Operational Aspects
The European Council Resolution of December 5, 1978 established the framework for the European Monetary System and the Agreement of March 13, 1979 between the EC central banks provided the operating procedures. The operation of the EMS from the beginning has been characterized by two elements: it has been flexible in many of its technical aspects, and it has been run efficiently and smoothly by the participating central banks. The Central Bank Agreement was never intended to lay down rigid and unalterable rules for the system. The wording of the Agreement leaves room for flexibility and thus allows a gradual evolution of the system over time as experience is gained. In some important aspects of the system (such as currency of intervention, maturity of credits, or means for settlement of such credits), the provisions of the Agreement apply only “in principle,” leaving open the choice of mutually agreed ways to be used in handling new problems.
This high degree of flexibility can be explained by a number of factors. The EMS—as is true for many aspects of EC policies—is primarily an institution not so much of a technical but mainly of a political nature, where technical means are used toward political objectives. It may be seen as an important step on the way to a much more ambitious but also fairly distant goal: monetary union. The flexibility of the system was seen as necessary because of varying institutional settings and different approaches to many common problems among the participating central banks; it is workable because the central banks concerned, which are limited in number, have a long-standing tradition of close cooperation in foreign exchange matters, stemming, in particular, from the operation of the “snake.” The Committee of Governors of the central banks of the EC and its working parties meet regularly and frequently, normally in Basle, in connection with the monthly meetings of the Bank for International Settlements (BIS). All questions that arise in connection with the functioning of the system, be they of a more technical nature or touching upon more fundamental questions of external or domestic monetary policy, are discussed during these meetings. The four daily telephone consultations between all EC central banks on exchange rate matters usually provide, on the level of foreign exchange departments, an exchange of information on market developments and intervention activities and offer a readily available channel for consultation on higher levels as the need arises. (The central banks of Norway, Sweden, Switzerland, and the Federal Reserve Bank of New York, as well as, less directly, the central banks of Canada and Japan, are associated with this network of information.) The Monetary Committee of the EC regularly brings together high officials from central banks and finance ministries. The central bank governors also frequently attend meetings of economics and finance ministers of the EC member countries. The EC Commission is represented on the appropriate level on all the committees and at all the meetings which have been mentioned.
The following sections describe the evolution of some key operational features of the EMS. The survey is neither exhaustive, nor, in all respects, completely up-to-date because many aspects remain in flux and some details may not be known, due to their confidential nature.
Exchange Rate and Intervention Mechanism
The Central Bank Agreement stipulates that intervention be effected, in principle, in currencies of the participating central banks and be unlimited in amount at the obligatory intervention limits. This provision allows intervention in third currencies which normally takes place before the intervention limits are reached. It also allows intervention in participating currencies before the intervention limit is reached (“intramarginal intervention”), which is, however, subject to prior approval of the partner central bank in whose currency intervention is to take place.
In the first three years of the EMS, more than half of intervention by EMS participants has been effected in third currencies, predominantly in U.S. dollars. This is due to a number of reasons. At first, such intervention may have primarily, but neither necessarily nor exclusively, served the purpose of influencing the exchange rate relationship between the currency of the intervening central bank and the third currency in question. Secondly, a number of central banks prefer to intervene before their currency reaches its obligatory intervention limits because they want to avoid larger fluctuations, even within the band of the EMS. Or, the central banks may intervene because they consider it important to counter exchange rate movements at the beginning before they gather momentum; in this way, they may hope to achieve their exchange rate objectives with a smaller change in their reserve position. For such purposes, these central banks may use U.S. dollars for intervention instead of intramarginal intervention in participating currencies, since the latter requires the consent of the central bank whose currency is being used. The Italian authorities, in view of the wider fluctuation margins of 6 percent for the Italian lira, have consistently followed a policy of not allowing their currency to reach its intervention limits, and have consequently intervened in U.S. dollars (see Chart 2).
Of intervention in EMS currencies during the first three years, again more than half consisted of intramarginal intervention. Since the automatic financing provisions of the very short-term facility apply only to obligatory intervention at the limits, a central bank that would like to intervene intramarginally in another EMS currency would at first need to have a sufficient amount of the currency in question at its disposal. While this could be done by acquiring such currency on the market during earlier periods of strength, the Central Bank Agreement limits the holding of such currencies to working balances. Alternatively, the central bank would have to obtain the currency by agreement from the issuing central bank. Such agreements have at times been concluded when the interests of the two central banks could be reconciled, for instance, in the form of swap agreements that allowed a reversal of the transaction after a certain time. For the creditor central bank, such an agreement had the desirable effect of also canceling the liquidity-creating effect of the initial intervention. The choice of intervention currency is, of course, of interest with regard to its consequences on the overall creation of international liquidity and its impact on the exchange rates of the currencies involved.
