XVII. INDIRECT EXCHANGE
15. Interlocal Exchange Rates
Let us first assume that there is only one kind of money. Then with regard to money's purchasing power at various places the same is valid as with regard to commodity prices. The final price of cotton in Liverpool cannot exceed the final price in Houston, Texas, by more than the cost of transportation. As soon as the price in Liverpool rises to a higher point, merchants will ship cotton to Liverpool [p. 453] and thus will bring about a tendency toward a return to the final price. In the absence of institutional obstacles, the price of an order for the payment of a definite amount of guilders in Amsterdam cannot rise in New York above the amount determined by the costs involved by reminting the coins, shipment, insurance, and the interest during the period required for all these manipulations. As soon as the difference rises above this point--the gold export point--it becomes profitable to ship gold from New York to Amsterdam. Such shipments force the guilder exchange rate in New York down below the gold export point. A difference between the configuration of interlocal exchange rates for commodities and those for money is brought about by the fact that as a rule commodities move only in one direction, namely, from the places of surplus production to those of surplus consumption. Cotton is shipped from Houston to Liverpool and not from Liverpool to Houston. Its price is lower in Houston than in Liverpool by the amount of shipping costs. But money is shipped now this way, now that.
The error of those who try to interpret the fluctuations of the interlocal exchange rates and the interlocal shipments of money as determined by the configuration of the nonmonetary items of the balance of payments is that they assign to money an exceptional position. They do not see that with regard to interlocal exchange rates there is no difference between money and commodities. If cotton trade between Houston and Liverpool is possible at all, the cotton prices at these two places cannot differ by more than the total amount of costs required for shipment. In the same way in which there is a flow of cotton from the southern parts of the United States to Europe, gold flows from the gold-producing countries like South Africa to Europe.
Let us disregard triangular trade and the case of the gold-producing countries and let us assume that the individuals and firms trading with one another on the basis of the gold standard do not have the intention of changing the size of their cash holdings. From their purchases and sales, claims are generated which necessitate interlocal payments. But according to our assumption these interlocal payments are equal in amount. The amount that the residents of A have to pay to the residents of B is equal to the amount that the residents of B have to pay to the residents of A. It is therefore possible to save the costs of shipping gold from A to B and from B to A. Claims and debts can be settled by a sort of interlocal clearing. It is merely a technical problem whether this evening up is effected by an interlocal clearinghouse organization or by the turnovers of a special market for foreign [p. 454] exchange. At any rate, the price which a resident of A (or of B) has to pay for a payment due in B (or in A) is kept within the margins determined by the shipment costs. It cannot rise above the par value by more than the shipment costs (gold export point) and cannot fall below the shipment costs (gold import point).
It may happen that--all our other assumptions remaining unaltered--there is a temporal discrepancy between the payments due from A to B and those from B to A. Then an interlocal shipment of gold can only be avoided by the interposition of a credit transaction. If the importer who today has to pay from A to B can buy at the market of foreign exchange claims against residents of B as fall due in ninety days, he can save the costs of shipping gold by borrowing the sum concerned in B for a period of ninety days. The dealers in foreign exchange will resort to this makeshift if the costs of borrowing in B do not exceed the costs of borrowing in A by more than double the costs of shipping gold. If the cost of shipping gold is 1/8 per cent, they will be ready to pay for a three months' loan in B up to 1 per cent (pro anno) more as interest than corresponds to the state of the money-market interest rate at which, in the absence of such requirements for interlocal payments, credit transactions between A and B would be effected.
It is permissible to express these facts by contending that the daily state of the balance of payments between A and B determines the daily point at which, within the margins drawn by the gold export point and the gold import point, the foreign exchange rates are fixed. But one must not forget to add that this happens only if the residents of A and of B do not intend to change the size of their cash holdings. Only because this is the case does it become possible to avoid the transfer of gold altogether and to keep foreign exchange rates within the limits drawn by the two gold points. If the residents of A want to reduce their cash holdings and those of B want to increase theirs, gold must be shipped from A to B and the rate for cable transfer B reaches in A the gold export point. Then gold is sent from A to B in the same way in which cotton is regularly sent from the United States to Europe. The rate of cable transfer B reaches the gold export point because the residents of A are selling gold to those of B, not because their balance of payments is unfavorable.
