Forum S U M M E R

2000

Spotlights

DICE Reports

Focus

THE EUROPEAN

MONETARY UNION

Otmar IssingHermann RemspergerT. Padoa-SchioppaDaniel GrosGiancarlo Corsetti

Pro and Contra

INFLATION TARGETING Pro: Peter BofingerContra: Franco Reither

EURO-ZONE EXPORTS:FROM GLOOM TO BOOM

FISCAL MONITORING IN THE

EURO-AREA

NETHERLANDS: EMPLOYMENT

PRODUCTIVITY IS NOT EVERYTING

RECRUITMENT OF SKILLED LABOUR

ABROAD

A joint initiative of Ludwig-Maximilians-Universität and the Ifo Institute for Economic Research

TrendsECONOMIC SURVEY INTERNATIONAL

STATISTICS UPDATE

CESifo Forum ISSN 1615-245XA quarterly journal on European issuesPublisher and distributor: Ifo Institute for Economic ResearchPoschingerstr. 5, D-81679 Munich, GermanyTelephone ++49 89 9224-0, Telefax ++49 89 9224-1461, e-mail ifo@ifo.deAnnual subscription rate: Euro 50.00Editor: Heidemarie C. Sherman, Ph.D., e-mail sherman@ifo.deReproduction permitted only if source is stated and copy is sent to the Ifo Institute

CESifo ForumVolume 1, Number 2 Summer 2000_____________________________________________________________________________________

THE EUROPEAN MONETARY UNION

The Monetary Policy of the European Central Bank: Stategy and ImplementationOtmar Issing 3Future Challenges for the ECBHermann Remsperger 10

The Eurosystem in the International Monetary SystemTommaso Padoa-Schioppa 18One Euro from the Atlantic to the Urals?Daniel Gros 26

A Perspective on the Euro – Why is the external value of the euro currently low?Giancarlo Corsetti 32

INFLATION TARGETING

Pro: Peter Bofinger 37Contra: Franco Reither 39

Euro-zone exports: From gloom to boom 41Fiscal monitoring in the euro-area 42Netherlands: Employment performance 44

Productivity is not everything – An international comparison of the sources of prosperity 46Recruitment of highly skilled labour abroad – the practice of selected OECD countries 48

Economic Survey International 53Statistical update 54

Focus

Pro and Contra

Spotlights

DICE Reports

Trends

THE EUROPEAN MONETARY

UNION

CESifo Forum3

Focus

Inflation leads to amisallocation of resources

n 1 January 1999, responsibility for mone-tary policy in eleven of the Member States

of the European Union (EU) passed from theirrespective national central banks (NCBs) to theGoverning Council of the European Central Bank(ECB).1 This marked a fundamental change in thepolitical and economic environment, as responsi-bility for a key instrument of macroeconomic poli-cy passed into the hands of an independent supra-national authority. This article describes the frame-work for monetary policy in this new environmentand, in the context of this framework, describes theimplementation of monetary policy in 1999.

The benefits of price stability and implications formonetary policy

A broad consensus has emerged over recentdecades that the appropriate objective of monetarypolicy is the maintenance of price stability (Blinder,

1998). This consensus is built on the belief that both

inflation and deflation are costly in terms of gener-

al economic welfare and performance. Among oth-

ers, Barro (1997) provides macroeconomic evidence

to this effect. In large samples of cross-country data,

he has demonstrated a negative relationship

between inflation and economic growth.

Explanations of this macroeconomic evidence are

based on the view that inflation introduces or exac-

erbates distortions in the real economy. High rates

of inflation are typically associated with greater

volatility of inflation and the price level. This

volatility distorts the relative price signals on

which the market mechanism relies and raises the

inflation risk premium in long-term real interest

rates. Both phenomena may result in a misalloca-

tion of real resources, which is prejudicial to

growth performance (ECB, 1999a). High rates of

inflation may also distort money holdings

(Friedman, 1956) and exacerbate the distortions

introduced into economic allocation by the dead-

weight losses associated with tax and welfare sys-

tems (Feldstein, 1995). Furthermore, unexpected

movements in prices associated with high and vari-

able inflation may result in large and arbitrary re-

distributions of wealth between creditors and

debtors, inter alia. Such arbitrary redistribution

may threaten social and political instability if prop-

erty rights are perceived to have been violated.

The institutional framework for the single mone-

tary policy, described in the Treaty establishing the

European Community, reflects these economic

principles. First and foremost, the Treaty clarifies

the objective of the single monetary policy and the

Eurosystem. Article 105 of the Treaty states: “The

THE MONETARY POLICY OF

THE EUROPEAN CENTRAL

BANK:STRATEGY AND IMPLEMENTATION

OTMAR ISSING*

With the start of the year 1999 the European Monetary Union went into effect with its central institution,the European Central Bank, and its single currency, the euro. After more than a year in existence, the eurohas depreciated by more than 17% against the US dollar. In some quarters this has raised questions aboutthe wisdom of a common monetary system for Europe; more generally it has led to concerns about thepolicies of the European Central Bank, and about the international role of the Euro.The following contributions address these concerns. The editor is grateful to the authors for their gener-ous permission to make them the focus of the second edition of CESifo Forum.

O

* Professor Otmar Issing is a member of the Executive Board ofthe European Central Bank. The author is grateful to Huw Pill forhis valuable contribution.1 The Governing Council consists of the eleven NCB Governors ofthe countries that adopted the euro in January 1999, plus the sixmembers of the ECB’s Executive Board. The geographical areadefined by the countries that adopted the euro is called the euroarea. Collectively, the ECB and the NCBs of these eleven countriesare labelled the Eurosystem.

primary objective of the [Eurosystem] shall be to

maintain price stability.” The Treaty thereforeestablishes a clear hierarchy of objectives for thesingle monetary policy, with price stability unam-biguously assigned overriding importance.

Furthermore, recognising that monetary policydoes not operate in a vacuum, the Treaty alsorequires the Eurosystem – insofar as this does notprejudice the primary objective of price stability to“support the general economic policies in the

Community with a view to contributing to the objec-

tives of the Community laid down in Article 2”. Theobjectives of the Community include, inter alia,“sustainable and non-inflationary growth” and “a

high level of employment”.

The scope for monetary policy to pursue “growth”or “a high level of employment” is limited. Theseminal contribution of Friedman (1968) re-estab-lished the principle that monetary expansions areneutral in the long run. Other than by maintainingprice stability and thereby reaping its benefits interms of economic performance discussed above(as Friedman (1977) discussed in his Nobel lec-ture), there is no trade-off at longer horizonsbetween inflation, on the one hand, and economicgrowth or employment, on the other, that can beexploited by monetary policy makers. The bestcontribution that monetary policy can make to thefulfilment of the Community’s broader economicand social objectives is to maintain price stabilityin a credible and lasting manner, thereby securingits benefits and providing an environment in whichthe private sector and other policy authorities canoperate most effectively and efficiently.

Given the potential risk of political intervention inthe design and implementation of the single mone-tary policy (reflected in the time inconsistency litera-ture, e.g. Barro and Gordon (1983)), the Treaty alsomade the ECB and the NCBs independent of nation-al governments and political interference. This insti-tutional independence allows the Eurosystem to pur-sue its primary objective in an appropriate medium-term framework and thereby significantly enhancesthe credibility of monetary policy. In fact, because theindependence of the Eurosystem is guaranteed by aninternational treaty, it can claim to be among themost independent central banks in the world.

In a democratic society, institutional independenceof the central bank must be balanced by trans-

parency and accountability in monetary policymaking. The Treaty imposes stringent requirementson the Eurosystem in this regard, which exceedthose required of most other central banks in theworld. As is described in greater detail by Issing(1999), the ECB has committed itself to exceedingeven these stringent requirements. Foremostamong the vehicles used for communication withthe public is the President’s introductory statementat the regular monthly press conference. This state-ment provides unparalleled timeliness and open-ness regarding policy decisions and their rationale.The introductory statement is complemented bythe publication of a Monthly Bulletin, the AnnualReport and regular appearances by members ofthe Executive Board before the competent com-mittees of the European Parliament.

The Eurosystem’s monetary policy strategy

In October 1998 the Governing Council of the ECBannounced the Eurosystem’s stability-oriented mon-etary policy strategy. The strategy consists of a quan-titative definition of the primary objective of mone-tary policy and “two pillars” that are used to assessrisks to future price stability, namely a prominentrole for money and a broadly based assessment thatencompasses a wide range of indicator variables.

The quantitative definition of price stability

To quantify its primary objective given by the Treatymore precisely, the Governing Council announcedthe following definition: “price stability shall be

defined as a year-on-year increase in the Harmonised

Index of Consumer Prices (HICP) for the euro area

of below 2%”. Price stability according to this defi-nition “is to be maintained over the medium term”

(ECB, 1999b). The Eurosystem’s published defini-tion of price stability gives guidance to expectationsof future price developments, helping to build upthe credibility of the new strategy. Moreover, thisdefinition provides a yardstick against which thepublic can hold the Eurosystem accountable.

The phrase “below 2%” clearly delineates theupper bound for the rate of measured inflation inthe HICP which is consistent with price stability.At the same time, the use of the word “increase” inthe definition clearly signals that deflation, i.e. pro-longed declines in the level of the HICP index,would not be deemed consistent with price stabili-

CESifo Forum 4

Focus

In the long run,there is no trade-off

between inflationand growth

CESifo Forum5

Focus

Money is the “first pillar” of theEurosystem’s monetary policy

ty. Moreover, the definition identifies a specificprice index to be used in the assessment of whetherprice stability has been maintained. This index isharmonised across the various countries in theeuro area and its use is consistent with the public’susual focus on consumer prices. This focus is obvi-ously on the HICP for the euro area, since mone-tary policy decisions are based on an assessment ofdevelopments in the euro area as a whole.

The absence of an explicit numerical lower bound tothe published definition of price stability reflects, ina transparent and honest manner, the uncertaintiesfaced by the Eurosystem regarding the potentialexistence and unknown magnitude of so-called“measurement bias” (e.g. Boskin, 1996) in the HICPat the outset of Stage Three. Because the HICP is anew concept and long runs of back data do not exist,estimates of the HICP measurement bias are pre-liminary and inconclusive. The absence of an explic-it lower bound should not be viewed as suggestingthat the Eurosystem is indifferent to deflation. Inpractice, a “grey zone” exists – as expected inflationfalls towards zero, the Eurosystem becomes increas-ingly concerned about the development. At theupper bound, in order to ensure that price stabilityis maintained according to the Eurosystem’s defini-tion in the face of the inevitable shocks to pricedevelopments, policy makers need to ensure thatinflation is normally likely to be safely below 2%.

The statement that “price stability is to be maintained

over the medium term” reflects the need for monetarypolicy to have a medium-term orientation. Further-more, in response to some types of unforeseen eco-nomic disturbance with an impact on the price levelthat may threaten price stability, a medium-term ori-entation of monetary policy is important in order topermit a gradualist and measured response, which willnot introduce unnecessary and possibly self-sustain-ing uncertainty into the real economy.

In the context of the Eurosystem’s strategy, interestrates are set so as to achieve the primary objective(defined as described above) on the basis of informa-tion about the outlook for price developments overthe medium term revealed by analysis undertaken inthe context of the “two pillars” of the strategy.

The reference value – a prominent role for money

In the light of a large body of empirical and theo-retical evidence, there is general agreement in the

academic literature with – when it is properlyinterpreted – Friedman’s (1956) famous assertionthat “inflation is always and everywhere a mone-tary phenomenon”. This points to ensuring thatmonetary aggregates are thoroughly incorporatedinto the Eurosystem’s strategy. The GoverningCouncil therefore assigns a prominent role tomoney within the strategy. Money constitutes anatural “nominal anchor” for monetary policy aim-ing at the maintenance of price stability.

The prominent role for money – the “first pillar” ofthe Eurosystem’s monetary policy strategy –embodies a commitment to analyse monetarydevelopments in detail for the information thatthey contain about future price developments. Tosignal the prominent role assigned to money, theGoverning Council announced a quantitative refer-

ence value for monetary growth. The referencevalue is intended to help the Governing Councilanalyse and present the information contained inthe monetary aggregates in a manner that offers acoherent and credible guide for monetary policyaimed at the maintenance of price stability overthe medium term (ECB, 1999c).

Two characteristics of the quantitative referencevalue for monetary growth should be emphasised.First, the reference value is derived in a manner thatis consistent with – and serves the maintenance of –price stability over the medium term. To ensure thisconsistency, money must have a stable relationshipwith the euro area price level at this horizon. Second,substantial or prolonged deviations of monetarygrowth from the reference value would, under nor-mal circumstances, signal risks to medium-term pricestability. This feature of the reference value is basedon the evidence that monetary growth is normally aleading indicator of future developments in the pricelevel. The available empirical evidence suggests thatbroad monetary aggregates exhibit the propertiesrequired for the announcement of a reference value.Therefore, the Governing Council chose to announcea reference value for the broad aggregate M3.

The reference value for monetary growth wasderived using the well-known quantity relationshipbetween money, on the one hand, and prices, realGDP and the velocity of circulation, on the other.Using the Eurosystem’s definition of price stabilityand medium-term assumptions for real GDP (trendgrowth in the range 2% to 21/2% per annum) andM3 income velocity (a trend decline of between

– 1/2% and – 1% per annum), in December 1998 theGoverning Council decided to set its first referencevalue for M3 growth at 41/2% per annum. Monetarydevelopments relative to the reference value areassessed on the basis of three-month moving aver-ages of the annual growth rates. The reference valueof 41/2% was confirmed in December 1999.

A broadly based assessment of the outlook for

price developments

Although the monetary data contain informationvital to informed monetary policy-making, on theirown they will not constitute a complete summary ofall the information about the economy required toset an appropriate monetary policy which maintainsprice stability. Therefore, in parallel with the analy-sis of the monetary data, a broadly based assess-ment of the outlook for price developments and therisks to price stability in the euro area – the “secondpillar” – play a major role in the Eurosystem’s strat-egy. This assessment is made using a wide range ofnon-monetary economic indicators.

This broad range of indicators includes: labour mar-ket indicators, such as wages and unit labour costs;fiscal policy indicators; financial market indicators,such as asset prices, etc. One important indicator isthe exchange rate of the euro. It should be empha-sised that the Eurosystem’s strategy embodies nei-ther an explicit nor an implicit objective for the euroexchange rate. This does not mean that the euroexchange rate is treated with neglect. Rather it isclosely monitored and analysed and influences mon-etary policy decisions insofar as it has implicationsfor the outlook for price developments in the euroarea. Fluctuations in the exchange rate will affect theoutlook for price developments both directly(through their impact on import prices) and indi-rectly (through competitiveness effects and henceaggregate demand). Consequently, the euroexchange rate is, in fact, an important indicator with-in the second pillar of the Eurosystem’s strategy.

As with the reference value for monetary growth,the broadly based assessment is not intended todefine a quasi-automatic feed-back rule for interestrate decisions. Rather the second pillar is a frame-work or process for organising and analysing infor-mation, so that policy makers can make an assess-ment of the appropriate interest rate which will bestserve the maintenance of price stability. Althoughanalysis under the second pillar of the strategy

includes an assessment of macroeconomic forecastsfor the euro area, including those produced by theEurosystem itself, these should clearly not be seenas constituting the second pillar in its entirety, stillless a “sufficient” or “summary” statistic of all theinformation that policy makers require for takingappropriate monetary policy decisions. Forecastscertainly play a role in the monetary policy process,as one would expect given the forward-looking ori-entation of the Eurosystem’s strategy. They help tosummarise and synthesise a large quantity of infor-mation that may otherwise become too unwieldy toform a sensible basis for policy discussions.

However, forecasts also suffer from a number ofdrawbacks. For example, they may quickly becomeoutdated, since its is difficult for them to incorpo-rate all information in a timely manner. Forecastsare based on assumptions for certain variables,such as oil prices or developments in the worldeconomy, which can quickly become out-of-date.The final projection is then an incomplete summa-ry of the outlook for price developments.Furthermore, precisely because they synthesiseinformation, forecasts can sometimes obscureinformation about the individual threats to pricestability on which the appropriate monetary policyresponse would normally depend.

The operational framework for the single monetary policy

The operational framework for monetary policy pro-vides the instruments to guide the level of marketshort-term interest rates to the level that theGoverning Council deems best serves the mainte-nance of price stability (see also ECB, 1998). Thisframework fulfils three basic functions. First, it pro-vides instruments that signal clearly the stance ofmonetary policy. Second, it ensures that policy mak-ers can steer money market interest rates and containtheir volatility. Finally, the operational frameworkprovides basic refinancing to the financial system,ensuring that sufficient liquidity is available. Threetypes of monetary policy instruments are available tothe Eurosystem: open market operations, standingfacilities and a minimum reserve system.

The major open market operation is the weeklymain refinancing operation (MRO), which takesthe form of a reverse repurchase transaction with amaturity of two weeks. The main refinancing oper-

CESifo Forum 6

Focus

Assessment of the outlook for prices is

the “second pillar”

CESifo Forum7

Focus

The operational framework includesopen-market operations, a monthly longer-termrefinancing opera-tion, two overnightstanding facilities,and a minimumreserve system

ation is based on a tender procedure. The tender

may be a fixed rate tender, with counterparties bid-

ding amounts at an interest rate pre-specified by

the ECB (as has been the case during 1999), or a

variable rate tender, where counterparties propose

bids including both amounts and interest rates.

The operational framework also includes a regular

monthly longer-term refinancing operation. This has

a maturity of three months and normally takes the

form of an interest rate tender with a pre-

announced absolute amount for the allotment. This

ensures the Eurosystem does not signal its mone-

tary policy stance through these operations. The

Eurosystem is also equipped to conduct fine-tuning

operations (through the national central banks of

the euro area or, in exceptional circumstances, cen-

trally) and structural operations (which may take

the form of outright purchases or sales of securities

or the issuance of debt certificates by the ECB).

The ECB also operates two overnight standing

facilities, the deposit facility and the marginal lend-

ing facility, which are available to all credit institu-

tions at national central banks of the euro area.

The rate of the marginal lending facility constitutes

the upper bound of collateralised overnight money

market rates. The deposit facility is remunerated at

a rate that constitutes the lower bound of

overnight money market rates.

When using the marginal lending facility, or, for that

matter, when entering in liquidity-providing open

market operations in the form of reverse transac-

tions, counterparties have to post assets as collater-

al. These assets are meant to act as guarantees for

credits received from the Eurosystem. A list of eli-

gible assets has been drawn up for this purpose. So-

called tier one assets have been selected by the ECB

according to uniform criteria and contain mar-

ketable paper of high quality. Tier two assets have

been selected by the ECB because they are of par-

ticular importance for certain national banking sys-

tems in the euro area and promote a certain degree

of continuity at the start of the Monetary Union

with the national operational frameworks that exist-

ed prior to the introduction of the euro. Tier two

assets need to meet similar quality standards as tier

one assets. Both tier one and tier two assets may be

used by any credit institution in the euro area, irre-

spective of its location. A set of risk control mea-

sures has been elaborated to ensure that, for any

counterparty, the amount of assets provided as col-lateral is always sufficient.

The ECB also applies a minimum reserve system tocredit institutions in the euro area. This system helpsto stabilise money market interest rates through anaveraging mechanism, whereby the fulfilment ofminimum reserve requirements is based on averagereserve holdings over a month-long maintenanceperiod (normally ending on the 23rd of the month).During the maintenance period, averaging allowsbanks to absorb liquidity shocks without the need touse the standing facilities, thereby stabilising interestrates and reducing the need for frequent fine tuningoperations by the Eurosystem.The minimum reservesystem also increases the demand for central bankmoney and thus enlarges the liquidity deficit of thebanking system vis-à-vis the Eurosystem, ensuringthe role of the Eurosystem as a provider of liquidityto the banking system.

Reserve requirements are calculated by applying areserve ratio of 2% to the deposits, debt securitiesand money market paper issued by credit institu-tions, excluding those instruments with maturitygreater than two years. Although repurchaseagreements are included in the reserve base, theyare subject to a zero reserve ratio. Inter-bank lia-bilities and liabilities vis-à-vis the Eurosystem arenot subject to reserve requirements. A lump sumallowance is deducted from the reserve require-ments of each individual institution, implying cred-it institutions with a small reserve base do not haveto hold minimum reserves. Reserve holdings up tothe required reserve level are remunerated at themarginal rate of the main refinancing operation(averaged over the maintenance period).

Implementation of monetary policy during 1999

Although the single monetary policy officially cameinto force on 1 January 1999, monetary policy co-ordination was extensive prior to the formal transi-tion to Stage Three. This assessment of monetarypolicy implementation therefore starts with the co-ordinated interest rate reduction by the NCBs inwhat is now the euro area on 3 December 1998. Thisco-ordinated interest rate cut took note of theimplications for future price developments of theapparent weakening of economic activity in theeuro area, revealed first by declining industrial con-fidence and then by the emergence of the first signs

of a slowdown in industrial production. Turmoil infinancial markets associated with the devaluationof the Russian rouble in August 1998, at a timewhen financial markets were already unsettled bythe Asian financial crisis, spread concerns of a cred-it crunch. The wealth effects of a potential assetprice collapse and financial instability in the UnitedStates also threatened the global outlook. Againstthis background, projections for world outputgrowth were revised downwards in late 1998, weak-ening prospective economic growth and inflation inthe euro area. Therefore, at a moment when actualinflation in the euro area was around 1% and mon-etary growth and other indicators were in line witha subdued outlook for price developments, it wasdeemed appropriate to reduce the level of keyinterest rates in the euro area to a common level of3% prior to the start of Stage Three.

