The Fund Agreement in the Courts—XVI
  • 1 0000000404811396 Monetary Fund

This sixteenth installment in the series dealing with the effect of the Articles of Agreement of the International Monetary Fund on litigation 1 examines a proceeding before an arbitral tribunal established under the Agreement on German External Debts in which the main issue was the meaning of the expression “the least depreciated currency” in relation to two revaluations of the deutsche mark. An important facet of this issue was whether the expression had to be understood within the context of the international monetary order established by the Articles.


This sixteenth installment in the series dealing with the effect of the Articles of Agreement of the International Monetary Fund on litigation 1 examines a proceeding before an arbitral tribunal established under the Agreement on German External Debts in which the main issue was the meaning of the expression “the least depreciated currency” in relation to two revaluations of the deutsche mark. An important facet of this issue was whether the expression had to be understood within the context of the international monetary order established by the Articles.

This sixteenth installment in the series dealing with the effect of the Articles of Agreement of the International Monetary Fund on litigation 1 examines a proceeding before an arbitral tribunal established under the Agreement on German External Debts in which the main issue was the meaning of the expression “the least depreciated currency” in relation to two revaluations of the deutsche mark. An important facet of this issue was whether the expression had to be understood within the context of the international monetary order established by the Articles.

“The Least Depreciated Currency”

On May 16, 1980, the Arbitral Tribunal for the Agreement on German External Debts, consisting of seven members, delivered a decision in the case of The Government of the Kingdom of Belgium, the Government of the French Republic, the Swiss Federal Council, the Government of the United Kingdom of Great Britain and Northern Ireland, the Government of the United States of America, Applicants v. the Government of the Federal Republic of Germany, Respondent. 2 The decision, which was concurred in by a majority of four members with three members dissenting, dealt with the interpretation of Article 13 of the Agreement on German External Debts, London, 27 February 1953. 3 The Agreement will be referred to as the LDA (London Debt Agreement) in this discussion of the majority and minority opinions of the arbitrators. The opinions and submissions of the parties take account of various aspects of the Fund’s Articles.

background of the case

Under the Versailles Peace Treaty of July 28, 1919, Germany agreed to pay compensation for losses sustained as a result of the First World War. In April 1921, the amount of reparations was fixed at 132 billion gold marks, and a plan for payments was imposed in May 1921. Germany was unable to perform its obligations under this plan. A scheme—the Dawes Plan—was worked out to reduce the war debt. To help stabilize the German currency and pay the debt, a foreign loan of 800 million gold marks was negotiated in various currencies. Reparation and interest payments on U. S. dollar bonds, which were issued in the United States, were to be made in gold. In September 1928, a new effort was made to solve the problem of reparations. The result was the Young Plan, under which the reparations debt was to be paid in annual installments. A proportion amounting to US$600 million was to be payable in foreign currency, but payments under the Dawes loan were to be set off against this sum. A further agreement to carry out the purpose of the Young Plan took effect in 1930. It provided for the issue in international markets of loan bonds in nine currencies, for a total value of $300 million. The Bank for International Settlements (BIS) was formed to ensure observance of commitments under the Young Plan. It acted as Trustee for creditors under the loan, and it drafted the General Bond, which was signed on June 10, 1930. The General Bond, which set forth the conditions for the loan bonds, including a gold clause, took the form of an agreement between the German Government and the BIS as Trustee for the holders of issued and outstanding bonds. Only part of the obligations were met in accordance with the terms of the General Bond before the Second World War.

In order to integrate the Federal Republic, which was founded in 1949, into the world economy, it was considered necessary to settle the problem of outstanding foreign debts. The Government of the Federal Republic accepted liability for outstanding prewar debt and expressed its willingness to join in a plan for settlement of the debts of the Government and of public authorities, including the economic aid received after 1945, and all arrears of interest. In April 1951, the three Western occupying powers (hereinafter referred to as the Three Powers) set up the Tripartite Commission on German Debts to act on their behalf in the negotiation of a settlement. The Commission was to prepare a plan for settlement, which would be incorporated in an agreement among participating governments. The Three Powers made it clear that the settlement would not provide unfair or privileged treatment for any group of creditors and that the terms of settlement would include no variation based on the currencies in which obligations were expressed.

The meetings of the Tripartite Commission with representatives of the Federal Republic were attended on occasion by observers from the Belgian, Dutch, Swedish, and Swiss Governments and representatives of German debtor interests. The Tripartite Commission announced the willingness of the Three Powers to forgo substantial proportions of their claims in respect of economic aid given after the Second World War.

The Conference on German External Debts opened in London on February 28, 1952 and continued until August of that year. The Tripartite Commission represented the Three Powers; the private creditors of these countries were represented by separate delegations; 22 creditor countries sent national delegations composed of governmental representatives and, in many cases, representatives of private creditors; 3 countries sent observers; the BIS was represented as a creditor in its own right; and the delegation of the Federal Republic included representatives of the government and of private debtors.

The German representatives pointed out that a gold clause would be invalid under both U.S. and German law and deduced that creditors would have to be satisfied with payments in currencies in their depreciated state. The Tripartite Commission also objected to a gold clause as a term of the settlement, but did not object to a protective clause in some other form. Other delegations insisted on a gold clause, because they considered it to be the most important feature of their existing rights. Various protective clauses were proposed as compromises, but, until a solution was found, they were resisted as favoring one group of creditors over another.

The LDA gave effect to the understandings reached in the London Conference. It was signed by the Federal Republic of Germany, the states represented by the Tripartite Commission, and 15 other creditor countries. Subsequently, other countries acceded to the agreement. The three texts—English, French, and German—in which the LDA was written were equally authentic.

clause in dispute

Article 2 of Annex I, Section A of the LDA dealt with outstanding obligations resulting from the Young Plan. Article 2(e) provided that the amounts due in respect of the various issues of the 1930 loan were to be payable only in the currency of the country in which the issue was made. The basis for calculating the amount of currency payable was to be the amount in U. S. dollars to which the payment due in a currency would have been equivalent at the rates of exchange ruling when the loan was issued. The nominal amount of U. S. dollars arrived at in this way was to be reconverted into a currency of payment at the rate of exchange on August 1, 1952. Conversion Bonds were issued for an amount in the original currency arrived at in this way. The nominal value of the American issue was the same in Extension Bonds as the nominal value of the bonds issued in 1930. Bonds issued originally with a nominal value of DM 1000 became Conversion Bonds with a nominal value of DM 1000.50.

The second part of Article 2(e) is the protective clause on which agreement was reached for the benefit of Young Loan bondholders, and which became the subject of controversy. The text in the three languages is as follows:

Should the rates of exchange ruling any of the currencies of issue on 1st August, 1952, alter thereafter by 5 per cent, or more, the instalments due after that date, while still being made in the currency of the country of issue, shall be calculated on the basis of the least depreciated currency (in relation to the rate of exchange current on 1st August, 1952) reconverted into the currency of issue at the rate of exchange current when the payment in question becomes due.

Au cas où les taux de change en vigueur le 1er août 1952 entre deux ou plusieurs monnaies d’émission subiraient par la suite une modification égale ou supérieure à 5%, les versements exigibles après cette date, tout en continuant à être effectués dans la monnaie du pays d’émission, seront calculés sur la base de la devise la moins dépréciée par rapport au taux de change en vigueur au ler août 1952, puis reconvertis dans la monnaie d’émission sur la base du taux de change en vigueur lors de l’échéance du paiement.

Sollte sich der am 1. August 1952 für eine der Emissionswàhrungen massgebende Wechselkurs später um 5 v.H. oder mehr ândern, so sind die nach diesem Zeitpunkt fälligen Raten zwar nach wie vor in der Wàhrung des Emissionslandes zu leisten; sie sind jedoch auf der Grundlage der Währung mit der geringsten Abwertung (im Verhältnis zu dem Wechselkurs vom 1. August 1952) zu berechnen und zu dem im Zeitpunkt der Fälligkeit der betreffenden Zahlung massgebenden Wechselkurs wieder in die Emissionswährung umzurechnen.

changes in external value of currencies

The majority of the tribunal noted that there was no “agreed rate” for the French franc from January 26, 1948 to December 29, 1958. This reference was to the period that began with the adoption of an “unauthorized change” of par value, to use the language of Article IV, Section 6 of the Fund’s original Articles, 4 and terminated with the establishment of a par value in which the Fund concurred. The initial par value of the French franc established under the Articles was 0.00746113 gram of fine gold per franc or 0.839583 U.S. cent per franc; the corresponding figures for the new par value were 0.00180000 and 0.202550. In 1957, France adopted a system under which a 20 per cent surcharge was imposed on all outgoing payments and a 20 per cent premium was paid on all incoming payments. The Fund regarded this system as equivalent in economic terms to a devaluation. 5

The protective clause was applied to the changes in external value of the French franc in both 1957 and 1958. The U. S. dollar was treated on both occasions as “the least depreciated currency,” and the new external value of the French franc on each occasion was taken to exceed the test of 5 per cent in Article 2(e) of Annex I, Section A of the LDA. The BIS, as Trustee for the bondholders, pointed out that the implication of applying the clause was that the expression “the least depreciated currency” covered a currency that had not depreciated. It appeared to the BIS that the expression was not quite adequate. 6

The first revaluation of the deutsche mark took place in 1961. The majority cited the following exchange rates and percentage changes between that currency and the U. S. dollar:

article image

The exchange value of one deutsche mark had increased by 1.1905 U. S. cents, or 5.000105 per cent of the exchange rate on August 1, 1952; the price of one U. S. dollar had been reduced by DM 0.20, or 4.761905 per cent of the exchange rate on that date. The majority referred to the “exchange rates” on the two dates, but the numbers represent the parity between the two currencies that resulted from the par values that were established for them under the Fund’s Articles. Parities as a concept did not correspond to exchange rates, because rates were free to fluctuate provided that they did not move outside margins around parity that were consistent with the Articles. 7 The references by the majority to “exchange rates” were dictated no doubt by the terminology of Article 2(e) of Annex I, Section A of the LDA, but under Article 13 (a) of the LDA an exchange rate for the purposes of the LDA meant, with some exceptions not relevant here, par values “in force … as agreed” with the Fund on the relevant date.

It followed from the values mentioned above that a change of 5 per cent or more from the “exchange rates” ruling on August 1, 1952 had occurred if the change in the parity between the two currencies was expressed as an “appreciation” of the deutsche mark against the U. S. dollar but not as a “depreciation” of the U. S. dollar against the deutsche mark. The majority did not discriminate between this terminology and the terminology of “revaluation” and “devaluation.”

