Open Issues in European Central Banking

Open Issues in European Central Banking Lorenzo Bini Smaghi and Daniel Gros

Open Issues in European Central Banking Lorenzo Bini Smaghi Director for International Affairs Italian Treasury Rome

and Daniel Gros Deputy Director Centre for European Policy Studies Brussels

9&

in association with THE CENTRE FOR EUROPEAN POLICY STUDIES

M

First published in Great Britain 2000 by

MACMILLAN PRESS LTD Houndmills, Basingstoke, Hampshire RG21 6XS and London Companies and representatives throughout the world A catalogue record for this book is available from the British Library. ISBN 0-333-77913-4

First published in the United States of America 2000 by

ft!

ST. MARTIN'S PRESS, INC., Scholarly and Reference Division, 175 Fifth Avenue, New York, N.Y. 10010 ISBN 0-312-22845-7 Library of Congress Cataloging-in-Publication Data Bini Smaghi, Lorenzo. Open issues in European central banking / Lorenzo Bini Smaghi and Daniel Gros. p. cm. Includes bibliographical references and index. ISBN 0-312-22845-7 (cloth) 1. Banks and banking, Central—Europe. 2. Monetary policy-Europe. I. Gros, Daniel, 1955- . II. Title. HG2976.B56 1999 332.1T094-dc21 99-40985 CIP

© Lorenzo Bini Smaghi and Daniel Gros 2000 All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No paragraph of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1P 0LP. Any person who does any unauthorised act in relation to this publication may be liable to criminal prosecution and civil claims for damages. The authors have asserted their rights to be identified as the authors of this work in accordance with the Copyright, Designs and Patents Act 1988. This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. 10 9 8 7 6 5 4 3 2 1 09 08 07 06 05 04 03 02 01 00 Printed and bound in Great Britain by Antony Rowe Ltd, Chippenham, Wiltshire

CONTENTS Foreword

ix

Introduction

1

1. Who is in charge of the Eurosystem?

5

Introduction

5

1.1. The decision-making process

6

What the treaty states (and what it doesn't state) The balance of power The risks of a weak ECB 1.2. Implementation The interpretation of the treaty The underlying reasons for decentralization The consequences of decentralization

6 9 11 17 17 20 23

1.3. Is (de)centralization sustainable?

25

2. Who takes care of financial stability in Europe?

29

Introduction

29

2.1. Preventing financial crises The instruments Financial regulation in the EU Prudential supervision

30 30 31 35

2.2. Financial crisis management

40

Solvency and liquidity crises Bailing out banks in monetary union The lender-of-last-resort in monetary union Who (in the Eurosystem) should lend in the last resort?

40 41 45 49

2.3. Should the Eurosystem be granted supervisory powers?

51

Conclusions

55

Annex 2.1. Does prudential supervision interfere with monetary policy?

58

Contents

VI

3. Are European central banks over-capitalized?

67

Introduction

67

3.1. The Eurosystem and its components

68

3.2. A comparative G-3 perspective

70

Liabilities and size: A comparison with the US and Japan The composition of the asset side: Foreign exchange assets Balance sheet size: Does it matter? 3.3. Cleaning up the Eurosystem balance sheet

70 73 76 80

A lean balance sheet

80

Excess foreign exchange reserves?

82

Conclusions

84

Annex 3.1. An example of simplicity: The allocation of assets and returns in the Federal Reserve System

86

Annex 3.2. Estimating the revaluation reserves using a VaR approach 88 4. Who gets the seigniorage?

94

Introduction

94

4.1. What is seigniorage?

95

4.2. The treaty's provisions

96

4.3. The problems

98

4.4. The agreed non-solution

103

4.5. Alternative approaches to phasing-in

105

Conclusions

108

Annex 4.1. Legal Issues

111

Annex 4.2. The distribution of seigniorage in the context of the fiscal constitution of the EU

112

Contents

vn

5. How can the Eurosystem become (and remain) independent?...

118

Introduction

118

5.1. The independence provided by the treaty Statutory independence Functional independence Economic independence Personal independence

120 120 122 123 124

5.2. Measuring the ECB's independence against that of other central banks

125

5.3. The limits of measurement

129

5.4. Potential threats to the ECB's independence

132

The exchange rate regime NCBs' activities 'outside' the ESCB Conclusions

132 136 138

6. ...and accountable?

144

Introduction

144

6.1. Democratic control, independence and accountability

144

6.2. The economics of central bank accountability

147

6.3. How to assess central bank accountability?

151

6.4. How accountable is the ECB?

157

Conclusions

166

7. 3 X 2 Concrete Proposals for Strengthening the Eurosystem

171

Introduction

171

7.1. Transform the Eurosystem into a true European Central Bank

171

7.2. Clarify who is responsible for financial stability

172

7.3. Clarify financial relationships within the Eurosystem

174

Contents

Vlll

Annexes I. Chapters 2 and 3 of Title VI on Economic and Monetary Policy of the Treaty on European Union

177

II. Protocol (No. 3) on the Statute of the European System of Central Banks and of the European Central Bank

186

Index

205

Names Index

208

List of Tables and Figures Tables 1.1 1.2 1.3 1.4 2.1 2.A.1 3.1 3.2 3.3 4.1 5.1 6.1

Staff levels at the NCBs and the ECB, 1998 Eurozone policy indicators The Eurosystem's monetary policy operations Operational steps in the tender procedure EU bank assets, 1996 (percentage of total assets) Results of regressions Central bank balance sheets (liability side) Index of central bank capitalization The Eurosystem's balance sheet Basic data for seigniorage distribution in EMU Ranking of central bank independence Accountability criteria

10 15 18 19 34 61 71 72 81 100 127 153

Figures 3.1 3.2 3.3 4.1 4.A. 1

Ratios of foreign assets and gold over monetary base 74 Average inflation and capitalization indicator 77 Indicator of central bank independence and lean capitalization 78 The ratio of currency in circulation to GDP 102 The difference between GNP ratios and the ECB capital key 114

FOREWORD We have both been directly involved for many years in European monetary integration and have benefited, in our frequent exchanges of views, from our complementary positions, as official and independent academic, respectively. This book puts forward a common reflection on some key issues in European central banking that we think, from our privileged viewpoint, represent important challenges for the near future of Europe. The opinions expressed in this book reflect only those of the authors. The experience gained in our respective institutions have nevertheless been essential in providing us with a deeper understanding of all the aspects related to the economic and monetary unification of Europe. We are thus grateful to all our colleagues with whom we have worked in the course of the last few years. Special thanks are owed to Pierre Jaillet and Ulrich Ramm who took the time to comment on the manuscript. Daniel Gros' contribution to this volume would not have been possible without his many conversations with colleagues at CEPS and from its large European network. We also wish to express our gratitude to Franziska Schobert, who co-authored Chapter 3 and contributed to Chapter 4, to Luigi Cucca and Anne Harrington who edited the text with great speed and to Karin Godyns, who followed the project faithfully through its different stages.

Brussels and Rome June 1999

IX

This page intentionally left blank

INTRODUCTION After a seven-year transition period, Economic and Monetary Union (EMU) was finally launched on 1 January 1999. Exchange rates were irrevocably fixed and the European System of Central Banks (ESCB) began to conduct the single monetary policy for the 11 EU countries participating in the Eurozone. The primary objective of the policy is clearly established in the Maastricht Treaty and has since been restated on numerous occasions: price stability. The ECB, or rather the Eurosystem, has announced its interpretation of price stability (maintaining inflation below 2 per cent) and its strategy for achieving this goal. The instruments and procedures through which the Eurosystem conducts its operations are working smoothly. In spite of a general impression that the institutional framework for Europe's monetary union is complete, several issues remain unresolved. These issues are usually overlooked by the media, which tend to concentrate on more tangible news items, such as whether the Eurosystem will lower or increase interest rates. They are also neglected by academic and political commentators. These unresolved issues concern deeper, and generally less well known problems related to the peculiar institutional context in which EMU takes place. The unresolved issues arise ultimately from the fact that creating a monetary union of sovereign states is an experiment sui generis. Since the EU is not about to transform itself into something even remotely resembling a State, these issues are unlikely to be resolved in the near future. For that reason, they will exert a strong influence on the way in which the single monetary policy is established and implemented for some time to come. This book examines these key unresolved issues concerning the European System of Central Banks. We have selected six main questions, which can be roughly grouped under three headings. 1

2

Open Issues in European Central Banking

The first group of issues concerns the allocation of responsibility between the national and the European elements of the European System of Central Banks. The attentive reader will haive noticed that we have used the terms ESCB, Eurosystem and ECB as if they were equivalents. Some clarification is needed. The Maastricht Treaty assigned responsibility for conducting the single monetary policy of the Eurozone to the European System of Central Banks. The ESCB consists of the EU national central banks (NCBs) plus the European Central Bank (ECB). The latter is one element, admittedly the central one, of the entire system. Since not all EU member countries participate in monetary union, the ESCB also comprises NCBs that do not implement the single monetary policy. The ESCB thus has no practical role. It is the Eurosystem, a sub-set of the ESCB that excludes the NCBs of countries not participating in EMU, that is actually responsible for carrying out the single monetary policy. In order to understand the nature of central banking in the Eurozone, one must take into account the important role that is played by the NCBs within the Eurosystem, both in the definition and the implementation of the single monetary policy. Indeed, one of the basic principles underlying the conduct of the single monetary policy is that of decentralization. A question that naturally arises is what decentralization means and how it affects the execution of the single monetary and foreign exchange policy. Will it lead to suboptimal decisions and inefficient implementation? The first chapter examines such inefficiencies and the extent to which the decentralized structure of the ESCB can be maintained over time. Chapter 2 examines an important issue that arises in the context of the decentralized structure of the ESCB, namely the responsibility for financial stability. This is usually discussed under the heading of the lender-of-last-resort (LOLR) function in the event that financial institutions or markets experience a liquidity crisis. A clear answer to the question of who performs that function and how it is coordinated with the other tasks of the Eurosystem has not yet been provided. This is not surprising, in light of the policy of 'constructive ambiguity' generally adopted by central banks to address this issue. Could this state of uncertainty lead to sub-optimal policies in the face of the financial fragility?

Introduction

3

Having set out the institutional framework for the common monetary policy, the second group of questions examines the balance sheet structure of the Eurosystem. Although the latter focuses primarily on price stability, it cannot ignore the implications for the financial position of its member NCBs in defining its day-today policies. The allocation of profits and losses in the Eurosystem is governed by a different set of principles than that determining the common monetary policy. Two major questions arise in this context. The first, examined in Chapter 3, concerns the general issue of why, in a monetary union, NCBs should hold large amounts of assets, such as foreign exchange reserves or gold. The specific issue for Europe in this context is whether monetary union and the creation of the Eurosystem provide the opportunity to dispose of some of the excess assets on the balance sheets of NCBs. These assets, which after all represent real wealth of the member states, might be better employed elsewhere. What would be the implications if NCBs relinquished part of their assets? Would this impair the ability of the Eurosystem to conduct its policy? The chapter examines these questions and offers a model balance sheet for the Eurosystem and its components. Chapter 4 analyzes how the revenues obtained from the conduct of the single monetary policy are distributed among the member states. The Treaty defined broad principles, with a view to ensuring long-term neutrality, but the Eurosystem has postponed implementation of a system for income-sharing. The result is that the amount of profits to be allocated annually to each NCB is uncertain which may lead to substantial income redistribution across countries. This may affect the decision-making process for the single monetary policy. The third group of issues concerns the question of independence and accountability of the Eurosystem. Independence is crucial for establishing the credibility and reputation of the European Central Bank in terms of its ability to maintain a monetary policy that is oriented towards price stability. There is a broad literature on central bank independence, but it cannot be fully applied to the Eurosystem, and in particular to the ECB, since there is no precedent of an independent central bank being in charge of monetary policy for a group of sovereign countries. The independence of the ECB is

4

Open Issues in European Central Banking

usually examined in a rather theoretical way, based in particular on Treaty provisions. A more pertinent approach would be to assess the challenges that may arise to such independence. These challenges derive from the simple fact that monetary policy produces consequences not only for inflation - the containment of which is indeed the primary objective of the ECB - but also for growth and employment, for which political authorities at both national and the European level are responsible. It is thus inevitable that the ECB will be challenged in the conduct of its policies. Chapter 5 examines the sources of these challenges, whether and in what ways the ECB may defend itself and the role of the NCBs in this context. The counterbalance to central bank independence is accountability. The Statute of the ESCB and of the ECB imposes a series of reporting requirements on the system. Chapter 6 examines the extent to which these requirements are sufficient to ensure that the single monetary policy is understood by the public and the political institutions, both at the national and the European level. Proposals have already been put forward in academic and political fora for enhancing these requirements. These proposals are discussed and critically assessed. We conclude with a chapter that presents some proposals that we hope will contribute to the resolution of some of the open issues that cast a shadow over the smooth and efficient execution of the single monetary policy in Europe. For ease of reference, relevant portions of the Maastricht Treaty and the Statute of the Europeam System of Central Banks and of the European Central Bank are reproduced in Annexes I and II, respectively. The issues examined in this book may appear to be purely institutional and to concern problems internal to the world of central banking, but this is not the case. Instead, they are of interest to financial market participants and economic agents in general, since they have a direct bearing on the implementation of monetary policy. These issues are not likely to be resolved in the short term, unless a major shock occurs that makes institutional reform possible. They will remain open for some time and are thus worth examining carefully with a view to acquiring a better understanding of the factors that explain the conduct of monetary policy in Europe.