The interest rates on the ECB’s monetary policyinstruments applying at the start of Stage Threewere then officially set on 22 December 1998 andfollowed those prevailing at the euro area centralbanks at the end of Stage Two. The rate on themain refinancing operation was set at 3%, the rateon the marginal lending facility at 4.5%, and thaton the deposit facility at 2%. However, in order tosmooth the transition for the banking sector, theGoverning Council of the ECB set a “narrow cor-ridor” for short-term interest rates during the firstthree weeks of January 1999 by setting the interestrates on the marginal lending facility and thedeposit facility at 3.25% and 2.75% respectively.On 22 January 1999, these transitional arrange-ments expired and the marginal lending rate andthe deposit rate were effective at 4.5% and 2.0%.

In the first few months of 1999, signs emerged thatthe extent of the slowdown of economic activity inthe euro area was stronger than had been antici-pated in December 1998. In line with this, pricepressures continued to be weak. Headline HICPinflation in December 1998 was only 0.8%, andremained at that level in January and February1999. Figures on economic activity that becameavailable in the first months of 1999 all pointed toa significant economic slowdown in late 1998. RealGDP growth had weakened significantly in the lastquarter of 1998; industrial production was alsoweakening and business confidence continued todecline. It thus became increasingly clear thateffects stemming from the slower than projectedgrowth of the euro area economy – mainly caused

by weaker external demand – represented a seri-ous downward risk to price stability.

A monetary policy reaction to these downwardrisks to price stability was complicated, however,by the fact that some indicators appeared to pointin the opposite direction in early 1999. In particu-lar, M3 growth at the start of 1999 was slightlyabove the reference value. The January datashowed a significant increase in overnight depositswhich was only partially corrected in February. Itwas also notable that credit to the private sectorwas growing relatively fast.

The Governing Council was thus faced with a verydifficult situation in early 1999. With respect tomonetary developments it had, however, many rea-sons not to be too concerned about upside risks toprice stability. Monetary growth was then still closeto the reference value (the three-month average ofthe annual growth rates for the period December1998 to February 1999 was 5.1%). In addition, itappeared that the changeover to Stage Three hadcontributed significantly to the high increase inovernight deposits in January 1999 and it could notbe ruled out that institutional factors, such aschanges in the statistical reporting systems or inreserve requirements, had played a role in the risein monetary growth in that month. Given the mod-eration of monetary growth in February 1999, theGoverning Council did not regard monetary devel-opments in early 1999 as implying upward risks toprice stability.

Against this background, in an environment wherecurrent inflation rates were significantly below theupper limit of the Eurosystem’s definition of pricestability and in view of downward pressures onfuture price developments associated with the cur-rent and anticipated weakening of economic activi-ty, the Governing Council decided on 8 April 1999to reduce the main refinancing rate by 50 basispoints to 2.5%. On the same occasion, the Councillowered the rate on the marginal lending facility to3.5% and that on the overnight deposit facility to1.5%. These moves were regarded by the Council asappropriate to preserve price stability in the medi-um term by contributing to improved business con-fidence and the better exploitation of the growthpotential of the euro area economy.

Following this move, during the summer of 1999business sentiment improved and the outlook for

CESifo Forum 8

Focus

At the start of StageThree, interest rates

were set at 3% onthe main refinancing

operation, 4.5% onthe marginal lending

facility, and 2% onthe deposit facility

CESifo Forum9

Focus

As the balance ofrisks increased, theECB raised its interest rates in late1999

real activity recovered in the euro area. At thesame time, the external environment strengthenedas the Asian economies stabilised and then startedto recover, and financial markets stabilised.Overall, it became progressively more evident thateconomic activity in the euro area was set to accel-erate significantly in the second part of 1999 and inthe year 2000. Further analysis under the secondpillar also pointed towards a shift towards theupside in the balance of risks to price stability.More positive figures coming from confidenceindicators were accompanied by evidence thatindustrial production had stabilised in the firstquarter of 1999 and slightly increased in the secondquarter, while data on euro area real GDP indicat-ed that some positive developments had alreadyoccurred in the first quarter of 1999. In addition,the continued weakening of the effective exchangerate and the further rises in oil prices were gradu-ally feeding their way through to consumer prices.

With regard to the first pillar of the ECB’s strate-gy, an upward shift of risks to price stability overthe summer of 1999 was indicated by monetarydevelopments. Annual M3 growth was on a moreprolonged upward trend in 1999, with the three-month average of annual growth rates for July toSeptember 1999 reaching around 6%. Even whenexcluding the exceptional developments in Januaryand February 1999, monetary growth over the sum-mer was significantly above 41/2% when examiningannualised growth rates of seasonally adjusted M3data at shorter horizons. In parallel, credit to theprivate sector continued to expand at a fast rate, ofabout 10%. As the balance of risks continued tomove upwards towards the autumn of 1999, on4 November 1999 the Governing council of theECB decided to raise the rate on the main refi-nancing operations to 3%. On the same occasion,the rates on the deposit facility and the marginallending facilities were raised to 2% and 4% respec-tively. This level of interest rates was then main-tained until the end of 1999.

Concluding remarks

The success of the Eurosystem’s strategy can onlybe assessed over the medium term. Given the lagsin monetary policy transmission to the price level,it is far too early to assess whether monetary poli-cy based on the Eurosystem’s strategy has success-fully maintained price stability. Nevertheless, the

currently available information on the outlook forprice stability gives a very positive judgement onthe single monetary policy in the first year.

ReferencesBarro, R.J. (1997). Determinants of economic growth:

A cross-country empirical study, MIT Press.Barro, R.J. and R. Gordon (1983). “Rules, discretionand reputation in a model of monetary policy”.Journal of Monetary Economics, vol. 12, pp. 101–21.Blinder, A.S. (1998). Central banking in theory and

practice, MIT Press.Boskin, M. (1996). Toward a more accurate measure

of the cost of living. Final report to the U.S. SenateFinance Committee from the Advisory Commissionto Study the Consumer Price Index.ECB (1998). “The single monetary policy in StageThree: General documentation on ESCB monetarypolicy instruments and procedures”, Frankfurt amMain.ECB (1999a). “Stability-oriented policies anddevelopments in long-term real interest rates inthe euro area”. ECB Monthly Bulletin, November,pp. 31–40.ECB (1999b). “The stability-oriented monetarypolicy strategy of the Eurosystem”. ECB MonthlyBulletin, January, pp. 39–50.ECB (1999c). “Euro area monetary aggregates andtheir role in the Eurosystem’s monetary policy strat-egy”. ECB Monthly Bulletin, February, pp. 29–46.Feldstein, M. (1995). “The Costs and Benefits ofGoing from Low Inflation to Price Stability” ineds. C. Romer and D. Romer Reducing inflation:

Motivation and strategy, Chicago University Press.Friedman, M. (1956). “The quantity theory ofmoney: A restatement” in Studies in the quantity

theory of money, University of Chicago Press.Friedman, M. (1968). "The role of monetary poli-cy”. American Economic Review, vol. 58, pp. 1–17.Friedman, M. (1977). “Nobel lecture: Inflation andunemployment”. Journal of Political Economy, vol.85 (3), pp. 451–72.Issing, O. (1999). “The Eurosystem: Transparentand accountable”. Journal of Common MarketStudies, vol. 39 (3), pp. 503–19.

ith the launch of European monetaryunion, a new era began for Europe last

year. On 1 January, 1999, the euro became themonetary standard for eleven of the fifteen mem-ber states of the European Union, and thus for afinancial and economic area which, in terms of itseconomic performance, is second only to theUnited States of America. On the same date,responsibility for the single monetary policy wastransferred to the Governing Council of theEuropean Central Bank, which has decided theappropriate level of official interest rates in theeuro area ever since.

The Governing Council has been responsible forEuropean monetary policy for only a year. Hence itis still much too early to pass a mature verdict onthe success of the secular project of European mon-etary union. Looking back on the first year, howev-er, it can be said that the Eurosystem has alreadypassed quite a number of significant tests: cross-bor-der payments and money market management areoperating almost smoothly, the new range of mone-tary policy instruments has proved its worth, andinterest-rate policy measures have manifestly beengeared to maintaining the high degree of price sta-bility that has already been reached. The membersof the Governing Council have demonstrated thattheir attention is focused on the euro area as awhole and not on individual countries.

Even if the Eurosystem has already successfullyaddressed quite a number of problems, today and inthe years ahead it still has to meet a lot of chal-lenges. At the top of the agenda is grappling withthe criticism of the monetary policy strategy it ispursuing and with the charge of a lack of trans-

parency. In the present article, I shall ask whethersuch charges are warranted, and what the EuropeanCentral Bank can do to clear up any remaininguncertainties and problems of understanding. In thisconnection, I should like to address both, the debateabout the objectives of monetary policy and the dis-cussion on the strategy adopted to achieve thoseobjectives. Although my remarks will focus mainlyon the strategic challenges, I should also like to takethis opportunity of outlining, in the final section ofmy article, the logistical and institutional challengeswith which the Eurosystem will likewise have tocontend in the next few years.

The objectives of monetary policy

In most industrialised countries it is generally recog-nised, in the light of the experience gained duringthe seventies and early eighties, that safeguardingprice stability is the best contribution which mone-tary policy can make to long-run macroeconomicwelfare. The Maastricht Treaty spells out that objec-tive with exemplary clarity for the European Systemof Central Banks. The assignment of responsibilitiesis likewise suitably unambiguous: in pursuing theirprimary objective, the monetary decision-makers areindependent of instructions from national govern-ments or comparable EU authorities.

Despite these unequivocal stipulations, theMaastricht Treaty leaves unresolved a number ofissues which have increased in significance, given thehigh level of price stability achieved in recent years.For instance, in the last few years there has been alively discussion as to whether the optimal level ofinflation should be set at zero or at slightly abovezero. Moreover, the controversy on the contributionthat monetary policy can and should make to foster-ing growth and employment in an environment offairly low inflation rates has flared up time and again.

The optimal rate of inflation

The Treaty of Maastricht does not give a precisedefinition of what is to be understood by price sta-

CESifo Forum 10

Focus

The Eurosystem has already passed

several tests

W

FUTURE CHALLENGES

FOR THE ECB

HERMANN REMSPERGER*

* Professor Hermann Remsperger is a member of the Board of theDeutsche Bundesbank.

CESifo Forum11

Focus

Should output stabilisation be asecondary target ofmonetary policy?

bility. In order to fill this gap, the ECB GoverningCouncil published a quantitative definition of pricestability in the autumn of 1998. According to thatdefinition, price stability is to be equated with ayear-on-year increase of less than 2% in the har-monised consumer price index for the entire euroarea with price stability to be maintained over themedium term. Since this definition of price stabilityrules out both a decline in consumer prices and anincrease of more than 2%, it can be interpreted as atarget corridor of between just over zero and under2%. This objective has been criticised as being tooambitious by some observers. Other critics wouldhave preferred a point target to a corridor.

One reason for setting the floor somewhat abovethe zero level is the supposed overstating of theactual inflation rate by the measured rate. Most ofthe empirical studies which address this problemcome to the conclusion that official price indicesare apt to slightly overstate the “true” inflationrate (as measured by a cost-of-living index).1

Whether there are other reasons for setting a posi-tive optimal inflation rate, besides the problem ofmeasurement errors, is open to debate. Some well-known US economists have claimed that down-wardly rigid nominal wages may give rise to a per-manent trade-off between inflation and unemploy-ment in the event of inflation rates near zero.2

However, the assumption underlying this argu-ment, that labour-market players are subject tomoney illusion, is contrary to experience – at leastin Germany. Instead, it is likely that nominal wagereductions are easier to push through under condi-tions of price stability than in an inflationary envi-ronment. Moreover, improvements in productivity,even at times of constant nominal wages, makeroom for reductions in unit labour costs. The refer-ence to an empirically verifiable connectionbetween disinflation and an increase in unemploy-ment in a number of European countries likewisefails to convince.3 It is unlikely that opposition tonominal wage cuts might have played a major rolehere, if only because inflation rates have not beenvery close to zero for a long while in any countryduring the periods under consideration.

A further argument is that only a positive inflationtarget would enable the central bank to stimulateaggregate demand by setting negative real interestrates. Since the level of nominal interest rates hasits floor at zero, that is also where the limit for thereal interest-rate level is, given an expected infla-tion rate of zero. Thus, it is feared, monetary policywould lose all scope for exerting influence in theevent of a profound and sustained recession.Against this argument, it may be objected, firstly,that a forward-looking policy that responds ingood time to prospective distortions should nor-mally be able to generate sufficiently expansionaryeffects by means of low positive real interest rates.4

Secondly, the empirical evidence suggests thatmoney demand is not infinitely elastic when theinterest-rate level is zero. Hence, even when cen-tral bank rates reach zero, a central bank still hasthe option of stimulating demand and output byenlarging the monetary base.5

The role played by output stabilisation

The priority commitment of the Eurosystem to thefinal goal of price stability is generally accepted.But there is far less agreement about the extent towhich the objective of output stabilisation shouldbe included in the monetary decision-makingprocess as a “secondary target” if the prime objec-tive of price stability is not at risk. In principle, ananticyclical monetary policy is not inconsistentwith the preservation of price stability in the medi-um run. In practice, however, the stabilisationoptions of central banks are quite limited. This isbecause, in the first place, our knowledge about thetransmission mechanisms of monetary stimuli vis-à-vis real activity continues to be very sketchy.Furthermore, uncertainties of diagnosis are espe-cially large in the real sector of the economy.

If the disturbance of macroeconomic equilibrium isdue to an unexpected increase or decrease inaggregate demand, then there is no trade-off any-way between output stabilisation and safeguardingprice stability. If, however, the cause is a supply-side shock, the central bank theoretically has achoice between a swift stabilisation of inflation at

1 On this point see Hoffmann, J.: Problems of InflationMeasurement in Germany, discussion paper 1/98 of the EconomicResearch Group of the Deutsche Bundesbank, 1998.2 See Akerlof, Dickens and Perry (1996):The macroeconomics of lowinflation, Brookings Papers on Economic Activity, No.1, pp. 1–76, andKrugman (1996): Stable prices and fast growth: just say no, TheEconomist, August 31, pp. 19–21.3 See IMF:World Economic Outlook – Prospects and Policy Issues,autumn 1999, p. 124.

4 In recent years there have been quite a number of simulation stud-ies on the question of how great the likelihood is that the zero limitwill be reached. An overview of such literature is provided byJohnson et al. (1999): Monetary Policy and Price Stability, Board ofGovernors of the Federal Reserve System, International FinanceDiscussion Papers, No. 641.5 See King (1999): Challenges for monetary policy: new and old,Bank of England Quarterly Bulletin, November 1999, p.408 and thestudies cited there.

the price of higher output variability, on the onehand, and a reduction of output fluctuations at thecost of higher inflation variability, on the other. Anoptimisation of this trade-off which seems feasiblein theoretical models founders in the real world onuncertainty about the transmission mechanismsand about the reactions of the market playersaffected by monetary policy.

Simple stabilisation rules, such as the well-knownTaylor rule, may at first sight create the impressionof greater robustness. However, their use likewisepresupposes that the “true” level of output relativeto trend is known at the time of the policy decision.6

As a more recent study by Orphanides (1998)shows, the relatively good performance of theserules with regard to output stabilisation turns out tobe an illusion if the unreliability of real-time GDPdata and potential estimates is taken into account.7

Hence there are good reasons why the GoverningCouncil of the ECB opted for a monetary strategythat underlines the responsibility of monetary poli-cy for medium-term price movements. However, theunambiguous orientation towards the preservationof price stability does not mean that interest-ratedecisions in the euro area are taken in total disre-gard of real economic activity. Price prospects in theeuro area are influenced to some extent by trends inbusiness activity. If those trends, together with otherindicators – above all the money stock – suggest thatprice prospects are changing, action must be taken.Viewed in these terms, references to the overall eco-nomic situation by no means call into question theEurosystem’s orientation towards the objective ofprice stability. Instead, they form a building block inthe second pillar of its monetary policy strategy.

Moreover, with the orientation towards the growthof the production potential in the derivation of thereference value for monetary growth, the first pil-lar of the Eurosystem’s monetary policy strategyalso includes a “quasi-automatic” anticyclical com-ponent: if the actual increase in GDP falls short oftrend growth, ample money will tend to be sup-plied. If, by contrast, actual economic growthexceeds the medium-term growth potential, thenthe expansion of the money stock will tend to betight.

Monetary policy and the “new economy”

However, the growth of the production potential isnot easy to ascertain. This is shown not least by thevigorous debate on whether the pronounced expan-sion of the US economy since 1991 marks the startof a new era distinguished by consistently fastergrowth at lower rates of inflation. Against the back-ground of the sharp increases in productivity sincethe mid-nineties, accompanied by diminishing ratesof inflation, the advocates of the “new economy”approach assume that faster, non-inflationarygrowth is to be expected in future as well. This ideais substantiated, in particular, by reference to theprice-curbing effects of global competition and tothe strong growth potential of new, computer-aidedtechnologies. On the other hand, sceptical observerswarn against overrating the growth potential of thenew technologies. They point out that the impres-sive performance of the US economy in recent yearsowes much to a combination of particularlyfavourable circumstances, such as the strength ofthe dollar due to its “safe haven” function in thewake of the east Asian and Russian crises and to theconsumption-boosting boom (which is unsustain-able in the long run) on the US stock exchange.

With regard to the euro area, the question arises asto how good the chances are that the upswing nowtaking shape will usher in a new era of consistentlyhigher non-inflationary growth. This question is rel-evant to monetary policy because the derivation ofthe reference value for the growth of the monetaryaggregate M3 calls for an appraisal of the trend ofreal growth. When deriving the reference values for1999 and 2000, the ECB Governing Councilassumed that the trend growth rate in the euro areais currently running at between 2 and 21/2%. Thiscorridor is in line both with the results of internalforecasts and with the estimates of other institu-tions. But, when announcing the reference value for2000, the Governing Council of the ECB explicitlypointed out that the growth rate in the euro areathat was compatible with price stability could beenhanced distinctly by structural reforms on thelabour and goods markets. The Council envisagedtaking due account of any such changes.

However, this approach is not uncontroversial.Critics have accused the Governing Council of undu-ly great caution. They have suggested that monetarypolicy should make an active contribution to foster-ing growth and employment by refraining from rais-

CESifo Forum 12

Focus

Stabilisation rulessuffer from theunreliability of

real-time GDP dataand estimates of

potential GDP

6 On this point, see the article ”Taylor interest rate and MonetaryConditions Index” in: Deutsche Bundesbank, Monthly Report,April 1999.7 See Orphanides (1998): Monetary Policy Evaluation with NoisyInformation, Working Paper, Board of Governors of the FederalReserve System, October.

CESifo Forum13

Focus

ECB monetary policyis neither requirednor able to remedyinflation differentialsin EMU

ing interest rates for the time being. However, thatsuggestion disregards the fact that, over the longerterm, there is no trade-off between price stability andreal growth. Any attempt to increase economic activ-ity over and above the pace of potential GDP will inthe short run lead to a rise in aggregate demand andoutput but, at given capacity levels, the pressure onwages and prices will likewise rise. In order to getaccelerating inflation back under control again, thecentral bank will have to step on the brakes sooner orlater; the upshot is a “boom and bust” cycle, whichactually impairs the medium-term growth andemployment outlook. Against this background, suchwell-known US economists as John B. Taylor andMartin Feldstein have drawn attention to the factthat it was specifically the early adoption by the Fedof a more restrictive policy stance during 1994 thatwas a key prerequisite of the longevity of the currentupswing in the US economy.8

On the other hand, a policy of “testing” the limits togrowth by refraining from raising interest rates forthe time being would imperil the advantages associ-ated with price stability. Moreover, it would be atvariance with the assignment of responsibilities laiddown in the Maastricht Treaty, which flatly rejectsany involvement of monetary policy in other areas ofeconomic policy if such involvement poses any riskto price stability. On closer inspection, it should beobvious that monetary policy cannot solve the gravestructural problems that mark the employment situ-ation in many member states of the monetary union.It is up to national governments and the parties towage settlements to remedy such problems.

Inflation differentials within the euro area

In recent months, the question of whether nationalgrowth and inflation differentials pose a risk to thesingle European monetary policy has also attractedsome attention. This discussion was triggered by adrifting-apart of the growth and inflation perfor-mances of particular countries, which has been dis-cernible since mid-1998. For instance, the rise inconsumer prices in Ireland and Spain in autumn1999 was distinctly above the 2% mark, whereasthe inflation rate in Germany, France and Austriaremained below 1% into the late autumn.However, the spread visible at present, of about2%, is much narrower than it was during the eight-ies. Besides, a comparison with the United States

shows that regional divergences on this scale arequite normal, even in a long-standing monetaryunion.9

The current inflation differentials in EMU aremainly due to two factors, according to studies bythe ECB.10 In the first place, the differentials owemuch to differences in the cyclical positions of thevarious economies. Secondly, the harmonisation ofprice levels between the member countries (onaccount of greater market integration, the enhancedtransparency of prices and real economic conver-gence) likewise plays a role. The Eurosystem’s mon-etary policy can only be geared to the objective ofprice stability throughout the euro area. TheCouncil is therefore neither required nor able toremedy inflation differentials resulting from themerging of the markets or from differences in levelsof business activity. If economic disequilibria arisefrom persistent inflation differentials, it is, rather,the responsibility of national economic policy toadopt counter-measures, for instance in the fields offiscal policy or structural policy.

Monetary policy strategy

Even more than the definition of its final goal, themonetary policy strategy of the Eurosystem hasbeen criticised more or less vigorously right fromthe beginning. That is understandable inasmuch asthe decision in favour of a two-pillar strategy rep-resented a new departure. Unlike what wouldhave happened in the event of the mere adoptionof one of the previously practised alternatives ofmonetary targeting or inflation targeting, thatdecision meant that a certain period of “learningto understand” and of critical assessment of theEurosystem’s strategic approach was foreseeable.

Hence it hardly comes as a surprise that the “ECBwatchers” have mostly expressed their criticismagainst the background of the “pure” strategywhich they have always preferred. To this extent,the proposed therapy was likewise foreseeable:whereas some called for a more prominent role ofthe money stock, others expressed doubts as to therelevance of monetary growth and demanded astrengthening of the second pillar, by means of thepublication of detailed inflation forecasts.

