52000DC0277

Report from the Commission - Convergence report 2000 (prepared in accordance with Article 122(2) of the Treaty) /* COM/2000/0277 final */


REPORT FROM THE COMMISSION - CONVERGENCE REPORT 2000

(prepared in accordance with Article 122(2) of the Treaty)

TABLE OF CONTENTS

Tables

Graphs

1. INTRODUCTION AND MAIN FINDINGS

1.1. Introduction

1.2. Main findings

1.2.1. Greece

1.2.2. Sweden

2. GREECE

2.1. Compatibility of national legislation with the Treaty and the Statute of the European System of Central Banks

2.1.1. Assessment of compatibility in 1998

2.1.2. Legislative action taken since 1998 and overview of the legislation in force

2.1.3. Assessment of compatibility

2.2. Price stability

2.2.1. Inflation developments

2.2.2. Underlying factors and sustainability of inflation

2.3. Government budgetary position

2.3.1. Excessive deficit procedure

2.3.2. Current budgetary situation and prospects

2.4. Exchange rate stability

2.5. Long-term interest rate

2.6. Additional factors

2.6.1. Results of the integration of markets

2.6.2. Balance of payments on current account

2.6.3. Unit labour costs and other price indices

3. SWEDEN

3.1. Compatibility of national legislation with the Treaty and the Statute of the European System of Central Banks

3.1.1. Assessment of compatibility in 1998

3.1.2. Legislative action taken since 1998 and overview of the legislation in force

3.1.3. Assessment of compatibility

3.2. Price stability

3.2.1. Inflation developments

3.2.2. Underlying factors and sustainability of inflation performance

3.3. Government budgetary position

3.3.1. Excessive deficit procedure

3.3.2. Current budgetary position and prospects

3.4. Exchange rate stability

3.5. Long-term interest rate

3.6. Additional factors

3.6.1. Results of the integration of markets

3.6.2. Balance of payments on current account

3.6.3. Unit labour costs and other price indices

ANNEX A: COMPATIBILITY OF NATIONAL LEGISLATION

ANNEX B: INFLATION CRITERION

B.1. Treaty provisions

B.2. Reference value for inflation

B.3. Recent methodological developments in HICP

B.4. Additional tables on other price and cost indicators

ANNEX C: GOVERNMENT BUDGETARY DATA

C.1. Shift to ESA 95 and revision to Regulation (EC) No 3605/93

C.2. Additional tables on the public finances

ANNEX D: EXCHANGE RATE CRITERION

D.1. Treaty provisions and ERM II

D.2. Application of exchange rate criterion in Stage 2 of EMU

D.3. Application of the exchange rate criterion in Stage 3 of EMU

D.4. Fulfilling the exchange rate criterion in the current examination

ANNEX E: LONG-TERM INTEREST RATE CRITERION

Tables

2.1 Greece: average inflation rate (HICP) and the reference value

2.2 Greece: other inflation and cost indicators

2.3 Greece: government surplus/deficit , debt and investment expenditure

2.4 Greece: composition of budgetary consolidation between 1995 and 1999

2.5 Greece: updated convergence programme projections for GDP growth, government surplus/deficit and debt

2.6 Greece: spread of GRD against the median ERM currency and against the euro

2.7 Greece: long-term interest rates

2.8 Greece: product markets

2.9 Greece: external indicators

3.1 Sweden: average inflation rate (HICP) and the reference value

3.2 Sweden: other inflation and cost indicators

3.3 Sweden: government surplus/deficit , debt and investment expenditure

3.4 Sweden: composition of budgetary consolidation between 1995 and 1999

3.5 Sweden: updated convergence programme projections for GDP growth, government surplus/deficit and debt

3.6 Sweden: long-term interest rates

3.7 Sweden: product markets

B.1 Inflation convergence-HICP

B.2 Development of average HICP inflation rates in EU Member States and the reference value

B.3 Evolution of the inflation reference value and the three best performers

B.4 Price deflator of private final consumption expenditure in EU Member States

B.5 Labour costs in EU Member States

B.6 Import prices in EU Member States

C.1 Government surplus/deficit in EU Member States

C.2 Government debt in EU Member States

C.3 Updated stability/convergence programme projections for government surplus/deficit in EU Member States

E.1 Long-term interest rates in EU Member States

Graphs

2.1 Greece: annual inflation rate (HICP)

2.2 Greece: comparison of average inflation rate (HICP) with reference value

2.3 Greece: inflation and wage trends

2.4 Greece: government deficit and debt

2.5 Greece: spread of GRD against the median ERM currency and against the euro

2.6 Long-term interest rates - Greece and EUR-11

2.7 Greece: comparison of average long-term interest rate with reference value

3.1 Sweden: annual inflation rate (HICP)

3.2 Sweden: comparison of average inflation rate (HICP) with reference value

3.3 Sweden: inflation and wage trends

3.4 Sweden: government surplus/deficit and debt

3.5 SEK/EUR exchange rate

3.6 Long-term interest rates - Sweden and EUR-11

3.7 Sweden: comparison of average long-term interest rate with reference value

1. INTRODUCTION AND MAIN FINDINGS

1.1. Introduction

The move into the third stage of economic and monetary union (EMU) and the introduction of the single currency, the euro, on 1 January 1999 was a major step forward in European economic integration. It followed several years of successful but often difficult adjustment efforts by the Member States during the second stage of EMU to achieve the high degree of sustainable convergence required for EMU participation and needed for the stability and success of the new currency. The decision [1] by the Council (of Heads of State or Government) on 3 May 1998 in Brussels on the 11 Member States ready to participate in the single currency from the beginning had, in accordance with the Treaty (Article 121(4); ex Article 109j(4)) [2], been prepared by the Ecofin Council on a recommendation from the Commission and was based on the two convergence reports made by the Commission [3] and the European Monetary Institute (EMI). [4] These reports, prepared in accordance with Article 121(1) of the Treaty (ex Article 109j(1)), examined in considerable detail whether the Member States satisfied the convergence criteria and met the legal requirements [5].

[1] OJ L 139, 11.5.1998, pp. 30-35.

[2] With the entry into force of the Amsterdam Treaty on 1 May 1999, the numbering of the articles of the Treaty was changed.

[3] Report on progress towards convergence and recommendation with a view to the transition to the third stage of economic and monetary union, COM(1998)1999 final, 25 March 1998.

[4] Convergence Report, European Monetary Institute, March 1998.

[5] Denmark and the United Kingdom were not the subject of a formal assessment because of their opt-out arrangements.

Those Member States which were assessed in 1998 as not fulfilling the necessary conditions for the adoption of the single currency are referred to as "Member States with a derogation". Two Member States fell into this category, Greece and Sweden, and they are the subject of this new report. Article 122(2) (ex Article 109k(2)) of the Treaty lays down provisions and procedures for re-examining the situation of Member States with a derogation (see Box: Article 122(2)). At least once every two years, or at the request of a Member State with a derogation, the Commission and the European Central Bank (ECB) are required to prepare new convergence reports on such Member States. Greece submitted a request on 9 March 2000 for its convergence situation to be re-examined. Two years have elapsed since the last reports were made by the Commission and EMI (25 March 1998) and since the Council decided on which Member States would initially adopt the euro (3 May 1998), and so Greece and Sweden are both due for re-examination. Article 122(2) additionally sets out the procedure by which the Council shall decide on the admission to the single currency of a Member State with a derogation now judged to fulfil the necessary conditions (see box); the steps of this procedure differ somewhat from those for the decision of 3 May 1998, which was based on Article 121(4).

BOX: Article 122(2) (ex Article 109k(2))

At least once every two years, or at the request of a Member State with a derogation, the Commission and the ECB shall report to the Council in accordance with the procedure laid down in Article 121(1). After consulting the European Parliament and after discussion in the Council, meeting in the composition of the Heads of State or Government, the Council shall, acting by a qualified majority on a proposal from the Commission, decide which Member States with a derogation fulfil the necessary conditions on the basis of the criteria set out in Article 121(1), and abrogate the derogations of the Member States concerned.

Two other Member States do not participate in the euro. Denmark and the United Kingdom negotiated opt-out arrangements before the adoption of the Maastricht Treaty (Protocols No 26 (ex 12) and No 25 (ex 11), respectively). Until these Member States indicate that they wish to participate in the third stage and join the single currency, they are not the subject of an assessment by the Council as to whether they fulfil the necessary conditions. Although the 1998 convergence report gave a considerable amount of information about the convergence situation in these two countries, the Commission made no judgement on whether they fulfilled the criteria and achieved a high degree of sustainable convergence. The present report by the Commission is limited to Greece and Sweden and does not deal with Denmark and the United Kingdom.

The reports to be prepared by the Commission and the ECB are, like the earlier reports, governed by Article 121(1) (see Box). This requires that the reports shall examine the compatibility of national legislation with the Treaty and the Statute of the European System of Central Banks (ESCB) and shall also examine the achievement of a high degree of sustainable convergence by reference to the fulfilment of the four convergence criteria dealing with price stability, the government budgetary position, exchange rate stability and the long-term interest rate as well as in the light of some additional factors. [6]

[6] Among the factors of which the reports also have to take account is "the development of the ECU". On 1 January 1999, every reference to the ECU was replaced by a reference to the euro at a rate of one euro to one ECU. Since there are no country-specific elements of this factor, it is not examined further in this report.

The four convergence criteria and the relevant periods over which they are to be respected are developed further in a Protocol annexed to the Treaty (see Box: Protocol (No 21 (ex 6)) on the convergence criteria). Detailed explanations of the way in which the criteria were being interpreted and applied were given in the 1998 convergence report.

BOX: Article 121(1) (ex Article 109j(1))

1. The Commission and the EMI shall report to the Council on the progress made in the fulfilment by the Member States of their obligations regarding the achievement of economic and monetary union. These reports shall include an examination of the compatibility between each Member State's national legislation, including the statutes of its national central bank, and Articles 108 and 109 (ex Articles 107 and 108) of this Treaty and the Statute of the ESCB. The reports shall also examine the achievement of a high degree of sustainable convergence by reference to the fulfilment by each Member State of the following criteria:

- the achievement of a high degree of price stability; this will be apparent from a rate of inflation which is close to that of, at most, the three best performing Member States in terms of price stability;

- the sustainability of the government financial position; this will be apparent from having achieved a government budgetary position without a deficit that is excessive as determined in accordance with Article 104(6);

- the observance of the normal fluctuation margins provided for by the exchange rate mechanism of the European Monetary System, for at least two years, without devaluing against the currency of any other Member State;

- the durability of convergence achieved by the Member State and of its participation in the exchange rate mechanism of the European Monetary System being reflected in the long term interest rate levels.

The four criteria mentioned in this paragraph and the relevant periods over which they are to be respected are developed further in a Protocol annexed to this Treaty. The reports of the Commission and the EMI shall also take account of the development of the ECU, the results of the integration of markets, the situation and development of the balances of payments on current account and an examination of the development of unit labour costs and other price indices.

One of the principles which has been followed in the preparation of this report is that Member States with a derogation not yet participating in the euro area are to be assessed as far as practicable in the same way as the first wave of euro area participants. This is the principle of equal treatment which implies that, wherever possible, the Treaty provisions on the convergence criteria should be interpreted and applied in the same way as in 1998. However, the assessment of some of the convergence criteria has to take into account the introduction of the euro. This is particularly the case for the exchange rate criterion, where, with the establishing of the euro in place of the ECU and the replacement of the original exchange rate mechanism by the new ERM II at the beginning of 1999, there is a changed frame of reference. Similarly, the existence of a single monetary policy in the new euro area may have implications for the assessment of inflation developments. Where there have been some necessary changes in approach from the 1998 report, these are spelled out in detail in this report and its annexes.

BOX: Protocol (No 21 (ex 6)) on the convergence criteria referred to in Article 121 of the Treaty establishing the European Community

THE HIGH CONTRACTING PARTIES,

DESIRING to lay down the details of the convergence criteria which shall guide the Community in taking decisions on the passage to the third stage of economic and monetary union, referred to in Article 121(1) of this Treaty.

HAVE AGREED upon the following provisions, which shall be annexed to the Treaty establishing the European Community.

Article 1

The criterion on price stability referred to in the first indent of Article 121(1) of this Treaty shall mean that a Member State has a price performance that is sustainable and an average rate of inflation, observed over a period of one year before the examination, that does not exceed by more than 1½ percentage points that of, at most, the three best performing Member States in terms of price stability. Inflation shall be measured by means of the consumer price index on a comparable basis, taking into account differences in national definitions.

Article 2

The criterion on the government budgetary position referred to in the second indent of Article 121(1) of this Treaty shall mean that at the time of the examination the Member State is not the subject of a Council decision under Article 104(6) of this Treaty that an excessive deficit exists.