1. WHO IS IN CHARGE OF THE EUROSYSTEM? Introduction Who is responsible for the euro? The obvious answer should be the European Central Bank (ECB), but things are not so simple. The ECB is just one component of a broader entity, the Eurosystem, which comprises also the National Central Banks (NCBs) of the initially 11 - countries that have adopted the euro.1 The main tasks of the Eurosystem are to define and implement the monetary policy of the Eurozone, conduct foreign exchange operations, hold and manage the foreign exchange reserves of the member states that have adopted the euro and promote the smooth operation of the payment system (see Article 3 of the Statute in Annex II). This first chapter examines the Eurosystem and the appropriateness of its structure for the efficient management of the euro. Does the way in which the Eurosystem has organized itself really matter? At first sight, the question seems of little relevance for monetary analysis, but this is not the case. The internal organization of the Eurosystem, in fact, has implications that go far beyond simple considerations of efficiency. Important issues and priorities are imbedded in the way in which the decision-making process and the implementation of monetary policy have been organized in Europe. As will be discussed below, the evident lack of clarity on several issues will create serious problems for the conduct of the single monetary policy in the years to come. The questions addressed in this chapter have been raised in other contexts. The history of central banks, in particular those with a federal structure such as the Federal Reserve System in the United States or the Bundesbank in Germany, shows that organizational matters have broader implications, in particular on the way in which 5

6

Open Issues in European Central Banking

policies are decided and enacted. They are often a source of conflict, which leads to structural evolution and new developments in central banks. The transition, however, is not always smooth. It may entail inefficiencies and policy mistakes that burden society. These considerations apply also to the establishment of the Eurosystem. This chapter opens with an examination of the powers and responsibilities assigned to the Eurosystem. It then analyses the decision-making process, the relative weight of centrifugal and centripetal forces. Section 1.2 examines the way in which the implementation of the single monetary policy by the Eurosystem has been designed. Its efficiency, high degree of decentralization and sustainability in an integrated money and financial market are questioned. Section 1.3 draws some possible scenarios for the future. 1.1

The decision-making process

What the treaty states (and what it doesn 't state) The ECB has two decision-making bodies: the Executive Board and the Governing Council. The Executive Board has six members, including the President and the Vice-President. The Governing Council comprises the six Executive Board members and the Governors of the participating NCBs.2 The Governing Council is vested with the responsibility to 'formulate the monetary policy of the Community including, as appropriate, decisions relating to intermediate monetary objectives, key interest rates and the supply of reserves in the system, and shall establish the necessary guidelines for their implementation' (Article 12.1 of the ECB Statute). The Executive Board is assigned the task of implementing monetary policy; in particular, it 'shall give the necessary instructions to NCBs'. In addition, 'the Executive Board may have certain powers delegated to it when the Governing Council so decides'. The formulation of the treaty leaves no doubt as to the hierarchical structure of the system: NCBs are subordinate to the ECB. The way in which the decision-making bodies operate in practice depends on a series of arrangements characterizing the dayto-day functioning of the Eurosystem. This is defined in particular in the rules of procedures of the ECB.3

Who is in charge of the Eurosystem?

7

The Executive Board has an important role in the decisionmaking process. First, it prepares the agenda of the Governing Council meetings, proposing the main items to be discussed. The other members of the Governing Council can add or remove items. Second, the preparation of the meetings of the Governing Council is largely made on the basis of documents prepared and agreed first by the Executive Board. The frequency of the meetings of the Governing Council (every fortnight) makes it difficult for the NCB Governors to submit documents for discussion. Finally, if it acts in a cohesive way, as one should expect it would, the Executive Board has a strong voting bloc, six out of seventeen. It needs just three more votes to achieve a majority. The powers of the Executive Board are limited in several ways. First, while the Board has the task of preparing the meetings of the Governing Council, it does not have an exclusive power of initiative, similar to the one that the EU Commission enjoys with respect to the Council of Ministers. The ECB Governing Council can also decide on the basis of proposals that do not originate from the Executive Board. Second, several committees have been created in all major areas of ECB competence, with a view to 'assist in the work of the Eurosystem'. These committees are mandated and report to the Governing Council, via the Board. The chairmanship of these committees is normally, but not always, held by ECB staff. Part of the preparation of the meetings of the Governing Council is thus shifted from the Executive Board to these committees. This practice follows the one prevailing during the preparatory work conducted by the European Monetary Institute (EMI), in which the search for broad consensus was an important priority, at the cost, however, of slowing down the decision-making process. Third, and most important, the ECB, and hence the Executive Board, has no budgetary autonomy. The budget, in particular expenditure for staffing, is approved by the Governing Council under a special voting procedure in which the vote of each NCB Governor is weighted by the capital subscription key and from which the members of the Executive Board are excluded. As confirmation

8

Open Issues in European Central Banking

of the lack of trust that the NCB Governors place in the Executive Board, control over the ECB budget has been further tightened through the creation of a Budget Committee, composed of NCB representatives. Finally, the ECB has no control over a series of activities performed by NCBs. NCBs 'may perform functions other than those specified in this Statute unless the Governing Council finds, by a majority of two-thirds of the votes cast, that these interfere with the objectives and tasks of the ESCB. Such functions shall be performed on the responsibility and liability of NCBs and shall not be regarded as being part of the functions of the ESCB' (Article 14 of the ECB Statute). Given the composition of the Governing Council, where NCB Governors have about 65 per cent of the votes, it is not easy to find a two-thirds majority in favor of abolishing or regulating the functions performed by NCBs that are not directly in the realm of the Eurosystem. NCBs have thus been allowed to continue performing most of the functions they had performed prior to EMU. The most notable one is that of banking supervision. Other tasks, less known to the outside world but closely related to monetary policy, are the management of the foreign exchange reserves that have not been transferred to the ECB, the management of those domestic assets that are not the direct result of monetary policy operations and the holding of government accounts. Overall, over 50 per cent of the Eurosystem balance sheet is de facto outside the direct control of the ECB Governing Council because it allegedly resorts to activities unrelated to the conduct of monetary policy. NCBs are also allowed, or are not forbidden, to perform functions such as maintaining the stability of domestic financial markets, promoting their financial center or payment system (even when the latter is competing with the one managed by the Eurosystem), and giving advice to the government. The principle adopted so far seems to be that anything that is not explicitly forbidden is allowed. The implications of this freedom are further explored in Section 1.2 below and in subsequent chapters. To summarize, the drafters of the treaty envisaged that the ECB would function as the single brain of the Eurosystem, to be obeyed by all the other members (the NCBs). Translation of the treaty into practice, however, has given birth to other little brains,

Who is in charge of the Eurosystem?

9

with corresponding bodies that may at times behave independently and potentially conflict with or undermine the workings of the ECB. The balance of power The way in which the treaty has been applied raises the question of the potential conflicts between the center (ECB) and the periphery (NCBs). In what way will the balance of power tilt between the ECB, and in particular the Executive Board, and the NCBs? A comparison with the Federal Reserve System and the Bundesbank suggests that the center of the Eurosystem is relatively weak. First, in terms of voting power, the Executive Board is in a minority, exercising control over six votes out of, at present, 17. The Fed Open Market Committee (FOMC), which meets once every six weeks and has functions similar to those of the ECB Governing Council, is composed of the seven Governors and the twelve Presidents of the regional Federal Reserve Banks. Of the latter, however, only five have the right to vote. The majority thus lies with the Board of Governors sitting permanently in Washington. In the Bundesbank, after the 1992 reform, the Direktorium, which has an equivalent role to that of the ECB Executive Board, has eight members, against nine LandesZentralbank Presidents. The center is in a minority position, but a stronger one than that of the Executive Board in the ECB. A second difference is the role of the members in the decisionmaking bodies. The Presidents of the Fed District Banks or of the LandesZentralbank have often a managerial (in the US) or political (in Germany) background, which differs from, and to some extent complements, the expertise of the Board. There is an implicit division of labor in the Fed and the Bundesbank, whereby the center elaborates the policy line while the periphery provides the political support or the managerial skills for implementation. This division of labor is absent in the Eurosystem. It even faces explicit opposition, as some NCB Governors have publicly stated their aversion to letting the Executive Board play a leadership role in the preparation of monetary policy decisions.

10

Open Issues in European Central Banking

A further difference with the Bundesbank and the Fed is that the staff employed by the ECB is comparatively smaller than that of the rest of the system (Table 1.1). It will take time before the imbalance is corrected since, as mentioned above, staffing issues, which have a direct bearing on the financial accounts of the ECB, are decided by the Governing Council, the Executive Board members having no right to vote. Any attempt by the Executive Board to strengthen its staff has to be submitted for approval to those that believe that they have the most to lose, namely the large NCBs. The opposite situation prevails in the US, where the Board of Governors in Washington has to approve the budgets of the District Banks. ITie numerical imbalance between the staffs at the ECB and the NCBs makes it necessary to continue the practice followed in the EMI, namely to have most important decisions prepared by committees. Table 1.1

Staff levels at the NCBs and the ECB, 1998 Total staff

ECB

Headquarters 500**

Analytical functions*

Research

100-150

20-50

70

NCBs

Germany France Italy Spain Netherlands

17,632 16,917 9,307 3,269 1,721

2,770

1,500

360 750 300 350 165

S

23,727

1,700

350

2,000

280 150 60

* Research, statistics, economics, etc. ** It is planned that the number of staff members of the ECB will increase to about 700 by the end of 1999. Source: Morgan Stanley, Central Banking Directory, 1997, national central banks, 1995 data.

Who is in charge of the Eurosystem?

11

Another difference, in terms of staffing, between the Eurosystem and the Fed is that the regional banks in the latter case recruit from a common US market for economists and market practitioners. NCBs on the other hand continue instead to hire economists almost exclusively from their respective countries, although their outlook should now be area-wide and EU regulation forbids any discrimination against other EU nationals. The relative weakness of the center is made more acute by the absence, at least at the start of monetary union, of a well established analytical framework on which to base policy decisions. Eurozone statistics are constructed on the basis of national statistics, which are often produced by the respective NCBs. The forecasts for the Eurozone are largely computed from the aggregation of NCBs' forecasts, the quality of which varies widely and whose reliability cannot be verified in detail by the ECB. In sum, a comparison between the Eurosystem, the Bundesbank and the Federal Reserve System suggests that the ECB Executive Board starts from a relatively weak position in the monetary policy decision-making process and has to rely heavily on contributions and input from NCBs. The risks of a weak ECB Does it matter if the power and leadership for the conduct of euro monetary policy lies with the center (the Executive Board) or the periphery (NCB Governors)? Is it really a problem if the ECB is so dependent on the NCBs? An exceptionally strong Executive Board may entail some risks. In the absence of full political union, European citizens may feel that monetary decisions are being made far away from them without adequate attention being paid to their concerns. As a result, the accountability of the ECB and thus its independence may be weakened.4 NCB Governors have thus an important role to play, in particular to explain the single monetary policy to their national audiences and to support a price stability-oriented culture in their respective countries. This was certainly a concern of the drafters of the treaty.

12

Open Issues in European Central Banking

An excessively weak center also has its dangers. A short review of the experience of the two other federal central banks may be useful. During its early years, the Federal Reserve functioned under a rather decentralized structure.5 Friedman and Schwartz (1963) attribute the severity of the depression during the 1930s to serious policy mistakes made by the Federal Reserve, caused by its excessively decentralized decision-making structure: A committee of twelve men, each regarding himself as an equal of all the others and each the chief administrator of an institution established to strengthen regional independence, could much more easily agree on a policy of drift and inaction than on a coordinated policy involving the public assumption of responsibility for decisive and large-scale action. There is more than a little element of truth in the jocular description of a committee as a group of people, none of whom knows what should be done, who jointly decide that nothing can be done. And this is especially likely to be true of a group like the Open Market Policy Conference, consisting of independent persons from widely separated cities, who share none of that common outlook on detailed problems or responsibilities which evolves in the course of long-time daily collaboration. The structure of the Federal Reserve System has been changed since then, such that the Board of Governors in Washington exercises a clear majority in the decision-making body, as mentioned previously. In the Bundesbank, prior to 1992, the Board (Direktorium) consisted of seven members6 most of the time, in a Council of 18, the remaining 11 being Presidents of the LandesZentralbanken. The perception that the Board played a more important role than the one granted by its votes derived from the fact that it was based in the financial center where the major money market participants were located and on the fact that all foreign exchange operations took place in Frankfurt. It was nevertheless felt, in 1992, that the Board had to be further strengthened, especially after five new Lander joined. The 16 members representing Lander were reduced to nine. At present, eight Board members are balanced against nine LandesZentralbanken Presidents.

Who is in charge of the Eurosystem?

13

Experience suggests that the main risk of a weak center is that the decision-making process becomes too cumbersome and slow, possibly resulting in sub-optimal decisions. How relevant is this risk for the Eurosystem? We must first recall that all members of the Governing Council have one vote. This provision is very important and has to be seen in parallel with the high degree of independence of the ECB and its explicit mandate to preserve price stability.7 Together with provisions for a long tenure in office, this voting rule aims at fostering a high degree of cohesiveness and collegiality in the policy-making body. The main argument against weighted voting is that it could lead the NCB Governors to think that they are primarily representing national interests (as the ministers in the Council). By giving each member of the Governing Council the same voting power as the Governor of the Luxembourg NCB as well as the Governor of the German or French NCB, the treaty aims at entrusting them with independence and responsibility for achieving a common European objective. The principle of one-man one-vote requires that all members think in a European way. If this requirement is not met, that is if NCB Governors, voluntarily or not, continue to give priority to national objectives, as they did before EMU, the decision-making within the Governing Council will be distorted, much in the sense described by Friedman and Schwartz. This risk cannot be discounted. The lack of a fully-fledged euro-wide analytical framework underlying the single monetary policy makes it difficult for NCB Governors to become completely 'European' from the start of EMU. They can hardly be expected to change their way of thinking overnight. It is unavoidable that they will continue to be strongly influenced by their national background for some time yet. They will spend most of their time and effort in their home country, dealing with national problems and addressing national audiences. Their diversity is fueled by strong expertise on the local economy available in their respective research departments. Unless this expertise is broadened to become euro wide in scope, there is a risk that NCB Governors will arrive at the Council

14

Open Issues in European Central Banking

meetings equipped with briefing notes that are distinctly 'colored' by national views. The fact that national data become available before euro area data further increases this risk. A good indicator of future developments is the set of topics addressed in NCBs' staff working papers and in regular reports. The first reports published in 1999 tend to confirm that the focus of most NCBs remains national, while the ECB monthly report concentrates on Eurozone developments. The fear that national interests would prevail over European interests has been fueled by several decisions taken during the preparatory work, some of which are described in the next section. What would be the consequences of a persistent nationality bias in the voting behavior of the members of the ECB Governing Council? First, a one-man one-vote system, if polluted by national considerations, is likely to give excessive weight to smaller countries. This could lead to distorted decisions if smaller countries' economies diverge from those of the larger ones. If each NCB Governor looked only at his own economy, the Eurozone's economy would resemble a simple average of national developments rather than the weighted average, the weight corresponding to the relative size of the respective economies. A comparison of economic indicators for the Eurozone based on simple-average and weighted-average calculations may give an idea of the role that the 'national bias' could play in the ECB decision-making process. Looking at the 1997-2000 period on the basis of a few relevant indicators for monetary policy decisions (Table 1.2), the fictitious simple-average Eurozone grows relatively faster than the true (weighted) Eurozone, by about 1 percentage point in 1998-99, and 0.7 points in 2000. Inflation is higher in the simple-average Eurozone, expected to reach the 1.9 level by 1999, close to the ECB's tolerance level. A first conclusion that one may reach from this simple exercise is that a nationalistic bias within the ECB Governing Council in the early years of EMU could lead to a more restrictive monetary policy than is required. This may not necessarily materialize, but the risk cannot be ignored, as suggested by the experience in the run-up to EMU when some NCBs kept interest rates higher than in the core countries, even after the decision to adopt the euro in early May 1998. A good example of such behavior

Who is in charge of the Eurosystem?

