An Outline of International Price Theories

CHAPTER VI

DEVELOPMENTS SINCE 1918: THEORIES OF THE EXCHANGES UNDER DEPRECIATED CURRENCIES

1. Preliminary Remarks

The development of the theory of international price relationships since the Great War has been associated mainly with two important controversies: that in regard to the exchanges of depreciated currencies and that in regard to the transfer problem. The controversy concerning the rates of exchange of depreciated currencies has thrown light on the question of the adjustment of a violent disturbance due to a monetary cause, while the controversy concerning the transfer problem has clarified thought on the question of the correction of a disturbance due to an unfavourable balance of payments. We shall describe the former controversy in the present chapter and devote the following chapter to the latter controversy.

With the commencement of the Great War in 1914, the inflation of currencies became general in Europe, and the effects that it would produce on the foreign exchanges engaged the minds of the economists of the time. During the War the exchanges of the principal European currencies were artificially maintained at parity, but in the Spring of 1919, the “pegging” of the exchange values of the pound, the franc, and the lira to the American dollar was abandoned, and to the depreciation of the currencies of Central Europe was added the depreciation of the British, French, and Italian currencies. The exchanges of the depreciated currencies became one of the gravest problems of the time.

The various theories which were advanced to explain the exchanges of depreciated currencies have sometimes been classified in three groups: the balance of payments theory, according to which the state of foreign trade, international capital movements or other items of foreign indebtedness are the sources of exchange dislocation; the purchasing power parity theory, according to which the initiating cause of exchange depreciation is inflation; and the doctrine that disequilibrium in the finances of the governments is the source of all monetary and exchange troubles. Strictly speaking, however, the third doctrine is not an independent explanation of the subject. Disequilibrium in governmental finances by itself does not cause exchange troubles. It is only when an unbalanced budget effects a real, or an anticipated, change in the volume of the circulation that the exchanges are affected.

We may begin with a brief review of the balance of payments theory. According to the theory, the balance of payments was the source of the exchange troubles and the causal sequence runs from the balance of payments through the exchanges to prices and the volume of money. The theory has been widely held by the spokesmen of governments, by representatives of the business world, and by Knapp and his school. Since the views of the former two groups contain no recognition of any relation between currency inflation and exchange depreciation, they are clearly erroneous. Similarly, Knapp and his school are wrong when they refuse to recognize the existence of any relationship in the prices of trading countries. A more correct view would be that the balance of payments is the immediate cause governing the exchanges, and that other factors act on the exchanges only through the balance of payments. Even in that view there is a serious difficulty, namely the difficulty of defining the term, the balance of Payments.

The international accounts of a country may be divided into two groups. The first group of items would include all imports and exports of goods and all imports and exports of services. The second group would consist of the payments of interest, profits, dividends and so on, on the one hand, and of international capital movements on the other hand. Under the latter, “one should distinguish between long term and short term investments. Long term capital exports consist in the purchase of shares in foreign undertakings, the repurchase of home securities or repayment of loans contracted abroad, the purchase of foreign holdings in property located at home, etc., etc. Short term capital exports include any increase in the volume of bank balances held abroad, or in the holdings of foreign bills, and any decrease in the volume of commercial indebtedness to foreign countries”.1

Now the question arises as to whether in drawing up the balance of payments we are to include the item of short-term capital movements. As Professor Aftalion very ably points out, if we exclude that item we neglect a very important factor; if we include that item, then accounts are always balanced and that gives us no theory of foreign exchanges.2

The only way out of the difficulty is to discard the idea that the balance of payments is a fixed quantity and to interpret the term, the balance of payments, in the sense of demand and supply of foreign currencies in the foreign exchange markets.3 We shall consider both the quantity demanded of foreign currencies (or the total debt the country owes to foreign countries) and the quantity supplied of foreign currencies (or the total debt the foreign countries owe to the country) as functions of the rates of the exchanges. Those functional relations give us the demand and supply curves of the exchanges, and they will give us the equilibrium rate of the exchanges.4 The balance of payments theory would thus be that in equilibrium the rates of the exchanges equalize the quantity of the currencies demanded and the quantity of the currencies supplied in the foreign exchange markets.

The balance of payments theory in that form, though correct, is not a complete theory of foreign exchanges. We must go behind the demand and supply and see how they are determined.5 The purchasing power parity theory, which we are going to examine in the following pages, will show how monetary factors may influence the demand and supply of the exchanges. The general theory of international trade will help us to see some of the factors determining the imports and the exports of goods and services. The theory of transfer, which forms the subject of the following chapter, deals with the question how international capital movements or reparation payments may lead to changes in the foreign exchanges.

2. The Purchasing Power Parity Theory and Cassel’s Version of it

Before we examine the purchasing power parity theory it is necessary briefly to state the meaning of the term, purchasing power parity theory. As pointed out by Professor Pigou,6 the theory of the foreign exchanges consists of two parts, namely first, the positive part to which the problem of exchange equilibrium belongs; and second, the comparative part which deals with the causes of disturbance of the norm of the exchanges and its restoration. Likewise, the purchasing power parity theory had two forms. In its positive form the purchasing power parity theory is that the rate of exchange would be in equilibrium when the “purchasing power of the moneys” is equal in all trading countries. If the term purchasing power refers to the power of purchasing commodities, which are not only similar in technological composition, but also in the same geographical situation, the theory becomes the classical doctrine of comparative values of moneys in different countries and is a sound doctrine. But unfortunately the term purchasing power in connection with the theory sometimes implies the reciprocal of the general price level in a country. While so interpreted the theory becomes that the equilibrium point for the foreign exchanges is to be found at the quotient between the price levels of the different countries. That is, as we shall see, an erroneous version of the purchasing power parity theory. It seems better to retain the term, “the positive form of the purchasing power parity theory” for the former or correct interpretation. In its comparative form, the purchasing power parity theory is that when we start with the condition of equilibrium in the foreign exchanges, an increase or decrease in the prices of one country will cause, other things being equal, a proportionate fall or rise in the rate of exchange. As a doctrine of comparative statics the theory is a correct one. Let us turn now to the evolution of the theory since the Great War.