Article 3

The criterion on participation in the exchange-rate mechanism of the European Monetary System referred to in the third indent of Article 121(1) of this Treaty shall mean that a Member State has respected the normal fluctuation margins provided for by the exchange-rate mechanism of the European Monetary System without severe tensions for at least the last two years before the examination. In particular, the Member State shall not have devalued its currency's bilateral central rate against any other Member State's currency on its own initiative for the same period.

Article 4

The criterion on the convergence of interest rates referred to in the fourth indent of Article 121(1) of this Treaty shall mean that, observed over a period of one year before the examination, a Member State has had an average nominal long-term interest rate that does not exceed by more than 2 percentage points that of, at most, the three best performing Member States in terms of price stability. Interest rates shall be measured on the basis of long-term government bonds or comparable securities, taking into account differences in national definitions.

Article 5

The statistical data to be used for the application of this Protocol shall be provided by the Commission.

Article 6

The Council shall, acting unanimously on a proposal from the Commission and after consulting the European Parliament, the EMI or the ECB as the case may be, and the Committee referred to in Article 114, adopt appropriate provisions to lay down the details of the convergence criteria referred to in Article 121 of this Treaty, which shall then replace this Protocol.

The introductory section of this chapter is followed by a summary of the main findings about convergence in Greece and Sweden. The report then continues with two main chapters, one on Greece and one on Sweden. Within each of these chapters, fulfilment of each the convergence criteria and other requirements is examined in the order that they appear in Article 121(1). Common material and other background information not specific to Greece and Sweden but relevant to the assessment are contained in a set of annexes which completes the report. Throughout this report there are frequent references to the 1998 Commission convergence report, and the current assessment of Greece and Sweden has to be seen in the context of the earlier report. The current report does not repeat in full the explanations of the 1998 report, preferring to focus on the specific situations in Greece and Sweden and to indicate any necessary changes in the way the criteria are applied and in methods for the provision of data.

1.2. Main findings

1.2.1. Greece

In the 1998 convergence report the Commission assessment was that Greece fulfilled none of the four convergence criteria. However, legislation in Greece was considered to be compatible with the Treaty and the ESCB Statute, despite an imperfection relating to the timing of the central bank's integration in the ESCB.

During the last two years Greece has achieved striking progress towards convergence and the assessment in this report is positive.

Legislation in Greece continues to be compatible with the Treaty and the ESCB Statute, and once further draft legislation is adopted by parliament even the imperfection identified earlier will have been removed.

The average inflation rate in Greece during the 12 months to March 2000 was 2.0%, below the reference value of 2.4%. The Greek inflation rate has been equal to or below the reference value since December 1999. The improvement in price stability is based on sound foundations, but there are risks associated with the reduction in short-term interest rates and movement of the exchange rate to its conversion rate in the approach to adoption of the euro; it will be necessary to maintain a tight budgetary policy stance and to secure continued wage moderation to prevent a possible resurgence of inflationary pressures. Greece fulfils the criterion on price stability.

The Council decision of 26 September 1994 on the existence of an excessive deficit in Greece was abrogated in 1999 (Council decision of 17 December 1999). On the latest available figures, the government deficit was brought down from 10.2% of GDP in 1995 to 1.6% in 1999, below the 3% reference value. The government debt ratio reached its highest level in 1996 at 111.3% of GDP and has since declined every year to 104.4% in 1999; the debt ratio is expected to continue declining and to fall below 100% of GDP in 2001. Greece fulfils the criterion on the government budgetary position.

The Greek drachma participated in the ERM from March 1998 until December 1998 and in the ERM II since January 1999, a total period which is longer than two years at the time of this examination, and has not experienced severe tensions during the period under review. The central rate of the Greek drachma was revalued against the euro in January 2000. During the review period the Greek drachma traded most of the time beyond a ±2.25% fluctuation range around its central rate (initially against the median currency in the ERM, and subsequently from January 1999 against the euro). However, the deviation of the Greek drachma was above its central rate. It reflected, inter alia, the higher interest rates in Greece and was not indicative of severe tensions in the examination period. Greece fulfils the exchange rate criterion.

The average long-term interest rate in Greece in the year to March 2000 was 6.4%, below the reference value of 7.2%. The narrowing of interest rate differentials in 1998 and 1999 brought the average rate in Greece down gradually and it fell below the reference value from October 1999 onwards. Greece fulfils the criterion on the convergence of interest rates.

In the light of its assessment on the fulfilment of the convergence criteria the Commission considers that Greece has achieved a high degree of sustainable convergence.

1.2.2. Sweden

In the 1998 convergence report the Commission assessment was that Sweden already fulfilled three of the convergence criteria (on price stability, the government budgetary position [7] and the convergence of interest rates) but that it did not fulfil the exchange rate criterion. Furthermore, legislation in Sweden was considered not compatible with the Treaty and the ESCB Statute.

[7] Subject to approval by the Council of the Commission recommendation, made at the same time as the adoption of the 1998 convergence report, for abrogation of the excessive deficit decision for Sweden.

In November 1998 Sweden adopted legislation amending the Constitution and Acts dealing with the Riksbank that was not significantly different from the drafts on which the examination in the 1998 convergence report was based. Legislation in this field has remained unchanged since then in Sweden. Consequently, the assessment on legal convergence in the 1998 report still stands, i.e. legislation in Sweden is not compatible with the Treaty and the ESCB Statute.

The average inflation rate in Sweden during the 12 months to March 2000 was 0.8%, below the reference value of 2.4%; indeed Sweden was one of the three best-performing Member States used for the calculation of this reference value. The Swedish inflation rate has been below the reference value throughout the period from December 1996. Sweden continues to fulfil the criterion on price stability.

The Council decision of 10 July 1995 on the existence of an excessive deficit in Sweden was abrogated in 1998 (Council decision of 1 May 1998). On the latest available figures, the government deficit was brought down from 7.9% of GDP in 1995 to 2.0% in 1997, and a government surplus of 1.9% was achieved in 1998 and 1999. The government debt ratio peaked in 1994 and has since declined every year to reach 65.5% of GDP in 1999; the debt ratio is expected to continue declining in 2000 and in future years. Sweden fulfils the criterion on the government budgetary position.

The Swedish krona has never participated in the ERM nor in the ERM II; in the two years under review the krona has fluctuated against the ERM currencies and the euro, reflecting, inter alia, the absence of an exchange rate target. Sweden does not fulfil the exchange rate criterion.

The average long-term interest rate in Sweden in the year to March 2000 was 5.4%, below the reference value of 7.2%. The reference value has been respected throughout the period since December 1996. Sweden continues to fulfil the criterion on the convergence of interest rates.

In the light of this assessment the Commission concludes that there should be no change in the status of Sweden as a Member State with a derogation.

2. GREECE

2.1. Compatibility of national legislation with the Treaty and the Statute of the European System of Central Banks [8]

[8] See Annex A for a brief description of the Treaty requirements in this area, in particular for central bank independence.

2.1.1. Assessment of compatibility in 1998

Legislation in order to comply with the Treaty and statute requirements was adopted by the Greek parliament in December 1997. The amendments in the new law were incorporated in the statute of the Bank of Greece by the shareholders of the Bank of Greece in December 1997. The amended statute was finally adopted by parliament in May 1998.

In its 1998 convergence report the Commission concluded that legislation in Greece was compatible with the requirements of the Treaty and the ESCB Statute.

However, an imperfection noted was that the amended statute on the central bank included some powers of the Bank of Greece which it would only have as long as Greece had not adopted the euro and the bank was not an integral part of the ESCB. This concerned two points: first, the power of the Bank of Greece to impose minimum reserves and penalties in the case of non-compliance, a provision which did not recognise the ECB's competence in this field; and second, the participation of the central bank in international monetary and economic organisations without a reference to the ECB's right of approval.

2.1.2. Legislative action taken since 1998 and overview of the legislation in force

The legislation on which the assessment was based in 1998 has remained in force since then.

On 25 April 2000 the General Meeting of the shareholders of the Bank of Greece agreed to a number of draft amendments to the Statute of the Bank of Greece. Most of them concern technical adaptations to align with the monetary policy framework of the ESCB. Among the draft amendments are the two points which were identified as an imperfection in the 1998 convergence report. The new statute will recognise the powers of the ECB in these matters as from the date when Greece is part of the euro area. The proposed amendments are expected to be adopted by parliament well before the end of 2000.

- Objectives

The primary objective of the Bank of Greece shall be to ensure price stability. Without prejudice to this primary objective, the Bank shall support the general economic policy of the government. As from when Greece adopts the single currency, the bank as part of the ESCB shall pursue the primary objective of maintaining price stability in accordance with the terms set out in Article 105(1) of the Treaty.

- Independence

The central decision making body is the Monetary Policy Council, which shall "define and implement monetary policy and decide on matters pertaining to the conduct of exchange rate policy, the operation of payment systems and the issue of banknotes". The Monetary Policy Council consists of the governor, two deputy governors and three other members. Their term of office is six years. The General Council, a second decision making body, assumes the other tasks conferred upon it by the statute of the central bank, except for matters falling within the duties of the ESCB for which the governor is responsible.

Article 5A of the statute of the Bank of Greece stipulates that "...neither the Bank of Greece nor any member of its decision-making bodies shall seek or take instructions from the government nor any other political authority shall seek to influence the decision making organs of the bank in performance of their duties".

- Integration in the ESCB and other legislation

As from when Greece adopts the single currency, the central bank shall act in accordance with the guidelines and instructions of the ECB as stipulated in Article 105(2) and (3) of the Treaty and Articles 3 and 14.3 of the ESCB Statute.

Article 12.17 of the law of December 1997 includes a catch-all clause in order to ensure compatibility. It reads as follows: "As from the date of adoption of the euro legal provision which contravenes primary or secondary EU legislation on the operation of the ESCB and/or of the ECB shall cease to be valid".

2.1.3. Assessment of compatibility

Legislation in Greece is compatible with the requirements of the Treaty and the ESCB Statute.

The imperfection noted in the convergence report of 1998 will have been removed if the draft amendments to the central bank statute are adopted in their present form.

2.2. Price stability [9]

[9] See Annex B for the calculation of the reference value, a discussion of other inflation standards, and a short description of improvements in the harmonized indices of consumer prices (HICP).

2.2.1. Inflation developments

- Situation in the 1998 convergence report

Greece did not fulfil the criterion on price stability in the 1998 convergence report. The average inflation rate (HICP) in Greece during the 12 months to January 1998 was 5.2%, well above the reference value of 2.7%. The Greek inflation rate had exceeded the reference value throughout the period from December 1996, although the differential had narrowed.

- Recent trends

>REFERENCE TO A GRAPHIC>

The downward trend in inflation in Greece, which had been evident on the basis of HICP data since 1996, has continued in the last two years. This trend was interrupted briefly in mid-1998 following the devaluation of the drachma on entering the ERM in March 1998. Since the final months of 1999 the annual inflation rate has increased, influenced mainly by the rise in oil prices as has also been the case in other Member States.

- Respect of the reference value

>REFERENCE TO A GRAPHIC>

The 12-month average inflation rate which is used for the convergence assessment has been steadily declining since the end of 1996, when it was as high as 8% (see Table 2.1). The differential from the reference value has progressively narrowed and fell to zero in December 1999. In the succeeding three months to March 2000, the average inflation rate in Greece has remained below the reference value (see Graph 2.2).

In March 2000, the reference value was 2.4%, calculated as the arithmetic average of the 12-month average inflation rates in the three best-performing Member States (France, Austria and Sweden) plus 1.5 percentage points. The corresponding average inflation rate for Greece was 2.0%, below the reference value (see Table 2.1).

- Performance relative to other inflation standards

The favourable inflation performance in Greece is confirmed by reference to other possible standards of price stability (see Annex B). For example, a reference value calculated on the basis of the three best-performing Member States in the euro area (i.e. excluding Sweden and including Germany) would be 2.5%, implying a slightly improved relative performance in Greece. The Greek average inflation rate is currently just at the upper limit of the ECB's definition of price stability and is 0.6 of a percentage point above the euro-area average. Indeed, it is worth noting that two of the euro-area Member States currently have average inflation rates above the rate in Greece.

2.2.2. Underlying factors and sustainability of inflation

Since early in the 1990s, stability-oriented economic policies, steadily pursuing nominal convergence, have played a central role in disinflation in Greece. From close to 20% in 1990, the rate of increase of consumer prices (deflator of private consumption) was more than halved by the mid-1990s. Until 1996, the anti-inflation strategy relied primarily on the so-called hard-drachma policy: the primary objective of monetary policy was the reduction in inflation using an intermediate target of maintaining a broadly stable average exchange-rate for the drachma against the ECU. At the same time budgetary consolidation was pursued with the assistance of lower debt servicing costs as well as measures to enhance fiscal revenues and combat tax evasion: the general government deficit, which stood at almost 16% of GDP in 1990, was reduced to 7.8% of GDP in 1996.

