From the Gold Exchange Standard to the Gold Standard: The Crucial Role of Edwin Walter Kemmerer

Transcription

1 25ème Journées Internationales d Economie Monétaires et Bancaires Luxembourg, Juin 2008 From the Gold Exchange Standard to the Gold Standard: The Crucial Role of Edwin Walter Kemmerer Rebeca Gomez Betancourt PHARE, University of Paris I, Panthéon Sorbonne Abstract When Edwin Walter Kemmerer established central banks in several Latin American countries in the 1920s, what he had in mind was to establish the gold standard as an effective system. He developed expertise in metallic standards along with a political interest in implementing them during his early experiences in American colonies. This paper analyzes the gold exchange standard and gold standard systems as they were formulated and developed by Kemmerer throughout his career as a money doctor ( ). The first part deals with the emergence of the concept of Gold Exchange Standard. I look at Kemmerer s early experiences as an economic advisor, his proposals for establishing the gold exchange standards and his explanation of the system itself. Then, I examine the debates over the Gold Standard (GS) and the Gold Exchange Standard (GES). The second part analyzes the various stages in the establishment of central banks and the problem of the choice of the monetary system after the Genoa Conference. Keys words: Kemmerer, Lindsay, Gold Standard, Gold Exchange Standard, Central Banks. Classification J.E.L.: B16, E31, E40, N11. rebeca.gomezbetancourt@malix.univ-paris1.fr. Adresse: PHARE, Université Paris I, Panthéon-Sorbonne boulevard de l'hôpital Paris, France. Tel.: +33 (0) I would like to thank M. Jérôme de Boyer for his careful reading and inspired suggestions. I also thank the members of my laboratory PHARE for giving me access to Kemmerer s archives, which turned out to be so precious for the present research. Any remaining error or omission is mine. This article has been presented at the H2M Seminar at the University of Paris I, Panthéon- Sorbonne on February 1, 2008 and in the International Congress on the Caribbean Economic History, Port au Prince, Haiti, on February 28, This work is still in progress: please do not quote without the author s permission.

2 A gold standard is the rule now in all parts of the world; but a gold currency is the exception. Keynes, 1913: Introduction The American economist Edwin Walter Kemmerer ( ) stands out as a key figure in the World s and more specifically Latin America s history of money and banking for the role he played as a money doctor. In the first thirty years of the 20 th century he traveled to over fifteen countries, often accompanied by a group of experts known as the Kemmerer missions. Their aim was to solve problems related mainly to the exchange standard and central banking. In his autobiography, he explains how he became a money doctor. 1 From the very beginning of his academic career Kemmerer chose to work on the quantity theory of money. It was the subject of both his master thesis and his doctoral dissertation, which he started in 1903 and published in Kemmerer states that this choice, as will be seen later, was largely responsible for my becoming a money doctor (Autobiography, n.d.: 23). In 1903, Kemmerer received a telegram from his former doctoral supervisor, Jeremiah Jenks, asking him to become a member of an American commission in order to implement monetary reforms in the Philippines. Jenks and Conant recommended Kemmerer for the job. This was the starting point for Kemmerer s career as a money doctor. He first served as an economic advisor in various colonies, namely, Sierra Leone, the Philippines and Puerto Rico. In the 1920s he worked as the head of commissions aimed at developing central banks and gold standard systems in Latin America. The aim of the present article is to examine the nature of the metallic standard systems proposed by Kemmerer in a number of Latin American countries and colonies. The role he played in establishing central banks in the Andean countries has already been studied 3 and is relatively well-known. However, his involvement in establishing monetary systems in colonies and small sovereign economies had received less attention. I argue that Kemmerer played a central part in developing and maintaining the gold standard (GS) in 1 Kemmerer s Papers. Mudd Library. Princeton University. Box N It was his first publication. See Kemmerer (1907) and Gomez Betancourt (2006). 3 It has been documented by Paul Drake (1989). 2

3 Latin America in the 1920s and in establishing the gold exchange standard (GES) in the American colonies prior to the First World War. More broadly, my aim is to identify Kemmerer s position on the monetary standards (GS for the core countries and GES for the peripheral countries) in the years following the First World War a period which was greatly affected by many changes in the international monetary system and with respect to the proposals drafted at the Genoa Conference. 4 Given my objective, the analysis will focus on the explanations of the specificities of the GS and GES systems as they were presented by Kemmerer throughout his career as a money doctor. When Kemmerer worked as financial advisor for the governments of some colonies or protectorates, he usually recommended the establishment of a GES. By contrast, he advocated the GS when he defined his plan to establish central banks in Latin American countries. Kemmerer believed that one of the main prerogatives of central banks was to ensure the effectiveness of the GS. Kemmerer s missions in the colonies and in Latin America will be studied in order to identify theoretical issues. I argue that in Kemmerer s economic theory there is a continuum between his early works as a financial advisor and his last missions, which made him famous as money doctor. 5 Section 2 deals with the emergence of the GES concept, from the works of Ricardo to those of Kemmerer. I reject the traditional reading according to which Ricardo became regarded as the father of the GES. Section 3 examines the early proposals made by Kemmerer when he worked as an economic advisor, 6 the specificities of the GES and the ways in which it differs from the GS. I then analyze the debates surrounding the GS and 4 One of the conclusions reached at the Genoa Conference (1922) was the necessity to establish GES and central banks for countries which had not yet created them. See Kemmerer (1944), the original text of the Genoa Conference (Conférence de Gênes et documents politiques, In particular, see pages 146 and for the financial commission) and Fink (1984: ). 5 Although Kemmerer conducted thorough studies and implemented monetary reforms in several economies, it can be claimed that he did not do history of economic thought. Kemmerer did not analyze former theories and debates. He disregarded important discussions on monetary theory. Kemmerer was therefore not a historian of economic thought. Even though; he was acquainted with the works of some important authors such as Cannan, Hawtrey, Jevons, Ricardo and Thornton. 6 I argue that his missions in American colonies and his examination of the reforms in India had a significant impact on his career. It could explain why his role shifted from an economic theoretician to a reformer of monetary institutions or Money doctor. 3

