EMU and the W orld Economy - International Monetary FundEMU and the Delors Committee Report of 1989 then set in train the three-stage process leading to EMU laid out in the Maastricht

The advent of European Economic and MonetaryUnion (EMU) scheduled for the beginning of

1999 will constitute a major change in the interna-tional monetary system. Involving as it will the re-placement of the national currencies of a number ofhighly developed economies of substantial size by asingle common currency, EMU will in fact have noparallel in history. But while EMU will be a majorchange, it will not be a sudden change. When the lead-ers of EU countries decide in May 1998 which coun-tries will participate in EMU from its outset, they willbe putting some of the final touches to the scene set forthe third and final stage of EMU, following the designlaid out in the Treaty on European Union, agreed byEU governments at Maastricht in December 1991.More fundamentally, the achievement of EMU will re-flect over 40 years of progress in strengthening eco-nomic, monetary, and political ties within Europe.

The Treaty of Rome of 1957, which founded theEuropean Economic Community, identified the ex-change rates of member states as a matter of commonconcern. Five years later, in 1962, the Commission ofthe European Communities set out a plan for monetaryunion. The Werner Report of 1970, endorsed by lead-ers of the European Communities in March 1971, en-visaged monetary union within a decade, but the planfoundered with the collapse of the Bretton Woods ex-change rate system of fixed but adjustable parities.Monetary integration was revitalized by the establish-ment of the European Monetary System (EMS) in1979, with an exchange rate mechanism (ERM) at itscore that was aimed at providing a zone of monetarystability in Europe. With the launch in the mid-1980sof the program to complete the EU’s internal market,attention refocused on the objective of a single cur-rency as part of the framework that would enhance thefunctioning of an increasingly integrated market(Figure 19).29 The greater stability of exchange rateswithin the ERM seemed to confirm that this was alsoa feasible goal. The reaffirmation by the EuropeanCouncil in June 1988 of the EU’s commitment toEMU and the Delors Committee Report of 1989 thenset in train the three-stage process leading to EMU laidout in the Maastricht Treaty.

51

IIIEMU and the World Economy

29The increasing economic integration of Europe was discussed inthe October 1994 World Economic Outlook, Annex I, pp. 98–103.

Figure 19. Trade Within the European Union1

1Imports plus exports of goods.

In percent of totalEuropean Unioncountries' trade

In percent of totalworld trade

1950 55 60 65 70 75 80 85 90 95

70

60

50

40

30

20

10

Since the early postwar period, intra-European Union trade has risento more than 60 percent from about 40 percent of total EuropeanUnion countries' trade.

©1997 International Monetary Fund

III EMU AND THE WORLD ECONOMY

The first stage, the initial preparations, began in July1990 and saw the abolition by all EU countries of re-maining restrictions on international capital move-ments. Stage 2 commenced on January 1, 1994, atwhich point the European Monetary Institute (EMI)—the forerunner of the European Central Bank (ECB)—came into being, monetary financing of budgetarydeficits and privileged access of the public sector to fi-nancial institutions were prohibited, and proceduresfor the surveillance of economic policies by EU insti-tutions were strengthened. At the beginning of Stage 3,exchange rates of participating countries will be irrev-ocably locked, a common currency, the euro, will beintroduced, and the ECB will take over responsibilityfor monetary policy. At the December 1995 meeting ofthe European Council in Madrid, it was agreed thatStage 3 would begin on January 1, 1999, though thepossibility of determining an earlier date had been en-visaged by the drafters of the treaty.30 The 1992–93crises in the ERM had threatened to derail the process,while the prolonged period of economic weakness thatensued impeded efforts to meet the criteria for qualifi-cation for EMU set out in the treaty, particularly in thefiscal area. These problems made it difficult to beginStage 3 before 1999.

This chapter addresses a number of questions raisedby the prospect of EMU. First, what is the nature ofthe change in policy regime that EMU will entail forthe participating countries: how will monetary and fis-cal policies be conducted in the euro area, and howwill participating economies adjust to diverse devel-opments in the absence of exchange rate flexibilityand domestic monetary policy? Second, to what extenthave prospective participants established the condi-tions needed for EMU to function successfully, andwhat more do they need to do? Third, what are thepossible pitfalls in the path to a smooth transition?Fourth, how will EMU affect the rest of the world?And fifth, how critical are fiscal and labor market re-forms for the success of EMU?

EMU: A New Policy Regime

Monetary and Exchange Rate Policy

EMU will represent a change in the monetary policyregime in Europe in two key respects. First, the scopefor national monetary policies, which countries haveretained in the ERM, albeit constrained by the centralobjective of exchange rate stability, will disappear.And second, there will be a shift in the geographic ori-entation of monetary policy. In the ERM, the anchor

role of the deutsche mark has meant that the policy ofthe Deutsche Bundesbank, which has aimed at pricestability in Germany and which has been set primarilyon the basis of domestic considerations, has had adominant influence on the stance of monetary policyin the system as a whole. By contrast, policy in EMUwill in principle be attuned to conditions throughoutthe euro area.

The design and implementation of monetary policyin the euro area will be the exclusive preserve of theEuropean System of Central Banks (ESCB), compris-ing the ECB and the national central banks (NCBs).The ECB will come into being soon after the decisionis taken on which countries will participate in Stage 3,and the EMI, which has prepared the technical ground-work for conducting monetary policy in the euro area,will then be dissolved. The Maastricht Treaty statesthat the primary objective of the ESCB shall be tomaintain price stability, which official statements sug-gest means inflation in the range of 0–2 percent ayear.31 Without prejudice to that objective, the ESCBshall also support the general economic policies inthe EU, with a view to contributing to such objectivesas sustainable growth, high employment, a high de-gree of convergence of economic performance, andeconomic cohesion and solidarity among memberstates.32 Monetary policy decisions will be made bythe ECB’s Governing Council, made up of its presi-dent, vice president, the other members of its Execu-tive Board (who may number up to four), all appointedby the heads of state or government of the countries inthe euro area, and the governors of the NCBs. TheExecutive Board will implement monetary policy inaccordance with the guidelines and decisions of theGoverning Council and give the necessary instructionsto the NCBs. The Executive Board may have certainpowers delegated to it where the Governing Councilso decides.

Among the issues that have arisen concerning theoperation of monetary policy in EMU, three are par-ticularly notable in the context of the change inmonetary policy regime: the independence and ac-countability of the ESCB; the strategy and intermedi-

52

30The treaty provides that “If by the end of 1997 the date for thebeginning of the third stage has not been set, the third stage shallstart on 1 January 1999” (Maastricht Treaty, Article 109j (4)).

31Thus, the EMI has observed that “While theory does not providea precise definition of price stability and while measurement prob-lems exist, there has been a broad consensus among central banksfor several years that a range of 0–2 percent inflation per annumwould be appropriate.” See The Single Monetary Policy in StageThree: Elements of the Monetary Policy Strategy of the ESCB(Frankfurt: European Monetary Institute, February 1997), p. 12.

32The precedence given to the objective of price stability in themandate of the ESCB is similar to that in the charter of theBundesbank. It may be contrasted with the statutory objectives ofthe U.S. Federal Reserve, which is charged by the Full Employmentand Balanced Growth Act of 1978 (U.S. Code, Title 12, Chapter 3,Subchapter I, paragraph 225a) “to promote effectively the goals ofmaximum employment, stable prices, and moderate long-term inter-est rates”—a mandate that places high employment on a similarfooting to price stability.

ate targets to be used by the ECB in pursuit of its ob-jectives; and differences among member countries inthe way the single monetary policy may affect theeconomy.

The ESCB will be vested with a high degree of in-dependence, which may be expected to help it pursuethe objective of maintaining price stability.33 Thetenure of members of the Executive Board of theECB—eight-year nonrenewable terms—is designed toinsulate them from political pressure, and the gover-nors of the NCBs must have terms of at least fiveyears.34 The Maastricht Treaty also precludes membergovernments from attempting to influence ECB deci-sions, and members of the ECB’s decision-makingbodies from seeking or taking instructions from gov-ernments or EU institutions. Additional features thatwill help insulate the ESCB from political influenceare treaty provisions that prohibit the ESCB from fi-nancing either governments or EU institutions or fromassuming their commitments. Moreover, changes tothe key provisions of the Statute of the ESCB must beratified by all EU countries.35 Although there are anumber of procedures for channeling the views ofthe Council,36 the Commission, and the EuropeanParliament to the ECB, there is no obligation on theECB to act on these views.

One concern about the ESCB’s independence hasbeen that it could be limited by the Council’s right toenter into formal agreements relating to exchange ratearrangements for the euro vis-à-vis non-EU curren-cies or to formulate “general orientations” for its ex-change rate. While this opens up the possibility thatthe ECB might have to take measures jeopardizing itsown objectives, action by the Council to enter intoformal agreements would require unanimity, wouldhave to be on a recommendation from the Commis-sion or the ECB, and in the case of the former wouldbe after consultation with the ECB “in an endeavor toreach a consensus consistent with the objective ofprice stability.” Moreover, the ECB will not beobliged to follow any “general orientations” for theexchange rate specified by the Council if it believesthem incompatible with achieving price stability. The

Council and the Commission, in cooperation with theEMI, have been requested by the European Council(i.e., the heads of state or government of EU membercountries), at its meeting in Amsterdam in June 1997,to study effective ways of implementing all provi-sions of the treaty relating to possible exchange ratearrangements for the euro, and to report to its meetingin Luxembourg in December 1997.

A formal exchange rate arrangement already agreedwithin the EU is “ERM2,” by which EU countriesoutside the euro area that wish to participate in a newexchange rate mechanism will have central rates fortheir currencies expressed in euro, with 15 percentfluctuation bands on either side as in the currentERM.37 This arrangement provides explicit protectionfor the ECB’s monetary policy independence in that itwill be party to the agreement on central rates and inthat it can suspend its intervention and financingobligations if it believes that these threaten the in-tegrity of its monetary policy. The ECB (along withthe other parties to the arrangement) will also havethe right to initiate the procedure for considering cur-rency realignments.

The independence of the ESCB therefore seemswell safeguarded. In fact, more questions have beenraised about the adequacy of its accountability to thepublic and parliamentary representatives. The long-run success of the ESCB in delivering low inflationwill to some extent, as with any monetary authority,depend on the public’s support for that objective. Inthis context, mechanisms to ensure public understand-ing of the ESCB’s policy decisions, in addition tothose set out in the treaty (which relate to the ECB’sAnnual Report and presentations to the EuropeanParliament and its committees) would be likely to en-hance its ability to pursue the requisite policies. Thisissue is related to that of the appropriate monetary pol-icy strategy.

While the ECB will make the final decision onwhich particular monetary policy strategy is best suitedfor achieving the objective of price stability, prepara-tory work done by the EMI has narrowed the options.Thus, the EMI has ruled out exchange rate targeting asa viable nominal anchor for monetary policy in theeuro area. This seems appropriate in view of the muchsmaller importance of international trade for the euroarea as a whole than for individual EU countries, andhence of the exchange rate as an influence on domesticinflation. The euro area will in this respect resemble therelatively closed economies of Japan and the UnitedStates, where monetary policy decisions are based pri-marily on domestic considerations (Figure 20). Formost countries participating in EMU, this will requirea significant change in the way monetary policy is


53

33Evidence on the relationship between central bank indepen-dence and inflation is discussed in the October 1996 WorldEconomic Outlook, Box 11, pp. 128–29.

34When the ECB is established, the terms of the initial appointeesto the Executive Board will range from four to eight years, so as tostagger subsequent appointments.

35In this respect, the ESCB will be more independent than eitherthe Bundesbank or the U.S. Federal Reserve, legislation governingwhich can be altered by the respective national legislatures.

36In this chapter “the Council” refers to the Council of theEuropean Union, also known as the Council of Ministers, whichrepresents all member governments and is the principal decision-making body of the EU. In Stage 3, for matters relating to the euroarea only, the voting rights of member states outside the euro areawill be suspended.

37There is provision for individual countries to negotiate a nar-rower band vis-à-vis the euro.


viewed, since in the ERM it has focused, except inGermany, on exchange rate objectives. On the otherhand, given that the ERM currencies have been float-ing collectively against other currencies, the magnitudeof the regime shift for the euro area as a whole will ineffect be smaller than may appear.

Two options—targeting inflation and targeting amonetary aggregate—have been left open, and theEMI has made arrangements to ensure that the techni-cal infrastructure is in place to conduct monetary pol-icy on the basis of either of these frameworks.38

Having a monetary aggregate as an intermediate tar-get might in principle provide an objectively measur-able means whereby the appropriateness of monetarypolicy, and effects of the ECB’s actions, could be mon-itored by the monetary authority and the public on atimely basis. With an inflation target, by contrast, thereis a substantial lag before the actual effects of policyon the targeted variable can be seen; in the meantime,assessments would depend on inflation forecastswhose reliability may be questioned, especially until atrack record is established. A monetary target wouldalso provide the ECB with the advantage of continuitywith the framework used by the dominant central bankof the ERM, the Bundesbank, and it might thereby fa-cilitate the transfer of credibility. However, monetarytargeting is the exception rather than the rule amongthe central banks of advanced economies. In the UnitedStates and most other advanced economies, the stabil-ity of demand-for-money relationships that is neededfor successful monetary targeting has been questionedsince such targeting began in the mid-1970s, and inmost countries the rapid pace of financial innova-tion subsequently has led to the conclusion that mone-tary aggregates cannot on their own provide an ade-quate guide for monetary policy. Even in Germanythere appears to be less empirical basis for anchoringpolicy to a money target now than in the past.39

54

38The main operating instrument of the ESCB will be a short-terminterest rate, and open market operations are expected to be con-ducted through repurchase agreements. The NCBs will play an im-portant role in implementing open market operations. The ESCBwill also have standing facilities—a lending and a deposit rate—andreserve requirements are also expected to figure among the set of instruments available to the ESCB. See The Single MonetaryPolicy in Stage Three: Specification of the Operational Framework(Frankfurt: European Monetary Institute, January 1997).

39See J. Von Hagen, “Inflation and Monetary Targeting inGermany,” in Inflation Targets, ed. by Leonardo Leidermanand Lars E.O. Svensson (London: Centre for Economic PolicyResearch, 1995), pp. 107–121; and Richard Clarida and MarkGertler, “How the Bundesbank Conducts Monetary Policy,” NBERWorking Paper No. 5581 (Cambridge, Massachusetts: NationalBureau of Economic Research, May 1996), for discussion of the sta-bility of demand for money relationships in Germany. Benjamin M.Friedman, “The Rise and Fall of Money Growth Targets asGuidelines for U.S. Monetary Policy,” NBER Working Paper No.5465 (Cambridge, Massachusetts: National Bureau of EconomicResearch, February 1996), discusses the experience with monetarytargets in the United States.