15

is Ireland. Its economy was growing faster than the rest of the euro area, with clear signs of local overheating in the labor and the real estate markets. The Irish Central Bank maintained higher interest rates, several basis points above the German rates, until very close to the start of monetary union. How easy is it for the Irish Governor to change policy orientation in a few months, especially if inflation remains higher in his country than in the rest of the Eurozone? Table 1.2

Eurozone policy indicators 1997

1998

1999

2000

Real GDP growth 2.9 2.6 Weighted average 2.5 3.0 3.6 Simple average 3.4 4.0 4.2 Industrial production growth 3.1 3.6 Weighted average 3.8 4.5 4.4 Simple average 3.7 5.6 5.3 Unemployment rate 10.4 9.8 Weighted average 11.9 11.0 8.2 7.6 Simple average 9.7 8.8 Inflation 1.9 1.5 Weighted average 1.4 1.7 1.9 1.9 Simple average 1.8 1.7 Wage growth 2.8 2.7 Weighted average 2.5 1.8 3.3 3.2 Simple average 3.0 2.8 Government deficit (as a % of GDP) -1.7 -1.9 Weighted average -2.5 -2.3 -0.4 -0.3 Simple average -0.9 -0.7 Source: European Commission data. For 1999 and 2000, October 1998 forecasts.

Another reason for divergent views may come from the differences in monetary policy transmission mechanisms between the various countries. Several studies have documented such differences and their underlying reasons.8 In some countries (for example in

16

Open Issues in European Central Banking

Germany), firms rely more on long-term capital than in others (for example, Italy). This implies that interest rate changes produce different effects across countries. It is difficult to assess the empirical importance of such differences. Moreover, differences in the monetary policy transmission mechanism are not exogenous to the monetary policy regime. Convergence may nevertheless be slow, as shown by the experience of countries that have had the same macro policies for over a decade (e.g. Germany, France, Austria and Belgium), but have maintained quite different financial structures.9 If the differences between transmission mechanisms are taken into account in the ECB decision-making process and if NCB Governors look also at their national transmission processes in addition to the one functioning in the Eurozone, monetary policy decisions may turn out to produce interest rate changes that are too large and that take effect too late. The construction of a simpleaverage transmission mechanism, compared to one based on weighted-average indicators, may provide some insight into such a bias. An additional weakness of the Eurosystem's decision-making process is the size of the Governing Council, which initially numbers 17 but may potentially grow to above 20 within a few years. Research on decision-making in large and heterogeneous bodies suggests that efficiency and consistency are likely to suffer. The Governing Council is unlikely to take decisions with small majorities (say nine to eight) on relevant issues. Experience with standing committees and practices in central banking suggest that much energy is devoted to trying to achieve consensus. A no-decision might in some cases constitute the smallest common denominator. A strong 'status quo bias' is likely to influence the decision-making process. In summary, a comparatively weak Executive Board runs the risk that undue weight will be given to national considerations and that the decision-making process will be inefficient and slow.10 Similar considerations led to changes in the Federal Reserve System in 1933 and the Bundesbank in 1992, with a view to providing more weight to the center. It is not unlikely that the EU will be confronted with a similar need for reform in a few years, especially in view of the adoption of the euro by the other EU countries (UK, Sweden,

Who is in charge of the Eurosystem?

17

Denmark and Greece) and of enlargement. When the EU increases to 20 member countries and the ECB Governing Council numbers over 25, reform will become unavoidable. 1.2

Implementation

The interpretation of the treaty The treaty indicates that 'to the extent deemed possible and appropriate, the ECB shall have recourse to the national central banks to carry out operations which form part of the tasks of the ESCB' (Article 12.1). The formulation suggests that it has to be shown that a decentralized option is possible and more appropriate than a centralized one in organizing the implementation of the single monetary policy. In reality, the burden of the proof has been reversed. As a result, implementation of monetary policy has been assigned almost exclusively to NCBs, while the ECB is given a very limited role. The ECB can, under exceptional circumstances, conduct fine-tuning operations. These operations have to settle through the NCBs, given that commercial banks are not allowed to hold accounts with the ECB. The only operations that can be directly effected on the account of the ECB are exceptional foreign exchange interventions. The foreign exchange reserves transferred to the ECB are managed by the NCBs: a complex risk management structure has been put in place, by which the ECB sets the benchmark and monitors the investment performance of the NCBs. All other monetary policy operations are implemented in a decentralized way (Table 1.3), but the margins for NCB discretion are practically nil. NCBs mainly act as mailboxes for the ECB. Concerning open market operations, for instance, the criteria for making allotment decisions are set by the ECB. The NCBs collect the bids from the banks located in their respective countries and transmit them to the ECB. The ECB takes the decision on the allotment of the tender and communicates it to the public and to the NCBs, which in turn settle the transactions on their respective accounts (Table 1.4).

Table 1.3

The Eurosystem's monetary policy operations Types of transactions Provision of liquidity Absorption of liquidity

Maturity

Frequency

Procedure

Reverse transactions

Two weeks

Weekly

Standard tenders

Longer-term refinancing operations

Reverse transactions

Three months

Monthly

Standard tenders

Fine-tuning operations

Reverse transactions Foreign exchange swaps

Non-standardised

Irregular

Quick tenders Bilateral procedures

Irregular

Bilateral procedures

Regular and irregular

Standard tenders

Irregular

Bilateral procedures

Monetary policy operations Open market operations Main refinancing operations

Foreign exchange swaps Collection of fixed-term deposits Reverse transactions Outright sales

-

Outright purchases Structural operations

Standing facilities The marginal lending facility The deposit facility

Reverse transactions

Issuance of debt certificates

Outright purchases

Outright sales

Reverse transactions Deposits

Source: EMI, The Single Monetary Policy in Stage Three of EMU, September 1997.

Standardised/ non-standardised

Overnight

Access at the discretion of counterparties

Overnight

Access at the discretion of counterparties

Who is in charge of the Eurosystem? Table 1.4

19

Operational steps in the tender procedure

Steps Procedure Step 1 Tender announcement: • Announcement by the ECB through public wire services • Announcement by the national central banks through national wire services and directly to individual counterparties (if deemed necessary) Step 2 Counterparties' preparation and submission of bids Step 3 Compilation of bids by NCBs; transmission to the ECB Step 4 Tender allotment and announcement of tender results by the ECB Step 5 Certification of individual allotment results by NCBs Step 6 Settlement of the transactions by NCBs Source: EMI, The Single Monetary Policy in Stage Three of EMU, September 1997.

Concerning the two standing facilities (a lending facility at which banks can obtain overnight re-financing and a deposit facility for investing excess liquidity overnight), it is the ECB that decides the respective interest rates. Each NCB records the use of these facilities by the banks located in its own country and informs the ECB. A further element of unnecessary decentralization derives from the fact that required reserves have to be held only with the NCB where the bank is located. Banks with subsidiaries and branches throughout the euro area are not allowed to rationalize their holdings with only one NCB. In sum, the decentralized implementation of monetary policy has no impact on its integrity. Contrarily to what is sometimes believed, NCBs have no leeway in deciding any aspect related to the creation of central bank liquidity, neither in terms of price or quantity. One may thus wonder why so much importance has been given to decentralization, sometimes going beyond rational justification.

20

Open Issues in European Central Banking

The underlying reasons for

decentralization

Three main reasons have been put forward to justify decentralization. The main one is that monetary policy requires indepth knowledge of the local financial system. The second reason is that the NCBs already had the infrastructure in place to implement operations. It would thus have been inefficient to create a new one. And the third argument in favor of decentralization is to prevent Frankfurt, already a main financial center, from receiving an unfair advantage from the fact that the ECB would be located there. These arguments are not convincing. First, the knowledge of local financial systems might be useful for understanding possible differences in the transmission mechanism of monetary policy, which would be relevant in conducting analyses and preparing policy decisions. It is not relevant, however, for the implementation of the single monetary policy. The latter does not require any special knowledge of local banking conditions. As explained above, the operating procedures are the same throughout the entire area and leave no margin for NCBs to react to local liquidity conditions or to exert moral persuasion on banks. NCBs cannot influence the way in which banks located in their own country participate in monetary policy operations. The knowledge of local counterparties thus gives no particular advantage to an NCB. Furthermore, the concept of local banks and local markets has lost any practical meaning in monetary union. There is only one euro money and financial market. Forcing market participants to maintain their accounts with their respective NCB does not impart to the latter any additional knowledge on how the market functions nor on the role that any particular participant plays in it. Concerning the second argument put forward in favor of decentralization, the availability of the NCBs' infrastructure, it has not yet bee validated in practice. For sure, in some areas, such as the management of foreign exchange reserves, the adaptation and integration of NCBs in a single unified communications framework has turned out to be much more costly than a completely new system at the ECB. The last argument, concerning the potential distortion in favor of one financial center, is also irrelevant. With modern

Who is in charge of the Eurosystem':

21

communication systems, market participants do not have to be located in the same town, or country, as their trading counterparts. This also holds true for the relationship between commercial banks and the central bank. A transaction, such as a repurchase agreement, between the ECB and a bank located in Paris or Madrid entails the same procedure as a transaction between the ECB and a bank located in Frankfurt. The latter enjoys no special advantage from being in the same town as the ECB. If there were really an informational advantage as a result of being located in Frankfurt, this advantage would exist in any case, since all the relevant decisions are taken by the ECB in Frankfurt. The implementation itself, centralized or decentralized, adds no relevant information for the counterpart. Experience in several countries (for example, Italy) shows that financial centers may develop independently of where the central bank is located and operates. Only a protectionist-dirigistic view of the role of central banks has led to thinking otherwise. If the three above reasons are unconvincing, one may wonder what are the true reasons behind decentralization. Three come to mind. The main reason for decentralization is the desire to protect the employment of specialized staff in the NCBs. The 15 central banks in the EU currently employ altogether over 60,000 persons. The total employment in the Federal Reserve System, which performs similar tasks to those assigned to the ECB and the national central banks, including supervisory and reporting functions, is less than half this number. The number of staff in national central banks is clearly difficult to compare. Differences in geography, financial structure and historically inherited tasks can explain some of the striking differences in levels of staffing. The numbers of central bank staff per million inhabitants range from a high of 294 for Belgium (and 292 for France) to a low of 83 for Spain, with an EU15 average of 160. The latter figure is in contrast to 82 for the US and 50 for Japan. Some NCBs undertake extensive printing activities beyond those related to the issue of banknotes; others - of which the Banque de France, with about 17,000 employees, is the prime example - are heavily involved in the production of economic statistics, the analysis

22

Open Issues in European Central Banking

of company financial statements and so on. Some even still offer financial services to domestic private clients and banks. Aside from these caveats, there is no point in denying the fact that there is an excess of staff in the Eurosystem. It is understandable that such a problem cannot be tackled in one shot. Nevertheless, by constructing a highly decentralized and costly structure, with a view to continuing to employ the same number of staff members as were on the payroll prior to EMU, NCBs have put themselves in a rather difficult position for the future. The second reason behind decentralization is to give NCBs the opportunity to pursue other private objectives. One of these objectives, which is not always so well hidden, is to give preferential treatment to their domestic banking and financial systems. If, for example, foreign exchange interventions were centralized, there would be little chance for banks that are not particularly active in the foreign exchange market, especially those located in small countries, to operate. Decentralization ensures that more banks, independently of their efficiency, have a chance to conduct such transactions. Another example is the choice of financial instruments to be used as eligible collateral to obtain central bank credit. Two types of assets can be used for open market operations: i) the so-called 'tierone' assets, that are standardized throughout the euro area, and ii) the so-called 'tier-two' assets that are used only in some national financial markets. The selection of the second tier is left to the NCBs, although subject to approval by the ECB. The choice of having the second tier follows the same logic as above, aimed at subsidizing assets that are not easily tradable within the euro financial market. This issue is further examined in the next section. The possibility of performing activities that are not directly related to the single monetary policy has also been a reason behind decentralization. NCBs can continue, for instance, to manage large amounts of assets, domestic and foreign, with little or no notification to the ECB except when the flow of transactions is larger than a preset level. The exclusion of these assets from the regular monetary policy operations ensures that NCBs can continue to grant a privileged treatment to national assets and national banks. Other examples can be given of the attempt made by NCBs to give some form of advantage to domestic markets or banks in the context of the

Who is in charge of the Eurosystem?

23

implementation of the single monetary policy. Unfortunately, no institution has been charged with the responsibility to ensure that the way in which monetary policy is implemented does not interfere with the functioning of the euro financial market. There is also little information on the transactions conducted by central banks to enable an assessment of such distorting practices. * The third reason for decentralization is that, in spite of the provisions foreseen by the treaty, no agreement has been reached with respect to the redistribution of seigniorage between the NCBs. The decentralization of monetary policy implementation has been seen by NCBs as a way to retain income without having to redistribute it. This issue is examined in Chapter 4. The consequences of decentralization Does it really matter for the regular conduct of monetary policy if implementation is decentralized? Decentralization can produce several forms of distortion. The first concerns the use of monetary policy instruments and procedures. Reserve requirements, for instance, have been imposed mainly with a view to safeguarding the ability of NCBs to play a role in the practical implementation of the single monetary policy. Since reserves are remunerated, no wedge is created between the return on deposits and that of other financial assets. Reserves are thus of no use for the purpose of controlling monetary aggregates. The main use of reserves is to create a liquidity buffer that banks can use in order to absorb liquidity shocks. Why is this buffer so important? One reason is that it avoids frequent fine-tuning interventions by the central bank in the money market to stabilize the short-term interest rate. What is wrong, one may ask, with frequent interventions in the money market? The Federal Reserve and other central banks act in this way without any problem. The answer is that these operations can hardly be conducted in a decentralized way, through NCBs. The procedure for conducting open-market operations in a decentralized way normally takes more than 12 hours, and can be brought down to two hours in cases of emergency. This is still too long to tackle unforeseen liquidity shocks. Only operations effected directly by the

24

Open Issues in European Central Banking

ECB, with quick tenders or bilateral interventions, are rapid enough to effectively counteract liquidity shocks during the day. Allowing the ECB to conduct these frequent operations would give it a prominent role in the Eurosystem for the implementation of the single monetary policy. This is viewed as undesirable by NCBs. By imposing reserve requirements, banks are forced to hold a buffer of liquidity that they can mobilize to absorb unforeseen liquidity shocks. The need for frequent central bank intervention is thereby reduced. To sum up, reserve requirements have been imposed mainly with a view to preventing the ECB from being operational. Another distortion concerns the functioning of the financial markets. By allowing for the second tier of assets to be used as collateral for open market operations or marginal lending, some instruments that are only used locally are provided with a subsidi2:ed chance to survive in the new euro money market. For example, the Wechsel (a bill of trade dating from the 19th century), which has constituted the base for discount operations in Germany, has been introduced by the Bundesbank in its tier-two list. In theory, the existence of this type of collateral should not lead to discrimination as even tier-two collateral can be used in the entire area. It is nevertheless unlikely, for example, that a French bank would deal in Wechsel and would thus be in a position to use them as collateral. The ECB must approve all securities included in the tier-two list proposed by NCBs and can thus limit any abuse. The ECB does not have sufficient staff to thoroughly assess all these instruments, however, and must therefore rely on NCBs' self-restraint. Furthermore, since most NCBs try to promote second-rank domestic assets, the ECB does not have a majority voting in the Governing Council to ban their use. An additional distortion is produced by the fact that these assets do not have a market price. It is very difficult to adopt an appropriate non-distorting risk-management policy, in the form of initial margins and haircuts. It is thus unavoidable that the Eurosystem will create distortions in the process of allocating central bank money. This is not surprising, in light of the attempts by the NCBs to defend the prerogatives of domestic banking systems, even at the cost of introducing distortions into the euro financial market.