4 GES before and after the First World War. I will look closely at both systems and specify the strengths and weaknesses of each. Section 4 focuses on the various stages in the establishment of central banks and the choice of the monetary system in some of the countries visited by the money doctor in the 1920s. I will present some concluding remarks in section The GES forty years before the Genoa Conference The GES was the prevailing monetary system in the late 19th and early 20th century in a number of British and American colonies. After the First World War it was adopted in different versions by several countries following the recommendations of the Genoa Conference. 7 The GES was essentially implemented by three groups of countries before the war. The first one included countries which were on a silver standard such as India, the Straits Settlements, Indochina, Siam, the Philippines, Mexico and Panama. 8 The second one consisted of countries using paper money such as Argentina and Brazil. 9 The last group included Russia, the Austro-Hungarian Empire and the Scandinavian countries. 10 In this paper, I will focus on the first group only. 7 At the time of the outbreak of the First World War, the gold exchange standard was in operation in India, the Philippine Islands, Java, the Straits Settlements, Nicaragua, and a few other countries (Kemmerer 1933: 1). In the 1920s the Gold Exchange Standard had been proposed, as we will see, by Lindsay for India and some American economists (Jenks, Conant and Kemmerer) for American colonies, China and Mexico. 8 The steady decrease of silver was responsible for great fluctuations in the exchange of these countries. In order to stabilize the exchange, many economists proposed to create a reserve of currency (Abozzia, 1936: 29). 9 The system was extended, after some modifications, to countries using paper money for circulation. The currency board created in Argentina in 1899 and in Brazil in 1905 was also inspired by the system used in India. In both countries, the aim was to prevent inflation and the depreciation of paper money. The reform was very successful. In addition to stabilizing the exchange, both countries created a metallic fund which was designed in order to establish the gold standard later (Abozzia, 1936: 27-29). The establishment of the system was seen as a possible remedy to inflation and the depreciation of the exchange. 10 Before the War, the GES had been adopted in Europe by Russia in In this case, the currency board was not created. The Imperial Bank, which had accumulated a large stock of German currencies without creating a gold fund, declared itself ready to buy and sell gold currencies at a fixed rate. The experiment gave excellent results and the exchange remained stable until the First World War (Abozzia, 1936: 27-29). They adopted the GES in order to avoid creating the metallic fund which was required for the establishment of the gold standard. 4

5 According to Icard (1912) and Le Branchu (1923), the GES was the result of practical rather than theoretical considerations. 11 However, Kemmerer drew on the theoretical analysis of Alexander Martin Lindsay (1867), who worked at the Bank of Bengal, and elaborated it from 1903 onwards. The Origins of the GES concept: the forerunners Traditionally the GES has been associated with the Genoa Conference. It should be noted, however, that the system had already been established in several countries and in different conditions for over forty years. A careful reading of secondary literature on the subject reveals that the theoretical definition of the Gold Exchange Standard should not be directly attributed to David Ricardo (1816) and his Proposals for an Economical and Secure Currency, as it is sometimes claimed, 12 but to A. M. Lindsay. Although some researchers present the GES as the brainchild of Ricardo, Kemmerer (1904, 1905, 1906, and 1916) and Keynes (1913) amongst other economists argue that it had first been elaborated by A. M. Lindsay in 1876 for the Indian monetary system. Instead of a GES, Ricardo proposed a GBS. He presented the GBS in 1816 as a solution to the inconveniencies of the gold circulation. First, Ricardo believed that the GS was too costly for the government: the circulation of coins would deteriorate too quickly. Furthermore, gold was also hard to collect by central banks because it was often hoarded or exported, according to Ricardo. 13 Second, the central power of the bank could be threatened not only by too many depositors, but also by people holding small banknotes. To solve these problems, Ricardo recommended limiting monetary circulation to notes issued by the Bank of England, convertible into gold bullion but not into gold coins 11 The theory of the GES was elaborated from the results of the monetary reforms carried out in the countries. The Gold Exchange Standard was born from a short-term measure planned after reforms based on quantity theory failed in countries with paper money circulation and silver circulation (Icard, 1912: 162). It is somewhat like a medicine used without really knowing if it would have an effect, or what the effect will be, which proved efficient and was then used on a larger scale (Le Branchu, 1933: 63). 12 Among those attributing the paternity of the GES to Ricardo are Sayers (1953 and 1957) and Bordo & Schwartz (1984). 13 In Keynes words: A preference for a tangible gold currency is no longer more than a relic of a time when governments were less trustworthy in these matters than they are now, and when it was the fashion to imitate uncritically the system which had been established in England and had seemed to work so well during the second quarter of the nineteenth century. (Keynes 1913: 51). See also Lindsay Plan and the GES in Vol. XV of Collected Writings. Pages Kemmerer is in favor of gold coin circulation. 5