The BIS, on May 27, 1961, notified the Federal Debt Administration (FDA), the appropriate agency of the Federal Republic, that payments subsequent to the revaluation of the deutsche mark had to be adjusted because all currencies of issue had depreciated against the deutsche mark and because the test of the minimum of 5 per cent had been met in 1957 and had become irrelevant thereafter. The FDA rejected this claim on the basis of its understanding of the protective clause. The BIS informed the five governments that were the applicants in this proceeding of the FDA’s reaction and recommended recourse to the arbitral tribunal that, under the LDA, had exclusive jurisdiction in all disputes regarding the interpretation or application of the LDA.

The deutsche mark was revalued again in 1969. The majority described this action as a change of 12.587 per cent in the 1952 rate of exchange for the U. S. dollar. 8 The German Government again refused to adjust payments. Negotiation did not dispose of the difference of opinion, and the five applicant governments instituted proceedings against the German Government in May 1971.

principal contentions of the parties

The main arguments of the applicants were, first, that Article 2(e) of Annex I, Section A of the LDA applied not only to the depreciation but also to the appreciation of a currency of issue. The English and French texts referred to “depreciation” and “dépréciation which, unlike “devaluation” and “dévaluation” comprised the loss of a currency’s value even if it was not the result of a governmental act. An appreciation of a currency automatically meant the depreciation of all other currencies in this broader sense.

Second, the object, purpose, and history of the LDA supported this analysis because the LDA was intended to achieve a fair settlement of German indebtedness in circumstances in which settlement had been made possible only because private creditors and their governments had waived a substantial part of their claims. The respondent’s obligation to permit all creditors to have the benefit of an appreciation of the deutsche mark followed from the prohibition of discrimination among classes of creditors that was imposed by Article 8 of the LDA and was a fundamental principle of the LDA. If this interpretation was not adopted, bondholders entitled to receive payment in deutsche mark would receive more favorable treatment on a revaluation of that currency because they would not receive less than the principal amount as calculated on the basis of exchange rates ruling on August 1, 1952.

The applicants asked the tribunal to adjudge that the revaluations of the deutsche mark gave rise to the application of the protective clause and entitled the holders of each issue made in a non-German currency to have payments falling due after a revaluation made on the basis of the rate of exchange between the deutsche mark and the currency of issue on the due date.

The respondent moved to have the tribunal reject the applicants’ submissions on the ground that the LDA should be interpreted according to the circumstances at the time when the LDA was negotiated. At that time, no participant thought of an appreciation of the deutsche mark. The clause was intended only as protection against devaluation of a currency of payment (i.e., a reduction in value by governmental action) as was demonstrated by the language of the German text. The revaluation of the deutsche mark had not led to a change under the Articles in the par values of the currencies of the creditor countries, and therefore an increase in the amounts due in those currencies was not justified according to Article 13 of the LDA. The respondent, in arguing that the protective clause did not apply on the occasion of the appreciation of a currency, conceded that the expression “the least depreciated currency” applied to a currency that had neither appreciated nor depreciated as well as to a currency that had depreciated.

The respondent also argued that the condition of a change of 5 per cent or more in the disputed clause was not satisfied for all time by the depreciation of a single currency on a single occasion. The clause was applicable each time that a currency of issue depreciated by 5 per cent or more, compared with its position on August 1, 1952, or when, as a result of several depreciations of less than 5 per cent each, a cumulative depreciation of 5 per cent or more was reached, compared with the position on August 1, 1952. The tribunal did not rule on this question because its conclusion on the meaning of “the least depreciated currency” made a ruling unnecessary.

majority decision

The majority 9 approached the problems of interpretation by invoking provisions of the Vienna Convention on the Law of Treaties of May 23, 1969. The Convention did not apply in the proceedings, but the view was widespread that the Convention, at least in regard to interpretation, was a codification of existing customary law. The majority relied mainly on Article 31 (1) of the Convention: “A treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose.” The decisive words to be interpreted were “Abwertung” “depreciation,” and “dépréciation.”

The majority declared that it had to give equal effect to all three texts. The ordinary and everyday sense of the crucial terms might lead to one interpretation of “Abwertung” and a different interpretation of “depreciation” and “dépréciation.” Although “Abwertung” in proper technical language meant a reduction by governmental act in the external value of a currency in relation to a yardstick such as gold, some uncertainty was introduced by the tendency to describe this phenomenon as “formelle Abwertung” in everyday usage. The English and French words, however, were used normally to describe a loss of value however it came about, while the result of governmental action was usually termed “devaluation” or “dévaluation.” Even if “depreciation” and “dépréciation” could be distinguished from “devaluation” and “dévaluation” in this way in the two languages, they were also used interchangeably in both everyday and technical language. The International Monetary Fund, 1945-1965: Twenty Years of International Monetary Cooperation, which was published by the Fund, was among the sources cited for this indiscriminate practice. The majority concluded that the possibility of different meanings for the three texts eould not be ruled out, which meant that the wording of the disputed clause offered no clear guidance in the attempt to arrive at the intended meaning and to give concordance to the interpretation of the texts.

No clearer guidance could be obtained by concentrating on the normal meaning of the terms at the time that the LDA was concluded. The words “depreciation” and “dépréciation” were often used interchangeably with “devaluation” and “dévaluation” even when the par value system was still in operation in accordance with the Fund’s Articles. The fungibility of these terms could be demonstrated by reference to Article I (iii) of the original Articles, which referred to “competitive exchange depreciation.” In the system of par values agreed with the Fund, this concept could refer only to dévaluation.

Textual interpretation could not resolve the vagueness of the terms in the English and French texts and the possible discrepancy between them and the German version, and it was necessary, therefore, to examine them “in their context” as required by Article 31 (1) of the Vienna Convention. “Context” meant the language of the disputed clause as a whole and the LDA as a whole. The applicants argued that the word “alter” (“ändern” “subir une modification”) in relation to exchange rates in the first part of the disputed clause was appropriate for both a rise and a fall in an exchange rate, with the consequence that the clause would apply if a currency were revalued. The applicants argued from this premise that the word “depreciated” (and its equivalents) in the latter part of the clause must be given the larger meaning. The majority did not accept the logic of this argument: the word “alter” did not necessarily control the meaning of “depreciated.”

The applicants contested the respondent’s argument that the word “depreciation” and its equivalents meant only devaluation on the ground that this construction could lead to inequitable results. If a single currency were revalued and all others were equally devalued, or if only one currency were devalued and all others remained unchanged or were revalued, no upward adjustment could be made because there would be no “least depreciated currency” to serve as a standard for payment in a devalued currency. The majority rejected this argument on the ground that the basis for adjustment in these cases would be the nondevalued currency before its revaluation. A nondevalued currency could be considered as “least depreciated” even though a revalued currency could not. The position would be unreasonable if payment in a devalued currency could not be adjusted because another currency had not been devalued but instead had been revalued. 10

The majority, in considering the LDA as a whole, gave special weight to Article 13 because that provision would have to apply if an adjustment had to be made under the disputed clause. There was a direct link, therefore, between the two provisions. The text of Article 13 is as follows:

Wherever it is provided in the present Agreement and the Annexes thereto that an amount shall be calculated on the basis of a rate of exchange, such rate shall, except in the cases provided for in Annex III and in Article 8 of Annex IV of the present Agreement, be___

  • (a) determined by the par values of the currencies concerned in force on the appropriate date as agreed with the International Monetary Fund under Article IV, Section 1, of the Articles of Agreement of the International Monetary Fund; or

  • (b) if no such par values are or were in force on the appropriate date, the rate of exchange agreed for current payments in a bilateral payments agreement between the Governments concerned or their monetary authorities; or

  • (c) if neither par values nor rates in bilateral payments agreements are or were in force on the appropriate date, the middle rate of exchange generally applicable for transactions ruling for cable transfers in the currency of the country in which payment is to be made in the principal exchange market of the other country on that date, or on the last date before that date on which such rate was ruling; or

  • (d) if there is or was no rate of exchange as specified under (a), (b) or (c) at the appropriate date, the cross-rate of exchange resulting from the middle rates of exchange ruling for the currencies in question in the principal exchange market of a third country dealing in those currencies on that date or the last date before the said date upon which such rates were ruling.

Under this provision, the majority declared, par values “in force … as agreed” with the Fund had precedence in the determination of exchange rates. In 1952, all the parties concerned, except Switzerland, were members of the Fund, and all of these members, except France, had par values in force as agreed with the Fund. The importance of Article 13, the majority held, was that by means of it the LDA in matters of monetary law was expressly fitted into the structure of the Bretton Woods system. The essential framework of this system was the par value agreed between a member and the Fund and the bonding of exchange rates to the par value.

A consequence of the system was the unacceptability of the applicants’ argument that the revaluation of one currency automatically meant a depreciation of all currencies not simultaneously revalued. If the deutsche mark was revalued, the purchaser of this currency for Belgian francs had to spend more Belgian francs than would have been necessary before the revaluation, but this result was not the consequence of a devaluation of the Belgian franc in terms of gold or the U. S. dollar of fixed gold content under the Articles. The relationship of the Belgian franc to currencies other than the deutsche mark remained unchanged. The value of the Belgian franc in terms of them would be reduced only if it were devalued. There was no depreciation of the Belgian franc in the sense in which the disputed clause used this term within the framework of monetary law as it stood when the LDA was negotiated.

The majority attached importance also to the fact that in 1961 and 1969, when the revaluations of the deutsche mark occurred, all of the applicants, except Switzerland, had par values that came within the terms of Article 13(a) of the LDA. The par values of members other than the Federal Republic of Germany had not been changed by the revaluations and would have to be applied under Article 13(a) in any necessary determinations under that provision.

The applicants had questioned recourse to Article 13 as an aid to interpretation, arguing that it was a purely technical provision and that there were different German expressions in the texts of Article 13 and the disputed clause. Article 13 referred to “Umrechnungskurs” (“conversion rate”) while the disputed clause referred to “Wechselkurs” (“exchange rate”). The latter expression, they argued, usually described the relation between or among currencies. The exchange rate for a currency could change without action by the government of the issuer. The majority rejected this argument, in part because there was no similar discrepancy in the English and French texts, which used “rate of exchange” and “taux de change” in both provisions, and in part because a study of the German text of Article 13 as a whole showed that “Umrechnungskurs” was used in that context to mean exchange rate.

To support the view that the disputed clause had nothing to do with gold or the U. S. dollar of fixed gold content as the common denominator of par values under the Articles, the applicants relied on the English case Lively Ltd. and Another v. City of Munich. 11 That case has been discussed in the thirteenth installment in this series of The Fund Agreement in the Courts. 12 A loan to the City of Munich in 1928 fell within the ambit of the LDA, under which the maturity date became December 1, 1973 instead of December 1, 1953. The issue in the case was whether the amount payable at the later date was to be calculated according to Article 13(a) or Article 13(c) of the LDA. Par values were still in existence under the law of the Fund, but the par value system had collapsed. The court preferred Article 13(c) because of the collapse. The majority held that the case did not support the applicants because the situation in 1973 was different from the situations in 1961 and 1969 with which the tribunal was concerned. The case did support the respondent’s view, however, that the word “depreciated” in the disputed clause could be interpreted by reference to Article 13(a) in circumstances in which the relevant facts antedated the time at which it became appropriate to apply Article 13(c).

