standard described in § 4. The same system is now in successful operation in the Philippines, in Mexico, and in Panama.*89 It withstood a severe test in India when, in 1908, the trade balance was "adverse" and required the sale of over £8,000,000 of bills on London before the Indian currency was sufficiently contracted to stem the tide.

§ 8 VII.45 Among the nations which now have the limping standard is the United States. In 1792, Congress adopted complete bimetallism. Full legal-tender quality was given to both gold and silver coins; both were to be coined freely and without limit at the ratio of 15 ounces of silver to 1 of gold.

VII.46 This soon came to be below the market ratio as affected by conditions abroad and especially in France. In consequence, gold tended to leave the country. It is impossible to state with exactness how soon this movement began, but Professor Laughlin sets it as early as 1810 and concludes that by 1818 little gold was in circulation.*90 America, although nominally a bimetallic country, became actually a silver country.

VII.47 Influenced partly by the desire to bring gold back into circulation, and partly also, perhaps, by the supposed discoveries of gold in the South, Congress passed acts in 1834 and 1837 establishing the ratio of "16 to 1,"—or, more exactly, 16.002 to 1 in 1834 and 15.998 to 1 in 1837. Whereas silver money had been over-valued by the previous laws, by these new laws gold was overvalued. That is, the commercial ratio continued to be near 15½ to 1, while the monetary ratio was slightly greater. This remained the case up to 1850; consequently, in accordance with Gresham's Law, gold money, now the cheaper, drove out silver money, and the United States became a gold-standard country. In 1853, to prevent the exportation of our subsidiary silver coins, their weight was reduced.

VII.48 The United States continued to be a gold-using country until the period of the Civil War, during which "greenbacks," or United States notes, were issued in considerable excess. Again Gresham's Law came into operation. Gold was in turn driven from the currency, and the United States came to a paper standard.*91 For some years after the close of the war the country remained on a paper standard, little gold being in circulation except on the Pacific coast, and not much silver anywhere.

VII.49 In 1873 Congress passed a law (called by bimetallists the "Crime of '73") by which the standard silver dollar was entirely omitted from the list of authorized coins.

VII.50 Of course this could not have had any immediate effect on the value of gold and silver, because the country was at the time on a paper basis. But when specie payments (i.e. gold and silver payments) were resumed in 1879, this repeal of the free coinage of silver brought the country to a gold standard, not to a silver one. Had it not been for the law of 1873, the United States, when it returned in 1879 to a metallic basis, would have been a silver country with a standard considerably below the gold standard it actually reached. Our monetary problems would then have been very different from what they actually became.

VII.51 But in returning to a gold basis we reintroduced the silver dollar in a minor rôle. Although the free coinage of silver was not resumed, the advocates of silver, through the "Bland-Allison Act" of 1878 and the "Sherman Act" of 1890, which replaced it, succeeded in pledging the government to the purchase of large, but not unlimited, amounts of silver and the coinage of a large, but not unlimited, number of silver dollars. The Bland-Allison Act required the Secretary of the Treasury to purchase every month from $2,000,000 to $4,000,000 worth of silver and to coin it into standard silver dollars. The Sherman Act required the purchase every month of 4,500,000 ounces of silver.

VII.52 Under these acts 554,000,000 silver dollars were coined, although less than 20 per cent of them have ever been in actual circulation. Silver certificates redeemable in silver dollars on demand, and, for a time treasury notes, have circulated in the place of this immense mass of silver. The silver dollars (and therefore the silver certificates) maintain their value on a parity with gold primarily because they are limited in amount. If any question were raised as to their parity with gold, the treasury would probably offer to specifically redeem them in gold. No law directly provides for the redemption of silver in gold, but it is made the duty of the Secretary of the Treasury to take such measures as will maintain its parity with gold.

VII.53 In 1893 the Sherman Act was repealed, and in 1900 a law was passed specifically declaring that the United States shall be on a gold basis.