At the end of 1996, a new phase was initiated, when it appeared that a tighter and more balanced policy-mix was required. Using the exchange-rate as a nominal anchor had proved to be a successful strategy in reducing inflation in Greece; yet, with accelerating activity, labour costs pressures were rising entailing a large appreciation of the effective exchange rate in real terms and an ensuing loss of competitiveness. In the framework of the budget for 1997, the budgetary strategy was decisively oriented towards retrenchment measures to control current primary expenditure. By 1999 the government deficit was reduced to 1.6% of GDP and the primary surplus was increased significantly to 5.8% of GDP (see section 2.3). Monetary policy remained relatively tight after 1996. Inflation, as measured by the deflator of private consumption, was reduced by a further 5.7 percentage points to 2.5% in 1999.

- Wages and labour costs

Following a period of moderate increase in 1990-93, unit labour costs accelerated in the years up to 1997. Public sector wage increases well above ex-ante norms in that period spilled over into wage developments in the private sector, which was benefiting from accelerating activity and healthy profits.

>REFERENCE TO A GRAPHIC>

In March 1998, following the entry of the drachma into the ERM, more emphasis was placed on the role of incomes policy as a key component of the anti-inflation strategy. The restrictive stance of wage policy in the public sector was significantly strengthened in 1998 with the implementation of a wage increase norm of 2.5% for that year and increases related to expected inflation for the following years. In May 1998, a two-year national wage agreement for the private sector was signed; it provided for increases in minimum wages that would not compensate for productivity gains during 1998 and 1999 and included a compensation clause for inflation in excess of the announced targets for the end of each year. This agreement was viewed as an important step towards the establishment of a culture of wage moderation in Greece. Indeed, the agreement resulted in a deceleration in nominal compensation of employees and unit labour costs in 1998 and 1999, despite relatively buoyant activity and the activation of the compensation clause (see Table 2.2 and Graph 2.3).

- External influences on domestic prices

The policy of targeting a stable drachma exchange rate has helped to ensure that imported inflation was not a major source of inflation pressure in the last decade. More recently, the disinflation process has been assisted by low or even negative import price inflation due largely to falling commodity prices (excluding oil). Also, the inflationary impact of the drachma devaluation in March 1998 on prices proved to be only temporary, being fully absorbed in the course of the second half of 1998. Rising oil prices pushed consumer price inflation higher at the end of 1999, but tended to affect the price of goods rather than services (partly reflecting some deregulation of more sheltered sectors of the economy).

- Changes in indirect taxation

In the last quarter of 1998 and during 1999, the government introduced several cuts in indirect tax rates, which brought down the measured inflation rate through their mechanical impact on consumer prices. These measures were not intended to be removed at a later stage, implying a permanent impact on the level of prices. The overall direct impact of the indirect tax rate cuts, under the assumption of a full pass-through to consumer prices, is estimated on an annual basis at 0.7 percentage points in 1998 and 0.95 percentage points in 1999. Part of the impact of the measures adopted in 1999 is still influencing measured inflation rates. It is important to note that these measures, which were adopted for the most part towards the end of each year, also exerted an indirect favourable impact on wage developments, as they helped to reduce the inflation rate used for the calculation of the compensation clause mentioned above.

- Medium-term prospects

The considerable progress in disinflation achieved in recent years gives evidence that the foundations for price stability seem to be established in Greece.

Monetary conditions are likely to ease in the run-up to adopting the euro, as interest rates converge to euro-area levels, the exchange rate moves to the conversion rate and reserve requirements are lowered. While this easing of monetary conditions might be expected to give a stimulus to domestic demand, its impact should be diminished by the extent to which the movement in domestic interest rates and the exchange rate has been discounted by economic agents. Indeed, longer-term interest rates have already declined significantly in anticipation of euro adoption. Furthermore, as the income effect of lower interest rates is significant, the implied decline in household disposable income should dampen the stimulus to demand. Indeed, a large share of government debt is held by domestic households at variable rates. According to estimates by the Bank of Greece, interest rate convergence will reduce income of households by 3% of GDP. However, in the longer term, the impact of low interest rates, in combination with access to wide and deep euro-based financial markets, will probably have a demand-stimulating effect.

In this environment, other economic policies should contribute with determination to safeguarding price stability. The Council, in its opinion on the updated convergence programme of Greece, covering the period 1999-02 [10], urged the Greek government to strengthen the anti-inflationary stance of the policy instruments at its disposal, including budgetary and incomes policies. The 1999 updated convergence programme makes explicit the commitment of the Greek authorities to continue to pursue stability-oriented policies in the medium term in order to curb inflation further. The role of budgetary policy is enhanced: a tightening in the budgetary stance is projected in particular for 2001, when the effects of the monetary easing are more likely to emerge. The updated convergence programme also builds the anti-inflationary strategy on continuation of wage moderation in both public and private sectors; in the 2000 budget, the wage norm for the public sector has been maintained at the level of 2.3%. In the private sector, wage negotiations covering the next two years are still under way. An appropriate bi-annual national wage agreement in 2000 would help moderate unit labour costs. The government is clearly committed to promote such a settlement; the tax and benefit package decided in September 1999 allowing an increase in disposable income may facilitate moderate wage agreements.

[10] OJ C 60, 2.3.2000, p. 4.

Structural policies are expected to complement efforts towards maintaining price stability. The updated convergence programme restates the government's commitment towards accelerating the pace of reform with a view to enhancing competitive conditions and the operation of labour, goods and capital markets. Although structural reforms generally take time to produce tangible results, the liberalisation of the electricity and telecommunications markets in early 2001, following the implementation of Community Law, will affect short-term price developments and also reinforce the lasting character of price stability.

In all, some acceleration in consumer prices is to be expected in Greece in coming quarters; however, such a development is likely to be transient and should not reach an order of magnitude which might seriously undermine price stability in the medium term. The Commission currently forecasts consumer prices in Greece, as measured by the HICP, to accelerate from 2.1% in 1999 to 2.3% in 2000 and 2001.

The much improved inflation performance of Greece appears sustainable, provided budgetary policy remains tight and wage moderation is pursued. Greece has respected the reference value for inflation since December 1999. Greece fulfils the criterion on price stability.

2.3. Government budgetary position

2.3.1. Excessive deficit procedure

In the 1998 convergence report Greece did not fulfil the criterion on the government budgetary position. Greece was still the subject of a decision on the existence of an excessive deficit (Council decision of 26 September 1994). While there had been a very large reduction in the government deficit from 13.8% of GDP in 1993 to 4.0% in 1997, the deficit was still well above the 3% reference value. The government debt ratio was high and had reached a peak of 111.6% of GDP in 1996 before declining for one year to 108.7% in 1997.

In the last two years Greece has made further progress in reducing its government deficit and debt ratios. On the basis of the data available [11] in the autumn of 1999, the government deficit was estimated to have narrowed to 2.5% of GDP in 1998 and was expected to decline further to 1.9% of GDP in 1999. The government debt ratio declined further to 106.0% of GDP in 1998 (some 6 percentage points below its peak in 1996) and was expected to fall to 104.5% in 1999.

[11] Still on an ESA 79 basis, the same as used for the abrogation of the decisions on the existence of an excessive deficit in other Member States. These data took account of the information reported by the Greek authorities in September 1999 in accordance with Council Regulation (EC) No 3605/93. The compliance of these data with ESA rules and other Eurostat recommendations was examined by Eurostat, which validated the data reported by Greece.

On a recommendation from the Commission, the Council decided on 17 December 1999 to abrogate its former decision on the existence of an excessive deficit in Greece. [12] As Greece is no longer the subject of a Council decision under Article 104(6) of the Treaty that an excessive deficit exists, Greece now fulfils the criterion on the government budgetary position.

[12] OJ L 12, 18.1.2000, p. 24.

The remainder of this section reviews the current budgetary situation and prospects in Greece using the latest available data.

2.3.2. Current budgetary situation and prospects [13]

[13] As from March 2000, data on the basis of ESA 95 are being used for the purposes of the excessive deficit procedure and budgetary monitoring in general (see also Annex C). The changeover to ESA 95 implies an upward revision of the deficit ratio in Greece in 1998, estimated by Eurostat at 0.7 percentage point. The revision is mainly due to a clearer treatment in ESA 95 of general government transactions with public enterprises; some flows previously treated as financial transactions have been reclassified as capital transactions, with an influence on the deficit.

- Government deficit

In 1999, the general government deficit was reduced further to 1.6% of GDP. The stance of fiscal policy was tightened in order to contain inflation pressures stemming from the exchange rate adjustment of the drachma on entering ERM in March 1998 and additional retrenchment in primary expenditure was planned. However, final results showed that the improvement in the budgetary position in 1999 mostly resulted from buoyant budget revenues, while a decline in debt servicing costs partly offset higher than expected general government investment; primary current expenditure declined marginally as a share of GDP and the primary surplus reached 5.8% of GDP.

The Budget for 2000 targets a government deficit of 1.2% of GDP, in line with the updated convergence programme projections. The Commission services forecast the government deficit at 1.3% of GDP for 2000. The primary surplus is expected to remain at a high level. Lower debt servicing costs are expected to compensate less buoyant budget revenues.

>REFERENCE TO A GRAPHIC>

- Influence of the cycle

In the period from 1995 to 1999, the largest part of progress in reducing the general government deficit in Greece resulted from discretionary tightening and lower interest payments rather than from cyclical influences. According to calculations made by the Commission services, the output gap has been closing during the period to 1999 and is expected to become positive from 2000. When adjusted for the influence of the cycle, the deficit reduction was marginally smaller than the change in the actual balance. The positive impact of the business cycle on the government deficit in the period 1995-99 is less than 1 percentage point of GDP out of a total of 8.6 percentage points improvement in the actual balance.

- Size and composition of budgetary adjustment

The reduction in the government deficit from 1995 has largely benefited from the continuous decrease in interest payments as percentage of GDP: this was made possible by the progressive decline in interest rates, stemming from reduced inflation, active debt management operations and the beginning of reduction in the debt ratio. However, this effect was flanked by renewed stabilisation efforts which have provided substantial adjustment, particularly from 1996 onward.

Since 1995, fiscal consolidation has been building on corrective measures, some of them adopted in preceding years, and others at various stages and implemented with different promptness. First, measures were adopted in 1994 aiming at widening the tax base and combating tax evasion as well as improving the efficiency of tax assessment and collection. Then in 1996-97, an important package of measures, considered to be of structural nature, were adopted. Among these fiscal revenue enhancing measures aimed at broadening further the tax base, including a reduction or elimination of a number of tax breaks, increasing the corporate tax rate of financial institutions, introducing a real property tax and a 15% withholding tax on interest from government paper.

Other important measures included in the package were oriented at containing primary expenditure and at improving spending efficiency. A more transparent wage structure and a stricter hiring norm in the public sector were aimed at limiting current expenditure. At the same time, it was attempted to reduce the size of the public sector and rationalize spending in this area while limiting and controlling the growth of State guarantees.

Since 1997 the consolidation effort has fallen primarily on current expenditure retrenchment. In particular a strict wage policy in the public sector adopted in 1998 and based on expected inflation resulted in a virtual freeze in real terms until 2000 and possibly beyond.

An acceleration in budgetary consolidation resulted from these policies: the government primary surplus increased from 1.0% of GDP in 1995 to 5.8% in 1999. However, this improvement mainly proceeded from higher budget revenues, particularly buoyant in 1998 and 1999, and an increased tax burden. These were partly offset by greater investment expenditure only partly financed by faster absorption of EU structural funds resources. Wages and other current primary expenditures made almost no contribution to the overall deficit reduction in the period 1995-1999, showing some rigidity.

Fiscal consolidation in Greece builds primarily on reductions of the central government deficit. The other sub-sectors of the general government, and in particular the social security funds, show surpluses over 2% of GDP. This situation results from the 1990-1992 reform of the social security system which provided for a progressive increase in contributions and the rationalisation of expenditure.

Government investment expenditure in Greece, largely co-financed by EU Structural Funds financial resources, has been high and increasing in recent years. In terms of GDP, the share of government investment has increased from 3.2% in 1995 to 4.2% in 1999. Furthermore, since 1998, government investment has been greater than the government deficit.

- Medium-term prospects

The 1999 update of the Greek convergence programme restates the strategy defined by the previous programme aimed at achieving the completion of nominal convergence at the beginning of 2000 and to allow Greece to qualify for entry to the euro area on 1 January 2001. It also aims at achieving in the medium term a budgetary position meeting the requirements of the Stability and Growth Pact.

Starting from the favourable results achieved in 1999, budgetary consolidation is expected to reduce the government deficit to 0.2% of GDP in 2001; the government balance should then turn into a surplus of 0.2% of GDP in 2002.

During the period to 2002, reductions in interest payments are expected to be the main source of the improvement in the government balance; the primary surplus should remain broadly unchanged at a high level close to 6% of GDP.