Figure 20. Selected Advanced Economies:Openness1

(In percent of GDP)

1Half the sum of exports and imports of goods and services.

France

Germany

Italy

United Kingdom

United States

Japan

0 5 10 15 20 25

European Union includingintra-EU trade

European Union excludingintra-EU trade

The openness to trade of the European Union (EU) as a whole ismarkedly less than that of its member countries, and comparable withthat of the United States and Japan.

Indeed, the emphasis placed on monetary aggregatesby the Bundesbank should not be exaggerated. TheBundesbank normally interprets developments inmonetary aggregates in light of a variety of other indi-cators, and in fact it has missed its monetary targetabout half the time over the past twenty years, relyingon its inflation performance above all else to sustainits credibility.

While some recent studies have found evidencethat relationships linking monetary aggregates toeconomic activity and inflation are more stable andpredictable when money is aggregated across EUcountries, the significant restructuring of financialmarkets that will accompany the introduction of theeuro, and the likely initial instability of demand forthe new currency itself, seem almost bound to makesuch relationships unreliable, at least for a number ofyears.40

Apart from concerns relating to the stability of thedemand for money, the case for inflation targeting isstrengthened by the recent experience of a number ofcountries that have targeted inflation with some suc-cess after abandoning their exchange rate and mone-tary anchors. A significant challenge for the ECB inadopting an inflation targeting framework would be indealing with credibility problems that could arise inthe initial stages of EMU, given the lags between pol-icy action and the inflation outcome. Moreover, theregime shift is likely to increase, at least for a time, un-certainties about methods of forecasting inflation andof predicting the effects on inflation of changes in thesettings of monetary policy instruments—methodsthat are essential to the implementation of a fully spec-ified inflation targeting strategy.

As a practical matter, therefore, it seems unlikelythat the ECB will be in a position to rely solely on apure, fully specified framework, whether monetarytargeting or inflation targeting, when EMU begins.41 It

is likely to have to use a high degree of discretion inits conduct of monetary policy, including by monitor-ing a variety of indicators, in which monetary aggre-gates would be likely to play a significant role, butwhich would also include indicators of movements inprices, wages, economic activity, the exchange valueof the euro, and developments in asset markets gener-ally. This would not be unusual; in fact, it is how mostcentral banks, including the Bundesbank, operate.42

The present subdued inflation in the EU area shouldprove helpful in allowing the ECB to pursue such amonetary policy while building its reputation. It will,however, be important for the ECB to make its deci-sions transparent, as it seeks to establish its credibilityfully: for instance, it could publish inflation reportsthat would be forward looking, but which would alsocontain explanations of why the ECB took the deci-sions it did, and why its assessment of the requiredmonetary stance may have changed from one period toanother.43

Another issue regarding the prospects for successfuloperation of a common monetary policy in EMU isthe extent to which there will be differences amongcountries in how changes in monetary policy aretransmitted to the real economy. This transmissionmechanism may vary because of institutional differ-ences or differences in the process of financial inter-mediation—for instance, differences in the relativeimportance of floating-interest-rate versus fixed-ratedebt instruments, in the importance of bank lendingrelative to other forms of financing, and in householdnorms regarding indebtedness.44 Consequently, outputand inflation in some countries could be more sensi-tive to changes in short-term interest rates than inothers. The evidence indicates that there have beendifferences in the response of activity to monetary pol-icy across the EU countries, but that they are not assignificant as to suggest that substantial problemswould arise in the operation of a common monetary


55

40The greater stability of money demand relationships when ag-gregated across EU countries appears to be related to the internal-ization of currency substitution shocks to which individual countriesin the EU have been subject as part of the process of deepeningEuropean integration. For a more detailed discussion of this issue,see Carlo Monticelli and Luca Papi, European Integration,Monetary Coordination, and the Demand for Money (Oxford:Clarendon Press, 1996). See also Marcel Cassard, Timothy D. Lane,and Paul R. Masson, “Core ERM Money Demand and Effects onInflation,” Manchester School of Economic and Social Studies,Vol. 65 (January 1997), pp. 1–24; and Ramana Ramaswamy,“Monetary Frameworks: Is There a Preferred Option for theEuropean Central Bank?” IMF Paper on Policy Analysis andAssessment 97/6 (June 1997).

41Monetary targeting and inflation targeting, although differentmonetary policy strategies, share common ground. Indeed the EMIand others have argued that in practice the similarities are greaterthan the differences. Both serve the same objective, and both are for-ward looking. Furthermore, monetary targets are based partly on as-sumptions or objectives for inflation, while monitoring monetaryaggregates provides important information in the implementation ofinflation targets.

42See October 1996 World Economic Outlook, Chapter III. VonHagen, “Inflation and Monetary Targeting in Germany”; Claridaand Gertler, “How the Bundesbank Conducts Monetary Policy”; andBen S. Bernanke and I. Mihov, “What Does the BundesbankTarget?” European Economic Review, Vol. 41 (June 1997),pp. 1025–53, all argue that the Bundesbank has in practice followeda monetary policy rule similar to the Taylor rule, whereby adjust-ment in interest rates depends on the output gap and deviations ofinflation from its target rate.

43David Begg, “The Design of EMU,” IMF Working Paper 97/99(August 1997), discusses how a central bank’s credibility evolvesthrough interaction between the legal provisions for its inde-pendence and the success of its policy actions in building itsreputation.

44For instance, the United Kingdom has a high proportion of ad-justable-rate mortgage loans, while France has a high proportion offixed-rate loans. See Claudio E.V. Borio, “The Structure of Credit tothe Nongovernment Sector and the Transmission Mechanism ofMonetary Policy: A Cross-Country Comparison,” BIS WorkingPaper No. 24 (Basle: Bank for International Settlements, 1995), fora discussion of differences in credit markets in the EU.


policy.45 Furthermore, the operation of the commonmonetary policy and the creation of a single financialmarket may be expected over time to lead to greater har-monization in the transmission mechanism in the euroarea. Thus, any problems relating to differences in thetransmission mechanism are most likely to be transitory.

Nevertheless, differences in the transmission mecha-nism are likely to add to the challenges facing theESCB in its early years, as it conducts monetary policyin the face of what may be a particularly unstable de-mand for money and an uncertain inflation forecastingmethodology, as it learns to take policy decisions ef-fectively and in a timely way as a multinational centralbank, and as it seeks to establish its credibility fully.

Fiscal Policy

Unlike most monetary unions, the euro area will nothave a central fiscal authority. There is a budget for theEU as a whole, but it is relatively small. Spendingamounts to only a little over 1 percent of GDP, devotedmostly to the common agricultural policy and the struc-tural funds, and deficit financing is prohibited. In viewof the principle of subsidiarity46 and the fact that theEU is not likely in the foreseeable future to evolve intoa federation, the EU budget can be expected to remainrelatively small, and there is little reason to expect it tobecome a significant instrument of macroeconomicpolicy. Thus, budgetary decisions in the euro area willremain almost exclusively the province of memberstates, albeit subject to surveillance by the Union as awhole in the context of the requirements set out in theMaastricht Treaty and subsequent agreements.

The key fiscal concern of the Maastricht frameworkwas to avoid excessive budgetary imbalances. MostEU members suffered from chronic fiscal problems, asindicated by rising government debt ratios, public sec-tors of unprecedented size, and heavy and growing taxburdens. Most countries also faced the prospect ofworsening budgetary positions in the medium to longterm as a result of demographic developments.Moreover, there was increasing awareness of the neg-ative effects of fiscal imbalances on medium-termgrowth, including through higher real interest rates. Itwas widely understood that these problems needed tobe tackled anyway, irrespective of the EMU project. Itwas also viewed as particularly important for the suc-cess of EMU to avoid negative spillovers from the fis-cal policies of individual states to other members,which might weaken the credibility of the anti-infla-tionary thrust of the common monetary policy. In par-ticular, there were concerns both that excessive na-tional debts could, in EMU, produce pressure forundue relaxation of monetary policy, and perhaps evenfor a bailout of an overindebted government, and thatlarge deficits would result in a strained policy mix andcomplicate the task of monetary policymakers.

One option would have been to leave it solely to fi-nancial markets to discourage imprudent fiscal poli-cies by penalizing with larger borrowing spreads gov-ernments that followed such policies. But this was notseen to be a sufficiently reliable solution. While thereis empirical evidence that risk premiums increase withdebt levels, it is questionable in many cases whetherthey have risen quickly enough to act as a deterrent;and clearly, in Europe, substantial premiums have notin practice discouraged some governments from build-ing considerable indebtedness, albeit in a contextwhere monetization and exchange rate depreciation re-mained a possibility.

To address these concerns, the treaty introduced theexcessive deficit exercise, an annual examination offiscal positions whereby a country is deemed to be inexcessive deficit if it violates either of two criteria re-lating to the deficit and debt of the general government.These criteria provide reference values for the generalgovernment deficit of 3 percent of GDP and for generalgovernment gross debt of 60 percent of GDP, to beused in judging whether there is sufficient fiscal disci-pline. There is some scope for allowing performancethat is in breach of these reference values, notablywhere the deficit is elevated owing to exceptional andtemporary circumstances or the debt is declining at asufficiently fast pace (see note to Table 11).47 The cri-

56

45Fernando Barran, Virginie Coudert, and Benoit Mojon, “TheTransmission of Monetary Policy in the European Countries,” CEPIIWorking Paper No. 96-03 (Paris: Centres d’études prospectives etd’information internationales, February 1996), find that the standardeffects of monetary policy have been operative in most EU countriesand that the impact of monetary policy on GDP has also been fairlysimilar in terms of lags, though not in magnitude. Erik Britton andJohn Whitley, “Comparing the Monetary Transmission Mechanism inFrance, Germany, and the United Kingdom: Some Issues andResults,” Quarterly Bulletin, Bank of England, Vol. 27 (May 1997),pp. 152–62, obtain results suggesting no marked difference betweenthe three economies in the response of output or inflation, but alsorefer to these results as inconclusive. Using a larger set of EU coun-tries, Ramana Ramaswamy and Torsten Sloek, “The MonetaryTransmission Mechanism in Europe: How Important Are the Differ-ences?” IMF Working Paper (forthcoming), find that while the impactof monetary shocks on GDP differs somewhat across EU countries,the differences do not seem very large. They find that EU countriesfall broadly into two groups, in one of which (Austria, Belgium,Finland, Germany, the Netherlands, and the United Kingdom) activ-ity responds relatively slowly but ultimately by more than in the othergroup (Denmark, France, Italy, Portugal, Spain, and Sweden).

46According to the principle of subsidiarity established in theMaastricht Treaty (Article 3b), a government function should be as-signed to the community “only if . . . the objective of the proposedaction cannot be sufficiently achieved by the Member States and cantherefore, by reason of scale or effects of the proposed action, bebetter achieved by the community.”

47In the most recent application of the excessive deficit exercise,Denmark, Ireland, and the Netherlands each passed the examinationdespite having debt levels in excess of the reference value, as theirdebt ratios were judged to be falling sufficiently quickly. The othertwo countries that passed, Finland and Luxembourg, complied withthe reference values for both the deficit and debt, although revisedfigures for Finland published more recently show the deficit as

teria that the treaty provides for the assessment of acountry’s readiness to participate in monetary unioninclude that it not have an excessive deficit. The treatyalso requires that, once monetary union commences,participating countries avoid excessive deficits or besubject, potentially, to financial sanctions.

At the June 1997 meeting in Amsterdam of theEuropean Council, the EU member states concludedtheir negotiation of a Stability and Growth Pact, whichmakes more precise how surveillance of fiscal posi-tions will be carried out in Stage 3 (see Box 3). It setsa specific time frame for the various steps of the ex-cessive deficit procedure, introduces clearer guidanceas to when a deficit larger than 3 percent of GDPmight not be considered excessive because of excep-tional and temporary circumstances, and establishespresumptions as to when financial sanctions would beimposed and what the size of those sanctions would


57

having exceeded its reference value slightly in 1996. There has beenno case to date where an exemption has been given for a deviation,known at the time of the exercise, from the deficit reference value.The Stability and Growth Pact makes more specific the exceptionalgrounds on which such a deviation might be justified.

Table 11. European Union: Convergence Indicators for 1996, 1997, and 1998(In percent)

Consumer General Government Gross Government Long-TermPrice Inflation Balance/GDP Debt/GDP2 Interest Rates3________________________ __________________________________ _________________________ _____________

1996 1997 1998 1996 1997 19971 1998 1996 1997 1998 August 1997

Germany 1.5 1.9 2.3 –3.6 –3.1 –3.0 –2.9 60.7 62.2 62.7 5.7France 2.0 1.1 1.3 –4.1 –3.2 –3.0 –3.2 55.4 57.7 59.2 5.6Italy 3.9 1.8 2.1 –6.7 –3.2 –3.0 –3.0 123.8 122.9 121.2 6.6United Kingdom4 2.9 2.6 2.7 –4.7 –2.0 . . . –0.6 53.8 54.5 52.4 7.1

Spain 3.5 2.0 2.2 –4.4 –3.0 –3.0 –2.6 69.8 69.0 68.2 6.2Netherlands 2.1 2.3 2.3 –2.3 –2.1 . . . –1.8 78.0 73.6 71.2 5.5Belgium 2.1 1.6 1.9 –3.2 –2.8 –2.9 –2.6 127.4 125.1 122.8 5.7Sweden 0.8 1.0 2.0 –2.5 –2.1 –2.1 — 77.7 77.1 73.9 6.5Austria 1.9 1.5 1.6 –3.9 –2.5 –3.0 –2.5 70.0 68.0 67.6 5.7Denmark 2.2 2.5 2.6 –1.4 0.5 0.7 0.5 69.9 66.4 63.2 6.2Finland 0.6 1.3 2.3 –3.1 –1.9 –1.4 –0.4 58.8 59.4 57.9 5.8Greece5 8.2 5.7 4.7 –7.4 –4.7 –4.2 –4.1 111.8 108.0 104.2 9.6Portugal 3.1 2.2 2.3 –4.0 –2.9 –2.9 –2.9 66.0 62.9 61.7 6.3Ireland 1.6 1.7 2.1 –0.9 –0.8 –1.5 –0.8 72.8 67.5 65.0 6.3Luxembourg 1.8 2.0 2.0 –0.1 –0.1 . . . –0.1 5.9 5.7 5.5 6.0

All EU6 2.5 1.9 2.2 –4.3 –2.8 . . . –2.3 73.9 74.0 73.2 6.2

Reference value7 2.5 2.6 3.1 –3.0 –3.0 . . . –3.0 60.0 60.0 60.0 8.0

Sources: National sources; and IMF staff projections.Note: The table shows IMF staff estimates of the convergence criteria mentioned in the Maastricht Treaty, except for the exchange rate. The

data and projections shown for consumer price inflation are based on national statistics rather than on the harmonized consumer price indicesbeing constructed by Eurostat that will be used in applying the Maastricht criteria. The three relevant convergence criteria are (1) consumerprice inflation must not exceed that of the three best performing countries by more than 1!/2 percentage points; (2) interest rates on long-termgovernment securities must not be more than 2 points higher than those in the same three member states; and (3) the financial position mustbe sustainable. In particular, the general government deficit should be at or below the reference value of 3 percent of GDP. If not, it shouldhave declined substantially and continuously and reached a level close to the reference value, or the excess over the reference value should betemporary and exceptional. The gross debt of general government should be at or below 60 percent of GDP or, if not, the debt ratio should besufficiently diminishing and approaching the 60 percent value at a satisfactory pace. The exchange rate criterion is that the currency must havebeen held without severe tensions within the normal fluctuation margins of the ERM for at least two years, in particular without a devaluationat the initiative of the member state in question.

