Читать книгу Readings in Money and Banking - Chester Arthur Phillips - Страница 62

INTERNATIONAL BIMETALLISM

Оглавление

Table of Contents

[15] … There are natural and commercial causes which may operate to produce either an incessant fluctuation in the relative value of silver and gold, or a wide and increasing divergence, from year to year, through a long period, from the ratio of exchange existing between the two metals at the commencement of the period. So far are the sources and conditions of supply of the one different from those of the other that, notwithstanding the influence of the durableness of the metals in giving steadiness of value to either by turns, and hence to the two in their relation to each other, it would be in the highest degree unreasonable to assume that the ratio of exchange between gold and silver would remain unaltered through any considerable term of years. The annual or monthly variations may take the form of oscillations, now on one side and now on the other of any historical ratio, or they may be cumulative on one side of that ratio, producing a divergence increasing from month to month, and year to year; but variations in some degree, in some direction, are to be expected under the unrestrained operation of causes influencing the demand for, or the supply of, each metal.

The conditions, natural and commercial, which determine the ratio of exchange of the two metals being such, we have seen that government may enter, and, by making the two indifferently legal tender for debts at a ratio fixed by law, may, for the time, counteract the operation of any and all forces tending to produce divergence. So long as any country establishing such a principle holds a considerable amount of that metal which, under the natural and commercial conditions of supply and demand prevailing at the time, tends to become the dearer of the two, it is impossible that the cheapened metal should there, or in any market, fall far below that ratio. By the force of the bimetallic law, the substitution of the cheapened for the dearer metal will at once begin; and so long as that continues, the divergence of the market ratio from the mint ratio can never be wide. Why should any one in London or New York pay much more than fifteen and a half ounces of silver for an ounce of gold, when gold can, at any time and in any amount, be obtained for silver at the rate of fifteen and a half in Paris?

This operation of the bimetallic system can not be denied; but there is ground for dispute as to the degree of the advantages to result, and as to the cost at which those advantages are to be obtained. The monometallist, or advocate of the so-called single standard, is disposed to disparage the benefits to be expected, and to magnify the expense of this system. He points to the fact that the two metals do not actually circulate in the same country, at the same time, in any considerable degree; that it is always the one metal or the other which is used as money, according as the market ratio diverges to the one side or the other of the mint ratio, while the coin made from the dearer metal acquires a premium, and is exported or hoarded. Hence it is said bimetallism really means the use of but one metal in a country at a time. It is not a double standard, but an alternate standard.

To this the bimetallist replies that the concurrent use of the two money metals, side by side, in the same markets, is a matter wholly of indifference. The merit of the bimetallic scheme does not depend on this at all.

The object of bimetallism is, by joining the two metals together in the coinage, at a fixed ratio, to diminish the extent of the fluctuations to which the value of each would be separately liable, by generating a compensatory action between the two, by which the cheapening metal shall receive a larger use, while the appreciating metal drops partially out of its former demand, thus making the two fall together, if there must be a fall, or rise together, in the opposite case: or, conceivably, making the tendency of one to fall precisely counteract the tendency of the other to rise.

Thus we may suppose four successive cases to illustrate the working of this principle.

The first is, where the demand for the use of either metal in trade remaining the same, a large increase in the supply of one metal, A, takes place, the supply of the other, B, remaining unchanged. In this case, without the bimetallic system, the value of A would tend to fall rapidly through a considerable space, while the value of B would stand fast. With the bimetallic system, the joint supply of the two metals would be applicable to meet the joint demand for the two. Now, as the joint supply has been increased without any change in the joint demand, there must be a fall in value; but the fall will be in the two indistinguishably, except for a slight degree of delay and friction in exchange. Both will fall, but the depth of the fall will be diminished as the surface over which it is to take place has been enlarged.

The second is where, the demands of trade for both metals remaining the same, a diminution occurs in the supply of A, while the supply of B remains unchanged. Here, by the operation of the same principle, a rise in the value of money will take place, since the joint supply has been reduced without any corresponding change in the joint demand. The rise will be a rise of the two metals indistinguishably, the height of the rise being diminished as the surface over which it is to take place has been enlarged.

The third case is where, demand remaining the same, the supply of both metals undergoes a change in the same direction, either of increase or of diminution, at the same time. In this event, the fall or rise will again be of the two indistinguishably, the point reached being a mean between the points which would have been reached by the two severally.