Professor Cassel is generally recognized as the leading exponent of the purchasing power parity theory, and for that reason we shall begin with his particular version of the theory. Cassel starts from the quantity theory and makes it the basis of his theory of the foreign exchanges. He recognizes that the quantity of money not only determines the level of domestic prices but also, through prices, the rate of exchange:—

The rate of exchange is primary an expression for the value in the money of one country put upon the money of another country. If we consider two countries, A and B, with independent paper currencies, the money of A can have value in B only on the ground that it represents buying-power, or more generally paying-power, in A. The price in B of the money of A will, therefore, be broadly proportional to the buying power of the money of A and will consequently stay in inverse proportion to the general level of prices in A. Further, the price in B will, of course, tend to be proportional to the general level of prices in B. Thus the rate of exchange between the two countries will be determined by the quotient between the general levels of prices in the two countries. Now, according to the quantitative theory of money the general level of price varies, other things being equal, in direct proportion to the quantity of the circulating medium in a country. If this be true, the rate of exchange between the two countries must vary as the quotient between the quantities of their respective circulating media.7

That is the purchasing power parity theory as stated by Professor Cassel during the War. It is what we describe as the erroneous version of the positive form of the theory.

Three points may be noted in connection with the theory as expounded in the statement of Professor Cassel which we have quoted above. First by “buying power of money” in a country is meant the general purchasing power of money against all goods, but not the purchasing power merely in terms of internationally traded goods. The point is clearly stated in one of his later writings. In opposing the second interpretation he says:—

If we assume that prices of all the export commodities of country B are doubled, whilst all other prices in B remain unchanged, it would not be possible for the rate of exchange to be reduced by a half, as a much smaller fall in the rate would bring out the latent export possibilities of a mass of other commodities in B, and would prevent a further fall in the exchange.8

Secondly, the purchasing power parity is simply a commodity price parity, because Cassel does not, as pointed out by Mr. Keilham, “take into account that there is not only an international market for wholesale articles, but for capital also.”9 Finally, the effect of Professor Cassel’s doctrine thus interpreted is, as Mr. Keynes says, “that, even with the hindrances to free movements of goods which prevail in war time, real price levels in different countries tend to equality. That is to say, the index numbers of local prices corrected by the world-value of the local money, as measured by the exchanges, tend to equality.”10

In that absolute form the purchasing power parity doctrine is certainly wrong. It neglects transport costs and other hindrances to trade, which were of special importance during the period in which Cassel wrote. It fails to take into consideration the element of truth which is contained in the classical doctrine of international price relationships.

In his later writings, i.e. writings of the post-war period, Professor Cassel makes many important revisions of the original, dogmatic version of his doctrine. Thus in 1918 he recognized that the actual rate of exchange may permanently deviate from the purchasing power parity, because of obstacles to trade. “If the trade between the two countries is hampered more severely in one direction than in the other, the rate of exchange will be deviated from its purchasing-power parity. If the imports of a country are more severely restricted than its exports, the consequence will be that foreign money will sink in value, as claims on such money will be comparatively easy to procure, but difficult to make use of.”11 In 1919 he accounted for the abnormal deviations of the German mark from the parity by reference to the magnitude of her exports of capital and the sale of mark notes at substantial concessions of price, in addition to her limitation on the exportation of goods.12 Finally, in 1922, he made a further modification, that is he admitted that changes in the conditions of foreign trade may occasion changes in the purchasing power parity.13 Moreover, he pointed out:—

In judging the value of a country’s money, a foreign country will naturally not only go by trade prices, but also by the height of wages. For the foreign country can buy labour for the money in question, as, for example, by having raw materials worked up in that country. But even if the country utilizes home-acquired raw material to make its export products nevertheless, the latter’s value represents, to a considerable extent, labour expended on them. The level of wages in the country, therefore, is always a very important factor—in the long run may be the predominating one—in determining the international value of the country’s currency.14

So much for his absolute version of the purchasing power parity theory. Let us turn now to his comparative version and see how he has used the doctrine to explain the variations of exchange rates. It was not until 1920 that he abandoned the concept of “purchasing power parity” as the quotient of price levels and began to consider it as equal to “the old rate multiplied by the quotient between the degrees of inflation of both countries”.15 In 1922 he even went so far as to remark that “there is no need whatever to pre-suppose that the parity rate of exchange corresponds to the quotient of absolute price levels”.16 In other words, starting from an equilibrium rate, the exchanges would tend to equal the original rate multiplied by the quotient between the degree of internal inflation in one trading country and that of another, if other things remained unchanged and if the prices of home-trade products moved together with the prices of foreign trade products.

Let us turn now to the causal sequence of events as assumed by the purchasing power parity theory. In general, we may say that it assumes the following sequence.17

Quantity of currency—rise of prices—the depreciation of the exchanges.

3. The Importance of Cost of Conveyance and the Role of International Demands: Pigou, Viner, and Heckscher

Although Professor Cassel must be remembered as one who has called attention to the danger of inflation, his version of the purchasing power parity theory is on the whole an incorrect one. We shall see in the present section how his version of the positive form of the theory should be modified or corrected and in the following section how his dynamic theory should be improved. It should be noted, however, that as a part of a doctrine of comparative statics his version of the comparative form of the theory is a correct one. But Professor Cassel never limits his version to the case of comparative statics.

The best way to show how Cassel’s positive version of the theory should be corrected is perhaps to present the important criticisms raised by Professors Pigou, Viner, and Heckscher.