8 See G. Baker: “Did a brave Fed kill off inflation, or was it luck?”,Financial Times, December 17, 1999, p. 4.

9 See ECB Monthly Bulletin, October 1999, pp. 39–49.10 See the articles in the ECB Monthly Bulletins of July andOctober, 1999.

The role of monetary aggregates

Among the theorists and practitioners of monetarypolicy alike, it is generally agreed that, in the long run(i.e. after the expiry of all adjustment processes),inflation is a monetary phenomenon. The correlationbetween the money stock and prices is deemed to beone of the most robust “stylised facts” in economics.11

Econometric studies which were conducted ahead ofmonetary union bore out this assessment for the euroarea. They came to the conclusion that, in the past,the necessary empirical preconditions for a promi-nent role of the broad monetary aggregate M3 in theEurosystem’s monetary policy strategy were met.12

On the other hand, it was not safe to assume that sucha far-reaching regime shift as EMU would leave thepayment and investment behaviour of householdsand businesses in the euro area unchanged.Furthermore, problems were to be expected in thechangeover from national monetary statistics to thenew, uniform standards. For that reason, theGoverning Council of the ECB decided to announcea less binding reference value, rather than a puremoney-stock target. It also decided to supplementthat benchmark by a second pillar in the shape of abroadly-based analysis of price prospects.

In the light of experience to date, those decisionsseem to have been right. For once, the movementof the money stock M3 in the euro area last yearwas largely consistent with the traditional determi-nants of money demand. There was no evidence ofa collapse of the underlying relationships. On theother hand, statistical problems, reflected in fre-quent revisions of the data, caused some discomfi-ture. Furthermore, there were major differences inmoney stock movements in the various countries.

Whether the empirical conditions for pure mone-tary targeting are met in the euro area will only beascertainable on the basis of an estimation periodencompassing a much longer period of time withthe single currency. Considering that uncertaintyabout the underlying behaviour patterns cannot beeliminated within one or two years, a fundamentalreorientation of the policy strategy in the directionof monetary targeting must definitely be regardedas premature at the moment.

The role of inflation forecasts

Besides the money stock, the ECB analyses – as partof the second pillar of its strategy – a wide range ofother indicators of the underlying trend in the infla-tion rate. Critics have accused the ECB of leaving itunclear what importance the decision-makers attachto individual indicators of price prospects. Thecharge is that it is hardly possible for outsiders toreconstruct how the assessment of the risks to pricestability in the second pillar of the strategy is effect-ed. Hence the ECB should publish its overallappraisal in the form of an explicit inflation forecast.

Hitherto, the ECB has rejected any publication of itsinflation forecast, drawing attention, among otherthings, to the substantial uncertainties in the euroarea and to the associated risk of large forecasterrors. However, President Duisenberg has alreadyhinted that the ECB may publish its forecasts in theforeseeable future. However, the publication of fore-casts poses a number of problems which should beaddressed carefully before such a decision is taken.

One potential problem is that the general publicmay attach more weight to a published forecast thanis justified by its actual significance in the monetarydecision-making process. The accuracy of inflationforecasts over the time-horizon of one to two yearsthat is relevant to monetary policy is highly uncer-tain. That applies all the more to the first few yearsof monetary union, when structural discontinuitiesare to be expected. Much thought still has to begiven to the question of how the uncertainty associ-ated with inflation forecasts can best be brought tothe attention of the outside world.

A high degree of uncertainty exists, for instance,regarding the effects of the progressive structuralchanges in the financial system. The elimination ofexchange risk within the euro area has increasedinvestors’ propensity to diversify their portfoliosacross national borders. The growing demand forhigher-yielding securities facilitates enterprises’direct access to the capital markets. The longer-term repercussions of these trends on the transmis-sion of monetary stimuli are not yet foreseeable.

But forecasts are based on estimates of equationsand models that reflect the laws and behaviour pat-terns applying in the past. Conjectures on recentchanges in such behaviour cannot be incorporatedsystematically into these models. To this extent,

CESifo Forum 14

Focus

The GoverningCouncil’s decision

to supplement areference value for

M3 with an analysisof price prospects

seems to have beenright

11 See Robert E. Lucas, jr. (1996): Nobel Lecture: MonetaryNeutrality, Journal of Political Economy, Vol. 104, No. 4, p. 336.12 See Fagan and Henry (1998): Long-run money demand in theEU: evidence from the area-wide aggregates, in: EmpiricalEconomics, 23, pp. 483–506, and the studies cited there.

CESifo Forum15

Focus

Verbal explanationsof ECB policy decisions at thepress conferencesare more informativethan publishedminutes

forecasts cannot provide a comprehensive summa-ry of all the data relevant to monetary policy. Norshould forecasts be expected to curtail the discre-tionary leeway of decision-makers.

It should, moreover be borne in mind that everyinflation forecast includes an assumption about thelevel of interest rates. Most of the central bankswhich publish their inflation forecasts assume con-stant central bank rates.13 This assumption, howev-er, poses certain problems: if the published forecastis outside the envisaged target corridor for theinflation rate, a need to take interest-rate actioncan be derived therefrom. If the central bankresponds by making such a move, that move, inturn, changes the inflation forecast.14 This consid-eration would seem to suggest publishing two ormore forecasts, subject to alternative assumptions,which, in its turn, might impair transparency.

The transparency of interest-rate decisions

With respect to the criticism of its monetary policystrategy, the Eurosystem is faced with something ofa dilemma: on the one hand, it is in the best interestsof the decision-makers to comply with the call forgreater transparency, thereby enhancing their credi-bility. On the other hand, the complexity of the cho-sen strategy cannot, in itself, be a sufficient reasonfor altering that strategy. Thus, the supporters of thetwo alternative proposals – either pure monetarytargeting or inflation targeting – have so far failedto prove that a one-dimensional strategy is not onlymore transparent but also more promising in termsof attaining the final goal of price stability. On thecontrary, given the change of regime and the associ-ated uncertainty, the two-pillar strategy continues toappear superior to the aforementioned alternatives.

That raises the question of what the Eurosystemand its decision-makers can do to comply with thecall for greater transparency in the context of thechosen strategy. A key role in this debate is playedby information policy: the Governing Council mustinform the general public at an early date, regularlyand in a comprehensible way about its assessment

of the economic situation and the background to itsinterest-rate decisions. In the Eurosystem, this isdone, firstly, through the Monthly Bulletins, inwhich the ECB gives a detailed appraisal of themonetary, financial and economic conditions inEMU, and secondly through the press conferenceswhich are held once a month immediately after themeetings of the Governing Council.

At these press conferences, President Duisenbergexplains the interest-rate-policy decisions of theGoverning Council and then answers questions putby the journalists present. His introductory remarksmay well be compared to an instant summary of theminutes of the meeting. Those comments may evenbe more informative for the markets and the mediathan carefully edited minutes which are publishedwith a delay of several weeks. The GoverningCouncil of the ECB has decided against publishingminutes of its meetings, in order to prevent the deci-sion-makers from coming under pressure at thenational level. Moreover, the experience gained inother countries suggests that the information con-tent of published minutes is limited anyway.

With regard to the transparency of interest-rate pol-icy, the question of how far the Governing Councilshould prepare the markets for forthcoming inter-est-rate measures has likewise been discussed recur-rently in the past few months. That discussion wassparked off by the decision taken in December 1998by the US FOMC to publish, in particular cases, astatement on its bias regarding future interest-ratepolicy immediately after the meeting. In principle,the announcement of such a “bias” provides anopportunity to steer the interest-rate expectationsof market players in the direction desired by thecentral bank. However, as the US experience hasshown, markets tend to interpret such bias state-ments not so much as a snapshot of current thinkingbut rather as a fairly certain prediction of the nextmove.. That gives rise to the danger of the decision-makers being compelled to take action, because amodification of their verdict, once passed, wouldcall into question their credibility and competence.That applies with particular force to a newly-estab-lished institution such as the ECB, which does notyet have a “track record” of its own. To this extent,great caution is required regarding the publicationof bias statements.

In order to stabilise market expectations, monetarypolicy must be strategically calculable: it must pro-

13 An exception is the Reserve Bank of New Zealand which pub-lishes inflation projections based on a monetary policy reactionfunction that brings inflation back into the middle of the targetrange one to two years ahead. See Mayes, D.G. and W.A. Razzak(1998):Transparency and accountability: Empirical models and pol-icy making at the Reserve Bank of New Zealand, EconomicModelling, 15, p.380, footnote 6.14 Ex post, it is thus precisely a successful policy that arouses theimpression of having been unduly restrictive or expansionary. SeeAlan Blinder (1999): Central Banking in Theory and Practice, MITPress.

vide the markets with reliable guidance. However,that does not imply that every single policy mea-sure must be predictable. Instead, the decision-makers must preserve the freedom to disagree withthe markets, and to surprise market players onoccasion, if necessary.

The role of exchange rates and asset prices

Among the strategic challenges, the ECB also hasto face up to the issue of the significance ofexchange rates and asset prices for monetary poli-cy. Last year, public attention time and againfocused on the weakness of the euro against theUS dollar. That development owed a great deal tothe growth differential between the euro area andthe US economy. An additional factor was the mis-trust of the markets with regard to the quality ofEuropean economic, structural and fiscal policy.However, the Eurosystem is not pursuing anexchange-rate target, nor is it aiming at a specificlevel of the euro exchange rate against the US dol-lar or any other currency. But since sustained fluc-tuations in the exchange rate are reflected in themovement of consumer prices, the Eurosystemcannot afford to ignore the exchange rate either.Instead, exchange-rate movements constitute amajor building block in the second pillar of themonetary policy strategy.

Although the problem of the valuation of asset-pricemovements is less pressing at the moment for theECB than it is for the US Fed15, in principle neitherthe Fed nor the ECB can afford to disregard strongmovements of asset prices because they affect thesaving, investment and consumption decisions of eco-nomic agents. At the same time, however, it is gener-ally agreed among central bankers that equity pricesand real-estate prices cannot be target variables formonetary policy because the uncertainty about thecorrect valuation of businesses or real property is toogreat. Nobody can say for sure whether a sustainedupward movement of prices in a given period is war-ranted by the fundamentals, or whether it is a case ofa speculative bubble, whose bursting might havegrave implications for the economy.

Hence it must be examined under which conditionsprices may go on rising, or corresponding expecta-tions may strengthen. In my opinion, the axiomthat inflation is always, in the long run, a monetary

phenomenon applies to asset prices, too. The risksthat asset-price inflation may involve are a weightyargument in favour of curbing monetary growth.The simple idea behind this, that even a speculativebubble must be “fed” (i.e.: financed), is borne outby past experience. In Japan, for instance, the cre-ation of a bubble in the late eighties was accompa-nied by a very strong expansion of the monetaryaggregates.

In general, a monetary policy oriented towards themedium term which manages to keep the increasein the money stock in line with the real growthpotential will hardly allow asset prices to lead “toovigorous a life of their own”. After all, keeping themoney stock tight is part of the avowed strategy ofthe Eurosystem. That is why I consider it unlikelythat asset prices will pose a major threat to stabili-ty in the euro area in the years ahead.

New means of payment (e-money)

Last among the strategic challenges is grapplingwith possible changes in non-banks’ payments. Inparticular, the monetary policy implications of afurther dissemination of electronic money must becontemplated in this connection. As catchwords, Imay mention, for instance, the problems posed bydefining the monetary aggregates, and the impactof a (possibly sizeable) substitution of electronicmoney for currency.

Specifically in the context of the timetable forintroducing euro banknotes and coins, it is nowoften argued that the possibility of the cross-bor-der use of electronic money (e-money) will give animpetus to its more widespread employment. Thecharges for exchanging national currencies withinthe euro area have made this payment innovationseem comparatively attractive.

However, the empirical observations recorded todate fail to confirm this scenario. With a money cardissuance total of Euro 63 million in November 1999,the amount of e-money issued in Germany wasequivalent to just 0.05% of the volume of Germancurrency in circulation. In the other countries of theeuro area, the situation is similar or even lessfavourable to e-money.

Moreover, the central banks addressed this topic atan early date, and noted that unambiguous rules forthe issuance of this new payment medium are desir-

CESifo Forum 16

Focus

Exchange-ratemovements are part

of the “second pillar” of monetarypolicy, but there isno exchange-rate

target

15 The ratio of equity-market capitalisation to GDP in the euro area, at63%, is much lower than in the USA, where it amounts to some 172%(at the end of 1998, see ECB Monthly Bulletin, August 1999, p. 36).

CESifo Forum17

Focus

EU membership imposes obligationson the accessioncountries with respect to futuremembership in EMU

able. That relates primarily to the range of issuers,which is to be confined to credit institutions. As longas these rules are complied with, and as long as theprocess of the substitution of e-money for conven-tional payment media continues to proceed gradual-ly, no serious disruptions of the single monetary pol-icy are to be expected, at least in the foreseeablefuture. Even so, vigilance is imperative.

Logistical and institutional challenges

The exchange of currency in 2002

Despite its great significance, I shall allude to thelogistical challenge only briefly: the introduction ofeuro currency in the first half of 2002 will call foryet another distinct tour de force on the part of thenational central banks and credit institutions. Togive an idea of the magnitudes involved: by1 January 2002, a total of 13 billion euro banknotesmust be printed and 70 billion coins minted in theeleven participating countries; in Germany alone,4 billion euro banknotes and 17 billion euro coinsmust be produced.

In Germany, the period for the parallel circulationof currencies has been shortened to zero by law:from 1 January 2002 onwards, the euro will be thesole legal tender (the statutory “big bang”). But atotal exchange of all currency on a single day is notpossible. Hence the associations of the bankingindustry, of traders and of vending-machine opera-tors have reached agreement on a transitional peri-od of two months, i.e. up to 28 February 2002. TheBundesbank, however, is assuming that theexchange will be virtually complete after two tothree weeks. During that period, it is expected thatsome 2.5 billion DM banknotes and 28 billion DMcoins (with a face value of about DM 9.5 billionand a weight of some 98,500 tonnes) will bereturned to the Bank.

EU enlargement

Among the institutional challenges facing theEurosystem is the possible accession of those EUmember states which have hitherto – for a wide vari-ety of reasons – not participated in monetary union.The EU heads of state and government have agreedto Greece’s request to join the monetary union inthe year 2001. The situation of Denmark, Swedenand the United Kingdom is more complex, since in

those countries a decision on joining the monetaryunion depends on the result of a referendum.

Yet another focus of attention is the countrieswhich wish to join the European Union. Accessionnegotiations are currently being conducted with sixcountries; with six others the “accession process”has been initiated.16 At the European Councilmeeting in Helsinki, finally, Turkey was also recog-nised as a thirteenth candidate.

From the present perspective, it is to be expectedthat at least the six countries of the “first wave”will join the European Union in the foreseeablefuture (i.e. after the conclusion of the intergovern-mental conference on institutional reform and theratification of the results). For those countries,joining the EU also involves the obligation to givetheir central banks an independent status. This fol-lows from Article 109 of the EC Treaty. That clauseincludes, inter alia, the ban on the financing of pub-lic sector deficits by the central bank, and the banon public authorities being given access to finan-cial institutions on preferential terms.

It is to be expected that most of the new membercountries will join the EU with the firm intentionof adopting the euro at a later date. One of the con-ditions for so doing is that the currency of thecountry in question should have been pegged tothe euro successfully – i.e. without major exchange-rate fluctuations – for a certain length of time.Nowadays a wide variety of exchange-rate systemsstill exist.

Finally, the possible accession of a fairly large num-ber of new participants to the monetary union willnot fail to have an impact on the decision-makingstructures of the Eurosystem. If it is assumed thatthe number of members of the Executive Boardwill remain unchanged, the Governing Council ofthe ECB, after the accession of six new memberstates and the present “outs” to the monetaryunion, will comprise 27 members. After the acces-sion of further member states, that figure will riseaccordingly. In order to ensure effective decision-making, institutional reforms are inescapable herein the long run.

16 The six countries of the “first wave” are Cyprus, the CzechRepublic, Estonia, Hungary, Poland and Slovenia; the other sixcountries are Bulgaria, Latvia, Lithuania, Malta, Romania andSlovakia.

he introduction of the euro will have far-reaching implications not only for the evolu-

tion of central banking and monetary policy, butalso for international monetary and financial rela-tions and the world economy at large. It will modi-fy the environment in which players outside theeuro area, the large as well as the medium andsmall ones, will operate in the years to come. It willtherefore entail new responsibilities for policy-makers in Europe.

The purpose of this article is to offer the reader anoverview of how the Eurosystem views its role as amajor actor in international monetary relations. Thearticle suffers from a very important limitation in thatit is confined to only one aspect, albeit a crucial one,of “monetary relations”, i.e. exchange rate relations.Although for decades the latter were seen as the solemonetary issue generated by international interde-pendence, the reader should be aware that this is nolonger the case. Over the last quarter of a century, withthe emergence of a global monetary and financialmarket, the agenda for international monetary co-operation has expanded considerably. It now includessuch key central banking fields as the efficiency andsoundness of the payment systems and the stability ofthe banking industry and securities markets. Co-oper-ation in these areas has in fact progressed at a fasterpace than in the field of exchange rates.

The next section outlines the background againstwhich the euro came into being, showing how itscreation constitutes a rather unique answer to theproblems and contradictions of international eco-nomic interdependence. After highlighting the size

factors that contribute to the international respon-sibility of the euro area (Section 3), the articleexamines the possible implications of the euro forthe relationships among the three key currencies(Section 4). It then discusses the scope and meth-ods of international co-operation in a tripolar sys-tem (Section 5) as well as its institutional frame-work (Section 6). The role of the euro in relation tonon-key currencies (Section 7), particularly thoseseeking an external anchor for their monetary pol-icy, is then considered. Finally, the article ends witha short conclusion (Section 8).

The underlying rationale

With the stipulation of the Maastricht Treaty in 1992(the “Treaty”), Europe has adopted an entirely newapproach to the choice of the monetary order of agroup of interdependent but sovereign countries.Going beyond fixed exchange rates between nationalcurrencies, it has created a single currency and estab-lished a single central bank. The importance of thisstep is apparent if one looks back over the last 50years and considers the repeated attempts to stabiliseexchange rate relations through adjustable pegregimes. The evolution of international monetaryrelations illustrates that i) open trade and ii) freeinternational movement of capital render iii) fixedexchange rate systems hardly compatible with iv)

independent national monetary policies. Exchangerates pegs have repeatedly been chosen to stabilisetrade and financial relationships between economi-cally interdependent countries, at both the global andthe European scale. If, however, the policies of thecountries involved diverge, markets question whetherexchange rate commitments can and will be main-tained. As the doubt grows, pressure may becomeenormous and eventually lead to the abandonment ofthe peg. The four elements mentioned above formwhat is sometimes called an “inconsistent quartet”.

In the Bretton Woods System the inconsistencybecame apparent in the 1960s when the develop-ment of an international capital market made it pos-sible to circumvent official restrictions and controls.

CESifo Forum 18

Focus

The crucial role ofthe Eurosystem

in exchange-raterelations

T

THE EUROSYSTEM IN THE

INTERNATIONAL MONETARY

SYSTEM

TOMMASO PADOA-SCHIOPPA*

* Tommaso Padoa-Schioppa is a member of the Executive Boardof the European Central Bank.

CESifo Forum19

Focus

A single currency at the European level was combined witha floating-ratesystem at the globallevel

It became increasingly difficult to subordinate eco-nomic policies to the exchange rate objective. Whenthe policy stance in the United States (the anchorcountry) came into conflict with anti-inflationarypreferences of other important players (such asGermany), the dollar became increasingly overval-ued and tensions grew. With economic fundamentalsin the United States deteriorating sharply specula-tive activity in the foreign exchange market reacheda level far exceeding the defensive capacity ofnational authorities. A devaluation of the Frenchfranc and a temporary floating of the German Markwere, at the end of the 1960s, the first signs of thecoming breakdown of the system. Eventually, theBretton Woods regime collapsed as a result of thewaning of the two conditions that had ensured itsinitial success: a balanced macroeconomic situationin the anchor country and limited capital mobilitybetween national financial markets.

Intra-European relations went through a similarexperience. Europe’s exchange rate peg movedfrom a US dollar to a Deutsche Mark standard,which took various forms and lasted until theadvent of the euro. Its main manifestation, theEuropean Monetary System (EMS), was designedin full awareness of the two drawbacks of theBretton Woods regime: inflexibility and asymme-try. Flexibility was pursued, for more than half ofthe ERM “narrow band” life, through timelyrealignments (11 between 1979 and 1987). As tosymmetry, no currency was explicitly given a lead-ing role and parities were decided by “commonaccord” in relation to the ECU currency basket. Inpractice, however, the Bundesbank led the mone-tary policies of other members and the GermanMark was de facto the anchor of the system.

However, even intra-European relations came underincreasing strains as a result of capital liberalisationand the creation of a single market in banking andfinancial services. Economic and Monetary Union(EMU) was decided upon and enshrined in theTreaty, not only for reasons of “high politics”, butalso to remove the inconsistency of the quartet andto firm up the Single Market. When uncertaintiesabout the ratification of the Treaty pointed to therisk that Monetary Union might not come about, anexchange rate crisis erupted because markets redis-covered domestic imbalances and policy contradic-tions. Only the substantial widening of the fluctua-tion band to ± 15% in August 1993 restored calm inthe market. In the years that followed, economic

convergence in Europe was achieved through con-siderable fiscal consolidation and successful disinfla-tion. Financial markets rewarded these efforts withlower interest rate differentials and relatively stableforeign exchange relations.

EMU is not just an evolution or a tightening ofprevious arrangements such as the European“snake” and the EMS. It is not a binding interna-tional agreement to co-ordinate monetary policies.Rather, it is a complete change of regime, theestablishment of the same monetary order thatnormally exists within a nation: one currency, one

central bank, and one monetary policy. The coun-tries that have transferred their monetary compe-tencies to a new supranational institution, howev-er, retain their separate sovereignties in manyareas, and even in the economic field. This is theunprecedented feature of EMU.

