Chapter

IX European Monetary Unification and the EFTA

Author(s):
International Monetary Fund
Published Date:
December 1990
Share
  • ShareShare
Show Summary Details

Although the establishment of an economic and monetary union (EMU) is not part of the internal market program, its completion is sometimes seen as a precondition for the program’s success because of the far-reaching implications of complete financial integration (Schinasi, 1989). At the same time, the increased likelihood of EC-EFTA free trade in financial services has raised the issue of whether this will also require EFTA countries to participate in the EC’s arrangements on monetary and economic cooperation. This would entail joining the European Monetary System (EMS), which in turn would have major implications for the conduct of economic policies in the EFTA countries. The benefits and costs of joining the EMS would, to a large extent, depend on the policy adjustments membership would require.

Evolution of Monetary Integration in the EC

The EEC Treaty does not include precisely defined commitments regarding monetary and exchange rate policies. Monetary unification is not a stated objective, although the Treaty does cover most of the conditions necessary for monetary integration (Emerson, 1979, p. 23). However, when the fixed exchange rate system came under pressure in the late 1960s, the Heads of State and Government of the EC decided to seek the creation of an EMU in Europe. The Werner Plan, presented in 1971, envisaged fixed exchange rates, free movement of capital, and coordinated monetary policies supervised by a common monetary institution. It also favored harmonized economic policies, especially in the fiscal field. However, events in the world economy prevented the Plan’s implementation.

In April 1972, after the suspension of U.S. dollar convertibility and the Smithsonian Agreement of December 1971, EC member states concluded a common margin agreement in Basle aimed at reducing the fluctuations between their currencies. Under this agreement, the EC central banks set up an arrangement known as the “snake in the tunnel,” because the allowed margin in the arrangement was only half the 4.5 percent allowed under the Smithsonian Agreement.72 In May 1972, the Norwegian krone became associated with the snake. In March 1973, pressure from speculative capital flows forced the participants to jointly float vis-à-vis the dollar, causing the tunnel around the snake to disappear. Both Norway and Sweden participated in this new arrangement. However, conditions in the snake remained unstable, causing several currencies—including the British and Irish pounds, the Italian lira, and the French franc—to leave. Sweden and Norway left the snake in August 1977 and December 1978, respectively, as they regarded monetary policy in the other participating countries as too restrictive and incompatible with the overriding domestic objective of high employment (Gylfason, 1990, p. 5). Although limited progress was made toward monetary integration during this period, the snake arrangement served as a useful framework of cooperation for the remaining countries and laid the foundation for the EMS.

In December 1978, a resurgence of interest in monetary cooperation led to the adoption of the resolution of the European Council, which formed the basis of the EMS. In the hope of creating a zone of monetary stability in Europe, the EMS was based on three cornerstones: the exchange rate mechanism (ERM); the ECU; and several credit facilities (Ungerer, 1979). 73 The EMS, which became operational in March 1979, has shown much flexibility as well as the ability to evolve gradually over time (Ungerer, 1990; and Ungerer and others, 1983, 1986). The success of the EMS in fostering monetary convergence to secure increasingly stable exchange rates within the ERM has created favorable conditions for further European economic and monetary integration (Giavazzi and others, 1988; Gros and Thygesen, 1988; and DeCecco and Giovannini, 1989).

As the initial goals of the EMS have progressively been achieved with full capital liberalization ahead of schedule, there are concerns that the demands on the EMS may eventually exceed the system’s capabilities.74 At the Hanover summit in June 1988, the European Council appointed a committee chaired by EC Commission President Jacques Delors to propose steps leading to monetary unification within the EC. The Delors Committee presented its report in April 1989 (Committee for the Study of Economic and Monetary Union, 1989). The report proposed a three-stage process to achieve a monetary union based on total and irreversible convertibility of currencies, completely free capital movements, fully integrated financial markets, and irrevocably fixed exchange rates.

In the first stage, economic and monetary coordination would be strengthened to secure greater convergence of economic performance. New forms of multilateral surveillance would be used where policy corrections were of joint interest, and Community-level consultations would lead to proposals for changes in national policies. Member states would also be expected to consult with the Community before adopting key monetary targets or policies. A single financial area would be formed, and all EC member currencies would participate in the ERM.

In the second stage, a European System of Central Banks (ESCB) would be established and would gradually absorb the functions of the present European Monetary Cooperation Fund, and the Committee of Central Bank Governors and its subcommittees (which were established in the first stage). Budgetary coordination would be further tightened, and exchange rate realignments would be permitted only in exceptional circumstances.

The third stage would entail the transition to a single currency. At this point, the ESCB—which would already be implementing Community exchange rate policy—would assume control over exchange market intervention in third currencies. Within the Union, Community bodies would be given enforceable powers in such areas as the binding coordination of members’ budget policies and, if necessary, conditional use of EC resources to secure structural and regional policy objectives or to intensify members’ adjustment efforts. At this stage, the Community would assume its full role in international policy coordination and international monetary negotiations.