6 directly. This system would not only prevent the minting of gold coins but would also guarantee that the money in circulation had a counterpart in gold. According to Ricardo s proposal, only banknotes would circulate, while gold would be excluded from circulation. Despite the fact that gold coins did not circulate, the monetary unit would correspond to a specific weight of gold. Gold bullions were not transformed into coins. Gold would only circulate outside the country as an international means of payment. Domestic monetary circulation was therefore different from international circulation. Additionally, money hoarding would be reduced to a minimum as all gold coins were melted down and cast into bullion. The high value of bullion made it inaccessible to the public. 14 Central Banks were obliged to buy and sell gold at a legal price in bullion form in exchange for the banknotes issued. In the GBS, money reserves are therefore in the form of gold bullion. In proposing such a system, Ricardo s purpose was twofold. First, it was an attempt to ensure the security of the economic system. The currency was only convertible into gold bullion. In cases of a crisis, the bank had to comply with the requests to convert banknotes only for a sum of money at least equal to the value of bullion. In short, Ricardo s plan was to ensure the convertibility of banknotes into bullion. Second, he was aiming to establish a more economical system. Ricardo was aware that in a paper money system no labor was required in the circulation of banknotes 15. From Ricardo to Lindsay The GES concept was first proposed by A. M. Lindsay (1876) in an article entitled Ricardo s Exchange Remedy. 16 It was then developed and reinterpreted by American economists in order to be implemented in their colonies from 1903 onwards (Kemmerer, 16: 79-92). 14 Of course, large bars may be cut up into small pieces by private individuals and sold to the public as merchandise, and this is done, to some extent. No gold is used in the form of coins for hand-to-hand circulation. The gold-bullion Standard is, therefore, much more economical in its use of gold than is the gold-coin Standard, although it is less economical than the gold-exchange Standard. (Kemmerer, 1944: 175). 15 Gold money is a commodity like all the commodities, produced at a cost that can be reckoned in terms of the labour used in the process of production. Paper money, on the contrary, when it is not kept convertible into gold (as Ricardo himself showed was possible by using a gold exchange standard) is produced at virtually no cost (Sayers 1957: 7). See also: Bonar, J. (1923). Indeed, this idea was proposed earlier by Adam Smith (1776). 16 Lindsay himself claimed that he drew on proposals made by Ricardo in 1816, which makes him initially responsible for the confusion between his own and Ricardo s plan. 6

7 The GES in the form in which it has been adopted in India is justly known as the Lindsay scheme. It was proposed and advocated from the earliest discussions, when the Indian currency problem first became prominent, by Mr. A. M. Lindsay, deputy secretary of the Bank of Bengal, who always maintained that they must adopt my scheme despite themselves. His first proposals were made in 1876 and They were repeated in 1885 and again in 1892, when he published a pamphlet entitled Ricardo s Exchange Remedy. Finally, he explained his views in detail to the Committee of (Keynes 1913: 24). Lindsay s work has remained relatively unknown in the history of economic thought. Some references to his articles appear however in Conant (1909), Mises (1912), Laughlin (1927) and Dimand (1991), themselves quoting from Kemmerer s and Keynes works. Nevertheless, I believe that more attention should be paid to Lindsay s proposals for monetary reforms and, in particular, that the primary sources should be more closely examined. 17 Of a much greater importance was the second plan for securing a gold standard without a gold currency. This Plan, which provided for a system of redemption in drafts on a gold fund located outside of the country, was named after his chief advocate, A. M. Lindsay, deputy secretary and treasurer of the Bank of Bengal. It appears to have been first proposed for India by Mr. Lindsay in It was discussed by him in the Calcutta Review for October 1878, and July 1885, also in the Bankers magazine (London) of August and September In the latter month Mr. Lindsay published a pamphlet on the subject, entitled Ricardo s exchange Remedy, a Proposal to Regulate the Indian Currency by making it Expand and Contract automatically at fixed Sterling Rates with the Aid of the Silver Clause of the Bank Act. The scheme was described and supported by Charles MacDonald in November 1892, in his testimony before the Herschell Committee. (Kemmerer, 1916: 80). Lindsay (1878) advocated a system in which gold did not circulate. Hence, on the one hand no minting of gold coins could occur, and, on the other, gold reserves in the form of coins or bullion were negligible. It was designed to provide India with a circulation means that was suitable to its trading needs. Also from an economical perspective it was worth supporting this system since it did not call for large gold reserves. The circulation means was exclusively fiduciary money, i.e. coins and banknotes convertible into the monetary unit of the country on the GS. Convertibility into gold was not required but convertibility into drafts and currencies had to be ensured. Lindsay recommended the establishment of a GBS in Great Britain and a GES in India. For the latter, the country s monetary gold was to be centralized in a fund known as the Gold Standard Fund. A large part of this fund was kept abroad to raise interests. Reserves 17 The analysis of Lindsay s archives is under way. 7