The fourth case is where, demand remaining the same, the supply of the two metals undergoes a change at the same time, but in opposite directions, A through diminution, B through increase. In this case, the opposite tendencies will counteract each other. If of equal force, the value of money will be stable; if of unequal force, there will be movement in the direction of the stronger to the extent of the difference between the two. Instead of one falling and the other rising in value, the change will be wrought in the two indistinguishably.

It will appear from the foregoing statements that, under the bimetallic system, the value of money will be liable to vary more frequently than under the monometallic system. That is, a change in respect to either constituent of the money mass will produce a change of value; and it is apparent that the chances of change are greater with two constituents than with one. On the other hand, the variations under the bimetallic system are likely to be less extensive. Indeed, it is a matter of practical certainty that they will be far less extensive than they would be under the monometallic system, whichever metal were adopted as the standard of deferred payments.

But, again, the monometallist interposes the objection that the bimetallic system is only to be supported at great expense to the States maintaining it; that they lose by the exchange of the dearer for the cheapened metal, even though they acquire a certain premium in doing so, and that sooner or later the stock of the dearer metal in the bimetallic countries will become exhausted, and the system will collapse, the price of the two metals no longer being held closely or nearly at the former ratio by the possibility of exchanging them at that ratio, freely, in any amount.

How far a bimetallic country loses by the alternation of the metals in circulation, as now one and now the other becomes the cheaper at the coinage ratio, is a nice question.

That the service rendered to the commerce of the world by establishing a normal price for each metal in terms of the other, and thus creating and maintaining a par-of-exchange between gold countries and silver countries, is worth far more than its cost, seems to me beyond a rational doubt. It would, in my view, be as reasonable to doubt whether London Bridge repays the expense of its erection and repair. Were the cost of this bimetallic service, whatever it is, properly assessed upon and collected from each commercial nation of the world by turns, according to the proportion in which it derives advantage therefrom, I think it might safely be said that no one of these nations would sustain a single other charge which so fully justified itself in the return it made, whether that other charge were for works of construction, for the administration of justice, or for any other strictly necessary purpose.

But there is no assurance that the cost of the bimetallic system will be thus equitably assessed. If the whole charge of erecting and repairing London Bridge were thrown upon the merchants of the two or three streets nearest thereto, while yet the whole population were allowed to use the bridge, free of toll, there would not unnaturally arise a strong sense of injustice on the part of those who bore this burden for the public benefit; it might even become a question whether the undoubted advantages derived by them from the use of the bridge repaid the disproportionate expense which it caused them. If the maintenance of the bimetallic system involves a certain burden on the nations which sustain it, as I am disposed to think is the case, it fairly becomes a question whether those individual nations are compensated for bearing the whole expense of the service by their share of the advantages resulting therefrom to the trade and industry of the world.

That England could well have afforded, throughout the present century, to maintain this system for her own benefit, whatever it cost, even though other nations profited by it in greater or less degree, is clear as the light. That France, a country of far less extended international trade, has been compensated for bearing so large a part as she has done of the burden of maintaining a par-of-exchange for the commerce of the world, by her share of the resulting advantages, I make no question; but it must be admitted to be fairly a matter of dispute.

On such a point it is evidence of no small value that the French people themselves and the French statesmen, though singularly acute and sagacious in matters of finance, have apparently not doubted that the bimetallic system was for the interest of their country. Certain of the French political economists—MM. Chevalier, Levasseur, Bonnet, Mannequin, Leroy Beaulieu—from their theory of the subject have held that France lost by her policy in this respect; but the financiers of that remarkable nation held firmly to the "double standard" from 1785 to 1874. And though France at the latter date restricted her silver coinage, and two years later stopped it altogether, it was not done as the result of any change of views. Partly it was from deference to her monetary allies, Belgium and Switzerland, but chiefly because the demonetization of silver by Germany and the sale of the discarded metal of that empire brought a sudden strain upon the bimetallic system which threatened to break it violently down. Hence France closed her mints to silver, but not with any confession that her policy had been erroneous under the conditions previously existing; not from any desire to abandon that policy should the future offer conditions which would admit the resumption of bimetallism. It was the declaration of M. Léon Say, the French Minister of Finance, the President of the International Monetary Conference of 1878, that France, in suspending the coinage of silver, had taken no step towards the single gold standard, but had placed herself in a position to await events, a position which she would not leave till good reasons for action should appear, and then most probably to re-enter on the system of the double standard. …

The objection that the stock of the dearer metal in the bimetallic States must, if the drain be indefinitely continued, become after a while exhausted, and that the system will then lose all its efficiency in holding the two metals together, is unquestionably valid; but an altogether unreasonable weight has been assigned to it in the discussion of bimetallism as a scheme of practical statesmanship.