At its Madrid meeting in June 1989 the European Council endorsed the Delors Report. Restating that the EMU would be indispensable for the completion of the internal market, the EC decided that the first stage would begin on July 1, 1990. For the second stage, the EC has set a target date of January 1, 1994. As this stage would include the transfer of monetary sovereignty—thus requiring adjustments of the Treaties of Rome—it was agreed to set up an inter-governmental conference for negotiating the exact terms of these adjustments. While this conference resumed its work in December 1990, the start of the third stage is still under consideration.

Exchange Rate Arrangements in EFTA Countries

In contrast to the EC, where national monetary and exchange rate policies aim at convergence, exchange rate policies in the EFTA countries differ markedly, especially between the Alpine and Nordic member countries. Austria has pegged the external value of the schilling to the deutsche mark, requiring the authorities to use monetary policy exclusively to maintain exchange rate stability.75 Switzerland has adopted similar monetary targets as in Germany, leading to a de facto harmonization of monetary policies.76 It has not set a formal exchange rate target, but since the monetary policies of the Deutsche Bundesbank and the Swiss National Bank have been guided by similar targets, the deutsche mark/Swiss franc exchange rate has remained remarkably stable.77

In contrast to the Alpine EFTA countries, all Nordic EFTA countries but Norway peg their currencies against baskets of currencies in which the ERM currencies have a substantial weight. Currencies included in the baskets generally reflect the relative importance of trade with other countries (Table 5). However, in the case of Sweden, the U.S. dollar has been given double weight since a substantial share of Swedish exports is denominated in dollars. The dollar has also a disproportionately large weight in the Icelandic currency basket, reflecting the importance of the dollar for capital account transactions. The currency weights in the baskets are reviewed periodically.

Table 5.Currency Basket Arrangements in the Nordic EFTA Countries1
Finnish

markka
Icelandic

króna
Swedish

krona
Composition of the basket2
Austrian schilling1.60.41.3
Finnish markka2.16.7
Norwegian krone3.85.28.7
Swedish krona19.15.3
Swiss franc2.42.32.1
Total weight of EFTA
currencies26.915.318.8
Belgian franc3.11.73.7
Danish krona4.47.07.6
Deutsche mark19.113.416.4
Netherlands guilder4.84.74.6
French franc6.54.75.2
Italian lira5.03.23.8
Spanish peseta1.92.21.5
Total weight of ERM
currencies44.836.942.8
Canadian dollar0.41.2
Japanese yen6.37.93.6
Portuguese escudo4.5
Pound sterling13.615.211.1
U.S. dollar8.420.022.5
Organized foreign exchange
marketYesNoYes
Weighting scheme3Bilateral4Bilateral5Bilateral6
Margins±3.0±2.25±1.5
Adjustment frequencyquarterlyannuallyannually
Sources: National authorities.

As of October 31, 1990.

Weights in percent of total.

In percent of the basket index.

The index includes all convertible currencies of those countries accounting for at least 1 percent of Finland’s foreign trade in commodities during the previous three-year period.

The trade index contains 15 currencies of those countries that are most important for Iceland’s foreign trade.

The index includes all convertible currencies of those countries that account for at least 1 percent of Sweden’s foreign trade during the previous five-year period.

Sources: National authorities.

As of October 31, 1990.

Weights in percent of total.

In percent of the basket index.

The index includes all convertible currencies of those countries accounting for at least 1 percent of Finland’s foreign trade in commodities during the previous three-year period.

The trade index contains 15 currencies of those countries that are most important for Iceland’s foreign trade.

The index includes all convertible currencies of those countries that account for at least 1 percent of Sweden’s foreign trade during the previous five-year period.

Until recently, Norway too pegged its currency to a basket. However, shortly after the United Kingdom decided to join the ERM, Norway decided to peg the krone to the ECU. The central value of the krone was set at NKr 7.9940 per ECU. As before, the Bank of Norway will keep the value of the krone within a fluctuation range of 2.25 percent around its central value.

EFTA’s Interest in Monetary Consultations with the EC

The EFTA’s countries—while willing to participate in financial integration in Western Europe—recognize the implications of free trade in financial services, particularly as they relate to monetary independence, currency substitution, fiscal policy, and the ability to deal with country-specific shocks (Folkerts-Landau and Mathieson, 1989). There is growing concern whether the EFTA countries—with the EC as their largest trading partner—can achieve full integration of financial services without participating in the EC’s economic and monetary arrangements. For example, the report of the EFTA’s Consultative Committee on the Creation of a European Financial Area states:

Having in mind the interdependence of the EFTA and the EC countries and the goal of the European Economic Space, the avoidance of divergent exchange rate developments between the countries is an important task… Instabilities … will undermine the creation of a European Financial Area (EFTA, 1989, p. 20).