Divergence Indicator
In the early phases of discussions about the EMS, there was a proposal that the exchange rate mechanism should be based not on a grid of bilateral parities—as it was in the “snake”—but directly on the ECU. Under such a system, for each currency a central rate in terms of the ECU would be declared, and fluctuation margins would be defined against the ECU, instead of against other currencies; countries would be obliged to keep their currencies within these margins. In the event, this idea was not accepted, mainly for two reasons. The first one is of a general nature. While such a system, requiring any central bank to intervene whose currency diverges by a given margin from its ECU central rate, would support efforts to achieve a higher degree of convergence, it would do so irrespective of the desired direction of convergence. Secondly, as a technical point, though with important policy implications, there would be the problem of determining the “partner” currency for intervention. Since, under such a system, frequently only one currency would reach its intervention point in terms of the ECU, the selected or designated intervention partner would become an “involuntary” creditor who would have to accept the creation of more liquidity in its own currency, or an “involuntary” debtor who would have to suffer the loss of reserves. In the end, the parity grid became the basis for the exchange rate and intervention mechanism, and the ECU-based divergence indicator became a supplementary device, functioning as a warning system. By inviting consultations and creating a “presumption” for corrective action, the indicator became a factor in promoting policy coordination.
Nevertheless, the role of the divergence indicator remains ambiguous. While being in line with the desire expressed in the Resolution of December 5, 1978, to balance the burden of both deficit and surplus countries, the indicator has a built-in tendency to promote convergence not necessarily toward monetary stability but rather toward some average level of monetary and price developments. This ambiguity may have been one reason why the divergence indicator appears not to have played the role it was expected to play. There are other reasons, some of them of a more technical nature. First, signals of the divergence indicator cannot automatically trigger action, but can only attempt to induce such action. Second, the very design of the indicator, based on a composite of all the currencies, causes it to respond only in situations where one currency is clearly divergent from the average of the other currencies. If two currencies would strongly move in opposite directions, it is most likely that neither currency would cross its divergence threshold. This explains also why, in certain cases, a currency may reach its intervention limit under the parity grid before reaching the divergence threshold, thus diminishing the ability of the divergence indicator to act as an “early warning system.” Lastly, the inclusion of the pound sterling, which does not participate in the exchange rate mechanism of the EMS, in the ECU as well as the application of wider margins for the Italian lira, have at times resulted in some distortions, in spite of the adjustments in the calculation of the divergence indicator eliminating the movements of these currencies beyond their notional 2.25 percent margins.24
During a review of the EMS conducted in September 1979, it was decided not to modify the functioning of the divergence indicator. It would have been possible to increase the likelihood of the indicator signaling a currency as being divergent before it would reach its bilateral intervention limits by lowering the divergence threshold significantly, for example, from 75 to 50 percent. However, this would have meant that the indicator’s warning bell would ring quite frequently—often without justification—causing the authorities to pay less attention to it.
Settlement of Intervention Debts
According to the Central Bank Agreement, the time for the settlement of claims and debts from obligatory intervention in participating currencies of debts under the very short–term financing facility is 45 days after the end of the month in which intervention took place. This can be extended automatically by three months, subject to certain limitations, and by another three months, subject to the agreement of the creditor central bank. Settlement is to be effected:
-
—in the first place by using holdings in the creditor currencies;
-
—subsequently in part or wholly in ECUs, with the proviso that a creditor central bank is not obliged to accept settlement in ECUs of an amount exceeding 50 percent of its claim;
-
—for the remaining balance in other reserve assets in accordance with the composition of the debtor central bank’s reserves, excluding gold.
The Agreement, however, explicitly leaves room for other forms of settlement as agreed between creditor and debtor central banks.