- Debt trends

The government debt ratio in Greece peaked at the high level of 111.3% of GDP in 1996 and has since been declining. It reached 104.4% of GDP in 1999 and is forecast by the Commission services to decline to 103.7% in 2000. The updated convergence programme expects the debt ratio to fall by 6.2% of GDP in the period 1999-2002, moving below 100% of GDP from 2001 onwards. Thus a clear downward trend has been established, but the debt ratio remains very high and well above the 60% reference value.

The decline in the debt ratio has been and is projected to continue at a slower pace than might be expected given the now lower size of the deficit and the relatively strong growth of GDP. This reflects the impact, often called the "stock-flow adjustment", of a series of factors which can add to the stock of debt, such as changes in the value of foreign currency denominated debt, changes in net holdings of financial assets, and other statistical adjustment. This debt-increasing adjustment was very high in Greece in the early 1990s and still averaged 3% of GDP per year in the period 1994-98.

Since 1998, the reduction in the stock-flow adjustment is partly due to the greater use of privatisation proceeds to pay off public debt. Privatisation efforts have been planned since the beginning of the current stabilisation phase, but were delayed until 1998. The package of accompanying measures announced on the entry of the drachma into the ERM in March 1998 included a wide ranging privatisation plan, to be implemented by the end of 1999. Consequently, financial resources amounting to 2.5% of GDP, according to the estimates of the 1999 update of the Greek convergence programme, were allocated to pay off public debt in 1999.

The maturity structure of the debt has been progressively lengthened in the last few years, with a view to making the public finances less vulnerable to interest rate shocks. The average maturity of new debt issues increased from 1.6 years in 1994 over 6 years in 1999.

From the information presented in this section, it can be seen that the budgetary position of Greece, which is not at present the subject of an excessive deficit decision, has improved further in 1999 and is likely to do so again in 2000. There is no reason for the Commission to reopen the excessive deficit procedure for Greece.

2.4. Exchange rate stability

In the 1998 convergence report Greece was assessed not to fulfil the exchange rate criterion. While the Greek drachma had entered the ERM in March 1998, it had not participated in the mechanism during the two years under review (up to February 1998). The drachma had been relatively stable against the ERM currencies in the review period but had at times experienced tensions which were counteracted by increases in domestic interest rates and by foreign exchange intervention.

The assessment of the exchange rate criterion in this report is carried out on the basis of the analytical framework presented in Annex D, which takes into account the regime change which occurred with the introduction of the euro at the beginning of 1999. As in the 1998 convergence report, exchange rate stability in the period prior to the introduction of the euro is assessed by examining the behaviour of the drachma vis-à-vis the median ERM currency [14]. From 1 January 1999 onwards, the spread against the euro becomes the relevant indicator. Graph 2.5 and Table 2.6 summarise the behaviour of these spreads in the review period.

[14] The median currency within the ERM is defined as the currency closest to its central rate against the ECU at the exchange rate fixing on any given day.

>REFERENCE TO A GRAPHIC>

The two-year period relevant for the present assessment extends from April 1998 to March 2000. The Greek drachma entered the exchange rate mechanism (ERM) on 16 March 1998, hence just before the two years covered by the present assessment. Since 1 January 1999, the drachma has participated in the new exchange rate mechanism (ERM II).

On its entry into the ERM, the central rate of the drachma against the DEM was set some 12.4% below the average market rate in the ten preceding days. As the market exchange rate did not decline to the central rate, the de facto devaluation was somewhat smaller.

In the period that followed, the exchange rate of the drachma was supported by tight budgetary and monetary policies aiming at consolidating public finances and bringing down inflation, and thus accelerating macroeconomic convergence towards the euro area performance. In particular, the Bank of Greece kept nominal and real interest rates well above the rest of the prospective euro area in order to rein in price inflation [15]. Despite short-lived speculative pressures in April 1998 due to rumours of a possible drachma devaluation to coincide with the May decision on EMU participation, the shift in policy rapidly gained credibility in the financial markets, leading to large capital inflows which bid up the exchange rate. By early August 1998, the drachma was some 8% above its central rate against the median ERM currency (see Graph 2.5), and during the summer the Bank of Greece started to sterilise part of the capital inflows in order to prevent a further appreciation of the drachma.

[15] The average for the Greek three-month money market rate was 13.9% in 1998 and 10.3% in 1999, as compared to 3.9% and 3.0% in the euro area.

The appreciation was temporarily interrupted by the turbulence in international financial markets in the summer and autumn of 1998, triggered by the Russian debt moratorium in August. Pressures originated mainly in the Greek bond market, but were then rapidly transmitted to the money and exchange rate markets. Market interest rates rose sharply, in contrast with the prevailing trend in the rest of the EU, and the drachma fell by some 4% in the course of August 1998. The Bank of Greece did not raise interest rates, but it intervened in the foreign exchange market in order to avoid too wide fluctuations of the currency. [16]

[16] In the aftermath of the Russian crisis in August 1998, Greek market interest rates rose considerably. The three-month interest rate went up by 5-6 percentage points in the course of the month, temporarily approaching 18.5% at the end of August. Correspondingly, the spread on three-month rates over Germany widened to nearly 15 percentage points at the end of August. Such increases were however short-lived, and the spread soon came down to the pre-crisis level of around 9.5 percentage points. The rise in long-term yields was smaller, about 2 percentage points, but more persistent. The long-term spread over Germany widened in August from 2.9 to 4.2 percentage points, it peaked at 4.7 percentage points in October 1998, and it came back to pre-crisis levels only at the beginning of 1999.

By the beginning of 1999, the Greek drachma had fully recovered and it started the year some 7% above its ERM II central rate against the euro of GRD/EUR 353.109. The appreciation continued in January 1999, taking the drachma up to a maximum of 9.2% above the central rate at the end of that month. From then, the drachma started moving towards its ERM II central rate, in line with a narrowing of the short-term interest rate differential with the euro area. On 14 January 2000, with the market exchange rate still more than 6% above the central rate, the Greek authorities requested a revaluation of the central parity. The new parity was set at GRD/EUR 340.75 as of 17 January 2000, implying a 3.6% revaluation of the drachma vis-à-vis its central rate.

Following the revaluation and favourable inflation data, the Bank of Greece lowered the key 14-day deposit rate by 100 basis points to 9.75% on 28 January 2000 and then by a further 50 basis points on 8 March 2000. The three-month spread over the euro area fell correspondingly from 6.1 percentage points at the start of 2000 to 5 percentage points in mid-March, supporting the rapprochement of the drachma to its new ERM II central rate. At the end of the review period the drachma was trading 1.8% above the central rate.

During the review period, the average deviation of the drachma market rate from its central rate against the median currency and then the euro was 6.4%. The exchange rate of the drachma (against the DEM until 31.12.1998 and against the euro thereafter) has generally experienced relatively low and diminishing volatility during the review period, with the exception of the period of financial market turmoil in the latter part of 1998.

Summarising the evidence presented above:

* the drachma has participated in the ERM and subsequently in the ERM II for a total period of more than two years;

* the drachma traded above its ERM central parity against the median currency in the whole period April-December 1998. Since the introduction of the euro and the ERM II, the drachma has always remained above its ERM II central rate against the euro;

* during the review period, there has been no devaluation of the drachma's central parity in either the ERM or the ERM II;

* although the drachma underwent some tensions at the height of the Summer 1998 turbulence in international financial markets, such pressures were only temporary and did not significantly alter the position of the Greek currency in the ERM.

Greece therefore fulfils the exchange rate criterion.

2.5. Long-term interest rate [17]

[17] See Annex E for information on the calculation of the reference value for the long-term interest rate and on the data employed.

In the 1998 convergence report Greece did not fulfil the criterion on the convergence of interest rates. The average long-term interest rate in Greece in the year to January 1998 was 9.8%, clearly above the reference value of 7.8%.

>REFERENCE TO A GRAPHIC>

During the 1990s, the Bank of Greece has been pursuing a disinflationary strategy based on a rigorous monetary policy stance and a strong exchange rate. After the entry into the ERM in March 1998 short-term interest rates were kept relatively high, both in nominal and real terms, and the drachma stood above its bilateral central rates in the ERM grid. As the disinflationary process gained credibility and fiscal consolidation progressed, long-term interest rates started declining towards the levels prevailing in the euro area countries, although the spread remained sizeable (see Graph 2.6).

The period of turbulence in international financial markets in Summer/Autumn 1998, temporarily interrupted convergence. After the Russian debt moratorium in August 1998, tensions emerged in the Greek bond market, leading to a sharp widening of the yield spread between Greek and the average for euro area ten-year benchmark bonds from 2.9 percentage points at the end of July to 4.2 percentage points at the end of August. The spread peaked in October, when it reached 4.7 percentage points. By January 1999, the spread was back to pre-crisis levels, and since then, long-term interest rate convergence has progressed. In March 2000, the ten-year spread between Greece and the euro area was down to around 80 basis points.

>REFERENCE TO A GRAPHIC>

Graph 2.7 shows the development in the twelve-month average long-term interest rate in Greece, which is relevant for the assessment of the Treaty criterion, in comparison with the evolution of the reference value (the upper bound of the corridor in the graph). Having declined steadily since 1996, the average rate in Greece reached the reference value in October 1999 and has since been below it. In March 2000, the latest date for which data are available, the reference value, given by the average of long-term interest rates in France, Austria and Sweden plus 2 percentage points, stood at 7.2% (see Annex E). The twelve-month average of the yield on ten-year Greek benchmark bonds stood at 6.4%, hence below the reference value. Therefore, Greece fulfils the criterion on long-term interest rate convergence.

2.6. Additional factors

2.6.1. Results of the integration of markets

- Development in product markets

Trade and foreign direct investment (FDI) statistics show that the Greek economy is not as integrated in European product markets as other Member States. The Greek intra-EU trade to GDP ratio [18] of 10% for the period 1996-98 is the lowest in the EU, which may be attributed in part to Greece's peripheral location. Moreover, the trade to GDP ratio has declined slightly since the early 1990s (from 11.3 % in 1990 to 9.7% in 1997). This decline may well be explained by the political problems in the neighbouring former Yugoslavia. In 1998, however, the ratio rose again to 10.4%, which is indicative of the recent economic reforms undertaken by the Greek government. FDI inflows are relatively low as well. The intra-EU FDI to GDP ratio for the period 1996-98 equalled 0.7%, which was well below the EU average of 1.7%. It is also a reflection of the fact that cross-border merger and acquisition activity involving Greek companies is still of relatively little importance. However, again an increasing trend may be emerging. While the intra-EU FDI to GDP ratio was consistently below 0.5% over the period 1993-96, it rose to 0.7 % in 1997 and 0.9% in 1998.

[18] Defined as: (intra-EU imports + intra-EU exports)/(2*GDP).

The relatively low level of integration of the Greek economy in European product markets may be associated with a lack of competitive pressures on Greek product markets and result in relatively high price levels. Aggregate pre-tax price levels in Greece are higher than would be expected for a country with a comparatively low standard of living. While manufactured goods prices in 1997 were just below the EU average, an indication that the Greek manufacturing sector was nevertheless exposed to competition from the other EU countries, price levels [19] for some equipment goods such as office machinery and computers were relatively high. Price levels in the services and construction sectors were well below average EU levels, as expected. The price of purchased transport services, an extreme case, was less than half the EU average.

[19] Greek price levels are not administratively controlled except for refined oil products and pharmaceuticals.

Greece has also been relatively late in launching the economic reform process. As a result, there are a number of obstacles that still need to be removed to allow the full integration of Greece in European product markets:

- Greece has made slow progress in transposing Single Market legislation and the November 1999 non-transposition rate was higher than in any other EU Member State.

- A liberalisation process is taking place opening up the network industries to competition, albeit at a slower pace than in other Member States. However, within the domestic market, publicly owned companies continue to play an important role.

However, recently, a number of reforms have been initiated which will improve the situation:

- A relatively high share of public procurement calls for tenders are published in the Official Journal.

- The privatisation of public enterprises has progressed, mainly within telecommunications, petrol refineries, water, supporting services to air transport and banking. Also, public enterprises are increasingly exposed to market forces and the provision of services has been gradually rationalised.

- Developments in financial markets

The Greek government securities market has developed rapidly in recent years. A system of primary dealers was introduced in 1997, which includes domestic as well as foreign banks. An electronic secondary market trading system was introduced in 1998 which from January 1999 also includes the primary market (government auctions). In February 2000, Greece launched its first bond since the establishment of a new debt management agency; this was also Greece's first bond issue in euro. This marked a new development in strategy, using targeted international bonds to attract more international investors into its government bond markets ahead of an eventual adoption of the euro. The euro-denominated bond issue is Greece's first "parallel" bond, indicating that it will become fungible with the drachma benchmark bond, carrying identical terms and conditions, issued in March 2000. In the event that Greece adopts the euro, Greece intends to re-denominate its domestic government bonds into euro, thereby creating large and liquid bond issues.