The applicants argued that the prohibition of discrimination among classes of creditors under Article 8 of the LDA was part of the context that had to be considered in the interpretative process. The principle of equal treatment would be violated if other classes of creditors under the Young Loan did not receive as much value as the holders of the German issue after the revaluations of the deutsche mark. The tribunal agreed there was inequality, but it was countenanced by Article 8, which provided that differential treatment was not proscribed if it was the result of a “settlement in accordance with the provisions of the present Agreement and the Annexes thereto.” The inequality in this case followed from Article 13 and the disputed clause. Moreover, the prohibition of Article 8 ensured only pari passu treatment of bondholders and other creditors in the servicing of loans, as provided for by the General Bond, and not equality in the value of redemption payments. 13 The prohibition meant that other creditors were not to rank ahead of bondholders in the receipt of repayment if the debtor was unable to repay both classes fully and uniformly at the same time.

The majority, pursuing its approach under Article 31(1) of the Vienna Convention, turned to the “object and purpose” of the LDA. The object was the settlement of German external debts, and the purpose a fair compromise, in order that the German economy could recover and the plan of settlement could be honored. This end could be achieved only if foreign creditors were prepared to waive a substantial part of their claims and refrain from placing an intolerable burden on the German economy. Problems of interpretation had to be resolved in this spirit and from the standpoint of the circumstances that existed in 1952. Neither the spirit nor the circumstances could be ignored because of the strength and speed of the economic recovery of the Federal Republic. The majority did not accept the applicants’ argument that a broad interpretation was appropriate because of the applicants’ concessions and the continuing strength of the German recovery.

The majority concluded, finally, that the approach it had taken in accordance with Article 31(1) of the Vienna Convention resolved the problem of reconciling the three texts of the LDA. Even if the view were accepted that there was an irreconcilable discrepancy between the German text and the other two texts, the German text and the strict meaning of “Abwertung” would prevail because of the principle of Article 33(4) of the Vienna Convention. According to that principle, when several authentic texts cannot be reconciled and a discrepancy remains, the meaning to be preferred is the one that “best reconciles the texts, having regard to the object and purpose of the treaty.” The German text most closely approached the object and purpose of the LDA.

The majority, who began by referring to the word ‘ ‘Abwertung” as “relatively clear” and as affected by “at least, some uncertainty,” ended by describing the word as having “quite unequivocal meaning.”

dissenting opinion

The minority also relied for guidance on Articles 31 and 33 of the Vienna Convention. The opinion begins by emphasizing that the object and purpose of the LDA was a fair and equitable settlement for debtors and creditors. This intention was expressed in Article 8 of the LDA, which prohibited “any discrimination or preferential treatment among the different categories of debts or as regards the currencies in which debts are to be paid or in any other respect.” Any provision of the LDA that required interpretation and did not come within the exception to the general principle of Article 8 had to be interpreted to give effect to this principle. It would not accord with the object and purpose of the LDA to give the disputed clause an interpretation under which the tranche issued in German currency and tranches issued in other currencies would not receive equality of treatment.

The minority’s understanding of the text of the disputed clause, confirmed by its reading of the travaux préparatoires, was that a fixed pattern of relationships was established among all the currencies of issue on the basis of exchange rates as of August 1, 1952, with the intention that each bondholder when paid would receive equality of treatment in terms of U. S. dollar value as of August 1, 1952. The disputed clause referred to exchange rates, which meant the relationships among currencies, and not relationships to some external and fixed standard. The clause applied therefore to both rises and falls in a rate of exchange. If the exchange rate for one currency rose, the exchange rates for other currencies fell. The drafters had intended to replace a gold clause with a general or multiple currency exchange guarantee against changes in the exchange rates of all currencies of issue and not simply protection against the devaluation (“Abwertung”) of a currency of issue in which payment was to be made. The guarantee came into play whenever an alteration of 5 per cent or more occurred in the exchange rate, whether that alteration was the result of devaluation or revaluation or some other cause of depreciation or appreciation. The term “the least depreciated currency” should not be understood as a technical expression that restricted the ordinary and broad meaning of the opening part of the disputed clause (“alter,” “ändern” “subiraient . . . une modification”). The search for the least depreciated currency arose as a basis for calculation only after it had been determined that an alteration had brought the guarantee into play.

The fallacy in the respondent’s argument, the minority held, was its concentration on gold as the standard and on each currency separately in relation to it. The argument was based on the assumption that the guarantee under the LDA gave no more protection than was given by the original gold clause, namely, protection against the formal devaluation of the currency of issue and payment. This assumption was incorrect because the guarantee was intended to give bondholders the benefit of the most favorable currency among all the currencies of issue. A multiple currency exchange guarantee of this kind was normal, and it was not surprising that the way it operated on the revaluation of one of the currencies was not spelled out when the guarantee was adopted.

The majority, recalling that the Vienna Convention permitted recourse to travaux préparatoires to confirm an interpretation arrived at with the aid of the rules of Article 31, examined the preparatory work of the LDA. The majority and the minority took different views on the positive support they found for their conclusions in the travaux préparatoires.

A similar difference of opinion appears in the reaction to the practice that had been followed after the entry into force of the LDA. Under the Vienna Convention, interpretation must take account of subsequent agreements among contracting parties on interpreting the treaty and of subsequent practice in applying it, from which a consensus on interpretation might be inferred. The majority examined statements by the parties to the LDA in the period between the signing of it in 1952 and the first revaluation of the deutsche mark in 1961 but found them too contradictory to provide any assistance. The minority attached importance to what happened after the de facto alteration of the exchange rate for the French franc in 1957. The alteration was taken into account when subsequent payments were made in French francs, even though France had made no change in the official rate of the franc in relation to other currencies or in the par value of its currency under the Fund’s Articles. This action was an unequivocal acceptance of the view that “Abwertung” did not have the strict meaning of devaluation as contended by the respondent. Moreover, the adjustment was calculated by reference to the U. S. dollar, which was accepted as “the least depreciated currency” although it had not depreciated and on the contrary had appreciated against the franc in the exchange markets. The depreciation of all currencies against the deutsche mark in 1961 and 1969 corresponded to the depreciation of the French franc against the U. S. dollar and all other currencies of issue in 1957. The deutsche mark in 1961 and 1969 corresponded to the U. S. dollar in 1957 as a nondepreciated currency that, just as cogently as the dollar, qualified as “the least depreciated currency.”

The minority rejected the respondent’s assertion that bondholders entitled to payment in other currencies had not been deprived of any value by revaluation of the deutsche mark. The Bundesbank itself had reassessed its reserves in foreign currency and shown a loss after the two revaluations. What was true for the Bundesbank was true for the bondholders of the non-German tranches, and even more obviously for those who resided inside the Federal Republic.

The minority held that, if reconciliation of the texts of treaties written in more than one language was not possible except by preferring one of them, then equal authenticity was not a bar to an order of precedence among them on specific points of interpretation. Nor was equal authenticity a bar in these circumstances to recognition of the superiority of the language in which negotiations were conducted or an original version from which the other versions were translated. Only the English version was employed during the negotiation of the LDA, while the other two versions were translations from the English. The negotiators did not participate in the process of translation. It should be possible to show that a translation does not correctly reflect the meaning of an original text. The affairs of sovereign states cannot be influenced by the fortuitous choice of words selected by a nameless translator. The minority, by means of this reasoning, was led to favor the word “depreciated” in the English text and its broad meaning in that language.

The contrast in technique followed by the majority and the minority is as strong here as elsewhere. The majority refused to apply a number of interpretative approaches that were pressed upon them as valid and appropriate. Among them were various approaches that would favor the German text: the lowest common denominator of obligation, the clearest text, the interpretation least favorable to the parties that had primary responsibility for offering the text. Similarly, the majority rejected the approach, followed by the minority, of establishing an order of precedence among texts. This approach would have given primary weight to the English and French texts. To all of these approaches, the majority said nay because they would not necessarily have been compatible with the intention of the negotiators and would not have been directed toward fulfillment of the object and purpose of the treaty.

The minority did not agree that the alteration of a rate of exchange in Article 13(a) meant an alteration of par value under the Fund’s Articles. “Rate of exchange” was not defined in the LDA, and therefore it could not be assumed that the expression was used in an abnormal sense. Each of the paragraphs of Article 13 was based on the same ordinary meaning of “exchange rate.” They dealt with different situations of fact. Paragraph (a) did not give the concept a special meaning in the light of which all paragraphs had to be interpreted. Nor did the minority accept the argument that, because of the reference in Article 13(a) of the LDA to Article IV, Section 1 of the Fund’s Articles, or for some other reason, the provisions of the LDA were subordinated to the provisions of the Fund’s Articles. These reactions of the minority led it to reject the conclusion that “depreciated” meant “devalued” and to hold that a revalued currency could be “the least depreciated currency.”

language of the lda

A major problem in the case was created by the word “depreciated” in the LDA. The confusion with which the words “depreciation,” “appreciation,” “devaluation,” and “revaluation” have been and are used is widespread in both official and nonofficial publications. The first pair of words are regarded sometimes as equivalent to the second pair, sometimes as wholly different from them, and sometimes as more extensive. These three versions appear in the opinions of the tribunal and the written arguments, but in them a preferred version is not always maintained with consistency. Discrete meanings for the four words would be desirable, but it should not be assumed that the definition of them is a simple matter. There would be complications, for example, in the treatment to be accorded to governmental action. A case discussed in an earlier installment in this series illustrates the problems that can arise with a definition of “devaluation” in terms of “governmental action.” 14

The majority concluded that Article I (iii) of the Fund’s Articles, in referring to “competitive exchange depreciation,” was referring only to devaluation (i.e., a change in par value by a member). This argument cannot be sustained. Article IV, Section 6 of the original Articles dealt with unauthorized changes of par values—which by hypothesis were not agreed with the Fund—and Article IV, Section 8 dealt with those cases in which a member’s currency depreciated in the exchange markets in breach of the member’s obligations. Furthermore, a member might adopt a new fixed rate of exchange for its currency without calling it a par value or attempting to establish it as such under the Articles. This action also would be in breach of the member’s obligations if it had already established a par value under the Articles. The purpose of the Fund expressed in Article I (iii) would have been frustrated by competitive changes in these circumstances as effectively as it would have been by competitive changes of par value. It should be noted that Article I (iii) remains unchanged by the Second Amendment even though a par value system is not now in operation. Competitive depreciation is possible, therefore, even in the absence of a par value system.