Professor Pigou begins by distinguishing the positive form and the comparative form of Cassel’s doctrine of foreign exchange. He rightly points out that Cassel’s positive doctrine is false, for Cassel neglects the importance of the cost of transmitting goods from one country to another:—

If it were the fact that all commodities produced in either of the two countries flowed without cost between them, Professor Cassel’s positive doctrine would follow immediately. But a large number of commodities not only fail to flow without cost but do not flow at all. There is no necessity, in order that exchange equilibrium may be established, for the internal purchasing power of sterling in respect of these commodities to be the same as its external purchasing power. . . . There is no reason to expect that the prices of various sorts of non-traded and partially-traded goods will bear the same ratio to the prices of traded goods in different countries. Consequently, there is no ground for assuming that, even in the absence of one-sided obstacles to trade, the rate of exchange which confirms to purchasing power parity, as defined by Professor Cassel, will be identical with, or even in the close neighbourhood of, the equilibrium rate. The positive doctrine of purchasing power parities cannot, therefore, be maintained without reservations and qualifications so extensive as practically to destroy it.18

According to Professor Pigou, if there do not exist monopolistic discriminations or import quota systems, exchange equilibrium simply means “that nobody can gain by diverting a unit of any kind of product that might have been sold in the English market, to the American market, or vice versa”—England and America being two trading countries.19 Or, to put the same thing in other words, exchange equilibrium requires in general that a unit of international goods “in the country of export shall buy a claim in the country of import to a unit minus the cost in transport, taxes, loss of interest, and so forth involved in sending a unit there “For any commodity that does not flow between the two countries exchange equilibrium requires that a unit in one country shall exchange for a claim on a number of units in the other, not less than one unit minus the cost of transportation (including taxes) onwards, and not more than one unit plus the cost of transportation inwards.”20 If no article flowed between the two countries, then any rate of exchange would be an equilibrium rate and there would be no connection between the prices in one country and those in the other. As soon, however, as one or more articles enters into international trade, the prices of domestic goods in one country are connected with the prices of domestic goods in the other country through the channel of international goods.21

While recognizing the importance of the cost of conveyance in the determination of the norm of the exchanges,22 Professor Pigou23 does not, in this connection, analyse the role of reciprocal demands. The importance of the factor of reciprocal demands has been fully dealt with by Professor Viner and Professor Heckscher. In January, 1928, Professor Viner had a debate with Professor Cassel at Chicago, in which Professor Viner showed that changes in the conditions of demands for imports and exports must lead to changes in the barter terms of trade and consequently to changes in the rate of exchange (even though the volumes of currencies remain unchanged in the trading countries). In reply Professor Cassel admitted that the rate of exchange was determined partly by the conditions of reciprocal demands.24 Similar criticism is raised by Professor Heckscher in his study of Swedish economic history during the Great War.

According to Professor Heckscher, Cassel’s version of the positive form of the purchasing power parity theory “is correct only upon the never existing assumption that all goods and services can be transferred from one country to another without cost”.25 Upon that assumption, Heckscher says, “not only the general level of prices, but also the prices for each particular commodity, will be the same in both countries, when all prices are computed on the basis of the exchanges,”26 and consequently the proposition that “the rate of exchange between the two countries will be determined by the quotient between the general levels of prices in the two countries” becomes correct.27 Even in that case, however, the part played by reciprocal demands is not unimportant. “An increased demand for the export goods of a country, as compared with its import goods, will influence the monetary system of the country in the direction of increasing the amount of its circulating medium. The reason is, (a) from the point of view of general economic life that the national income has increased; and (b) from the point of view of the machinery of the exchanges that exports would otherwise become incessantly greater than imports and an equilibrium consequently become impossible.”28 If the amount of its medium of circulation remains unchanged, it should be added, the rate of exchange must change so that at the new rate of exchange the international account of the country will again be balanced.29

In reality there always is a cost of carrying goods from one country to another. As soon as we take into consideration the cost of transferring goods two important consequences will follow. First, the identity in the price both of the particular goods and of goods in general in different countries will disappear. The identity in the price of the particular goods disappears, because “goods stand higher in the importing than in the exporting country, by just so much as is represented by the cost of carriage”.30 The general price levels, compared on the basis of the exchanges, will not be the same in two trading countries, because the cost of transmission is never absolutely symmetrical, i.e. never influences import and export prices in the same way. There is consequently “a higher level of prices in a country the import goods of which have to pay a higher cost of carriage than the export goods, and vice versa”.31 Otherwise the international exchange of goods will not be in equilibrium. The second consequence is this—

that not all sorts of goods and services can be exchanged between countries. . . . Consequently the prices of a great number of goods and services are independent of international markets; and these prices will stand in no direct connection with the rate of exchange. Even if the cost of carriage were the same for imports and exports, and the prices of the goods which were being interchanged were, on an average, at the same level in both countries (when computed on the basis of the rate of exchange), it would consequently by no means follow that general price levels would be the same (computed on the same basis).32

Although there is no direct connection between the price of wholly domestic goods in one country and that in other countries, yet indirectly he recognizes they are connected. “There is always a connection between all different prices within the same community, on the ground that the same means (or factors) of production are used for different products and must influence their prices, because they must themselves in the long run command the same price for all their different uses.”33 Consequently, through the price relations of the international goods between different countries, the prices of wholly domestic goods in different countries are also related.

Heckscher then shows how in this case changes in international demands may lead to changes in the rate of exchange and in international price relationships. He says:—

Such goods, which for their production use much of some factor whose cost has been increased through a change in international demand, will rise in price relatively to other goods; on the other hand goods produced by factors which have been cheapened, will move in the opposite direction. . . . If Swedish export goods, taken as a whole, have found an increased foreign demand, prices will rise for all such Swedish goods as are produced by factors of production which are used for export goods, as compared with the prices of other goods. . . . The final result of this, quite necessarily, will be a rise in price, as compared with foreign prices, for Swedish goods which are not exchanged, taken as a whole. For they are wholly produced by Swedish means of production. . . . Consequently, a country which is subject to an increased foreign demand for its goods and (at an unchanged rate of exchange) a rise in prices for its export goods will, almost certainly, also experience an average rise in prices in the case of those domestic goods which are not exchanged internationally; while the opposite development takes place in a country which finds less demand for its export goods.34

The value of Swedish, as compared with foreign, factors of production has permanently risen, which is the same thing as an increase in the national income of Sweden, created by a Swedish right to more foreign goods, in return for Swedish, than before.35

The Swedish authority may either maintain a fixed rate of exchange and increase the amount of its circulation, or maintain the amount of its circulation and let the rate of exchange vary. There is, of course, a third alternative, namely that Sweden lends to foreign countries so that the excess of debts due to her is wiped out.36

From the arguments described in the preceding paragraphs, Heckscher has succeeded in refuting Cassel’s proposition that the point of equilibrium for the foreign exchanges is to be found at the quotient between the price levels of different countries.37 It does not follow, however, that Cassel’s version of the comparative form of the purchasing power parity theory should be entirely rejected. As a part of a doctrine of comparative statics, that version is, we think, unassailable. But as a dynamic doctrine it needs much modification. In the following section, we shall show some of the factors which we must take into account when we enter the domain of the dynamics of the theory of foreign exchanges of depreciated currencies.