Important structural reforms have been implemented in the stock exchange in the last year. In particular, the terms and conditions for listing shares have been made more flexible and harmonized with those of other EU countries. Other reforms include an extension of trading hours, the introduction of an electronically integrated trading system, the integration of the derivatives exchange and the planned creation of a new stock market for high growth SMEs. In 1999 twice as many companies (156) raised funds through the Athens stock exchange as the year before, with the total amount raised rising four-fold to over GRD 3 300 billion.

The last two years have seen considerable structural adjustments in the banking sector. A number of state-owned banks have been privatised, and further privatisation is planned for this year. In addition, the banking sector is undergoing continuous rationalisation as a result of mergers and acquisitions. Profits in the Hellenic banking sector have risen in the last year, partly responding to the strong growth of the sector during the latter part of the 1990s.

2.6.2. Balance of payments on current account

In 1999, the Greek current account balance showed a deficit of 1.4% of GDP. [20] The trade balance was in a deficit of 13.1% of GDP while the services balance was in surplus to the tune of 5.7% of GDP. Net factor income and transfers totalled 6% of GDP. Net transfers include Structural and Cohesion Funds from the EU of almost 3% of GDP.

[20] The data in this section are based on national accounts statistics according to ESA 95.

The current account performance deteriorated in 1996 and 1997 due to real exchange rate appreciation and comparatively strong domestic demand growth relative to trading partners. Following the devaluation of the Greek drachma and ERM entry in March 1998, export performance strengthened. Despite economic growth above the EU average, the current account balance improved in 1998 and 1999. With continued strong domestic demand growth and the effects of the devaluation having come through, the deficit on current balance is expected to widen to around 2% of GDP this year.

Gross fixed capital formation in Greece has risen markedly to reach some 23% of GDP in 1999 (this compares with an average investment rate in the euro area of about 21% of GDP). Greek government saving turned positive in 1999 and private sector saving amounted to 19% of GDP in 1999, just below the EUR-11 average. Based on these indicators, the Greek current account deficit now seems related to strong investment rather than to a deficiency of national saving (see Table 2.9).

The large deficit on the trade balance, the substantial surplus on the services balance and the positive net balance of income and transfers are all long-standing features of the Greek current account. In a longer-term perspective, as the Greek economy approaches the EU average in terms of productivity and living standards, there is likely to be a decline of EU transfers from the level secured in part under the Third Community Support Framework 2000-06. Meanwhile, the structure of Greece's export earnings differs from other euro economies due to heavy reliance on tourism and very modest manufacturing exports. Manufacturing exports to the euro area correspond to little more than 3% of GDP. Better Greek export performance over the medium term requires improvements on the supply-side in terms of labour productivity and the range of products produced.

2.6.3. Unit labour costs and other price indices

The examination of the development of unit labour and other price indices, required by Article 121(1), is included in the section on price stability (see 2.1.2).

3. SWEDEN

3.1. Compatibility of national legislation with the Treaty and the Statute of the European System of Central Banks [21]

[21] See Annex A for a brief description of the Treaty requirements in this area, in particular for central bank independence.

3.1.1. Assessment of compatibility in 1998

In its 1998 convergence report the Commission concluded that legislation in Sweden was not compatible with the requirements of the Treaty and the ESCB Statute. The following incompatibilities were noted:

- the legislation did not provide for independence of the central bank;

- legislation under preparation included an amendment of the Constitution, which for constitutional reasons had to be finally approved by the new parliament after the general elections of September 1998. Therefore it was clear that the central bank of Sweden would not be independent at the date of the establishment of the ESCB;

- the draft legislation did not ensure full integration of the central bank in the ESCB; in particular, the provisions regarding the Riksbank's competence in the monetary policy area did not recognise the ESCB's competence in this field.

Apart from these incompatibilities, the Commission also noted two imperfections with respect to the draft legislation in the spring of 1998:

- the draft Constitution retained the exclusive right to issue banknotes to the Riksbank; this provision failed to recognise the ESCB's competence in this field as laid down in Article 106 of the Treaty;

- the prohibition imposed on the Riksbank by draft legislation to seek or take instructions only covered monetary policy and did not extend to all ESCB related tasks as specified in Article 105 of the Treaty.

3.1.2. Legislative action taken since 1998 and overview of the legislation in force

The government had put forward to parliament in November 1997 a proposal to amend the Constitution, the Riksdag Act and the Riksbank Act. The amendment of the Constitution required approval by two consecutive parliaments. The parliament adopted the amendments to the Constitution for the first time in March 1998 and for the second time in November 1998, after the general elections On this occasion, parliament also adopted the amended Riksdag Act, Riksbank Act and a new Act on exchange rate policy by which the competence for exchange rate policy was transferred from the central bank to the government. All new legislation entered into force on 1 January 1999.

The legislation adopted in November 1998 was not significantly different from the drafts on which the examination in the 1998 convergence report was based.

- Objectives

The objectives of the Riksbank are defined as follows:

"The objective of the Riksbank's operations shall be to maintain price stability. In addition, the Riksbank shall promote a safe and efficient payment system." (Riksbank Act, Chapter 1, Article 2).

- Independence

Pursuant to the amended Riksbank Act, an Executive Board was established with the task of defining monetary policy. The Executive Board comprises six members including the governor who are all appointed for a six-years period by the General Council. The latter comprises eleven members appointed by the Riksdag. The General Council has a supervisory function with no monetary policy competence.

The reform of 1998 included further essential elements in order to make the central bank independent. In particular, public authorities are prohibited from giving the central bank instructions on monetary policy through a provision in the amended Constitution. The Riksbank Act in turn prohibits members of the Executive Board of the Riksbank from seeking or taking instructions in monetary policy matters. Prior to the reform of 1998, the Riksbank was obliged to consult the Government before taking policy decisions; this has been changed in the amended Riksbank Act to a requirement to inform the government before taking important monetary policy decisions.

- Integration in the ESCB and other legislation

The competence for monetary policy is attributed to the Riksbank without reference to the competence of the ESCB in this field. The right of the bank to impose minimum reserves for monetary purposes is also defined without such a reference.

3.1.3. Assessment of compatibility

Legislation in Sweden is not compatible with the requirements of the Treaty and the ESBC Statute. The following incompatibility is to be noted:

* the legislation applicable since 1 January 1999 does not ensure full integration of the Riksbank in the ESCB; in particular, the provisions regarding the Riksbank's competence in the monetary policy field do not recognise the ESCB's competence in this field.

Furthermore the following imperfections remain:

* the definition of the objectives of the Riksbank includes price stability but omits a reference to supporting the general economic policies in the Community, an objective which according to Article 105(1) of the Treaty the ESCB shall pursue, without prejudice to the objective of price stability;

* the Riksbank's exclusive right to issue banknotes does not recognise the ESCB's competence in this field;

* the prohibition of public authorities to give and of the members of the Riksbank's decision-making bodies to seek or take instructions in the Riksbank Act itself only covers monetary policy issues. The application of this rule to all ESCB-related tasks is only mentioned in the Explanatory Memorandum to the Act.

3.2. Price stability [22]

[22] See Annex B for the calculation of the reference value, a discussion of other inflation standards, and a short description of improvements in the harmonized indices of consumer prices (HICP).

3.2.1. Inflation developments

- Situation in 1998 convergence report

>REFERENCE TO A GRAPHIC>

Sweden already fulfilled the criterion on price stability when assessed in the 1998 convergence report. The average inflation rate in Sweden during the 12 months to January 1998 was 1.9%, below the reference value of 2.7%. The Swedish inflation rate had been below the reference value throughout the period from December 1996, and for the months from December 1996 to September 1997 Sweden had been one of the three best-performing Member States in terms of price stability.

- Recent trends

After rising during 1997, the annual inflation rate in Sweden (as measured by the change in the monthly HICP index from 12 months earlier (see Graph 3.1)) fell back during 1998 and reached zero in the autumn of 1998. In 1999, the annual rate remained subdued during the first half of the year, but accelerated towards the end of the year and reached 1.4% in March 2000.

- Respect of the reference value

The 12-month average inflation rate which is used for the convergence assessment has fluctuated in Sweden but has continued to be below the reference value. After rising to just over 2% in the first half of 1998, the average inflation rate declined sharply to a very low level. Since February 1999 Sweden has been one of the three best-performing Member States in terms of price stability (see Graph 3.2).

>REFERENCE TO A GRAPHIC>

In March 2000 the three best-performing Member States in terms of price stability were France, Austria and Sweden itself, and the simple arithmetic average of the 12-month average inflation rates in these three countries was 0.9%. The reference value, calculated in the same way as in the 1998 convergence report, was 2.4%. The average inflation rate in Sweden in March 2000 was 0.8%, well below the reference value (see Table 3.1 and Graph 3.2).

- Performance relative to other inflation standards

The good inflation performance in Sweden is also apparent when comparison is made with other possible standards for inflation (see Annex Table B.1). Excluding Sweden from the calculation of the reference value and using the three best performers in the euro area gives a slightly higher value of 2.5%. Furthermore, in March 2000 Sweden was the best performing Member State in the EU as a whole. Sweden's average inflation rate was well below the 2% upper limit of the ECB's definition of price stability.

3.2.2. Underlying factors and sustainability of inflation performance

>REFERENCE TO A GRAPHIC>

A successful reduction in the inflation rate was achieved in Sweden during the first half of the 1990s, and since 1995 inflation has been maintained at a low level. The disinflation process was helped by the very slow increase in unit labour costs, especially in 1993 and 1994 (see Graph 3.3). Since then compensation per employee has increased at a moderate pace, and coupled with favourable productivity growth, the increase in unit labour costs has remained low (see also Table 3.2). As in other Member States there has been some upward pressure on inflation in recent months, coming from the impact of higher oil prices on import prices, but in general in recent years the cost of imports has also been a restraining factor for inflation.

Since 1992 Sweden has had an explicit inflation target for monetary policy and a flexible exchange rate regime. Commitment to price stability as the objective of monetary policy was further underlined by the new legislation on the status of the Riksbank that came into force in 1999. The Riksbank's inflation target is 2%, based on the national consumer price index (CPI), with a tolerance interval of ±1 percentage point. Since October 1999 the Riksbank has specified that it bases its current monetary policy on the assessment of underlying inflation as measured by UND1X [23].

[23] UND1X is defined as the CPI excluding interest expenditure and direct effects of altered indirect taxes and subsidies.

The underlying inflation rate has been below the Riksbank's inflation target for a number of years, but within the tolerance interval. Underlying inflation has been 1.5% on average since 1996. Medium-term inflation expectations are firmly anchored around the inflation target and thereby close to 2%. In March 2000, the UND1X stood at 1.6% year-on-year.

It should be noted that the framework for achieving price stability in Sweden is different from that in the euro area. Whereas the Governing Council of the ECB has defined price stability as the year-on-year increase in the HICP for the euro area of below 2% in the medium term, the Swedish definition, by contrast, is symmetric around 2% with a tolerance interval of ±1% point. The target indicators are also different from the HICP, being either the CPI or, as at present, the UND1X. Although so far this has not been a source of difficulty, the issue will need to be addressed by the Swedish authorities at some stage, particularly if the existence of the two targets were to give rise to an appreciable inflation differential between Sweden and the euro area.

Sweden has respected the reference value for inflation throughout the period from December 1996 and since February 1999 has been one of the three best-performing Member States in term of price stability. Sweden therefore continues to fulfil the criterion on price stability.

3.3. Government budgetary position

3.3.1. Excessive deficit procedure

In the 1998 convergence report the Commission considered that the excessive deficit situation in Sweden had been corrected. On the data then available, there had been a very large and continuous reduction in the government deficit from 12.2% of GDP in 1993 to 0.8% in 1997, below the reference value, and a surplus was expected in 1998. The government debt ratio had peaked in 1994 at 79.0% of GDP and had since declined every year to reach 76.6% in 1997 and was expected to continue declining in 1998 and in future years. In the light of this assessment and in parallel with the adoption of the report, the Commission made a recommendation to the Council that the decision on the existence of an excessive deficit in Sweden (Council decision of 10 July 1995) should be abrogated.

Acting on this recommendation the Council adopted on 1 May 1998 a decision abrogating the decision on the existence of an excessive deficit in Sweden. [24] As Sweden is not the subject of a Council decision under Article 104(6) of the Treaty that an excessive deficit exists, Sweden fulfils the criterion on the government budgetary position.

[24] OJ L 139, 11.5.1998, p. 19.

The remainder of this section reviews the current budgetary situation and prospects in Sweden using the latest available data.

3.3.2. Current budgetary position and prospects [25]

[25] As from March 2000, data on the basis of ESA 95 are being used for the purposes of the excessive deficit procedure and budgetary monitoring in general (see also Annex C).