The majority may have assumed that Article I (iii) would not refer to actions inconsistent with the Articles, from which the majority may have inferred that the provision must be taken to refer to governmental acts to change par values. But a currency might have been “floating” in the days of the par value system without being inconsistent with the Articles. This was the legal position before the Fund called on a member to establish a par value under a Membership Resolution of the Board of Governors. The Fund would have regarded a competitive depreciation in these circumstances as inconsistent with the Articles. To prevent such a situation from arising, the Fund included a term in Membership Resolutions under which, in the period between the acceptance of membership and the establishment of an initial par value, a member was required not to change the exchange rate for its currency without consulting the Fund in advance of a change and without obtaining its agreement. 15

Perhaps, because of the provisions referred to above, Article IV, Section 4(a) of the original Articles, in giving effect to Article I (iii) in the form of an obligation of members, required them to avoid “competitive exchange alterations.” The language of the Articles does not confirm the proposition that in 1952 “depreciation” in the Articles meant “devaluation.” The only proposition that can be derived from the Articles is either that the applicants’ understanding was correct or that precision was lacking.

Neither the majority nor the minority discussed Article IV, Section 8(b) of the original Articles, in which the reduction in a par value is expressly distinguished from a depreciation. That provision would have cast doubt on the majority’s conclusion that “depreciation” in Article I (iii) had to mean “devaluation.” Similarly, the provision would not have supported the conclusion of the minority that “depreciation” covered any reduction in the value of a currency, whether or not it resulted from a change in par value.

A discussion of Article IV, Section 8 was relevant to another aspect of the case. The respondent argued that the French action of 1957 was a “formal devaluation” and that it confirmed the view that Article I (iii) meant devaluation in referring to depreciation. On October 2, 1957, however, the Fund decided that France was required to adjust the Fund’s holdings of French francs under Article IV, Section 8(b)(ii) as a result of the action France had taken. The decision meant that the adjustment was necessary not because of a reduction in par value under Section 8(b)(i) but because, as described in Article IV, Section 8(b)(ii), the currency had “depreciated to a significant extent within that member’s territories.”

The expression “the least depreciated currency” was interpreted by the majority to include within its reach a currency that had neither depreciated nor appreciated. An appreciated currency is even further removed from depreciation than an undepreciated currency. It is not surprising, however, that there should be such a strong division of opinion about the resonance of the expression. A more surprising conclusion of the majority was the view that in certain circumstances, adduced by the applicants to illustrate the unreasonableness of the respondent’s argument, the least depreciated currency could be taken to be a revalued currency at its value before revaluation. The relationship between two currencies on this basis would correspond to nothing in the real world when payments were made after the revaluation. In the Lively case, 16 a rate of exchange that the defendants argued should be applied under Article 13(a) of the LDA was rejected by the court on the ground that it had no relation to any market or commercial rate of exchange.

The minority could have made more of the drafting of Article 13 of the LDA. The majority and the minority differed on what was to be drawn from the reference to par values in Article 13(a). The majority gave predominant importance to this paragraph and held that the reference to par values showed an intention to fit the LDA into the Bretton Woods system and its basic relationship to gold. The minority rejected this analysis and gave predominant importance to paragraphs (b), (c), and (d) because they did not refer to par values. These paragraphs could not have referred to par values, because they dealt with situations in which par values were not agreed with the Fund or were not in force. Nothing was made of the argument, however, that gold was the substratum of the par value system and that even in these situations it would have been possible to determine the foreign exchange value of currencies in terms of gold. It is true that the minority pointed out that paragraphs (b), (c), and (d) had no link to “gold or the gold dollar.” It attributed the absence of that link to the unacceptability of a gold clause in the negotiations leading to the LDA, but the explanation is not persuasive in view of paragraph (a). The argument could have been reinforced by demonstrating that even paragraphs (b), (c), and (d) could have been fitted into the Articles, and that the negotiators, in refraining from following this course, had shown that they were not adapting their solution to the monetary law of the Articles.

The drafters of the LDA could have taken cognizance of Article IV, Section 8 of the original Articles of Agreement if such a solution had been acceptable to them. Under that provision, the “gold value” of the Fund’s assets had to be maintained by adjusting the Fund’s holdings of a member’s currency in accordance with changes in the par value or the foreign exchange value of the currency. If the Fund found that the foreign exchange value of a member’s currency had depreciated to a significant extent within the member’s territories, which meant that the gold value of the currency had been reduced, the member had to pay more currency to the Fund in an amount “equal to the reduction in the gold value” of its currency held by the Fund. The provision did not mention appreciation, but the Fund interpreted the provision to require it to return to the member an amount of currency equivalent to the appreciation. As noted already, “depreciation” and “appreciation” under Article IV, Section 8 were phenomena distinguishable from changes in par values.

Article IV, Section 8 was both appropriate and necessary for the Fund. It was appropriate because the Fund’s functions included its jurisdiction over par values and exchange rates, from which the conclusion was drawn that the Fund should not make profits or suffer losses as the result of changes in par values or exchange rates. The provision was necessary because, if the Fund’s holdings of currencies were not adjusted in accordance with changes in gold value, the Fund could avoid profits and losses by standing ready to conduct its operations and transactions on the basis of unreal (i.e., unadjusted) exchange rates. In those circumstances, however, the Fund would face a run on an appreciated currency, and it would be unable to dispose of a depreciated currency. If, alternatively, the Fund chose to conduct its activities on the basis of current (i.e., adjusted) exchange rates without the adjustment of its holdings, it could not avoid profits or losses. In short, Article IV, Section 8 was necessary for the effectiveness of the Fund’s financial activities. In this respect, it could be regarded as analogous to Article 13 of the LDA because that provision also was intended to ensure effectiveness, in this case the protection of bondholders. It should be noted, furthermore, that the adjustment of the Fund’s holdings of a currency under Article IV, Section 8 did not depend on the legality of the conditions that made adjustment necessary.

Not until 1954, when the possible floating of sterling was in the air, did the Fund decide how it would apply Article IV, Section 8 to what were then called “fluctuating currencies.” The expression meant that exchange transactions in a currency were not conducted within the legal limits around parities (i.e., around the ratios resulting from the par value of that currency and the par values of other currencies). It was obvious how the gold value of a fluctuating currency would have to be determined in a system based on gold in which the United States was the only member of the Fund that had undertaken to maintain the value of its currency in relation to gold by means of purchases and sales of gold for U. S. dollars with the monetary authorities of other members of the Fund. The foreign exchange (gold) value of a member’s fluctuating currency would be determined by the exchange rates between that currency and the U.S. dollar in the member’s main financial center. What was meant by the exchange rates for the purpose of the decision was spelled out in detail by the decision. 17

The Fund’s solution for the application of Article IV, Section 8 could have been anticipated by the negotiators of the LDA had they intended to base all paragraphs of Article 13 of the LDA on gold value as the substratum of the Bretton Woods system, but they did not. If in 1952 the negotiators had wanted such a solution, nothing would have been simpler than a provision under which the protection of bondholders would be based on the relationship of the currencies of issue and payment to the U. S. dollar, with some caveat for the situation in which there would be a change in the par value of the U. S. dollar. Such a solution could have covered all cases, whether or not a par value agreed with the Fund was in force. This solution was proposed, with the understanding that there would be new negotiations if the par value of the U. S. dollar were changed, but the proposal was not acceptable to the United States because it conflicted with U.S. policy on the gold clause. It is unnecessary to dwell on the discrepancies between paragraphs (c) and (d) of Article 13 of the LDA, which were based on exchange rates in the market, and the Fund’s 1954 decision on fluctuating currencies. Paragraph (b) alone is sufficient to show how far Article 13 as a whole was from the solution of gold value. The exchange rate for a currency specified by a bilateral payments agreement might have been inconsistent with the exchange rate between the currency and the U. S. dollar. The specified exchange rate would have been a multiple currency practice and discriminatory currency arrangement and, even worse, might not have received the approval of the Fund. 18 The implication of Paragraph 13 is that the negotiators of the LDA had not intended to fit that agreement into international monetary law as established by the Articles of the Fund.

The majority and minority had different reactions to the circumstances that, when the LDA was negotiated, France did not enforce a par value agreed with the Fund and Switzerland was not (and still is not) a member of the Fund. The majority regarded these circumstances as exceptional and not as proof that the LDA had not been, or could not be, fitted into the monetary law of the Bretton Woods system. The minority held that the circumstances of these two countries disposed of any argument that the negotiators of the LDA had intended to fit it into the framework of the Bretton Woods system. On this point also the minority could have strengthened its opinion by showing that the currencies of even these two countries would not have been beyond the reach of a protective clause based on the gold value of the currencies. In the case of the French franc, the Fund was required to determine the gold value of that currency under Article IV, Section 8. In the case of the Swiss franc, the exchange rate between it and the U. S. dollar in the Swiss exchange market could have been considered the gold value of the Swiss franc by analogy to Article IV, Section 8.

some further reflections on the two opinions

The narrow margin of difference in the size of the majority and the minority is evidence of the difficulty of the basic issues. The reader of the two opinions must be impressed by the quality of each analysis and the wealth of cited authority. Is it possible to venture a statement of some underlying attitudes that guided the two groups to their different conclusions? The different responses to the cases of France and Switzerland have been mentioned already, but some broader differences can be discerned.

First among these differences is the influence attributed to a system of international monetary law. The majority regarded the Articles of the Fund as such a system and was disposed to hold that the negotiators of the LDA intended to work within that system. 19 The minority did not share this disposition.

Second, the two groups probably held different views on what was the main objective of the LDA. The majority seems to have given predominant weight to the recovery of the German economy as the main objective. Therefore, they assumed that the negotiators of the LDA had wanted to avoid obligations that would overburden the economy. The minority gave greater weight to the protection of bondholders as the main objective of the LDA.

Third, the two groups might have had different attitudes to the terms of the settlement compared with the original gold clause. The majority seems to have assumed an intention to provide comparable protection for bondholders, and it found that there was comparability between the gold clause and the par value system of the Articles. The minority attributed less importance to comparability between the old and the new terms and was willing to assume that agreement had been reached on a better form of protection than had been represented by the gold clause.