If we look at almost any treatise written from the monometallic point of view, we shall find that it is taken as conclusive against that scheme, that conditions of supply and demand can be assumed for the two metals separately which would result in the complete exhaustion of the dearer metal, and the consequent loss of all virtue in the bimetallic scheme. The bimetallist is confronted with a series of adverse conditions, taken each at its maximum and piled one above the other without the least regard to the modesty of nature, or the experience of the past; and is then challenged to say whether the system he proposes could be maintained under such circumstances. If he is candid enough to admit that bimetallism would fail there, it is taken for granted that the whole question is disposed of.

Now, human institutions are not to be judged of, and approved or disapproved, by such methods. The folly of reasoning like this would be seen at once were it applied to ordinary political matters. No government on earth could stand against one-fourth or one-tenth of the elements of hostility which might conceivably be arrayed against it. Mankind do not, therefore, refuse to form governments.

Bimetallism is a political institution for practical ends, and is entitled to be judged with reference to reasonable probabilities. It may claim the benefit of the chance that adverse conditions will be offset by conditions favourable, and that the adverse conditions will not prove so severe at the start as they may be conceived, and that their force will be more quickly spent than might be feared.

It would be perfectly legitimate ground on which to establish European bimetallism, that the French system, with so little of support from other States, passed within a quarter of a century through the three successive shocks of the gold discoveries of Siberia, the gold discoveries of California, and the gold discoveries of Australia, and yet was not brought to the ground.

With Germany, France, and England joined in a monetary union, no changes reasonably to be anticipated in the conditions of supply of the one metal or the other would succeed in moving the market ratio far apart from the mint ratio thus supported by maintaining over so wide a surface a legal equivalence between the two metals in payment of debts.

And, moreover, while bimetallism is entitled to be judged like any other political institution, with reference to the reasonable probabilities of the future, the allowance which requires to be made for error and extraneous force is less than in most political institutions, inasmuch as the failure of bimetallism involves no disaster to industry or society.

When an engineer designs a bridge which is intended to sustain a weight of eighty tons, he introduces a "factor of safety," say three or five, and makes the bridge strong enough to bear two hundred and forty or four hundred tons. The greater the calamity which would result from the breaking down of the bridge—the deeper the chasm which it spans, the swifter the torrent below—the larger the factor of safety. With many political institutions, likewise, the consequences of failure would be so disastrous that the statesman seeks to introduce a high factor of safety; but in the case of bimetallism no catastrophe whatever is to be anticipated, even in the event of failure. At the worst, after the drain of the dearer metal, in consequence of changes in the conditions of supply, is completed, the bimetallic country is simply in the same position with the countries of the single standard using the cheapened metal. While the process of substitution is going on, it sells the dearer metal at a premium; when the process is over, it is no worse off than it would have been had it originally selected as its sole money of full legal-tender power the metal which it has bought at a discount, and which other countries, perhaps its immediate neighbours, are still using. It is not the case of a country seeking to reject the cheapening metal, and to supply its place with the metal which is continually becoming scarcer and dearer. … There is all the difference, in the two cases, between going down hill and going up hill.

Not only is no catastrophe involved in the failure of bimetallism through the exhaustion of the dearer metal, but it is always in the power of the Government to arrest the drain at any point without shock.

Thus, in 1874, France and her monetary allies, seeing the prospect of a considerable drain of gold through the importation of the discarded and cheapened silver of Germany, and having decided, whether wisely or unwisely, not to prevent that drain, restricted the coinage of silver without repealing or suspending the law which made gold and silver legal tender indifferently at a fixed ratio. Two years later, finding that the forces operating to lower the value of silver were powerful and persistent, the coinage of silver was peremptorily stopped.

Can one point to any sign that France has suffered any special injury to her trade and production from this act? …

We now have to note … that every additional State which joins the bimetallic group, having the same mint ratio between gold and silver, does not only share the cost or the burden with those already in the system, but diminishes the aggregate cost or burden to be borne, and this, not in a slight, but in an important degree, so that should the monetary league become general, the total cost or burden to be divided among the many allies would be inappreciable; while, should the system come to embrace all commercial States, there would, in theory, be no burden at all to be borne by any one.

Thus let us suppose the commercial world to be divided into sixteen States, A to P, inclusive, the first six having the single gold standard, four, G to J, the so-called double standard of gold and silver, say at 15–½:1; the remaining six States having the single standard of silver, thus:

A, B, C, D, E, F (G, H, I, J), K, L, M, N, O, P.