§ 9 VII.54 The system of the limping standard, now obtaining in the United States, logically forms a connecting link between complete bimetallism and those "composite" systems by which any number of different kinds of money may be simultaneously kept in circulation. The manner in which most modern civilized states have solved the problem of concurrent circulation has been to use gold as a standard, and to use silver, nickel, and copper chiefly as subsidiary money, limited in quantity, with, in most cases, limited amounts of paper money, the latter being usually redeemable. The possible variations of this composite system are unlimited. In the United States at present we have a system which is very complicated and objectionable in many of its features—especially (as we shall presently see) in its lack of elasticity. Gold is the standard and is freely coined. A limited number of silver dollars, worth, moneywise, more than double their value bullion-wise, are a heritage of former bimetallic laws long since rendered inoperative by the paper money of the Civil War and expressly repealed in 1873. The two attempts of 1878 and 1890 to return halfway to bimetallism by the purchase of silver—attempts discontinued in 1893—have greatly swollen the volume of coined silver. The attempt to force silver dollars into circulation was not acceptable to the business world, and Congress therefore issued instead the two forms of paper mentioned. The chief form is the "silver certificate." For each silver certificate a silver dollar is kept in the vaults of the United States government.

VII.55 The absurdity of the situation consists in the fiction that somehow the silver keeps paper at par with gold. The paper would keep its parity with gold just as well if there were no silver. A silver dollar as silver is worth less than a gold dollar just as truly as a paper dollar, as paper, is worth less than a gold dollar. The fact that the silver is worth more than the paper will not avail in the least to make the paper worth a whole dollar so long as the silver is not itself worth a whole dollar. A pillar which reaches only halfway to the ceiling cannot hold the ceiling up any more than a pillar an inch high.

VII.56 The paper representatives of silver would continue to circulate as well as they do now, even if the "silver behind them" were nonexistent, although the absurdity of the situation would then be so apparent that they would probably be retired. Whether the half billion dollars of new currency, which came into circulation with the Bland and Sherman Acts, are of silver, overvalued to the extent of 50 per cent, or of paper, overvalued to the extent of 100 per cent, does not really affect the principle of the limping standard which keeps silver dollars at par with gold. The idle silver in the treasury vaults represents mere waste, a subsidy given by the government to encourage silver mining. Its only real effect to-day is to mislead the public into the belief that in some way it keeps or helps to keep silver certificates at par with gold;*92 whereas they are kept at par by the limitation on its amount. The silver and its paper representatives cannot fall below par without displacing gold, and they cannot displace gold because there are not enough of them.

VII.57 Another and equally useless anomaly is the existing volume of "greenbacks." These are United States government notes. Under the law of 1875, the greenbacks were by 1879 retired in sufficient numbers to restore parity with gold; but by a counterlaw of 1878, 347,000,000 of them were kept in circulation and are in circulation now. As soon as redeemed, they must be reissued; they cannot be retired. They are a fixed ingredient in our money pot pourri, neither expansive nor shrinkable. They have been kept at par with gold because: (1) they are limited in amount; (2) they are redeemable in gold on demand; (3) they are receivable for taxes and are legal tender. But it is absurd to redeem but not retire—in fact, almost a contradiction in terms. This absurdity has at times seriously embarrassed the government.

VII.58 The next feature of our currency to be considered is the bank note. Although the National Bank acts wiped out the old, ill-assorted state bank notes, they tied the new notes up with the war debt, and they have remained so tied ever since, in spite of the fact that the advantages of the connection have long been terminated and the disadvantages have grown acute. National bank notes cannot legally be issued in excess of the government debt, however urgent the need for them; nor can the government pay its debt without thereby compelling national banks to cancel their notes.

VII.59 One of the curious anomalies of the situation is that the prices of United States bonds are so high, and therefore the rate of interest returned on these bonds so low that there is actually less inducement to issue bank notes in regions where the rate of interest is high, as in the West, than in regions where it is low, as in the East.

VII.60 The result is an inelastic currency which, instead of adjusting itself to the seasonal fluctuations in trade, and thus mitigating the ensuing variations in the price level, remains a hard and solid mass to which the other elements in the equation of exchange must adapt themselves.*93

VII.61 The remaining features of our currency system, such as fractional and minor coins, adjusted to public demand, are satisfactory. The gold and currency certificates of deposit are scarcely independent features, as they are simply government receipts for public convenience representing the deposit of gold or greenbacks.