It appears from this glance at the past that two dif-ferent solutions have been chosen, at the global andthe European level respectively, to resolve the sameproblem. In both cases the inconsistent quartet hadcome about as a result of rising capital mobility. Toovercome the inconsistency, one element of the quar-tet had to be dropped. The international communitydecided to drop the fixed exchange rate element andmoved to a regime of floating rates. Europe decidedto drop independent monetary policy and founded asingle currency to complement the Single Market.The solution adopted for one area would not havebeen possible, or would not have worked, in the otherarea. Europe, in particular, could – and, in a way, wasalmost compelled to – move to the single currencybecause its economic and financial integration hadgone extremely far and because it was politicallyready to establish common institutions. Using the jar-gon of today’s debate on exchange rate regimes, theapproaches taken by the world and by Europerespectively correspond to the two “corner solu-tions” that are sometimes seen as the only viableones in a world of capital mobility. The unresolvedproblem of exchange rate arrangements for non-keycurrencies will be discussed below.

A new player on the field

The advent of the euro has brought a new playeronto the field. “Euroland” (a name that was quicklyadopted for the euro area) is, in economic terms,around the same size as the United States and twice

as large as Japan. With almost 300 million people, itproduces 15% of world GDP (the United States pro-duces 20% and Japan 8%). Accounting for approxi-mately 20% of world exports (the United Statesaccounts for 15%, Japan 8.5%), it is the largest trad-ing partner in the world economy.

In the monetary and financial field the euro, assum-ing the mantle of the German mark and its otherpredecessors, has from the start been the secondinternational currency. Euro/dollar trading is themost active and liquid segment in the foreignexchange market at the global level, followed by thedollar/yen currency pair. As for international bonds,the US dollar and the euro now jointly account for80% of new issuance, making the international bondmarket increasingly a two-currency market. As areserve currency, the share of the euro is expectedto have contracted somewhat in 1999 from the 15%of its legacy currencies, because German markreserves held by euro area central banks are nolonger recorded as foreign reserves. However, thereare signs that official authorities outside the euroarea, in particular in Asia where the bulk of globalofficial reserves are held, are gradually diversifyingtheir portfolios towards the euro.

The new economic player created by the single cur-rency has come hand in hand with a new institu-tion, the Eurosystem. This institution is the world’ssecond most important central bank.

These “size” factors obviously generate an interna-tional responsibility for the euro area policy-mak-ers, particularly the Eurosystem. Three questionscan be asked in order to explore the issues con-fronting the Eurosystem. First, what kind ofexchange rate regime will be established betweenthe major currencies? Second, what implicationswill the euro have for international co-operation?Third, what role will the euro play in relation tonon-key currencies?

Three floating currencies

With regard to the exchange rate regime, the intro-duction of the euro has coincided (it was indeedlargely a coincidence) with proposals for tighterexchange rate arrangements among the three maincurrencies. In particular, the concept of target zones,first advocated by Williamson in 1986, seemed to gainrenewed favour in Germany and in some other quar-

ters. Establishing such a system would have requiredthe leading industrial countries to agree on desirableexchange rate levels and to act in order to keep mar-ket rates within a range of permitted fluctuation.

In the debate that followed those proposals, theEurosystem expressed serious reservations aboutthe feasibility and desirability of any scheme thatwould attempt to enforce stability between thedollar, the euro and the yen. The Eurosystemwarned that as long as major players conduct inde-pendent and domestically-oriented monetary poli-cies such schemes would be impossible to reconcilewith the increasing mobility of capital, as theoccurrence of the inconsistent quartet has provedin the past. In today’s highly integrated andextremely liquid international capital markets, theamounts of funds that can be mobilised to push acurrency out of the target zone far exceed theamounts that proved sufficient to destabilise boththe Bretton Woods regime and the EMS. None ofthe leading central banks would now be willing toforego domestic policy objectives in order toabsorb or create the large amounts of liquidityneeded to defend the exchange rate objective.

It is true that pressures may be generated by unjus-tified market sentiments, uncertainty or mispercep-tions about the conditions and prospects of an econ-omy, as has been the case with the euro in late 1999and early 2000. The belief that “markets are alwaysright” is indeed naïve and unjustified. Even whenthey are wrong, however, markets are stronger thanpolicy-makers. Moreover, and most importantly,they tend to correct their own mistakes and toreflect, over time, the so-called “fundamentals”, bethey divergences in relative cyclical positions, differ-ent patterns of monetary policies, changes in com-petitiveness, or macroeconomic imbalances.

In theory, a fixed rate system allows one country orarea to benefit from a higher degree of freedom if theother two countries or areas were willing to follow inline. In practice, however, the hierarchical structuredisplayed by both Bretton Woods and the EMS – inthe sense that in both systems one country was lead-ing the monetary conditions of the others – could notbe reproduced in the new environment. Given thecomparable economic weight of the United Statesand the euro area, a hierarchy would be politicallyunacceptable. The three regions taken together aretoo far from conforming to the usual economic crite-ria for an optimum currency area. Furthermore, the

CESifo Forum 20

Focus

As a reserve currency, the shareof the euro is likely

to have fallen belowthe 15% of its

legacy currencies in 1999

CESifo Forum21

Focus

Two opposing factors are at workregarding the futureexchange-rate variability of thethree major currencies

increasing relevance of emerging markets for globaltrade, capital flows and also crisis potential meansthat co-operation among the three main global play-ers alone would be difficult to sustain.

The truth of the matter is that it would be neitherrealistic nor desirable to attempt to establish a kindof European Monetary System, or a less ambitiousvariant thereof, at the global level. The EMS owedmuch to factors specific to the European Union(EU), such as relatively homogenous economic struc-tures, a very high level of economic and political inte-gration and a comprehensive institutional edifice.Even so, the EMS itself came under great pressurewhen full mobility of capital was established andmarkets questioned its credibility and sustainability.The degree of co-ordination and political commit-ment that would be required for a system of more orless fixed exchange rates to function on a global scale– and thus to overcome the inherent inconsistencyproblem mentioned before – is so great that it cannotbe realistically expected from the major players inthe foreseeable future. This is why floating exchangerates among the major currencies are bound to stay.

Co-operation: Scope and Method

What requirements will be laid down, for interna-tional co-operation and the Eurosystem, by athree-currency system where exchange rates aremarket-determined?

Before addressing the specifics of this question, twobroader observations should be made. Firstly, in thepresent state of the world the preservation of eco-nomic order requires nation-states to be consciousof their international responsibilities. This is a mat-ter of both enlightened self-interest and “interna-tional public spirit”. Indeed, in a world where coun-tries or regions are economically and financiallyinterdependent there are, as in any “single econo-my”, certain public goods which are “public” withrespect to the world itself. Global financial stabilityor the maintenance of open trade are prominentexamples. As long as countries are sovereign in theconduct of their policies, such international publicgoods cannot be expected to automatically resultfrom the spontaneous behaviour of market partici-pants and national governments, because such spon-taneous behaviour tends to ignore the numerousexternalities that arise at the international level. Inprinciple, international institutions and forums exist

to address these externalities, and indeed they tryto. However, since they have been given only verylimited instruments, it is the task of countries them-selves to internalise global externalities.

Secondly, compared to its predecessor centralbanks, the Eurosystem is a much stronger interna-tional player, is much less vulnerable to externalshocks and influences, produces much greaterexternal effects with its own actions. Owing to itssheer size and to the size of Euroland, theEurosystem makes a considerably larger contribu-tion to world affairs and has a correspondinglylarger role and responsibility. These observationssuggest that the Eurosystem should adopt an activeand positive attitude towards international mone-tary co-operation.

Coming to more specific considerations, the firstquestion is whether the three main currencies willcontinue to exhibit the high degree of exchange-rate variability witnessed over the past 25 years.Looking ahead, two factors linked to the introduc-tion of the euro may be at work and pushing inopposite directions.

The first factor is the fact that Euroland is far lessopen than its national components, although its ratioof exports of goods and services to domestic GDP, at17.1%, is well above that of the United States (11.0%)or Japan (11.5%). Euroland, Japan and the UnitedStates are all large and rather closed economies. Thestrong and unsurprising positive correlation betweena country’s openness and its willingness to take theexchange rate into account in policy-making suggeststhat attitudes towards exchange rate developmentswill be relatively neutral in the coming years. Thismight lead to greater exchange-rate variability.

The second factor is the fact that all the threecountries or regions clearly gear their monetarypolicies towards medium-term price stability andhave an independent central bank. Although, ofcourse, price stability is never a permanent acquisi-tion, some of the special circumstances behind thehigh inflation of the 1970s and 1980s in manyindustrialised countries are no longer there. The“culture of stability” and the conviction that mon-etary policy can best contribute to economicgrowth and employment by ensuring an environ-ment of stable prices now characterise the worldeconomy. This positive combination should lead tomore stable exchange-rate relations.

On the whole, we cannot expect, for the years tocome, a significant decline in exchange-rate vari-ability, in the form of both short-term volatility andprolonged misalignments. We shall continue to seelarge day-to-day movements in exchange rates,often driven by one-day market reactions to eventsthat are quickly forgotten the next day. We shallalso continue to see deep and long exchange-ratewaves that, over quarters and years, seriously affectcompetitive positions.

Should a wide variability of exchange rates be acause for concern? My answer is a clear yes.Although little can be done about it, we should notfail to notice the costs and damages it may inflicton economic activity, financial stability and thesmoothness of international relations. If large andprolonged, variability may negatively affect macro-economic stability and distort the allocation of realand financial resources. It may determine shifts incompetitiveness with distribution and even politi-cal implications. As a result, it may fuel trade con-flicts and protectionist pressures.

Even limiting the observation to the last twodecades, several examples of harmful consequencesof wide exchange-rate variability can be recalled.The overvaluation of the US dollar in the mid-1980sand its subsequent sharp weakening was the firstprominent example of a major exchange-rate mis-alignment since the collapse of the Bretton Woodssystem. The reversal was finally triggered by thePlaza and Louvre accords, which ended a period of“extreme neglect” of the exchange rate and markedthe return to active co-operation. Since the mid-1980s, the movements of the yen – which exhibitedhigh variability along a rising trend from 250 yenper US dollar in mid-1985 to 85 yen per US dollar inmid-1995 – have been a continuous concern to poli-cy-makers in Japan and its trading partners andhave at times also given rise to protectionist ten-sions. This exchange-rate variability also partlyinteracted with the rise and subsequent bursting ofthe Japanese asset price bubble in the late 1980s.

A third example is the global financial crisis of1997-98, which was caused, inter alia, by unsustain-able exchange rates in emerging market economies.In many of the Asian economies concerned, the cri-sis was partially due to movements in the dollar/yenexchange rate. As these countries had linked theirexchange rates to the dollar, they suffered majorlosses in competitiveness when the yen sharply

depreciated against the dollar from the mid-1990sonwards. The resulting exchange-rate variation inAsia – in some cases exceeding 50% – has con-tributed to banking crises and deep recessions inthese countries, and it has been one of the mainmechanisms of global contagion and systemic risks.

These considerations illustrate that wide swings inthe world’s major exchange rates have indeed haddamaging consequences not only for the threemajor countries or regions but for the internation-al economy as a whole. The latter consequencesarise because the United States, Japan andEuroland influence the world business cycle andhave strong trade, financial and exchange rate linksto third countries.

However if, as argued above, exchange-rate stabilityis unlikely to become an end in itself for the threemajor economic players in the global economy, notleast because it would lead to a re-occurrence of theinconsistency problem, what kind of objectives couldbe set for co-operation arising from those concerns?Unfortunately, rather limited ones. When largeexchange-rate movements result from inadequatemacroeconomic and structural policies within thethree major economic areas, international co-opera-tion may generate peer pressure for the adoption ofappropriate corrections. When volatility arises frommarket uncertainties or misperceptions of actual orfuture policies, co-operation conducted in a trans-parent manner may be supportive in correcting mar-ket perceptions. A very recent example is the posi-tion the G7 expressed – in October 1999 and again inJanuary 2000 – on the yen. Although an adjustmentof the large and opposite current account imbal-ances of the United States and Japan arguablyrequires a rise in the yen/dollar rate, too rapid a risewas seen by the G7 as detrimental to the strengthen-ing of the long awaited Japanese recovery.

What instruments are available to pursue suchobjectives? Very few, and they are rather inade-quate for the task of removing “undesirable” vari-ability from the foreign exchange market.Inadequacy pertains, in the first place, to diagnostic

instruments. We have a sufficiently precise quanti-tative measure of price stability (the value ofmoney in terms of goods and prices) and there is lit-tle controversy about its desirability. Much less canbe said, however, for the value of money in terms ofanother money: assessing the “equilibrium”exchange rate and hence undesirable exchange rate

CESifo Forum 22

Focus

Wide variability ofexchange rates is a

cause for concern

CESifo Forum23

Focus

At the institutionallevel, the eurobrings a simplifica-tion, a complication,and innovations

variations is much more difficult and controversial.Second, prevention, i.e. measures aiming at buildinggreater stability into the market mechanism itself, isalso controversial and hardly effective. Restrictionson capital movements or “sand in the wheels” in theform of “Tobin taxes” on transactions, which belongto this category of measures, are technically diffi-cult to enforce and would only work if all countriesagreed to adopt them, which is quite unlikely. Thus,only symptomatic cures are left, in the form of dec-larations and occasional interventions. This is whatkey industrial countries have resorted to in the lastquarter of a century. On some occasions these cureshave proved effective. On the whole, however, theyhave not fundamentally corrected the imperfec-tions of the market mechanism.

The institutional framework

Since the breakdown of the dollar-based adjustablepeg, exchange rate co-operation among the main eco-nomic areas has completely changed. It used to bebased on firm rules (for the dollar, convertibility togold; for the other currencies, pegging to the dollar)and on an institution, the IMF, empowered to ensuretheir implementation. Now, there are no rules, nordoes the institution play a significant role. Discussionson macroeconomic and exchange rate developmentstake place within the small group of G7 FinanceMinisters and Central Bank Governors. In the G7 noformal decision-making procedures have been estab-lished, the IMF has been relegated to the modest roleof a technical secretariat, the agreed conclusionsrarely modify the policy that would otherwise be cho-sen and implementation of the conclusions is volun-tary. No mechanism of this kind would allow policy tobe conducted effectively within a country. So, nobodyshould be surprised by its great weakness for ordinaryinternational policy-making. The mechanism hardlyfunctions other than as a tool for crisis management.Indeed, only a crisis or a near-crisis provides the extraincentive to reach agreements that go beyondexchanges of views and information.

What does the introduction of the euro bring to thissystem? Paradoxically, it brings both a simplificationand a complication; but it also brings innovationsthat may exert an influence in the years to come.

First, let us consider the simplification. By reducingto three the number of relevant players, the adventof the euro makes the process of co-operation

more efficient and perhaps facilitates the formula-tion of common understandings. It is true that noG3 has come into being as yet as a result of theeuro, nor is this likely to come in the near future.However, the debates within the G7 have rapidlyevolved from a round table of seven countries to afocused discussion on the three major economies,their situations, how they interact and the implica-tions for the rest of the world.

Second, the euro leads to complication. This needs afew words of explanation. Euroland is a currencyarea that corresponds not to a state, but to a region-al entity formed by several largely sovereign states.From an economic policy point of view, this pecu-liarity complicates the co-operative game because itintroduces an entity, a player, where different poli-cies (monetary, fiscal, structural) are conducted atdifferent levels (European, national, sub-national).In other words, while Japan and the United Statesare single-tier entities, Europe is a multi-tier one.From an institutional point of view, the complica-tion relates to the fact that all international organi-sations and forums are built on the twofold pre-sumption that their members are countries and thatpolicy responsibility rests with the countries. Tofully accommodate the Eurosystem in organisationssuch as the IMF, the BIS, the G7 or the OECDwould require rather difficult adjustments becauseEuroland is neither a country nor a single-tier poli-cy-maker. In the present international institutionalframework, both the Eurosystem and the EU stillhave a somewhat special position, because they can-not claim the status of full members. At the sametime, the countries that have adopted the euro havealso changed their positions, because they are nolonger responsible for one of the key policies thatform the object of international co-operation.Furthermore, in forums such as the G7 or the G10,the Eurosystem also speaks on behalf of numerouscountries (eight for the G7, six for the G10) that arenot members of those same forums. For proceduresinvolving consultation and the circulation of infor-mation this is clearly a complication.

Third, the euro has brought about innovation. In thesecond half of the 20th century the foundation anddevelopment of the EU has brought two major nov-elties in the field of international relations: thesupranational and the regional character of its con-struction. The euro has galvanised and furtheradvanced such innovations. Europe, once the theatreof tragic conflicts originated by unfettered nation-

states, has created and successfully applied two for-mulas that, in my view, will also be increasingly need-ed in the organisation of global relations. A degreeof supra-nationalism is a sine qua non condition foran effective provision of international public goods,just as a national power is indispensable for the pro-vision of national public goods. A degree of region-

alism in organising relations between countries is anindispensable intermediate layer in a world wherethe number of sovereign countries has grown toalmost 200. Only the future will show whether thesetwo innovations will find their way into the institu-tional profile of global forums and organisations.Working to this end should be regarded as a specialtask for the EU, Euroland, the Eurosystem, and themember countries.

Key currencies as anchors

The implications of the advent of the euro for theinternational monetary system are not confined tothe relationships with the two other key currencies.They also derive from the decisions of non-keycurrencies seeking to anchor their monetary policyto one of the major currencies.

Unlike the three major economies, many of thealmost 200 countries of the world do assign, and willprobably continue to assign, a central role to theexchange rate in the formulation of their monetarypolicies. The reasons for this range from the small-ness and openness of the economy to a need tobuild credibility rapidly to strong trade links with,and financial dependence on, a large neighbouringcountry. Over the past decades, the means most fre-quently chosen by “third” countries to integrate theexchange rate in their monetary policy strategy hasconsisted in pegging the national currency to amajor currency, often the US dollar. In a world ofcapital mobility, this may no longer be the only, orthe most effective, means. Indeed, in the aftermathof the Asian crisis, those who view pegs as beinginherently unstable argued that either free floatingor firm fixing in the form of currency boards or“dollarisation” should be preferred to any interme-diate regime. The Mexican and the Asian crisis haveactually shown that the requirements for sustainingpegged exchange rates have become increasinglydemanding. Nevertheless, in the discussion, which isstill ongoing, a consensus is emerging that no singleformula meets the needs of all countries at any onetime. Any strategy has to be consistent with coun-

try-specific characteristics, such as the size of theeconomy, its trade and financial linkages, and thedevelopment and soundness of its financial sector.Intermediate solutions, including pegs and managedfloating, will remain the preferred option for a num-ber of countries unwilling to go for a more radicalsurrender of their monetary sovereignty and yetseeking an external anchor. Regional links andpolitical objectives will also play a role in the deci-sion on the appropriate strategy in any specific case.

Even now, a significant number of countries, espe-cially in central and eastern Europe and Africa, havemonetary or exchange-rate regimes involving theeuro in an exclusive or partial role.1 These arrange-ments are mainly a legacy of past links to the formernational currencies of the euro area countries. In thefuture, the euro can be expected to gain furtherimportance as a reference or anchor currency for themore than 80 countries located in what could becalled the “European hemisphere”, i.e. theEuropean, Mediterranean and African regions. Inparticular, small open economies entertaining signif-icant trade and financial links with the EU mayincreasingly resort to the euro as a reference curren-cy. The group of accession countries is a prominentcase. In their efforts to achieve economic and finan-cial integration with the EU these countries willdevote special attention to the exchange rate.Moreover, accession to the EU will at some point befollowed by participation in the exchange rate mech-anism ERM II and, eventually, the adoption of theeuro. As to non-accession countries within theEuropean hemisphere, for virtually all of these theEU is by far the largest trading partner, the base oftheir financial system and a counterpart in importantbilateral agreements in the fields of trade, technicalassistance and support for economic development.To the extent that these countries seek an externalmonetary anchor, the euro is the natural choice.

For the United States, and even more for Japan, asimilar process of regional clustering around themajor economy of the area seems less likely. In theAmerican hemisphere, comprising 35 countries,regional co-operation is at an early stage of devel-opment, is mainly confined to trade arrangements,and has a weak institutional structure not involvingany binding legislation or supranational powers.The NAFTA agreement is mainly a free trade

CESifo Forum 24

Focus

The euro will also serve as an anchor

for the monetarypolicy of other

Countries

1 For a description of these arrangements see the article entitled“The international role of the euro”, in the August 1999 issue of theECB Monthly Bulletin, pages 31–53.

CESifo Forum25

Focus

The advent of theeuro and the eurosystem impliesa new reality towhich all playersneed to adapt

arrangement without a monetary or exchange-ratedimension. The same holds for the Mercosur coun-tries, whose trade links (sometimes closer with theEU than with the US) make it difficult to pegexclusively to the dollar. Of course, the US dollarnevertheless plays and will continue to play a dom-inant role in the American hemisphere, as the cur-rent issue of dollarisation in Argentina andEcuador illustrates.

With regard to the East Asian and Pacific region,the prospects for the Japanese yen to play the roleof reference currency seem remote. While theEuropean economies moved away from the dollarstandard as soon as the Bretton Woods system col-lapsed, Asian economies generally remained onthe dollar standard for another 25 years, until thecrisis of the late 1990s (partly resulting from theexchange rate regimes themselves) severed thatlink. Most emerging market countries in Asia haverather diversified trade connections with allregions of the world. Their external trade withJapan is about the same as that with Europe andsomewhat less important than that with the UnitedStates. This configuration would hinder a pegexclusively to the yen, even though the crisis hasclearly shown these countries that the yen cannotbe excluded from their currency arrangements. Asa result, many small open Asian-Pacific economiesmay continue to face difficulties in choosing a sin-gle external anchor. The financial crises of theseemerging market economies in 1997-98 illustratedtheir vulnerability to exchange rate variabilityamong key currencies.