Who is in charge of the Eurosystem?

25

The experience of the first few months confirms this assessment. Banks have made an increasing use of tier-two assets in their operations with the Eurosystem, while keeping tier-one assets on their books for possible transactions with other market participants. This has created an advantage for banks located in countries with large amounts of tier-two assets. 1.3. Is (de)centralization sustainable? In the decision-making area, two opposite trends may emerge. On the one hand, experience in dealing with euro-wide issues may lead NCB Governors to progressively adopt a euro-wide approach. This would reduce the nationality bias in the ECB Governing Council. This may require some time, however, as well as significant adaptation to the structures of NCBs. Greater attention has to be devoted to developing a euro-wide view and cooperation, rather than competition, with the ECB. On the other hand, the increasing number of participants in the euro area is deemed to weaken the relative power of the Executive Board within the ECB Governing Council and thus to increase the importance given to national objectives, as opposed to European ones. Quite soon, the effectiveness of the ECB decision-making will become an issue to be tackled, as it once was both for the Fed and the Bundesbank. In the field of implementation, the rationalization of financial markets within the euro area is likely to fuel a trend towards more centralization of monetary policy and foreign exchange operations. Only a few NCBs are likely to really be 'on the market', while most others may become unemployed or at best under-employed. Most local foreign exchange markets, for example, are bound to disappear rather quickly, as the euro is traded against the dollar in only a few financial centers. What sense does it make, in this context, to decentralize foreign exchange market interventions? It would only lead to inefficiencies or the subsidization of some small local banks. The same reasoning applies to monetary policy operations. The integration of the euro money market is leading to the

26

Open Issues in European Central Banking

establishment of a few market-makers. With the squeezing of bid-ask spreads associated with the increasing volumes of trade, only a few counterparts are able to offer competitive bids, both for central bank money and for money market funds. For smaller banks, there is no advantage to participating in monetary policy operations with the Eurosystem. The management of their liquidity requirements can be more efficiently done through market-makers. In sum, while several thousands banks in the euro area are eligible to act as counterparts to the Eurosystem for monetary policy operations, only a few effectively play a role. The implementation of monetary policy operations will become progressively concentrated in NCBs of countries where market-makers are located. In summary, the attempt to establish a fully decentralized system for the implementation of monetary policy is unlikely to be successful, since only a few NCBs will remain operational over time. What will happen to the NCBs of countries where there are no market-makers? Will they keep their trading rooms open, after having spent so much money to modernize and link them with the rest of the Eurosystem, even if only a very few operations are conducted? What is the result of the two contrasting forces of more decentralized decision-making and more centralized implementation? Given the importance that NCBs have assigned to decentralization, it is unlikely that the market-induced tendency towards more centralized monetary policy implementation will occur smoothly. The greater the de facto centralization, the greater will be the pressure within the ECB Governing Council to adopt measures in favor of retaining decentralization, even if this increases the distortions. The underlying reason for these pressures is very simple. No NCB will accept a substantial curtailment in its activities, even those limited to implementation, or in the number of its employees. The forces at play are unpredictable, and so is the ultimate outcome. The way in which the Eurosystem will evolve is likely to give rise to tensions and potential conflicts in the years to come. The costs will be paid in terms of effective decision-making and efficiency of the euro financial market. Sooner or later, some refoim of the Eurosystem will be necessary.

Who is in charge of the Eurosystem?

27

References to Chapter 1 Borio, Claudio (1995), The Structure of Credit to the Non-Government Sector and the Transmission Mechanism of Monetary Policy: A Cross-Country Comparison, Bank for International Settlements, Working Paper No. 24. Dornbusch, Rudiger, Carlo Favero and Francesco Giavazzi (1998,), A Red letter Day?, CEPR Discussion Paper No. 1804 , February. De Grauwe, Paul, Hans Dewachter and Yunus Aksoy (1998), The European Central Bank: Decision Rules and Macroeconomic Performance, Discussion Paper No. 98-34, Catholic University at Leuven. Eichengreen, Barry (1992), 'Designing a Central Bank for Europe: A Cautionary Tale from the Early Years of the Federal Reserve System', in Establishing a Central Bank: Issues in Europe and lessons from the US, Matthew Canzoneri, Vittorio Grilli and Paul Masson (eds.), Cambridge University Press, pp. 13-48. European Commission (1997), The Impact of the Introduction of the Euro on Capital Markets, July. European Commission (1997a), 'Advancing financial integration', Background Paper, European Economy, Supplement A, No. 12, December. European Commission (1998), Official Journal of the European Communities. European Monetary Institute (1996), First Progress Report on the TARGET Project, Working Group on EU Payment Systems, August. European Monetary Institute (1997), The Single Monetary Policy in Stage III of EMU - Specification of the Operational Framework, January. Friedman, Milton and Anna Jacobson Schwartz (1963), A Monetary History of the United States 1867-1960, Princeton University Press, Princeton, N.J. Gerlach, Stefan and Frank Smets (1995), The Monetary Transmission Mechanism: Evidence from the G-7 Countries, Bank for International Settlements, Working Paper No. 26.

28

Open Issues in European Central Banking

Gros, Daniel and Karel Lannoo (1998), Capital Market and EMU, Centre for European Policy Studies (CEPS), CEPS Paperback, Brussels, May.

Notes to Chapter 1 1. The Eurosystem differs from the ESCB (European System of Central Banks) in that the latter includes the NCBs of all EU member countries, while the former comprises only the NCBs of those countries that have adopted the euro. For the purposes of this chapter, and the rest of this book, the distinction between the ESCB and the Eurosystem is not relevant, as we do not address the relationship between the 'ins' and 'outs' of EMU. We thus use the terms Eurosystem and ESCB interchangeably throughout. 2. There is also a General Council, which comprises all NCB Governors, the President and the Vice-President, but not the other members of the Executive Board, which addresses issues related to the cooperation between the euro area and those countries that have not yet adopted the euro. 3. Published in the Official Journal of the European Communities, 15 December 1998. 4. See Chapters 5 and 6 on these issues. 5. See Eichengreen (1992). 6. The law on the Bundesbank of 1957 envisaged that the Direktorium would consist of up to ten members, the same number as there were LandesZentralbanken at the time (the Saarland joined the Federal Republic later). But the maximum possible number of members of the Direktorium was never reached. 7. See Chapter 5 for further discussion about independence. 8. See for, instance, Borio (1995), European Commission (1997), Dornbusch et al. (1998) and Gerlach and Smets (1995). 9. See Gros and Lannoo (1998). 10. See De Grauwe et al. (1998).

2. W H O TAKES CARE OF FINANCIAL STABILITY IN EUROPE? Introduction Financial stability is a fundamental requisite for sustainable economic growth. It is at least as important, if not more so, than price stability. Without financial stability, economic activity and price stability may be at risk, as shown by the 1929 crisis or more recently, by the repercussions following the 1997-98 crisis in the Asian markets. The Maastricht Treaty cites price stability on several occasions as the primary objective of the European Central Bank, but it fails to assign responsibility for maintaining financial stability to any particular institution. In fact, financial stability receives only two references. First, Article 105 states that the Eurosystem 'shall contribute to the smooth conduct of policies pursued by the competent authorities relating to prudential supervision of credit institutions and the stability of the financial system'. Secondly, Article 25 of the Statute of the ESCB and of the ECB allows that the ECB 'may offer advice and be consulted by the Council, the Commission and the competent authorities of the Member States on the scope and implementation of Community legislation relating to the prudential supervision of credit institutions and to the stability of the financial system'.1 Thus, neither the Eurosystem nor any other European institution is directly held responsible for financial stability or for the conduct of prudential supervision.2 The general principle that holds the EU should act only in areas where it has explicitly been assigned a competence implies that the responsibility for maintaining financial stability in Europe rests with the national authorities. The question is whether national authorities have at their disposal the appropriate instruments to pursue this objective in a monetary union with fully integrated 29

30

Open Issues in European Central Banking

financial markets. If this is not the case, there is a risk that a vacuum of competencies will develop. This chapter examines the problems that may arise from such a vacuum and their implications for monetary policy. The first section examines the instruments available to prevent financial crises in the EU, in particular banking regulation and supervision. Section 2.2 analyzes the instruments used to manage crisis situations and how they are affected by the move to monetary union. Section 2.3 considers the possible concentration of supervisory powers in the ECB as a possible way to fill the vacuum. 2.1

Preventing financial crises

The instruments Aside from sound macroeconomic policies, which are an important pre-requisite for financial stability, two instruments are available to promote the sound behavior of financial market participants: regulation and prudential supervision. The literature has widely considered the case for financial regulation and supervision, as instruments to minimize systemic risks resulting from the interactions between financial market participants and to protect consumers from the consequences of asymmetric information.3 Regulation and supervision are expensive instruments for market participants to fund, but their costs can be justified from a social point of view by the benefits obtained from the reduction of financial risks. Authorities must try to find a balance between, on the one hand, the objective of minimizing the regulatory and supervisory burden for financial institutions, to facilitate the development of efficient financial markets, and, on the other hand, the objective of minimizing risks of misconduct with potentially systemic effects. The integration of financial markets worldwide and the spillover effects of banking crises have led national authorities to increase cooperation in the field of financial regulation and supervision, in particular among the major industrialized countries. In the European Union, the creation of an internal market for

Who takes care of financial stability?

31

financial services has also pushed towards greater integration of regulation and supervision. Financial regulation in the EU The European Union has tried to create an integrated financial market by providing a single regulatory framework through minimum harmonization and mutual recognition of national regulations and the expectation that market forces would over time lead to greater uniformity of rules and practices. A set of EU directives defines the minimum requirements for the regulation of the banking system. These cover wide-ranging issues such as capital adequacy, concentration of risks, deposit protection, rules of conduct and assignment of supervisory responsibilities. The assessments of the results obtained through this approach differ widely. According to regulators, the experience has been successful in the banking sector, as EU directives cover nearly all activities, but it is felt that progress is still needed in the other sectors.4 In the view of market participants, however, the European Union is still far from having a unified regulatory framework. The differences in opinion lie in the parameters used in making the assessments. Regulators use a quantitative parameter, based on the number of directives and areas covered. Market participants refer to the effective content of the legislation and its implementation. According to the latter, the directives address a wide range of aspects related to banking regulation, but leave too much discretion to national authorities in interpreting the directives and thus in translating their provisions into national legislation. Several examples can be cited. One is the definition of a bank, what it is allowed to do and what it is not allowed to do. These concepts have not been fiilly harmonized across the European Union. Another example concerns the definition of own funds, which is sufficiently broad in the directive to enable very different interpretations by national regulators. The same applies to the valuation of asset risks for calculating capital adequacy ratios, which may differ from country to country. Differences in accounting practices are another

32

Open Issues in European Central Banking

factor that provides substantial leeway for national differences in implementing directives. In implementation, especially in the area of prudential supervision, integration is even less advanced. Regulators often interpret and implement the EU-wide framework in a quite different manner. At times, this might be justified owing to differences in national market structures, but at times, national regulators have used their power to protect their national fiefdoms or their own institutions. Overall, according to market participants, the European financial system is not yet based on a unified regulatory framework, even in the banking sector. This assessment is unlikely to change in the first years of monetary union. The views and concerns of market participants have to be taken seriously. A single market for financial services will not be able to emerge and function efficiently if substantial differences remain in the way in which regulation is implemented. The absence of a harmonized system gives rise to regulatory arbitrage by banks in an effort to reduce the costs that regulation imposes on them. Part of this arbitrage may be beneficial, to the extent that it leads to a reduction of excessive regulation and to partial harmonization via regulatory competition to modify national regulations with the aim of providing advantages for domestic financial institutions. Competition between national regulations does not necessarily lead to equal treatment between banks located in different countries, however, which is the ultimate objective of the single market. Differences may even increase as a result of the attempt to compensate for existing regulatory divergences. There are several examples of distortions produced by the existence of different regulations in a single market. For instance, rules on deposit protection that are more stringent than the minimum EU requirements increase the costs for a foreign bank to enter a national market. Another example is the authorization by the supervisory authorities that is required in some countries to allow the foreign acquisition of a bank. In short, despite the attempt to create

Who takes care of financial stability?

33

minimum standards, the present system of biannual European market regulation has not led to harmonization. The counter argument, provided by those that are satisfied with the existing situation, is that there is no evidence that the current regulatory framework creates significant distortions to the European banking and financial market. This view is justified by the fact that banking activity has remained mainly national. PanEuropean banking groups have not developed at the pace expected. Under these circumstances, it is argued that differences in regulations may not have a serious impact on the European banking system. These arguments neglect at least two important factors. The first is that banking activity may have remained predominantly within national borders precisely because of the different national regulations, which entail high costs for banking institutions to intensify cross-border activity or to develop into pan-European groups. The second is the fact that, in spite of the remaining obstacles, monetary union has increased the integration of crossborder banking activity.5 The pan-European dimension of a bank is generally examined from the point of view of its ownership structure or the geographical distribution of its loans. The first factor is irrelevant to our analysis, and the second considers only one part of European banking activity. Loans represent about one-half of European banks' assets. The remaining 50 per cent are composed of inter-bank deposits and securities (see Table 2.1). With monetary union, the geographical distribution of these two components becomes indeterminate. In the euro money market there is no advantage from trading or taking position with banks located in any one rather than another country of the euro area. The same applies to securities, not to mention offbalance sheet activities. For instance, there is no reason why, from a risk-return point of view, a French bank should hold OATs, rather than Bunds or BTPs or trade only with other French banks rather than Austrian, Belgian or Spanish banks.