Reflecting these concerns, the EFTA heads of state, in March 1989, expressed an interest in intensified consultations with the EC on economic and monetary policy (“EFTA Summit …, ” 1989, para. 12). However, they did not say whether the talks should be informal or within the framework of the EMS. This aim, stated even before the Delors Report, has been interpreted as being “not far from what is foreseen as part of the first step of the EMU” (Kleppe, 1989). Some have argued that—considering the challenges arising from financial integration—there is no alternative but for the EFTA countries to join the EMS, since only through formal economic and monetary cooperation with the EC can the EFTA countries hope to defend their interests within the European Economic Space.78

Mr. Duisenberg, head of the Dutch central bank, has also argued that having the EFTA countries join the EMS might be in both their interest and that of current EMS members:

Coherence within [the EMS] has grown and will continue to grow now that policies are increasingly converging and the planned full liberalization of capital markets is gradually coming nearer. This raises the question whether it would perhaps be better if all European countries, and that includes all EFTA countries, were to assign greater significance to the stabilization of intra-European exchange rates. This would create a broader ‘base of stability’ against the non-European currencies, such as the U.S. dollar. This would be in perfect harmony with EFTA’s declared intention to seek participation in the formation of the EC internal market (Duisenberg, 1988, pp. 40 and 42).

These arguments appear to assume that the formation of the EES would create an optimum currency area for the EC and EFTA. If this is the case, the enlargement of the EMS and EFTA’s participation in the Delors Plan toward complete monetary unification in Western Europe could indeed be beneficial for both blocs.

Concept of the Optimum Currency Area

There is considerable debate as to what constitutes an optimum currency area,79 with the literature going back to a classic study by Mundell (1961). In this study, Mundell argued that an optimum currency area must have a high degree of mobility of factors of production requiring a large measure of openness for both labor and capital. McKinnon (1963) also viewed openness as an essential feature of a currency area. However, he focused on goods and services, arguing that openness should be measured by the size of the tradables sector. Kenen (1969) argued that the degree of diversity of the product mix in an economy was also important in determining a currency area. Finally, shifting the emphasis from the goods to the capital markets, others have argued (Ingram, 1969) that an optimum currency area requires a high degree of financial integration. Capital mobility alone, as suggested by Mundell, would not be a sufficient condition.

All these approaches seek to establish objective criteria for the formation of a currency area. The extent of factor mobility, product diversification, and financial integration—as well as the share of tradables in production—are all seen as indicators of whether a country’s economic policy interests would be furthered by joining a given currency area. Based on these criteria, the EES—with the EFTA countries fully participating in all four freedoms—may indeed form an optimum currency area.80 As observed earlier, the formation of the EES, combined with the internal market program, would imply free movement of capital and labor between the EC and EFTA. With the removal of capital controls and the deregulation of cross-border restrictions on the provision of financial services, financial markets would become integrated. Moreover, all EFTA countries are very open economies and most have a relatively high degree of product diversification.

In addition to the approaches that emphasize objective criteria, other literature focuses on “subjective” criteria. This approach can be divided into two broad categories, one focusing on commonality of aims and one on similarity of policy attitudes. The former includes Haberler (1970) and Fleming (1971)—who stressed the importance of broadly equivalent rates of inflation—and De Grauwe (1975), who emphasized the need for similar inflation-output or inflation-employment trade-offs. The latter category includes Tower and Willett (1976), who argue that an optimum currency area requires a high degree of policy integration.81

The “subjective” approaches are not very different from a third approach that focuses on the costs and benefits of participating in a currency area. Although the cost-benefit approach takes a different perspective, there are obvious similarities with the subjective approaches. In principle, it can be assumed that if differences between national economic policy objectives and attitudes exist, the costs of joining an existing currency area will depend on the degree of necessary policy adjustments. The more economic priorities align with those in the area, the less the need for adjustments and hence the lower the costs of joining (and vice versa).

Based on this approach, Canzoneri and Rogers (1990) and Végh and Guidotti (1990) discuss the issue from a fiscal point of view. Given that optimal taxation requires that the marginal disutility from different forms of taxation including the inflation tax is equalized, they argue that an optimal currency area would include countries with the same level of inflation. However, if national tax structures differ, requiring divergent inflation rates, the harmonization of inflation rates necessary to preserve stable exchange rates would be a sub-optimal solution. The welfare costs involved have to be weighed against the benefits of having a single currency, especially the absence of valuation and currency conversion costs, which may impede trade between regions.

Despite similarities between the subjective approaches and the cost-benefit approach, they may result in different conclusions. Assume, for example, that economic policies differ sharply in the countries concerned. The subjective approaches imply that these countries would not form an optimum currency area. In contrast, the cost-benefit approach may achieve the opposite result, if the benefits of adjusting economic policies exceed the cost associated with these adjustments. Since not only the cost but also the benefits of forming a currency area may rise with the adjustments, sharp differences in economic policies do not preclude the possibility of an optimum currency area. In turn, relatively similar policies do not necessarily imply formation of a currency area. Although the costs may be relatively small, they may still exceed the benefits.