4. The Dynamic Factors

As a dynamic theory Cassel’s version of the purchasing power parity theory is defective, because he fails to give due consideration to the problem whether a change in the volume of money would lead to secondary changes in the fundamental data of the price system and to the place of inertia and anticipation in the determination of the rate of exchange. It is incorrect to say, as Cassel did, that an increase in the volume of paper money in a country will lead to a proportionate fall in both its internal and external values. The increase in money may lead to the following four changes in the data. First of all it may, to a certain extent, increase the employment of labour and the volume of production and consequently bring about a change in the supply conditions of the exports of the country. Its supply conditions being changed, there will be a change also in the barter terms of trade and the rates of foreign exchange. Secondly, an increase in the volume of the circulating medium usually means a redistribution of incomes. As the conditions of ownership are not the same, the demand conditions of the country cannot remain unchanged, and consequently the barter terms of trade and the rate of foreign exchange will be affected. Thirdly, when a country depreciates its currency the other countries may find it desirable to adopt a policy of protection for the purpose of preventing what has sometimes been vaguely called the evil effects of “exchange dumping”. The result will be a change in the cost of conveyance of goods from one country to another. Consequently the rate of exchange will be affected. Finally, during the process of inflation, the velocity of circulation of the currency is likely to change. The problem of the velocity of circulation is closely connected with the problem of confidence and loss of confidence and hence may be better discussed together with the latter problem, to which we shall now turn.

The problem of confidence and loss of confidence is essentially a psychological problem. Earlier writers on paper currencies had not failed to notice the importance of the psychological elements. Thus both Professor Mitchell and Professor Subercaseaux, in their studies of the paper currencies of the pre-War period, called attention to the importance of the psychological factor in the working of a paper standard.38 They found that if confidence in the stability of the exchanges was established, it would suffice to maintain the stability of the exchanges. If, on the contrary, people lost all their confidence in a paper money, the paper money must necessarily cease to work as an independent standard of value. Usually, however, the confidence of the people in a paper currency is between the two extremes.39 It becomes necessary, therefore, to investigate what are the elements which are likely to influence the confidence of the people. Both Mitchell and Subercaseaux found that the political and financial situations of the government were the major factors entering into the psychological estimates of the people. If the government was participating in a war, then the course of military events, or news concerning it, would have much effect on confidence in the paper money.40

In the post-War period, the idea that the financial situation of the government of a paper-standard country may affect the rate of exchange through the degree of confidence of the people in the paper standard has appeared in the form of the doctrine that disequilibrium in the finances of the government is the source of all monetary and exchange troubles. Professor Young and Professor Rist are among the ablest who hold that view. According to Professor Young, the characteristic sequence of events in the post-War period of inflation was: unbalanced budget—disordered exchanges—inflation.41 The connecting link between budgetary disequilibrium and the dislocations in external and internal values of money was found to be “speculation” or “anticipation By “speculation” he meant “something much more far-reaching than the technical processes of organized speculation or the transactions of professional speculators. Just as the housewife who buys sugar in advance of her needs because she thinks that if she waits she will have to pay a higher price is a speculator in sugar, so every person who spends or accumulates money or who borrows or lends with a view to securing gains or avoiding losses attaching to changes in its purchasing power is a speculator in money. The most striking instances of speculation in money occur when a sudden loss of confidence in the stability of its value, arising from the disclosure of a disappointing budgetary situation or from other developments which are likely to affect the national finances adversely leads to a ‘flight’ from a country’s money in the form of the precipitate buying of foreign exchange”.42 Similarly, Professor Rist, in his study of inflation in the first half decade of the post-War period in England, the United States, France, and Czechoslovakia,43 finds that budgetary equilibrium is both a necessary and precedent condition for deflation and stabilization of currency. He is of the opinion that, given a balanced budget, speculation will turn favourable, foreign capital will flow in, the exchange will naturally improve, internal prices will fall, because of the increase in the supply of gold, which is a result of the improvement of the exchanges, and finally the quantity of money will contract as a natural consequence.44

Two comments may be made in connection with that doctrine of foreign exchange. First, while we should give budgetary disequilibrium its right place in the theory of the exchanges, we must recognize that budgetary equilibrium may exist together with a dislocation in the exchanges and that an unbalanced budget is not a necessary condition, nor is it always the initial cause, of exchange disturbance.45 Secondly, that doctrine is not, as it is often said to be, a “synthesis” of the purchasing power parity theory and the balance of payments theory. It is true that the acceptance of that doctrine makes the former two theories appear closer together, because it admits of the sequence of events running either from exchange to prices or from prices to exchange. But instead of an independent or a synthetical doctrine, it may best be considered as an addendum to the purchasing power parity theory. We are in agreement with de Bordes when he says that

this would suggest an amendment to the purchasing power parity theory in the sense that during periods of depreciation the rates of exchange tend to coincide, not with the actual, but with the expected purchasing power parity. In other words, the foreign exchanges market may be so completely dominated by speculation and movements of capital that the rates of exchange move independently from the purchasing power parity, i.e. from the price levels.46

In accordance with the degree of confidence or distrust, we may divide currency inflation into three stages, namely:—

(a) the stage of mild inflation or moderate depreciation;

(b) the stage of extreme inflation or severe depreciation; and

(c) the stage of collapse.