Table 2.1

EU bank assets, 1996 (percentage of total assets)

Q>

Austria

Belgium

Finland

France

Germany

Italy

Luxembourg

Netherlands

Portugal

Spain

Interbank deposits

29.17

32.86

2.78

35.52

22.41

6.75

55.69

15.55

26.53

16.25

Secunties

15.30

29.92

23.84

18.94

20.43

14.23

21.90

16.25

21.95

19.63

g

Loans

50.90

31.88

50.20

36.28

53.52

40.19

18.46

63.99

32.70

44.83

§

On

Co

^ ^

f^

Memo:

Capital & reserves

-^

4.33

2.52

5.27

4.12

4.06

9.20

2.47

4.10

8.30

8.60

£ ^

I " Source: OECD, Bank Profitability, 1998.

Who takes care of financial stability?

35

Cross-border diversification of securities holding and trading is a natural consequence of monetary union, unless regulatory obstacles stand in the way. De facto, over 50 per cent of banking activity has thus become cross-border or pan-European after monetary union. Even traditional retail banking activities are developing cross-border at an increasing pace, as a result of the reduction in payment fees and the tendency towards securitization that is also taking place in Europe. In short, the assertion that European banking system is still nationally segmented is only half true. Nevertheless, the argument is sometimes made that the differences in national structures, history, culture, vested interests and viewpoints within Europe are such that any process of harmonization will inevitably result in unsatisfactory compromises, is bound to be incomplete, and will be difficult to amend.6 These arguments certainly reflect the reality of an incomplete European political union but they should not be used as a way to hide the welfare costs and the risks entailed in maintaining different national regulations within a single currency area. Prudential

supervision

Prudential supervision is largely decided upon and implemented at the national level. Three main arguments are generally put forward to justify the current situation. The first is that since regulation is made at the national level (even if it has to conform to EU directives), it is the national authorities that best know the national legislation and can therefore most effectively monitor its implementation. The second argument is that national authorities are in a better position to assess the risks taken by financial institutions, their systemic implications and the possibility of contagion. The third argument is that the system has worked well until now. The existing cooperation between national authorities, in particular through the exchange of information on activities of branches and subsidiaries, is contended to be broadly sufficient to tackle crossborder issues. In sum, the current situation is justified on the principle of subsidiarity, according to which only those competencies

36

Open Issues in European Central Banking

that cannot be exercised more efficiently by the national authorities should be allocated to the European level. These arguments are not fully convincing, for at least four reasons. First, the present degree of cooperation cannot be considered adequate.7 The exchange of information organized between national authorities is restricted to the activity of foreign branches and subsidiaries; it is not envisaged for parent banks, whose activities are certainly more relevant to financial stability in Europe. Supervisors indicate that in case of need, they may obtain the necessary information from other countries' supervisory authorities, as foreseen by the BCCI Directive. There is no evidence, however, that such an exchange of information has taken place in the past. Furthermore, once a crisis has emerged, it is too late to process information, especially on financial institutions not previously known. In fact, supervisors jealously guard the information they collect on financial institutions located in their own countries and are not keen to disclose it, even to their colleagues in other EU member states. The proposal of a multilateral information network in the EU has been rejected. Moreover, any exchange of information is foreseen to take place only on a bilateral basis. Confidentiality is the argument put forward to restrict the exchange of information to only what is strictly necessary, which only proves that supervisors do not trust their counterparts in other countries. Second, the risk that a certain competition in laxity will develop between supervisors increases in monetary union. The benefits of tight supervision on domestic banks decrease in a single financial market with multiple supervisory systems, as participants are exposed to contagion from banks located in countries with possibly lighter supervision. There is an incentive to alleviate supervisory costs and to adopt self-disciplining risk control systems. This is clearly a case of an externality, made more acute by the intense competition among the major European financial centers. The lighter the burden imposed by local supervisory authorities, the greater is the incentive to locate business in that financial center. Third, the home-country control principle, according to which banks should be supervised by the authority where the parent

Who takes care of financial stability?

37

company is incorporated, is problematic. There is a serious question mark over the motivation that bank supervisors may have in conducting their activities in a member state where they are not directly accountable, in particular to depositors. There may even be a conflict of interest between the home-country supervisors and the way the host deposit insurance scheme is designed.8 The minimum EU standards for deposit insurance may have mitigated the problem but they have not eliminated it. These problems are exacerbated in dealing with banks with multinational activities. The home-country control principle loses any practical meaning for the supervision of banking groups where cross-border or foreign activity becomes as important as domestic business. The task of the home-country supervisors, even when they are motivated by the best of intentions, becomes extremely difficult to perform. To be sure, the supervision exercised by a foreign supervisory authority cannot be compared, in terms of quality, intensity, and thus in terms of costs, to that exercised by the local authority. As long as the foreign bank is a small branch, with limited activity, the difference in treatment may not be too worrisome. If instead the foreign bank is large, and competes with the local banks, the difference in treatment may create substantial distortions. The development of pan-European banking groups and the acceleration of cross-border banking activity magnify these distortions. It undermines the effectiveness of prudential supervision conducted at the national level. In these circumstances, the implementation of the home-country control principle results in a weakening of the quality of prudential supervision in Europe. Finally, the view according to which the risks incurred by banks, and their potential systemic implications, are better assessed by national supervisory authorities is flawed. National authorities may be in a better position to assess the credit risk of loans made to non-banks, to the extent that banks continue to lend largely to customers located in their own territory. They have no advantage, however, when the assessment entails cross-border banking activity. Concerning other types of risks, national authorities do not have a clear competitive advantage. For instance, in the euro money and

38

Open Issues in European Central Banking

financial market, most counterparties available to any bank are located outside the national borders and the local supervisory authorities have no information on them. National supervisors thus cannot assess counterparty risk. The same applies to market risk and settlement risk. Therefore, the potential for cross-border contagion cannot be monitored and assessed by national authorities alone. It certainly does not make much sense to continue supervising the main banking groups that act as market makers in the euro money and financial market according to national practices only. Even for banking groups of minor importance, it is simply wrong to think that supervision is more appropriately done at national level just because the retail banking activity takes place at that level. This ignores the fact that with a single euro money and financial market these banks can easily become heavily exposed to foreign banks in the euro inter-bank market. The importance of a euro-wide view on risks faced by banks has been recognized at least with respect to settlement risk, as the Eurosystem has the task of promoting the smooth functioning of the payment systems. The Eurosystem has put in place and manages TARGET, a real-time gross settlement system, precisely with a view to reducing settlement risk. TARGET increases the possibility of contagion throughout the euro area. A bank experiencing difficulties may well have as much exposure with other European banks as with national banks. This is not linked to the size of the bank but rather to its trading patterns and relationships within the euro money market. National supervisory authorities do not have sufficient information about the inter-bank exposure of banks located in their respective territories. They may thus overestimate or underestimate the potential for systemic risk. This is the reason why the Eurosystem has been assigned the responsibility for overseeing cross-border payment systems such as TARGET. As mentioned above, however, no multilateral exchange of information between national supervisory authorities is envisaged to effectively support this oversight function. The settlement of securities traded throughout the euro area has become an issue of increasing relevance in Europe. The

Who takes care of financial stability?

39

integration of the European financial market and the increase in cross-border trading activity broaden the range of financial institutions that can be potentially affected by difficulties, even of a technical nature, experienced in one national security settlement system or by a bank custodian. The incident involving the Bank of New York in 1985 showed the wide-ranging systemic implications that may arise from securities settlement problems. The ECB has set a series of minimum requirements for national and international settlement systems, but their oversight continues to be largely conducted by national authorities and no clear framework for crossborder cooperation yet exists. Overall, the analysis suggests that the effectiveness of the instruments available to national authorities to promote financial stability has been weakened by the move to monetary union.9 While the interaction between market participants located in different countries has intensified, and the potential for cross-border contagion effects has become more important, the regulatory framework and the conduct of prudential supervision have remained mainly in national hands. Maintaining regulatory powers and prudential supervision at the national level while national borders have been blurred within the European financial market is not only inefficient but also potentially dangerous for financial stability. It is inconsistent to try to promote financial stability in an integrated market for financial services with free movement of capital and of establishment while maintaining the responsibility for banking regulation and supervision at the national level. It is an inconsistency similar to that arising from the conduct of national monetary policies in a fixed exchange rate system with free capital movements. This inconsistency was the reason for moving to monetary union. The dangers of the current situation are particularly serious in a phase of transition, such as the one that the European banking system is experiencing in the first years of monetary union, characterized by increased competition, falling profit margins, restructuring and concentration. These developments provide a background in which financial turmoil may occur.10 The euro area appears poorly prepared to tackle this transition.

40 2.2

Open Issues in European Central Banking Financial crisis management

Solvency and liquidity crises Financial crises emerge when one or several market participants experience difficulties that prevent an adequate functioning of market mechanisms. Two types of crises can be distinguished, according to their origin. The first are crises produced by market failures. A typical example is a technical failure, which precipitates a liquidity shortage for one or several financial institutions. Intervention by the policy authorities helps to restore adequate information among market participants and avoids the failure of sound institutions. The intervention has no pecuniary costs for society. The second type of crisis arises as a consequence of unsound behavior by one or several market participants. The choice of whether to intervene or not has to be made on the basis of two contrasting factors. No intervention may lead to a financial crisis with social costs and contagion to other sectors of economic activity. Intervention, on the other hand, means that market participants' unsound behavior is not punished, thereby fueling moral hazard. Three broad forms of intervention can be identified: the government budget, the banking system and the central bank. Ultimately, the cost of the intervention is largely borne by citizens: through public finances: if public funds are provided to re-capitalize the bank; through tax credits in favor of the banking system, if the latter is asked to contribute to the bailout; and through the inflation tax or lost seigniorage, if the central bank is called upon to provide liquidity to the bank experiencing difficulties. The two types of crises are known in the literature as liquidity and solvency crises, respectively. The first type of crisis is addressed through emergency lending, or lending-of-last-resort. If intervention is needed to address the second problem, it takes the form of rescuing the institution in difficulty. The lender-of-last-resort function addresses problems of liquidity, by transforming illiquid assets into liquid ones. Bailout operations address problems of

Who takes care of financial stability?

41

solvency and thus have an impact on the size of a bank's balance sheet. The choice of the instrument to be used for managing crises should be made by the authority responsible for banking supervision, which should in theory be able to distinguish between solvency and liquidity crises. In practice, this is not always so simple. First, since the intervention often entails the use of public funds, government authorities can hardly be excluded from the decision-making. Second, since solvency crises may partly emerge as a result of ineffective supervision, it may not be appropriate that the choice of the instrument to bail out a bank is made only by the institution that is in charge of the regular performance of banking supervision. Last, but not least, it is difficult to distinguish ex ante a solvency crisis from a liquidity crisis. One must ascertain whether the bank is indeed solvent, and is only facing a problem of liquidity, or whether its liquidity problems are hiding a solvency problem. Different instruments may be used to tackle financial crises, depending on specific circumstances, the cause of the crisis, and the institutional framework in which the policy authorities exercise their functions.11 There is no established doctrine for managing a financial crisis. Several aspects have to be taken into consideration by policy authorities to design the appropriate instrument of intervention. Systemic implications and moral hazard are two of the most important factors. Authorities should act in a way that avoids inducing market participants to expect that, whatever the circumstances, any bank experiencing difficulties will always be helped. Finally, equal treatment and competition rules are also relevant in choosing the appropriate mix of instruments. These principles can be summarized under the concept of 'constructive ambiguity'. Bailing out banks in monetary union 'Constructive ambiguity' is a guiding principle also in monetary union. In theory, national authorities maintain the responsibility of choosing and using the appropriate combination of instruments to

42

Open Issues in European Central Banking

manage financial crises. This is not without problems, however. Let us consider, in turn, the three instruments mentioned above. Concerning the use of public funds, responsibility resides with national governments. Nevertheless, resort to this instrument is constrained by the budgetary rules contained in the Treaty and in the Stability and Growth Pact (SGP). Furthermore, the use of public funds to re-capitalize banks is subject to the monitoring of European competition policy authorities. The rescue of banks such as Credit Lyonnais or Banco di Napoli has shown that the margins for maneuver are narrow in principal, but rather wide in actual practice. The resort to national rescues should be more limited in monetary union, where the potential for market distortion has to be assessed throughout the euro area and financial markets. Furthermore, the bailout of an insolvent bank can hardly be justified on systemic, euro-wide grounds if the decisions are based on national information and considerations alone. Finally, there is a problem of incentive compatibility in the use of public funds. For instance, national authorities would be reluctant to use public funds to bail out the parent of a pan-European banking group that conducts a large part of its activities abroad. Rather, there would be the incentive to shift the cost of the bailout to the rest of the Community. One way to do this would be to resort to the deposit insurance schemes of the other countries. Another way to shift the burden to the Community is to resort to the inflation tax, or the reduced seigniorage, which in monetary union is paid by all citizens. Central bank liquidity is a very powerful instrument for addressing financially critical situations. It produces two main effects. The first is a quantity effect, as the increase in central bank money enhances financial institutions' liquidity position. The second is a price effect, as the lower level of interest rate induced by the monetary expansion increases asset prices. Both effects improve the balance sheet position of financial institutions, thereby counteracting financial shocks. The fact that monetary policy can be a rather powerful instrument for addressing problems experienced by the financial sector raises at least three questions.

Who takes care of financial stability?

43

First, there is the temptation to entrust monetary policy with the task of safeguarding the financial soundness of the banking system. If this temptation is yielded to, there is the risk that price stability, which is normally the priority for a central bank, would be jeopardized.12 The reason is that financial stability and price stability do not necessarily require the same type of monetary policy. The Federal Reserve, for instance, was under severe pressure to abandon its restrictive policy in the early 1980s, because of the fragile financial conditions of the US banking system.13 The Bank of England was reluctant to increase interest rates to defend the pound in the EMS during the September 1992 crisis, because of severe problems that such a policy would have created for some UK banks.14 The second problem that may occur if monetary policy also aims at financial stability is the interference with the functioning of market mechanisms. The way central bank liquidity is created may have an impact on the functioning of the money market. Monetary policy is generally conducted through transactions, such as repurchase agreements, that attempt to avoid interfering with market mechanisms.15 Central bank intervention to bail out a bank, however, causes substantial distortionary effects. Even if the intervention is sterilized, re-distributive effects occur across all market participants. If central bank money is allocated directly to one or a few credit institutions, possibly at a subsidized rate, the remaining amount of liquidity has to be allocated at a higher rate than otherwise would be the case. The other market participants are indirectly called upon to pay for the subsidy to the institutions in difficulty. Thus, the provision of liquidity in periods of crises raises the question of equal treatment of market participants. A further source of interference arises from the fact that if a central bank is engaged in bailout operations, its capital base may be undermined. If each NCB decided independently on such operations in a decentralized way, there would be an excessive incidence of rescue operations. Over time, the soundness of the Eurosystem balance sheet would deteriorate, with serious repercussions for the credibility of the currency.