Of the three key currencies, the euro may there-fore be the one which develops the most importantinternational role as an anchor. Active promotionof the role of the euro as an anchor by the EU andthe Eurosystem should be ruled out, as it would beinconsistent with the key policy mandate definedby the Treaty. This role should only come about asa result of unilateral decisions by third countries.The Eurosystem will have to follow it closely andalso to define its position. A fear of potentialimplicit constraints deriving from an expansion ofthe international role of the euro would be neitherjustified nor appropriate. Euroland is largeenough, and the Eurosystem is sufficiently strongand independent, to have little to fear. After all,this role was not refused by Germany vis-à-vis theERM countries, whose GDP in 1990 was triple thatof Germany, while the total GDP of the 85 non-EU

countries of the European hemisphere is less thana third of that of Euroland!

Conclusion

This article specifically centres on a review of theimplications of the euro and the Eurosystem forwhat has been, over around half of the last50 years, the key aspect of the international mone-tary system, i.e. exchange-rate relations. On theone hand, it shows that for the relationshipsbetween the three key currencies these implica-tions are likely to be limited. On the other hand, itgives an idea of how far-reaching the implicationscould be in other fields, such as the institutionalframework of international organisations andforums or the monetary strategies of the manythird countries which seek an external anchor fortheir policies. This is new ground where predictionsand the formulation of a policy are very difficult.

The advent of the euro and the Eurosystem opensa new chapter in the history of national and inter-national monetary regimes. Both Euroland and itsinternational partners, be they the majoreconomies of the world or smaller players, willneed time to adapt to the new reality, to under-stand in full its implications, and to design the bestpolicies to address them.

As Otmar Issing explains in another article of thisjournal, the Eurosystem quickly understood that amonetary strategy mechanically repeating the highlysuccessful approach of the Deutsche Bundesbankwould not have been appropriate for the newly cre-ated euro area. This is all the more true for thedesign of an international policy for the Eurosystem.The economic and financial conditions as well as thehistorical and political constraints of the pre-euroworld were so different from what is already emerg-ing and will become ever more visible, that usingpast schemes as the paradigm for the future wouldbe misleading. A combination of firmness on the keymission to preserve price stability and of opennesson the strengthening and improvement of interna-tional monetary governance is the way forward.

he EU has started accession negotiations with 10countries in Central and Eastern Europe (CEE).

Moreover, a number of other countries, mostly fromthe area of the Former Republic of Yugoslavia, but alsoincluding Turkey, have also been recognised as having a“European vocation”. It is thus likely that practicallythe whole of Central and Eastern Europe will soonhave a concrete perspective of EU membership andthus, eventually, also of participation in the eurozone.

To become member of the EU does, of course, notimply immediate EMU memberhip. On the contrary,the official doctrine on the path to EMU is: “First jointhe EU, then converge towards the Maastricht crite-ria, then join the ECB”. This is the way the initialgroup of countries had to go. This doctrine might stillbe useful if one thinks about full participation in theeurozone, which implies that the country concernedwill have a seat on the Board of the ECB and can thusinfluence the policy of the eurosystem. However, inan environment characterised by the constant threatof currency crisis and unstable capital flows, a differ-ent approach might be needed for Central andEastern Europe. Moreover, the euro might alsobecome the anchor for countries that are not candi-dates for EU membership. The purpose of this contri-bution is thus to discuss the stabilising role the eurocould play in the larger Europe from the Atlantic tothe Urals. The main thesis is that a single currencyeuro zone from the Atlantic to the Urals should notbe looked as a distant perspective, but a concrete pos-sibility for this first decade of the new millennium.

Who would benefit from the euro?

In thinking about the optimal exchange rateregime for Central and Eastern European three

groups of countries should be distinguished,according to their relative strength:

1. Fiscally and institutionally very strong countrieswith Maastricht-conforming policies;

2. countries that are not yet at the Maastrichtlevel, but have the necessary institutions and areheading in that direction;

3. countries in states of acute financial crisis, withvery weak institutions.

1) The very strong countries, i.e. countries that couldbecome members of the EU at any time and that ful-fil the Maastricht criteria most of the time (e.g.Switzerland and Estonia) gain from pegging to theeuro because the EU is anyway their major tradingpartner. Moreover, pegging to the euro gives finan-cial markets an anchor for longer-term expectations,thus reducing the impact of financial shocks. Forthese countries the classic criteria of the OptimumCurrency Areas approach are close to being fulfilledas their economic structures are close to that of theEU. However, even if they have a strictly economicinterest in joining the euro area, these countries canafford the luxury to wait and see. Given theirstrengths they can comfortably survive outside. Evenfor these countries, however, the step from a tighteuro-peg to full euroisation is likely to bring benefitsas they would thus experience considerable savingsin transactions costs. This applies with particularforce to the case of Estonia, which has already fore-gone the use of the exchange rate with its currencyboard, which linked the Kroon originally to the DM.As this link is not to be changed anyway, the intro-duction of the euro in cash form can only yield addi-tional benefits that will be important for an economyin which exports amount to close to 100% of GDP.

2) The middling countries, with moderate inflationrates (now usually below double digit) and fiscaldeficits, for example the countries of centralEurope. These countries are also in an intenseprocess of structural change whose outcome is dif-ficult to foresee. They might therefore need someflexibility in their real exchange rate for some time.

But the cost of retaining some flexibility in theexchange rate is that this leaves open the threat of

CESifo Forum 26

Focus

There is an alternative

approach to the official path to

EMU membership

T

ONE EURO FROM THE

ATLANTIC TO THE URALS?

DANIEL GROS*

* Daniel Gros is Director of the Centre for Economic PolicyStudies (CEPS) in Brussels.

CESifo Forum27

Focus

Euroisation hasmany advantages for the weakestgroup of accessioncountries

speculative attacks, especially given their ratherlarge current account deficits, as the example of theCzech Republic has recently shown. However, theCzech experience also suggests that the cost of suchan attack is serious, but limited, and does not resultin outright catastrophe as in the case of Russia. Thecosts and benefits of different exchange rateregimes are thus often finely balanced and must beconsidered case by case. It is sometimes argued thatEU membership, and later ERM membershipshould be a sufficient protection against speculativeattacks for the new member countries from Centraland Eastern Europe – and that they thereforewould not need to adopt extreme measures likeeuroisation. The experience with the currency crisesof 1992–1995 in the EU shows, however, that evencountries with stable institutions, that were mem-bers of a tight ERM, can at times be buffetedseverely by shocks coming from financial markets.

For the “middling” countries one can thereforeconclude that they would benefit from retaining anational currency and some freedom to adjust theexchange rate if foreign exchange markets do notbecome a source of instability themselves. In anideal world the exchange rate would be a policyinstrument to adjust to country-specific shocks.But experience has shown that this is no longer thecase when capital markets are opened. For thisgroup of countries the key consideration mightthus be the timetable for the liberalisation of capi-tal movements.

3) The very weak cases, namely countries that arevery far from fulfilling any of the requirements forEU membership in general (and the Maastricht cri-teria in particular). These countries usually havelarge fiscal deficits and high inflation, their curren-cies are often under pressure and real interest ratesare very variable, often patently unsustainably highwhen the government tries to stabilise the economy.These countries would gain from being able to enterthe euro-area, because that would be a way toimport sensible macroeconomic policies and deci-sively gain the confidence of financial markets. Sincethe alternatives are hyperinflation and/or enormousrisk premia on foreign debt, the benefits of this con-fidence effect and of a stable currency can far out-weigh any potential costs of not being able to reactto asymmetric shocks with exchange rate changes.

In countries with a weak fiscal administrationseigniorage can be an important source of revenue

for the government, but high inflation rates alsohave economic costs (and erode the real value of taxrevenues). Seigniorage considerations are thus not asufficient argument for a high inflation regime. Buta credible low inflation regime cannot be createdovernight by conventional means. This is why onehas to resort to unconventional measures.

Another key advantage of full euroisation would beits systemic impact, in transforming the politicaleconomy inside the country and thus the chances ofhealthy economic growth. In some parts of EasternEurope the banking system is still a conduit forlarge scale corruption and political intervention inthe economy. A political class that cannot run largedeficits and that cannot control the banking systemwill be forced to leave more room for really pro-ductive private enterprise. Supporting loss-makingstate enterprises, or just favouring politically wellconnected “business men”, will become more diffi-cult and the cost will become more apparentbecause it would have to go through the budget.Entrepreneurs will learn to concentrate on manag-ing their enterprises more efficiently because thatwill become the main avenue for success. Politicalconnections will count for less. Petty corruption andfavouring some enterprises through tax breaks etc.will of course remain, but the sums that can be allo-cated this way pale in comparison with the wealththat can be controlled through the banking systemand large scale inflationary finance.

The main argument usually advanced againsteuroisation (or currency boards) is that this makesit more difficult to adjust the real exchange rate.This argument is based on the observation that inwell established economies nominal wages andprices are usually rigid in the sense that they arevery difficult to lower. However, this argumentdoes not apply to many of the countries in CEE. Inmany of them wages are not set in national agree-ments and can thus adjust much more easily tomarket conditions. Moreover, in the less stablecountries in Central and Eastern Europe wages areoften anyway set either explicitly or implicitly inDM. In these cases devaluation cannot achieve animprovement in competitiveness. These countriestherefore risk ending up with the worst combina-tion of unstable financial systems and rigid labourcosts. Any attempt to have national monetary poli-cies would face, at any rate, considerable problemsas the DM is already playing an important role inthe banking system of many countries in CEE.

There is only one concrete example of a countrythat has given up its currency for an extended peri-od. This is the case of Panama. This country is usu-ally considered to constitute a special case thatcould not serve as a model for countries in Centraland Eastern Europe. However, the annex showsthat this view is mistaken. Panama is similar in sizeto a number of countries in this region, and its eco-nomic structure is also not that different. Panamahas used the US dollar for almost 70 years and hasfared rather better in economic terms than itsimmediate neighbours.

How to anchor to the euro?

One way for non-EU countries to enter the euroarea is to opt for a currency board, as is alreadybeing done by Bosnia, Bulgaria, Estonia andLithuania. The first three chose the DM as theanchor and are now, in macroeconomic terms, defacto members of the euro area. A currency boardcan deliver the benefits of credibility with financialmarkets and low inflation as these examples haveshown. However, as the experiences of Argentinaand Hong Kong also show, even currency boardsthat are run very conservatively can come underattack. While the mechanism of the currency boarditself is usually technically unassailable, theseattacks are costly because they lead to increases indomestic interest rates which have a negative effecton demand. Defending a currency board is thustechnically easy, but can have a high price in termsof unemployment. Financial markets know this andthis is why a currency board is never 100% credible.This weakness of currency boards has recentlyprompted the Argentine government to considerplans to switch totally to the US dollar.

The economic benefits of full dollarisation or euroi-sation are similar to those resulting from a currencyboard, but they are more certain. They can be reapedeven by countries whose institutional and politicalweakness would leave doubts in the market that itcould follow the rules of the game of a currencyboard (e.g. Russia and the Ukraine). Under a cur-rency board regime the country still has a domesticmonetary authority, which might cede to govern-ment pressure and violate the rules of the currencyboard, e.g. by giving credit to the government.

Therefore the radical solution of unilateral, totaladoption of the euro as the domestic currency

offers even more benefits, compared to the curren-cy board, for countries with very weak institutions.The case of Argentina suggests that the idea is oneof practical, not just theoretical interest. But it isnot to be thought of as something more foradvanced emerging markets such as Argentina. Onthe contrary, in the case of Central and EasternEurope euroisation meets the objection that thenational authorities would still be quite free toabandon their commitment to the monetary rule ofthe currency board, which in present circumstanceswould mean costly interest rate risk premia.Abandonment of the euro would be much morecostly in political terms. Total euroisation shouldthus be a more credible regime.

Implementation and cost

Euroisation is also technically straightforward. Thekey is that the government of the country in ques-tion would just declare that at a certain date thenational currency is substituted by the euro and theold national cash will be exchanged into euro notesand coins at the current market rate.

The national monetary authorities (which couldanyway be abolished) would, of course, have no seaton the Governing Council of the ECB, which wouldthus not be affected, but it would still be preferableto have an explicit agreement on the details ofeuroisation. For example, there should be an under-taking to radically liberalise the domestic bankingsystem (in particular allowing EU banks to acquirelocal ones) and a proper banking supervision. EUauthorities, including the national central banks inthe Eurosystem should be able to provide therequired technical assistance.

One consequence of euroisation is that the countryconcerned loses its seigniorage, which wouldinstead accrue to the ECB. This is one of the rea-sons why countries that are on a currency boardregime hesitate to switch to the full adoption of theeuro. However, in the case of countries that arecandidates for membership this obstacle can andshould be overcome. The most direct way to avoida transfer of seigniorage from the rich EU to itspoor future member countries in Central andEastern Europe would be to provide the initialendowment in euro via a zero interest rate loanfrom the EU to the countries concerned. In thisway the countries which adopt the euro do not lose

CESifo Forum 28

Focus

One way to anchor acurrency is through

a currency board,but the radical solution is the

adoption of the euroas the domestic

currency

CESifo Forum29

Focus

Lessons from Panama

their seigniorage (their central banks can investtheir present reserves or they could be used toretire public debt). The loan would have to berepaid upon accession as full member, or if thecountry de-links from the euro. The net cost for theEU would be zero. The debt service on the EUbudget would be offset by the higher monetaryincome of the eurosystem.

The sums involved would anyway be modest, com-pared to the EU budget, and the size of the overalleuro money supply.The combined mass of cash in cir-culation of the 10 applicants is about 20 billion euro.With a long term interest rate of about 5% this wouldlead to a debt service for the EU budget of about1 billion euro, about 1% of the overall EU budget.Member countries would anyway receive a similar1

amount in the form of their shares in the higher mon-etary income distributed by the Eurosystem.

What is in it for “us”?

Is there any danger for the stability of the euro?This is difficult to conceive. First of all, theEurosystem would continue to gear its policy onlytowards price stability in the euro area. But couldsevere economic difficulties in countries that haveeuroised not create difficulties for the euro area aswell, so that the ECB would be under strong indi-rect pressure to relax? The combined GDP of the10 applicant countries is about the same as that ofSweden, or around 5% of the GDP of the eurozone. Even in the unlikely event that all of themadopt the euro – and experience difficultiesbecause of a tight policy pursued by the ECB – itis thus difficult to see how their situation couldinfluence significantly the stance of theEurosystem.

It is actually more likely that countries that haveeuroised would prefer a tighter policy by the ECBbecause they are likely to grow faster than mostcountries in the euro zone. As their price levels con-verge to the EU average their measured inflation

rates are also likely to be higher so that real inter-est rates are likely to be much lower than in theeuro area. Countries that have euroised are thuslikely to find themselves in a similar position assome of the current euro zone countries that aregrowing fast and would therefore prefer higherinterest rates. To the extent that countries that haveeuroised exert any pressure on the ECB at all, itwould probably go in the direction of a tighter pol-icy. All this suggests that euroisation by a large partof Central and Eastern Europe cannot endangerthe stability of the euro.

The real gain for Europe would be that euroisationwould contribute to stabilising the part of Europethat has so far not participated in the political andeconomic stability European integration hasbrought to its Western half.

Annex

Euroisation: Any Lessons from the PanamanianExperience?2

The main example of a country that has adopted aforeign currency is Panama. A brief review of theexperience of that country is therefore instructive.Panama uses the US dollar, has no central bankand no official foreign exchange reserves (they arenot needed). This arrangement has by now workedwithout technical problems for almost a century.The “Balboa” is mainly used for accounting pur-poses and exists only in the form of silver coins.

In the following we will address three key issuesthat arise when discussing euroisation forSoutheast Europe. 1) Is Panama just an uninterest-ing special case? 2) Are perfect labour markets apre-requisite for the adoption of a foreign curren-cy? 3) Can euroisation protect against financialshocks? The answers, in short, are: no, no and yes.

1) Is Panama a useful example for SoutheastEurope (SEE)?

Panama is often perceived as a special case whichhas no relevance for other situations. Even a super-ficial examination reveals, however, that in threekey areas there are close parallels to the countriesin SEE.

1 Not exactly zero if the interest the EU institutions pay on theirborrowing is higher than the return the ECB obtains from itsinvestments. Another complication is that all 15 governments con-tribute to the EU budget, but only the 11 euro area governmentswould receive a share of the revenues of the ECB. However, as thesums involved are minor this issue can be neglected.While it is straightforward to account for the initial endowment ofcash it is more difficult to deal with subsequent developments froman economic point of view. If a Euroised economy grows itsdemand for cash should also grow. Unless the EU makes furtherloans of its currency the country in question would have to run abalance of payments (either current account or capital account)surplus. However, the magnitudes would again be so small that thisis not an important issue.

2 This section is based on information from the IMF and Moreno-Villalaz (1997).

i) Size. Panama has a population of about 2.7 mil-lion, larger than a number of countries in CEE. Itstotal GDP, around $ 9.000 million is in the sameorder of magnitude as that of Bulgaria.

ii) Weakness of institutions. Panama also resemblesthe countries of SEE in that is has rather weakadministrative structures and democratic institu-tions.

iii) Importance of transfers from abroad. Panama isoften perceived as an artificial country that livesoff the canal, but has no significant economic activ-ities of its own. But this impression is wrong. Thecountry does have a sizeable industry. Exports ofgoods are at the same level, per capita, as in itslarger neighbour, Colombia, and agriculture occu-pies a similar share of the population.

Moreover, the transit fees from the Canal makePanama actually more comparable to the accessioncountries in CEE for which the “euroisation” optionshould be considered because these countries willalso receive substantial transfers from the EU.These transfers will at first increase (upon acces-sion) and then decline, which will require substan-tial changes in relative prices. However, these prob-lems are not insurmountable. Panama had to dealwith large swings in the revenues from the canal, butthere was never any question that the countrywould need a flexible exchange rate to deal withthem. The importance of the revenues from thecanal should also not be exaggerated. In 1998 totalrevenues from the canal amounted to about $ 650million, equivalent to about 7% of GDP. The canalis not a free lunch, however; operating costs were ofa similar magnitude (with about one half forlabour). The (net) rent the country receives fromthe canal is thus probably only around 3% of GDP,which is about what the countries in CEE canexpect from the EU structural funds alone. Thecanal is relatively more important for publicfinances. The Canal authorities and user chargescontribute about $ 150 million to an overall budgetof about $ 1.1 billion. But again, a similar effectwould operate through the structural funds.

2) Are perfect labour markets a prerequisite forthe adoption of a foreign currency?

It is often argued that a national currency is need-ed as a safety valve in case domestic price andwage pressures mount, and that such pressures

arise more often in poor countries. This concern isnot borne out in the case of Panama. There hasbeen no long-term price pressure on thedollar/Balboa link. On the contrary, over the lastthirty years prices have actually increased less inPanama than in the US, on average by 1.7% eachyear. Over the entire period (between 1967 and1997) the US CPI increased by about 370%,whereas in Panama the increase was only 170%.There is also absolutely no indication thatPanamanian labour priced itself out of the market.Unemployment in Panama hovers presentlyaround 13%, but this compares well with otherLatin American countries and is close to the expe-rience of many countries in CEE. The data fromoverall employment is even more encouraging.Despite its young population, the overall employ-ment rate (employment/population) is 33%, whichis much higher than its Latin American neighbours(Colombia 15%, Guatemala 8%) and again closeto the values of the countries in CEE with theirmuch older population.

This is not to say that the labour market in Panamais perfect. On the contrary, it was actually themodel for the Harris Todaro model of a duallabour market in which there are two sectors: aninternational and modern one and a traditionalrural one. The first sector pays above-market clear-ing wages to reduce the incentive to shirking.(High transactions costs make it impossible toenforce contractual behaviour by labour.) The sec-ond, traditional sector absorbs surplus labour at awage rate that is determined informally throughhousehold and other non-market activities (subsis-tence agriculture, small-scale commerce, etc.). Inequilibrium there is substantial unemployment incities until the cost of moving there just equals theexpected wage differential (which in turn dependson the probability of finding a job, i.e. the unem-ployment rate). This is exactly what is likely to hap-pen in parts of CEE. A different monetary regimedoes not change the fundamental reasons for thisdual economic structure, which characterises manyCentral American countries. But the experience ofPanama shows that a stable currency regime doesnot increase the problems that result from such adual labour market. On the contrary, each timePanama had to face a major political or economiccrisis (e.g. US embargo, oil price increase) theinformal sector was able to absorb the surpluslabour released by the formal sector without exces-sive unemployment problems in the cities.

CESifo Forum 30

Focus

Many parallels with CEE countries

CESifo Forum31

Focus

Euroisation limitsthe swings in thereal exchange rate

Most countries in CEE have imperfect labour mar-kets whether or not they adopt the euro quickly.Their problems are likely to be similar to the onesfaced by other countries, like Panama, with a similarincome per capita (and weak institutions), but differ-ent from the ones facing EU members with theirhighly developed social systems. The experience ofPanama suggests that not having a national monetarypolicy does not worsen the unavoidable problemsthat arise from dualistic labour market structures.

3) Can euroisation protect against financialshocks?

The clear answer here is yes. Adopting the dollarhas protected Panama against most financialproblems and allowed it to survive the recentglobal financial crisis much better than othercountries in the region. The latest IMF reportnotes that its economy was affected by the fall indemand in the rest of the region, but there wereno signs of financial instability in Panama itself.Deposit and lending rates remained essentially atthe average of the 5 preceding years (around 7and 10% respectively) whereas dollar equivalentrates in other Latin American countries wereoften 20 percentage points above Libor.Dollarisation thus protects the domestic economyagainst external financial shocks.