When the depreciation is a mild one, there is normally no distrust of the currency and the degree of confidence is high. We should expect the rate of exchange to lag behind the prices of commodities and the latter behind the volume of the circulating medium. The existence of those lags is due partly to inertia or, in other words, adjustments in the price system are a time-taking process. But there are more fundamental reasons why the internal and external values of the currency do not fall proportionately to the increase in the volume of money. They are the secondary changes in data that are the results of the very increase of money. One of the most important secondary changes is that of the velocity of circulation of money. Practically all writers on the subject agree that at the first stage of depreciation, the liquidity preference becomes greater and the velocity of circulation becomes smaller. One quotation may be sufficient to present that view:—

The public is so much accustomed of thinking of money as the ultimate standard that, when prices begin to rise, believing that the rise must be temporary, they tend to hoard their money and to postpone purchases, with the result that they hold in monetary form a larger aggregate real value than before.47

Inertia is also a reason why the exchange rate lags behind the movement of internal prices. At the stage of mild inflation, an increase in the volume of money cannot affect the rate of foreign exchange until it directly or indirectly affects the balance of payments. If the increased money is used to buy domestic goods, then the balance of payments will be affected only after the prices of goods produced at home and/or the incomes of the owners of factors of production have been increased. In that case, the exchanges will lag behind both the increase of money and the movement of internal prices.48

As soon, however, as confidence in the paper money is declining, we have the stage of severe inflation. Severe depreciation, as pointed out by de Bordes, differs from mild depreciation both quantitatively and qualitatively.49 The following are some of the conclusions reached by the various studies of the severe inflation in Austria and Germany in the period from the War to the years 1922 and 1923. First, when currency depreciation is great and rapid, people must sooner or later discover that the currency is declining in value, that is to say that to hold money would mean a loss to the holders. As soon as they discover that fact they become less willing to hold money and try to get rid of it quickly. In other words the velocity of the circulation of money is increasing. The rate of the increase in the velocity of circulation depends upon the speed and the scale of the depreciation.50 Since the velocity of circulation is increasing, the rise in prices must be at a quicker rate than the increase in the volume of money. Second, in the foreign exchange market, speculation or anticipation is in full play. People watch closely the situations of all factors (including the situation of the budget), which may affect the future rate of increase of the paper money or the demand and supply of foreign currencies. Whenever there is a tendency for the volume of money to increase or people think that there is such a tendency, the rate of exchange immediately falls. The foreign exchange market becomes so sensitive that the external value of the money falls at a quicker rate than its internal value. Moreover, the very fall in the rate of exchange may lead to a further fall in the rate. Thus Professor Graham says of the demand and supply of the bills of exchange during the severe inflation of Germany in the years 1920 to 1923:—

Reciprocal demand schedules for an inconvertible paper currency are by no means smoothly sloping curves. It is possible that a rising exchange value of a depreciated paper currency, instead of reducing the amount of that currency demanded by holders of foreign moneys, may increase it and, instead of increasing the amount of the cheapening foreign currencies demanded by holders of the said currency, may reduce it. . . . This phenomenon, which is common on all speculative markets, very much increases the range of paper currency exchange rate fluctuations.51

Therefore, during a great part of the stage of severe inflation, we should expect the exchanges to fall at a quicker rate than the increase of both the internal prices and the volume of money. Third, causes and effects here are often reciprocal. Various factors react upon one another. The rate of foreign exchange, for instance, is affected by the prices of commodities in different countries, while the latter is also affected by the rate of exchange. Similarly, the condition of the budget affects the rate of exchange through the anticipation of the future volume of paper money, while the rate of exchange may affect the budgetary condition if the government concerned has unconditional obligations in foreign currency and, consequently, affect the actual volume of money.52

A severe inflation cannot last very long without leading to a collapse of the currency. In the stage of collapse, the paper money no longer remains an independent standard of value. All business is directly transacted in a stable foreign currency. All prices are actually calculated in terms of that stable foreign currency and are simply expressed nominally in terms of the local paper money. Prices are therefore nothing but the product of prices in terms of the stable foreign currency multiplied by the rate of exchange between that currency and the local currency. It follows, therefore, that prices must rise in proportion to the fall in the rate of exchange.53 Experience also shows that in the stage of collapse, the lags either way between the volume of currency and the prices of commodities “became either completely or largely eliminated”.54

We may now turn to the question of causal relationship and sequence of events. The question of causal relationship has often been misunderstood, especially by those who attempt at an inductive study of the problem. Firstly, causal relationship must be distinguished from chronological sequence. When the purchasing power parity theorists say that an increase in the volume of the circulating medium of a country (cause) will lead both to an increase in its internal price level and to a fall in the rate of exchange (effect), or when they say that the causal sequence would run from money through prices to the exchanges, they do not mean that the chronological sequence will also follow the causal sequence. People may anticipate an increase in the volume of money and that anticipation is usually sufficient to cause a rise in prices and a fall in the exchanges. Secondly, causes and effects are, as already indicated, often confused and reciprocal. What the purchasing power parity theory says is simply that currency inflation is practically the most important initial cause of disturbance. It does not deny that the effects of the initial currency inflation may, in turn, become causes for further currency inflation. Thirdly, causal sequence should be further distinguished from actual sequence. In the real world the actual course is governed by various causes. The actual course of events may not be the same as that expected by the theory, because new disturbances may enter in during the process of adjustment. In other words, other things may not be equal.55 Professor Cassel is wrong when he expects the actual course of events to turn out in accordance with a doctrine of comparative statics. As a doctrine of dynamics, his doctrine must be corrected by the doctrines summarized in the present section. As an application to an actual case, his version of the purchasing power parity theory is again incorrect, because he fails to pay sufficient attention to other changes in the price system.