8 were kept in the form of currencies, bonds convertible into gold, and for a very small part in gold. Most of it was at no cost because it produced a yield. The GES was the most economical version of the different kinds of Gold Standards known before the First World War. It was designed for countries which did not have sufficient gold resources at their disposal. These countries were unable to provide the convertibility of money issued and to ensure gold circulation. The main purpose was to guarantee the stability of the exchange rate (Lindsay, 1878: ). 18 I am indeed puzzled why some economists have failed to differentiate Lindsay s plan from the one of Ricardo. Even if Lindsay used Ricardo s material to draw the main principles of his reform plan, the two plans called for distinct monetary reforms with different goals. However, certain similarities between the two plans may explain this common confusion. Both Lindsay s and Ricardo s plans were characterized by the absence of gold coinage and gold coin circulation. Moreover, both were in favor of the most economical system. 19 As Keynes wrote, Ricardo s proposals for a sound and economical currency were based on the principle of keeping gold out of actual circulation (Keynes 1913: 51). The main feature that both proposed systems hat in common was that banknotes were not convertible into coins. Although Lindsay s plan was adapted to India s economic situation, it required disposing the English GCS, a proposal which was not well-received at the time. He was harshly criticized by members of the government and financiers Speaking as a theorist, I believe that it (of the GES) contains one essential element the use of a cheap local currency artificially maintained at par with the international currency or standard of value (whatever that may ultimately turn out to be) - in the ideal currency of the future (Keynes 1913, 25). 19 The reasons for this change are easily seen. It has been found that the expense of a gold circulation is insupportable and that large economies can be safely effected by the use of some cheaper substitute; and it has been found further that gold in the pockets of the people is not in the least available at a time of crisis or to meet a foreign drain. For these purposes the gold resources of a country must be centralized (Keynes 1913: 50-51). Note that Smith had used the same argument of the cheapest system as early as Lord Farrer described it as far too clever for the ordinary English mind with its ineradicably prejudice for an immediately tangible gold backing to all currencies. Lord Rothschild, Sir John Lubbock (Lord Avebury), Sir Samuel Montagu (the late Lord Swaythling) all gave evidence before the Committee that any system without a visible gold currency would be looked on with distrust. Mr. Alfred de Rothschild went so far as to say than in fact a gold standard without a gold currency seemed to him an utter impossibility. Financiers of this type will not admit the feasibility of anything until it has been demonstrated to them by practical experience. It follows, therefore, that they will seldom give their support to what is new. (Keynes 1913: 24-25). 8

9 From Lindsay to Kemmerer In 1903, the governments of Mexico and China called for the services of the most competent American economists to conduct a study on the stability of the exchange rates between currencies of countries using different monetary standards. Appointed by the President of the United States, Theodore Roosevelt, professors H. H. Hanna, J. V. Jenks and Ch. A. Conant promptly commissioned a group of experts known as the Commission of Gold Exchange Standard 21. They met to determine their tasks and headed for various capitals in Europe and Asia, taking with them disciples and Kemmerer was one of them. This was how he made his debut as a money doctor. The most active members of the commission soon became the theoreticians of the GES and attempted to implement it in their new colony, i.e. the Philippines 22. The aim of the Commission was to unify monetary systems. By opting for a new relation between gold and silver, they meant to seal the decline of silver, thus cutting their losses in order to prevent any further decline. (Le Branchu 1933: 75). Kemmerer s first patient: the Philippines Islands The chief of the Insular Bureau at Washington, Colonel Clarence P. Edwards, was at first quite skeptical about Kemmerer. He considered him too young and not experienced enough for the job he had been assigned for. However, he eventually called Kemmerer, who subsequently left for Manila in June 27, 1903 with his wife. During the thirty-day-long ship journey to the Philippines, Kemmerer spent his time reading recent studies on the 21 The Commission members summarized the results of their investigation in two reports presented to the American Parliament. The reports advocated the adoption of the Gold Exchange Standard in countries with a depreciated currency. It was the first time that the empirical measures experimented in India were approved of by monetary experts (Abozzia, 1936: 26). Fisher (1911: ) shows the importance of the results of the studies conducted by the GES Commission in which Kemmerer took part. 22 Since the Indian system has been perfected and its provisions generally known, it has been widely imitated both in Asia and elsewhere. In 1903 the government of the U.S introduced a system avowedly based on it into the Philippines. Since that time it has been established, under the influence of the same government, in Mexico and Panama. The government of Siam has adopted it. The French have introduced it in Indo-China. Our own Colonial Office has introduced it in the Straits Settlements and are about to introduce it into the West African colonies. Something similar has existed in Java under Dutch influences for many years. The Japanese system is virtually the same in practice. In China, as is well known, currency reform has not yet been carried through. The GES is the only possible means of bringing China on to a gold basis, and the alternative policy is to be content at first with a standard, as well as a currency, of silver. A powerful body of opinion, led by the US, favours the immediate introduction of a GS on the Indian model. (Keynes, 1913: 25). 9