44

Open Issues in European Central Banking

In monetary union, the responsibility for the conduct of monetary policy, and thus for the inflation tax, is in the hands of the ECB, which is independent from the national authorities. It is thus not easy to convince the ECB to expand monetary policy with a view to increasing the inflation tax so as to bail out one or a few banks located in one country. It may nevertheless be questioned whether the ECB is really protected from pressures coming from other policy authorities to step in to address solvency problems. The Treaty attributes to the ECB the task of conducting monetary policy, with the primary objective of price stability. Neither financial stability nor prudential supervision is mentioned as the ECB's main responsibilities. NCBs, however, which are part of the Eurosystem, are permitted to exercise other tasks, such as being a fiscal agent for the government or managing domestic and foreign assets that are not part of the regular monetary policy operations.16 NCBs can thus in principle be involved in the bailout of banks, as long as this does not interfere with the conduct of the single monetary policy. Several NCBs are allowed to perform this function. Financial stability is inscribed in their Statute as one of their primary objectives. In this capacity, they may use central bank money to bail out banks. The fact that the responsibilities of the Eurosystem are limited to monetary policy, while banking and financial policies are excluded, can be viewed as an opportunity for NCBs to maintain an active role to pursue financial stability. We turn to this issue in the next section. The third instrument to address solvency crises, the involvement of the banking system, faces the difficulty that no European policy authority has been given responsibility for organizing lifeboat operations at the European level.17 Such an institution is necessary if the bank in difficulty conducts Europeanwide business and is too large to be helped by banks of one country alone. In fact, the continuation of national practices, by which foreign banks are often kept away from lifeboat operations, is not in line with European competition rules.

Who takes care of financial stability?

45

In sum, the traditional instruments available for managing financial crises have become more difficult to use under EMU. In addition, national authorities often do not have sufficient information on the overall nature and the systemic extent of a crisis. Furthermore, they do not necessarily feel any incentive to use the most appropriate instrument for their motivation is to try to shift the burden to the rest of the Community. If a crisis emerges, pressure will be placed on the Eurosystem to bear the burden, through the creation of central bank liquidity. The lack of clarity surrounding the role of lender-of-last-resort increases the possibility for such pressures. The lender-of-last-resort

in monetary union

The literature has examined the conditions under which the central bank should intervene in the face of a financial crisis. The doctrine developed by Bagehot (1873) can be summarized along the following principles: the central bank should provide liquidity, even in emergency situations, to the whole financial system, but only to solvent institutions, at penalty rates and against good collateral. These principles should be made known to the public, with a view to avoiding moral hazard problems.18 Two prescriptions are clear from this doctrine. The first is that the central bank has to be involved in the lending to the banking system as a last resort to avoid liquidity crises. The central bank cannot shirk from this responsibility. Central bank money is the ultimate redeemable payment instrument in fiat money systems that everyone accepts, even in crisis situations. Entrusting another institution with the task of providing emergency lending, that is, providing central bank liquidity, would create an interference with the conduct of monetary policy. The second clear prescription is that the central bank should not create liquidity with a view to bailing out any specific bank. Solvency problems, if they have to be solved, should be tackled through other instruments, such as lifeboat operations or the re-capitalization of the bank with public funds, as mentioned above. The central bank may help in the use of such instruments, but should not ultimately be responsible. In sum, the

46

Open Issues in European Central Banking

lending-of-last-resort function should not be confused with, nor lead to, the bailout of insolvent banks. In monetary union, the two above prescriptions suggest that the Eurosystem should be the lender-of-last-resort but should not deal with solvency crises. The fact that NCBs are not prevented from being involved in the rescue of financial institutions, however, may seriously interfere with the smooth conduct of monetary policy. This problem has been overlooked by the drafters of the Treaty, who thought that by restricting its objective to price stability, the Eurosystem would suffer no interference from other policy objectives. The result is that some parts of the Eurosystem may adopt policies aimed at financial stability in an uncoordinated way and with possible repercussions on the single monetary policy. At this stage, the only possible way to reduce interference is to ensure, for instance through an intergovernmental agreement, that central bank money is not used, in any EMU country, to bail out banks. Let us turn now to the three principles contained in Bagehot's doctrine. The first concerns the instrument to be used to provide emergency lending. Bagehot suggests providing liquidity to the whole financial system. This recommendation may be interpreted as using standard open market operations to inject liquidity into the financial system.19 According to this view, a central bank should not provide single institutions with privileged access to central bank money, not even in a crisis situation. The reason is that the central bank generally does not know any better than the rest of the market whether the bank experiencing liquidity problems is solvent or not. The market should itself allocate liquidity among its participants in the most efficient way. The problem with this view is that if liquidity is provided through market mechanisms, it may not ultimately get to the institution experiencing difficulties. Liquidity crises emerge because of asymmetric information about the balance sheet situation of a bank, which leads the other banks not to lend to it. Furthermore, the crisis, although affecting one or a few institutions, may rapidly spread to the whole system. The asymmetry of information creates

Who takes care of financial stability?

47

an externality, which prevents the market from efficiently allocating funds. To prevent contagion, the central bank has to intervene at an early stage to directly provide the bank in difficulty with the necessary liquidity to overcome the crisis.20 This policy rule has been followed for instance by the Federal Reserve to tackle the problems faced in 1985 by the Bank of New York, which was solvent but unable for technical reasons to get hold of the necessary liquidity to balance its positions. The short-term loan by the Fed was reimbursed as soon as the technical problem was resolved. Concerning the Eurosystem, it is not clear what instrument it may use to inject central bank money into the euro money market so as to address liquidity problems faced by one or a few banks. No special instrument has been devised and market participants are not aware of any that they may use in case of difficulty, and what are the conditions for accessing such an instrument. The second of Bagehot's principles is that the central bank should inject liquidity through standard monetary policy instruments and procedures, in particular against good collateral. This policy is difficult to implement in practice. In a crisis situation, standard open market operations can hardly be used. A crisis situation is generally characterized by the fact that normal eligibility criteria are not met by financial institutions. The latter, even if solvent, may not have at their immediate disposal a sufficient amount of collateral to bid for central bank liquidity. If the central bank wants to provide liquidity to the banking system, it has to modify, at least temporarily, some of the eligibility criteria, such as the range of collateral or the margin requirements. Even the range of counterparties eligible for central bank credit may have to be extended in a crisis situation, with a view to ensuring the highest possible liquidity of financial instruments. Such measures were implemented for instance by the Federal Reserve during the 1987 stock market crash. The difference between lending in normal times, through standard monetary policy instruments and procedures, and lending under emergency or crisis situations is clear. Appropriate procedures are generally put in place to organize the latter, although they may

48

Open Issues in European Central Banking

not be revealed to the outside world for purposes of avoiding moral hazard. It is not known whether the Eurosystem has actually put these procedures in place. The Statute of the ESCB and of the ESC defines very precisely the operations that can be conducted to provide liquidity to the market. A clear requirement is that all credit operations have to be made against adequate collateral. In theory, it is not possible for the Eurosystem to grant uncollateralized credit, as the Federal Reserve may do. The ECB Governing Council determines the range of eligible collateral. It can decide to broaden the range of eligible assets in order to facilitate access to emergency lending by institutions experiencing difficulties. Alternatively, the risk control measures applied by the ECB may be relaxed. Appropriate procedures have to be adopted to enable the ECB Governing Council to respond swiftly to a crisis, or to delegate power to the Executive Board. The third principle set by Bagehot for lending in emergency situations is to lend only to solvent institutions. How can the central bank know that an institution is solvent? Either it finds out for itself or it trusts someone to make the assessment. The problem in the latter instance is how to ensure that the necessary incentives to provide the correct information are present. One incentive might be to require the source of information to 'put his money where his mouth is'. If the central bank is responsible for banking supervision, and thus for assessing banks' solvency, it should incur the costs of possible errors of judgement in assessing the solvency of a particular bank.21 If another institution is in charge of assessing a bank's solvency, separate from the central bank, it should be required to put up the extra collateral required to access central bank credit. This is the case in countries where prudential supervision is exercised by a government agency that is separate from the central bank. In Germany, for instance, a bank that does not have sufficient collateral to access the normal credit operations with the central bank may address itself to the Liquidity Consortium, whose capital is financed by banks, the State and the Bundesbank, to borrow the necessary

Who takes care of financial stability?

49

collateral. The bank in difficulty uses the collateral to obtain credit from the central bank. Who (in the Eurosystem) should lend in the last resort? If one or few banking institutions experience a liquidity shortage in the Eurozone, the decision to provide liquidity can hardly be taken only by the NCB of the country where the banks in difficulty are located. This is true for several reasons. First, in an integrated European financial market, an individual NCB is not in a position to assess the systemic implications of a crisis that first emerges at the national level, for instance in the local payment system. The system for exchanging information between supervisory authorities, as currently designed, is not adequate to give a full picture of the problem at the European level. Even when the liquidity crisis starts in a bank of secondary importance, it may have an impact on the whole euro money and financial market. Second, as mentioned above, the provision of central bank money to one or a few institutions affects the distribution of funds across the whole banking system, even if its overall liquidity impact is sterilized. If a sum of liquidity is allocated to some banks outside the regular conduct of monetary policy, the other banks will have to pay a higher rate of interest than otherwise would be the case. A decentralized system of liquidity assistance cannot ensure equal treatment across the banking system. A minimum requirement is that the terms and conditions under which emergency lending is granted, such as the rate of interest, the type of collateral and the margin requirements, are harmonized within the Eurosystem. We are not aware that such harmonization has taken place and that the differences prevailing prior to monetary union have been eliminated. The US experience shows that liquidity support to banks in difficulty may be much larger than expected. In the Bank of New York crisis in 1985, the Fed had to provide in a few hours an amount of uncollateralized overnight credit equivalent to more than 10 per cent of the US money base. The sterilization of such an operation cannot be considered a trivial operation.

50

Open Issues in European Central Banking

In the Eurozone, NCBs may be tempted to provide liquidity assistance to banks experiencing solvency, rather than liquidity, problems. Such a temptation may be relevant for NCBs that are also in charge of banking supervision, on grounds that solvency and liquidity problems can hardly be distinguished.22 Decentralized lender-of-last-resort induces a potential competition for excessive lending, with disastrous effects in terms of moral hazard. The decision to provide emergency lending to one or a few local banks cannot be taken by the ECB alone either, since the latter is not in a position to assess their solvency. The ECB may rely on the supervisory expertise of national authorities. Two problems nevertheless emerge in this case. The first is that national authorities have until now refused to provide the ECB with prudential information on a systematic basis. The second is that even if the latter agree to provide the necessary information, the ECB may not be fully satisfied with it. Indeed, national authorities have an incentive to falsely declare that a bank is solvent, with a view to enabling it to obtain temporary assistance from the Eurosystem. This would be a way for insolvent banks to be bailed out through the LOLR operation, whose costs would be borne by the entire Community. The incentive to shift the burden of the rescue effort from the national to the Community level could be reduced if the risk of insolvency was borne by the national supervisory authority. This might nevertheless not represent a sufficient deterrent against biased information. Even if NCBs were to bear the burden of the bailout, the weakening of their financial position would inevitably affect the soundness of the overall Eurosystem balance sheet, with repercussions for the soundness of the currency. A reasonable solution might be that the national Treasuries provide the collateral.23 Further work should be conducted in this area. In summary, there is an inconsistency between the performance of the lender-of-last-resort function and the way in which banking supervision is organized in monetary union. To restore consistency, either the LOLR is exercised in a decentralized way, with all the problems mentioned above, or the responsibility for prudential supervision is transferred to the Community level and the

Who takes care of financial stability?

51

ECB has access to the relevant information. Although the latter solution appears more rational, it may not necessarily be the one immediately chosen by European authorities. The Eurosystem may thus be faced with sub-optimal solutions in the short and medium term. In the longer run, as greater supervisory responsibilities are likely to be transferred to the Community level, the problem arises of whether the latter should be exercised by the Eurosystem or by a separate European institution. A European solution does not necessarily entail full centralization of both decision-making and implementation. Only the former should be envisaged, as in the case of monetary policy, while implementation can, to the extent possible, be decentralized. 2.3

Should the Eurosystem powers?

be granted

supervisory

The question of whether a central bank should be responsible for both banking supervision and monetary policy or whether, on the contrary, the two tasks should be exercised by separate institutions, has recently received renewed attention, following in particular the reforms of the Central Banks of Australia and England. If the need arises to transfer greater responsibility in the field of banking supervision to the European level, the problem of whether this task should be entrusted to the ECB or to a separate, possibly new, European institution has to be tackled. The question of whether or not prudential supervision interferes with the conduct of monetary policy is crucial in this debate. The literature has examined the advantages and disadvantages of combining the two functions.24 The main disadvantage is that the performance of banking supervision may interfere with the conduct of monetary policy. If the central bank is responsible for banking supervision, and is in particular concerned with the stability of financial institutions, it may be tempted to create central bank liquidity with a view to avoiding that one or more financial institutions in difficulty would experience losses that could trigger financial instability. This situation may occur especially if the underlying macroeconomic conditions require the central bank to

52

Open Issues in European Central Banking

tighten monetary policy, for instance to counter inflationary pressures. This issue was examined in Section 2 above. The second type of problem arises from the fact that in the case of major bank rescue efforts, the government is bound to play some role, given the impact on taxpayers' money.25 Since the different instruments to address the crisis entail different costs for different sectors of society, the choice of the instrument assumes a political connotation from which a central bank should stay away.26 Finally, negative experiences in the resolution of banking crises may jeopardize the credibility of the central bank, even with respect to its monetary policy responsibilities. Goodhart emd Schoenmaker (1993) suggest, for instance, that the credibility of the Bank of England has been impaired after the BCCI crisis. Market participants may interpret the emergence of crises as a failure of the supervisory capacity of the central bank. The latter may thus be induced to cover up such a failure through the injection of central bank liquidity. If market participants anticipate that central banks with supervisory responsibilities may be more prone to use the monetary levy, the problem of moral hazard may be exacerbated. Moral hazard may also increase as a result of the lack of transparency arising from the fact that the central bank has two objectives which are not clearly distinguished and which may at times conflict.27 On the other hand, central banks have in general been keen to establish Chinese walls in their internal organization between the departments in charge of monetary policy and prudential supervision, although these walls are not always well known to the public. The main argument for combining the functions of monetary and supervisory management within the central bank is that the latter facilitates the performance of the lender-of-last-resort function.28 As mentioned in the previous section, the central bank has to be able to assess the solvency of financial institutions, to avoid lending to insolvent institutions. If the central bank does not have access to micro-prudential data, the lender of last resort function is difficult to perform and mistakes can easily be made.