* * * * * *

The rate of exchange, as already indicated, may sometimes take the lead and leave the prices of commodities to follow. We may now see how a change in the rate of exchange, which is not the result of changes in prices of commodities, may lead to changes in the latter. Professor Mitchell is perhaps the first economist who subjects that question to an elaborate analysis. According to Professor Mitchell, when the rate of exchange falls the wholesale prices of imports are immediately raised56 and the wholesale prices of goods exported in large quantities are also directly affected.57 Those upward movements in the prices of international goods will cause similar movements in the prices of domestic goods:—

In a modern community the prices of different goods constitute a complexly organized system, in which the various parts are continually being adjusted to each other by intricate business processes. Any marked change in the prices of imported goods disturbs the equilibrium of this system, and business processes at once set going a series of readjustments in the prices of other goods to restore it. Thus a course which directly affects the price of one good only may indirectly affect the prices of many other goods.

[When the prices of imported and exported goods are raised by the fall of the exchanges] prices of commodities made from materials exported or imported were adjusted to compensate for the change in the expense of producing them. Prices of domestic commodities produced in competition with foreign goods had an opportunity to rise as the cost of the latter increased, and were forced down as the cost of the latter declined. Prices of goods capable of being substituted for wares imported or exported were kept more or less in harmony with the prices of the latter by shifting demand. And in general, men of all classes found that their business and living expenses were increased when they bought imports or exports, goods made from them, goods made in competition with them, or substitutes for them. These increasing expenses they sought to recoup by charging higher prices for whatever they had to sell.58

Ultimately, the influence of the depreciation in the exchanges will spread over the whole field of wholesale prices, “though with varying degrees of celerity and completeness.” The average of retail prices moves more sluggishly than the average of wholesale prices, while the average wage moves more slowly than the retail prices.59

Similar conclusions have been reached by Professor Subercaseaux. According to him, a fall in the exchanges immediately raises the wholesale prices of those imported goods which are produced only in foreign countries. If the imported goods are also produced at home, the depreciation of the exchanges acts as a sort of protection to the manufacturers of those goods at home, because normally domestic costs of production lag behind the movement of the exchanges. The average price of the latter kind of imported goods will be affected by the fall of the exchanges in a degree somewhat less than that of the price of imports which are only produced abroad.60 Wholesale prices of exports are also directly influenced by the fall of the exchanges. They will rise in proportion to the fall in the latter.61 There Professor Subercaseaux tacitly assumes that the country with the depreciated currency is relatively unimportant in the world market so that the quantities of its exports offered could not affect the world prices of those exports. If that condition is not fulfilled, then the rise in the prices of those exports will not be proportionate to the fall of the exchanges. Finally, like Mitchell, Subercaseaux points out that the wholesale price of domestic goods lags behind that of international goods,62 the retail prices behind the former63 and the average wage behind the retail prices.64

In the post-war discussions, a similar doctrine was advanced. Professor Graham, for instance, has expressed similar views. But he, unlike Subercaseaux, is of the opinion that one should expect the export prices to lag behind the import prices. Graham says:—

When the value of a currency is relatively depressed on the exchange markets . . . all domestic commodities, both those exported and those consumed at home, become low in price in comparison with the price of imports, but the prices of exportable goods become high relative to those of commodities which find their market solely within the country. . . . The price of imports is normally established in foreign exchange markets and increased pari passu with a rise in the cost of foreign exchange. The prices of commodities produced and consumed at home are directly affected by neither the prices prevailing in foreign markets nor by the exchange rate. The prices of exports are affected, on the demand side, by foreign monetary conditions and by exchange rate while, on the supply side, the domestic situation is determinative. They therefore tend to assume an intermediate position below those of commodities produced abroad but above those of commodities produced and consumed at home.65

Graham is assuming that the prices of the exported goods in the world market change as the quantities supplied by the country change. Upon that assumption his conclusion naturally follows. It is clear, therefore, that there is no real contradiction between the view of Subercaseaux and that of Graham. They differ from each other because the underlying assumption concerning the relative importance of the country under consideration is not the same to them. Which is the more correct assumption depends upon the nature of each case.

The movements in the sectional price levels consequent upon a fall in the exchanges of a country, as described in the preceding three paragraphs, normally lead to an expansion of its exports. If the exports of the country are stimulated, it is reasonable to presume that, “by the parity of reasoning,” the imports are checked. It is also possible, however, that the imports are not checked or are not checked to a similar extent as the exports are stimulated. Firstly the exports of the country may be elastic and its imports inelastic. In that case the rise in the prices of the imported goods will not lead to a decrease in the total value of the goods imported. Secondly, when the exports of the country are manufactures, which require for their production raw materials imported from other countries, an increase in its exports would necessarily cause an increase in the imports of the raw materials. Thirdly, if the exchange depreciation is quick in raising the incomes of the people of the country and if the exchange depreciation is slow in influencing the economic conditions of other countries, the income of the country would increase so that the same quantity or even a greater quantity of the imported goods are purchased in spite of the fact that their prices are higher in terms of local currency. What would be the actual effects of an exchange depreciation upon international trade depends upon the conditions of demand and supply, the nature of the monetary and financial structure of the trading countries, etc., and no definite rule can be laid down.66

Finally, it should be noted that the increase or decrease in the rate of exchange apart, the very possibility of changes or fluctuations in the exchanges means an additional element of uncertainty, and consequently would have injurious effects upon international trade.

In fact, Viner’s criticism is applicable only to the doctrine “that the exchange rates will vary in exact inverse proportion with the relative variations in the index number of prices of international commodities in the two countries”, but not to the doctrine of Heckscher. The former doctrine is one concerning change and the measurement of change, while Heckscher’s doctrine is one of economic statics. Heckscher’s proposition as summarized in the text remains unscathed in spite of Viner’s criticism.