10 Philippine monetary issue. 23 He stayed there for two and a half years and made several monetary reform proposals. He was recommended to Governor Taft by Jenks (his former doctoral advisor) to present a plan aimed at solving the monetary issues in the Philippines. The core feature of his plan was to substitute the silver-standard system (in operation at this time) for a goldstandard currency system. The Philippines then used as money the Mexican peso whose value in gold was approximately the same as that of the silver it contained. Nevertheless, the silver value fluctuated continuously, affecting the Philippines economies. To partially solve this problem, the new Philippines peso (called the Conant) contained less pure silver than the Mexican peso. The day following his arrival, Kemmerer proposed his first plan, which was rejected. He then started drafting a new plan aimed at establishing a gold exchange standard. This bill, with a few minor changes, became the Philippine Coinage Act of October 10, It provided the nearest approach to a «pure» gold-exchange standard that the world had yet seen (Autobiography: 40). This law provided the Philippines with a monetary system similar to the one used in India. The Philippines were facing the same monetary problems as India, with the only difference that the balance of payments showed a large deficit: an essential prerequisite for a better financial situation was missing. As a result, the monetary reform in the Philippines was a blatant failure in the year it was implemented. In late 1904, the United States had a significant increase in military spending, and the balance of payments became favorable to the Philippines (Abozzia, 1936: 26). Kemmerer s intellectual sources for his definition of the Gold Exchange Standard were Ricardo (1816), Lindsay (1878), Jenks (1904), Conant (1909) and Laughlin (1927). 24 Nonetheless, Kemmerer himself was the forerunner and theoretician of metallic standards, along with his fellow economists, and one of those, who most contributed to the definition and diffusion of the concept of the GES. 23 To the best of my knowledge, it was during his journey to Manila where Kemmerer appears to have studied the reforms that Lindsay had proposed for India. 24 The three monetary experts Conant, Jenks and I, who had assisted the government in its currency reform plans, were subjected to much public criticism and ridicule (Autobiography: 33). 10

11 3. Different types of Gold Standards according to Kemmerer For Kemmerer (1938: 1-2) the gold standard is a monetary system in which the unit of value, be it the dollar, the franc, the pound, or some other unit in which prices and wages are customarily expressed and in which debts are usually contracted, consists of the value of a fix quantity of gold in a free gold market. Elsewhere, he defines the gold standard as the monetary system in which units of value, whether prices, wages or debts, are expressed as a fixed quantity of gold (Kemmerer, 1944). 25 This definition does not include certain elements usually associated with the gold standard, such as the gold coin circulation, the definition of the legal tender, free coinage, and the convertibility of banknotes into gold. Respectively, Kemmerer writes: these things are all customary accompaniments of the gold standard. They are useful devices for maintaining it. However, the gold standard could exist without any or all of them Furthermore, a currency might have all these attributes and still not be a true gold standard (Kemmerer, 1934: 4). Kemmerer (1944) explains that what remains fixed on the gold standard system is the weight of the metallic content of the monetary unit and not its value. The law has to provide for carefully determined units of measure, i.e. the amounts of gold corresponding to each monetary unit. However, its value is assimilated to its purchasing power. He uses the example of distinct units of measure, i.e. the pound or the kilogram, to illustrate measures with fixed weights. The gold dollar, for instance, is the value assigned to a given weight of gold at a given moment. The main problem that may arise is when this value is modified. The monetary unit is thus defined with respect to a fixed weight but not a fixed value. According to Kemmerer, two conditions are necessary in order to reach what he called the automatic mechanism of the gold-standard. First, the state must set the amount of gold contained in the monetary unit. For a dollar, for instance, it would be 25.8 grams of gold 9/10ths fine or grains of pure gold. Kemmerer argues against all kinds of restrictions or taxes on gold imports and exports: the authorities must not impose seigniorage. Second, coinage must not be limited, i.e. minting houses must be obligated to use all the gold they receive A fixed amount of gold has the advantage of enabling the Central Bank to make up for the temporary or seasonal fluctuation of the domestic monetary demand with a smaller quantity of gold (Hawtrey, 1919: 102). 26 The authorities must be mindful also of the relationship between the gold value and the amount 11