All this is valid with regard to any payments to be transacted between various places. It makes no difference whether the cities concerned belong to the same sovereign nation or to different sovereign nations. However, government interference has considerably changed the conditions. All governments have created institutions which make [p. 455] it possible for the residents of their countries to make interlocal domestic payments at par. The costs involved in shipment of currency from one place to another are borne either by the treasury or by the country's central bank system or by another government bank such as the postal savings banks of various European countries. Thus there is no longer any market for domestic interlocal exchange. The public is not charged more for an interlocal order to pay than for a local one or, if the charge is slightly different, it no longer has any reference to the fluctuations of the interlocal movements of currency within the country. It is this government interference which has sharpened the difference between domestic payment and payment abroad. Domestic payments are transacted at par, while with regard to foreign payments fluctuations occur within the limits drawn by the gold points.
If more than one kind of money is used as a medium of exchange, the mutual exchange ratio between them is determined by their purchasing power. The final prices of the various commodities, as expressed in each of the two or several kinds of money, are in proportion to each other. The final exchange ratio between the various kinds of money reflects their purchasing power with regard to the commodities. If any discrepancy appears, opportunity for profitable transactions presents itself and the endeavors of businessmen eager to take advantage of this opportunity tend to make it disappear again. The purchasing-power parity theory of foreign exchange is merely the application of the general theorems concerning the determination of prices to the special case of the coexistence of various kinds of money.
It does not matter whether the various kinds of money coexist in the same territory or whether their use is limited to distinct areas. In any case the mutual exchange ratio between them tends to a final state at which it no longer makes any difference whether one buys and sells against this or that kind of money. As far as costs of interlocal transfer come into play, these costs must be added or deducted.
The changes in purchasing power do not occur at the same time with regard to all commodities and services. Let us consider again the practically very important instance of an inflation in one country only. The increase in the quantity of domestic credit money or fiat money affects at first only the prices of some commodities and services. The prices of the other commodities remain for some time still at their previous stand. The exchange ratio between the domestic currency and the foreign currencies is determined on the bourse, a market organized and managed according to the pattern and the commercial customs of the stock exchange. The dealers on this special [p. 456] market are quicker than the rest of the people in anticipating future changes. Consequently the price structure of the market for foreign exchange reflects the new money relation sooner than the prices of many commodities and services. As soon as the domestic inflation begins to affect the prices of some commodities, at any rate long before it has exhausted all its effects upon the greater part of the prices of commodities and services, the price of foreign exchange tends to rise to the point corresponding to the final state of domestic prices and wage rates.
This fact has been entirely misinterpreted. People failed to realize that the rise in foreign exchange rates merely anticipates the movement of domestic commodity prices. They explained the boom in foreign exchange as an outcome of an unfavorable balance of payments. The demand for foreign exchange, they maintained, has been increased by a deterioration of the balance of trade or of other items of the balance of payments, or simply by sinister machinations on the part of unpatriotic speculators. The higher prices to be paid for foreign exchange cause the domestic prices of imported goods to rise. The prices of the domestic products must follow suit because otherwise their low state would encourage business to withhold them from domestic consumption and to sell them abroad at a premium.
The fallacies involved in this popular doctrine can easily be shown. If the nominal income of the domestic public had not been increased by the inflation, they would be forced to restrict their consumption either of imported or of domestic products. In the first case imports would drop and in the second case exports would increase. Thus the balance of trade would again be brought back to what the Mercantilists call a favorable state.