• 1It is the practice of some economists to divide economic analysis into three departments, viz., statics, comparative statics, and dynamics. Economic statics deals with the conditions of equilibrium, if they exist at all. By conditions of equilibrium we mean those positions in which the price system will remain at rest, once it reaches them, provided that there is no fresh disturbance. That, however, does not imply that conditions of equilibrium must necessarily be reached ultimately by the price system. Comparative statics is concerned with the differences in the conditions of equilibrium in two or more price systems that are due to differences in the given data in the different systems. Economic dynamics deals with the problem, how a change in the fundamental data disturbs the price system and how the disturbances are corrected. It shows us, as Professor Robbins has pointed out, “the actual process of change—the path followed through time between one equilibrium position and the other.” In economic dynamics we have to see whether it is necessary or likely that the original change in data would lead to secondary changes in data. It is mainly because of those secondary changes in data that we have to differentiate between a doctrine of dynamics and one of comparative statics. In the latter we always suppose that when the original data change all other things will remain the same. In dynamics, however, we are not allowed to make the assumption of ceteris paribus.
• 2In dividing history into periods, we find it convenient to use certain dates, but the selected dates are only approximately accurate indicators. Nobody can point to a certain date as marking the exact point of time at which a historical stage or period either terminated or began. Nevertheless, we have to draw some lines of demarcation, however inexact they may be.
• 3The term, quantity theory of money, is sometimes defined as the theory that, in a given period of time, the product of the total volume of goods bought and sold multiplied by the average price of those goods is equal to the volume of the circulation of the medium of exchange. In such general terms, the theory does not indicate whether an increase in the circulation of money would lead to an increase in the volume of the goods or to a rise in their average price. The causal sequence may run either from the volume of the circulation to the average price of the goods transacted, or from the circulation to the volume of those goods. In the strict sense of the term, however, the quantity theory of money assumes the causal relation to run only from the circulation to the average price. We shall, in this essay, use the term in that strict sense. As to the theory that an increase (or decrease) of money would lead to an increase (or decrease) of the volume of goods and transactions, we shall call it “the doctrine that money stimulates trade and production”.
• 4Ibid., p. 34.
• 5Ibid., p. 69.
• 6Ibid., p. 105.
• 7For example, St. Thomas Acquinas, Nicholas Oresme, and Molinæus had developed, or at least indicated, a similar theory.
• 8Hales, op. cit., pp. 87, 115 f.
• 9That “theory of the store of value”, as it might be called, has been wrongly credited to Petty by Suviranta (The Theory of the Balance of Trade, 1923, pp. 52 f.). That theory is the natural outcome of the financial conditions in the sixteenth century and the greater part of the seventeenth century when gold and silver were alone easily exchangeable values, because foreign investment was practically unknown in those days and goods were not so easily exportable as at later periods.
• 10People having “plentie of things” and wanting to store up the surplus would naturally be willing to exchange these things for such wares as would lie in lesse Romes, and contineweth longest with out perishinge, and be carried to and fro with lesse charge, and be most currant at all times and at all places”. Gold and silver were “most of that sort” (Hales, op. cit., p. 72). See also p. 73.
• 11Discourse, Lamond’s ed., pp. 86–7; see also p. 113.
• 12Ibid., p. 35.
• 13That is in essence what has been called after Thomas Mun the balance of trade theory. The theory, in fact, had a long process of development. As early as the last quarter of the fourteenth century we find very definite statements of it in English by one of the officers of the Mint in his evidence before Parliament. The statement is as follows: “. . . As to the fact that gold and silver come not to England, whilst that which is in England is carried abroad, . . . if the merchandise exported from England be well and justly governed, the money which is in England will remain, and great plenty of money will come from abroad. It must be ascertained that no more foreign merchandise come within the realm than the value of the merchandise of this country that goes out of the realm.” However, the exact term balance of trade itself seems to have been coined in 1615 and the term favourable balance of trade in as late as 1767. See W. H. Price, “The Origin of the Phrase ‘Balance of Trade’,” Quarterly Journ. of Economics, xx (1905), pp. 157–167.
• 14Hales, op. cit., p. 102.
• 15Discourse, Lamond’s ed., pp. 88, 65.
• 16Ibid., pp. 47, 51 passim.
• 17Ibid., p. 40, see also pp. 43 ff., 47 f.
• 18It should be noted that the domestic demand for imported goods could not be permanently inelastic. When the balance of payments of the country turns unfavourable and specie flows out, then, other things being equal, the total income of consumers falls. The fall in consumers’ income will sooner or later lead to a reduction in the total value of its imports.
• 19Seigneur de Malestroit (second half of the sixteenth century) was a “member of the royal council of France and comptroller of the Mint”.
• 20“Lon ne peut dire qu’vne chose soit maintenant plus chere qu’elle n’estroit il y a trois cens ans, sinon que pour l’achepter il faille maintenant bailler plus d’or ou d’argent que l’on n’en bailloit alors. Or, est-il que pour l’achapt de toutes choses, lon ne baille point maintenant plus d’or n’y d’argent que Ion en bailloit alors. Doncques puis ledict temps rien n’est encbery en France. Les maxims sont clairs.” (Les Paradoxes sur le faict des Monnyes, Paris, 1578 ed. used, p. 4). Compare Paul Raveau, Précédé d’un Étude sur le Pouvoir d’Achat de la Livre, Paris, 1926.
• 21Les Paradoxes sur le faict de Monnyes, Paris, 1578, p. 15.
• 22Other writers, especially Spanish writers, must have more or less noticed the effects of the influx of the American treasure in raising prices. But there can be no doubt that Bodin is entitled to the credit for first giving due emphasis to the importance of the supply of the money metals and that it was from Bodin and not from others that later writers got that explanation of the price revolution. For that reason we feel ourselves justified in choosing Bodin as the representative exponent of that view. See Earl J. Hamilton, American Treasure and the Price Revolution in Spain, 1501–1650 (Cambridge, Mass., 1934), pp. 283–306 and Bernard W. Dempsey, “The Historical Emergence of Quantity Theory,” Q.J.E., 1935.