1Official targets or intentions. The IMF staff’s fiscal projections shown in the preceding column are in some cases based on different growth,inflation, or interest rate assumptions from those used by national authorities and do not take into account further consolidation measures thatare planned by EU governments in accordance with their convergence programs but which have not yet been announced. See Box 1 for theIMF staff’s fiscal assumptions.

2Debt data refer to end of year. They relate to general government but may not be consistent with the definition agreed at Maastricht. Forthe United Kingdom, general government consolidated debt evaluated at the end of March.

3Ten-year government bond yield or nearest maturity.4Retail price index excluding mortgage interest.5Long-term interest rate is 12-month treasury bill rate.6Average weighted by GDP shares, based on the purchasing-power-parity (PPP) valuation of country GDPs for consumer price index, gen-

eral government balances, and debt.7The treaty is not specific as to what methodology should be used to calculate reference values for inflation and the interest rate beyond not-

ing that they should be based on the three lowest inflation countries or a subset of them. For illustrative purposes, a simple average for thethree countries is used in calculating the reference values. The reference value for long-term interest rates in August 1997 is based on yieldsin the three countries with the lowest projected inflation rates for 1997 as a whole.


58

The Stability and Growth Pact clarifies how the sur-veillance of national fiscal policies will be carried out inStage 3 of EMU, pursuant to the requirements of theMaastricht Treaty.1 The treaty pays particular attention tothe avoidance of “excessive deficits” and allows for theimposition of financial sanctions in Stage 3 when a coun-try found to have an excessive deficit does not respondadequately to the advice of the Council of Ministers. Butthe treaty leaves some key questions open: What is meantby the exceptional and temporary circumstances in whichthe general government deficit can exceed 3 percent ofGDP without being judged excessive? What time frameis envisaged for the various steps in the excessive deficitprocedure? And in what circumstances would financialsanctions be invoked and what would be their size? Byaddressing these questions, the Stability and Growth Pactgives greater precision to how the excessive deficit pro-cedure will operate in the euro area. At the same time, thepact strengthens the Council’s surveillance of medium-term fiscal policies with a view to avoiding excessivedeficits, and focuses this surveillance on the need formedium-term fiscal positions to be close to balance or insurplus.

Operation of the Excessive Deficit Procedure in Light of the Pact

A general government deficit in excess of 3 percent ofGDP will be considered exceptional and temporary if itresults from an unusual event outside the control of themember state in question or from a severe economicdownturn, provided also that, should the unusual event orsevere downturn have passed, Commission projectionsfor the following year envisage the deficit falling back to3 percent or less. To avoid an excessive deficit finding,the deficit would also have to remain close to the refer-ence value. A decline in GDP of 2 percent in the year inquestion would as a rule be regarded as a severe down-turn. The pact allows member states to argue that asmaller output decline was exceptional, on the basis ofevidence such as the abruptness of the downturn or thecumulative loss of output relative to past trends.Countries have agreed not to claim exceptional circum-stances for annual output declines of less than #/4 of1 percent.

The excessive deficit exercise normally commenceswith the national authorities submitting by the beginningof March each year data on the prior year’s fiscal out-come.2 The Council would hand down any excessivedeficit findings, together with its advice, by the end of

May and would impose financial sanctions by the end ofthe year on countries judged not to have responded adequately to its advice (or subsequently on countriesthat failed to follow through on their initial response). To avoid sanctions, a country would normally be ex-pected to bring the deficit down to 3 percent of GDP orless in the year following that in which an excessivedeficit was identified. However, the pact allows theCouncil to set a longer adjustment period if there arespecial circumstances (which are not defined in thepact). Sanctions would initially take the form of nonre-munerated deposits, amounting to between 0.2 percentand 0.5 percent of GDP, depending on the size of thedeficit. The deposit would be returned if the excessivedeficit was corrected within two years; otherwise itwould be converted into a fine. Additional deposits (alsosubject to conversion into fines) may be required eachyear following the initial deposit until the excessivedeficit is corrected.

Strengthened Surveillance of Fiscal Positions

Under the pact, each country will aim for a medium-term fiscal position that is close to balance or in surplusso as to allow an adequate safety margin to avoid exces-sive deficits in the face of normal cyclical fluctuations.Countries will submit stability programs annually speci-fying how their medium-term fiscal goals will be real-ized. The Council will examine the initial programs andmay choose to review the later submissions. It will pro-vide advice where a program is judged inadequate, andprogram implementation will be monitored in the regularmultilateral surveillance exercises. This should normallyprovide countries with an early warning from the Councilif there is risk of an excessive deficit.

Implications for the Management of Fiscal Policy

While the pact reinforces the framework for fiscal dis-cipline, a question arises as to how it might constraincountercyclical fiscal policy. A number of factors, in ad-dition to any exercise of discretionary fiscal policy, arerelevant to this, including the sensitivity of the fiscal po-sition to the cycle; the underlying stance of fiscal policy;and the size and duration of cyclical fluctuations in theeconomy and the budget.

Estimates by IMF staff indicate that, on average in theEU, a 1 percent shortfall in output from potential worsensthe fiscal balance by 0.6 percent of GDP, with most ofthis effect occurring in the year of the shortfall. For somecountries, the impact of the cycle is larger; in Denmark,the Netherlands, Sweden, and the United Kingdom, theresponse parameter of the fiscal balance to GDP fluctua-tions is on the order of #/4 or higher. Where governmentfinances are in structural balance and the cyclical re-sponse parameter is of average size, the operation of au-tomatic stabilizers would accommodate an output gap of5 percent without the deficit exceeding 3 percent of GDP.In a country with a response parameter of #/4, an underly-

Box 3. The European Union’s Stability and Growth Pact

1The pact consists of two European Council Regulations—one on the excessive deficit procedure and the other on surveil-lance—which have the force of law, and a European CouncilResolution, which gives guidance to the Commission, theCouncil, and member states in applying the pact.

2The exercise can also address planned breaches of the refer-ence values.


59

ing surplus of about 1 percent of GDP would be neededto provide the same buffer.

Estimating the extent to which cyclical fluctuationsalone might in the past have caused breaches in theMaastricht reference value, even with underlying fiscalbalance, is complicated by uncertainty in the estimationof output gaps. A European Commission study foundonly three countries for which the largest cyclical deficithad exceeded 3 percent of GDP over 1961–96 (by a sig-nificant margin in Finland and Sweden, with a minor ex-cess in Luxembourg).3 The authors noted that theirmethodology tended to yield smaller estimates of outputgaps than that used by the IMF or the Organization forEconomic Cooperation and Development (OECD).Another study, using World Economic Outlook data for12 EU countries from the late 1970s to 1995, estimatedthat for most countries the standard deviation of thecyclical deficit was less than 2 percent of GDP.4However, the largest cyclical deficit exceeded 3 percentof GDP in five cases: for Denmark, the Netherlands, andthe United Kingdom, the excess was in the range of !/2–#/4of 1 percentage point, but it was considerably larger forFinland and Sweden. All these countries except Finlandare marked by above-average sensitivity of the fiscal po-sition to the cycle.

It is important to note that, under the standard proce-dures for the excessive deficit exercise, a country wouldnot incur financial sanctions (initially a nonremunerateddeposit) unless the deficit exceeded 3 percent of GDP fortwo consecutive years, with neither year classifiable as anexceptional and temporary circumstance, and the Councilhad concluded that the country had not taken effective ac-tion following its recommendations to eliminate the ex-cessive deficit.5 Even then, special circumstances couldallow an extended adjustment period without sanctions.The aforementioned Commission study, examining theactual changes in fiscal positions at times of recession orcyclical slowdown (including the effects of any discre-tionary measures), concluded that if countries had startedfrom balanced fiscal positions prior to the cyclical weak-ening, deficits in excess of 3 percent of GDP would havepersisted in the year after the recession in only 5 cases ofsevere recession (out of 24 cases of output decline greaterthan #/4 of 1 percent) and in 1 of mild recession.6 (In this

last case, the country had failed to reverse discretionarycountercyclical policies despite a strong recovery in thepostrecession year).

In general, it appears that the Stability and GrowthPact will not pose a great problem for the operation ofautomatic stabilizers if countries maintain balancedmedium-term (structural) fiscal positions, or small sur-pluses in the case of countries whose fiscal positions arecharacterized by above-average sensitivity to cyclicalfluctuations. This is not to deny that complications areinevitable in the implementation of the pact. Three canbe noted here. First, deep and protracted recessions arelikely to require recourse to the special circumstancesclause. For such recessions, there may also at times beneed for discretionary countercyclical policies, espe-cially in the case of asymmetric shocks, where monetarypolicy would not be able to provide support. Second, thespecification of the exceptional circumstances clause isless well suited to countries with relatively high trendoutput growth rates, which are less likely to experienceoutput declines during periods of cyclical weakening.Some recognition of this is implicit in allowing accumu-lated output losses to be taken into account in assessingwhether an output decline of between #/4 of 1 percent and2 percent is exceptional, but additional recognition maybe needed in implementing the special circumstancesprovision. Third, many countries at present fall short ofthe medium-term fiscal goal, placing them potentially indifficult positions if a cyclical weakening should occurearly in Stage 3. In a different vein, one needs also torecognize that uncertainties in the measurement of out-put gaps, inter alia, can make it difficult to achieve a par-ticular goal for the structural balance with a high degreeof precision.

In any event, concerns about the potentially constrain-ing effects on countries’ abilities to pursue countercycli-cal fiscal policies need to be put into the perspective ofthe constraints imposed by large deficits in most EUcountries over much of the past twenty-five years. Fromthis viewpoint, the increased discipline involved in ad-hering to the pact may well permit a greater stabilizingrole for fiscal policy than has been possible in most ofthese countries for many years. At the same time, theachievement of a high degree of price stability, togetherwith the focus of the European System of Central Bankson conditions throughout the euro area (in contrast to thedominant influence of German economic conditions onmonetary policy in countries participating in the ex-change rate mechanism (ERM)), should allow monetarypolicy to play a greater stabilizing role than in the past forthe euro area as a whole.

3M. Buti, D. Franco, and H. Ongena, Budgetary Policies dur-ing Recessions: Retrospective Application of the Stability andGrowth Pact to the Post-War Period, Economic Paper No. 121,European Commission (May 1997).

4Paul R. Masson, “Fiscal Dimensions of EMU,” EconomicJournal, Vol. 106 (July 1996), pp. 996–1004. The EU countriesexcluded from the sample were Greece, Luxembourg, andPortugal.

5An expedited procedure could be used in the case of adeliberately planned deficit that the Council judged to beexcessive.

6Of course, when a recession starts from a position of over-heating, a balanced structural fiscal position would be asso-

ciated with an actual prerecession surplus. Additional calcu-lations by the authors (not included in the published paper)show that, even when starting from an actual prerecession sur-plus corresponding in size to the cyclical budget component, thedeficit still remained above 3 percent of GDP in the year afterthe recession in four out of the five cases of severe recession.


be. The agreement also calls for fiscal positions to beclose to balance or in surplus in normal times so as toallow room for automatic stabilizers to operate overthe cycle.48 In support of this goal, surveillance overmedium-term fiscal policies is being strengthened toprovide an early warning system in which recommen-dations for corrective action are made by the Councilwell before the deficit exceeds the reference value.

The framework for fiscal policy in Stage 3 thereforehas a clear emphasis on avoiding excessive deficits.While this is understandable in the context of fiscalperformance over the past two decades, it does raisequestions as to whether there is adequate scope for theoperation of automatic stabilizers and as to whethersufficient weight is placed on ensuring an appropriatefiscal policy stance for the union as a whole.

Concerning the operation of automatic stabilizers, akey question is whether countries keep their underly-ing fiscal positions close to balance or in surplus. Ifthey accomplish this, historical experience suggeststhat the reference values of the Stability and GrowthPact would allow adequate room for stabilization inmost circumstances (see Box 3). Deep or protractedrecessions would cause greater difficulties, but herethe pact allows flexibility. In particular, a deficit largerthan 3 percent of GDP in the year of a large output de-cline would not be deemed excessive, because of ex-ceptional circumstances, provided it was temporaryand the excess was deemed small. Moreover, if adeficit is deemed to be excessive, the Council can tai-lor to the circumstances the speed at which the deficitshould be corrected if financial sanctions are to beavoided. Thus while the pact requires that normally anexcessive deficit be corrected no later than the yearafter that in which it is identified, the Council may ex-tend the adjustment period to take account of “special”circumstances. While the meaning of “special” in thiscontext has not been elaborated, it would seem appro-priate that it should encompass situations where an ex-cessive deficit has emerged even though the countryconcerned has achieved underlying balance or surplusconsistent with the medium-term objectives of thepact; in these circumstances, an excessive deficitwould usually reflect a deep or protracted recession.

A particularly difficult case—though not one envis-aged in the IMF staff’s projections—would be a weak-ening cyclical position soon after the start of EMU,when many countries may not have moved closeenough to the medium-term goal for fiscal balances.Applying the full force of the Stability and GrowthPact to such a situation might generate significant pub-lic and political discontent at a time when unemploy-ment would still be high. On the other hand, to short-

circuit the operation of the pact might erode the cred-ibility of the commitment of EU governments to fiscaldiscipline.

Regarding the need to ensure an appropriate stanceof fiscal policy in the euro area as a whole—and hencean appropriate mix of fiscal and monetary policies—inthe absence of a substantive central fiscal function, ef-forts will need to rely on policy coordination. Thetreaty provides for the Council to issue annually a setof broad guidelines for the economic policies of EUmember states; this has been done since 1993. Buthow effective these guidelines could be in calibratingthe euro area’s fiscal stance is not clear. To date, theguidance has largely paralleled that of the excessivedeficit exercise, and it remains to be seen how muchindependent influence the guidelines might have onpolicies, particularly as adherence to them is volun-tary. Moreover, if the euro area is significantly smallerthan the union as a whole, addressing the fiscal needsof the euro area within these guidelines may be a com-plex task, as the guidelines are adopted by a qualifiedmajority of the Council as a whole. At its meeting inAmsterdam in June 1997, the European Council re-quested the Council and the Commission to examinehow to improve the processes of coordination of eco-nomic policies in EMU, and to report to its meeting inLuxembourg in December 1997.