A final word is necessary. The case dealt only with the effect of the revaluations of the deutsche mark in 1961 and 1969 on the application of the protective clause. The case did not deal with the effects of the breakdown of the par value system after August 15, 1971 or with the consequences of the Second Amendment of the Articles. 20


Some Economic Aspects of Tax Administration—richard goode (pages 249-74)

The paper discusses the application of economic analysis and statistics to several questions of tax administration and legislation bearing on it. The principle of equalizing marginal administrative costs and marginal tax revenue is suggestive, but persuasive arguments can be made for spending more or less than it indicates. The marginal principle, nevertheless, is valuable as a guide for allocating a fixed appropriation that is too small to cover all remunerative activities. Human capital theory corroborates the productivity of staff training but implies that different financing arrangements may be appropriate for general and specialized training. Economic studies can help in establishing criteria for identifying cases for detailed examination or audit and in making administrative or alternative assessments where accounts are inadequate. A lump-sum, first-year depreciation allowance equal to the discounted value of normal allowances would have merit as a simple method of adjusting for inflation. Proposals for interlocking, self-enforcing tax systems and for market-enforced self-assessment of taxable property are judged to be unpromising. Suggestions for economizing on administrative costs by imposing heavy monetary penalties for noncompliance are subject to theoretical objections and appear to be inconsistent with widely accepted beliefs about justice. The concept of economic externality has some general relevance for tax administration but few direct applications. Tax administrators and economists could benefit from acquaintance with the attitudes and methods of each other’s discipline.

A Revised Version of the Multilateral Exchange Rate Model—jacques r. artus and anne kenny mcguirk (pages 275-309)

This paper gives a description of the current version of the Multilateral Exchange Rate Model (MERM), a model first presented in the November 1973 issue of Staff Papers. The purpose of the model—the estimation of the medium-term (two- to three-year) effects of changes in the exchange rates of the various industrial countries on their trade balances—has not changed since it was first presented in 1973, but numerous technical improvements have since been made in the specification of its structure. Simulations of exchange rate effects are presented for each of the industrial countries on the basis of the new model.

The MERM is a static general equilibrium model of the Walrasian type, simplified to a large extent by introducing input-output relationships into the demand and supply equations. In the revised model, there are 18 industrial countries and 2 regional groups of countries. The commodity classification is somewhat more disaggregated than in the 1973 version, so that the model now includes six goods (shown with the Standard International Trade Classification (SITC) number): agricultural commodities (SITCs 0-1), raw materials (SITCs 2 and 4), mineral fuels (SITC 3), semifinished manufactures (SITCs 5-6), finished manufactures (SITCs 7-9), and nontraded commodities. Each good satisfies both intermediate demand and final demand, and detailed data on the input-output structure of each country are taken into account in the respecification of the system of demand equations.

The specification of the supply equations has remained unchanged from the previous version; however, the shifts in the supply schedules, which represent changes in the costs of production induced by the effects of exchange rate changes on the prices of material inputs and the returns to primary factors, are fully endogenous and take into account the whole input-output structure of each country. The model is closed, as was done before, by assuming that the level of total real output of each country is kept constant.

There are three major assumptions upon which the model is based. The first—that costs and prices are somewhat “inflexible,” at least over the medium-term horizon considered—implies that a change in the exchange rate may facilitate establishment of a relative price structure that will restore external equilibrium. The second—that industrial countries produce differentiated goods that face finite price elasticities of demand in world markets—implies that the major relative price effects of exchange rate changes are the reallocations of demand among similar, but differentiated, goods produced by various countries. The third assumption—that the level of nominal final demand can be influenced by the central authorities—is the basis for the proposition that the central authorities use monetary and fiscal policies to offset whatever effect an exchange rate change may have on the overall level of economic activity in real terms.

In the analytical framework discussed here, the exchange rate changes of interest are those that far exceed, or fall far short of, the concurrent inflation rate differentials. These may occur because of discrete exchange rate adjustments among countries maintaining pegged exchange rates, or because of disturbances in asset markets in countries with floating exchange rates. Analogously, it is possible to analyze the trade balance effects of divergent inflationary tendencies among countries maintaining pegged nominal exchange rates.

An Analysis of Factors Influencing the Level of SDR Holdings in Non-Oil Developing CountriesRobert g. murphy and george m. von fursten-berg(pages 310-37)

During the period 1973-78, no SDR allocations were made. This period therefore presents researchers with an opportunity to study how countries adjust their SDR holdings in the absence of supply shifts caused by new allocations. Non-oil developing countries are particularly useful to study, because most of them have been consistent net users of SDRs and have been able to initiate transfers of SDRs, while net acquirers or the designated recipients of SDRs have been more passive. After showing that non-oil developing countries generally did not reduce their SDR holdings to the lowest possible level permitted under reconstitution limits, this paper identifies some economic factors that may influence the level of SDR holdings in relation to cumulative allocations.

We found an indication that a rise in the SDR interest rate relative to the combined market rate used in the SDR interest rate basket encouraged a greater degree of retention of SDRs by non-oil developing countries as a group toward the end of the past decade. Since the SDR interest rate was fixed at approximately 60 per cent of the combined market rate from the middle of 1974 through 1978, the influence of other factors could be examined separately over this period through the pooling of annual data for 27 major non-oil developing countries. This analysis showed that higher burdens from external debt, adjusted for differences in the normal rates of countries’ economic growth, do not have a statistically significant effect on SDR holdings, measured as percentages of allocations. A portfolio balance factor, the positive effect of the ratio of non-gold reserves to imports, is statistically significant in most regressions, with SDR holdings tending to rise somewhat less than in proportion to reserves.

The Equivalence of Product Tax Changes and Public Enterprise Price Changes— Robert h. floyd (pages 338-74)

This paper analyzes, in a variety of partial and general equilibrium settings, the effects on both resource mobilization and allocation of (a) an increase in nonfinancial public enterprises’ prices and (b) the application of a tax to certain transactions and/or enterprises. It is often assumed that if nonfinancial public enterprises are operated efficiently, pricing policies and practices that would result in the generation of operating surpluses (or losses) may be regarded as equivalent in some senses to taxation (or subsidization) measures. However, when the concept of equivalence is defined to include the levels of government and public sector revenues as well as real economic variables, such as production and consumption levels, this study finds that an excise or product tax is often not equivalent to an increase in the price of a public enterprise’s output. From the viewpoint of raising revenues for the government’s budgetary purposes, a tax appears to be more reliable and more predictable than a price increase. However, if the alternative revenues were comprised of a dividend equal to the incremental profits resulting from a price increase, there would be circumstances in which a price increase might generate more revenues for the government and for the public sector as a whole. These results suggest that the objectives of public policy should be clearly defined and taken into account when determining whether a tax or a price change is the preferred policy instrument.

Testing the Impact of Value-Added and Global Income Tax Reforms on Korean Tax Incidence in 1976. An Input-Output and Sensitivity Analysispeter s. heller (pages 375-410)

This study estimates the incidence of each tax in the Korean tax system in 1976. Input-output analysis is used to trace the incidence of taxes levied on intermediate inputs. On balance, the Korean tax system appears progressive for nonfarm households and regressive for farm households and thus narrows the after-tax income differential between the two sectors. The global income tax, the gift and inheritance tax, and the farmland tax are the principal progressive elements of the tax system; indirect taxes—such as the value-added tax (VAT), the special consumption tax, and customs duties—are the regressive elements. Some reduction in the degree of inequality in the before-tax income distribution has also been achieved through the tax system.

The principal impact of the two major tax reforms—the introduction of a global income tax and the value-added tax—has been to increase the relative tax burden of the nonfarm sector. The increase in the nonfarm tax burden principally affected households in the lower-income deciles and in the highest decile. The shift from a schedular to a global income tax lowered the tax burden of all but the richest households in the nonfarm sector. The simplification of the indirect tax system with the VAT reform increased the burden of nonfarm households, principally lower-income ones.

The results suggest that to ensure the accuracy of the incidence results, one must take account of the detailed structure of any tax in terms of its rates and exemptions. In cases where a tax has been deliberately formulated to achieve distributional objectives, the use of a general tax allocation criterion may yield incidence results that are misleading in their policy implications.

The Fund Agreement in the Courts—XVI—Joseph gold(pages 411-36)

The Arbitral Tribunal for the Agreement on German External Debts has delivered a decision, by four votes to three, in a case in which the Governments of Belgium, France, Switzerland, the United Kindgom, and the United States were applicants and the Government of the Federal Republic of Germany the respondent. The issue was whether the expression “the least depreciated currency” in the London Agreement on German External Debts, 1953, as the standard for calculating payments, applied to the deutsche mark when it was revalued in 1961 and 1969. The majority concluded that the expression could not be understood in this way. A major reason was the assumption that the negotiators of the Agreement had intended to fit it into the international monetary system as it existed in those years. The expression had to be understood, therefore, in the light of the par value system. Under it, the appreciation of a currency in terms of gold as the common denominator did not [mean] result in the depreciation of all other currencies against the common denominator. The minority concluded that there had been no intention to fit the Agreement into the Fund’s Articles, and the appreciation of one currency did [mean] result in the depreciation of all other currencies, so that it was possible to hold that an appreciated currency was the least depreciated.


Quelques aspects économiques de l’administration fiscalerichard goode (pages 249-74)

Le texte porte sur l’application de l’analyse et de la statistique économiques à diverses questions d’administration et de législation fiscales. Le principe de l’égalisation des coûts administratifs marginaux et des recettes fiscales marginales est stimulant, mais on peut invoquer des arguments convaincants pour dépenser plus ou moins qu’il ne l’indique. Le principe marginal est cependant valable comme guide pour la répartition d’un crédit budgétaire fixe trop modique pour couvrir toutes les activités rémunératrices. La théorie du capital humain corrobore la productivité de la formation des cadres, mais elle implique une différenciation probable des dispositifs de financement quant à la formation générale et quant à la formation spécialisée. L’analyse économique peut aider à l’établissement de critères permettant d’identifier les cas qui nécessitent un inventaire détaillé ou une vérification sur place; elle peut faciliter aux autorités un jugement optimum ou une évaluation administrative avec variantes pour les cas d’absence ou de vice de données. Un forfait pour amortissement de première année, d’un montant égal à la valeur escomptée des provisions normales, servirait bien de méthode simple d’ajustement eu égard à l’inflation. On ne voit pas de promesse d’avenir aux propositions tendant à l’interconnexion de systèmes fiscaux autonomes, ni à l’évaluation par le contribuable de sa fortune imposable selon les cours du marché. Les suggestions d’économies réalisables sur les frais administratifs par l’imposition de lourdes amendes en cas d’infraction prêtent le flanc à des objections théoriques et ne semblent pas cadrer avec les principes de justice généralement acceptés. La notion d’extériorité économique intéresse quelque peu l’administration fiscale, sur un plan général, mais ne comporte que de rares applications directes. Les administrateurs fiscaux et les économistes gagneraient peut-être à se familiariser mutuellement avec les attitudes et les méthodes propres à leurs disciplines respectives.

Une version révisée du modèle multilatéral de taux de change— Jacques r. artus et anne kenny mcguirk (pages 275-309)

Ce document donne une description de la version courante du modèle multilatéral de taux de change (MMTC), qui a été présentée pour la première fois dans le numéro de novembre 1973 de la série Staff Papers. L’objet du modèle—l’évaluation des effets à moyen terme (deux à trois ans) des variations des taux de change sur la balance commerciale des pays industriels—n’a pas changé depuis qu’il a été présenté pour la première fois en 1973, mais on a apporté par la suite de nombreuses améliorations techniques à la spécification de sa structure. Les simulations des effets exercés par le taux de change sont présentées pour chacun des pays industriels sur la base du modèle révisé.