• 23Jean Bodin (1530–1596) was bora and educated at Angers. He spent his college days in studying law; became, after his graduation, a lecturer on jurisprudence and later practised law in Paris. Not being very successful at the bar, he entered government service. In 1568 he published his famous Reponse aux paradoxes de M. de Malestroit touchant l’encherissement de toutes les choses et des monnaies, which was followed in 1578 by his Discours sur le rehaussement et diminution des monnaies, pour reponse aux paradoxes de sieur de Malestroit. It is, however, as political philosopher that he is chiefly noted. In political theory he was sometimes considered as “one of the most important figures after Aristotle”. His famous work, Les six livres de la République, was published in 1576. In it he incorporated the substance of his Réponse. The Réponse has an English translation in A. E. Monroe’s Early Economic Thought. The reference is to Monroe’s edition, pp. 123–6.
• 24Ibid., p. 127.
• 25Ibid., pp. 128–131.
• 26Ibid., p. 131.
• 27As early as the Saxon coinage controversy around 1530, the opponents of depreciation put forward arguments suggestive of the quantity theory.
• 28By the Quantity Theory of Money we mean the doctrine according to which, to use the words of Wicksell, “the value or purchasing power of money varies in inverse proportion to its quantity, so that an increase or decrease in the quantity of money, other things being equal, will cause a proportionate decrease or increase in its purchasing power in terms of other goods, and thus a corresponding increase or decrease in all commodity prices.” One may, of course, use the term, quantity theory of money in a broader sense, viz. in the sense that
• 29Réponse (Monroe’s ed.), pp. 138 ff.
• 30Bernardo Davanzati (1529–1606), A Discourse upon Coins (1582). We have used the English translation by John Toland (London, 1696), p. 18.
• 31Ibid., p. 15.
• 32Antoyne de Montchrétien, Traicté de l’économie politique, 1615, Funck-Brentano ed. (Paris, 1889) used, p. 257.
• 33See Professor R. H. Tawney, Introduction to a new edition of Thomas Wilson’s “Discourse upon Usury” (London, 1925).
• 34Ibid., p. 60.
• 35See Memorandum prepared for the Royal Commission on the Exchanges, London, 1564, reprinted in Tudor Economic Documents (edited by Tawney and Power, London, 1924), vol. iii, pp. 346–359.
• 36Ibid., p. 350.
• 37“The Exchange is the gouernere of prices of all warres enterchangablye vented betweene this Realme and the Low Contreyes, because the greateste quantetye of wares transported ether outeward or inwarde is boughte by money taken vppe by Exchange, and also because, althoughe the wares he boughte with his owne moneye, in sellinge of his hondred poundes worthe of wares he Consydered what gaynes he myght have made by Exchange of so muche moneye, and he maketh the price of his wares accordingly or to some Convenyent overpluse.” (Ibid., p. 347.)
• 38See Tawney, op. cit., pp. 138–154.
• 39“Protest against the State Control of Exchange Business,” 1576, reprinted in Tudor Economic Documents, vol. ii, pp. 169–173.
• 40Tudor Economic Documents, p. 172.
• 41See Thomas Milles, The Customers Replie or Second Apologie (London, 1604); The Customers Apologie (London, (1601) 2nd ed., 1609); An out-port-Customers Accompt (London, 1609); and especially Customers Replie, pp. 4–5; ch. x, pp. 14–32.
• 42Thomas Gresham went even so far as to say: “As the exchainge is the thinge that eatts ought [eats out] all princes, to the wholl destruction of ther common well [weal], if itt be nott substantially looked unto, so likewise the exchainge is the cheffest [chief] and richest thinge only above all other, to restore your Majestie and your reallme to fine gowld and sillvar . . .” (“Letter to Queen Elizabeth on the Fall of the Exchanges,” 1558—quoted from Tudor Economic Documents, ii, p. 148). Gresham, it may be noted, has been known for his excellent exposition of the doctrine that bad money drives out good. The doctrine was given the name “Gresham’s Law” by H. D. Macleod in 1858. It had, however, been formulated also by other writers.
• 43Gerrard de Malynes (fl. 1586–1641), the son of an English Mint-master, himself became an Assay Master of the Mint and in 1609 was appointed commissioner on Mint affairs. He had wide experience in trade and was frequently employed by Elizabeth and James I as an adviser on commercial questions. His writings, which bear on the present section, are: A Treatise of the Caker of England’s Commonwealth (London, 1601) and England’s View in Unmasking of two Paradoxes (by De Malestroit) with a Replication unto the answer of Jean Bodin (London, 1603). For his later writings see Sec. 5.
• 44Treatise of the Caker, p. 2.
• 45Ibid., p. 3.
• 46To him the causal sequence ran as follows: “Foreign exchanges—money—commodities.” See also The Maintenance of Free Trade, p. 2.
• 47See Lex Mercatoria, p. 415; The Maintenance of Free Trade, pp. 14–15; The Center of the Circle, pp. 41 f.
• 48Caker, p. 35; see also The Maintenance of Free Trade, p. 61.
• 49Caker, p. 18.
• 50Caker, p. 34.
• 51Ibid., pp. 38–9.
• 52Ibid., pp. 42–3.
• 53Ibid., p. 99.
• 54Ibid., p. 10.
• 55Ibid., p. 11.
• 56England’s View, pp. 55, 60.
• 57“The property of the money . . . is, that plentie of money maketh generally things deare, and scarcitie of money maketh likewise generally things good cheape. Whereas things particularly are also deare or good cheape, according to plentie or scarcitie of the things themselves, or the use of them” (Caker, p. 10). See also England’s View, pp. 7 and 9. In discussing the value of “money” Malynes and his contemporaries (e.g. Mun) had used the term “money” in two senses: firstly in the modern sense, and secondly in the sense of money delivered to (or taken from) foreign countries, i.e. bills of exchange. But in the quotation cited the word “money” is unmistakably used in the modern sense.
• 58Professor J. W. Angell has said of Malynes that he was one who considered the value of money as “something absolute and unchanging” (The Theory of International Prices, p. 18). That is certainly incorrect.
• 59England’s View, p. 65.
• 60Ibid., pp. 65 f.
• 61Caker, p. 35.
• 62England’s View, pp. 65–6.
• 63Ibid., p. 70.
• 64Although he recognized that an inflow of specie could produce opposite results, he was of the opinion that a high rate of exchange did not necessarily cause an inflow of specie. He said that “it would be so, were it not for the tolleration of the monies to go currant farre above their value beyond the seas” (Caker, p. 48). In other words, his theory would apply to countries, the value of whose coins were identical with their metallic contents.
• 65Caker, pp. 117–118.
• 66See, for example, ibid., pp. 106 f.