VII.62 The status in the United States July 1, 1912, is represented in the table on the opposite page taken from Comtrollers' Reports and Treasurers' Reports.

VII.63 We see here a currency system in which gold, the basis of it all, enters as between one third and one half the circulation outside of the Treasury and banks and a little over half the money in banks used as reserves for deposits (and for bank notes, though these are also guaranteed by the government). The remaining money in circulation consists almost wholly of inelastic and almost constant elements. Consequently, a change in the quantity of gold in circulation will not cause a proportional change in the quantity of all money in circulation, but only about one third as much. Since, however, almost all the money can be used as bank reserves, even national bank notes being so used by state banks and trust companies, the proportionate relations between money in circulation, money in reserves, and bank deposits will hold true approximately as the normal condition of affairs. The legal requirements as to reserves strengthen the tendency to preserve such a relationship.

MONEY IN THE UNITED STATES (in Millions)

	GOLD	SILVER	U.S. NOTES	BANK NOTES	SUBSIDIARY AND MINOR COINS	TOTAL

In U.S. Treasury 264*94 26*97 9 40 26 365 In banks 801*95 217*98 253 108 38 1417 Outside both 752*96 323*99 85 597 107 1864 Total 1817 566 347 745 171 3646

VII.64 In the United States, then, we have a currency system in which, as has been observed, the only really adjustable money is gold. As gold requires time for minting or transportation, the adjustment is slow and clumsy as compared with the prompt issue or retirement of bank notes practiced in other countries. The seasonal changes in the purchasing power of money, as well as the changes connected with crises and credit cycles, are therefore greatly and needlessly aggravated. As this second edition goes to press, there seems a likelihood that Congress may at last enact legislation designed to remedy this condition.

Endnotes

66. Henry Dunning Macleod, The History of Economics, New York (Putnam), 1896, pp. 37 and 38.

67. P. 477.

68. Macleod, The History of Economics, pp. 38 and 39.

69. 893-898, Frere's translation.

70. Sometimes custom is less strict than law. For instance, in California worn gold coin below the mint limit of tolerance continues to circulate. It is called "bank gold."

71. Irving Fisher, "The Mechanics of Bimetallism," (British) Economic Journal, September, 1894, pp. 527-536.

72. In its present application it is somewhat like a symbolism suggested by Jevons in his Money and the Mechanism of Exchange, New York (Appleton), 1896, p. 140.

73. The possibility of government's fixing any ratio between gold and silver to which the market ratio will conform has been so bitterly disputed that in addition to the positive argument contained in the text, a negative criticism of what are believed to be the chief fallacies underlying these disputes is inserted in § 1 of the Appendix to (this) Chapter VII.

74. Of course a unit of water represents gold and silver at their coining weights. If the bimetallic ratio is 16 to 1, the cisterns must be so constructed that a cubic inch of water shall represent an ounce of gold or 16 ounces of silver and that the number of inches separating the surfaces of liquid from OO shall represent the marginal utility of an ounce of gold and of 16 ounces of silver, respectively.

75. But it does not necessarily lie between the level of the currency under gold monometallism and its level under silver monometallism.

76. See § 2 of Appendix to (this) Chapter VII.

77. As long as the premium lasts, the cheaper metal will doubtless circulate somewhat faster, and the dearer somewhat more slowly, than when there is no premium. This may, if desired, be represented by conceiving the thickness of the currency reservoir to decrease on the one side of f and increase on the other, making one metal "go farther" (cover more area in the diagram) than normally, and hasten the motion of f to the equilibrium point. Sluggishness (increase of thickness), of which "hoarding" is the chief application, is referred to below.