Le MMTC est un modèle d’équilibre général statique de type walrarien, simplifié, dans une large mesure, par l’insertion des relations entrées-sorties dans les équations de la demande et de l’offre. Dans le modèle révisé, il y a 18 pays industriels et 2 groupes régionaux de pays. Le classement par produits est un peu plus désagrégé que dans la version de 1973: le modèle comprend maintenant six rubriques (présentées avec le numéro de la Classification type pour le commerce international—CTCI): produits agricoles (CTCI 0-1), matières brutes (CTCI 2 et 4), combustibles minéraux (CTCI 3), articles manufacturés semi-finis (CTCI 5-6), articles manufacturés finis (CTCI 7-9) et produits non échangés. Chaque produit satisfait la demande intermédiaire et la demande finale, et, dans la respécification, du système d’équations de la demande, on tient compte des données détaillées sur la structure des relations entrées-sorties de chaque pays.

La spécification des équations de l’offre est restée la même que dans la version antérieure; mais les déplacements des courbes d’offre, qui correspondent aux modifications des coûts de production induites par les effets de variations du taux de change sur les prix des inputs matériels et les revenus des facteurs primaires, sont pleinement endogènes et tiennent compte de la totalité de la structure des relations entrées-sorties de chaque pays. Le modèle est clos, comme antérieurement, par l’hypothèse que le niveau de la production réelle totale de chaque pays est maintenu constant.

Le modèle est basé sur trois principaux postulats. Le premier—que les coûts et les prix sont quelque peu «rigides», au moins dans l’horizon du moyen terme considéré—implique qu’une modification du taux de change peut faciliter la mise en place d’une structure des prix relatifs qui rétablira l’équilibre extérieur. Le deuxième—que les pays industriels produisent des articles différenciés qui sont soumis à des élasticités de prix limitées de la demande mondiale—implique que les principaux effets de prix relatifs des variations de taux de change sont une reventilation de la demande entre produits de même nature, mais non identiques, fabriqués par les divers pays. Le troisième postulat—que les autorités centrales peuvent influer sur le niveau de la demande finale nominale—sert de base à la thèse selon laquelle les autorités centrales utilisent les politiques monétaire et budgétaire pour neutraliser tout effet qu’une variation du taux de change peut exercer sur le niveau global de l’activité économique en valeur réelle.

Dans le présent cadre analytique, les variations de taux de change présentant un intérêt sont celles qui dépassent de beaucoup les différentiels de taux d’inflation concomitants, ou leur sont très inférieures. Cela peut se produire en raison des ajustements discrets des taux de change entre pays qui multintiennent des taux de change liés, ou à cause de perturbations sur les marchés des actifs financiers dans les pays à taux de change flottants. Par analogie, on peut analyser les effets, sur la balance commerciale, des tendances inflationnistes divergentes parmi les pays qui maintiennent des taux de change nominaux liés.

Analyse des facteurs qui ont une influence sur le niveau des avoirs en DTS des pays en développement non exportateurs de pétrolerobert g. murphy et

george m. von furstenberg (pages 310-37)

Pendant la période 1973-78, aucune allocation de DTS n’a eu lieu. Cette période offre donc aux chercheurs la possibilité d’étudier la manière dont les pays ajustent leurs avoirs en DTS en l’absence de variations de l’offre sous la forme de nouvelles allocations. Il est particulièrement utile à cet égard d’étudier le cas des pays en développement non exportateurs de pétrole, car la majorité de ces pays sont, régulièrement, utilisateurs nets de DTS, et sont en mesure de procéder à des transferts de DTS, alors que les acquéreurs nets ou les bénéficiaires désignés jouent un rôle plus passif. Après avoir montré que les pays en développement non exportateurs de pétrole, en général, n’ont pas réduit leurs avoirs en DTS au niveau le plus bas que permettaient les limites de reconstitution, ce document identifie certains des facteurs économiques qui peuvent avoir une influence sur le niveau des avoirs en DTS relativement aux allocations cumulatives.

D’après les constatations faites, il semble que, vers la fin de la dernière décennie, toute augmentation du taux d’intérêt du DTS par rapport au taux combiné du marché utilisé dans le panier servant à la détermination du taux d’intérêt du DTS a eu tendance à encourager ces pays à détenir davantage de DTS. Comme le taux d’intérêt du DTS a été fixé à environ 60 % du taux combiné du marché du milieu de 1974 à la fin de 1978, l’influence des autres éléments a pu être examinée séparément pendant cette période par voie du regroupement de statistiques annuelles concernant les 27 pays. Cette analyse a démontré que l’alourdissement de la charge constituée par la dette extérieure, ajustée en fonction des différences entre les taux normaux de croissance économique des pays en question, n’a pas d’effet statistiquement significatif sur les avoirs en DTS, mesurés en pourcentage des allocations. Facteur relatif à l’équilibre des portefeuilles, l’effet positif du ratio des réserves autres que d’or aux importations est statistiquement significatif dans la plupart des régressions, les avoirs en DTS ayant tendance à ne pas augmenter tout-à-fait proportionnellement aux réserves.

L’équivalence entre la modification des taxes sur les produits et le relèvement des prix des entreprises publiquesrobert h. floyd (pages 338-74).

L’auteur du présent document analyse dans diverses situations d’équilibre partiel ou général les effets sur la mobilisation et la répartition des ressources a) d’une hausse des prix des entreprises publiques non financières et b) de l’application d’une taxe à certaines transactions et/ou à certaines entreprises. On présume fréquemment que, dans la mesure où des entreprises publiques non financières sont bien gérées, les règles et pratiques servant à fixer les prix et qui pourraient être à l’origine d’excédents (ou de déficits) d’exploitation peuvent être considérées à certains égards comme équivalant à des mesures fiscales (ou à des subventions). Toutefois, lorsque la notion d’équivalence englobe les recettes des administrations et du secteur public, ainsi que certaines variables économiques réelles (niveaux de production et de consommation, par exemple), la présente étude constate qu’une taxe d’accise ou une taxe sur les produits n’est souvent pas équivalente à un relèvement du prix de la production d’une entreprise publique. S’agit-il de se procurer des recettes à des fins budgétaires, la taxe est, semble-t-il, un moyen plus sûr, dont les conséquences sont plus faciles à prévoir, que la majoration des prix. Toutefois, si les autres recettes consistaient en un dividende égal aux bénéfices supplémentaires imputables à une hausse de prix, cette hausse pourrait, dans certains cas, procurer davantage de recettes à l’Etat et au secteur public dans son ensemble. Ces résultats donnent à penser que les pouvoirs publics doivent définir clairement leurs objectifs et les prendre en considération avant de décider s’il est préférable d’intervenir par une modification des impôts ou des prix.

Détermination de l’effet des réformes fiscales (impôt général sur le revenu et taxe à la valeur ajoutée) sur la pression fiscale en Corée en 1976: analyse entrées-sorties et analyse de sensibilitépeter s. heller (pages 375-410)

Dans son étude, M. Heller estime l’incidence de chacun des impôts dans le système fiscal de la Corée en 1976. Il détermine, en procédant à une analyse entrées-sorties, l’incidence des impôts prélevés au niveau de la consommation intermédiaire. Dans l’ensemble, le système fiscal de la Corée est apparemment régressif pour les ménages agricoles et progressif pour les autres, et permet ainsi de réduire l’écart entre les revenus nets des deux secteurs. L’impôt général sur le revenu, l’impôt sur les donations, les droits de succession et la contribution foncière agricole sont les principaux éléments progressifs du système fiscal; les contributions indirectes (taxe à la valeur ajoutée, taxe spéciale de consommation et droits de douane, par exemple) en sont les éléments régressifs. Le système fiscal permet, en outre, de réduire dans une certaine mesure les inégalités qui accompagnent la répartition des revenus avant impôt.

Ces deux grandes réformes fiscales—adoption d’un impôt général sur le revenu et de la taxe à la valeur ajoutée—ont eu essentiellement pour effet d’accroître le fardeau fiscal relatif sur le secteur non agricole, principalement sur les ménages dont le revenu se situe dans le premier et le dernier décile. Le passage d’un système cédulaire à un système général d’impôt sur le revenu a réduit le fardeau fiscal, sauf pour les ménages les plus riches du secteur non agricole. La simplification du système des contributions indirectes par l’adoption de la TVA accroît le fardeau fiscal des ménages non agricoles, et surtout de ceux dont le revenu est faible.

L’étude donne à.penser que, pour obtenir exactement l’incidence recherchée, il est indispensable de tenir compte de la structure détaillée (taux et exemptions) des impôts. Dans les cas où un impôt est délibérément formulé à des fins de répartition, l’emploi d’un critère en matière de répartition de l’impôt risque de donner des résultats dont les implications peuvent induire en erreur quant aux mesures qu’il convient d’adopter.

Les statuts du Fonds devant les tribunaux—XVI—joseph gold (pages 411-36)

Le tribunal d’arbitrage de l’accord sur les dettes extérieures allemandes a rendu, à la majorité de quatre membres contre trois, une décision dans une affaire où les demandeurs étaient les gouvernements de la Belgique, des Etats-Unis d’Amérique, de la France, du Royaume-Uni et de la Suisse, et le défendeur, le gouvernement de la République fédérale d’Allemagne. Le tribunal était saisi de la question de savoir si la formule “la devise la moins dépréciée” qui est retenue, dans l’Accord de Londres de 1953 sur les dettes extérieures allemandes, comme base de calcul des montants dus, s’appliquait au deutsche mark lorsque celui-ci a été réévalué en 1961 et en 1969. La majorité des membres du tribunal a conclu que la formule ne pourrait pas être entendue de cette façon. Une des principales raisons qui motivent cette conclusion est l’hypothèse selon laquelle les négociateurs de l’Accord avaient eu l’intention de placer celui-ci dans le cadre du système monétaire international tel qu’il existait à l’époque. La formule devrait donc être comprise à la lumière du système des parités fixes. Dans ce système, l’appréciation d’une monnaie par rapport à l’or, pris comme commun dénominateur, n’entraînait pas la dépréciation de toutes les autres monnaies au regard du commun dénominateur. La minorité a estimé que les négociateurs n’avaient pas eu l’intention de faire entrer l’Accord dans le cadre tracé par les statuts du Fonds et que l’appréciation d’une monnaie entraînait la dépréciation de toutes les autres monnaies, de sorte qu’il était possible de soutenir qu’une monnaie qui s’était appréciée constituait la devise la moins dépréciée.