Exchange Rate Behavior, Monetary Growth, and Inflation

A close examination of the EFTA countries’ exchange rate and monetary policies yield several conclusions. First, the short-term exchange rate variability of the currencies of the Alpine states against the ERM currencies is remarkably low (Guitián, 1988, p. 20). In fact, the variability of the Austrian schilling is lower than for the currencies in the ERM. While relatively limited exchange rate movements vis-à-vis the ECU can be observed for the Finnish markka, the Norwegian krone, and the Swedish krona over the medium and long run, the movement of the frequently devalued Icelandic krona differed markedly from all ERM currencies (Charts 4 and 5). The Alpine EFTA states’ currencies, on the other hand, have maintained their values vis-à-vis the ECU at least as well as most of the de jure ERM participants. Over time, both currencies have followed the appreciation of the deutsche mark and the Dutch guilder against the ECU.82

Chart 4.EMS: Exchange Rates Against the ECU

(third quarter, 1979 = 100)

Source: IMF, International Financial Statistics

Chart 5.EFTA: Exchange Rates Against the ECU

(third quarter, 1979 = 100)

Source: IMF, International Financial Statistics

The differing exchange rate developments in the EFTA countries may to a large extent be explained by differing monetary growth rates. For example, monetary growth rates in the Alpine members of the EFTA seem to have followed developments in the EMS relatively closely (Table 6). A tendency toward lower money growth is also observed in Sweden, while monetary growth in Finland and Norway has remained faster than in the EMS. However, the more rapid monetary growth may be partly explained by the financial reform that took place in these countries during the mid-1980s.

Table 6.Monetary Growth in the ERM and in EFTA Countries1
19791980198119821983198419851986198719881989
ERM countries2
Average increase313.08.710.49.513.210.48.28.98.77.67.4
Average deviation from
mean45.92.93.72.35.17.04.53.02.92.02.0
Average deviation from
lowest45.24.95.73.98.05.73.73.24.74.42.9
EFTA countries
Austria
Increase311.19.511.411.08.63.96.76.98.86.77.0
Deviation from ERM
mean44.30.70.61.55.410.73.63.50.31.41.1
Deviation from ERM
lowest44.06.06.64.93.40.92.21.25.03.52.5
Finland
Increase315.916.015.415.013.713.417.313.510.714.118.3
Deviation from ERM
mean40.57.23.45.50.31.27.03.12.26.010.2
Deviation from ERM
lowest48.812.510.68.98.28.612.87.86.910.913.8
Iceland
Increase355.555.278.558.076.949.241.643.936.126.326.0
Deviation from ERM
mean440.146.466.548.562.934.631.333.527.618.218.1
Deviation from ERM
lowest448.451.773.751.971.744.437.138.232.323.121.5
Norway
Increase313.112.112.710.710.314.116.29.021.05.55.5
Deviation from ERM
mean42.33.30.71.23.70.55.91.412.52.62.6
Deviation from ERM
lowest46.08.67.94.65.19.811.73.317.23.31.0
Sweden
Increase315.49.79.713.621.84.03.713.310.68.410.9
Deviation from ERM
mean40.92.34.17.410.66.62.92.10.32.8
Deviation from ERM
lowest48.36.24.97.516.40.80.87.66.85.26.4
Switzerland
Increase39.56.84.33.38.67.45.92.59.38.35.7
Deviation from ERM
mean44.92.07.76.25.47.24.47.90.80.22.4
Deviation from ERM
lowest42.43.30.52.83.42.61.43.25.55.11.2
Source: IMF, International Financial Statistics; and IMF staff calculations.

Monetary growth (changes over preceding year) refers to broad money.

During 1979–88, ERM countries include Belgium, Denmark, France, Germany, Ireland, Italy, Luxembourg, and the Netherlands. While Spain is included for 1989, the United Kingdom has been omitted as it joined the ERM only in October 1990.

In percent.

In percentage points.

Source: IMF, International Financial Statistics; and IMF staff calculations.

Monetary growth (changes over preceding year) refers to broad money.

During 1979–88, ERM countries include Belgium, Denmark, France, Germany, Ireland, Italy, Luxembourg, and the Netherlands. While Spain is included for 1989, the United Kingdom has been omitted as it joined the ERM only in October 1990.

In percent.

In percentage points.

On the other hand, money growth in Iceland greatly exceeded that in the EMS.

Since a currency area with fixed exchange rates requires converging inflation rates at least over the medium run, it is useful to consider the extent to which price developments in the EFTA countries have deviated from those in the EMS (Table 7). Inflation in the EMS also seems to have converged at a lower level.83 Again, Austria and Switzerland exhibit relatively low inflation rates, whereas rates in the Nordic countries have remained higher.