There has been a significant improvement in the public finances in Sweden in recent years with a government deficit of 11.9% of GDP in 1993 turned into a surplus of 1.9% in 1998 [26] and 1999. A larger surplus of 2.4% of GDP is forecast by the Commission services for 2000 (see Table 3.3 and Graph 3.4). In the years between 1995 and 1998 almost all of the improvement was structural in nature, and the influence of the cycle has been relatively limited. However, the underlying improvement came to an end in 1999 and, in a context of strong economic growth, there are indications of a relaxation in the fiscal stance in 2000, with some narrowing of the cyclically adjusted surplus.

[26] A one-off sale of pension fund real estate, amounting to 0.9% of GDP, increased the surplus in 1998.

Between 1995 and 1999, the cyclically adjusted government balance improved by 9.4% of GDP (see Table 3.4). A small part of the improvement was due to declining interest payments, but the largest share was due to an increase in the primary balance of 7.8% of GDP; more than half of this resulted from a reduction in the primary expenditure ratio (including a fall in fixed capital spending), but the rise in the revenue ratio up to 1998 also made a strong contribution. The substantial budgetary consolidation in recent years has thus been primarily based on lower expenditure but has also relied significantly on increases in tax pressure.

>REFERENCE TO A GRAPHIC>

The updated convergence programme for Sweden was adopted by the government in November 1999 and assessed by the Council in January 2000 [27]. It covers the years to 2002 and foresees government surpluses of about 2% of GDP in each year, in respect of Stability and Growth Pact requirements (see Table 3.5). The budgetary strategy continues to be based on strict control of government spending and a further decline in the expenditure/GDP ratio. At the same time cuts in taxes are being introduced, reversing the trend in the years up to 1998. Budgetary discipline is reinforced by nominal ceilings approved by the parliament for central government expenditure for three years ahead, and by a legal obligation on local authorities from 2000 onwards to balance their budgets. Furthermore, the recently introduced pension reform - which involves a higher degree of individual funding - will enable the pension system to cope better, in the medium term, with business cycle fluctuations and, in the long term, with the ageing of the population.

[27] OJ C 60, 2.3.2000, p. 5.

After rising strongly in the early 1990s the government debt ratio in Sweden reached a peak in 1994 and has since been declining. There was a sharp fall in the ratio in 1999, to 65.5% of GDP and a further decline in 2000 is forecast by the Commission services to 61.3%, close to but still slightly above the reference value (see Table 3.3 and Graph 3.4). With the government balance remaining in surplus, the updated convergence programme projects a further rapid decline in the debt ratio to 52.0% by 2002.

From the information presented in this section, it can be seen that the budgetary position of Sweden, which is not at present the subject of an excessive deficit decision, has remained strong in 1999 and is likely to improve further in 2000. There is no reason for the Commission to reopen the excessive deficit procedure for Sweden.

3.4. Exchange rate stability

In the 1998 convergence report Sweden was assessed not to fulfil the exchange rate criterion. The Swedish krona had not participated in the ERM, and in the two years under review (up to February 1998) the krona had fluctuated against the ERM currencies, reflecting among other factors the absence of an exchange rate target.

Sweden has continued not to participate in the ERM during the current review period (two years up to March 2000). The monetary framework of explicit inflation targeting implies that the Swedish krona is left floating almost freely in the exchange rate market. The absence of an explicit exchange rate commitment means that the stability of the exchange rate of the krona cannot be assessed in the same way as for the Member States participating in the ERM (i.e. on the basis of deviations from that commitment).

>REFERENCE TO A GRAPHIC>

Between Spring 1996 and the outbreak of the international financial crisis in late Summer 1998, the exchange rate of the krona did not show any clear trend. The krona was severely hit by the international financial crisis. Against the DEM, it depreciated from SEK 4.50 at the end of July 1998 to about SEK 4.90 in October, a level not seen since autumn 1995. Since the beginning of 1999, the krona has largely been on an appreciating trend against the euro, rising by more than 12% from SEK 9.47 to SEK 8.28 against the euro at the end of March 2000 (see Graph 3.5).

The volatility of the krona exchange rate has diminished in the second half of the 1990s, as credible and sustainable fiscal and monetary policy provided the preconditions necessary for a stable exchange rate in the medium run. The krona has been more stable and less volatile against EMU currencies and the euro than against non-EMU currencies. Volatility was generally limited during the review period, although there was a steep but temporary increase during the Summer/Autumn 1998 financial crisis. Nevertheless, the absence of an exchange rate anchor was reflected in relatively higher volatility than in the exchange rate of the two ERM II currencies.

The Swedish krona has not participated in the ERM and subsequently in the ERM II during the review period and it has fluctuated against the ERM currencies and the euro, also reflecting the absence of an exchange rate commitment. Hence, as was the case at the time of the 1998 assessment of convergence, Sweden does not fulfil the exchange rate criterion.

3.5. Long-term interest rate [28]

[28] See Annex E for information on the calculation of the reference value and on the data employed.

>REFERENCE TO A GRAPHIC>

Sweden already fulfilled the criterion on the convergence of interest rates in the 1998 convergence report. The average long-term interest rate in Sweden in the year to January 1998 was 6.5%, below the reference value of 7.8%.

Sweden's good inflation record and impressive consolidation of public finances in the past years have been reflected in declining long-term interest rates (see Graph 3.6). More recently, Swedish long-term interest rates (the yield on the 10-year benchmark bond) have followed the upward trend in international bond markets, increasing by some 150-200 basis points since the trough reached in January 1999 at about 4%. Over the last year, the spread over euro area long-term yields has fluctuated between 30 and 70 basis points. The main contributing factors to the variation in the spread may have been changes in the perceived probability of Sweden joining EMU within the near-term future and expectations on the cyclical position of Sweden relative to the euro-area economy. However, on 9 March 2000, the spread relative to Germany fell to a low of 12 basis points as the Swedish National Debt Office announced a buy back of government bonds due to substantially higher public sector net savings this year partly resulting from privatisation revenues.

Sweden has continued to have a twelve-month average interest rate below the reference value. In fact, Sweden has had long-term interest rates below the reference value ever since December 1996, the first date for which this comparison can be provided (see Graph 3.7 and Table 3.6).

>REFERENCE TO A GRAPHIC>

In March 2000, the latest month for which data are available, the reference value, given by the average of long-term interest rates in France, Austria and Sweden plus two percentage points, stood at 7.2% (cf. Annex E). The twelve-month average of the yield on ten-year Swedish benchmark bonds stood at 5.4%, hence below the reference value. Therefore, Sweden continues to fulfil the criterion on long-term interest rate convergence.

3.6. Additional factors

3.6.1. Results of the integration of markets

- Development of product markets

Sweden is increasingly integrated in the EU economy. Sweden's intra-EU trade to GDP ratio for the period 1996-98 of 18.2% was comparable to that of Austria and Finland, which entered the EU at the same time. Moreover, the trade ratio has been rising, although it remained below the 1998 average for the smaller EU Member States. The intra-EU FDI to GDP ratio for the period 1996-98 was above the average for the smaller Member States. Sweden's share in EU cross-border mergers and acquisitions (M&As) is well above what one would expect for a country of its size.

Over the recent years, a number of structural policy reforms have been introduced facilitating the rising integration of Sweden in EU product markets:

- Sweden has an excellent record in transposing and applying Single Market legislation.

- Measures have been taken to open up public procurement.

- A lot of effort has been made to deregulate the network industries. Telecommunications prices are amongst the lowest in the EU. Nevertheless, mergers and acquisitions combined with only limited entry have created a highly concentrated supply structure in some of these industries.

- Sweden is well integrated in the European research area and it is one of the forerunners in the developing information society. R&D spending is the highest in the EU and more than 70% of this was conducted by businesses.

In spite of the significant progress made in opening up the product markets to competition, both from within and from outside the country, price levels in Sweden remain high compared to those in most other Member States. The high relative price levels may be attributed in part to the high levels of indirect taxation in Sweden, but more importantly to a lack of competition in certain sectors such as construction, pharmaceuticals, food retailing and some of the network industries. The market domination of few retailers is among the highest in the EU.

- Developments in financial markets

The Swedish money and bond market has been well established for almost two decades with high liquidity and active participation of both domestic and foreign financial intermediaries. The major part of government debt, around 70%, is in SEK. Of the remaining 30%, approximately half is in euro with about one fifth each in US Dollar and Japanese Yen.

In August 1999, the Oslo Stock Exchange decided to join Norex, the common market for shares established through collaboration between the OM Stockholm Exchange and the Copenhagen Stock Exchange. A shared trading platform and harmonized trading rules should help to promote a more efficient stock market with greater liquidity.

Structural changes in the financial system are well underway, driven by increased competition, globalisation and the introduction of the euro. In the past years, several mergers and acquisitions have taken place, both within Sweden and across borders. The Swedish banking system is characterised by a high degree of concentration, but competition appears to remain strong. Furthermore, Swedish banks are at the forefront of the development of internet banking.

3.6.2. Balance of payments on current account

In the 1970s and 1980s Sweden emerged as a significant international net debtor as a combination of low private and public saving and recurring competitiveness problems created substantial current account deficits. Sweden's position changed after the deep recession at the beginning of the 1990s. The large depreciation in November 1992 and continued depressed private domestic demand in combination with consolidation of public finances generated large current account surpluses. Since 1997, when the current account reached its highest surplus at about 3% of GDP, it has fallen somewhat to about 2% of GDP as the recovery in private domestic demand accelerated. Current and prospective external surpluses in combination with high growth suggest that the competitiveness of the Swedish economy is currently strong.

3.6.3. Unit labour costs and other price indices

The examination of the development of unit labour costs and other price indices, required by Article 121(1), is included in the section on price stability (see 3.1.2.).

ANNEX A: COMPATIBILITY OF NATIONAL LEGISLATION

According to the second sentence of Article 121 of the Treaty, the report drawn up under this article "shall include an examination of the compatibility between each Member State's national legislation, including the statutes of its national central bank, and Articles 108 and 109 of this Treaty and the Statute of the ESCB".

According to Article 109 of the Treaty, each Member State shall ensure, at the latest at the date of the establishment of the ESCB, that its national legislation including the statute of its national central bank is compatible with this Treaty and the Statute of the ESCB.

Following the approach chosen in the convergence report of 1998, the examination is divided into three areas:

- objectives of national central banks;

- independence;

- integration in the ESCB and other legislation.

Objectives of national central banks

The objective of a national central bank must be compatible with the objectives of the ECB as formulated in Article 105(1) of the Treaty (and Article 2 of the Statute of the ESCB):

"The primary objective of the ESCB shall be to maintain price stability. Without prejudice to the objective of price stability, the ESCB shall support the general economic policies in the Community with a view to contributing to achievement of the objectives of the Community as laid down in Article 2".

References in national law to the policy of the government or to specific macroeconomic objectives are not incompatible provided that the primacy of the first and second objectives of Article 105 of the Treaty is respected.

Independence

Article 108 of the Treaty ensures that the ESCB will operate free from instructions from third parties:

"When exercising the powers and carrying out the tasks and duties conferred upon them by this Treaty and the Statute of the ESCB, neither the ECB nor national central bank, nor any member of the their decision-making bodies shall seek or take instructions from Community institutions or bodies, from any government of a Member State or from any other body. The Community institutions and bodies and the governments of the Member States undertake to respect this principle and not to seek to influence the members of the decision making bodies of the ECB or of the national central banks in the performance of their tasks".

The various aspects of independence are described in more detail in the convergence report of 1998.

Integration of national central banks in the ESCB and other legislation

According to Article 9.2 of the Statute of the ESCB, the ECB shall ensure that the tasks conferred upon the ESCB are implemented either by its own activities or through the central banks. Furthermore, according to Article 14.3, the central banks are an integral part of the ESCB and shall act in accordance with the guidelines and instructions of the ECB. Therefore, provisions in the statutes of the central banks, which stand in the way of the central banks assuming their role, need to be adapted under Article 109.

The examination in this report recalls the result of the examination in 1998, describes legislative action since the earlier examination, summarises the essential elements of central bank legislation in force and concludes with an assessment of compatibility. The statement on compatibility concludes that a country's legislation either is or is not compatible with the Treaty and the ESCB Statute. Legislation is not compatible where incompatibilities exist which infringe upon principles of the Treaty. In some cases, imperfections have been identified which are either of technical nature or are ambiguities rather than obvious inconsistencies. The judgement on compatibility is not affected by the reference to such imperfections.

ANNEX B: INFLATION CRITERION

This Annex discusses implications of the third stage of EMU for the operational definition of the price convergence criterion and describes recent methodological developments in the harmonized consumer price indices.