Algunos aspectos económicos de la administración de impuestosrichard goode (páginas 249-74)

En este trabajo se examina la aplicación del análisis económico y la estadística a varias cuestiones de administración de impuestos y la legislación con ella relacionada. Resulta interesante el principio de la igualación del gasto marginal de administración y del ingreso fiscal marginal, pero pueden aducirse argumentos persuasivos a favor de un gasto mayor o menor que el indicado. No obstante, el principio marginal es valioso como orientación para distribuir una asignación fija que resulte demasiado pequeña para abarcar todas las actividades remunerativas. La teoría del capital humano corrobora la productividad de las actividades de adiestramiento de los empleados pero implica que pueden necesitarse distintos sistemas de financiamiento para el adiestramiento general y el especializado. Los estudios económicos coadyuvan a formular criterios con los que identificar casos para ser examinados detalladamente o sometidos a auditoría y para efectuar evaluaciones administrativas o de otro género cuando sean deficientes las cuentas. Sería interesante una asignación por depreciación del primer año a tanto alzado igual al valor actualizado de las asignaciones normales, como método sencillo de ajuste por inflación. Se juzgan como poco prometedoras las propuestas de sistemas tributarios inter-conexos de cumplimiento propio y las de autodeterminación del valor imponible de la propiedad con cumplimiento por medio del mercado. Se presentan objeciones teóricas a algunas sugerencias para reducir gastos de administración imponiendo fuertes sanciones monetarias por falta de cumplimiento, pues parecen ser incongruentes con las creencias ampliamente aceptadas sobre lo que es justo. El concepto de externalidad económica tiene cierta pertinencia general para la administración tributaria, pero pocas aplicaciones directas. Los administradores tributarios y los economistas saldrían ganando si adquirieran un conocimiento mutuo de las actitudes y métodos de sus disciplinas respectivas.

Versión revisada del modelo multilateral de tipos de cambio—jacques r. artus y anne kenny mcguirk (páginas 275-309)

En este trabajo se describe la versión que actualmente se emplea del modelo multilateral de tipos de cambio (MMTC), presentado por primera vez en el número de noviembre de 1973 de Staff Papers. Su finalidad de estimar los efectos a medio plazo (de dos a tres años) de las variaciones de los tipos de cambio de varios países industriales en sus balanzas comerciales no ha variado desde que se presentó el modelo en 1973, pero desde entonces se han hecho numerosas mejoras técnicas en la especificación de su estructura. En este estudio se presentan simulaciones de los efectos de los tipos de cambio en cada uno de los países industriales tomando como base el nuevo modelo.

El MMTC es un modelo estático de equilibrio general de tipo walrasiano, en gran medida simplificado mediante la introducción de relaciones de insumo-producto en las ecuaciones de demanda y oferta. En el modelo revisado se han incluido 18 países industriales y dos grupos regionales. La clasificación de productos es algo más desagregada que la de la versión de 1973, distinguiendo el modelo actual seis clases de bienes (que aparecen con el número correspondiente de la Clasificación Uniforme para el Comercio Internacional (CUCI)): productos agrícolas (CUCI 0-1), materias primas (CUCI 2 y 4), combustibles minerales (CUCI 3), manufacturas semiterminadas (CUCI 5-6), manufacturas terminadas (CUCI 7-9) y productos que no se comercian. Cada una de estas clases de bienes satisface tanto la demanda intermedia como la demanda final, y en la nueva especificación del sistema de ecuaciones de demanda se tienen en cuenta datos detallados de los cuadros de insumo-producto de cada país.

La especificación de las ecuaciones de oferta no ha variado respecto a la de la versión anterior; sin embargo, los desplazamientos de la curva de oferta que representan variaciones de los costos de producción provocados por los efectos de las variaciones de los tipos de cambio en los precios de insumos materiales y en la retribución de los factores primarios son completamente endógenos y reflejan la estructura de insumo-producto de cada país. Se ha construido un modelo cerrado, al igual que el anterior, suponiendo que se mantiene constante el nivel del producto real total de cada país.

El modelo se basa en tres supuestos importantes. El primero—que los costos y precios son algo “inflexibles”, por lo menos durante el plazo medio considerado—implica que toda variación del tipo de cambio puede facilitar la formación de una estructura de precios relativos que restablezca el equilibrio externo. El segundo—que los países industriales producen distintos bienes que tropiezan en el mercado mundial con una elasticidad finita de demanda en relación con el precio–implica que los principales efectos en cuanto a precios relativos, causados por variaciones de los tipos de cambio, son los de redistribución de la demanda de bienes semejantes, pero distintos, producidos por los diversos países. El tercer supuesto—que las autoridades centrales pueden influir en el nivel de la demanda final nominal—constituye la base de la proposición de que las autoridades centrales emplean las políticas monetaria y fiscal para contrarrestar cualquier efecto que una modificación del tipo de cambio pueda tener en el nivel global de la actividad económica en términos reales.

En el marco analítico que se utiliza en este estudio, las variaciones de los tipos de cambio que nos interesan son las que quedan muy por encima, o muy por debajo, de las diferencias entre las tasas de inflación simultáneas. Tales discrepancias pueden ocurrir debido a ajustes discretos de los tipos de cambio entre países que mantienen tipos vinculados, o debido a perturbaciones en los mercados de activos en países con tipo de cambio flotante. Del mismo modo, se pueden analizar los efectos en la balanza comercial de tendencias inflacionarias divergentes entre países que mantienen tipos de cambio nominales vinculados.

Análisis de los factores que influyen en el nivel de tenencias de DEG de los países en desarrollo no petrolerosrobert g. murphy y george m. von furstenberg (pagínas 310-37)

En el período de 1973 a 1978 no se hicieron asignaciones de derechos especiales de giro (DEG). Por consiguiente, este período brinda a los investigadores la oportunidad de estudiar de qué modo ajustan los países sus tenencias de DEG cuando no hay variaciones de la oferta originadas por nuevas asignaciones. Los países en desarrollo no petroleros presentan un caso cuyo estudio resulta especialmente útil, porque la mayoría de ellos han sido con regularidad usuarios netos de DEG y han podido iniciar transferencias, en tanto que los adquirentes netos o receptores designados de derechos especiales de giro han sido más pasivos. Tras demostrar que los países en desarrollo no petroleros no redujeron en general sus tenencias de DEG al nivel mínimo posible permitido conforme a los límites de reconstitución, en este estudio se señalan algunos factores económicos que pueden influir en el nivel de tenencias de DEG en relación con las asignaciones acumulativas.

Hay indicios de que el aumento del tipo de interés del DEG en relación con el tipo de mercado combinado que se ha utilizado en la cesta de valoración del tipo de interés del DEG contribuyó a un mayor grado de retención de DEG por parte de los países en desarrollo no petroleros como grupo hacia fines de la última década. Dado que el tipo de interés del DEG se mantuvo fijo alrededor del 60 por ciento del tipo combinado de mercado desde mediados de 1974 hasta 1978, durante este período pudo examinarse por separado la influencia de otros factores mediante la recopilación de datos anuales relativos a 27 países en desarrollo no petroleros importantes. Ese análisis demostró que la carga más elevada de la deuda externa, ajustada según las diferencias de las tasas normales de crecimiento económico de los paises, no tiene un efecto estadísticamente significativo en las tenencias de DEG, medidas como porcentajes de las asignaciones. En la mayoría de las regresiones hay un factor de equilibrio de cartera, el efecto positivo de la razón entre las reservas no constituidas por oro y las importaciones, que es estadísticamente significativo, tendiendo las tenencias de DEG a elevarse algo menos que en proporción a las reservas.

Equivalencia de las modificaciones de los impuestos sobre bienes y las modificaciones de los precios de empresas públicas— robert h. floyd (páginas 338-74)

En este trabajo se analizan, en diversas situaciones de equilibrio parcial y general, los efectos en la movilización y en la asignación de recursos de a) una elevación de los precios de las empresas públicas no financieras y b) la aplicación de un impuesto a ciertas transacciones o empresas. A menudo se supone que si se administran con eficacia las empresas públicas no financieras, las políticas y medidas de precios que produzcan superávit (o pérdidas) de explotación pueden considerarse equivalentes en cierto sentido a medidas tributarias (o de subvención). Sin embargo, en este estudio se llega a la conclusión de que, cuando el concepto de equivalencia se define de modo que incluya los niveles de ingreso del gobierno y del sector público, así como también variables económicas reales—por ejemplo, los niveles de producción y consumo—un impuesto sobre bienes frecuentemente no es equivalente a una elevación del precio del producto de una empresa pública. Desde el punto de vista de la obtención de los ingresos fiscales para fines presupuestarios, parece que un impuesto es más fiable y previsible que una elevación del precio. Sin embargo, si los ingresos fiscales alternativos consistieran en un dividendo de magnitud igual al aumento de utilidades obtenido mediante una elevación del precio, existirían circunstancias en las que una elevación del precio produciría más ingresos para el gobierno y para el sector público en conjunto. Estos resultados indican que los poderes públicos deben definir claramente sus politicas en estas materias y tenerlas en cuenta cuando se trata de determinar si el instrumento de política preferible es una modificación de los impuestos o una modificación del precio.

Comprobación del efecto que las reformas tributarias del impuesto al valor agregado y el impuesto general sobre la renta tuvieron en la incidencia fiscal de Corea en 1976: Análisis de insumo-producto y de sensibilidadpeter s. heller (páginas 375-410)

En este estudio se estima la incidencia que tuvo cada impuesto en el sistema tributario de Corea en el año 1976. Se utiliza el anàlisis de insumo-producto para determinar la incidencia de los impuestos que gravan los insumos intermedios. En general, el sistema tributario de Corea parece progresivo en lo que respecta a las unidades familiares no dedicadas a la explotación agrícola y regresivo en el caso de las unidades familiares dedicadas a la misma y, por consiguiente, reduce la diferencia entre los ingresos, una vez deducidos los impuestos, de dichos sectores. El impuesto general sobre la renta, el impuesto sobre las donaciones y sucesiones y el impuesto sobre la renta agrícola son los principales elementos progresivos del sistema tributario; los impuestos indirectos—como, por ejemplo, el impuesto al valor agregado (IVA), el impuesto especial al consumo y los derechos de aduana—constituyen los elementos regresivos. Asimismo mediante el sistema tributario se ha logrado reducir en cierto grado la desigualdad de distribución de los ingresos antes de deducidos los impuestos.

El principal efecto que produjeron las dos reformas tributarias màs importantes—la implantación de un impuesto general sobre la renta y el impuesto al valor agregado—fue el de aumentar la carga tributaria relativa al sector no agrícola. Este aumento afectó principalmente a las familias encuadradas en los deciles de menor ingreso y en el decil más elevado. El remplazo del impuesto celular por el impuesto general sobre la renta hizo disminuir la carga tributaria salvo para las familias màs ricas del sector no agrícola. La simplificación del sistema tributario indirecto con la reforma del IVA aumentó la carga de las familias no dedicadas a la explotación agrícola, en especial las de menor ingreso.