Adjustment Without Exchange Rates

With its monetary and fiscal policy frameworks de-signed to foster price stability and fiscal sustainability,a key question is how well equipped the euro area willbe to absorb and adjust to economic shocks. For thelonger-term harmony of the euro area and the successof EMU, it is important that the benefits of monetaryunion not be clouded by perceptions in individualcountries either that the stabilization of prices and out-put in the national economy is weaker than prior tomonetary union or that the effectiveness of stabiliza-tion policies varies across the Union. In examiningthese issues, two benchmarks suggest themselves: theresponse that would have occurred under the regimeexisting prior to monetary union and the response thatis achieved in other large monetary areas, such as theUnited States. A number of different types of shocksalso need to be considered, including symmetricshocks, which have a similar impact throughout themonetary area, versus asymmetric shocks, which havedifferential effects across countries; and temporaryversus permanent shocks.

To the extent that shocks tend to be symmetricallydistributed across EU countries, the loss of the possi-bility of independent exchange rate or monetary policyaction at the national level should not impede the abil-ity of countries to adjust to them. Moreover, the con-siderations outlined above suggest that provided coun-tries adhere to the medium-term goals of the Stability

60

48The importance of strengthening fiscal positions over themedium term, perhaps to positions close to balance, was empha-sized in the first broad guidelines for economic policies issued bythe Council in 1993.

and Growth Pact, the degree of shock absorption pro-vided by fiscal policy in the euro area should not di-minish, and may indeed be greater than has been avail-able for some time.

There are nevertheless a number of factors thatmay impede areawide adjustment even to symmetricshocks. One relates to international differences inthe transmission mechanism for monetary policy.Although, as discussed earlier, these differences do notappear to be worrisome, it will take some time for thetransmission mechanism to develop the degree of uni-formity that facilitates the work of policymakers in theUnited States, for example.49 A second question re-lates to the external leakages of fiscal stabilization,which can be quite large, especially from the smaller,more open EU economies. Thus, following a shockthat results in a general economic downturn, in the ab-sence of policy coordination the amount of fiscal sta-bilization done in aggregate might be suboptimal. Theinadequacy of fiscal stabilization might be especiallypronounced if the shock was of such a magnitude as tothreaten a widespread breach of the Stability andGrowth Pact, as this would represent an additional dis-incentive to allow automatic fiscal stabilizers to workor to engage in discretionary fiscal action. These is-sues are less important in other large monetary areaswhere there is an important central stabilization role,and it remains to be seen how quickly the developmentof policy coordination in the euro area can compensatein this regard. Finally, countries’ abilities to adjust tosymmetric shocks will depend on the respectiveeconomies’ structural flexibility. Thus, the greater theflexibility of markets in all the national economies, theless likely it is that significant deviations in perfor-mance will develop.

In gauging the impact of EMU on countries’ abili-ties to absorb and adjust to shocks, a more critical setof issues revolves around asymmetric disturbances. Itis here that common policy responses are more likelyto be ineffective and the loss of independent monetaryand exchange rate policies at the national level islikely to be more constraining. The nature and magni-tude of the problems that could arise will depend interalia on the incidence of asymmetric shocks, onwhether shocks are temporary or permanent, on thescope for fiscal policy to cushion shocks and ease ad-justment, and on the availability of other adjustmentmechanisms, particularly through the flexibility of in-ternal markets.

The incidence of asymmetric shocks in the past ap-pears to have been more pronounced in the EU than inthe United States, but for a “core” group of EU coun-

tries, the difference does not appear to have been sub-stantial (Box 4). In fact, the incidence of asymmetricdemand shocks in the EU may diminish after the eurois introduced, as asymmetric developments induced bynational monetary policies or exchange rate move-ments within the euro area will no longer be a factor;and the discipline imposed by the Stability andGrowth Pact should reduce the prevalence of fiscalshocks in the medium term.50 On the other hand, thistendency toward a lower incidence of asymmetricshocks may be offset if member economies becomemore specialized in production.

The nature of the policy challenge presented byshocks depends on whether they are temporary or per-manent. In the case of temporary shocks, the challengeseems unlikely to be more difficult with EMU than inthe ERM.51 Monetary policy and exchange rate flexi-bility have typically not been used in the ERM asmeans of adjustment to such shocks,52 and, as notedearlier, the scope for fiscal stabilization should not beany less than in the past, if countries adhere to themedium-term strategy of the Stability and Growth Pact.

An important question, nevertheless, is whether eco-nomic stabilization could be strengthened if part of thisfunction were assigned to a central fiscal institution.The experience of the United States and other federalsystems with diverse economic regions points to thepotential effectiveness of a central system of automaticstabilizers.53 On the other hand, there seems to be noreason in principle why a similar degree of effectivestabilization could not be carried out by national gov-ernments in the euro area, as the principal stabilizationtools used by central governments in federal systemswill be the prerogative of national governments inEMU. Indeed, in the past, the stabilization performedby national governments in the EU seems to have beensimilar to that provided by the U.S. federal system.


61

49Monetary policy is also likely to have asymmetric effects on fis-cal balances, depending in particular on the size and term structureof government debt and on whether it is at fixed or floating rates.This may carry implications for the ability of different countries toadhere to the Stability and Growth Pact.

50In the short term, however, countries’ differing efforts to com-ply with the medium-term fiscal goals may well be a source ofasymmetric shocks.

51Of course, independent of EMU, it is often difficult to determinethe duration of a shock when it first occurs.

52The exception here is Germany, which will no longer be able tomarshall monetary policy to deal with shocks that particularly affectitself.

53It has been estimated by Xavier Sala-i-Martin and Jeffrey Sachs,“Fiscal Federalism and Optimum Currency Areas: Evidence forEurope from the United States,” in Establishing A Central Bank:Issues in Europe and Lessons from the U.S., ed. by MatthewCanzoneri, Vittorio Grilli, and Paul R. Masson (Cambridge:Cambridge University Press, 1992), pp. 195–219, that federal taxesand transfers in the United States may offset almost one-third of thedeviations of per capita income from the national average. This es-timate may, however, exaggerate the stabilization role of federal fis-cal policy by conflating it with its interregional income redistribu-tion role. Other studies that have taken account of this problem havearrived at lower estimates, in the 10–20 percent range. See T.Bayoumi and P.R. Masson, “Fiscal Flows in the United States andCanada: Lessons for Monetary Union in Europe,” EuropeanEconomic Review, Vol. 39 (February 1995).


62

A natural benchmark for assessing the likely magni-tude and effects of asymmetric shocks in the euro area isthe experience of the United States in this regard. TheUnited States and the EU are roughly comparable interms of population, economic size as measured by GDP,and openness to trade. Since the United States has been acurrency area for a long time, a comparison of the inci-dence of shocks between the United States and the EUshould provide an indication of some of the challengesthat the currency union in Europe may face.1

One way of gauging the likely effects of asymmetricshocks in the euro area is to compare the correlation ofoutput fluctuations across EU countries with that forU.S. regions. The table reports the correlation coeffi-cients of output growth fluctuations between westGermany and other EU countries, and between theMideast region of the United States and other U.S.regions, over 1964–90.2 Fluctuations in output growth,in general, were more highly correlated across U.S. re-gions than across EU countries over the period.Nevertheless, in both cases there appears to have beena “core” (in the EU: west Germany, France, the Nether-lands, Belgium, and Austria; in the United States: theMideast, New England, Great Lakes, and Southeaststates) where fluctuations in output growth tended to berelatively highly correlated.3 Output growth fluctuationsin the “peripheral” regions were much less highly corre-lated with their respective anchor areas in both theUnited States and the EU. An interesting finding shownin the table is that even though the correlation coeffi-cients of output growth fluctuations across the EU coreare lower than those across the U.S. core, the differencesare not very large: they are considerably smaller, in par-

ticular, than the differences within the United States be-tween the core and the periphery.

These results tend to be reinforced by analyses that de-compose the aggregate disturbances into supply and de-mand shocks. Supply shocks refer to unanticipated distur-bances, for instance, to technology or commodity prices,and tend to have relatively long-lasting effects on outputand prices. Examples of supply shocks are the oil price in-creases of the 1970s. Demand shocks arise owing to suchdevelopments as unanticipated disturbances to businessbehavior, consumer preferences, or export demand, aswell as significant changes in monetary and fiscal poli-cies. Demand shocks tend, in general, to have less long-lasting effects on output, but more permanent effects onthe level of prices. German unification, which changed themix of monetary and fiscal policies in Europe, is an ex-ample of an asymmetric demand shock. Following themethod used in the study by Bayoumi and Eichengreen,4

Box 4. Asymmetric Shocks: European Union and the United States

Correlation Coefficient of Output Growth withAnchor Areas, 1964–90

U.S. statesMideast 1.00New England 0.92Great Lakes 0.87Southeast 0.84Plains 0.74Far West 0.68Southwest 0.34Rocky Mountains 0.18

EU countriesWest Germany 1.00Netherlands 0.77France 0.71Belgium 0.71Austria 0.70Denmark 0.61Greece 0.61Spain 0.54Portugal 0.49Ireland 0.48Italy 0.47Sweden 0.43Finland 0.41United Kingdom 0.19

Note: The correlation coefficients of output growth with an-chor areas in the United States were found to be stable with re-spect to the choice of different time periods. The correlation co-efficients in the EU “core” were also found to be stable whendifferent time periods were chosen. However, the correlations ofoutput growth between the anchor area and Finland, Sweden,and the United Kingdom were found to vary with the choice oftime period. For instance, choosing 1973–90 as the reference pe-riod increased the correlation coefficients for the UnitedKingdom but reduced them significantly for Finland andSweden. Nevertheless, these three countries continued to be partof the “periphery” irrespective of the time period chosen. 1A number of studies have attempted to compare the incidence

of shocks and adjustment mechanisms in the United States withthose in the EU. See T. Bayoumi and B. Eichengreen, “ShockingAspects of European Monetary Unification,” in Adjustment andGrowth in the European Monetary Union, ed. by FranciscoTorres and Francesco Giavazzi (Cambridge, England; NewYork: Cambridge University Press, 1993); and T. Bayoumi andE. Prasad, “Currency Unions, Economic Fluctuations andAdjustment: Some New Empirical Evidence,” Staff Papers, IMF(March 1997), pp. 36–58.

2West Germany offers a natural standard for comparison in thecase of the EU for the period considered, being the largest econ-omy, and having played the anchor role in the ERM. TheMideast region of the United States, being the largest economi-cally, provides a corresponding standard for comparisons acrossthe U.S. regions. The choice of 1990 as the end-point for theanalysis has been motivated by the need to leave out the effectsof German unification on the correlations of output fluctuationsacross the EU countries. Using the average rates of growth forthe United States and the EU instead of those for the Mideast re-gion and west Germany do not alter the results significantly.These issues are discussed in more detail in Bayoumi andEichengreen, “Shocking Aspects.”

3The regional groupings of U.S. states are as defined by theBureau of Economic Analysis of the U.S. Department ofCommerce. 4Bayoumi and Eichengreen, “Shocking Aspects.”


63

the figure plots the correlations of the fluctuations inoutput arising from demand and supply shocks of thedifferent countries in the EU with those in Germany, andthe corresponding correlations for the different regionsof the United States with those in the Mideast statesover 1962–88. Fluctuations in activity due to bothsupply and demand shocks tended to be more highlycorrelated across U.S. regions than across EU coun-tries. However, as with aggregate disturbances, fluc-tuations in output due to supply shocks in the coresof both the United States and the EU were more highlycorrelated with their respective anchor areas than wasthe case with fluctuations in output due to supplyshocks in the respective peripheral regions. The same

is true of fluctuations in activity arising from demandshocks.

The broad picture that emerges from this analysis is thatthe asymmetric effects of shocks were relatively morepronounced in the EU than in the United States during1962–88. For the core group of EU countries, however,even though the asymmetric effects on output of shockswere larger than in the United States, the magnitudes ofthe differences do not appear to be substantial. However,as far as judgments about the economic viability of theeuro area are concerned, past instances of asymmetric ef-fects of shocks can provide only a partial guide to whatthe future holds. The introduction of the single currencywill constitute a major regime change that will havesignificant implications for the pattern of likely shocks inthe euro area. The incidence of asymmetric demandshocks, in particular, is likely to diminish after the singlecurrency is introduced, because the countries in the euroarea will not be able to pursue independent monetary poli-cies, and exchange rate fluctuations within the area will beeliminated.

The impact that the regime change will have on thepattern of supply shocks is harder to predict. The intro-duction of the single currency may lead to more special-ization in manufacturing among countries in the EU thanexists currently. Consequently, there is some potentialfor an increase in the incidence of asymmetric supplyshocks (as well as asymmetric demand shocks) dueto this factor after the single currency is introduced.5However, any such locational changes in manufacturingproduction are bound to take place only gradually.Moreover, research indicates that the United States is re-gionally more specialized than the EU only in the pro-duction of manufactures; there are no significant differ-ences in regional specialization as far as the rest of theeconomy is concerned.6 In sum, while increased special-ization in manufacturing may tend to increase asymmet-ric shocks in the euro area, it is likely to be offset by thereduction in asymmetric demand shocks implied by aunified monetary policy and tight constraints on nationalfiscal policies under EMU.

5Paul R. Krugman, Geography and Trade (Leuven, Belgium:Leuven University Press, 1991), points out that manufacturingproduction is currently more regionally specialized in theUnited States than in the EU and argues that the introduction ofthe single currency will lead to greater regional specializationas firms seek to maximize various network externalities.However, product market integration is unlikely to increasethe incidence of asymmetric shocks if such integration ischaracterized by intra-industry rather than inter-industryspecialization, and if the incidence of industry-specific shocksis more important than country-specific shocks. See the dis-cussion in Alan C. Stockman, “Sectoral and National Aggre-gate Disturbances to Industrial Output in Seven EuropeanCountries,” Journal of Monetary Economics, Vol. 21 (May1988), pp. 387–409.

6See Bayoumi and Prasad, “Currency Unions.”