B.1. Treaty provisions

The Treaty requires the achievement of a high degree of price stability as a prerequisite for entering the third stage of EMU. According to Article 121(1) of the Treaty, the convergence reports of the Commission and the ECB shall "examine the achievement of a high degree of sustainable convergence by reference to the fulfilment by each Member State of the following criteria", including "the achievement of a high degree of price stability; this will be apparent from a rate of inflation which is close to that of, at most, the three best-performing Member States in terms of price stability".

This rule is developed further in Article 1 of Protocol No 21 (ex 6) the first sentence of which reads as follows: "The criterion on price stability...shall mean that a Member State has a price performance that is sustainable and an average rate of inflation, observed over a period of one year before the examination, that does not exceed by more than 1½ percentage points that of, at most, the three best-performing Member States in terms of price stability".

Article 1 of Protocol No 21 also specifies that "Inflation shall be measured by means of the consumer price index on a comparable basis, taking into account differences in national definitions". To meet this requirement, Harmonized Indices of Consumer Prices (HICP) have been produced in each Member State since January 1997 using a harmonized methodology developed by national statistical offices and Eurostat.

B.2. Reference value for inflation

The assessment of the price convergence criterion as laid down in Protocol No 21 of the Treaty requires an operational definition of the reference value against which the price performance of Member States with a derogation will be assessed. In the Commission's Convergence Report of March 1998, the reference value was calculated as the arithmetic average of the inflation rates of the three best performing Member States plus 1.5 percentage points.

The Treaty makes no distinction between initial and later entrants to the euro area as regards the application of the criteria. The principle of equal treatment implies that, as far as possible, Member States joining later should not be confronted with additional hurdles, nor be allowed to join on looser terms than the first-round entrants.

Within this framework, it needs to be examined whether the fact that the third stage of EMU is underway - which was not the case at the time of convergence assessment in March 1998 - might have implications for the application of the price convergence criterion. For instance, the question arises whether countries not participating in the euro should be considered when the average of the best-performing Member States is used to calculate the reference values for the inflation and long-term interest rate criteria.

The context in which the inflation (and interest rate) criteria were devised must be recalled. When drafting the convergence criterion on inflation, the Treaty negotiators had to choose a reference performance which could be expected to satisfy the criterion of "a high degree of price stability". No widely accepted definition of price stability existed at the time. In referring to the "best performers in terms of price stability", it was assumed that these countries would provide the most appropriate standard available before the start of monetary union.

Since the start of the third stage of EMU, price stability in the eleven participating Member States is pursued at euro area level by means of the monetary policy of the Eurosystem. Moreover, the ECB has provided its interpretation of price stability: "price stability shall be defined as a year-on-year increase in the Harmonized Index of Consumer Prices (HICP) for the euro area of below 2%". [29]It therefore seems desirable that the assessment of "a high degree of price stability" should also take into consideration the price stability performance of the euro area as well as the ECB's definition of price stability. This is all the more so since the euro and the euro area economy constitute the economically relevant benchmarks to which countries aiming to join the euro should orient their convergence efforts.

[29] The ECB has made it clear that negative inflation rates, should they occur, would not be consistent with price stability. Price stability TTTABLEin the euro area is to be maintained over the medium term.

As an additional consideration, it is possible to envisage situations in which the average of the three best performers would not constitute an economically meaningful benchmark. This could be the case if the average inflation performance in the three best-performing Member States was very close to zero or substantially below the average of the euro area. In consequence, the calculated reference value could be below 2%, or it could lead to a reference value below the euro average. If such cases should arise, a careful appraisal would have to be made in the light of the ECB's definition of price stability and the inflation performance of the euro area with a view to not including countries with a non-representative inflation performance in the reference group in order to better reflect the Treaty's intentions in providing a yardstick for the achievement of a high degree of price stability.

Although the Treaty is not specific on whether Member States outside the euro area should be part of the group of reference countries or not, continuation of previous practice would speak in favour of doing so. From an economic point of view, however, there is no reason why the decision on whether a Member State fulfils the conditions for adopting the euro should depend on the inflation and interest rate performance of countries which are not in the euro area.

These considerations may be relevant in future applications of the convergence criterion but they do not have any practical bearing on the present assessment of inflation convergence in Sweden and Greece. Table B.1 shows the 12-month average rate of inflation in March 2000 for each Member State. The three best-performing Member States in the EU are Sweden (0.8%), France (0.9%) and Austria (0.9%). The unweighted arithmetic average of the inflation rates in these three countries is 0.9%. The resulting reference value, calculated according to the operational definition used in the March 1998 convergence assessment, is therefore 2.4%.

Given generally low inflation in the EU and a 12-month average inflation rate of 1.4% in the euro area, none of the three best-performing EU Member States is considered to have a non-representative inflation performance.

However, Sweden is not part of the euro area. On the basis of a definition which excludes non-euro Member States from the group of reference countries, the 12-month average inflation rate of the three best-performing euro area Member States is 1.0%, giving a reference value of 2.5%.

Table B.1 also shows the 12-month average inflation in the euro area for March 2000 as well as the upper limit of the ECB's definition of price stability, both of which are relevant in assessing the "achievement of a high degree of price stability".

Table B.2 shows how average inflation rates have evolved in the Member States since the beginning of 1999, and the derivation of the reference value, calculated using the same method as in the 1998 Convergence Report. Table B.3 shows for the period since the HICP became available (first comparison possible December 1996) the three best performers (in the EU) in terms of price stability, the average inflation rate in these three best performers and the corresponding reference value.

B.3. Recent methodological developments in HICP

The HICPs provide comparable measures of consumer price inflation in the Member States. The HICPs are used for the assessment of convergence in inflation across the EU and they form the main measure of price stability in the euro area. Since the Commission undertook its previous convergence assessment in March 1998, Eurostat and the statistical offices of the Member States have continued to refine the measurement of inflation provided by the HICPs.

HICPs are intended to cover all forms of household expenditure ("household final monetary consumption expenditure"). At the launch of the HICP in 1997, certain technically difficult areas were excluded because Member States did not yet agree on comparable methodologies. The initial product coverage included approximately 95% of all household final monetary consumption expenditure (this concept does not include owner-occupied housing). The product coverage of the HICPs has now been extended to include additional goods and services, and there have been changes to the population and geographic coverage. [30] This brings the coverage to virtually 100% of household final monetary consumption spending (still not including owner-occupied housing). The extension occurs in a two-stage approach: most are reflected from the January 2000 index but a few items will be included only from the index for January 2001.

[30] The extension takes place under the legal umbrella of Council Regulations (EC) Nos 1687/98 and 1688/98.

The extension of product coverage from January 2000 covers health, education, and social protection items, certain insurance and financial services, and certain tax-like charges in connection with housing. The extension was possible due to an agreed methodology on tax-like charges and subsidies. [31] The consumer prices of subsidised goods and services, e.g. pharmaceutical products and medical and dental services, are measured net of reimbursements and subsidies. Since the prices that consumers pay in the health, education and social protection areas are linked to the social welfare and tax policies in the different countries, the weights and price changes for these sub-indices may vary markedly between Member States. The insurance and financial services have been extended with e.g. health insurance and fees for investment counsellors. The few additional services which will be covered only from January 2001 are hospital services, social protection services provided within the home (e.g. cleaning, meals), retirement homes, and residences for the disabled. There has also been a harmonization of the treatment of tariff prices.

[31] As laid down in Council Regulation (EC) No 2166/1999.

The geographic and population coverage was harmonized in all Member States with the January 2000 index. Spending by tourists and cross-border shoppers, which previously was not covered by all Member States, are now always covered in the Member State in which the purchase is made. The HICP now has complete population and geographic coverage without omissions or overlaps.

The implication of the methodological improvements for headline inflation is likely to be very small in the early months of 2000. This is because of the way the data are linked: the annual rate of change for the all-items HICP index in, for example, March 2000 is based on the change from March 1999 to December 1999 (using the old coverage) combined with the change from December 1999 to March 2000 (using new coverage). Thus, in this example, the extension of the coverage only affects 3 out of 12 months. The impact may become progressively larger during the year but not enough information is available to provide a reliable estimate of the full-year impact of the methodological improvements.

Given that the data used for the price convergence criterion is the average rate of inflation observed over one year, the practical implications of this change in methodology for the assessment of inflation convergence in this examination is small. [32]

[32] However, in connection with the proper chaining of initial coverage items between December 1999 and 2000, the National Statistical Service of Greece replaced expenditure weights referring to the year 1994 with a new set of expenditure weights referring to the year 1998. The new weights are used in the HICP for January 1999 onwards. This is in compliance with Commission Regulation (EC) No 2454/97 concerning the quality of HICP weights and gives an HICP comparable to those of other Member States.

B.4. Additional tables on other price and cost indicators

ANNEX C: GOVERNMENT BUDGETARY DATA

C.1. Shift to ESA 95 and revision to Regulation (EC) No 3605/93

The government budgetary data used in the excessive deficit procedure, and hence for the assessment of convergence, are on a national accounts basis. From the year 2000 a new system of accounts, ESA 95, has been brought into use for these purposes. This section describes main features of the new system and key differences from the earlier system of accounts, ESA 79, which was used in the excessive deficit procedure until the end of 1999 and in the 1998 convergence report.

From April 1999, the new system of national and regional accounts ESA 95 [33] has become obligatory in the EU Member States. However, according to Regulation (EC) No 3605/93 [34] - which specifies the relevant definitions of government deficit and debt, and of GDP, and governs the twice-yearly transmission of government data by the Member State to the Commission - the former accounting system (ESA-2nd edition, or simply ESA 79) remained the applicable standard for the purpose of the excessive deficit procedure until the end of 1999.

[33] The ESA 95 accounting methodology and the transmission programme of national accounts data are specified by Council Regulation (EC) No 2223/93 of 25 June 1996 on the European system of national and regional accounts in the Community, OJ L 310, 30.11.1996, p.1.

[34] Council Regulation (EC) No 3605/93 of 22 November 1993 on the application of the Protocol on the excessive deficit procedure annexed to the Treaty establishing the European Community, OJ L 332, 31.12.1993, p.7.

This lag in the application of the new accounting system to the excessive deficit procedure had been planned since long, notably in the regulation that adopted the ESA 95 methodology. [35] It allowed the two reporting exercises of 1999 (in March and September) to be made on the same methodological basis. As a result, the Council Decision of 17 December 1999 to abrogate the Decision on the existence of an excessive deficit in Greece was taken on the same accounting basis as all previous Council decisions on the existence of, or correction of, excessive deficits.

[35] Article 8(3) of Regulation (EC) No 2223/96.

Regulation (EC) No 3605/93 was amended [36] to take into account the new accounting environment in February 2000, and Member States reported government data, in the context of the excessive deficit procedure according to the new methodology, for the first time at the beginning of March 2000. Therefore, unless otherwise stated, government data in this report comply with ESA 95.

[36] Council Regulation (EC) No 475/2000 of 28 February 2000 amending Regulation (EC) No 3605/93 on the application of the Protocol on the excessive deficit procedure annexed to the Treaty establishing the European Community, OJ L 58, 3.3.2000, p.1.

The implementation of ESA 95 is the result of a long-term exercise conducted within the EU (Eurostat and the national statistical offices) but also at international level (UN, IMF, OECD, World Bank). The new accounting system ensures consistency with the international System of National Accounts (SNA 93) and the 5th edition of the IMF Balance of Payments Manual. It culminates in the modernisation of concepts to reflect new economic and financial phenomena, increasing harmonized methodologies, greater precision in accounting rules and definitions and a wider coverage of the whole range of economic activities. In addition, the implementation of ESA 95 has been the opportunity for each Member State to improve its sources and its methods of compiling the accounts.

Government accounts, in particular the government deficit/surplus, have been affected by the changeover to the new system. However, for most Member States, the changes in the balance ratios are relatively small, especially if the magnitude of total government expenditure and revenue (in some cases above 50% of GDP) is taken into account. The methodological differences between ESA95 and the former system that have a larger impact on the government balance ratios are the following:

delimitation of general government. Some entities that were previously recorded outside the government sector are now treated as part of general government. Similarly, some units formerly included in government are now accounted as financial or non-financial corporations.

Transactions are, in principle, recorded at the time of the events that generate them (accruals principle), rather than when a payment is made. For example, interest is recorded as accruing continuously to the creditor and no longer when it is paid. The time of recording of taxes and social contributions has also been adjusted.

Changes in GDP (generally upwards) have also led to marginal revisions in the deficit and debt ratios.

The government debt (general government consolidated gross debt at nominal value) is not strictly-speaking an ESA 95 concept, just as it was not an ESA 79 concept. It is defined in Regulation (EC) No 3605/93 specifically for the purpose of the excessive deficit procedure. This is because the accounting system that existed in 1994, when the excessive deficit procedure entered into force, did not contain any definition of debt. [37] Moreover, the definition of financial liabilities at market prices that exists now in ESA 95 is not appropriate for the excessive deficit procedure. Nonetheless, the debt definition was also revised in February 2000 to take into account the new accounting environment. Among other changes, the new delimitation of government sector has been taken into account, the components of the debt have been further amplified (for example, financial leasing is now included in the liabilities constituting government debt) and the valuation of debt issued in foreign currency (including liabilities swapped to, or among, foreign currencies) has been clarified.