Table 7.Inflation in the ERM and EFTA Countries1
19791980198119821983198419851986198719881989
ERM countries2
Average inflation rate38.211.211.710.68.06.34.82.22.02.44.1
Average deviation from
mean33.85.35.03.62.92.01.81.91.71.51.4
Average deviation from
lowest44.15.95.45.35.23.92.52.42.71.73.0
EFTA countries
Austria
Inflation rate33.76.46.85.43.35.73.21.71.41.92.6
Deviation from ERM
mean45.86.96.45.85.50.92.51.40.61.03.0
Deviation from ERM
lowest40.41.00.50.10.53.31.01.92.11.21.5
Finland
Inflation rate37.511.612.09.68.47.15.92.94.15.16.6
Deviation from ERM
mean42.01.71.41.60.40.50.20.22.12.22.6
Deviation from ERM
lowest43.46.25.74.35.64.23.73.14.84.45.5
Iceland
Inflation rate345.358.650.951.084.329.232.421.318.824.720.8
Deviation from ERM
mean435.845.337.539.875.522.626.718.216.821.816.8
Deviation from ERM
lowest441.253.244.645.781.526.830.221.519.524.019.7
Norway
Inflation rate34.810.813.711.48.46.35.77.28.76.74.6
Deviation from ERM
mean44.71.50.30.20.40.34.16.73.80.6
Deviation from ERM
lowest40.75.47.46.15.63.93.57.49.46.03.5
Sweden
Inflation rate37.213.712.18.68.98.07.44.24.25.86.4
Deviation from ERM
mean32.30.41.32.60.11.41.61.12.23.92.4
Deviation from ERM
lowest43.18.35.83.36.15.65.24.44.95.15.3
Switzerland
Inflation rate33.64.06.55.73.02.93.40.81.41.93.2
Deviation from ERM
mean45.99.36.95.65.83.72.32.30.61.00.8
Deviation from ERM
lowest40.51.40.20.40.20.51.21.02.11.22.1
Source: IMF, International Financial Statistics; and IMF staff calculations.

Changes in consumer prices (over preceding year).

During 1979–88, ERM countries include Belgium, Denmark, France, Germany, Ireland, Italy, Luxembourg, and the Netherlands. While Spain is included for 1989, the United Kingdom has been omitted as it joined the ERM only in October 1990.

In percent.

In percentage points.

Source: IMF, International Financial Statistics; and IMF staff calculations.

Changes in consumer prices (over preceding year).

During 1979–88, ERM countries include Belgium, Denmark, France, Germany, Ireland, Italy, Luxembourg, and the Netherlands. While Spain is included for 1989, the United Kingdom has been omitted as it joined the ERM only in October 1990.

In percent.

In percentage points.

These findings imply that, at present, the subjective approaches do not support the view that the EMS is an optimum currency area for the Nordic EFTA countries, whereas Austria and Switzerland have behaved as if they had already belonged to the EMS.84 This suggests that if the EFTA countries were to join the EMS, major policy adjustments could probably be avoided only in the two Alpine countries. In the Nordic countries, however, joining would require—to varying degrees—changes in their economic policies. From the view of the cost-benefit approach to optimum currency areas, does an accession of the EFTA countries to the EMS appear desirable?

Cost and Benefits of Joining the EMS

With perfect capital mobility and fixed exchange rates, there is little scope for an independent monetary policy. The central bank controls neither the domestic interest rate nor the money stock. The only effect of an open-market operation is a change in the composition of its balance sheet. The speed with which central banks purchase bonds determines the speed of their loss of reserves. This is one of the main conclusions of the literature on the Mundell-Fleming model (Mundell, 1968; Fleming, 1962; Frenkel and Razin, 1987;Genberg, 1989; and Kenen, 1985).

The experience with the EMS—where capital mobility has been less than perfect and exchange rates not completely fixed—has shown that such a system severely limits autonomy in policy formulation (Guitián, 1988). Under present arrangements, monetary coordination involves using the Deutsche Bundesbank’s policy as a reference for the conduct of monetary policies in the other ERM member countries (Schinasi, 1989). However, with full financial integration and irrevocably fixed exchange rates as envisaged in the Delors Report, any remaining independence would be lost, requiring complete monetary coordination among the EMS members.85

All EFTA countries except Switzerland pursue exchange rate targets. These countries are aware of the constraints that fixed exchange rates impose on economic policymaking.86 Joining the EMS would not imply a change from a flexible exchange rate regime to a fixed rate regime, but one from a fixed but adjustable exchange rate regime to another.87 The important difference would be that under existing arrangements discretionary exchange rate changes are the Government’s decision, while, in the EMS, agreement must be reached with other participants. With exchange rate realignments in the EMS depending on a joint decision by all participants, the authorities are forced to follow the monetary policy stance in the other member countries, which, in effect, has been dominated by the anti-inflationary policy of the Bundesbank. This additional constraint on policy autonomy becomes more severe, the more monetary policy and inflation rates deviate from those in the EMS.