A key issue for many countries with large externaldebts is to what extent euroisation could lower therisk premium paid on foreign debt. For a countrywith an external debt to GDP ratio of 100% (e.g.Bulgaria) a risk premium of 10% (not unreason-able under current circumstances) implies anadditional annual transfer of 10% of GDP to for-eign creditors. The experience of Panama showsthat euroisation could be a big help in this area aswell. The public external debt of Panama is alsosubstantial, now around 60% of GDP (after apeak of 75% in 1995), but the government neverhad to pay a large risk premium on its indebted-ness. Why should dollarisation lead to radicallylower risk premia also on external debt? Per se,the monetary regime does not cure the chronicdifficulties of the public sector in countries withweak institutions to raise tax revenues (and limitpressures for more expenditure). But euroisationhas two important consequences.

First of all, it eliminates the difference betweenexternal and internal debt. This in turn has two

consequences: the government cannot discriminateagainst foreign creditors and it cannot rely on acaptive domestic market to finance deficits.

Secondly, it eliminates a key source of uncertaintyabout the capacity of the government to serviceexternal debt because with euroisation the largeswings in the real exchange rate that result fromthe large sudden depreciations which often ariseduring currency crisis are no longer possible.Russia is a case in point. When the exchange rateof the rubel quadrupled last year (going from 6 to24 rubles per dollar) the capacity of the Russiangovernment to service its foreign debt was cut toone fourth (and the price of Russian eurobondswent to 25% of their face value). After a year,domestic prices have doubled so that the realdevaluation is “only” 50%, but this still impliesthat the dollar value of Russian tax revenues hasbeen halved (and the price of Russian eurobondswent back to 50%). With euroisation swings of thereal exchange – and the price of foreign debt – ofthis order of magnitude will not happen.Euroisation thus improves debt service capacity ina number of ways. A much lower country risk pre-mium is therfore entirely appropriate.

A PERSPECTIVE ON THE

EURO

WHY IS THE EXTERNAL VALUE OF THEEURO CURRENTLY LOW?

GIANCARLO CORSETTI*

t the start of European Monetary Union in1999, the vast majority of analysts expected

the new currency to gain against the dollar. Therewere several reasons backing these expectations,but two elements were common to most views.First, very few analysts doubted that the EuropeanCentral Bank would establish itself as an indepen-dent central bank determined to pursue its institu-tional mandate. Despite criticisms ranging from alack of transparency to some inadequacy to copewith financial crises, it was hard to forecast for thenear future a change in Europe-wide monetarypolicy, which would compromise low inflation andmonetary discipline. Second, European growthrates were expected to catch up, and perhaps sur-pass, growth rates in the United States.

The first element did not prove wrong. There ishardly any evidence that the external weakness ofthe euro reflects current or prospective inflationdifferentials, let alone any lack of confidence onthe ability of European Monetary Union and itsmonetary institutions to guarantee price stability.As is well known, the same cannot be said asregards the second element – expectations of nar-rowing growth differentials.

The first graph shown in this page plots the evolu-tion over time of the dollar-euro exchange ratetogether with the difference between the consensusforecast for the euro area and the U.S. The top figurerefers to the consensus forecasts for the year 1999,the bottom figure to the forecasts for the year 2000.In both figures, a drop in the growth differentialsline at a particular date means that, at that date, ana-

lysts expected the gap between the US and the Euro

area growth rates to widen. The graph, originallypresented at the Brookings Institution in September19991, has been reproduced, updated, and comment-ed upon in several reports – by OECD, CEPS andother institutions.2

The graph shows that growth differentials betweenthe Euro-area and the U.S. increased during the firsthalf of 1999, flattening in the second part of the sameyear, but kept increasing afterwards, through thebeginning of May 2000. To many, this last piece ofinformation may come as a surprise, since over thesame period European growth rates have indeedpicked up. Most importantly, the graph shows that,over time, the growth differentials and the exchangerate have tracked each other quite closely. Any revi-sion in the rate of growth of the U.S. has corre-sponded with a drop of the external value of theeuro – this pattern is quite evident in the fall of theEuropean currency until quite recently.

To avoid misunderstanding, let me clarify that, per

se, the empirical evidence shown in the graph is nota proof that (surprises in) relative growth ratesbetween U.S. and Europe cause a weak euro. It istrue that many models that are taught in class-rooms and are well known to policy makers sug-gest a positive relationship between relativegrowth and the exchange rate. But, as many inter-national economists (including myself) stress, it isgenerally dangerous to rely on a single-variableinterpretation of exchange rate movements.Indeed, the evolution of growth differentials doesnot seem to fit bilateral exchange rates acrossother currencies – a notable case being the yen-dollar exchange rate. To be honest, it is hard to pro-vide a convincing interpretation of the recent evo-lution of the euro.

So, let me proceed by taking the evidence in thegraph as a working hypothesis, a perspective ratherthan a conclusion. My main goal is to draw atten-

CESifo Forum 32

Focus

Expectations of arising euro/dollar

exchange rate were based on a

disciplined monetarypolicy and acceler-

ating growth inEurope

A

* University of Bologna, Yale and CEPR.

1 G. Corsetti and P. Pesenti (1999). “Stability, Asymmetry andDiscontinuity: The Launch of European Monetary Union”,Brookings Papers on Economic Activity, 2, December, pp. 295–372.2 For a comprehensive analysis of euro-related issues, see the EuroHomepage at www.econ.yale.edu/~corsetti/euro/Euroit.htm.

CESifo Forum33

Focus

In the U.S. the high growth rates of supply have beenoutpaced by thegrowth in demand

tion to a potentially relevant piece of information.

From such perspective, I will address three ques-

tions. First, how can differences in growth rates

between the Euro-area and the U.S. affect the

euro-dollar bilateral exchange rate? Second, what

are the alternative interpretations of the current

euro weakness and how are they related to one

centred around growth differentials? Third, is

there any lesson for the future development of the

euro-dollar exchange rate?

Growth and demand differentials

The high growth rates in the U.S. reflect important

changes on the supply side of the economy –

including both technological innovations leading

to higher productivity, and a higher capital stock

resulting from the sustained investment rates in

the past few years. For thesake of argument, suppose atfirst that domestic demandwere fixed or increasing moreslowly than supply. For a givendemand, a higher supply ofdomestic goods must be soldin foreign markets – thus low-ering the relative price ofdomestic goods. In otherwords, given demand, growthin supply would cause the realexchange rate to depreciate,not to appreciate. This meansthat either inflation at home islower than abroad, or thenominal exchange rate depre-ciates, or both. Also, the cur-rent account should go intosurplus. This could be the rightscenario in the presence of atemporary boom in productiv-ity – that is, a boom that is notexpected to last in the future.

Clearly this is not what is goingon in the U.S. The high produc-tivity gains and growth rates ofoutput in this country are out-paced by the growth in demand.For instance, a few weeks agowe learnt that, in the first quar-ter of the year 2000, U.S. con-sumption demand grew at a

staggering annual rate of 8.3%, reflecting exuberantconsumer confidence. To many analysts, this figuremust have looked like a typographical error (indeed ithas been subsequently revised downwards to 7.7%).

The implications of strong demand growth for thereal exchange rate are, of course, quite differentfrom what we have discussed above. As internaldemand grows faster than supply, the realexchange rate needs to appreciate – either becauseof higher inflation or because of a nominal appre-ciation, or both. At the same time, the currentaccount of the booming country goes into deficit.

What does this tell us about the past and currentvalue of the euro? To address this question, it isappropriate to modify our graph, explicitly account-ing for relative demand dynamics. The second set offigures plots the evolution of the euro-dollar

Graphs 1 + 2

a) Every month, Consensus Forecasts surveys over 200 financial and economic forecasters in more

than 20 countries for their estimates of a range of variables.

exchange rate together with the differential consen-sus forecasts for consumption and investment in theyear 2000. As these variables are strongly correlatedwith GDP growth, it is not surprising to see that thegeneral message of this set of graphs is the same asthe previous one. More interesting is to focus on thedates showing a sharp revision of these differences– sharp revisions that are not visible in the previousgraph. The Euro-area – U.S. differences for bothconsumption and investment widen visibly in mid-October 1999 and in mid- April 2000. The eurodrops more rapidly when the differentials betweenthe U.S. and the Euro-area widen.

Misperception, portfolio adjustment and otherfactors

Some observers claim that the euro is currentlysuffering because markets misperceive fundamen-

tals, reflecting excessive exu-berance in the U.S. and/orexcessive concern with struc-tural problems in Europe. Ifthis is true, we should notehere that our graph suggestsa strong correlation betweenmarkets’ pessimism and rela-tive growth rates. Could mis-perception currently coin-cide with excessive reactionto news about the dynamismof the two economies? This isa topic surely requiring fur-ther investigation.

Another view focuses on theslow pace at which Europeangovernments are implement-ing fiscal reforms andreforms of social securityand the labour market.Supposedly, these reformsare expected to have a posi-tive impact on the externalvalue of the euro mainlythrough its effect on currentand prospective growthrates. Note that, since thestart of the euro, it is hard todetect important news, eitherpositive or negative, aboutthe pace of reforms in

Europe. One could argue, perhaps, that the “news”is the absence of news, leading to deterioration inmarket sentiment.

According to a third interpretation, the weaknessof the euro is a reflection of the strength of the U.S.

stock market – supposedly driven by anticipationsof future growth. A fall in this market should leadinvestors to increase their exposure in Europe,bringing the value of the euro back up. Recent evi-dence seems at odds with such an interpretation:the U.S. stock market occasionally fell, evensharply, without bringing about any reduction inthe value of the dollar vis-à-vis the euro. Of course,to the extent that the fall in the U.S. market wasdriven by expectations of monetary tightening bythe American central bank, anticipation of aninterest rate hike could explain both a weak stockmarket and a strong dollar – interest differentialsproviding an incentive to move from dollar-

CESifo Forum 34

Focus

The euro drops morerapidly when the

differentials betweenthe U.S. and theeuro area widen

Graphs 3 + 4

a) Every month, Consensus Forecasts surveys over 200 financial and economic forecasters in more

than 20 countries for their estimates of a range of variables.

CESifo Forum35

Focus

The future value of the euro will depend on the future growthperformance of Europe ...

denominated equity to dollar-denominated bonds.The recent evidence nonetheless provides a warn-ing against assuming a one-way adjustment (U.S.–Europe) of global portfolios.

A fourth thesis attributes the weakness of the euroto markets’ uncertainty about the evolution of thepolitical and institutional setting in the area, andthe ability of European institutions to cope withpossible political and economic tensions. Thisuncertainty, however, has not prevented an impres-sive growth of markets in euro-denominated assets(admittedly helped by the low interest rates in theEuro area as well as by expectations of a weak cur-rency). Would this growth be possible if marketslacked confidence in the future of the euro?

Looking ahead

What does our analysis tell us about the future ofthe euro? Some believe that most of the demandgrowth in the U.S. is driven by a bubble in thestock market. If so, U.S. demand will fall when thebubble bursts. The timing and modalities of thisadjustment are, however, hard to predict.Moreover, one should also note here that almosteverything is possible in “bubble land”. Forinstance, the burst of a bubble in the U.S. couldimply a strong speculative (that is, non-fundamen-tals related) asset sale in European markets aswell. In other words, at the burst of the U.S. stockmarket bubble, the euro could fall further.Alternatively, the European market may attract anew wave of “irrational exuberance” (why not?),pushing the euro to very high values – hasn’t thishappened to the yen before?

Whether or not there is a bubble, the high U.S.demand has to be driven by expectations of a high-er future output, leading U.S. consumers and firmsto spend today more than they produce, thus gen-erating a large current account deficit. As it is wellknown, a large current account deficit todayimplies that a country will have to increase itsexports in the future. This is true for the U.S., evenafter accounting for the advantages of being theissuer of an international reserve currency in termsof low borrowing costs and easy access to interna-tional lending.

Increasing exports will be easier for the U.S. in ascenario of sustained growth. A key question for

the U.S. is thus whether current growth rates willtend to persist in the future, or, rather, whether thecurrent performance is only an adjustment to ahigher level of output – so that, once this adjust-ment has taken place, growth is expected to slowdown substantially.

Since increasing U.S. exports will tend to lower theexternal value of the U.S. currency, could thedynamic of adjustment to a large current accountlead, per se, to a turnaround of the recent upwardtrend of the dollar? There are two reasons to besceptical. First, experience shows that the timing ofsuch adjustment is highly uncertain and may hap-pen far in the future. Second, even assuming thatadjustment is around the corner – perhaps helpedby U.S. monetary policies –, the euro-dollarexchange rate is a relative price, its value alsodepending on the future growth performance ofEurope.

To sum up

In 1998, many analysts predicted that investmentopportunities in the newly created euro economywould generate a large capital inflow – a scenarioleading to a strong currency and calling for atighter monetary policy. At the time, however,internal demand in the area was quite low, hardlysuggesting expectations of a coming boom.Consistently, already at the end of 1998, monetaryconditions in the Euro-area were quite relaxed –and became even more so with the large interestcut in April.

The expected large capital inflows never materi-alised. If anything, the flow of European foreigndirect investment and portfolio investment abroad– already large in 1998 – has been increasing in1999 and 2000. According to our perspective, acombination of strong growth in the U.S. and arelaxed monetary stance in Europe appear to haveconvinced investors that, at least in the short-run,the euro will not appreciate considerably vis-à-visthe dollar.

As the graphs illustrate, the correlation betweenEuro-area – U.S. growth differentials and the euro-dollar exchange rate has been quite strong and per-sistent over time. To my knowledge, there is noother factor that has a comparable empiricalrecord.

While, at the time of the writing, news confirms thepattern in the graph, the correlation betweengrowth differentials and the exchange rate maybecome less strong in the future, especially if theU.S. performance becomes more predictable. Thisis to stress that a growth-centred perspective of theeuro-dollar exchange rate is far from being simply“cyclical”. What mattered in the past few monthshave not been cyclical differences, but the uncer-tainty about the pace of two economies dealingsimultaneously with the unknown effects of techni-cal change, and unprecedented monetary reform.

CESifo Forum 36

Focus

... and on the relative pace at

which it deals withthe effects of

technical change

CESifo Forum37

Pro and Contra

PRO INFLATION TARGETING:YES, BUT ...

PETER BOFINGER*

nflation targeting was born “a child of need”. Inthe late 1980s and early 1990s monetary policy

strategies that had seemed to have worked quitewell in the past, suddenly became inapplicable. TheERM crises of 1992/93 ended the strategy ofexchange rate targeting for the United Kingdom,Finland and Sweden almost over night. In Canadaand New Zealand policy makers were becomingincreasingly dissatisfied with monetary targeting.

Confronted with the need to fill a vast conceptualvacuum, central bankers decided to focus on theobvious: a monetary policy approach that aims at alow inflation rate by pursuing an interest rate poli-cy that is mainly guided by available inflation fore-casts for two years ahead. Traditional intermediatetargets like the money stock and the exchange rateare no longer given a “prominent role”, but theyare still regarded important determinants of theinflation forecasts. For this policy framework theappealing name “inflation targeting” was invented.

In a very general form inflation targeting can beregarded as synonymous to what is normally (andespecially in Germany) understood as a stability-oriented monetary policy. Thus, almost all majorcentral banks could be regarded as inflation tar-geter. This applies even to those central bankswhich pretend to follow a strategy of monetary tar-geting. For instance, reaction functions for theBundesbank show very clearly that money growthwas not a dominant factor in the decisions of itsCentral Bank Council. For the ECB, the role of its“reference value” for the money stock M3 cannotyet be adequately assessed, but its first interest ratedecision, a reduction on 9 April 1999 was clearlyincompatible with an excessively high monetary

growth rate at that time. In the case of the FederalReserve System, no specific intermediate targethas been announced in the last decade and there isalso no doubt that Alan Greenspan is following aforward-looking approach in the Fed’s interest ratedecisions.

Of course, to many adherents of inflation targeting,this approach goes beyond a general commitmentto price stability as the main target of monetarypolicy. It is above all Lars Svensson1 who regardsinflation targeting as a clearly specified policy rule(“targeting rule”) and who emphasises the needfor central banks to publish their internal inflationforecasts.

The monetary policy rule of inflation targetingsounds relatively simple. A central bank comparesthe inflation forecast for two years ahead (which isconditional on the actual short-term interest rate)with its inflation target. If the forecast exceeds thetarget, the short-term rate must be raised and viceversa. Analogous to monetary and exchange ratetargeting, the inflation forecast is regarded anintermediate target of monetary policy; conse-quently Svensson has coined the term “inflation-forecast-targeting”. Upon closer scrutiny this ruleis far from being simple. While it is easy to comparetarget and forecast, it is by no means easy to arriveat an inflation forecast. As the proponents of infla-tion targeting only offer very general recommen-dations for this difficult task (“look at everythingwhich is needed as an input for a good economet-ric model”) , their “rule” provides a central bankwith little help in its difficult business of settingshort-term interest rates. Thus, compared with a“simple rule” like a Taylor rule or the rule of mon-etary targeting which both reduce the complexityof a central banker’s decision process, “rule” is sim-ply a misnomer when applied to inflation targeting.

The second key feature of narrowly defined infla-tion targeting is the publication of a central bank’sinflation forecast. It is argued that this (ideally

INFLATION TARGETING

* Professor of Economics at the University of Würzburg.1 Lars E.O. Svensson (1999), “Inflation Targeting as a MonetaryPolicy Rule”, Journal of Monetary Economics, forthcoming.

I

CESifo Forum 38

Pro and Contra

together with the minutes of a central bank’s deci-sion making body) would enhance the accountabil-ity of monetary policy. In this context it is neces-sary to make a distinction between two forms ofaccountability. Ex-post accountability means that acentral bank has to justify actual deviations of theinflation rate form its target rate. Ex-ante account-ability is a forward-looking concept according towhich a central bank is already to be held account-able for expected deviations of the inflation ratefrom its target value. Because of the long lags ofmonetary policy decisions this forward-lookingdimension of accountability is certainly veryimportant. The main question is whether the cen-tral bank’s own forecast is an ideal benchmark forthat exercise. Following the literature based on theBarro-Gordon model, one must be aware of therisk that a central bank that intends a policy of sur-prise inflation will not be providing an honest fore-cast. Therefore, ex-ante accountability should bebased on the external assessment of future pricechanges. Useful data for that purpose may beobtained from the forecasts of international insti-tutions (IMF,OECD), from consensus forecasts ofprivate institutions, from the forecasts that areimplicit in financial market prices, and from wageagreements. Thus an efficient ex-ante accountabili-ty normally does not require a forecast by the cen-

tral bank itself. The only exception is a situationwhere the central bank’s forecast deviates fromexternal forecasts. By publishing its forecast andexplaining how it has been derived the centralbank could try to correct an inadequate externalassessment of its policy.

All in all, if inflation targeting is understood in thevery general sense of a “stability-oriented mone-tary policy” there is little one could object to suchan approach. However, such a framework provideslittle concrete guidance for the conduct of mone-tary policy. This is somewhat different for the morenarrowly defined inflation targeting suggested e.g.by Lars Svensson. His “targeting rule” is useful asa very broad guideline, but for concrete interestrate decisions it simply states the obvious. Themost important element of this more narrowapproach consists of the inflation forecasts andtheir very useful contribution to the ex-anteaccountability of monetary policy. For that pur-pose, external forecasts of inflation are the key fac-tor, while the central bank’s own forecast is onlyneeded if it differs considerably from externalforecasts.

For more information on inflation targeting see myhomepage: www.geldpolitik-online.de

CESifo Forum39

Pro and Contra

CONTRA:NO, BECAUSE ...

FRANCO REITHER*

efore discussing some conceptual shortcom-ings of inflation targeting as a strategy for

monetary policy, it may be useful to characterisethe constitutional and macroeconomic frameworka central bank faces. The typical features of the sce-nario envisaged here are the following: The centralbank is provided with instrument independence,but its ultimate objective is determined externally.At whatever institutional level this ultimate objec-tive may be defined, it is assumed to consist inachieving and maintaining price stability. It is fur-ther assumed that the decision-making body of thecentral bank is made up of persons of differentminds and economic policy ideas.

The principal macroeconomic constraints the cen-tral bank faces are (1) the various lags governingmacroeconomic adjustment processes because oflimited flexibility of production, goods prices andnominal wages, (2) the possibility of boundedrationality of expectation formation by the public,and (3) excess volatility of flexible exchange rates.

As regards the purely theoretical viewpoint, LarsSvensson, in particular, has demonstrated innumerous papers1 the overwhelming appeal of thestrategy of inflation targeting, as well as the inferi-ority of monetary targeting. Svensson’s work isworth mentioning because it is one of the few thatexplicitly recognise the lag problem in monetarypolicy making. However, the lag structure is notthe only constraint on monetary policy. An impor-tant additional constraint is given by the fact that

neither market agents nor the central bank have

full knowledge of the ‘true’ macroeconomic model

(in theoretical analyses, this problem is simply

assumed away via the ‘rational expectations’

hypothesis). Since precisely this is the main feature

of the real world that substantially justifies a dis-

cussion of monetary policy strategy, analyses that

disregard this point are of limited value.

Because of the various lags in the transmission

process, the choice of the intermediate target is

critical. The intermediate target should enable

market participants to assess the performance of

monetary policy even in the short run. Addition-

ally, it is the main object of daily communication

between the central bank and the public. Under

inflation targeting, the forecasted inflation rate

serves as the intermediate target. This value is

based on a macroeconomic model which has to be

developed, estimated, and published by the central

bank itself. In terms of credibility and accountabil-

ity, the basic flaw of this procedure lies in the pub-

lic having to monitor monetary policy on the basis

of information which is supplied by the central

bank itself. This means that, in the short run, the

public will not be given any opportunity to com-

pare central bank announcements and actual mea-

sures with statistically objective data. Moreover,

under inflation forecast targeting the central bank

would run an additional credibility risk concerning

its performance as a forecasting agent.