78. To represent the steadying effect on a single fluctuation, we observe that under bimetallism the three reservoirs act as one. Therefore, compared with monometallism, the fluctuations are diminished in the inverse ratio of the liquid surfaces over which the fluctuations spread. Thus, if the combined breadths of the two left reservoirs at water level are two thirds of the combined breadths of the three, an influx of gold which, if distributed only over the two reservoirs, would make a layer an inch in depth, would, over the three, have a depth of two thirds of an inch. Likewise the right reservoir being one third the aggregate width, an inch fluctuation of silver, when merely merchandise, would be reduced to one third of an inch, when connection is maintained with the money reservoir. The breadths of the reservoir, which here play the important rôle, are the rate of increase of commodity relative to decrease of marginal utility. The law of inverse breadth applies with exactness only to static, or short-time readjustments.

79. Cf. Leonard Darwin, Bimetallism, London (Murray), 1897, 341 pp.; Bertrand Nogaro, "L'expérience bimétalliste," Revue d'êconomie politique, 1908.

80. In the mechanical interpretation, since the law would now be inoperative, the film would no longer yield to pressure from the right, and an appreciable difference of level on its two sides would ensue. Such a mechanism would illustrate the concurrent circulation of two metals at independent valuations, but experience shows that such a condition is too full of inconveniences to be maintained long. The silver will be progressively tabooed, that is, its velocity of circulation will gradually decrease. This, as we have seen, may be figured by supposing the currency reservoir (on the right of the film only) to be thickened, thus bringing the film to the right. If the silver is completely tabooed, the film will be completely to the right, and there will be gold monometallism. The monetary result of such discrimination against silver is thus the appreciation of gold.

81. Cf. Charles A. Conant, "The Gold Exchange Standard," Economic Journal, June, 1909, pp. 190-200.

82. W. A. Shaw, The History of the Currency, 3d ed., London (Clement Wilson), 1899, pp. 183.

83. Ibid., p. 184.

84. W. A. Shaw, The History of the Currency, 3d ed., London (Clement Wilson), 1899, pp. 183 and 184.

85. Cf. J. F. Johnson, Money and Currency, Boston (Ginn) 1905, p.227.

86. Although France did not entirely suspend the coinage of the silver five-franc pieces until 1876, yet limitation began with 1874. See W. A. Shaw, The History of the Currency, pp. 194, 196.

87. These figures are compiled from data given in W. A. Shaw, The History of the Currency, p. 159, and Reports of the Director of the Mint.

88. For a brief history and discussion of the Indian experience, see E. W. Kemmerer, Money and Credit Instruments in their Relation to Prices, pp. 36-39.

89. See Charles A. Conant, "The Gold Exchange Standard in the Light of Experience," Economic Journal, June, 1909, pp. 190-200; Hanna, Conant, and Jenks, Report on the Introduction of the Gold Exchange Standard into China, the Philippine Islands, Panama, and other Silver-using Countries and on the Stability of Exchange, Washington (Government Printing Office), 1904; Kemmerer, "Establishment of the Gold Exchange Standard in the Philippines," Quarterly Journal of Economics, August, 1905, pp. 600-605.

90. The History of Bimetallism in the United States, New York (D. Appleton and Company), 1901, 4th ed., p. 29.

91. See especially Wesley Clair Mitchell, History of the Greenbacks, Chicago (The University of Chicago Press), 1903; also his Gold, Prices, and Wages under the Greenback Standard, Berkeley, California (University of California Press), 1908.

92. Seager, in his Introduction to Economics, 3d ed., New York (Holt), 1908, p. 317, urges that the government should attempt gradually to dispose of this silver and substitute an equal value of gold.

93. The Aldrich-Vreeland Act of 1908 has not changed this situation. However helpful it may be in mitigating the evils of crises, it does not give elasticity of the currency in ordinary times.

94. "Free" gold (i.e. exclusive of gold held in Treasury for gold certificates held by the public).

97. "Free" (i.e. exclusive of silver held in the Treasury for silver certificates held by the public).

95. Including 563 millions of gold certificates for which gold was on deposit in the United States government vaults.

98. Including 194 millions of silver certificates for which silver was on deposit in the United States government vaults.

96. Including 380 millions of gold certificates for which gold was on deposit in the United States government vaults.

99. Including 275 millions of silver certificates for which silver was on deposit in the United States government vaults.