Who takes care of financial stability?

53

There are at least two possible objections to this argument. The first is that it is very difficult, even for supervisory authorities, to effectively assess the solvency of institutions. In most cases, a bank experiencing liquidity problems is also likely to experience some solvency problems.29 There is in general little time to examine the books of the supplicant borrower in a sufficiently careful way to tell whether the bank is indeed solvent or not. The second objection is that the rule of lending only to solvent institutions becomes less important in a world in which the central bank can only lend against collateral. In this case, the central bank can always recover its loan by selling the collateral if the counterpart fails. If the loan is not collateralized, the solvency of the institution is crucial for the safeguarding the central bank. In situations of crisis, the bank may not have access to sufficiently good collateral. The question can then be reformulated in terms of availability of collateral rather than solvency of the financial institution. In case of difficulty, the latter may get the collateral from someone else, for instance a government agency, which would be ready to intervene and provide the bank with good collateral, even if the latter ran into solvency problems. In Germany, for instance, the Bundesbank does not need to make an assessment of the solvency of a bank experiencing a liquidity shortage, as long as the latter can get some form of collateral from the Liquidity Consortium and access the special refinancing window. A similar instrument could be used at the national or European level to enable banks in liquidity shortage to borrow the necessary collateral. This would entail the creation of a lender of penultimate resort that would provide the bank in difficulty with the collateral to access central bank credit.30 Such a lender would clearly need to have adequate information about the solvency of the bank in difficulty. The existence of a lender of penultimate resort, to be agreed and harmonized at European level, would enable the separation of supervisory from monetary policy responsibilities. Another argument which is put forward in favor of combining the two functions within the central bank is that the exercise of prudential supervision provides relevant information for the conduct

54

Open Issues in European Central Banking

of monetary policy. Applied at the European level, this position would entail that prudential supervision be exercised by the Eurosystem. The experience of countries where the two functions are separate suggests, however, that monetary policy can be conducted independently from prudential supervision. The relevant information for the conduct of the former pertains to macro-prudential data, which is generally collected by all central banks (whether or not they are responsible for financial supervision), rather than microprudential data, which are relevant for the supervision of financial institutions. It is interesting to note that there is as yet no euro-wide macro-prudential data available to the public, and perhaps not even to the ECB. The empirical evidence comparing the relative performance of systems where the central bank exercises supervisory responsibilities with those where a separate authority exercises supervision is scarce. Goodhart and Schoenmaker (1993) examine the issue from the point of view of the effectiveness of prudential supervision in avoiding banking crises. On the basis of a sample made of 104 banking crises, the authors conclude that there is no evidence that one system has performed better than the other. Concerning the inflation performance of the two systems, Heller (1991) finds that countries where the central bank is also responsible for banking supervision have recorded on average a significantly higher rate of inflation. Goodhart and Schoenmaker (1993) obtain a similar result. These authors suggest, however, that this result may not be considered as a proof of a conflict of interest between monetary policy and banking supervision, but possibly of the fact that central banks with combined responsibilities are generally less independent. The correlation between the supervisory responsibility of the central bank and inflation performance would thus be a proxy of the (negative) relation between central bank independence and inflation.31 The annex at the end of this chapter examines whether the involvement of the central bank in supervisory activities has a significant impact on inflation performance, independently from the degree of central bank independence. The empirical results show that there is indeed a positive and statistically

Who takes care of financial stability?

55

significant relationship between the exercise of supervisory responsibilities and inflation performance in industrial countries. The result is robust to the use of different indicators of central bank independence and supervisory involvement. The exercise of prudential supervision by the central bank has instead no significant effects on growth performance or variability. These results have to be taken with care. They cannot be interpreted as evidence of a direct causality relation between the exercise of banking supervisory and inflation performance. 32 Moreover, these results cannot be used to draw immediate policy conclusions. Rather, they should be seen as a stimulus for further research, with a view to improving the understanding and assessment of the interaction between the task of conducting monetary policy and the exercise of responsibilities in the field of banking supervision. These results suggest, however, that although greater supervisory powers might have to be transferred to the European level, the Eurosystem is not necessarily the best institution to entrust with such responsibility. Any decision in this area has to take into account a broad set of considerations, also from a politico-economic point of view. It is often mentioned that the allocation of Europeanwide supervisory responsibilities to the ECB would have the advantage that the latter already exists as an institution. The costs and political difficulties of creating a new European institution would be avoided. Another consideration is that the transfer of responsibilities from the national to the European level is unlikely to take place in a quantum jump, but rather in a progressive way. In this context, the ECB solution may seem more promising than the creation of an empty shell to be progressively filled up with increasing responsibilities relinquished by national authorities. Conclusions The institutional framework designed to promote financial stability in Europe is inefficient and internally inconsistent. The inefficiencies and inconsistencies have increased with the move to monetary union. Furthermore, the instruments to face systemic crises have weakened.

56

Open Issues in European Central Banking

Should financial instability arise, the burden of dealing with its consequences would fall on the ECB, despite its being insufficiently equipped for this task. On the one hand, the ECB does not have the appropriate instruments and procedures to face liquidity crises. And on the other, it runs the risk of being exposed to strong pressures to bail out financial institutions in difficulty, a task it should instead stay away from. The resulting ambiguity, which does not seem constructive, may lead to strong interference in the regular conduct of monetary policy. Moral hazard is inevitably bound to increase. A worrying aspect of our diagnosis is that the conclusions are widely shared by academics, market practitioners and, surprisingly, by policy-makers and even by some central bankers. There is a generalized reluctance, however, to push the analysis a step further and to make concrete proposals for solving the problems. The political difficulties involved in any change to the status quo appear to most observers to be so formidable as to intimidate the proponents of even the most timid proposals Few economists would disagree with the conclusion that the present set-up for supervision and regulation is not viable in the long run, once financial markets have become integrated in Europe. It is often argued, however, that there is no need for radical change now because there are no signs of financial stability on the horizon. Moreover, the present system could evolve in the right direction should any problems arise. We would certainly agree that there is no need to elevate all responsibility for financial stability to the European level, but the strong resistance to even minimal adjustments to the status quo suggests that important vested interests will oppose any evolution. The 'learning by doing' approach that has worked well in other areas is not appropriate in the financial area. Problems can build up under a surface of apparent calm, and when they break into the open, the cost of dealing with them can become very large. Reforms should therefore start now. The risks of centralizing a bit too early seem well worth taking when compared to the high cost that might arise if the system is reformed too late. The sense of frustration with the current situation leads to the consideration, often whispered at conferences or seminars, that only

Who takes care of financial stability?

57

a major financial crisis can really push European policy authorities to redesign the map of responsibilities in the field of prudential regulation and supervision and to relinquish at least part of their national powers. After all, a major political impetus for monetary union was supplied by the EMS crises, which convinced political authorities of the impossibility of pushing informal monetary policy cooperation much further. The only hope can be that the learning process is quicker this time and that the damages from the crises can be contained.

58

Open Issues in European Central Banking Annex 2.1 Does prudential supervision interfere with monetary policy?

Empirical evidence on the impact of supervisory responsibility on monetary policy is scarce. Heller (1991) compares the inflation performance of groups of countries, depending on whether the respective central bank exercises banking supervisory responsibilities. He finds that in countries where the central bank is also responsible for banking supervision, a significantly higher rate of inflation has been recorded on average. Goodhart and Schoenmaker (1993) obtained a similar result. These authors suggest, however, that this result may not be considered as a proof of a conflict of interest between monetary policy and banking supervision, but possibly the fact that central banks with combined responsibilities are generally less independent. The observed correlation would thus be a proxy of the observed (negative) relation between central bank independence and inflation.33 However, de Haan and van't Hag (1995) show that the hypothesis that central banks that do not regulate financial institutions are more independent than those where the two responsibilities are combined can be rejected. They find that the correlation between the extent to which the central bank exercises prudential supervision and the degree of central bank independence, measured according to different criteria, is negative but not significant. This would suggest that there is a relation between inflation performance and the exercise of supervisory responsibilities by the central bank, aside from the degree of independence of the latter. No formal test has been conducted, however. We examine this issue by considering 21 industrial countries. Based on the analysis of the supervisory regime conducted by Goodhart and others (1998), we identify nine countries in which, in the period 1974-90, the central bank had clear responsibility over banking supervision: Australia, Greece, Ireland, Italy, the Netherlands, New Zealand, Portugal, Spain and the United Kingdom.34 In the others, the responsibility is either attributed to a

Who takes care of financial stability?

59

separate institution or shared. We then create a dummy variable SUPi for the above countries where the central bank has supervisory responsibility. We then run a regression to explain inflation performance in the 1974-90 period, using SUPi as the only explanatory variable. This corresponds to a test of the hypothesis that inflation performance over a long-term horizon is fiilly determined by central bank action and that the latter is fully dependent on whether it also performs supervisory responsibilities. This is clearly a restrictive hypothesis, but useful nevertheless as a first step in examining the issue. The result of the ordinary least square regression is: INF1974-9o = 6.53 + 5.75 SUP R2 = .421 SE = 331 (6.94) (3.94) The values in parenthesis are t-statistics. The coefficient on the SUPi variable is positive and significantly different from zero, suggesting that countries where the central bank had supervisory responsibilities recorded a significantly higher inflation than the others did. This result is consistent with that of Heller (1991) and Goodhart and Schoenmaker (1993). In the sample period, the nine countries where the central bank had responsibility over prudential supervision recorded an average inflation rate of 12.38 per cent (with a standard deviation of 4.53 per cent) against an average of 6.63 per cent (1.98 per cent standard deviation) for the other 12 countries. This result is unaffected if different indicators of the degree of central bank involvement in banking supervision are used, that allow, for example, for the distinction between cases in which banking supervision is exercised exclusively by a completely separate institution and those in which the competence is shared with the central bank, as proposed by Grilli, Masciandaro and Tabellini (1991). Using SUP2 in the above regression (where countries in which the central bank has exclusive responsibility over banking supervision have a value of 1, those in which the responsibility is shared with another institution a value of 0.5, and those in which it is exercised exclusively by a completely separate institution a value of 1), instead of SUPi, one obtains:

60

Open Issues in European Central Banking

INF1974-9o = 6.65 + 5.12 SUP2 R2 = .256 SE= 3.75 (5.58) (2.81) These results are subject to the criticism mentioned above, according to which the positive correlation between the degree of central bank involvement in prudential supervision and inflation performance is spurious. It only reflects the correlation between the former and central bank independence, which is found in the literature to be negatively correlated with inflation. To assess whether the exercise of prudential supervision by the central bank has a significant impact on inflation performance, one should thus control, in the above regression, for the degree of central bank independence. We use three indices of central bank independence (CBI): GMT produced by Grilli, Masciandaro and Tabellini (1991); CUK, produced by Cukierman (1992);35 and ES, produced by Eijffinger and Schaling (1993).36 A second version of the CBI index produced by Grilli, Masciandaro and Tabellini (MGMT) is also constructed, excluding the extra rating that the respective central banks were allocated by these authors in case they did not have competence over banking supervision.37 Both SUPi and SUP2 are negatively, but not perfectly correlated with the indices of CBI, as found by de Haan and van't Hag (1995). We perform the same regressions as before, adding the variable related to CBI, measured according to the four indices mentioned above. Table 2.A.1 reports the results of the regressions. The coefficients on the CBI indices all have a negative sign and are significantly different from zero, except for CUK, in line with the results found in the previous literature. These results hold when the index of CBI is used as only independent variable and when SUPi or SUP2 are added. The coefficients on the SUP variables are positive and significantly different from zero. The value of the coefficient is not significantly affected by the introduction of the CBI variable, except when GMT is used, as could be expected since the latter is calculated taking into account the participation of the central bank in supervisory activities.

Who takes care of financial stability? Table 2.A.1 CBI index used

GMT

MGMT

CUK

ES

61

Results of regressions c

CBI

SUPi

18.03 (10.59)

-1.187 (-5.61)

14.64 (6.89)

-.907 (-4.01)

16.11 (8.30)

-1.083 (-3.89)

12.66 (7.49)

-0.852 (-3.95)

4.54 (3.98)

13.50 (8.22)

-1.025 (-4.79)

-

11.52 (4.76)

-0.66 (-1.88)

8.077 (3.86)

-.037 (-.781)

5.56 (3.72)

8.87 (3.87)

-.059 (-1.13)

-

15.01 (8.91)

-2.366 (-3.89)

11.37 (5.85)

-1.581 (2.69)

4.70 (3.54)

12.13 (5.72)

-1.85 (-2.93)

-

SUP2

2.97 (2.31)

4.69 (3.77)

5.02 (2.77)

4.18 (2.66)

R2

SE

.603

2.73

.677

2.47

.444

3.32

.672

2.49

.695

2.95

.009

4.32

.409

3.34

.267

3.72

.414

3.32

.564

2.86

.469

3.16

Note: The figures in parenthesis are t-statistics.

We also examined whether the exercise of banking supervision, as measured by SUPi or SUP2, has any significant impact not only on inflation performance but also on other variables. We ran regressions with the same independent variables as in Table 2.A.1, with a view to explaining the rate of growth of GDP in the sample period and its variability. The coefficients were never found to be significantly different from zero.