Correlation of Demand and Supply Shocks withAnchor Areas

–0.4 –0.2 0 0.2 0.4 0.6 0.8 1.0–0.4

–0.2

0

0.2

0.4

0.6

0.8

1.0

–0.4 –0.2 0 0.2 0.4 0.6 0.8 1.0–0.4

–0.2

0

0.2

0.4

0.6

0.8

1.0

Denmark

New England

BelgiumFrance

Netherlands

SpainIreland

UnitedKingdom Italy

Portugal

Supply

Supply

Great Lakes

SoutheastPlains

Far West

Southwest

Mountain

Dem

andD

emand

Correlation with Germany

Correlation with U.S.Mideast Region

Source: T. Bayoumi and B. Eichengreen, “Shocking Aspects ofEuropean Monetary Unification,” in Adjustment and Growth in theEuropean Monetary Union, ed. by Francisco Torres and FrancescoGiavazzi (Cambridge, England; New York: Cambridge UniversityPress, 1993).


The principal case, in the context of temporaryasymmetric shocks, for supplementing national fiscalstabilization with a central fiscal stabilization func-tion would seem to be to provide a means of sharingrisk. Thus countries that experience an above-averageincidence or severity of temporary shocks would havethe costs of their bad luck shared by others. The de-sirability of such insurance might be increased bythe constraints on government borrowing associatedwith the Stability and Growth Pact, and it wouldbe particularly valuable to more open economies,given the fiscal leakages discussed earlier. The designof risk-sharing mechanisms is, however, quite com-plicated, owing partly to moral hazard problems, assuch mechanisms would lessen the need for indi-vidual economies to make themselves more adaptableto economic shocks. The approach chosen in theMaastricht Treaty was the adoption of quite restrictiveprovisions for community financial assistance tomember countries.54

In the case of longer-lasting asymmetric develop-ments, the role of fiscal policy is necessarily morelimited since adjustment rather than financing isrequired.55 Exchange rate changes, including withinthe ERM, have provided an important tool for EUcountries in such circumstances, by facilitating adjust-ment in the face of nominal rigidities. It is to be hopedthat more disciplined fiscal and monetary policies willhelp avoid misalignments of costs and prices in EMU.Nevertheless, with the possibility of exchange ratechanges among countries in the euro area no longeravailable, adjustments in wages and labor mobilityamong countries in the euro area will need to play alarger role. Here, present differences between the EUand the United States are striking. Real wage flexibil-ity in the EU has been estimated at about one-half ofreal wage flexibility in the United States.56 Further,labor mobility across and within EU countries is alsolow by U.S. standards.57 This adds to the need, which

exists anyway given Europe’s high levels of unem-ployment, for the EU to reform labor markets in waysthat would reduce rigidities, by increasing wage flexi-bility. While cultural and linguistic differences will in-evitably continue to limit migration, it is also impor-tant in this context to examine the scope for reducingfurther the barriers to labor mobility across nationalboundaries.58

Overall Assessment of the Policy Framework

The above analysis suggests that the failure to rec-ognize explicitly the importance of flexible labor mar-kets was an important omission from the Maastrichtframework. In the years since the treaty was signed,spurred by the persistence of high unemployment,there have been efforts to strengthen surveillance oflabor market policies by the EU, but this surveillancehas been relatively weak partly because of the tenetthat labor market policies are a matter principally fornational policymakers. Thus, for example, the broadeconomic guidelines set out each year by the Councilhave avoided country-specific references in this area,and the broad-ranging nature of the employment strat-egy that has been discussed at the EU level may havediffused the attention needing to be given to difficultquestions in the areas of social benefit systems andlabor market regulations.

In recognition of the lack of progress in solving theEU’s labor market problems, the draft Treaty ofAmsterdam, agreed in June 1997, makes employmentpolicies explicitly a matter of common concern andsets out procedures for the surveillance of these poli-cies. At the same time, the heads of state or govern-ment stressed the need for the Council to strengthenthe focus on employment in the broad economicpolicy guidelines and made plans for an employmentsummit later in 1997. While these steps may wellstrengthen the framework at the EU level, the reluc-tance of many EU member countries to embark onfundamental labor market reforms will need to beovercome. EMU itself may help in this regard, sincewith the exchange rate (particularly the escape hatchof depreciation) no longer available as an instrumentof adjustment at the national level, understanding bypolicymakers of the need for other means of adjust-ment may be expected to increase.

64

54Article 103a (2) of the treaty allows for community financial as-sistance only to a member state that “is in difficulties or is seriouslythreatened with severe difficulties caused by exceptional circum-stances beyond its control” and when the assistance is supportedunanimously by the Council, except that in the case of natural dis-asters the Council shall act by qualified majority.

55The EU’s structural funds provide assistance to foster structuraladjustment in EU regions generally, in addition to their more highlypublicized role of assisting the catch-up process in member stateswith per capita national incomes significantly below the EU average.

56See Layard, Nickell, and Jackman, Unemployment, Macro-economic Performance and the Labor Market.

57It has been estimated that labor mobility across the differentstates in the United States is almost three times more than labor mo-bility within both France and Germany. See B. Eichengreen, “LaborMarkets and European Monetary Unification,” in Policy Issues in theOperation of Currency Unions, ed. by Paul R. Masson and Mark P.Taylor (Cambridge: Cambridge University Press, 1993), pp. 130–62.Olivier J. Blanchard and Lawrence Katz, “Regional Evolutions,”Brookings Papers on Economic Activity: 1 (1992), pp. 1–75, estimatethat in the United States during 1970–90 more than half of the

adjustment in the first year following a shock occurred through mi-gration. In contrast, Jorg Decressin and Antonio Fatas, “RegionalLabor Market Dynamics in Europe,” European Economic Review,Vol. 39 (December 1995), pp. 1627–55, using a similar methodol-ogy, found that migration played very little role in the EU.

58Progress in this regard has already been made as part of the im-plementation of the single market program, for example, in the mu-tual recognition of educational and skill qualifications and the porta-bility of pensions, but more needs to be done in these and otherareas, including arrangements for taxation and welfare benefits forcross-border commuters.

In the monetary and fiscal areas, the emerging pol-icy framework appears to strike a good balance be-tween rules and the necessary scope for the exercise ofjudgment in the implementation of policies. The de-sign of the Stability and Growth Pact is notable in thisregard. Adherence by EU member states to themedium-term goals of the pact, together with the flex-ibility allowed to deal with “exceptional” and “spe-cial” circumstances, should provide adequate scopefor the use of budgetary policy to deal with shocks.Thus, the absence of a central fiscal function shouldnot pose the problems that are commonly perceived.Without such a central fiscal function, however, thetask of ensuring an appropriate stance of fiscal policyfor the euro area as a whole, and an appropriate mix offiscal and monetary policies in the euro area, will re-quire stronger procedures for policy coordination. Asto monetary policy, considerable discipline is estab-lished by the priority given to price stability in themandate of the ESCB, bolstered by other provisionsthat insulate decision makers there from political pres-sures. At the same time, the ESCB should have theflexibility it needs to deal with the uncertainties con-cerning the demand for financial assets and the trans-mission mechanism that are likely to be particularlyintense in the early part of Stage 3.

However, one should not underestimate the chal-lenges that will face policymakers in the early years ofEMU. In the area of monetary policy, it will be neces-sary for EU policymakers from diverse nationalbackgrounds to establish common positions and takeoperational decisions in a timely way on matters thathave hitherto been determined at the national level. Atthe same time, uncertainties noted in the previousparagraph will complicate the assessment of the im-pact of policies. These challenges are likely to be moreintense the greater the number of countries that partic-ipate initially.

While the low inflation prevailing in the EU pro-vides a good starting point for the single monetary pol-icy, the markets are likely to follow very closely thedecision-making process until the ESCB has fully es-tablished its credibility, to assure themselves thatmonetary policy is staying true to its anti-inflationmandate. To provide this assurance—while at thesame time maintaining a balanced approach to policythat allows the ESCB to take due account of othergoals such as stabilizing output—it will be critical forthe ESCB to explain carefully and clearly its strategyand the rationale for its actions.

In the area of fiscal policy, too, the coordinationneeded will be greater than has been the practice todate. With progress in establishing strong fundamen-tals not as good as in the monetary area, there will inaddition be special scrutiny given to the Council’s im-plementation of the Stability and Growth Pact. TheCouncil will be in a much better position to use thescope for flexibility in the pact, should it see the need,

if markets are convinced of the commitment to fiscaldiscipline. Thus, further durable progress in fiscal con-vergence during the remainder of Stage 2 is, as is dis-cussed below, imperative.

Establishing Conditions for a Successful EMU

It has been clear since monetary union became an as-piration of European policymakers that for EMU tofunction successfully, the participating economieswould need to satisfy a number of conditions relating toconvergence of economic performance. To provide abasis for assessing the readiness of countries to partici-pate in EMU, the Maastricht Treaty established conver-gence criteria in the areas of inflation, public finances,interest rates, and exchange rates (see note to Table 11),and prior to the decision on which countries are readyfor EMU, the Commission and the EMI will provide re-ports evaluating how countries have complied withthese criteria, and the European Parliament will alsogive its opinion. These reports will be based on actualoutcomes in 1997 but will also take account of budgetsfor 1998 in assessing whether fiscal positions are sus-tainable. Many observers have assumed that in comingto its decision on the start of EMU, the Council willalso consider other factors such as whether there aresufficient qualified countries to initiate EMU, the desir-ability of having as broad a monetary union as possible,subject to ensuring a sound footing for the euro, and thepolitical importance of the project.

In the period since the treaty was signed, the EUcountries have made considerable progress in macro-economic convergence (see Table 11). For 1997, av-erage consumer price inflation in the EU as a whole isprojected at 2 percent, less than half of what it waswhen the treaty was agreed in 1991, with all EUmembers, except Greece, expected to have inflationbelow 3 percent. EU inflation is now lower than atany other time in the last thirty-five years (Figure 21).Important improvements have also been made in thefiscal area. The EU-wide general government deficitin 1997 is projected at 2#/4 percent of GDP, some 3!/2percentage points lower than its peak in 1993.Corrected for the effects of the economic cycle, thegeneral government deficit has been narrowingsteadily since 1991 (Figure 22).

With significant convergence in inflation and inpublic finances, and with progress toward EMU seento be gathering momentum, long-term interest differ-entials vis-à-vis Germany have fallen to historicallylow levels. These differentials are now less than 1 per-centage point for all ERM countries, and in a numberof cases the differential has effectively disappeared.Paralleling convergence in interest rates, conditions inthe ERM have been relatively settled since the springof 1995 and, against the background of a strengthen-


65


ing trend in the U.S. dollar, there has been a significantconvergence of market exchange rates on central ratesagainst the deutsche mark.59

Countries’ efforts to improve inflation and fiscalperformance in recent years have not been withoutshort-term costs in terms of weaker activity, coming ata time when macroeconomic conditions were alreadyrelatively weak. But fiscal positions in most countriesneeded correction, independent of the requirements ofthe Maastricht Treaty. Moreover, the costs of fiscalconsolidation should not be exaggerated. Reducingfiscal deficits has facilitated sharp falls in interest pre-miums in a number of countries and an easing of mon-etary conditions more generally. Furthermore, the fal-tering of Europe’s recovery in 1995–96 also reflectedthe impact on confidence of inadequacies in structuralpolicies in many countries—notably the slow pace oflabor market reforms, but also the limited progress inthe reform of government spending programs and theconsequent overdependence on tax measures in fiscalconsolidation efforts.

Despite the efforts made, it has proven more diffi-cult than expected to meet all the reference values setin the treaty. All countries aiming to join EMU fromthe start already satisfy the treaty’s requirements forinflation and long-term interest rates. Nor is the ex-change rate criterion expected to pose an obstacle forthese countries.60 In the fiscal area, however, the threelargest of these countries—France, Germany, andItaly—are projected, in the absence of new measures,to slightly exceed the Maastricht reference value forthe deficit, with some other countries satisfying it withlittle margin to spare. Given the inherent uncertaintyof macroeconomic projections—fiscal positions mayturn out a little stronger than projected, without achange in policies, but they could just as easily beweaker—it is still unclear how many countries will in1997 have deficits of 3 percent of GDP or below. Thedifficulties encountered in deficit reduction have beenreflected also in slower-than-expected progress in debtreduction: only in 1997 has the EU-wide gross generalgovernment debt ratio stabilized, with the level formost countries in excess of the reference value.

66

59Excluding the Irish pound, the deviations of the strongest andweakest currencies from their bilateral central rates have typicallybeen in the range of 2–4 percent since the spring of 1996. SinceNovember 1996, the Irish pound has appreciated significantly rela-tive to its central rate but this has reflected country-specific factors(related partly to the appreciation of the pound sterling) rather thanbroader tensions within the system.

60Three EU countries, Greece, Sweden, and the United Kingdom,are not members of the ERM. Greece and Sweden do not plan toparticipate in Stage 3 at its inception and the government of theUnited Kingdom has stated that it is unlikely to participate at theoutset. While Finland (which joined the ERM in October 1996) andItaly (which rejoined in November 1996) will not have been mem-bers for a full two years when the decision on membership isreached, this is not expected, in itself, to present an obstacle to themembership aspirations of these countries.

Figure 21. European Union: Inflation and GeneralGovernment Balance

1Excludes Greece.

1960 65 70 75 80 85 90 95

1960 65 70 75 80 85 90 95

–8

–6

–4

–2

0

2

–5

0

5

10

15

20

25

General Government Balance1

(In percent of GDP)

Inflation(Annual percentchange)

Three countries withhighest inflation rates

Three countries withlowest inflation rates

European Union

The steady convergence of inflation since the 1970s contrasts with thecontinued growth of fiscal deficits up to the early 1990s.

The treaty, of course, does leave room for judgmentin assessing compliance with the convergence criteria.And it is important that countries’ performance beconsidered in the context of the economic rationale forthe criteria, which is that they are indicators of a coun-try’s ability to subscribe to macroeconomic policiesthat are conducive to the smooth working of a mone-tary union with low inflation, and that in particularhelp avoid spillovers that might weaken the credibilityof monetary policy. It is clear that the inflation perfor-mance of the EU countries in recent years provides apropitious starting point for the new central bank.What is less clear is how shortfalls from the fiscal cri-terion should be assessed.

The deficit and debt reference values specified inthe treaty do not represent definitive thresholds of dan-ger; no such precise thresholds exist. Nevertheless, de-spite an inevitable degree of arbitrariness, they do pro-vide reasonable reference points for assessing whethercountries are straying too far from their medium-termgoal of a balanced fiscal position, and thus a barome-ter of their commitment to that goal. A country’s per-formance in this respect needs to be interpreted in thecontext of its cyclical position. It is clear that countrieshave made important strides in strengthening theircyclically adjusted fiscal positions in the 1990s.Indeed, the EU-wide cyclically adjusted fiscal deficitis projected at around 1!/4 percent of GDP in 1997,some 2!/2 percent of GDP lower than in 1995, with allcountries that plan to participate in EMU from its startwell below 3 percent. Moreover, while most countriesstill have some way to go in establishing an underly-ing fiscal position that would allow the operation ofautomatic stabilizers in normal circumstances withoutbreaching the treaty’s reference values, they have un-derlined their commitment to sustained fiscal disci-pline through their Stability and Growth Pact.