In practice, it is impossible to judge current mone-

tary policy decisions by their effects on the ulti-

mate goal. This is, first, the consequence of the lag

problems. Second, actual inflation is measured as

the change of a price index in discrete time; this

statistical feature does not permit the breakdown

of price index movements into level changes and

trend changes. But price level shifts can result from

demand or supply shocks, induced by fiscal policy,

wage policy or international influences. Therefore,

the adequate assignment is to hold monetary poli-

cy responsible for the trend component alone. This

implies that, if this fact is disregarded, the concept

of inflation targeting will create an incentive for

B

* Franco Reither is Professor of Economics at the Universität derBundeswehr Hamburg.1 See, for example, Lars E.O. Svensson (1997), “Inflation ForecastTargeting: Implementing and Monetoring Inflation Targets”,European Economic Review 41, 1111–1146; Lars O.E. Svensson(1999) “PRICE STABILITY AS A TARGET FOR MONETARYPOLICY: DEFINING AND MAINTAINING PRICE STABILI-TY”, NBER Working Papers No. 7276, August.

moral hazard behaviour in the other economic pol-icy areas.

The final point concerns the decision makingprocess within the policy making body of the cen-tral bank. Here, the strategy is of great importancein that it serves as a kind of constitutional platformthat cannot be questioned on a case-by-case basis.Although this is a necessary requirement for atime-consistent monetary policy, it is not sufficient.This is because it would be wrong to treat the pol-icy making body as a homogenous group of indi-viduals who are always in agreement. The possibil-ity that some members have discretionary policyambitions cannot be ruled out a priori. A well-designed strategy will help to minimise the risk ofdiscretionary decisions. This risk is the higher, themore opportunities the strategy offers for debatingquantitative figures, as provided by the need to setforecasted values for particular macroeconomicvariables or to use parameter estimates. In thesedebates, discretionary policy ambitions may wellbe concealed by ‘judgements’.

CESifo Forum 40

Pro and Contra

CESifo Forum41

Spotlights

EURO-ZONE EXPORTS:FROM GLOOM TO BOOM

At the beginning of 1999 euro area exports(excluding intra-area trade) still suffered from the1998 crisis in the Asian emerging marketeconomies. In the first quarter they fell by 3.4%year-on-year, in the second quarter by another0.8%. During the second half of 1999, however,euro area exports witnessed positive rates ofgrowth of 6.7% in the third quarter and 11.9% inthe fourth quarter. No doubt, the weak euro washelpful but the main reason for the export recoveryhas been the upturn in the world trade cycle. Thegrowth of the euro area’s export markets1 acceler-ated from 2.7% year-on-year in 1999Q1 to 8.4% in1999Q4 (see chart).

The primary factor behind the rebound in worldtrade has been the recovery in Asia. As the IMFpoints out in its recent World Economic Outlook,global economic and financial conditions haveimproved dramatically during the past year. Theemerging economies in Asia have for the most partstaged a strong V-shaped recovery, and the transi-tion countries and the countries in Latin Americahave also begun to overcome the turbulence thatparticularly affected Russia and Brazil. Among thecrisis-hit countries, the pickup has been strongestin Korea, where GDP increased by 10.5% in 1999,and in Malaysia where growth was 5.5 percent.Thailand grew by 4.5% in 1999, and even inIndonesia – the hardest hit economy in the region– real activity was broadly stable for the year as awhole, a notable turnaround from the 13% con-traction in 1998. The driving forces of the Asianrecovery have been generallysimilar among the crisis-affected economies. With ini-tial support provided by fis-cal stimulus and a rebuildingof inventories, the upturn hasbeen driven more recently bya fast improvement in thevolume and value of exports.This has been supported byreal devaluations comparedwith pre-crisis values andexport market growth, thelatter reflecting continuedstrength in the United States.For the Asian countries in

question the U.S. are an important export market,absorbing between one fifth and one third of theirexports of goods.

Of the main trading blocks, Japan’s export marketshave been the first to be hit by the Asian crisis andare the first to come out of it. Europe and theUnited States are less exposed to trade with Asia(Table). The euro area does more than a third of itstrade with neighbouring countries in westernEurope. The share of exports to the US is almost asbig as the share of exports to Asia. Another impor-tant trading area is Eastern Europe which absorbsroughly one tenth of total exports. This balancedstructure of the euro area’s exports helped todampen the impact of the Asian crisis. But then italso takes more time for the Asian recovery to feedthrough to EU-11 exports.

The recent forecast of the six leading economicresearch institutes in Germany sees the growth ofexports of the euro area softening in the course of2000 as the effect of the weak euro fades and thegrowth of U.S. domestic demand and imports slowsdown. E.L.

Trade with Asia OtherUS EU-11 Japan Asia Other

Share of Asia’simports 15,0 13,2 18,4 41,3 12,1Exports to Asia/total exports 19,2 15,0a) 34,7 39,0 –

a) Excluding intra-EU trade.

Source: HSBC.

1 The weighted average of imports of goods in the euro area’s mainexport markets.

FISCAL MONITORING IN THE

EURO-AREA

Since the introduction of the euro, monetary policyhas been centralised while fiscal and wage policieshave not. For fiscal policy, however, the „Stabilityand Growth Pact“ of 1997 sets up a monitoring andintervention mechanism which aims at reducingbudget deficits and public debt levels in order tosupport (or at least not to counteract) the monetarypolicy measures of the European Central Bank. Thismechanism rests essentially on three pillars:

• A medium-term early warning mechanism(“Stability Programmes”),

• a short-term system for monitoring the imple-mentation of the Stability Programmes

• and an “Excessive Deficit Procedure”.

The procedure for the three parts of the monitor-ing and intervention system is described in Boxes 1– 3. The institutions charged with implementing themonitoring and intervention system are theEuropean Commission (EC), the Economic andFinancial Committee (EFC) and the ECOFINCouncil. The European Central Bank plays animportant role in critically analysing the mostrecent versions of the Stability Programmes as wellas the actual most recent budgetary performanceof the euro-area countries.

At the end of 1998, the euro-zone countries pro-duced their first stability programmes, at the end of1999 the second ones.

What can be said about the quality of the monitor-ing and intervention mechanism in the light of pastdevelopments and the near future? Let us firstlook at the facts.

• Budget deficits have not only been pro-grammed to be reduced step by step, but theyhave actually been reduced, in most cases moreor less in line with the programmed values.Certain exceptions are Austria and Portugal

where the EC forecasts higher deficits thanprogrammed for 2001. For 1999 and 2000 theirdeficits are, however, in line with the pro-grammes. The EC’s criticism last spring withregard to Austria led to an immediate – for thetime being verbal – response by Austriapromising to make fiscal adjustments. In spring1999 Italy applied for permission to surpass thedeficit programmed for that year (2.0%) by 0.4percentage points. It received permission(often publicly denounced as the first violationof the monitoring mechanism) – but finally didnot need it. The 1999 deficit was 1.9 %.

• Some countries (Finland, Ireland, Luxembourg,Netherlands) have a substantial budget surplus

CESifo Forum 42

Spotlights

Box 1Early Warning Mechanism

1. Annual submission (before 1 March) of “Stability Pro-grammes” (euro-area countries) and of “ConvergenceProgrammes” (non-euro-area EU member) to ECOFINCouncil, to EFC and to EC.

Content (for both types of programmes):• medium-term development of budgetary position; tar-

get according to “Pact”: close to balance or in surplus• main underlying assumptions• relevant budgetary and other economic policy measures• sensitivity analysis of changes in the main economic

assumptions• coverage of procedure, current and at least 3 follo-

wing years

2. EC and EFC elaborate (separately) assessments of theprogrammes.

3. ECOFIN Council examines – on the basis of these as-sessments and within at most 2 months after submis-sion of the programme –• whether the medium-term budget objectives provide

for a safety margin to avoid an excessive deficit• whether the assumptions are realistic• whether the measures are sufficient

4. If ECOFIN Council considers that the objectives andcontents of a programme should be strengthened, theCouncil “shall ... invite the member state concerned toadjust its programme.” (1466/97; Art. 5,2)

EC: European CommissionECOFIN Council: Council of Ministers of Economic andFinancial AffairsEFC: Economic and Financial Committee.

Journal “European Economy”, April 2000; Monthly report ofthe European Central Bank, May 1999 and March 2000;Weiland, R., Der Stabilitäts- und Wachstumspakt, in:Wirtschaftsstudium, Februar 2000; European Commission,Economic and Monetary Union – a Juridicial Compilation,Brussels, June 1999.

CESifo Forum43

Spotlights

which, except for the Netherlands, has evenbeen increasing.

• In no country is the budget deficit close to oreven higher than 3%. Thus, experience with theExcessive Deficit Procedure is still lacking.

• In all countries the public debt level will be lowerin 2001 (EC forecast) than in 1998. Only Belgiumand Italy are still far above the 60% ceiling.Whereas the surplus countries were able to reducetheir public debt substantially, progress of thedeficit countries is much slower. The public debthas been reduced in the deficit countries, however.

On the basis of these facts one is inclined to drawa favourable conclusion as regards the effective-ness of the monitoring mechanism. Account mustbe taken, however, of the fact that fiscal consolida-

tion has profited from an advantageous growthenvironment in many countries. Seen in this per-spective, the deficit reduction actually achieved orto be expected in 2000 and 2001 is not overly ambi-tious. This point is specifically stressed by theEuropean Central Bank. On the other hand, partof the deficit stems form the fiscal reform (taxreduction) efforts of some countries. Anotheraspect is the budgetary risk of ageing populations.If this is taken into account deficits should bereduced much faster than planned. R.O.

Box 2Monitoring the Implementation of the Stability

Programmes

1. Semi-annually (1 March and 1 September)submission of budgetary data by member states to EC

2. EC and EFC examine (separately) the budgetary data asto whether they comply with budgetary discipline, i.e.whether there is (a risk of) an excessive deficit.

Two cases of “budgetary discipline”• Budget deficit < 3 %

and: debt < 60 % or sufficiently diminishing or approa-ching the reference value (60%) at a satisfactory pace

• Budget deficit > 3 %,but: the deficit is not regarded as excessive, because it isexceptional and temporary and close to the referencevalueand: debt ratio is sufficiently diminishing or approa-ching the reference value at a satisfactory pace

3. If the EC sees (a risk of) an excessive deficit, the Excessi-ve Deficit Procedure (EDP) is started (see Box 3).

Journal “European Economy”, April 2000; Monthly report ofthe European Central Bank, May 1999 and March 2000;Weiland, R., Der Stabilitäts- und Wachstumspakt, in:Wirtschaftsstudium, Februar 2000; European Commission,Economic and Monetary Union – a Juridicial Compilation,Brussels, June 1999.

Box 3Excessive Deficit Procedure

1. If there is (a risk of) an excessive deficit, ECOFIN Coun-cil is informed by EC by EFC.

2. ECOFIN Council (11 euro-area countries + 4 non-euro-area countries; together 87 votes) decides by qualifiedmajority (i.e. 2/3 of 87) whether the deficit is excessive ornot.

3. If the deficit is held to be excessive, recommendations tothe member (not published) are made to the member sta-te concerned.

4. ECOFIN Council assesses the effectiveness of the mea-sures announced by the member state.

5. If the measures are regarded as ineffective or not im-plemented, ECOFIN Council may make its recommen-dations public and give notice to take own measures.

6. If the excessive deficit persists - due to non-implementa-tion or despite implementation - ECOFIN Council ap-plies sanctions on the member state concerned.

7. Sanctions will consist in a non-interest bearing deposit of0.2 % of GDP (fixed component) + a variable compo-nent of 1/10 of the difference between factual deficit/GDP and the reference value (3 %). 0.5 % of GDP is theupper limit for the deposit. The deposit might be conver-ted into a fine if the excessive deficit has not been correct-ed within 2 years.

Journal “European Economy”, April 2000; Monthly report ofthe European Central Bank, May 1999 and March 2000;Weiland, R., Der Stabilitäts- und Wachstumspakt, in:Wirtschaftsstudium, Februar 2000; European Commission,Economic and Monetary Union – a Juridicial Compilation,Brussels, June 1999.

THE NETHERLANDS:AN EMPLOYMENT SUCCESS

STORY DESPITE SOME

BLEMISHES

With the Wassenaar Agreement of 1982 theNetherlands introduced employment-conducivereforms. These reforms were made possible by theco-operation of labour unions, employer associa-tions and the government. The employment strate-gy consisted in longer-term wage moderation and areduction of working time with the aim of reallo-cating jobs in a cost-neutral way. It was supple-mented by reforms of statu-tory labour market regula-tion and of social security.

The success of this strategywas reflected in an increaseof Dutch employment by31% between 1982 and 1998.In the same period WestGerman employment onlyrose by 3% (see Fig. 1). Thecomparatively good employ-ment performance in theNetherlands is due in largemeasure to the expansion ofpart-time work. The share ofpart-time employment con-tracts in total jobs increasedfrom 21.4% in 1983 to 30%in 1998 (the figures for WestGermany are 12.6% and16.6%, respectively).

The fact that part-time jobs

were created faster than full-

time jobs hardly detracts

from the success of Dutch

employment policy. Part-

time work largely meets

workers’ needs. In 1994 only

8% of Dutch part-time work-

ers reported a preference for full-time work. In addi-

tion, the growth of part-time work was not merely a

redistribution of jobs. Rather, in the Netherlands

total hours worked have risen by 17% since 1982,

whereas in West Germany they have declined by 6%

(see Fig. 1). If total hours worked in West Germany

had grown at the same rate as in the Netherlands,

they would have risen from 46 billion hours in 1982

to almost 54 billion hours in 1998. They would thus

have exceeded the actual number of hours worked in

1998 by about 11 billion hours. Given the 1560 work-

ing hours per employed person calculated for 1998,

this difference would amount to an increase of

roughly 7 million jobs in West Germany alone.

German unemployment would have been erased.

CESifo Forum 44

Spotlights

Figure 1

CESifo Forum45

Spotlights

The good employment performance in theNetherlands has caused a decline in joblessness.The Dutch unemployment rate has fallen since

1983. It was not until 1993,however, that it came to liebelow the German unem-ployment rate which hasbeen calculated for pan-Germany since that year.Since then it has been con-siderably lower, reaching 4%in 1998 (see Fig. 2).

The decline in the Dutchunemployment rate has alsobeen caused by the fact,however, that members ofthe active labour force onwhich the calculation of theunemployment rate is basedbecame recipients of socialsecurity. The number of per-sons receiving disability ben-efits surged in the secondhalf of the 1970s and hassince remained at this levelwhich exceeds the interna-tional average. Furthermore,during the 1980s the numberof social assistance recipientsand early retirees increased.Only subsidised employmentprogrammes did not gain inimportance (see Fig. 3). Itmust be assumed that thegreater reliance on the socialsecurity system has benefitedthe labour market. If eligibil-ity requirements (especiallyfor disability benefits) hadbeen more restrictive, a larg-

er proportion of the labour force would have beenforced to take up a job.

W.O.

Figure 2

Figure 3

PRODUCTIVITY IS NOT

EVERYTHING – AN INTER-NATIONAL COMPARISON OF

THE SOURCES OF PROSPERITY

Economic growth in Germany has a low employ-ment intensity. Because average labour productivi-ty has risen relatively fast, employment has notshown the desired increase despite economicgrowth. Other countries like the United States,New Zealand and Switzerland, on the other hand,have demonstrated successful job creation via ahigher employment intensity of their economicgrowth. Their labour productivity has risen relati-vely slowly. Accordingly, their employment hasexpanded faster.

Successful job creation ...

In light of the successful employment performanceof other countries there have been repeateddemands in Europe for boosting employment, espe-cially in the low-wage segment, by wage, social, andtax policies, at the expense of lower aggregate pro-ductivity growth. Critics of these demands point outthat high productivity growth has been the source ofGerman prosperity, for example. This argument isfundamentally correct but does not take into consi-deration that productivity increases only measurethe change in output of those in employment butnot the degree to which people are included in gain-ful employment. The higher the participation rateand the lower the unemployment rate, however, thegreater is the number of people who contribute tothe creation of GDP and thus to the increase inaverage per capita income. The same applies to anextension of annual working hours.

This relationship can be expressed as:

Y • H • E • L • P15–64 = YH E L P15–64 P P

where:Y = Gross domestic product (GDP)H = hours worked per year (“ labour

volume”)E = Number of persons employedL = Total labour force (employed persons

+ registered unemployed)P 15–64 = Working age populationP = Total populationY/H = GDP per hour worked (1)H/E = Average annual working time per

employed person in hours (2)E/L = Employment rate (3)L/P 15–64 = Participation rate (4)P 15–64/P = Share of working-age population in

total population (5)Y/P = Per capita income (6)

The extent to which the individual componentscontribute to per capita income has been determi-ned by the Conference Board, New York, in aninternational comparison of the major OECDcountries.2 The country figures for 1997 presentedin the Table below show deviations from theOECD average.

Three country groups may be distinguished:

• Countries whose per capita income positionclearly lags behind their productivity position.Included in this group are Belgium, France, theNetherlands, Greece, Italy, Spain, Ireland,Germany, the United Kingdom, and Turkey.

• Countries whose per capita income positionroughly corresponds to their productivity positi-on (Column 6 – Column 1 ≤ |5|). Included hereare Finland, Sweden, Norway, Austria, Australia,Canada, and Portugal.

• Countries whose per capita income positionclearly surpasses its productivity position. Thisgroup is composed of the United States, NewZealand, Denmark, Sweden, and Japan.

CESifo Forum 46

DICE Reports

1 DICE = Database of Institutional Comparison in Europe.

2 B. van Ark, R. H. McGuckin, International Comparisons of LaborProductivity and Per Capita Income, in: Monthly Labor Review,July 1999.

DICE REPORTS1

CESifo Forum47

DICE Reports

The lag in per capita income in the first group isattributable to different causes depending on thecountry. The low income of the Belgians comparedto their hourly productivity is primarily the resultof a low participation rate to which the practice ofearly retirement is a major contributing factor.France makes poor use of its potential labour volu-

me (man-hours worked) in several ways: Annualworking hours are relatively low, unemployment isabove average, and the participation rate is belowaverage. In the Netherlands, annual working hoursare at a relatively low level due to the wide-spreadpractice of part-time work. Although the latter isan important element of the country’s successfuljob creation, it is achieved at the expense of percapita income. In Germany, the short workinghours, high unemployment and a below-averageparticipation rate contribute to the fact that percapita income is below the OECD average.

The countries in the third group display relativelyhigh per capita incomes because they use theirpotential labour volume (man-hours worked)more intensively than other countries. In Japan, alarger proportion of the population is in employ-ment than in other countries and the Japanese alsowork longer hours. In Switzerland, Denmark andthe United States, the participation rate is aboveaverage. In all three countries the female labourparticipation rate is high. The New Zealanders,finally, increase their income by working longerhours than is common in other countries.

... with lower productivity growth

In the countries in the last group, the potentiallabour volume is utilised to a large extent. Thesecountries display, among other things, a relativelylow unemployment rate. They are regarded ashaving successful employment policies. Unlikemany Continental European countries, they did nottry to solve their unemployment problems by redu-cing the labour supply. They did, however, achievetheir good employment performance at the expenseof a slowdown of productivity growth. W.O.

Components of per capita income in aninternational comparison, 1997

(in % of the OECD average)

GDP Effect of Effect of Effect of Effect of Per per annual employ- participa- age capita

Country hour working ment rate tion rate structure incomea)

worked hours(1) (2) (3) (4) (5) (6)

Belgium 128 – 5 – 3 – 19 – 1 101France 123 – 9 – 6 – 9 – 2 97Nether-lands 121 – 26 2 – 4 2 96Greece 75 – 4 – 2 – 11 1 58Italy 106 – 11 – 5 – 1 2 91Spain 84 13 – 14 – 13 2 71Ireland 108 5 – 4 – 12 – 3 95Germanyb) 105 – 5 – 3 – 4 2 96UnitedKingdom 100 – 9 0 3 – 2 92Turkey 36 2 0 – 8 – 1 29

Finland 93 0 – 7 2 0 88Sweden 93 – 3 – 3 6 – 4 88Norway 126 – 17 4 12 – 4 122Austria 102 – 4 3 – 2 1 100Australia 96 0 – 1 2 0 97Canada 97 2 – 2 2 2 100Portugal 56 2 0 1 1 60

United States 120 – 1 3 9 – 2 128New Zealand 69 8 1 3 – 1 79Denmark 92 0 1 9 1 103Switzer-land 94 0 3 12 1 111Japan 82 10 4 6 4 106

OECDc) 100 0 0 0 0 100

a) Total of the columns (1) to (5). – b) Including former East Germa-ny. – c) All listed countries.

Source: US Department of Labor, Monthly Labor Review, July1999, p. 36.

RECRUITMENT OF HIGHLY

SKILLED LABOUR ABROAD –THE PRACTICE OF SELECTED

OECD COUNTRIES

The dynamic development of the information tech-nology industry has led to bottlenecks in thelabour market for highly skilled IT specialists inmany OECD countries. To reduce these shortages,several countries are devoting intensive efforts tofinding specialists abroad. Other countries have arestrictive entry policy for foreign specialists. Thefollowing is a presentation of the extent to whichselected OECD countries recruit highly qualifiedspecialists and managers abroad and the condi-tions they attach to this recruitment.

The immigration policies of Australia, Canadaand the U.S. have become increasingly labour-market oriented

Countries such as Australia, Canada and the U.S.pursue an explicit immigration policy. They admitskilled labour not only on a temporary but also ona permanent basis.1

Since 1984 Australia’s immigration policy has beenincreasingly aimed at highly skilled workers. Theselection of specialists is done by means of a pointsystem which (since 1998) includes the chances theforeign specialists will have in the Australianlabour market and their possible contribution tothe economy of Australia. In 1997, nearly 20,000skilled workers (including family members)migrated to Australia, corresponding to 23% oftotal immigration to the country (Table 1).

Canadian immigration policy is also assigningincreasing importance to the immigration of highlyskilled workers and entrepreneurs who wish toestablish businesses. Here, too, a point system is used

to determine whether a foreign applicant can inte-grate successfully into the Canadian labour market.Potential entrepreneurs must show proof that theyhave the necessary prerequisites for the establish-ment of a business. There are no quotas in Canadianimmigration policy. In 1997, 50,000 specialists andentrepreneurs migrated to Canada (and an addition-al 75,000 family members). This was 58% of totalimmigration (Table 1). Most specialists and entre-preneurs came from Hong Kong, China, Taiwan,India, Pakistan and other developing countries.