In practice, various methods have been used to settle claims and debts arising from intervention. The very short–term financing facility, which applies only to obligatory intervention, has not been used extensively, and the formal settlement procedures as outlined above have been applied only to a relatively small part of intervention debts. The larger part of debts and claims resulting from obligatory intervention have been offset against each other, or settled by the debtor central bank buying the currency of the creditor. The latter method has the advantage of influencing the money supply of the creditor country only temporarily and was frequently facilitated by extending the settlement date automatically or by mutual agreement.
A substantial part of intervention in currencies of participating central banks was intramarginal and largely financed by using holdings of the intervention currency which had been acquired earlier in the markets or resulted from borrowing abroad. Another part was financed by spot settlement in U.S. dollars or ECUs.
Role of the ECU
At the start of the EMS, each central bank participating in the exchange rate mechanism was to contribute 20 percent of its gold holdings and 20 percent of its gross reserves in U.S. dollars to the European Monetary Cooperation Fund (EMCF). Against these contributions, the EMCF issued a corresponding amount of ECUs. These transactions took the form of revolving three-month swaps. For the purpose of these transactions, U.S. dollars are valued at the market rate, and gold at the average market price of the six previous months or of the two fixings on the penultimate working day, whichever is lower. Every three months, the necessary adjustments are made to ensure that contributions continue to represent at least 20 percent of the gold and U.S. dollar reserves of the participating central banks and to bring the amount of issued ECUs in line with changed valuations.
At the beginning of the EMS, ECU 23 billion were created. When, in July 1979, the United Kingdom decided to voluntarily contribute 20 percent of its gold and U.S. dollar reserves, the amount increased to ECU 27 billion. Subsequently, the amount of ECUs issued by the EMCF rose to nearly 50 billion in April 1981, but later fell to ECU 38 billion in July 1982. It increased again to ECU 42 billion in December 1982 (Table 27). Since July 1979, the quantity of gold contributed has remained virtually unchanged, while, since October 1979, the amount of U.S. dollars has fallen. The valuation of U.S. dollars fluctuated according to developments in exchange markets, keeping the ECU equivalent of U.S. dollar contributions, on balance, unchanged. The key to the growth of ECUs was the rise in the price of gold. As the shares of gold and U.S. dollars in the reserves (and hence in the contributions) of EMS central banks differ significantly, the distribution of ECUs among them is strongly affected by changes in the valuation of these two assets.
In sum, under the present provisions, the creation and distribution of ECUs is, apart from changes in international reserves of EMS member countries, determined by three variables that are outside the control of the system: the price of gold, the exchange rate of the U.S. dollar, and the respective share of these two assets. “Under these circumstances, the quantity of ECUs created cannot be expected necessarily to be consistent with the aims of the system….”25 Various proposals to stabilize the amount of ECUs created have been presented but are at this time not under active consideration.
The actual use of ECUs has been limited. In December 1981, the positive or negative net positions of central banks in ECUs (i.e., amounts above or below those received from the EMCF) reached less than 8 percent of the amount of ECUs created through the swap arrangements; this percentage rose in the first half of 1982, before declining to around 7 percent in August 1982. A number of reasons may account for this development. Debtor central banks may have been reluctant to use ECUs, because the transitional character of the EMS could eventually require them to clear negative ECU positions by acquiring ECUs from other participants. On the other hand, creditor central banks may have discouraged the use of the ECU in view of its limited attractiveness, largely due to the lack of convertibility and the constraints on usability even within the system.
Credit Facilities
The very short–term financing facility has already been discussed. While this facility is limited to participants in the exchange rate mechanism, other EC credit facilities are open to all member countries, and their establishment preceded the coming into existence of the EMS. The short–term monetary support, which is administered by the EC central banks, is designed to provide finance for temporary balance of payments deficits on the request of a debtor central bank within the limits of its debtor quota; credits beyond these amounts are discretionary. The granting of medium-term financial assistance in the case of balance of payments difficulties is subject to a decision by the EC Council of Ministers which lays down economic policy conditions.
Another facility for dealing with balance of payments difficulties, the Community loan mechanism, is also of a medium-term nature and was established in 1975 in connection with the first round of oil price increases. Under this facility, the Community can borrow in the market or from other sources and on-lend to member countries up to ECU 6 billion.26 As under the medium-term financial assistance facility, credits are subject to a Council decision and linked to economic policy conditions.