There are, however, some factors casting a shadowon fiscal performance. First, a number of countries haverelied to varying degrees on temporary measures in1997 to reduce their deficits, with some of these mea-sures being essentially cosmetic. More generally, byfailing to achieve their goals through structural reformsproducing durable cuts in spending, many countrieshave missed an opportunity to demonstrate their un-equivocal commitment to fiscal discipline. It is there-fore important that countries continue to follow throughon their fiscal policy commitments and introduce newreform-based measures that will ensure further substan-tive adjustment in 1998, while also strengthening con-fidence that adjustment will be lasting. The case for al-lowing countries to move to Stage 3 despite smalldeviations from the Maastricht reference values wouldbe stronger if there were assurances that sustainable fis-cal adjustment was being undertaken. Progress in re-ducing structural deficits is also essential to prevent po-tential complications in the operation of the Stabilityand Growth Pact at the beginning of Stage 3.


67

Figure 22. European Union: Inflation, GeneralGovernment Balance, and Gross Debt

–7

–6

–5

–4

–3

–2

–1General Government Balance(In percent of GDP)

General Government Gross Debt(In percent of GDP)

50

55

60

65

70

75

1991 92 93 94 95 96 97

1991 92 93 94 95 96 97

1991 92 93 94 95 96 97

0

4

8

12

16Inflation(Annual percent change)

Three countries withhighest inflation rates Three countries with

lowest inflation rates

European Union

Structural

Unadjusted

In the 1990s, continued convergence of inflation has beenaccompanied by substantial fiscal consolidation.

In coming to a decision on countries’ eligibility forStage 3, the Council will not focus explicitly onwhether countries have developed sufficiently flexiblelabor and product markets to make up for the loss ofthe exchange rate instrument. As noted earlier, this isclearly a critical issue for individual member states, notjust because of the social, economic, and budgetarycosts of the existing high levels of unemployment butalso because of the need for more flexible markets tohelp their adjustment to shocks and to make theireconomies more efficient. One should also not under-estimate its systemic implications for the euro area. Afailure to address labor market problems would preventEurope from realizing its full growth potential, andcould also weaken the credibility of the euro if finan-cial markets perceive that persistent unemployment iseroding support for prudent macroeconomic policies.

The persistently high structural unemployment inthe EU—estimated at 8–9 percent of the labor force in1997—points to the limited progress made in manyEU countries in addressing structural weaknesses inlabor markets. There is a need for more comprehen-sive and sustained efforts to increase incentives towork, to acquire skills, and to create productive jobs.This will require fundamental reform of overly gener-ous social benefits systems and the relevant aspects oftax systems, of minimum wage regulations, and ofother regulations that impede wage flexibility or dis-courage the creation of new jobs by limiting employ-ers’ ability to dismiss employees. Measures in theseareas need to be complemented by further steps to en-hance the competitiveness of product markets.61

Managing Uncertainties in the Preparations for EMU

The progress that EU countries have made inmacroeconomic convergence, together with the insti-tutional framework for macroeconomic policy that isbeing put in place for EMU, goes a long way to estab-lishing a solid basis for the euro. There is also a strongpolitical commitment to implement EMU at the begin-ning of 1999. Nevertheless, significant uncertaintiesremain that could ignite financial market tension.

A particularly difficult scenario would be one inwhich a delay in EMU’s start date was perceived as asignificant possibility.62 While interest differentialssuggest market confidence that EMU will start on time,there has been persistent concern that some key coun-tries may not satisfy the treaty’s reference values in the

fiscal area. This in turn has focused attention on ques-tions concerning how the scope for judgment providedin the treaty will be used by the Council in assessingfiscal performance, how large a deviation from thereference values would require a delay, and how muchweight will be placed on budgets for 1998. Moreover,it is unclear how much latitude in the interpretationof the convergence criteria would be viewed by theGerman Parliament or the German ConstitutionalCourt as a failure to comply with the treaty.63

A second set of uncertainties relates to whether spe-cific countries will qualify for initial participation ifEMU begins on time. In addition to doubts surroundingthe effectiveness with which these countries are carry-ing through their budgetary plans and how the scope forinterpretation of the treaty will be used, there are ques-tions as to whether a failure to qualify might weakenpolitical support for fiscal discipline, given that exclu-sion would entail failure to meet the strong politicalcommitment to being among the initial members.

The decision on which countries will initially partic-ipate in EMU will not remove all uncertainty. In the ab-sence of clear and credible guidance, there may be ques-tions concerning the exchange rates at which currencieswill be locked. The treaty provides little explicit direc-tion on this; its requirement relating to exchange ratestability applies only up to the time when the decisionto start EMU is made. There will also be uncertaintiesrelated to the central parities against the euro for thecurrencies participating in ERM2, as they will not be es-tablished formally until the start of Stage 3.

Concerns about whether the process as a whole willproceed as planned could lead to widespread turbu-lence in European financial markets, which woulderode confidence further, slow output growth, andwiden fiscal deficits. The dangers of such turbulencewould intensify if economic recovery in Europe wereto be slower than presently envisaged, unemploymentwere to rise further, and calls for misguided policy ac-tions aimed at the creation of jobs through public ex-penditure initiatives were to intensify. Moreover, anannouncement of a delay in the start of EMU that didnot offer credible assurance that EMU itself was not inquestion—and how such assurance could be providedis not obvious given that the treaty does not providefor a delay—would be likely to result in turmoil thatcould derail the entire process. This, in turn, couldhave adverse consequences for other projects directed


68

61See also Chapter I. Progress with labor market reforms inEurope is discussed in Chapter II, pp. 39–42.

62It is not clear how a delay of the starting date beyond January 1,1999 could be arranged. One view is that it would require an amend-ment of the treaty, which would have to be ratified by all memberstates.

63The German authorities have consistently emphasized the needfor a strict interpretation of the treaty, as did the German Parliamentwhen ratifying the treaty, out of concern that a softening of the cri-teria would undermine public acceptance of the euro. The parlia-ment will give guidance to the German government on the positionto take at the meeting of heads of state or government in spring1998, and the German government is committed to accepting thisadvice. The German Constitutional Court, in a 1993 decision, re-served the right to examine cases challenging whether decisions byEU institutions were consistent with the treaty.

at strengthening economic integration in Europe andcall into question the viability of the ERM. Even if theEMU process could be kept on track in face of suchpressure, market turbulence would have adverse con-sequences both for EU countries themselves and forother countries that have close exchange rate or tradelinks to the EU. Turmoil in the run-up to EMU couldalso weaken the initial credibility of the euro.

The best way to minimize uncertainties about thestart of EMU and to ensure that a sustainable andstrong EMU begins on time would be for countries totake fiscal actions in the remainder of 1997 and in 1998that demonstrate their commitment to the requirementsof the treaty and of the Stability and Growth Pact, par-ticularly through fundamental reforms that also addresssome of the root causes of Europe’s unemploymentproblem and excessive levels of public expenditure.The management of the EMU process would also befacilitated if governments reached early understand-ings that resolved the uncertainties noted above abouthow decisions on participation in EMU will be made.Such understandings should not necessarily be madepublic, but they should help to avoid potentially dam-aging public disharmony. Progress in these areas wouldin turn facilitate closer coordination of monetary poli-cies geared toward economic conditions in the ERM asa whole, which would further reinforce confidence thatEMU will start on time.

Uncertainties surrounding an individual countrythat did not call into question the EMU process itselfwould have less widespread effects but could never-theless damage economic performance both in thecountry in question and in countries with close ties toit. The dangers would be greatest where there weredoubts about the willingness of the country to continuethe pursuit of disciplined policies. A country in this po-sition would need to signal through its actions a re-solve to persevere with fiscal adjustment. This couldbe combined with clear statements by EU leaders thatthe country concerned would be admitted to the euroarea as soon as its economic performance permitted,on the same basis as countries in the first wave.64

Managing the Locking of Exchange Rates

The treaty states that the locking of exchange ratesshall not in itself modify the external value of the ECU(European currency unit), the official currency basketthat will be converted into the euro at a rate of one toone. While no further official interpretation has beengiven to this provision, it is generally expected that

currencies will be locked using the bilateral marketrates prevailing on the last working day prior to themonetary union; this would be sufficient to ensure thatcurrencies have the same value in terms of the euro atthe opening of its first day of trading as they had interms of the ECU at the close of its last day of trad-ing.65 Consistent with this interpretation of the treaty,however, there are different possibilities for managingexchange rates in the final months of the transition toEMU.66

One approach would be to continue with existingexchange rate arrangements. It could be argued thatthere would be only minimal risk in such a strategysince the macroeconomic fundamentals would be inplace and, as required by the treaty, exchange ratesamong participating countries would have been stablefor at least two years prior to the decision. Moreover,this approach would allow room for a realignment,should a large asymmetric shock occur prior to EMU.This latter consideration, however, seems out of tunewith an assessment that countries are ready to enter amonetary union. More generally, the experience of the1990s suggests that even if central rates were not per-ceived to be misaligned, exchange rates might never-theless be subject to speculative pressure, sparked bysuch factors as political developments and uncertain-ties about future economic policies. For example,market perceptions could emerge that a country wouldbe happy to see a depreciation of its currency tostrengthen its competitive position before entering theeuro area.

In view of the above, there is a strong case for pre-announcing the bilateral rates that will be used to lockin the participating currencies, by the time that entrydecisions are made in spring 1998.67 It is critical, how-

Managing Uncertainties in the Preparations for EMU

69

64The treaty specifies that the status of countries not participatingin Stage 3 would be reconsidered at least once every two years, butthat earlier consideration would be given at the request of the coun-try concerned. The entry criteria to be applied to countries joiningthe euro area after January 1, 1999 will be the same as those appliedto the initial participants.

65Under an alternative interpretation of the treaty, a different setof rates from those prevailing on the last day of Stage 2 could alsobe consistent with this requirement, as long as the implied depreci-ation of one participating currency against the ECU (weighted by itsECU-basket share) was exactly offset by the appreciation of one ormore of the other participating currencies (similarly weighted). Lastminute realignments, however, would not appear to be consistentwith the spirit of the treaty.

66Some of the issues that arise are discussed in David Begg,Francesco Giavazzi, Jürgen von Hagen, and Charles Wyplosz,EMU, Getting the End-Game Right (London: Center for Economicand Policy Research, February 1997); and by Maurice Obstfeld,“Strategy for Launching the Euro” (unpublished; University ofCalifornia, Berkeley, February 1997).

67Even if the bilateral rates among currencies of participatingcountries were to be preannounced, it would not be possible to setthe conversion rate of each currency to the euro until the start of1999 because the value of the nonparticipating currencies that are inthe ECU basket could fluctuate in the final part of Stage 2. It shouldalso be noted that the initial exchange value of the euro in terms ofnon-EU currencies will depend on the behavior of all currencies inthe ECU basket up to the end of Stage 2. As long as market rates onthe last day of Stage 2 are used, however, there will be no disconti-nuity between the exchange rates of participating countries againstnonparticipating countries at the end of Stage 2 and the implicitrates at the beginning of Stage 3.


ever, that the rates announced are broadly consistentwith economic fundamentals. Until the needed im-provements in the functioning of EU labor markets aresecured—and this will take some time, even assumingthat reform efforts intensify—downward inflexibilityof nominal wages, combined with low inflation in theeuro area, will make it difficult to achieve substantialand rapid adjustments to the pattern of real exchangerates within the euro area.68 It is also important thatany preannouncement be credibly backed by commit-ments from monetary policymakers.

It is widely assumed that any preannouncement ofbilateral conversion rates would be based on experi-ence in the period leading up to the announcement.For most countries, bilateral rates are likely to havebeen relatively stable over the previous two years.Moreover, since the choice of rates at which curren-cies of participating countries are locked requires aunanimous decision by the Council, it may be easier tosecure agreement based on the status quo than onsome other set of rates. One option would be to use theexisting bilateral central rates in the ERM. Selectingother rates might be seen as setting an awkward prece-dent for potential future members of the euro area, asit would cast doubt on whether the central parityagainst the euro in ERM2 was a good guide to the rateat which a country would enter. A close alternativewould be to use the average of the rates in a definedperiod prior to the announcement. For most countries,this would produce an outcome close to that involvedin the use of central rates, but it would allow some ac-commodation of special circumstances, such as the di-vergence of the Irish pound from its bilateral centralrates against other ERM currencies over the past year.

It is difficult to judge with any precision what riskswould be entailed in adopting the current pattern of bi-lateral rates as the basis for the conversion. Whilethere is no evidence that present market rates in theERM are widely out of line with fundamentals, esti-mates of equilibrium exchange rates tend to have widemargins of error. Moreover, it is inevitable that theevolution of economies in the euro area will give riseto the need for adjustments in the pattern of real ex-change rates, reflecting, inter alia, their relativesuccess in dealing with existing imbalances.69 Theseconsiderations make it all the more important to pur-sue structural reforms vigorously so that more flexiblelabor and product markets can facilitate requiredchanges in real exchange rates.

The question arises of whether a preannouncementof rates would require any fundamental change in howmonetary policy is conducted. One view is that a pub-licly stated commitment to intervene without limit onthe last day of Stage 2 would be sufficient to inducesupporting behavior in the markets in the precedingperiod. An advantage of this approach, it is argued, isthat the monetary consequences of any intervention onthe last day of Stage 2 would cancel out for the euroarea as a whole; accordingly it should be possible tomake the commitment to such a strategy credible with-out any liquidity impact for the euro area as a whole.However, the absence of any firm commitment to sup-port the agreed exchange rates prior to the last day ofStage 2 might raise questions about the degree of com-mitment to the agreement, especially if there are im-portant political uncertainties. Moreover, it wouldseem strange if the implications of intervention formonetary conditions were seen as an obstacle to moreextensive support for the agreed bilateral conversionrates, as it would be expected that countries about toenter a monetary union should be able to agree on thestance of monetary policy.

This clearly suggests an approach in which mone-tary policy coordination is intensified immediately fol-lowing the announcement of the bilateral conversionrates. This could be achieved through a pledge to in-tervene without limit on the part of all the centralbanks concerned so as to keep exchange rates close tothe announced rates, an understanding among centralbanks on the overall stance of monetary policy, and awillingness if needed not to sterilize intervention. Inmost circumstances, a prompt and determined re-sponse by central banks to any sign of a market testshould quickly convince markets of their commitmentto the chosen rates. An even more transparent strategywould be to coordinate monetary policy explicitlythrough the ECB, once it is established. As well asproviding a strong underpinning to the chosen bilateralrates, this would help smooth the change from sepa-rate monetary policies in Stage 2 to a single monetarypolicy in Stage 3.