12 Kemmerer (1944) identifies three types of gold standards depending on whether all or only some of the characteristics defining the GS are present. I will first give a brief definition of the various systems of metallic standards identified by Kemmerer: the gold coin standard system, the gold exchange standard system and the gold bullion standard system. Consequently, I will explain the differences in the way they worked before the First World War. The first system is the Gold Coin Standard (GCS), 27 also called Gold Specie Standard (GSS) or Gold Standard (GS). 28 In this system, gold coins circulate within the country, banknotes are convertible in gold coins without restrictions, and coinage is unlimited at costs generally determined by the law. The second type of metallic standard, on which my analysis is focused on, is the Gold Exchange Standard (GES). The main purpose of this system is to ensure the stability of the exchange rate. Following Lindsay, Kemmerer defines the GES as a system in which there is neither free coinage of gold nor domestic gold coins circulation, although the latter does not compromise the effectiveness of the mechanism. Currency reserves consist of short-term interest-raising investments on foreign stock markets. The third type represents the Gold Bullion Standard (GBS). It relies neither on domestic gold circulation nor on unlimited gold coinage. Gold is stored in the issuing bank and only serves to maintain the stability of the exchange rate. Banknotes are convertible of gold produced. Gold production is regulated through variations of production costs and not through variations of the gold prices. For other types of goods, production increases when prices increase (and falls when prices fall). Gold stands as an exception among all types of goods since its price cannot change: it is fixed by law. Kemmerer explains how the price of gold remained fixed at $20.67 per ounce from 1879 to 1916, even though gold production increased fourfold over the period. Provided that the price is fixed and not the value under the GS, gold producers will keep on receiving the same amount of money at the Mint. However, values may vary, in particular if the gold value falls. In this case, the prices of other goods increase. Gold production may be found to decrease as a result of a decrease in the producers benefits. By contrast, the gold value increases only if there is a decrease in the prices of other goods as well as in the cost of gold production. The production thus tends to increase as soon as the gold value increases. 27 Ricardo and other economists after him heavily criticized the GCS. Under this monetary system, a country was required to have a large stock of gold at its disposal since it was the base of the credit. Alternatively, gold money may be kept in a bank entitled to issue banknotes. It may then agree to grant a certain amount of credit by issuing extra banknotes or buy drafts with new banknotes so that the amount of credit granted may exceed the amount of gold in the reserve. 28 In the United States, the establishment of the gold standard came out of a theoretical controversy between the proponents of the gold standard and the proponents of bimetallism (and those in favor of the silver standard). This debate turned into major political arguments in the early 20th century. See among others Bordo, M. and A. Schwartz (1984), Eichengreen (1985) and Gomez Betancourt (2007). 12

13 only above a certain amount (equivalent of a bullion) and exclusively for exportation. This system was elaborated and presented by Ricardo (1816), 29 but implemented much later. It served as a transition between the GCS and the GES. In what follows, I compare the first system (GCS) with the second (GES), an analysis, which Kemmerer already undertook. Kemmerer s GES For Kemmerer, [t]he gold exchange standard is a variety of gold standard, since the unit of value is the value of a fixed quantity of gold in a free gold market. Its outstanding characteristic is that it provides for redemption of the various forms of fiduciary money drafts on gold funds located abroad, rather than in gold coin or gold bullion at home The GES is essentially a mechanism for providing a gold standard without a gold currency (Kemmerer 1933: 1). The GES ensures the convertibility of fiduciary money in drafts. The aim of Kemmerer and the other GES Commission members was to propose a system which retained the main advantages of the GCS but could be adopted by economies that either did not have the possibility to sustain gold circulation and gold reserves, or could not ensure convertibility in bullions. To explain how the system works, let us look at the first GES implemented by Kemmerer in the Philippines in The establishment of the GES in the Philippines Islands The case of the Philippines Islands between 1903 and 1910 represents one of the first illustrations of the GES. When the United States took control of its new colony, a group of economists was sent to develop a plan for monetary reforms and reorganize the financial currency system. In early 1903 the Philippine currency mainly consisted of silver pesos, silver coins and money certificates issued by the government. To 100% they were backed up by the reserves of silver pesos kept in the government fund and by banknotes. In addition, a small amount of American dollars were in circulation, but virtually none in the form of gold coins. Any Philippines gold coins were minted since the Philippines were an American colony, and the coins minted under the Spanish dominion were no longer in circulation. In 1903 the reform proposed by the American GES Commission was adopted by the Philippines. This reform provided a new monetary unit, which would be a theoretical 29 On the GBS see Charles Rist (1938: ). 13