The further loss in monetary autonomy would constrain the authorities’ ability to reduce the cyclical, or “demand deficient,” component of unemployment with an expansionary monetary policy.88 As a result, the reliance placed on structural and fiscal policies would increase. However, a reduction in monetary autonomy may also reduce the extent to which fiscal deficits can be financed by seigniorage from money creation (Table 8). As noted by Dornbusch (1988), part of this loss would come from a reduction in monetary expansion to a rate compatible with lower inflation rates. In addition, financial integration would bring pressure to harmonize reserve requirements for commercial banks. Those countries with relatively high reserve requirement ratios face, other things being equal, the largest prospective loss of seigniorage.89 Although the effects on financing fiscal deficits may be relatively small, it may still increase the burden on stabilization policies.

Table 8.Seigniorage in the EFTA Countries, 1980–8711
In Percent of GDPIn Percent of Total

Government Revenues
Austria0.41.2
Finland0.82.9
Iceland22.06.3
Norway0.40.9
Sweden0.40.9
Switzerland0.44.2
Source: IMF, International Financial Statistics; and IMF staff calculations.

Seigniorage is measured as the December-to-December change in reserve money.

Since required reserves have been indexed in Iceland, seigniorage is calculated as the increase in the nonindexed part of reserve money.

Source: IMF, International Financial Statistics; and IMF staff calculations.

Seigniorage is measured as the December-to-December change in reserve money.

Since required reserves have been indexed in Iceland, seigniorage is calculated as the increase in the nonindexed part of reserve money.

In addition, the harmonization of revenues from seigniorage could lead to a sub-optimal increase in tax distortions. Assuming that current inflation rates are consistent with optimal taxation, implying that the mar-consistent with optimal taxation, implying that the marginal disutility of raising taxes is the same for all revenue sources, the harmonization of inflation rates may be welfare-decreasing (Canzoneri and Rogers, 1990). The harmonization of conventional taxes could aggravate this problem as revenue losses in some countries could lead to a heavier reliance on revenues from money creation (Végh and Guidotti, 1990).

A loss of monetary autonomy can be seen as a disadvantage for economic policymaking. However, by joining a fixed exchange rate system under German leadership, more inflation-prone countries can gain credibility in the disinflation process, provided that realignments are not frequent and entail political as well as economic costs.90 This gain in credibility arises from the fact that maintaining fixed exchange rates requires the authorities to keep their policies in line with the policy stance in the EMS. In this case, participation in the ERM has a disciplinary effect and thus allows a country more easily to “buy” the anti-inflationary reputation of the Deutsche Bundesbank.91

The need for discipline concerns not only monetary policy, but also the labor market. Where cost-push factors are dominant in determining the inflation rate, moderate wage settlements are needed to maintain exchange rate stability. On the other hand, fixed exchange rates may provide a nominal anchor for inflation expectations and may thus contribute to moderate wage demands.

However, competitiveness will deteriorate if discipline is less than perfect—and differences between the anti-inflationary policy stance in Germany and in the country concerned remain—and if needed realignments are postponed.92 To determine the net impact of ERM participation, the credibility gains have to be weighed against the output loss caused by the real appreciation of the exchange rate.93 While this is a difficult task—requiring intertemporal judgment—it cannot, a priori, be excluded that ERM participation may reduce welfare.94

It should also be remembered that the present currency basket arrangements in Finland, Iceland, and Sweden are designed to provide the maximum average stability in the effective exchange rate. This implies that the adverse effects of global exchange rate fluctuations on competitiveness are minimized, which would no longer be the case if they joined the ERM. However, the basket arrangements do not appear so different from the ERM that this effect would be of dominant importance. In fact, with the accession of the pound sterling in the ERM in October 1990, the similarity between the basket arrangements and the ERM has become even greater. Presumably, this development has been a major contributor to Norway’s decision to peg the krone to the ECU.95

It has also been argued that financial integration—especially the removal of all existing capital controls—may increase the risk of large and destabilizing capital movements. However, as noted by Kleppe (1989), joining the ERM would have the advantage of having access to unlimited very short-term support when the currency is under pressure.

Views of Individual EFTA Countries

The preceding discussion does not provide a clear answer about whether participation in the ERM would yield benefits that would unambiguously exceed its costs. Such a judgment would depend on economic preferences in the country concerned, as well as their weighting of the potential costs and benefits. While the EFTA as a whole recognizes the need for closer economic and monetary cooperation with the EC, the views of individual member states vary as to whether this requires a formal accession to the EMS.

Having applied for EC membership in 1989, the Austrian authorities recognize that accession to the EC would also imply participation in the EC’s formal monetary arrangements. Consequently, Austria has initiated concrete steps toward an association with the EMS according to Article 5.2 of the EMS agreement (“Dr. Klauhs Presents an Austrian View …,” 1989). However, even before the Austrian National Bank was authorized to enter into detailed exploratory talks on Austria’s association with the EMS, several preliminary steps had been taken. At Austria’s request, the Austrian National Bank was granted the status of “other holders of ECUs” in July 1989. The Bank has also participated in the daily consultations of EC central banks, while the ECU has been quoted on the Vienna Foreign Currency Exchange since 1986.