External Effects of EMU

The effects of EMU on the world economy will de-pend on the external spillovers from its effects on eco-nomic performance in Europe and the extent to whichthe euro is used in international transactions. The lat-ter will be influenced by the strength and stability ofthe new currency, as well as developments over thetransition period among the currencies of potentialEMU participants. The recent weakness in EUeconomies and the associated depreciation of theircurrencies against the dollar and the yen have loweredthe demand for the exports of EU trading partners,while growth in EU net exports has supported the

70

68Real appreciation for one or two countries that do not have alarge weight in the euro area would be more manageable.

69The lagged consequences of German unification provide anillustration of such exchange rate dynamics. A theoretical analysisof the unification shock predicts an appreciation of the German realexchange rate in the short-to-medium term to help clear the goodsmarket, with a weakening in Germany’s net foreign asset position inthe long run warranting more than a full reversal of the earlier ap-preciation, other things being equal.

EU’s otherwise weak output and employment perfor-mance. These adverse spillovers of recent economicdevelopments in Europe should dissipate as more solidgrowth resumes in the EU.

Characteristics of the Euro

Over the medium to longer term, several factorssuggest a broader role and thus greater demand for theeuro than for the current EU currencies, which are un-derrepresented in global transactions relative to theU.S. dollar despite the similar size of their economicbases (Table 12). Even though the United States andthe EU both account for about 20 percent of world out-put and 15 percent of world exports, nearly half ofglobal trade flows are priced in dollars compared withonly about 30 percent in EU currencies. Global assetholdings, both private and public, are also dispropor-tionately denominated in dollars, including half thestock of debt issued by developing countries and 37percent of total international debt securities (equiva-lent to three times the value of debt issued in theUnited States itself), compared with shares of around16 percent and 34 percent, respectively, for EU cur-rencies. The dollar is held far more extensively thanthe EU currencies in official reserves and is the mostfrequently used currency in foreign exchange transac-tions, being involved in more than 40 percent of suchtransactions (after adjustment for the double-countingthat arises from the use of two currencies in eachtransaction), compared with 35 percent for all EU cur-rencies with intra-EU transactions included.70

The larger economic base of the euro and the elimi-nation of the transactions costs involved with multipleEuropean exchange rates are likely to increase gradu-ally the use of the new European currency as a unit ofaccount in the denomination of trade flows, with par-ticular growth in transactions between the euro areaand developing and transition countries. Increased in-tegration of Europe’s financial markets, with the re-placement of many currencies by one, should lowercosts of financial transactions, narrow interest ratespreads, and expand the supply of euro-denominatedassets as borrowers tap into the expanded European fi-nancial system. The euro is also likely to complementthe dollar increasingly as a major reserve currency,partly for intervention purposes but also because itsextensive use and the greater depth and breadth ofmarkets in financial assets denominated in euro, com-

pared with the multiple European currencies of today,will provide incentives for countries to diversify theirreserve holdings to be more in line with the currencycomposition of their trade and financial transactions;this will be particularly important for developingcountries that now hold largely dollars. Within theEMU countries, present reserve holdings of partners’currencies will be converted into euro and thus nolonger be reserves, although they will continue to be acounterpart of the monetary base. This will leaveEMU countries with foreign exchange reserve hold-ings primarily in dollars, and these are likely to be re-duced because trade within the euro area will nolonger need the backing of international reserves, al-though any reductions will be subject to the approvalof the ECB, which will examine them for consistencywith its monetary policy.

These changes point to increased demand for theeuro (relative to the overall demand for the corre-sponding pre-EMU currencies), but the associatedshift in official and private portfolios away from dol-lars will represent only a modest portion of U.S. inter-national assets and liabilities, which at the end of 1996totaled $3.35 trillion and $4.13 trillion, respectively.The shift will also occur gradually, as market partici-pants become familiar with the properties of the euroand as the incumbency advantages of other reserveand vehicle currencies linger. In fact, the effects of in-creased demand for the euro for portfolio purposes arelikely to be less important than such influences as thestance of policy and differences in economic perfor-mance across countries that affect returns to invest-ment and thus the attractiveness of holding differentcurrencies. In particular, the mix and stance of fiscaland monetary policies in the euro area, together with


71

70See Robert N. McCauley and William R. White, “The Euro andEuropean Financial Markets”; Alessandro Prati and Garry J.Schinasi, “EMU and International Capital Markets: StructuralImplications and Risks”; and Paul R. Masson and Bart G.Turtelboom, “Characteristics of the Euro, the Demand for Reserves,and Policy Coordination Under EMU,” in EMU and theInternational Monetary System, ed. by Paul R. Masson, Thomas H.Krueger, and Bart G. Turtelboom (Washington: IMF, 1997).

Table 12. United States, Japan, and the EuropeanUnion: Relative Economic Size and Relative Useof Currencies(In percent)

United States Japan EU15

Relative economic sizeShares of world GDP, 1996 20.7 8.0 20.4Shares of world exports

(ex-intra-EU), 1996 15.2 6.1 14.7

Relative use of currencies1

World trade, 1992 48.0 5.0 31.0World debt securities,

September 1996 37.2 17.0 34.5Developing country debt,

end-1996 50.2 18.1 15.8Global foreign exchange

reserves, end-1995 56.4 7.1 25.8Foreign exchange

transactions, April 19952 41.5 12.0 35.0

1Shares denominated in currency (or currencies) of country (or EU).2Shares adjusted for double-counting that arises from the fact that

each transaction involves two currencies.


the dynamism and efficiency of the participatingeconomies, are likely to be the most important factorsdetermining the strength or weakness of the newcurrency.

Independent of the question of the strength of theeuro relative to current exchange rates, the issue arisesof whether implementation of EMU may potentiallylead to somewhat more pronounced movements in theexchange value of the euro in terms of the other majorcurrencies—the dollar, in particular—than has beenobserved in the exchange rates of the deutsche markand the currencies linked to it in the ERM vis-à-visnon-EU currencies. The answer is uncertain. On theone hand, to the extent that a given change in the ef-fective exchange rate will have less of an impact ondomestic prices under EMU than previously, policy-makers at the ESCB may be less concerned about, andso less inclined to resist, exchange rate movementsthan have been pre-EMU policymakers. But on theother hand, a given change in the exchange rate of theeuro vis-à-vis the dollar is likely to have about thesame effect on prices in the euro area as a correspond-ing change in the dollar exchange rates of ERMcurrencies in the past, and the same appplies for anyother non-EMU currency. This suggests that policy-makers’ concerns about exchange rate fluctuations vis-à-vis the dollar or any other non-EMU currency maybe broadly unchanged.

With respect to cyclical factors, the extent to whichexchange rate fluctuations arise from cyclical diver-gences between the euro area and other economieswill depend on the nature and magnitude of the shocksto which they are exposed. Business cycles have be-come broadly more similar across the EU countries inrecent years, and increased trade and financial integra-tion under EMU are likely to increase the importanceof common factors across EU countries. But cyclicalfluctuations do not appear to have become more syn-chronized between the EU and the major advancedeconomies outside Europe.71 Differences in economicconditions resulting either from asymmetric shocks orfrom unsynchronized business cycles and divergentpolicy stances across the three major currency regionsmay then be reflected in relatively pronounced ex-change rate fluctuations. Such fluctuations should notin themselves be viewed as undesirable or destabiliz-ing, provided they are not excessive; indeed, with in-tegrated capital markets, changes in exchange rates

form an important part of the adjustment mechanismthat equalizes real rates of return and allocates capitalto the most productive uses across countries. As busi-ness cycles become more synchronized within theeuro area, cyclical conditions in the area could have anincreasingly important influence on other regions,through trade, financial conditions, and variations inthe exchange rate, much as the United States has a sig-nificant influence on the world business cycle.72

Implications for Advanced European CountriesOutside the Euro Area

Given their close economic ties to EMU members,advanced European economies outside the euro areawill derive important benefits in the medium to longrun from a well managed EMU, but for the same rea-sons they will also be closely affected in the short termby how well fiscal and monetary policies in the euroarea are adapted to the early challenges facing policy-makers. The effects of EMU on these economies willdepend on the initial size of the euro area, since asmaller EMU bloc will be more affected by externaltrade and thus less inwardly focused in policymaking,and also on whether they participate in ERM2.

As in the current ERM, participants in ERM2 willneed to demonstrate their anti-inflation credentials bypursuing cautious monetary and fiscal policies if theyare to sustain a stable exchange rate against the euro.For countries that seek to join EMU, there is likely tobe particularly close market scrutiny, with sharp reac-tions in terms of increased interest rate differentials ifdevelopments such as fiscal policy slippages move acountry farther from the convergence criteria neededto enter the euro area. Should the policy mix in theeuro area become a point of tension in the initial phaseof EMU (reflecting, for example, the inability of euroarea policymakers to deal adequately with some of thepolicy challenges that were noted earlier), this mayalso tempt markets to challenge some of the parities inERM2.

Countries that do not participate in ERM2 can, ingeneral, be expected to continue with the current ori-entation of their monetary policies. Such a strategywould allow them flexibility to accommodate asym-metric shocks between themselves and the euro area,including during the changeover to the euro.73 Some inthis group with important financial markets (e.g.,Switzerland, and also the United Kingdom if, as seemslikely, it does not participate initially in Stage 3) may

72

71Antonio Fatas, “EMU: Countries or Regions? Lessons from theEMS Experience,” European Economic Review, Vol. 41 (January1997), pp. 1743–51, provides evidence on the growing importanceof common shocks within EU countries, while Robin L. Lumsdaineand Eswar S. Prasad, “Identifying the Common Component inInternational Economic Fluctuations,” NBER Working PaperNo. 5984 (Cambridge, Massachusetts; National Bureau ofEconomic Research, April 1997), find that in the last two decadesbusiness cycles have not become more closely linked betweenEurope and the rest of the world.

72Stefania Fabrizio and J. Humberto Lopez, “Domestic, Foreignor Common Shocks,” IMF Working Paper 96/107 (September1996), find that fluctuations in U.S. output account for more than 20percent of the fluctuations in other countries’ output.

73See Torsten Persson and Guido Tabellini, “Monetary Cohabi-tation in Europe,” American Economic Review, Vol. 86 (May 1996),pp. 111–16.

also face policy complications should uncertaintiesabout the stance of policy in the euro area—includingin the run-up to Stage 3—spark large capital flows intoor out of the euro.74

Implications for Other Advanced Economies

As large economies, the United States and Japan arelikely to be affected only mildly by any change in thevariability of the euro (as compared with the averagevariability of exchange rates vis-à-vis pre-EMU cur-rencies), although of course their bilateral trade bal-ances with EMU countries will depend on the bilateraldollar and yen exchange rates vis-à-vis the euro.Additional effects on these countries would arise frommore extensive use of the euro as a reserve currencyand denominator for trade and financial transactions,though it is not clear how important these effects wouldbe. Most directly, a shift toward holdings of euro cur-rency as a store of value would reduce seigniorage tothe United States from U.S. currency holdings by non-U.S. residents. Foreign holdings of dollar bills are esti-mated to have accounted for more than half of the total$375 billion stock of U.S. currency held outside banksat the end of 1995, and these foreign holdings provideannual savings of $10–15 billion in interest on treasurysecurities.75 Potentially more important, however, isthat the desire of foreigners to hold dollar assets makesit easier for the United States to finance its balance ofpayments deficit; episodes in which even the hint of asell-off of treasury security holdings by foreign in-vestors have resulted in sharp increases in U.S. long-term interest rates clearly show the benefit to theUnited States derived from widespread acceptance ofthe dollar. The increased liquidity in global financialmarkets that may be expected to result from EMU,however, could lower borrowing costs for all countries,including the United States. The yen is not used nearlyas widely as the dollar in international transactions,with dollar use predominating even in trade betweenJapan and its trade partners in Asia. The emergence ofthe euro may well hamper the development of the yenas an international currency, although the plannedderegulation of the Japanese financial sector should in-crease the attractiveness of the yen for nonresidents.

Implications for Developing and Transition Countries

Movements in the values of the major currenciesand the degree of variability of the euro’s exchange

rate will affect developing and transition countrieswith economic ties to EMU countries. Possibly thelargest effects will be felt in countries in central andeastern Europe and the CFA franc zone that, partly be-cause of strong trade links, peg their currencies to EUcurrencies (and in future the euro), but pay contractualfinancial obligations on their debt in dollars or yen(Figure 23). With such a “mismatch,” a depreciation ofthe euro would lead to an increase in the domestic cur-rency cost of debt service, probably without a fullyoffsetting benefit in the trade account. This is likely tobe a particular concern for highly indebted countriesthat are constrained in their ability to diversify their li-abilities because of limited access to internationalcapital markets. For countries with access to privatecapital, EMU will provide a deeper market into whichto float euro-denominated bonds, allowing countriesthat currently borrow heavily in dollars to reduce themismatch between the currency of their exchange ratepeg and their financial obligations.

Significant exchange rate variability of the euromay also lead to terms of trade volatility if changes inexchange rates lead to divergent effects on the domes-tic currency prices of a country’s imports and exports.For example, an appreciation of the euro could possi-bly have an adverse effect on the terms of trade ofcountries that import manufactures from EU countriesbut export primary commodities to a more diversifiedgroup of partners; the domestic currency price of im-ports would rise and the price of exports would fall un-less shifts in global supply and demand resulting fromthe exchange rate movement lead to offsetting changesin euro and dollar prices.76 The incidence of mis-matched trade flows is fairly small, however, as coun-tries tend to have similar shares of imports and exportswith the EU (Figure 24).

Transition countries in central and eastern Europeand the Baltics that peg their currencies to thedeutsche mark or to a basket in which the deutschemark has a large weight are likely to switch to theeuro; for example, Bulgaria and Estonia have alreadystated this intention. Because the likely participants inEMU include these countries’ principal partners forboth imports and exports, the initial strength or weak-ness of the euro should have little effect on their termsof trade, although a weak euro might lead to higherprices for dollar-denominated energy imports and thusincrease domestic price pressures just as it would forenergy-importing EMU participants. The denomina-tion of the outstanding external debt of central andeastern European countries is somewhat more diversi-fied than that of their trade flows, with substantialobligations in both dollars and yen. Even so, fluctua-


73

74For a discussion of the implications of such capital movements,see IMF, Switzerland—Selected Issues and Statistical Appendix,IMF Staff Country Report No. 97/18 (March 1997).

75See Richard D. Porter and Ruth A. Judson, “The Location ofU.S. Currency: How Much of It Is Abroad?” Federal ReserveBulletin, Vol. 82 (October 1996), pp. 883–903.