Pressed hard, the Mercantilists cannot help admitting the cogency of this reasoning. But, they say, it applies only to normal trade conditions. It does not take into account the state of affairs in countries which are under the necessity of importing vital commodities such as food and essential raw materials. The importation of such goods cannot be curtailed below a certain minimum. They are imported no matter what prices must be paid for them. If the foreign exchange required for importing them cannot be procured by an adequate amount of exports, the balance of trade becomes unfavorable and the foreign exchange rates must rise more and more.
This is no less illusory than all other Mercantilist ideas. However urgent and vital an individual's or a group of individuals' demand for some goods may be, they can satisfy it on the market only by paying the market price. If an Austrian wants to buy Canadian [p. 457] wheat, he must pay the market price in Canadian dollars. He must procure these Canadian dollars by exporting goods either directly to Canada or to some other country. He does not increase the amount of Canadian dollars available by paying higher prices (in schillings, the Austrian domestic currency) for Canadian dollars. Moreover, he cannot afford to pay such higher prices (in schillings) for imported wheat if his income (in schillings) remains unchanged. Only if the Austrian Government embarks upon an inflationary policy and thus increases the number of schillings in the pockets of its citizens, are the Austrians in a position to continue to buy the quantities of Canadian wheat they used to buy without curtailing other expenditures. If there were no domestic inflation, any rise in the price of imported goods would result either in a drop in their consumption or in a restriction in the consumption of other goods. Thus the process of readjustment as described above would have come into motion.
If a man lacks the money to buy bread from his neighbor, the village baker, the cause is not to be seen in an alleged scarcity of money. The cause is that this man did not succeed in earning the amount of money needed either by selling goods or by rendering services for which people are prepared to pay. The same is true with regard to international trade. a country may be distressed on account of the fact that it is at a loss to sell abroad as many commodities as it would have to sell in order to buy all the food its citizens want. But this does not mean that foreign exchange is scarce. It means that the residents are poor. And domestic inflation is certainly not an appropriate means to remove this poverty.
Neither has speculation any reference to the determination of foreign exchange rates. The speculators merely anticipate the expected alterations. If they err, if their opinion that an inflation is in progress is wrong, the structure of prices and foreign exchange rates will not correspond to their anticipations and they will have to pay for their mistakes by losses.
The doctrine according to which foreign exchange rates are determined by the balance of payments is based upon an illicit generalization of a special case. If two places, A and B, use the same kind of money and if the residents do not want to make any changes in the size of their cash holdings, over a given period of time the amount of money paid from the residents of A to those of B equals the amount paid from the residents of B to those of A and all payments can be settled without shipping money from A to B or from B to A. Then the rate of cable transfer B in A cannot rise above a point slightly below the gold export point and cannot drop below a point slightly [p. 458] above the gold import point, and vice versa. Within this margin the daily state of the balance of payments determines the daily state of the foreign exchange rate. This is the case only because neither the residents of A nor those of B want to alter the amount of their cash holdings. If the residents of A want to decrease their cash holdings and those of B to increase theirs, money is shipped from A to B and the cable rate B reaches in A the gold export point. But money is not shipped because A's balance of payments has become unfavorable. What is called by the Mercantilists an unfavorable balance of payments is the effect of a deliberate restriction of cash holdings on the part of the citizens of A and a deliberate increase in cash holdings on the part of the citizens of B. If no resident of A were ready to reduce his cash holding, such an outflow of money from A could never materialize.
The difference between the trade in money and that in the vendible commodities is this: As a rule commodities move on a one-way road, viz., from the places of surplus production to those of surplus consumption. Consequently the price of a certain commodity in the places of surplus production is as a rule lower by the amount of shipping costs than in the places of surplus consumption. Things are different with money if we do not take into account the conditions of the gold-mining countries and of those countries whose residents deliberately aim at altering the size of their cash holdings. Money moves now this way, now that. At one time a country exports money, at another time it imports money. Every exporting country very soon becomes an importing country precisely on account of its previous exports. For this reason alone it is possible to save the costs of shipping money by the interplay of the market for foreign exchange.