[37] ESA 79 was a purely flow-based system that, as said above, contained neither balance sheets nor any other stock accounts.

C.2. Additional tables on the public finances

ANNEX D: EXCHANGE RATE CRITERION

D.1. Treaty provisions and ERM II

The relevant period for assessing exchange rate stability in this report is from April 1998 to March 2000, extending over the last nine months of Stage 2 of EMU and the first 15 months of Stage 3. Equality of treatment considerations require that application of the exchange rate criterion must reflect both the situation that existed in Stage 2 and the changes implied by the move to Stage 3 - notably the introduction of the euro and the establishment of the ERM II. Accordingly, the assessment of exchange rate stability in this examination will be based on the Treaty provisions, as elaborated by the relevant protocol, and by the Council Resolution establishing the ERM II with effect from 1 January 1999.

The relevant Treaty provisions are:

* The third indent of Article 121(1) (ex Article 109j(1)), which refers to the exchange rate criterion as:

"the observance of the normal fluctuation margins provided for by the exchange-rate mechanism of the European Monetary System, for at least two years, without devaluing against the currency of any other Member State";

* Article 3 of Protocol No 6 which states that:

"The criterion on participation in the exchange-rate mechanism of the European monetary system referred to in the third indent of Article 121(1) of this Treaty shall mean that a Member State has respected the normal fluctuation margins provided for by the exchange-rate mechanism of the European Monetary System without severe tensions for at least the last two years before the examination. In particular, the Member State shall not have devalued its currency's bilateral central rate against any other Member State's currency on its own initiative for the same period".

* The Council Resolution on the establishment of the ERM II (97/C 236/03 of 16 June 1997) [38] which states that:

[38] OJ C 236, 2.8.1997, p. 5.

"With the start of the third stage of economic and monetary union, the European Monetary System will be replaced by the exchange rate mechanism as defined in this Resolution....The exchange-rate mechanism will link currencies of Member States outside the euro area to the euro".

D.2. Application of exchange rate criterion in Stage 2 of EMU

In the convergence examination of March 1998, application of the exchange rate criterion was based on the median-currency approach. [39] This was a new analytical framework, necessitated by the changes that were made to the ERM following the exchange rate turbulence of 1992/1993. These changes had complicated the application of the criterion by creating uncertainty about how to define exchange rate stability in the context of ERM participation.

[39] The median currency within the ERM is defined as the currency closest to its central rate against the ECU at the exchange rate fixing on any given day.

The main difficulty in applying the exchange rate criterion related to the decision to widen the ERM fluctuation margins to 15% in August 1993. At the time of the convergence examination some five years later, it was unclear whether these widened margins or the former margins of 2.25% corresponded to the term "normal fluctuation margins" in the criterion. On the one hand, assessment of exchange rate stability on the basis of movements within the 15% margins was considered inappropriate, because (i) the Treaty had been drafted when the 2.25% margins were considered to be normal; (ii) the wider margins provided a very accommodative benchmark against which to measure exchange rate stability; (iii) the widening of the margins had been introduced as a temporary measure with the expectation of returning to the narrow margins; and (iv) the intention in widening the margins had not been to facilitate greater exchange rate variability but to counter speculation against ERM currencies. On the other hand, the absence of a formal commitment to observe the original 2.25% margins and the presumption that the wider margins could be exploited had to be taken into account. This meant that assessment of exchange rate stability on the basis of the original 2.25% margins was also considered to be inappropriate. The median currency approach was devised as a balanced solution to this difficulty in applying the criterion.

In the median currency approach, the exchange rate stability of ERM currencies was assessed in the context of fluctuation margins of ±2.25% against the currency at the centre of the mechanism on each day. While these margins may seem to correspond to the original narrow margins, the median currency approach was, in fact, less restrictive. Using the median currency as a basis meant that the stability of a currency was assessed against the currency at the middle of the mechanism rather than against the strongest currency. In consequence, the use of the fluctuation margins of ±2.25% implied that a 4.5% appreciation/depreciation against another currency would be tolerated, rather than the 2.25% appreciation/depreciation implied by the original normal margins. However, the attractive feature of the median currency approach was that it assessed favourably those currencies that were clustered at the centre of the ERM.

A breach of the ±2.25% fluctuation margins against the median currency was not automatically classified as indicative of severe tensions. In assessing whether a breach of the margins corresponded to severe tensions, a range of elements was taken into account. These included: (i) the duration and amplitude of the deviation; (ii) the nature and extent of any policy response with particular reference to foreign exchange intervention and/or changes in short term interest rates [40] and (iii) whether the pressure has been towards appreciation or depreciation of the currency. Indeed, a distinction was drawn between tensions in respecting the upper and lower margins, which was seen as corresponding, respectively, to relative strength and weakness of a currency. Given the implied linkage between severe tensions and devaluation in the wording of the Treaty, it was considered reasonable to exclude movements above the 2.25% margin against the median currency as a possible cause for non-fulfilment of the criterion. This interpretation was relevant in the case of the Irish pound, which had been far above the 2.25% range for most of the two-year assessment period.

[40] It should be recalled that the 1987 Basle-Nyborg 1987 agreement called for "...a more active, flexible and concerted use of the instruments available, namely exchange-rate movements within the fluctuation band, interest rates and intervention" (Press communiqué of the Committee of Governors of EC central banks). For completeness, any episodes of intervention within the ±2.25% limits were also examined.

D.3. Application of the exchange rate criterion in Stage 3 of EMU

Unlike the original ERM, the ERM II is not based on multilateral exchange rate commitments between all participant currencies but on a bilateral exchange rate commitment between the euro and the other participant currencies. A "standard" fluctuation band of ±15% has been established and this, in principle, corresponds to the "normal fluctuation margins" referred to in the Treaty. The standard fluctuation band of ±15% implies the possibility of a 30% appreciation/depreciation of a currency against the euro while remaining within the band, making it a very accommodative benchmark for assessing exchange rate stability. There exists the possibility of establishing narrower fluctuation bands within the ERM II to reflect progress in economic convergence, but there is no requirement to do so [41]. The main features of the ERM and the ERM II are compared in the box.

[41] Indeed, Greece has applied to adopt the euro while maintaining the standard fluctuation band for the GRD.

Despite the differences between the ERM II and the original ERM, equality of treatment can be reasonably assured by a modification of the framework for applying the exchange rate criterion in Stage 2. While the median currency approach no longer applies in the ERM II, a similar assessment of exchange rate stability can be made in the context of a fluctuation band of ±2.25% around a currency's central parity against the euro. Continuity between the two approaches is enhanced by the fact that the median currency in the original ERM on the final day of Stage 2 was fixed irrevocably against the euro from the first day of Stage 3. The "euro-based approach" would also imply that an appreciation/depreciation of 4.5% would be tolerated, although, once again, a breach of the band would not necessarily correspond to severe tensions but would be assessed by reference to the same range of elements as in the examination of 1998. As with the median-currency approach, a distinction would be drawn between movements above the 2.25% upper margin and movements below the 2.25% lower margin, with only the latter potentially indicating severe tensions within the ERM II.

BOX: Similarities and differences between the ERM and the ERM II

The common elements of the ERM and the ERM II are:

* central rates and fluctuation bands set by common procedure (involving Finance Ministers, ECB and NCB Governors and the Commission);

* standard fluctuation band is ±15%, while not excluding possibility of closer links;

* intervention support with appropriate financing will be automatic at the margin;

* realignments will be made by common procedure;

The main differences between the ERM and the ERM II are:

* bilateral links between the euro and the "pre-in" currencies (i.e. the so-called "hub and spokes" model) replace the multilateral links in the ERM; accordingly, intervention obligations will be bilateral between the ECB and each "pre-in" NCB. The ERM was characterised by multilateral intervention obligations;

* the euro is the formal anchor of the ERM II; while the DEM acted as de facto anchor in the ERM, this was not a formal role. With the euro as anchor, the operation of the ERM II will be clearly focused on the need to foster convergence among the "pre-ins" toward the standards of macroeconomic stability in the euro area;

* the ECB or any non-euro NCB will have a formal right to suspend intervention if its price stability objective is jeopardised; this "safeguard clause" did not formally exist in the ERM, although the events of 1992/1993 indicated that there were limits to the commitment to "unlimited intervention";

* realignments will take place in a timely manner. The experience of the ERM in 1992/93 revealed the dangers of a failure to adjust central parities before the emergence of speculative pressures. In the ERM II, all parties to the agreement, including the ECB, will have the right to initiate a procedure which may result in a realignment. In the ERM, a procedure for realignment of a currency's central rate could be initiated only by the Member State concerned.

D.4. Fulfilling the exchange rate criterion in the current examination

The principle of equal treatment requires that the exchange rate criterion should be applied as consistently as possible over time. For the purposes of this examination the criterion will be applied (i) using the median currency approach in respect of the months from April to December 1998 (Stage 2 of EMU); and (ii) using the euro-based approach in respect of the remainder of the examination period (Stage 3 of EMU). In summary, therefore, the conditions to be respected in fulfilling the exchange rate criterion would be as follows:

* Participation in the ERM II at the time of the assessment is mandatory.

* Participation in the ERM/ERM II for at least two years is expected, although exchange rate stability during a period of non-participation before entering ERM/ERM II can be taken into account.

* No downward realignment of the central parity either in the ERM or in the ERM II within the two-year examination period.

* Exchange rate to have been maintained within a fluctuation band of ±2.25% around the currency's central parity against the median currency in the context of the ERM and against the euro in the context of the ERM2. However, the extent to which a breach of the ±2.25% fluctuation band would correspond to severe tensions would take account of a range of relevant considerations. A distinction is to be made between exchange rate movements above the 2.25% upper margin and movements below the 2.25% lower margin.

ANNEX E: LONG-TERM INTEREST RATE CRITERION

The application of the interest rate criterion in this report is on the same basis as in the 1998 convergence report.

For the operational assessment of the criterion on the convergence of interest rates the yield on benchmark ten-year bonds has been used; details about the interest rates used for the Member States are given in the box. The long-term interest rates are averaged over periods of twelve months. The reference value is calculated from the simple average of the average long-term interest rates of the three best performing Member States in terms of price stability [42] plus 2 percentage points. As explained in Annex B, the three best performing Member States in terms of price stability are selected using the harmonized indices of consumer prices (HICPs); average inflation rates based on the HICPs can only be calculated from December 1996 and are not available beyond March 2000. This implies that the reference value for long-term interest rates can only be derived on a consistent basis for this same period, and that the March 2000 data are the relevant ones for the purpose of the assessment in this report.

[42] It should be noted that the best performing Member States in terms of price stability do not necessarily have the lowest interest rates.

BOX: Data for the interest rate convergence criterion

The fourth indent of Article 121(l) of the Treaty requires that the durability of nominal convergence and exchange rate stability in Member States should be assessed by reference to long-term interest rates. Article 4 of Protocol No 21 on the convergence criteria adds that these "Interest rates shall be measured on the basis of long-term government bonds or comparable securities, taking into account differences in national definitions".

Article 5 of Protocol No 21 requires that the Commission should provide the statistical data used for the application of the convergence criteria. However, in the context of the interest rate criterion, the European Monetary Institute developed the criteria for harmonising the series of yields on benchmark 10-year bonds on behalf of Eurostat and started collecting the data from the central banks, a task which has then been transferred to the European Central Bank. The selection of bonds for inclusion in this series has the following characteristics:

* a residual maturity close to 10 years;

* issued by central government;

* adequate liquidity, which is the main selection criterion; the choice between a single benchmark or the simple average of a sample is based on this requirement;

* yield gross of tax;

* fixed coupon.

For all the Member States, the representative interest rates used in this examination incorporate all of the above characteristics. This ensures cross-country comparability. Since December 1997, 11 Member States have been using a single benchmark bond and four a sample of bonds (Germany, Spain, Portugal, Sweden). The harmonized series for Greece starts in mid-1997, as a 10-year benchmark bond has been available only since June 1997. Before this date, the representative interest rate was based on available best proxies: the yield on a seven-year bond with fixed coupon from March to June 1997, rates at issue of seven-year bonds from February 1996 to January 1997, and rates at issue of five-year bonds from September 1992 to January 1996.

Table E.1 reports the annual averages for long-term interest rates in the years 1995-1999 and in March 2000, the latest date for which monthly data are available. Average long-term interest rates for the 12-month period from April 1999 to March 2000 are shown in the final column of the table. The reference value has tended to decline since December 1996, when it was 9.1%.

In March 2000, the reference value, derived from the average interest rates in France, Austria and Sweden, the three best performing Member States in terms of price stability, was 7.2%.