76For manufactured goods, however, there is substantial evidenceof “pricing to market,” so that domestic currency prices of these im-ports would be expected to vary less markedly than the exchangerate.


tions in the euro are unlikely to trigger debt-servicedifficulties since the affected countries have continu-ing access to international financial markets and gen-erally low-to-moderate levels of external indebted-ness. Hungary is an exception, with more substantialexternal debt, but it has been successful in reducing itsexposure to currency fluctuations by using forwardcurrency contracts to match its debt-servicing require-ments with the composition of its currency basket. Inthe long run, the transition economies of central andeastern European and the Baltics will unambiguouslybenefit from a successful EMU—both from highergrowth in their principal export markets and fromlower real interest rates that will facilitate the financ-ing of their continuing economic restructuring.Continued low inflation in the EMU countries will fur-ther provide a sound goal for the transition countries toaim for in order to fulfill their aspirations of accedingto the EU.

The countries of the CFA franc zone also maintainfairly close trade links with the EU, particularlyFrance, which guarantees the free convertibility of theCFA franc at the agreed peg. In contrast to the transi-tion countries, however, significant shares of CFAfranc zone exports and imports are from countries out-side the EU, so that even with a seamless transitionfrom the French franc to the euro, these countries’terms of trade will continue to vary with exchange ratemovements. Moreover, exchange rate fluctuationswould have potentially serious repercussions on theircapacity to service debt: a substantial portion of theirexternal debt is denominated in dollars and several ofthese countries are highly indebted and likely to havedifficulty obtaining access to international financialmarkets. Although the effects on trade and indebted-ness work in opposite directions, with, for example, adepreciating euro providing an improvement in theterms of trade but an increase in the burden of dollar-denominated debts, the substantially greater magni-tude of debt burdens relative to trade flows suggeststhat the latter effect will dominate.

Institutional Implications of EMU

The unavoidable uncertainty following the intro-duction of the new currency and the possibility oflarge movements in exchange rates may lead to greatercalls for international policy coordination betweenEMU and other countries, particularly to deal with anyperceived misalignment of exchange rates. Even so, alarge measure of caution is in order, since experienceshows that it is difficult to distinguish normal, albeitrelatively large, currency movements from misalign-ments that would call for concerted action.

The response to any such misalignment might in-clude altering the stance of both monetary and fiscalpolicies in the major economies. The central role ofthe ECB in determining EMU monetary policy will

74

Figure 23. Selected Developing and TransitionCountries: Trade and Financial Links with theEuropean Union(In percent)

0 20 40 60 80

Poland

Slovenia

Mali

Senegal

Gabon

Slovak Republic

Share of long-term debtdenominated in EU currencies

Share of trade withEU countries

Cameroon

Côte d'Ivoire

Hungary

Czech Republic

Countries of Centraland Eastern Europe

Countries of the CFA Franc Zone

Countries in Africa and central and eastern Europe that peg theirexchange rates to European currencies typically have close trade linkswith the European Union (EU), but service debt denominated in bothEU and non-EU currencies.

enable countries in the euro area to speak in unison(rather than adding a competing voice) and enablerapid and effective responses. To ensure that fiscalpolicies complement monetary policy in providing anenvironment consistent with medium-term stability, itmay be desirable to enhance the mechanisms for inter-national policy coordination, including at the IMF.This underlines the importance of putting in placemechanisms and institutions to improve the coordina-tion of fiscal policies among the countries participat-ing in the euro area. This would give the authorities ofthe euro area greater scope for changing the policymix, which should help correct misalignments in theexchange value of the euro.

The changes in the international monetary systemresulting from EMU will call for the IMF’s surveil-lance of euro area economies to be adapted in a num-ber of ways. Under the IMF’s Articles of Agreement,Fund membership is exercised by member states, andthus consultations under Article IV will continue to beheld with individual member countries. But followingthe efforts that have already been made to strengthenthe IMF’s regional surveillance, these will need to becomplemented by discussions with the institutions ofthe euro area, such as the ECB and other appropriatecounterparties, in order to address issues no longer inthe purview of national authorities.

Creating a single European currency has a numberof implications for the IMF’s financial operations, in-cluding changes in the composition and valuation ofthe SDR, in the media used in Fund operations, andthe interpretation of EMU members’ balance of pay-ments and reserve positions. These issues will be ad-dressed by the IMF Executive Board over the comingyear.

Reaping the Benefits of EMU: Importanceof Fiscal and Labor Market Reforms

While monetary union in Europe will be a devel-opment of great importance for the participatingcountries and for the world economy, the fiscal andstructural policies pursued by the countries in the euroarea will continue to play a critical role in their eco-nomic performance and its effects on the rest of theworld.

This section examines the implications of alterna-tive fiscal and structural policy scenarios in the euroarea, using a version of the Fund’s multicountryeconometric model (MULTIMOD). In one scenario(scenario 1), EMU is assumed to serve as a catalyst forfavorable change in both fiscal and structural policies.On the fiscal side, budget consolidation goes beyondthe adjustment assumed in the current WorldEconomic Outlook projections—with an additional re-duction of government expenditure equivalent to 2percent of GDP over the medium term—coming

Reaping the Benefits of EMU: Importance of Fiscal and Labor Market Reforms

75

Figure 24. Selected Developing and TransitionCountries: Shares of Trade with the EuropeanUnion(In percent of total trade)

0 10 20 30 40 50 60 70

Asia

Middle East andnorth Africa

Sub-Saharan Africa

Countries of theCFA Franc Zone2

Countries of Centraland Eastern Europe1

Baltics3

Russia and the othercountries of the

former Soviet Union4

Imports Exports

There is generally little evidence of a “mismatch” in countries’ tradelinks.

1Czech Republic, Hungary, Poland, Slovak Republic, and Slovenia.2Benin, Burkina Faso, Cameroon, Central African Republic, Chad,

Comoros, Congo, Côte d’Ivoire, Equatorial Guinea, Gabon, Mali,Niger, Senegal, and Togo.

3Estonia, Latvia, and Lithuania.4Armenia, Azerbaijan, Belarus, Georgia, Kazakhstan, Kyrgyz

Republic, Moldova, Russia, Tajikistan, Turkmenistan, Ukraine, andUzbekistan.


closer to the policy intentions often expressed by pol-icymakers in EU countries.77 On the structural side, itis assumed that product market liberalization, rein-forced by efficiency gains resulting from the introduc-tion of a common currency, leads to a rise in total fac-tor productivity of !/2 of 1 percent, while significantreforms of labor markets lead to greater real wageflexibility and a reduction in the natural rate of unem-ployment of 2 percentage points, to around 7 percent.

A second, “reform fatigue” scenario (scenario 2) il-lustrates the implications of lagging structural re-forms: here, real wage flexibility remains low and thenatural rate of unemployment gradually increases toabout the level of actual unemployment in the EUtoday. It is also assumed in this scenario that nationalgovernments attempt to offset the effects on outputand employment of deteriorating labor market condi-tions by additional government spending, although theroom for such expenditure programs is circumscribedby the Stability and Growth Pact. It is assumed in bothscenarios that the pact will be strictly enforced andthat all EU member countries will participate in EMUfrom the outset and according to the present timetable;the policy assumptions underlying scenarios 1 and 2diverge from the year 2000. The scenarios are con-structed as deviations from a hypothetical baselineprojection based on an extrapolation of underlyingtrends observed in or estimated from the 1990s (seeAnnex II for further details).

When EMU serves as a catalyst for reform (scenario1), the level of output in EMU member countries risesby almost 3 percent relative to the baseline over themedium term (see Table 13). This benefits the fiscalposition, and, when account is also taken of the as-sumed reduction in government expenditure, the fiscaldeficit in 2003 is about 2 percent of GDP lower than inthe baseline. This creates room for a declining tax bur-den over the medium term, with government debtlower by 12!/2 percent of GDP by 2010. The resultingrise in government saving is accompanied by higherprivate saving as potential output and households’wealth increase. The rise in national saving in the euroarea outweighs strong growth in investment demandso that the external current account improves over thefirst decade of reforms. The effect on other industrialcountries is quite small. Over the medium term, theiroutput increases somewhat, benefiting from lowerworld interest rates on account of higher saving in theeuro area and the world economy. The positive outputeffect for developing countries is somewhat largerthan for other industrial countries and reaches !/4 of1 percent over the medium term; these countries as agroup benefit not only from lower world interest rates,but also from a demand-pull effect as the stronger

growth performance in Europe is accompanied byhigher demand for their products.

The economic outlook for the euro area is likely tobe much less favorable if structural reform and fiscalconsolidation fall short of current expectations (sce-nario 2). With inflexible labor markets and risingstructural unemployment, the level of GDP is esti-mated to decline by 2!/2 percent relative to baselineover the medium term (or by more than 5 percent com-pared with the case of accelerated reform in scenario1). This contributes to a worsening of the fiscal deficitby 1!/4 percent of GDP by 2010, with the governmentdebt-to-GDP ratio rising by 10 percentage points. Theworsening economic outlook and fiscal position arelikely to result in a rise of the risk premium on euro-denominated assets, and this is assumed to be on theorder of 40 basis points in the simulation exercise. Adecline in household and government saving is morethan offset in this scenario by a sharp fall in invest-ment, and the external current and trade accounts im-prove by some !/4 of 1 percent of GDP by 2010. Thecounterpart is largely a deterioration of the trade bal-ance in other industrial countries. The overall outputof other industrial countries is nevertheless slightlyhigher than in the baseline, primarily reflecting a de-cline in interest rates in these countries. In contrast toscenario 1, however, where the decline in interest ratesin the other industrial countries reflects a rise in pub-lic and private saving in the euro area, the decline ininterest rates here is confined to the other industrialcountries and occurs because the increase in the riskpremium of the euro implicitly lowers risk premiumson other currencies. The reduction in activity in indus-trial countries and the consequent lower demand forimports from developing countries leads to some re-duction in output in the latter group, although thespillover effects are again quite small.

The different outcomes under scenarios 1 and 2 alsohave important implications for the euro area’s abilityto absorb adverse shocks, such as the demand andsupply shocks that typically characterize the businesscycle. First, fiscal policy will be severely constrainedif the deficit remains close to the limits imposed underthe Stability and Growth Pact. Under the “reform fa-tigue” scenario 2, this not only limits the authorities’ability to allow the automatic stabilizers to work dur-ing a cyclical downturn but can even require procycli-cal fiscal adjustments to meet the deficit limits (seeAnnex II). These constraints are unlikely to becomebinding with the stronger reforms undertaken in sce-nario 1. Second, apart from fiscal constraints, labormarket reforms can also play a central role in the abil-ity of the euro area to cope with adverse shocks, andlimit potential negative spillover effects to other re-gions. If reforms succeed in making labor marketsmore flexible, as envisaged in scenario 1, adverseshocks would tend to impose smaller economic costs(in terms of output losses and higher unemployment)

76

77The World Economic Outlook baseline projections assume thatstructural fiscal balances remain broadly unchanged after 1998.

within the euro area as well as in the rest of the worldeconomy. The effects would also be less persistent,with the economy returning relatively quickly to itslonger-run potential growth path.

The scenarios illustrate the critical role of the fiscaland structural policies pursued in the euro area for thesuccess of EMU. To the extent that EMU becomes acatalyst for economic reform, not only in the fiscal

area, but also in labor and product markets, there islikely to be substantial benefits for participating coun-tries, with positive, though typically not very large,spillover effects on the rest of the world economy. Butif EMU is not accompanied by further progress withstructural reforms and fiscal consolidation, there arelikely to be serious consequences for Europe, andother regions are likely to bear part of the cost.

Reaping the Benefits of EMU: Importance of Fiscal and Labor Market Reforms

77

Table 13. Implications of EMU for Europe and the Rest of the World—Simulation Results(Deviations from baseline, in percent, except when indicated otherwise)1

2000 2001 2002 2003 2010

Scenario 1: EMU with Additional Fiscal Consolidation and Labor Market Reforms

EMU members2

Real GDP 0.2 0.9 1.0 1.1 2.9GDP deflator –0.3 –0.7 –1.1 –1.4 –1.9Long-term real interest rate 0.1 –0.1 –0.3 –0.4 —Unemployment rate –0.2 –0.4 –0.6 –0.8 –2.0General government balance (in percent of GDP) 0.4 0.9 1.5 2.1 0.8

Net revenue — — 0.1 0.2 –1.1Expenditure –0.4 –0.9 –1.4 –1.9 –1.9

General government debt (in percent of GDP) –0.4 –1.4 –2.7 –4.5 –12.6Trade balance (in billions of U.S. dollars) –3.5 13.8 22.8 31.6 27.9

Non-European G-73

Real GDP –0.1 — — 0.1 0.1Trade balance (in billions of U.S. dollars) –0.6 –16.6 –25.5 –31.5 –31.9

Other industrial countries4

Real GDP –0.1 0.1 0.1 — 0.2Trade balance (in billions of U.S. dollars) –1.0 –1.9 –2.6 –3.5 –1.7

Developing countries5

Real GDP — 0.1 0.2 0.2 0.3Trade balance (in billions of U.S. dollars) 5.1 4.7 5.3 3.4 5.7

Scenario 2: EMU with Neither Additional Fiscal Consolidation Nor Labor Market Reforms

EMU members2

Real GDP 0.1 –0.3 –0.6 –0.9 –2.5GDP deflator 0.1 0.3 0.6 0.9 2.3Long-term real interest rate 0.2 0.3 0.4 0.5 0.5Unemployment rate 0.2 0.4 0.7 0.9 2.0General government balance (in percent of GDP) –0.2 –0.5 –0.7 –0.9 –1.3

Net revenue — –0.1 –0.2 –0.3 –0.7Expenditure 0.2 0.4 0.5 0.6 0.6

General government debt (in percent of GDP) 0.1 0.7 1.4 2.3 9.8Trade balance (in billions of U.S dollars) –1.7 22.1 31.7 38.8 67.3

Non-European G-73

Real GDP –0.1 — — 0.1 0.1Trade balance (in billions of U.S. dollars) –2.6 –22.6 –29.5 –33.7 –57.5

Other industrial countries4

Real GDP — –0.1 — — 0.1Trade balance (in billions of U.S. dollars) –0.8 –2.4 –3.1 –3.6 –6.1

Developing countries5

Real GDP — — — –0.1 –0.2Trade balance (in billions of U.S. dollars) 5.1 2.9 0.9 –1.5 –3.7

1Baseline is based on current World Economic Outlook database, with shocks starting in 2000. Adherenceto the Stability and Growth Pact is assumed in both scenarios.

2It is assumed that all current EU member countries participate in EMU from the start.3The United States, Canada, and Japan.4Australia, New Zealand, Norway, and Switzerland.5Rest of the world excluding transition economies.

Documents

EMU and the W orld Economy - International Monetary FundEMU and the Delors Committee Report of 1989 then set in train the three-stage process leading to EMU laid out in the Maastricht