The Swiss National Bank was the first central bank outside the EC to be granted the status of “other holders of ECUs” by the European Monetary Cooperation Fund. However, Switzerland does not plan to join the EMS, mainly for two reasons: First, President Lusser of the Swiss National Bank believes that the additional discipline provided by fixed exchange rates is not necessary to achieve price stability. EMS membership would hardly improve Switzerland’s situation.96 Second, it has been argued that the EMS is not an optimal currency area for Switzerland.97 Such an optimal currency area would exist if labor and capital were mobile. Unlike Austria, however, Switzerland has objected to removing labor mobility restrictions vis-à-vis EC members.

Among the Nordic countries, so far only Norway has indicated that it may seek to join the EMS.98 Recently this interest has intensified, as indicated by the pegging of the krone to the ECU. A report published in 1989 by a government committee on monetary policy suggested that EMS membership would be advantageous mainly for two reasons (Norges Offentlige Utredninger, 1989). First, the ERM would provide Norway with better protection against speculation than the current basket arrangement, by giving access to unlimited short-term support when the krone came under pressure. Although the krone has been quite stable since it was last devalued in 1986, there have been isolated incidents of speculation. Thus, the gains from access to the very short-term financing facility may outweigh the losses—in terms of exchange rate stability—from abandoning the present basket arrangement. Second, EMS membership may, it is argued, add credibility to Norway’s stabilization policies. Although participation in the EMS would also result in a loss of monetary autonomy, the overall benefits would still be positive.99 Therefore, Norway might seek accession to the EMS, even if its fellow EFTA members decide not to.

Finland, Iceland, and Sweden do not at present plan to join the EMS nor to forge closer links with the EC’s exchange arrangements. Although the authorities in these countries are aware that closer financial integration with Western Europe reduces policy autonomy—irrespective of whether they join the EMS—they are not yet willing to surrender the exchange rate instrument and to subordinate monetary policy to exchange rate stability (Moe, 1990). As Mr. Franzén, the Deputy Governor of the Bank of Sweden, has put it: “A necessary condition for seriously raising the issue of a Swedish association to the EMS is that there are clear indications showing that the overall domestic policy forcefully strive to reduce inflation. Today there are no such convincing indications. Consequently, a Swedish EMS association is not on the agenda today” (“Some Aspects of the EMS,” 1990). However, Mr. Franzén also stressed that a future Swedish EMS association was not to be ruled out. If other EFTA countries were to join or become associated with the EMS, closer ties might become more attractive. As Kotilainen and Peura (1989) have argued, this would also apply to Finland. In this respect, any step by Norway toward joining the EMS might lead to a change in the others’ position. As far as Iceland is concerned—which is subject to frequent supply shocks attributable to its heavy reliance on fisheries—participation in the EMS appears difficult at this time. Since Iceland’s overriding economic policy goal has been to maintain full employment, the costs of subordinating monetary policy to exchange rate stability would probably far exceed the benefits of EMS membership.