The usability of the short-term monetary support and medium term financial assistance facilities, which are schemes of mutual assistance, depends on the strength of the balance of payments and reserve position of the EC as a whole. The Community loan mechanism, with its reliance on outside sources of finance, not only supplements but to a certain degree also substitutes for the other credit facilities, in particular the mediumterm financial assistance facility. Apart from the very short-term financing facility, none of the credit facilities has been used since the start of the EMS. This may be explained by the relatively large reserves that at least some of the EMS countries have at their disposal. The easy access of EC countries to international markets may have made recourse to conditional credit facilities of the EC (or of the International Monetary Fund for that matter) less attractive, while the good credit rating of EC countries on the market may have been enhanced by the very existence of large credit facilities within the EC.
Proposals for Institutional Changes
In the European Council’s Resolution of December 5, 1978, which laid down the framework for the “initial phase” of the EMS, the intention stated was “to consolidate, not later than two years after the start of the scheme, into a final system the provisions and procedures” of the initial phase.27 In this second phase the EMS would “entail the creation of the European Monetary Fund [which would replace the European Monetary Cooperation Fund]… as well as the full utilization of the ECU as a reserve asset and a means of settlement.”28 The existing credit facilities were to be consolidated into a single fund.
Work on the next phase began soon after the start of the EMS in the Committee of Central Bank Governors and the Monetary Committee. The discussions concentrated on the following problems:
-
—The place and development of the ECU in the EMS. Of particular interest were the problems relating to a permanent transfer to the European Monetary Fund (EMF) of a certain portion of member countries’ reserves against ECUs (compared with the provisional transfer on the basis of revolving swap arrangements as practiced at present); and to the role of the ECU as a means of settlement and its attractiveness as a reserve asset, with emphasis on such questions as full usability within the system and convertibility into other reserve assets.
-
—The credit mechanisms and their consolidation in the EMF. A special issue was the fact that the various credit facilities (the very short-term financing facility, the shortterm monetary support, and the mediumterm financial assistance facility) are subject to different procedures governing their use. They have different maturities and objectives and may or may not involve conditionality. Credits are to be granted under the responsibility of different institutions, the central banks in the case of the very shortterm financing facility and the shortterm monetary support, and the Council of Ministers in the case of the mediumterm financial assistance facility.
-
—The role and structure of the EMF. The main questions were whether and to what degree the EMF should be autonomous from governments, and how its decision-making bodies would be composed; and what tasks and powers in the field of foreign exchange market intervention, granting of credits, and the creation of liquidity it should get.
Due in part to the worsening of the economic climate in the EC countries and the world at large, but mainly because of significant differences of opinion as to how these questions should be approached, the aim of limiting the initial, transitional phase to two years could not be observed. It became obvious that the economic, political, and legal problems would be formidable, going in scope far beyond technical considerations. Subsequently, a more gradual approach emerged, and, in early 1982, the EC Commission submitted to the Council of Ministers a set of proposals.29 These proposals were intended to further the step-by-step development of the EMS in certain key areas without waiting for a final, fully developed system, and dealt with the following problems:
-
—method of issue of ECUs, with the aim of limiting the volatility in the amount of created ECUs;
-
—abolition of the acceptance limit (now 50 percent) for the use of ECUs in intra-EMS settlement of intervention debts;
-
—increased private use of ECUs;
-
—automaticity for financing of intramarginal intervention (i.e., in EMS currencies);
-
—measures to advance convergence, mainly by way of consultation and recommendation; and
-
—coordination of attitude toward third currencies, in particular the U.S. dollar.
By their very nature, these proposals concentrated on certain operational and technical aspects which can more easily be formalized but nevertheless might have important policy implications (e.g., acceptance limit for ECUs, financing of intramarginal intervention), whereas they were less specific in other areas, such as the promotion of convergence in economic performance or the coordination of attitude toward third currencies. After thorough discussions in the competent bodies of the EC, no agreement on the package of proposals could be reached. Some member countries who favored the proposals concentrated on their operational aspects and emphasized that their implementation would help to consolidate and strengthen the system by enhancing its predictability. Other member countries opposed the proposals on the grounds that their acceptance would have adverse implications for the conduct of domestic monetary policy. They stressed the need to keep the system flexible; as long as there was no sufficient degree of convergence of economic performance, any attempt to make the features of the system more stringent would only weaken and ultimately endanger the system. There was a general consensus that one of the prime objectives for the countries participating in the EMS remained the pursuit of policies conducive to greater convergence in economic performance.