14 weight of 12.9 grain of gold, i.e. 50 cents of American dollar. Although no coinage was produced, it was clearly stated that all forms of money had to remain at par with gold. In order to maintain the parity between the Philippines silver coins and gold at that rate, a special reserve fund known as the Gold Standard Fund 30 was created. It was a very important institution since there were no central banks in the colonies, and not even, for that matter, in the sovereign country (i.e. the United States). The Gold Standard Fund consisted mainly of a loan allocated by the United States, but also included benefits resulting from the minting of new coins as well as other revenues related to the administration of the currency. It was a real reserve fund, distinct as it was from other government funds and used exclusively to maintain the Philippine currency at par with gold. Under the reform, part of the fund was be located in Manila, the other in New York City (Kemmerer, 1933: 312). 31 The gold points system under the GES 32 Kemmerer does not distinguish between the price-specie-flow-mechanism and the gold point mechanism. 33 He always refers to the exchange market when explaining gold movements between countries, but the causes for these movements are always associated with some monetary issue. I consider that he gives a thorough account of the gold points system, but his interpretation is limited to disruptions resulting from an excess or a deficit of the amount of money in circulation. Kemmerer explains that one of the consequences of trade between countries on the GS is that whenever the balance of payments of one country is in deficit, the exchange rate goes down, as a result of a relative abundance of the country s money supply and consequently gold is exported. Conversely, if one of the GS-countries engaged in trade has a favorable balance of payments, then the exchange rate goes up, resulting in a relative scarcity of the local currency, and gold is then imported. 30 See Kemmerer (1904), (1905), (1916), (1933) and (1944). 31 Lindsay (1878) also proposed to create a fund aimed at maintaining the currency in par with gold, but it was to be established only in London. 32 Kemmerer 1944 shows that he understood how the gold point system worked under a classical gold standard and under a gold exchange standard. We find that, from very early on Kemmerer 1904, 1905, 1906 and Under the gold exchange standard, as it existed in the Philippines in 1905, this same fundamental principle [gold points] applied (Kemmerer 1933: 314). 33 See de Boyer & Diatkine (2008) and de Boyer (2007). 14

15 He then considers exchange market to explain how the GES system works. Under the GES, gold physically remains outside the country (there is no gold in the colonies) but nonetheless the same mechanism is at work as in the gold point system. Redemption is now in the form of dollar drafts in New York City (Kemmerer 1916: 321). The Philippine funds deposited in American banks had to be used in case of a disturbance that had a chance to undermine the parity of the currency. Thus, for instance, when the exchange rate of the peso to the dollar fell to the Philippines gold-export point (of entry to the United States), gold was not literally exported on foreign markets as it would have been under similar circumstances in the United States and other countries on GCS or GBS. The Philippine Government would actually give to the gold exporter (in exchange for his Philippine money in Manila) a draft, which entitled him to a convertible deposit in New York. In other words, whenever the exchange rate reached the gold exit point, the Gold Fund would buy the Philippine silver pesos from the exporter and sell gold drafts in New York, or, to put it differently, the exporter would sell silver pesos and buy convertible dollars. In this operation, the Gold Fund took a commission equal to the hypothetical cost incurred in the export of gold bars from Manila to New York, had the exporter chosen to conduct the operation itself (Kemmerer 1933 : 314). The amount of silver pesos paid in exchange for the drafts in Manila was equal to the amount of gold it would have sent. 34 If there was a deficit in the balance of payment, Kemmerer adds, it would be settled by a transfer of drafts payable with gold deposited in the foreign country (in this case, New York City) thereby indicating that there was an excessive amount in circulation. As a result, it was necessary to export some gold in order to reduce the monetary supply. Silver pesos were often withdrawn from circulation by the government. This had the effect of reducing money circulation as effectively as if an equivalent amount of silver pesos had been exported. 35 When under the prewar Philippine type of the gold-exchange Standard a government sold for cash at home drafts on a reserve fund located abroad, it withdrew from circulation the purchase Money and contracted the monetary circulation by the amount withdrawn, and when foreign depository of the reserve 34 The tax charged by the Philippine government for the exchange of dollars in New York and the tax charged by the depositary bank in New York for the exchange of the peso in Manila represents the charges of the transportation of gold bars between the two cities. 35 For Kemmerer, the money supply meets the money demand and the parity of the currency is maintained by reducing circulation in times of relative abundance of money and increasing it in times of relative scarcity. 15

16 sold drafts on the reserve fund in the home country, the draft was paid in cash, increasing the monetary circulation by its full amount (Kemmerer, 1944: 167-8). Kemmerer considered that the GES had the same advantages in terms of regulating the monetary supply with respect to trade requirements as the GCS or the GBS. But unlike under the GCS, there was no circulation of gold coins in the Philippines, nor was it required to keep gold reserves, as it was necessary under the GBS (Kemmerer 1933: 315). If, on the other hand, the peso exchange rate reached the Philippines gold-import point (which meant that the exchange rate had risen toward the gold-export point in New York City), custodian banks in New York City gave the gold exporter a draft granting him the right to the equivalent in pesos in Manila. In other words, if the peso reached the gold-import point in the Philippines, then depositary banks in New York City sold silver pesos kept in the Gold Fund (in Manila), that is, they sold silver pesos and bought gold drafts. Moreover the Gold Fund was required to pay a commission in order to cover the expenses it would have incurred (on the gold coin standard) if it had transferred an equivalent amount of gold bars from New York City to Manila (Kemmerer 1933: 315). Once the drafts on the Gold Standard Fund reached Manila, they were paid in pesos and withdrawn from the fund, leading to an increase in the circulation of silver pesos. 36 These reforms and the establishment of the GES made it possible for the Philippines to stabilize the exchange between the gold-import point and the gold-export point. 37 Kemmerer s GES was not a Currency Board The system implemented in the Philippines was not quite an orthodox currency board. An orthodox currency board requires that the money issued is to 100% backed up by reserves of foreign currencies. More broadly, a currency board is a monetary system with a fixed 36 When the exchange rate fell to the gold-export point, the government announced: So we will sell on demand gold drafts on New York on unlimited quantities at these gold-export point rates, and will relieve the currency redundancy by withdrawing from circulation the pesos paid to us fore theses drafts (Kemmerer 1944: 159) By contrast, when the peso became stronger and the exchange rate reached the gold-import-point, the government stated: So far they may fall and no farther, a further decline would signify an appreciation of our peso above its legal gold par if we were on a gold coin standard, an it will be so interpreted under the Gold Exchange Standard, therefore, we will send on demand pesos laid down in Manila in unlimited quantities at these gold (or currency) import-point-rates, to be paid for by gold or its equivalent in New York. This will relieve the currency scarcity in the Islands by poring pesos into circulation from the Gold Standard fund (Kemmerer 1933: 316). 37 Countries with a depreciated currency could hope for a better situation with the adoption of the Gold Exchange Standard (Abozzia, 1936: 25). 16