72The Smithsonian Agreement re-established fixed exchange rates, with margins of ±2.25 percent, instead of ±1 percent under the Bretton Woods agreement.
73When the EMS became operational in 1979, all nine members but the United Kingdom participated in the ERM. Greece, Portugal, and Spain joined the EMS after acceding to the EC. Greece and Portugal have stayed out of the ERM, while Spain and the United Kingdom became participants in June 1989 and October 1990, respectively.
74With full capital mobility, speculation could lead to excessive swings in exchange rates and interest rates. Since financial integration is likely to increase the sensitivity of capital movements to changes in these variables, increased coordination of monetary policies may be necessary; see Russo (1988).
75Until 1978, Austria’s exchange rate policy was based on a currency basket, with the deutsche mark dominating. In 1981, the exchange rate was determined primarily by the deutsche mark. Austria’s strategy of pegging the schilling is often described as a “hard currency policy.” For details, see Genberg (1986) and Haltunen (1984).
76As did Germany, Switzerland switched to floating exchange rates in January 1973 in order to achieve effective monetary control. Monetary targeting was introduced in 1975 with the adoption of a target for M1. This policy was temporarily abandoned in 1978 in the face of strong demands to prevent an already poor competitive position from deteriorating. Instead, an explicit exchange rate target vis-à-vis the deutsche mark was set. However, monetary targeting was reintroduced in 1980 and targets have since been announced for the monetary base. This switch from M1 to the monetary base was related to exchange rate expectations and instabilities of the demand for M1.
77Although the external value of the Swiss franc is, in principle, endogenous, the Swiss National Bank occasionally modifies its policy of base money control by using the Swiss franc/deutsche mark exchange rate as a secondary intermediate target for monetary policy; see Genberg (1990).
78See “Keine Alternative zu EWS Beitritt …” (1988). Full membership in the EMS, with access to all credit facilities, seems only possible if the EFTA joins the EC. In contrast, accession to the ERM alone—where support through the very short-term facility for financing obligatory interventions is automatic—is also conceivable for non-EC members (Ungerer, 1986, pp. 5–6). Apart from this, the EFTA countries could also become associate members of the EMS. Such an association would, for example, consist of participation in the ERM, so that the associate member would undertake to keep its exchange rate within specified fluctuation limits fixed in relation to both the ECU and individual EMS currencies. However, in contrast to ordinary members, an associate member may not even have access to the very short-term facility of the ERM. This disadvantage could be offset through borrowing arrangements between central banks.
79This overview largely follows Guitián (1988), pp. 4–5.
80The position of Iceland, however, would appear somewhat tenuous.
81Of course, there is a close relationship between common aims and the degree of policy integration.
82According to Russo and Tullio (1988), the EFTA countries behaved as if they belonged to the EMS at least since 1983. However, the study does not include Finland or Iceland. In the Icelandic case, their findings would probably not apply.
83This did not necessarily occur because of the EMS. However, most empirical studies concluded that the EMS has provided a framework in which anti-inflationary policies could be pursued more effectively by entailing a binding constraint for high-inflation countries; see Russo and Tullio (1988).
84Whether the EMS is an optimum currency area for all participants in the system is not addressed here. However, Russo and Tullio (1988) found that some countries that are formally members of the EMS did not behave as “good” members by reason of their use of large margins of fluctuations, frequent realignments, or divergent economic policies.
85As argued, for example, by Giavazzi and Giovannini (1989), the present form of coordination—where monetary policy is set by Germany and the other countries may either go along with German monetary targets or realign—is inconsistent with fully integrated financial markets. Pointing to the Italian experience, where there was a major speculative attack when administrative controls on export credits were removed in May 1987, the authors emphasize the need for setting monetary targets by a supranational institution. Otherwise, speculative attacks could cause realignments that may be regarded as inconsistent with economic fundamentals.
86Capital controls in the EFTA countries have been almost completely removed and the remaining controls do not appear very effective. Akerholm and Tarkka (1987) argue that the Nordic EFTA countries could not conduct an independent monetary policy even when all controls are in effect. For empirical evidence, see Virén (1989).
87However, the degree of adjustability in the ERM may be open to dispute and, as noted, the Delors Report envisaged making ERM exchange rates irrevocably fixed.
88At the same time, financial integration may make it easier for European residents to switch among monetary assets denominated in different currencies so that the demand for key monetary aggregates may become less stable than in the past, which might complicate monetary control.
89The loss of revenue from money creation could be offset partly by two effects working in the opposite direction. First, lower inflation rates may lead to a greater willingness of the public to absorb money, thus reducing real interest rates. This effect will be particularly large in countries where the stock of interest-bearing public debt is large relative to the stock of base money. Second, “external” seigniorage gains could arise if assets were denominated in ECUs and grew in proportion to world trade; see Cohen and Wyplosz (1989).
90See Giavazzi and Pagano (1986) and Giavazzi and Giovannini (1989). However, as Lane and Rojas-Suarez (1989) have shown, the gain in credibility depends also on the degree of capital mobility and the margin within which the exchange rate is allowed to fluctuate.
91According to Kremers (1990), this could be observed for Ireland, where before 1979 Irish inflation expectations mainly followed expected movements in the United Kingdom. Upon entry into the EMS, Irish expectations converged toward the expected price behavior of ERM partners.
92If under these circumstances realignments are postponed, the currency of the country concerned will appreciate in real terms. As Russo and Tullio (1988, p. 49) argue, the size and duration of the real appreciation could be considered a measure of the extent to which the monetary authorities concerned have not been able to “borrow credibility” from the Bundesbank.
93Of course, the effects of the appreciation can be temporarily delayed by devaluing the exchange rate prior to entry in the ERM.
94Melitz (1988) argues that this may have been the case for France.
95Until recently, the Norwegian authorities maintained that EMS membership would hinge on that of the United Kingdom; see Skånland (1988), p. 46.
96See “Dr. Lusser warns against exaggerated expectations …” (1988) and “Schweizerische Notenbank: Absage an Beitritt zum EWS” (1988). However, as Dr. Lusser also stressed, Switzerland’s stabilization policy may nevertheless be beneficial for the monetary integration process in Europe as it enlarges the stability-oriented core of the EMS. See “Der Schweizer Beitrag zur Europäischen Währungsunion (1990).
99The Norwegian authorities recognize that the resulting loss in monetary autonomy may be particularly important for their country, which, as a major oil exporter, is subject to shocks that may be seen as country-specific relative to its trading partners. This has been one of the major reasons why the United Kingdom until recently has refused to join the ERM. See Ungerer and others (1986), p. 4.

    Other Resources Citing This Publication