It is evident that in the case of a change in the conditions of supply tending to cheapen silver relatively to gold, the new silver would pass into the countries of the double standard, G to J, be there exchanged for gold at the rate of 15–½: 1, with some small premium as the profit of the transaction, and the gold would go to the gold countries, A to F, in settlement of trade balances.

The rapidity with which this substitution of silver for gold will go forward will depend, first, on the force of the natural causes operating to cheapen silver, and, secondly, on the force of the commercial causes operating to maintain or advance the value of gold. The length of time during which the drain of the dearer metal can be sustained without exhaustion will (given the rate of movement) depend solely on the stock of that metal existing in the bimetallic States jointly when the drain begins.

But chief among the commercial causes operating to maintain or advance the value of gold is the exclusive power with which gold is invested by law to pay debts within States A to F; while the stock of the dearer metal available to sustain the drain described is made up, not of all the gold in the sixteen States A to P, or in the ten States A to J, but only of the gold in the four bimetallic States, G to J.

Hence we see that for every gold State which adopts the "double standard" the amount of gold available, in the case of a cheapening of silver, to meet the drain of the dearer metal (on which the virtue of the bimetallic system depends) is increased; while the demand for gold in preference to silver at 15–½:1 (the only cause which threatens the stability of the bimetallic system) is, in just so far, diminished. On the other hand, every silver State that adopts the "double standard" strengthens the bimetallic system in the case of a cheapening of gold.

Let us suppose the sixteen commercial States to be divided as four gold States, eight gold and silver States, and four silver States, as follows:

A, B, C, D (E, F, G, H, I, J, K, L), M, N, O, F.

We see that the bimetallic system is now not twice as strong merely as in the case first assumed, but many times as strong, since not only is the amount of the dearer metal (whichever that may at the time be) subject to drain greatly increased, but the demand for that metal, in preference to silver at 15–½:1, now comes from four countries only, instead of six, as formerly. The transfer of still another State from each of the two single-standard groups would vastly increase the stability of the bimetallic system, A, B, C (D, E, F, G, H, I, J, K, L, M), N, O, P. Not only would the base of the system be broadened by bringing the dearer metal of ten States, D to M, under tribute in the event of changes operating on the supply of either to affect its value; but the force of the causes threatening the equilibrium of the system would be reduced, since the demand for the dearer metal would now come from only three States: A, B, C, in the case of a cheapening of silver relatively to gold; N, O, P, in the case of a cheapening of gold relatively to silver.

Bring still another State from each group into the monetary union, and the danger of a breaking down of the system, under any change in the conditions of supply which it would be reasonable to anticipate, almost disappears.

A, B (C, D, E, F, G, H, I, J, K, L, M, N), O, P. Twelve States now supply the dearer metal; only two States will take it in preference to the other at the ratio of the mint. Those two States—whether A, B, or O, P—can not take the dearer metal indefinitely. They will soon be surfeited. A further increase of money in them would only be followed by a fall in its value, which would soon proceed so far as to bring the metals together again. What the one metal would tend to lose in value through increase of supply, the other would tend to lose through diminution of demand.

This is the Modern Bimetallic Scheme advocated by Wolowski and Cernuschi in France, Malou and de Laveleye in Belgium, Mees and Vrolik in Holland, Schneider in Germany, Haupt in Austria, Seyd and the Liverpool writers in England, Horton, Nourse, and George Walker in the United States.

It differs widely from the plan of the so-called "double standard," which was pronounced impracticable by Locke, Adam Smith, and Ricardo. Not the smallest presumption against the reasonableness of this scheme is created by the fact that eminent economists of the past century, and of the first half of the present, declared in favour of the single standard, whether of gold or of silver. Those writers contemplated a condition of international relations in which anything like general and permanent concert of action, in establishing and maintaining a ratio between the metals in the coinage, would have been wholly beyond reasonable expectation. …

A general or universal system of bimetallism would involve no machinery, no international accounts, no detail whatever. The simple agreement of governments to coin at a certain ratio would be sufficient for all the objects that have been discussed. If unification of coinage, identity of moneypieces, and mutual acceptance of coins by the several nations forming such a monetary league, were to be added, some machinery for the redemption of worn pieces might require to be brought into existence; but this is not a necessary feature of successful bimetallism, which would be entirely compatible with the retention by each State of its own devices and denominations, and with the exchange of moneys as at present effected. …

Readings in Money and Banking

Подняться наверх