Las conclusiones señalan que, para garantizar la exactitud de los resultados sobre la incidencia, debe tenerse en cuenta con detalle la estructura del impuesto en cuestión, en lo que se refiere a sus tasas y exenciones. Cuando el impuesto se ha concebido deliberadamente para lograr objetivos de distribución, la aplicación de un criterio de asignación tributaria de carácter general puede dar resultados de incidencia equívocos con respecto a sus implicaciones para la política.

El Convenio del Fondo ante los tribunales—XVI— joseph gold (páginas 411—36)

El Tribunal Arbitral para el Acuerdo sobre la Deuda Externa Alemana ha dictado un laudo, por cuatro votos contra tres, en un caso en el cual los Gobiernos de Bélgica, Francia, Estados Unidos, el Reino Unido y Suiza fueron demandantes, y el Gobierno de la República Federal de Alemania, demandado. Se trataba de dirimir si la expresión “la moneda menos depreciada” que aparece en el Acuerdo de Londres sobre la Deuda Externa Alemana, 1953, como la norma para calcular los pagos, era aplicable al marco.alemán cuando esté fue revaluado en 1961 y 1969. La mayoría concluyó que la expresiôn no podia ser entendida así. Una de las principales razones aducidas se basó en el supuesto de que los negociadores del Acuerdo habían querido que esté se conformara al sistema monetario internacional de aquellos años. La expresiôn, por consiguiente, ténia que ser entendida a la luz del sistema de paridades. De acuerdo con este sistema, el resultado de la apreciación de una moneda en relaciôn con el oro como denominador común no era la depreciaciôn de todas las demâs monedas frente al denominador común. La minoria, en cambio, concluyô que no habîa habido intenciôn de conformar el Acuerdo al Convenio del Fondo, y que el resultado de la apreciaciôn de una moneda si era la depreciaciôn de todas las demás monedas, de modo que se podía sostener que una moneda apreciada era la menos depreciada.

In statistical matter (except in the résumés and resúmenes) throughout this issue,

  • Dots (…) indicate that data are not available;

  • A dash (—) indicates that the figure is zero or less than half the final digit shown, or that the item does not exist;

  • A single dot (.) indicates decimals;

  • A comma (,) separates thousands and millions;

  • “Billion” means a thousand million;

  • A short dash (-) is used between years or months (e.g., 1977-79 or January-October) to indicate a total of the years or months inclusive of the beginning and ending years or months;

  • A stroke (/) is used between years (e.g., 1978/79) to indicate a fiscal year or a crop year;

Components of tables may not add to totals shown because of rounding.

International Monetary Fund, Washington, D.C. 20431 U.S.A.

Telephone number: 202 477 7000

Cable address: Interfund


No. 4. World Economic Outlook: A Survey by the Staff of the International Monetary Fund. June 1981.

This publication reviews recent developments in the world economy and examines its short-term prospects. Topics include the global perspectives for balance of payments adjustment and financing; the situations of industrial, oil exporting, and non-oil developing countries; and key policy issues. Appendices provide regional and country surveys, notes on the world oil situation and the situation in nonmember countries of Eastern Europe and the U.S.S.R., other supplementary notes, and statistical tables.

No. 5. Developments in Trade Policy in Industrial Countries, by S.J. Anjaria, Z. Iqbal, L. Perez, and W.S. Tseng.

No. 6. The Multilateral System of Payments: Keynes, Convertibility, and the International Monetary Fund’s Articles of Agreement, by Joseph Gold.

No. 7. International Capital Markets: Recent Developments and Short-Term Prospects, 1981, by a staff team headed by R.C. Williams.

No. 8. Tax Policy and Administration in Sub-Saharan Africa, by Carlos A. Aguirre, Peter S. Griffith, Vito Tanzi, and Zühtü Yücelik.

A package consisting of these five Occasional Papers (Nos. 4-8) is available for $20. The World Economic Outlook paper is available separately for $8. Special rate to university libraries, faculty, and students: package of five papers, $12; World Economic Outlook paper, $5. Advice on payment in other currencies will be given on request.

For information and to place orders, write to:

External Relations Department

Attention: Publications

International Monetary Fund

Washington, D.C. 20431 U.S.A


Sir Joseph Gold, Senior Consultant and formerly the General Counsel and Director of the Legal Department of the Fund, is a graduate of the Universities of London and Harvard. He is the author of numerous books, pamphlets, and essays on the Fund and on international and national monetary law.


The first seven articles in the series, together with another article, were issued in book form as The Fund Agreement in the Courts (Washington, 1962). The next four articles were issued as The Fund Agreement in the Courts: Parts VIII-XI (Washington, 1976). The twelfth, thirteenth, fourteenth, and fifteenth articles were published in Staff Papers, Vol. 24 (March 1977), pp. 193-231, Vol. 25 (June 1978), pp. 343-67, Vol. 26 (September 1979), pp. 583-611, and Vol. 27 (September 1980), pp. 601-24, respectively.


International Legal Materials, Vol. 19 (November 1980), pp. 1357-1408; Revue Générale de Droit International Public, Vol. 84 (1980), pp. 1157-1245.


United Nations Treaty Series, Vol. 333 (1959), pp. 4-449.


Joseph Gold, Légal and Institutional Aspects of the International Mone-tary System: Selected Essays (Washington, 1979), pp. 540-41, 554-56; Joseph Gold, “Unauthorized Changes of Par Value and Fluctuating Exchange Rates in the Bretton Woods System,” American Journal of International Law, Vol. 65 (1971), pp. 113-28.


International Monetary Fund, Annual Report, 1958 (Washington, 1958), p. 94.


Article 13(c) of the LDA was applied on both occasions because there did not exist at the relevant dates “a par value agreed” with the Fund or a rate of exchange under a bilatéral payments agreement between the United States and France.


Article IV, Sections 3 and 4 (original). The word original is used in these footnotes to denote the original (i.e., unamended) Articles.


The two changes in the par value of the deutsche mark and the effective dates of the changes were as follows:

article image


Issues involving the jurisdiction of the tribunal are not examined in this article.


The minority did not subscribe to the view that the respondent’s argument permitted the solution offered by the majority. The consequential unreason-ableness demonstrated that the respondent’s argument was wrong.


[1976] 3 All ER 851.


Staff Papers, Vol. 25 (June 1978), pp. 348-53.


The majority found further support in Article V(2)(b) of Annex II and Section B, Article 7 of Annex IV of the LDA for the conclusions it drew from the context.


Federal Maritime Commission v. AustralialU. S. Atlantic & GulfConference et al., 337 F. Supp. 1032 (1972); Joseph Gold, The Fund Agreement in the Courts: Parts VIII-XI (Washington, 1976), pp. 95-104. (Hereinafter referred to as Gold, Parts VIII-XI.) Note also that the Fund could decide to make uniform proportionate changes in the par values of the currencies of all members under Article IV, Section 7 (original) although a member could prevent the decision from applying to its currency.


Joseph Gold, Membership and Nonmembership in the International Monetary Fund: A Study in International Law and Organization (Washington, 1974), pp. 195-97.


[1976] 3 All ER 851.


For a full discussion of Article IV, Section 8 (original) and the Fund’s decision on computations involving fluctuating currencies, see Joseph Gold, Maintenance of the Gold Value of the Fund’s Assets, IMF Pamphlet Series, No. 6 (Washington, 2nd edition, 1971).


Article VIII, Section 3 (original).


Compare the discussion of the implications drawn from the conviction that there is or is not an international monetary system in Joseph Gold, The Rule of Law in the International Monetary Fund, IMF Pamphlet Series, No. 32 (Washington, 1980), pp. 13-17.


The BIS in its capacity as Trustee under the Young Loan issued an announcement dated May 30, 1980 (New York Times, Vol. 129 (May 30, 1980), p. D6) that included the following passage:

“It should be emphasized that this arbitral award only concerns the formal revaluations of the Deutsche mark in 1961 and 1969. The Arbitral Tribunal was not asked to rule on the more recent question, which has existed since 1971, of whether or not the exchange guarantee is applicable when currencies of issue of the Young Loan undergo important de facto alterations in their rates of exchange which does [sic] not result from formal changes in par values agreed with the International Monetary Fund.

“As a consequence of the floating of the exchange rates for the currencies of issue of the Young Loan, which commenced in May 1971 with the floating of the exchange rate for the Deutsche mark and has been the case for all the remaining currencies of issue for many years now, the Trustee notified the German Federal Debt Administration and the various paying agents that the reservations which the Trustee had already made regarding the rights of bondholders, and the precautionary steps to protect these rights being taken by the paying agents, extended to the new question of interpretation of the exchange guarantee to which floating exchange rates had given rise.

“If it is considered, in the light of the recent Arbitral Award, that the exchange guarantee does not apply in the event of the de facto appreciation in the exchange rate of a given currency of issue, it remains to be established whether the same is true when a given currency of issue suffers a de facto depreciation in exchange value.

“It should be recalled in this connection that the German Federal Debt Administration has applied the exchange guarantee, and made adjustments in the amounts payable to bondholders of the American, British, French and Swedish issues of the Young Loan, as from the due date for bonds and coupons of 1st June 1978. These adjustments resulted from the abolition of formal par values—which in fact had long since ceased to be respected in commercial transactions—to which the amendment of the articles of agreement of the International Monetary Fund on 1st April 1978 gave rise. The adjustments so made in the amounts payable to the bondholders in question were calculated on the basis of the Belgian franc, the currency of issue which, in the opinion of the German Federal Debt Administration, was, on each occasion, the “least depreciated currency” for the purposes of the exchange guarantee.

“No adjustments have, however, been made in respect of the preceding period of de facto floating exchange rates, which had commenced in 1971. In the view of the German Federal Debt Administration, a view not shared by the Trustee, par values remained in force for the purposes of the exchange guarantee until the above-mentioned amendment of the IMF articles of agreement.

“The Trustee has already drawn to the attention of the governments concerned the consequences for bondholders which result from floating exchange rates, and the further question of interpretation of the exchange guarantee to which this situation has given rise.

“In these circumstances, the Trustee has informed the German Federal Debt Administration that until the outstanding questions are settled, the Trustee’s earlier reservations of the possible rights of bondholders to additional payments remain in force irrespective of whether or not the final coupon and the bonds due for redemption are presented for payment on or after 1st June 1980. All the paying agents have been notified accordingly, and have been requested by the Trustee to maintain the protective measures which are necessary in the interests of bondholders. In addition, the Trustee may make a further approach to the governments concerned requesting them to take appropriate steps to reach the earliest possible solution of the outstanding questions to which floating exchange rates have given rise pursuant to the exchange guarantee.”