17 exchange rate in which the issuing of local money depends on the amount of foreign currencies in the reserves. A relatively stable foreign currency is chosen as the reserve money to support the issuing of money of the country. The country s amount of money in circulation may then increase (or decrease) only if the amount of reserve money kept in the country s central bank increases (or decreases). One difference between a currency board and the GES elaborated by Kemmerer for the American colonies was that the Gold Standard Fund did not issue money (whether in banknotes or deposits). Instead money was issued by private banks. The Gold Standard Fund was an exchange stabilization fund. It was neither a money-issuing bank nor a central bank. For Kemmerer, the fund was not meant to redeem banknotes and deposit from currency reserves. Reserves were used to provide a sufficient stock of currencies to meet the demands of local manufacturers and local salespeople. Thus Kemmerer considers the exchange stabilization fund of the GES as different from a central bank. In his view, a central bank may be used for exchange stabilization, but, as the analysis of the debates surrounding the creation of the FED made clear, it also serves other important functions, such as money elasticity and the stabilization of interest rates on the money market. 38 The advantages of the GES Kemmerer (1933: 316) argued that the GES had several advantages over other metallic standards. This system was better suited to the specificities of the Philippines and other colonies since it allowed the silver coins circulation, which was most adequate for the small transactions then prevailing inside the colonies. One important argument supporting the GES is the economical aspect it provides. Many countries cannot afford a GCS or a GBS. Under the GES the only type of money in circulation is fiduciary money, which is much more economical than gold coins. No gold coins or bars were hoarded, nor was any reserve kept in order to meet potential demands. Additionally, Kemmerer pointed to the advantage of concentrating all reserves in one fund: the Gold Standard Fund. This is reminiscent of his proposals for the FED when he also suggested centralizing the reserves scattered throughout the United States in order to make them available. In Kemmerer s words, where ever dollar can be immediately used in times of need, whereas gold coin in circulation and gold bars in private hands are difficult 38 See Gomez Betancourt (2007). 17

18 to mobilize in emergencies, for the very emergency that creates demand for the gold is likely to cause the public to cling to it more tightly (Kemmerer 1933, 316). There were advantages for colonies on the GES because it received a commission for the exchange when the government sold drafts through the Gold Standard Fund. Moreover, interests were raised by the fund deposited abroad. Finally, the GES facilitated exchanges between the sovereign country and its colonies, which promoted the development of trade between those countries. Importers and exporters are always certain to find the currencies they need at a fixed rate. Making the exchange process easier definitely gives a new impulse to business. 4. Central banks and monetary systems in the 1920s After the First World War and especially after the Genoa Conference in 1922, the GES came to be modified. One of the main differences is that the post-war GES was no longer administered by the government but by the central bank of each country. The establishment of the GES in the 1920s thus required the establishment of central banks in countries where they were not yet present. The characteristics of the GES after the War The GES serves to stabilize the exchange rates of countries with depreciated currencies relative to the monetary units of countries on the gold standard. For Kemmerer, the establishment of the GES therefore requires that a number a countries maintain the GCS. 39 Thus, countries on the GCS act as international reserve centers towards countries on the GES. 40 The cash balance was composed by a small part of gold and a large part in foreign currencies. These foreign currencies in the bank reserves must to be redeemable into gold. The money supply of countries on the GES in the 1920s essentially consisted of banknotes issued by the central bank. The reserve needed to guarantee the parity of the currency was mainly kept in the financial center abroad; in the form of deposits redeemed in the United States. 39 By definition, the GES can only exist within an international framework. 40 The 1913 reform is reminiscent of the GES. It allowed for the creation of the FED by superimposing 12 Federal Reserve Banks on affiliated banks, and imposing on member s banks to put a certain amount of their deposit in the Federal Reserve Banks. The GES also distinguishes issuing banks between those issuing gold currencies and those accepting them. They are also required to keep a small amount of these currencies. See Le Branchu (1933: 85) and Gomez Betancourt (2007). 18