U.S. immigration policy has traditionally placed thegreatest weight on family reunion. Since the immigra-tion law of 1990, however, the need for skilled work-ers has received greater emphasis. The U.S. sets annu-al immigration quotas. Workers are selected accord-ing to a five-stage scale of preferences indicating thedesired qualifications. The highest preference is givento professionals like outstanding professors and sci-entists, as well as managers of multinational corpora-tions. The second preference is given to academics ingeneral. In 1997, 40,000 skilled workers received aGreen Card for the U.S. (and an additional 50,000family members). Employment-based immigrationaccounted for 11% of total immigration (Table 1).

Temporary migration of highly skilled workers

Whereas only a few OECD countries pursue a tar-geted immigration policy, many countries recruit

CESifo Forum 48

DICE Reports

1 This article is based to a large extent on OECD/SOPEMI, Trendsin International Migration, 1999 Edition, Paris 1999.

Table 1Immigration of Skilled Labour to Australia,

Canada and the United States in 1997(thousands)

Workers Family Total Total (3):(4) Popu-mem- immi- in % lationbers gration

Country (3) =(1) (2) (1) + (2) (4) (5) (6)

Austra-lia . . . . . . 19.7 85.5 23.0 18,520a)

Canada 50.5 75.0 125.5 216.0 58.1 30,265b)

U.S. 40.3 50.3 90.6 798.4 11.3 264,500a)

a) 1996. – b) 1998.

Source: OECD, SOPEMI, Trends in International Migration,1999 edition, Paris (pp. 119, 224, 262, 263 and 266).

CESifo Forum49

DICE Reports

highly skilled workers abroad for temporaryemployment. This applies for the above-describedimmigration countries (Australia, Canada, U.S.),but it is also common practice in Japan and Korea,which strictly (or somewhat less strictly) rejectimmigration, as well as in the United Kingdom.France, the Netherlands and Germany, in contrast,only recruit very few highly skilled workers fromnon-EU countries.

The objective of recruiting foreign workers fortemporary employment is to increase labour mar-ket flexibility and to remove labour market bottle-

necks in the host country. This also facilitates thecross-border transfer of staff in multinational com-panies. Temporary admission means that the for-eign workers receive only a contract and/or a resi-dence permit for a limited period of time, andwhen the time has expired they must leave thecountry unless they obtain a change in resident sta-tus. The recruitment of foreign workers is primari-ly aimed at highly skilled workers – a category thatis very difficult to define.2 Correspondingly, the

Table 2

Country

Admission conditions Availability of Quotas Length of stay Restrictions Possibility Possibilitydomestic (possibility for renewal) on for for

workers as activity family changinggrounds for reunion status

refusal

Australia Nominated by employer Yes No 2 years (renewable once) Yes Yes Yes(exception: stay of less 4 years (renewable) for

than three months) teachers3 months to 4 years for

business specialists(TBE visa)

Canada Work permit required Preliminary authorisation Yes No Maximum 3 years Yes Yes No

(renewable) No work permit re- Bilateral agreements No No Maximum 9 months No Yes Yes quired (renewable)

United States H-1B (specialists) Prevailing wages required No Yes Inititial admission for 3 Yes Yes Yes

BA (4 years degree) years (renewable once) + practice in the

occupation O (extraordinary ability) Consultations with peers No No Up to 10 years Yes a) Yes Yes

depending on activity

Japan Ministry of Justice No No Depends on the contract Yes Yes b) Noapproval (thorough exami- (renewable with thenation of special abilities) same employer)

Korea Very strict conditions No No No ceiling Yes No Noimposed on skills

United Kingdom c) Employer must apply Yes d) No Up to 4 years (renewals Yes e) Yes f) Yes g)for work permit possible)

Restricted to highlyskilled persons ("key

workers"). Adequate

France Employer must apply Yes h) No 9 months (renewable Yes No Nofor work permit once, and in exceptional

cases twice)

Netherlands Employer must apply Yes i) No 1 year (renewable) Yes Yes j) Yesfor work permit

Germany k) Employer must apply Yes No 1 year (renewable) Yes Yes l) Nofor work permit

a) Must continue to work in the field of expertise.- b) Without work permit.- c) Under certain conditions (members of Commonwealth, etc.)no work permit required.- d) Exceptions for certain activities.- e) Change of employer only allowed under special circumstances.-f) With sufficient income and housing.- g) After 4 years working in the UK.- h) Exception: workers who earn more than FF22,000 a month.-i) Exception for employees in multinational firms.- j) After a stay of 3 years. k) According to Par. 5, Sect 2 Recruitment Stop Ordinance.-l) Dependent on the quality of the residence permit.

Source: OECD, SOPEMI, Trends in International Migration , 1998 and 1999 editions, Paris (pp. 202 ff and 26 ff.).

Residence conditionsRecruitment conditions

Conditions for the Recruitment and Temporary Residence of Skilled Foreign Workers in Selected OECD Countries

2 For the definition problems, see J. Salt, International Migration ofthe Highly Skilled, OECD Working Papers, No. 91, Paris 1997.

classification differs among countries, making aninternational comparison difficult.

The employer in the host country must usually sub-mit an application for a foreign worker to enter thecountry. The authorities then examine the extent towhich the worker meets the required qualifica-tions. In Australia, Canada, the United Kingdom,France, the Netherlands and Germany a furtherexamination is made as to the availability ofdomestic workers that meet the qualifications forthe job offered. Such a test is not made in the U.S.,Japan or Korea. The U.S. is the only country to setannual immigration quotas (for H-1B visas but notfor O visas).The residency ranges from severalmonths to four years, but longer stays are also pos-sible. Extensions are usually granted. In all coun-tries the activity to be performed is stipulated. Theoption of family reunion differs from country tocountry. It is possible in Australia, Canada, theU.S., Japan, the United Kingdom and (with limita-tions) in the Netherlands and Germany but notallowed in Korea and France. In Australia, the U.S.,the United Kingdom and the Netherlands the tem-porary work permit may be converted into a per-manent one. In other countries, a change in resi-dent status is not possible (Table 2).3

In the following, the recruitment policy of individ-ual countries will be briefly presented. Australiahas aimed its recruitment policy explicitly at high-ly skilled workers and business people. Theissuance of a temporary work permit requires thatthe applying company has made an effort to fill theposition from the Australian labour pool. The com-pany must prove that it advertised the position,must document the applications it received, andmust list the reasons for rejection (labour markettesting). As a rule, the work permit is granted forfour years, (including extension). As a result of theSkilled Temporary Resident Programme, 14,600skilled workers (and family members) moved toAustralia in 1992 (12,500 in 1997). These specialistscame mainly from the United Kingdom, Japan, theU.S. and Canada (Table 3).

Canada issues temporary residence permits forhighly skilled workers as long as the unemploy-ment rate in the corresponding labour-market seg-ment is low. For particular occupational groups,academics and scientists (on the basis of bilateral

agreements) and especially for urgently neededworkers, no permit is required. Employment thatrequires a permit can last for three years (exten-sions are possible). In 1992, temporary employ-ment in Canada was taken up by 84,900 foreignworkers and in 1995 by 60,000 (Table 3). To elimi-nate the shortage of computer and software spe-cialists, Canada developed a special programme forforeign recruitment in 1997. This programme issupplemented by a long-term education and train-ing programme for domestic IT specialists.

In the United States the temporary inflow of qual-ified workers is promoted even more than immi-gration. For temporary workers (six years includ-ing extensions), no Green Cards are issued. Asidefrom the possibility of recruiting workers underthe terms of the North American Free TradeAgreement (NAFTA), there are two ways torecruit highly skilled workers abroad: applicationfor H-1B visas for specialists and of O visas forworkers with extraordinary abilities. The issuance

CESifo Forum 50

DICE Reports

3 OECD/SOPEMI, Trends in International Migration, AnnualReport, 1998 Edition, Paris 1998. Part III.

Table 3Entries of Temporary Skilled

Workers in Selected OECD Countries1992 and 1997(thousands)

Country 1992 1997

Australiaa) 14.6 12.5

Canadab)

Permit requiredc) 66.4 42.2h)

Permit not requiredProfessionals 5.3 7.8h)

Academics, scientistsd) 5.6 5.0h)

Workers with significant benefits for Canada 4.6 5.0h)

United Statese)

Specialists (H-1B visa) 110.2 144.5i)

Specialists (NAFTA, TN visa)a) 12.5 27.0i)

Workers with distinguished abilities (O visa) 0.5 7.2i)

Japan 108.1 93.9

Korea . . . 42.2

United KingdomWith work permits of 1 year and overf) 12.7 18.7With short-term work permits 14.0 19.0

FranceWorkers on secondmentg) 0.9 1.0Researchers 0.9 1.1

Netherlands 1.9 1.5h)

Germany . . . 0.6j)

a) Including accompanying persons. – b) The figures refer toissued work permits. One person can receive several workpermits. – c) Criterion: low unemployment in the field of ac-tivity. – d) On the basis of bilateral agreements. – e) Every in-dividual border crossing in a year is counted. – f) Special-ists and senior managers. – g) Holders of provisional workpermits (APT). – h) 1995. – i) 1996. – j) 1998 (1999: 0.9).

Source: OECD/SOPEMI, Trends in International Migration,1998 and 1999 Editions, Paris (pp. 189 and 25).

CESifo Forum51

DICE Reports

of these visas requires that the applicants havecompleted a four-year university course and havegathered professional experience (H-1B visa) orthat their qualifications are positively assessed bysimilarly qualified Americans (O visa). Untilrecently, the state of the labour market had notbeen a factor in issuing the visas. Criticism hasbecome louder in recent years regarding the dis-placement of American by foreign workers andwage dumping in violation of the rules. In order tocorrect this “malpractice” the Labor ConditionApplication was introduced in 1998 according towhich an employer who is filing for a visa mustprove that no U.S. worker was dismissed and thatno suitable candidate could be found on the U.S.labour market. 110,200 H1-B visas and 500 O visaswere issued in 1992 and 144,500 H1-B visas and7,200 O visas in 1996 (Table 3). The followingannual quotas for H1-B visas were set up: 115,000(1999), 115,000 (2000), 107,500 (2001), and 65,000(2002). In light of the shortage of highly skilledworkers in information technology and biotechnol-ogy, a reform of the Immigration and NationalityAct is planned under which foreign graduates ofU.S. universities may receive working visas within90 days of graduation (T visas: T for technology),provided they can prove a job offer for $60,000 ormore a year.

Japan only allows foreign workers into the countryon a temporary basis. The Immigration Controland Refugees Recognition Act and other ordi-nances contain neither quotas nor explicit labourmarket reservations. Foreign specialists are select-ed according to qualification profiles formulatedand strictly enforced by the Ministry of Justice. Informulating the requirements, the state of thelabour market is implicitly taken into account. Thelength of the stay varies according to the employ-ment category. Family members may enter thecountry but are not allowed to work. In 1992108,000 and in 1997 94,000 highly skilled workersentered the country for temporary employment(Table 3). These were primarily North Americans,Europeans and Asians.

In view of its increasing integration in the worldeconomy, Korea is dependent on the inflow of for-eign specialists. Only in this way can it enhance itsattractiveness for foreign direct investment pro-jects and achieve the know-how transfer necessaryto increase its technological competitiveness.Korea issues residence permits for foreign workers

on the basis of a strict examination of qualifica-tions. According to SOPEMI data4, 42,000 skilledforeign workers took jobs in Korea in 1997 (Table3). But judging from other SOPEMI data, thisnumber probably contains a large proportion oftrainees. The actual inflow of highly skilled work-ers appears to have been much less, and has prob-ably fallen since the deep recession of 1998.

Like Japan and Korea, the United Kingdom hasalso opened its borders to highly skilled workersfrom non-EU countries. Job seekers with excellentuniversity degrees and at least two years of profes-sional experience as well as key workers may beissued work permits, provided the state of thelabour market allows this. Their stay can last for upto four years. After three years, they may apply forimmigrant status, which about 25% of these work-ers do successfully. The increased openness of theUnited Kingdom is reflected by the rise in tempo-rary work permits for highly skilled workers from27,000 in 1992 to 38,000 in 1997. About 30% ofthese workers came from the U.S., more than 10%from Japan, and the rest from numerous othercountries.

In contrast to the above-mentioned countries, sev-eral large continental European countries allowonly small numbers of highly qualified workersfrom non-EU countries to enter. One of the rea-sons for this is that as EU member states they mayemploy qualified workers from other EU memberstates (or, more precisely, the European EconomicArea). In 1997, only 1,000 highly skilled workers(and 1,100 scientists) from non-EU countries wereallowed to enter France on a temporary basis;including less qualified workers, a total of only4,700 temporary work permits were issued (Table3). The number of permits depends on the state ofthe labour market. This does not apply to employ-ees earning more than FF 21,000 a month. Themaximum stay is nine months; only one extensionis allowed.

In the Netherlands, work permits for highly skilledworkers are dependent on the state of the labourmarket unless the applicants are employees ofmultinational corporations. The permits are grant-ed for one year, but are also renewable. After threeyears of uninterrupted legal employment, foreignworkers no longer need a work permit. After five

4 The OECD migration observation system (Système d’observa-tion permanente des migrations).

years of legal residency, they may apply for immi-grant status. In 1995 1,500 highly skilled workersentered the Netherlands, including managers ofmultinational corporations (Table 3).

In Germany, highly skilled workers from non-EUcountries may be recruited as long as their employ-ment is in the “public interest”. Work permits aregranted for one year, with the possibility of exten-

sions. In 1998 and 1999, 560 and 800 specialists,respectively, came to Germany, many from formerEast Bloc countries and from India. In May 2000,Chancellor Schröder announced plans to recruit anadditional 20,000 IT specialists abroad. As a com-plementary measure, the education and furthertraining of domestic computer and software spe-cialists will be intensified.

Wolfgang Ochel

CESifo Forum 52

DICE Reports

CESifo Forum53

Trends

EUROPEAN FUNDAMENTALS

INCREASINGLY SOUND

In April 2000, 652 economic experts in 80 countrieswere polled by the Ifo Institute’s 68th EconomicSurvey International.

The world economy is expected to continue expanding,albeit at a somewhat slower pace than in past months.In April, the world economic climate remained at theall-time high reached at the beginning of the year, butassessments of future economic activity weakened.

In Western Europe, the upswing is likely to persistbecause of sound fundamentals. The weak euroand an easing of fiscal policy in some countriesshould prolong the expansion in the euro area. Theeconomic climate improved, although assessmentsof future economic activity worsened due to expec-

tations of slowing economic growth in the United

States and higher interest rates.

Economic prospects for Germany are very

favourable. The upswing is still fuelled by strong

export activity, but domestic demand has also picked

up. The planned tax cuts are likely to boost economic

activity in the coming six months. Tax relief is also

planned in France and Italy. In France, the economic

outlook remains very bright, and there are no signs of

inflationary pressure. Robust export activity, strong

private consumption and healthy capital spending

support growth. In Italy, growth is based on exports

and investment, whereas private consumption is still

unsatisfactory. It should benefit from the tax cuts. In

the United Kingdom prospects remain favourable but

survey results point to a slowdown in the second half

of the year, mainly due to the strong pound sterling

which is affecting exports. The trade deficit is expect-

ed to widen.

ECONOMIC SURVEY

INTERNATIONAL

Source: ESI 68, 2/2000.

Present and Expected Economic Situation

CESifo Forum 54

Trends

MONETARY CONDITIONS

IN THE EURO AREA

3-month interest rates in the euro-zone continued their rise in June. TheGoverning Council of the European Central Bank had decided, on 8 June, toraise the interest rate on the Eurosystem`s main refinancing operations from3.75% to 4.25%. The interest rates on the deposit and marginal lending facil-ities were raised to 3.25% and 5.25% respectively. At 6.3%, the3-month average of the annual growth rates of M3 (February to April) movedfurther away from the reference value of 41/2%. Liquidity in the euro areacontinued to be very generous.

Source: European Central Bank.

CESifo Forum55

Trends

* The industrial confidence indicator is an average of responses (balances) to thequestions on production expectations, order-books and stocks (the latter withinverted sign).** The consumer confidence indicator is an average of responses to the questionson the financial situation of households and their assessment of the general eco-nomic conditions, both in the past and future twelve months, and the question onbig-ticket purchases.

The EU industry confidence indicator rose one point in the EUand three points in the euro-zone. The main contributors wereItaly, Germany and France, in that order.Consumers, however, were not so optimistic in June as in the pre-vious month. A decline of one point was found in the EU where-as there was a two-point fall in the euro-area. Portugal, Greeceand Germany were the main sources of this slight correctionfrom a very high level of consumer confidence.

The assessment of order books in terms of balances moved intothe positive range in the past few months. In line with expectedorders, capacity utilisation rose above 83%. This is quite animprovement over late 1999.

Growth of real gross domestic product continued to accelerate,reaching 3.2% year-over-year in the first quarter 2000 both forthe euro-zone and for EU-15. In the euro-zone, in particular, thesmall countries have consistently outperformed the largereconomies. From the first quarter 1995 to the fourth quarter 1999GDP in the six small countries increased by 10.9% compared tothe GDP of the other five which grew by 7.3%. This difference isalso reflected in labour market performance.

a) The indicator of economic sentiment is a weighted average of the industrial con-fidence indicator, the construction confidence indicator, the consumer confidenceindicator and the share-price index. 1985 = 100.

Despite the improvement of the industry confidence indicator(and the construction confidence indicator which is also includ-ed along with the share price index), the economic sentimentindicator for the EU showed a slight decrease in June because ofthe decline in consumer confidence.

EU SURVEY RESULTS

CESifo Forum 56

Trends

a) BIS calculations; to December 1998, based on weighted averages of the euro areacountries’ effective exchange rates; from January 1999, based on weighted averagesof bilateral euro exchange rates. Weights are based on 1990 manufactured goodstrade with the trading partners United States, Japan, Switzerland, UnitedKingdom, Sweden, Denmark, Greece, Norway, Canada, Australia, Hong Kong,South Korea and Singapore and capture third market effects. Real rates are calcu-lated using national CPIs.Where CPI data are not yet available, estimates are used.

The real effective exchange rate of the euro, which had declinedcontinuously since even before the introduction of the commoncurrency, rebounded in June. The nominal USD/EUR rate rosefrom its low of an average 91 U.S. cents in May to 95 U.S. centsin June.

The general fiscal balances in the euro area have steadilydeclined from a deficit of 5.5% of GDP in 1993 to an estimateddeficit of 1% of GDP in 2000. Over the same period, the struc-tural deficit fell from 4.4% of potential GDP to an estimated0.9%. Consolidation is thus showing concrete results.

In Europe, the unemployment rate continued on its downwardtrend, levelling off in May, at 8.5%, in EU-15.A year earlier it hadstill stood at 9.2%. In the euro-zone, unemployment declinedmore, though from a higher level, from 10.0% to 9.2% during thesame one-year period. The pick-up in economic growth is at lasthaving a positive effect on the labour market.

In the euro-zone, the harmonised index of consumer prices rosesteeply in the second half of 1999, peaking at 2.1% in March2000, but receding to 1.9% in April and May. The rise was due tosurging energy prices as the oil price (Brent dated) went from$15.90 p/b in June 1999 to $27.15 in March 2000. During the sameperiod, core inflation, i.e. minus food and energy, was stable, fluc-tuating around 1%. In April, however, core inflation increased to1.2%, only to recede again in May, as food and energy pricesaccelerated again.

EU INDICATORS

ORDER FORM

If you wish to subscribe to CESifo Forum, please fill in the order form below and mail or fax(xx49 89 985369) to the Press and Publications Department of the Ifo Institute.

Subscriber’s name ...............................................................Department ...............................................................Company/Institution ...............................................................Street and number ...............................................................City, state, zip code ...............................................................Country ...............................................................

Please enter . . . subscriptions(s) to CESifo Forum

Annual subscription price: EUR 50.00 plus postage(Members of the ifo Institute: EUR 40.00 incl. postage)

Do not send payment. We will bill you.

............... ................. ......................................Date Reference Signature

Send to: Ifo Institute for Economic ResearchPress and PublicationsP.O. Box 86 04 60D-81631 MunichGermany

International Conferences

Conference Date Place Programme Chairman

Verein für Socialpolitik 19–22 Berlin Karlhans SauernheimerSept. 2000 Germany Univ. of Mainz

E-Commerce and Web 4–6 Sept. GreenwichTechnologies EC 2000 2000 UK

European Association 7–10 Lausanne Thomas von Ungern-Sternbergfor Research Sept. 2000 SwitzerlandIn Industrial Economics

11th ITS European Regional 9–11 LausanneConference Sept. 2000 Switzerland

Roayal Statistical Society 11–15 ReadingSept. 2000 UK

Legal and Valuation Issues of 13–15 Tilburg,e-Commerce and New Tech Com. Sept.2000 Netherlands

Intl. Society for New 22–24 Tuebingen isnie2000@wueconc.wustl.eduInstiutional Economics Sept. 2000 Germany

EMU and Asymmetries in 29 Sept. ParisLabour Markets 2000 France

Environmental Policy, Com- 7–12 Kerkradepetitiveness, Location Behaviour Oct. 2000 The Netherlands

Financial Developments in the 12–14 KavalaEra of Euro Oct. 2000 Greece

Atlantic Economic Society 15–18 Charleston, SCOct. 2000 USA

Chicago Board of Trade 13th 27–28 GlasgowEuropean Futures Symposium Oct. 2000 Scotland

Antitrust Issues in 30 Nov. – Berlin Ralph SiebertInternational Markets 2 Dec. 2000 Germany WZB

ASSA and AEA Annual Meeting 5–7 New OrleansJan. 2001 USA