62

Open Issues in European Central Banking References to Chapter 2

Bagehot, W. (1873), lombard Street, Reprint R. Irwin, Homewood 111., 1962. Begg, I. and D. Green (1995), Banking Supervision in Europe and Monetary Union, European Institute, South Bank University, London. Cukierman, A. (1992), 'Central Bank Strategy, Credibility and Independence: Theory and Evidence', MIT University Press, Cambridge Mass. De Bandt, 0. (1998), 'EMU and the Structure of the European Banking System', mimeo, European Central Bank, Frankfurt. De Beaufort Wijnholds, J. and L. Hoogduin (1994), 'Central Bank Autonomy: Policy Issues' in De Beaufort and Hoogduin (eds.), A Framework for Monetary Stability, pp. 75-95. De Haan, J. and G van't Hag (1995), 'Variation in Central Bank Independence across Countries: Some Provisional Empirical Evidence', Public Choice, No. 85, pp. 335-351. Eijffinger, S. and E. Schaling (1993), 'Central Bank Independence in Twelve Industrial Countries', BNI Quarterly Review, No. 192, pp. 39-83. Eijffinger, S. and J. De Haan (1996), 'The Political Economy of Central Bank Independence \ Special Papers in International Finance, No. 19, Princeton University Press. European Monetary Institute (1997), The Single Monetary Policy in Stage Three, Frankfurt. Folkerts-Landau, D. and P. Garber (1992), 'The ECB: A Bank or a Monetary Policy Rule', in Establishing a Central Bank: Issues in Europe and lessons from the US, in Canzoneri, Grilli and Masson (eds.), Cambridge: Cambridge University. Press. Friedman, B. (1986), 'Increasing Indebtedness and Financial Stability in the United States', in Debt, Financial Stability and Public Policy, Federal Reserve Bank of Kansas City. Giovannini, A. (1993), Central Banking in a Monetary Union: Reflections on the Proposed Statute of the European Central Bank", Carnegie Rochester Conference Series on Public policies, No. 38, pp. 191230.

Who takes care of financial stability?

63

Goodfriend, M. and R. King (1988), 'Financial Deregulation, Monetary Policy and Central Banking', Federal Reserve Bank of Richmond Quarterly Review, Vol. 74, No. 3, pp. 2-22. Goodhart, C. (1985), The Evolution of Central Banks, London School of Economics, London. Goodhart, C. and D. Schoenmaker (1993), Institutional Separation between Supervisory and Monetary Agencies, LSE Financial Markets Group, Special Paper No. 52, London. Goodhart, C , P. Hartmann, D. Llewellyn, L. Rojas-Suarez and S. Weisbrod (1998), Financial Regulation: Why, How and Where Now?, Routledge, London. Grilli, V., D. Masciandaro and G Tabellini (1991), 'Political and Monetary Institutions and Public Financial Policies in the Industrial Counties', Economic Policy, No. 13, pp. 341-392. Group of Thirty (1997), Global Institutions, National Supervision and Systemic Risk - A Study Group Report, Washington, D.C. Heller, R. (1991), 'Prudential Supervision and Monetary Policy', in J. Frenkel and M. Goldstein (eds.), International Financial Policy: Essays in Honor of J. Pollak, International Monetary Fund, Washington, D.C. Humphrey, T. and R. Keleher (1984), 'The Lender of Last Resort: A Historical Perspective' The Cato Journal, Vol. 4, No. 1, pp. 275318. Kaufman, H. (1986), 'Debt: The Threat to Economic and Financial Stability', in Debt, Financial Stability and Public Policy, Federal Reserve Bank of Kansas, pp. 15-26. Lannoo, Karel (1999), 'EMU, the ECB and Financial Supervision', Zentrum fur Europaische Integrationforschung (ZEI), Working Document. Melitz, J. and C. Bordes (1991), 'The Macroeconomic Implications of Financial Deregulation', European Economic Review, No. 35, pp. 155-178. Mishkin, F. (1994), Preventing Financial Crises: An International Perspective, NBER Working Paper No. 4636. Padoa-Schioppa, Tommaso (1999), 'Banking Supervision in EMU", speech given at the London School of Economics, 24 February 1999.

64

Open Issues in European Central Banking

Prati, A. and G Schinazi (1998), 'The ECB and Financial Stability in EMU', mimeo, International Monetary Fund, Washington, D.C. Swinburne, M. and M. Castelo-Branco (1991), 'Central Bank Independence and Central Bank Functions', in P. Dowes and R. Vaez-Zadeh (eds.)., The Evolving Role of Central Banks, International Monetary Fund, Washington, D.C. Vives, X. (1991), 'Banking Competition and European Integration', in C. Mayer and A. Giovannini (eds.), European Financial Integration, Cambridge University Press, Cambridge.

Notes to Chapter 2 1. Some have suggested that the task of promoting the smooth operation of the payment system attributed to the ESCB (Article 105.2 of the EC Treaty) is an indirect reference to financial stability. 2. The EC Treaty (Article 105.6) foresees that the Council may, acting unanimously on a proposal from the Commission and after consulting the ECB and after receiving the assent of the European Parliament, confer upon the ECB specific tasks concerning policies relating to the prudential supervision of credit institutions and other financial institutions with the exception of insurance undertakings. 3. See, for instance, Goodhart et al. (1998). 4. In October 1998, the European Commission produced a Communication on Financial Services: Building a Framework for Action, which underlines the main areas where further progress is needed. 5. See, for instance, De Bandt (1998) 6. See Goodhart etal. (1998). 7. This deficiency is well recognized at the international level (Group of Thirty, 1997). 8. See Vives (1991).

Who takes care of financial stability?

65

9. See also Giovannini (1993). 10. See Begg and Green (1995). 11. Goodhart and Schoenmaker (1993) examine 104 cases of bank crises and examine common patterns in the rescue operations. 12. Melitz and Bordes (1991) formalize this argument. 13. See Kaufman (1986) and Friedman (1986). 14. See Goodhart and Schoenmaker (1993). 15. This is the case of the ESCB. See, for instance, EMI (1997). 16. See Chapter 1. 17. Prati and Schinazi (1998) examine various possible scenarios and suggest that national Treasuries may be called to play a greater role in this field. 18. See also Humphrey and Keleher (1984). 19. See Goodfriend and King (1988). 20. See Folkerts-Landau and Garber (1992). 21. Goodhart and Schoenmaker (1995) find that central banks that exercise prudential supervision are often involved in the rescue of failed banks. 22. Goodhart and Schoenmaker (1993) provide some examples. 23. Although this might pose a problem with respect to the prohibition of monetary financing. 24. See in particular Heller (1991), Goodhart and Schoenmaker (1993, 1995) and de Beaufort Wijnholds and Hoogduin (1994). 25. This argument is made in particular by de Beaufort Wijnholds and Hoogduin (1994). 26. See Heller (1991) and Lannoo (1999).

66

Open Issues in European Central Banking

27. See Swinburne and Castelo-Branco (1991). 28. See Humphrey and Keleher (1984). 29. See Goodhart (1985) and Padoa-Schioppa (1999). 30. See Prati and Schinazi (1998). 31. This interpretation is also hinted at by Eijffinger and de Haan (1996). 32. Heller (1991) cites the case of the Netherlands, where the central bank also exercises prudential supervision but has recorded a very good inflation performance. The latter may however largely be due to the exchange rate policy, as suggested by de Beaufort Wijnholds and Hoogduin (1994), through which the Netherlands has 'imported' Germany's anti-inflationary policy. 33. This interpretation is also hinted at by Eijffinger and de Haan (1996). 34. This classification is consistent with the one of Grilli, Masciandaro and Tabellini (1991), with the exception of Spain, which is classified by the latter as having joint but not full responsibility. 35. This corresponds to the index of legal independence, named LVAU by Cukierman (1992). 36. The results do not significantly change when the regressions are conducted on the same sample on which the original indicators were calculated. 37. The rating that each central bank was allocated can be obtained from the difference between GMT and MGMT. A score of 2 implies no responsibility at all for prudential supervision, 1 means a shared responsibility between the central bank and another institution, 0 entails full responsibility for the central bank.

3. ARE EUROPEAN CENTRAL BANKS OVER-CAPITALIZED?* Introduction The job of central banks is to conduct monetary policy. This consists of buying and selling assets from and to the private sector, in exchange for central bank money, a rare commodity used by the financial sector to clear its daily position in the money market. The conduct of monetary policy has a direct effect on the balance sheet of both the banking system and the central bank. Central banks are not profit-maximizing institutions. Their primary objective is to maintain price stability. Central banks do nevertheless care about their balance sheets, in particular their profit and loss account. The avoidance of losses is indispensable to preserve financial independence. Moreover, if a central bank makes losses, other agents, presumably market operators, make corresponding profits. Loss-making operations thus imply a subsidy for the private sector, which is in contradiction with the conduct of monetary policy in an open market economy with free competition. The Eurosystem is no exception. Its balance sheet is of interest for at least three reasons. The first is that the allocation of profits and losses within the Eurosystem is governed by principles that are different from those underlying the single monetary policy. This issue is discussed in Chapter 4. What is interesting to note here is that there are no automatic mechanisms for ensuring financial solidarity, within the Eurosystem.1 The second reason is that, in contrast to any other central bank in the world, the governing body of the Eurosystem does not have control over the whole balance sheet. Both the Maastricht

With Franziska Schobert. 67

68

Open Issues in European Central Banking

Treaty and the Statute of the ESCB and of the ECB make an artificial distinction between assets and liabilities held by central banks for the conduct of monetary policy and other assets and liabilities unrelated to monetary policy. The latter set of assets and liabilities, which constitutes the largest part of the overall balance sheet, remain under the responsibility of the NCBs and can be managed by them independently to the extent that this does not interfere with the conduct of the single monetary' policy as already mentioned in Chapter 1. The third issue concerns the overall size of the Eurosystem balance sheet. NCBs have accumulated large amounts of assets in the past.2 In monetary union most of these assets are not needed any more and these could be dispensed with. Proposals have been put forward in particular to use excess foreign exchange reserves to fund large-scale European initiatives. We start by briefly examining the balance sheet of the Eurosystem and then look at how it relates to that of the component NCBs. We then compare it to the balance sheets of two other G-3 central banks (the United States and Japan) to assess whether the Eurosystem is over-capitalized. Finally, we present a proposal for a 'lean' Eurosystem balance sheet, discussing the implications for both the assets and the liabilities. 3.1

The Eurosystem and its components

Constructing a detailed balance sheet for the Eurosystem from national data should in principle be easy. It should be sufficient to add all assets and liabilities and account for the limited pooling of reserves with the ECB (39.5 billion euro so far). Constructing the balance sheet of the Eurosystem is difficult, however, because the accounting procedures of most national central banks are so opaque that it is very difficult to find out exactly what kind of assets and liabilities lie behind the various items recorded on their balance sheets. Since certain accounting rules vary from one country to another, one cannot make a straightforward comparison of balance sheets across countries.

Are European central banks over-capitalized?

69

For example, the Banca d'ltalia carries securities that are a residual of its past links with the Italian Treasury. Its claims on the government had a book value of 39 billion euro at the end of 1997, representing about 20 per cent of its total assets. Considering its 1 per cent return and the maturity (the last tranche will become due in 2044), the market value of these assets, even at today's low interest rates, is only about 30 per cent of its face value. Another example is Finland. In accordance with the Act establishing the Bank of Finland, fixed assets, shareholdings and long-term expenditures are entered in full as expenditures in the year of acquisition and thus do not appear on the balance sheet. Although the absolute amount of these items is not so important in relation to total assets, it is just one more reason why central bank balance sheets are not easily comparable.4 The case of gold is another example. Since gold will never again be used actively in the international monetary system, it may actually make sense to avoid the continuous revaluation of gains and losses derived from the use of market prices. If gold is no longer to be used, one may ask why central banks should still carry it on their balance sheets. The decision by the ECB to hold gold for 15 per cent of its reserves is difficult to understand from this point of view. The ECB has decided to present its balance sheet and part of that of the Eurosystem according to the 'mark to market' principle.5 This should make its accounting transparent. Moreover, the ECB's accounting rules apply only to that part of the NCBs' balance sheets that relate to the common monetary policy. For the remainder, which is as large again (see below), the NCBs may use national rules. Since the 'national' part of the NCBs' balance sheets is so important, it is also impossible to construct a full 'mark to market' balance sheet for the Eurosystem. The only point about the balance sheet of the Eurosystem that is clear at present is that it has no financial implications. Article 26.3 of the Statute of the ESCB and of the ECB makes clear that 'for analytical and operational purposes, the Executive Board shall draw up a consolidated balance sheet of the ESCB, comprising those assets and liabilities of the NCBs that fall within the ESCB'.

70 3.2

Open Issues in European Central Banking A comparative G-3 perspective

Liabilities and size: A comparison with the US and Japan Table 3.1 lists the standard broad liability items for the Eurosystem, the US Federal Reserve and the Bank of Japan. The first item shows that the Eurozone, Japan and the US have a monetary base of approximately the same size (about 430 to 450 billion euro). The third item shows that the total liabilities of the Eurosystem amount to about 700 billion euro, more than one and a half times its monetary base. In the case of the Fed, the difference between the monetary base and total liabilities is not so large. Japan's situation is more similar to that of the Eurosystem, following extended "dual" operations which, according to comments published by the Bank: of Japan, were aimed at stabilizing market interest rates. The 1997 situation was more similar to that of the Fed. An index of central bank capitalization can be constructed as the ratio of total liabilities over the monetary base in percentage. A value close to 100 indicates a balance sheet that is determined only by monetary policy operations, i.e. the provision of monetary base.6 In terms of this indicator, the Eurosystem scores 162 against 119 for the US Federal Reserve. The Japanese value rose to 144 in 1998, from 120 in 1997, for the reasons mentioned above. The consolidated data for the Eurosystem hide large differences between NCBs. Table 3.2 describes the situation for the component NCBs in terms of their degree of capitalization. The Bundesbank, with 142, while still far from the US or the Japanese values, has one of the lowest (leanest) capitalization levels. The Banco do Portugal stands out at the other extreme, because its total liabilities are over three times greater than its monetary base. The explanation for Portugal's situation is that when reserve requirements were reduced from about 16 per cent to the current level, the Banco do Portugal did not allow commercial banks to dispose of the deposits they had, but transformed them instead into medium-term liabilities which will mature in 2004. At that time, the balance sheet of the Banco do Portugal will shrink considerably, unless these liabilities are rolled over.

Table 3.1

Central bank balance sheets (liability side) (billion euro) Eurosystem

1. Monetary base Banknotes in circulation Deposits of financial institutions 2. Other liabilities and capital accounts 3. Total liabilities and capital accounts Memo: Total liabilities over monetary base

US Federal Reserve

Bank of Japan

429 342 87 268 697

437 421 16 85 522

455 422 33 200 654

162 %

119 %

144 %

Notes: For the Bank of Japan, the item "government securities held in custody and borrowed" is excluded for greater comparability. Gold and foreign exchange reserves are valued at market prices. Sources: Consolidated opening financial statement of the Eurosystem, 1.01.1999; Financial statement of the Bank of Japan, 31.12.1998; Statement of condition of all Federal Reserve Banks, 30.12.1998. (Details may not add to totals because of rounding.)

-0

Table 3.2

Index of central bank capitalization

Germany 131.3

France 187.8

Italy 221.1

Spain 160.6

Austria 164.2

Ireland 154.6

Netherlands 154.5

Portugal 306.2

Finland 185.8

Eurozone 162.6

Belgium 215.8

Note: The index is constructed on the basis of the ratio of total liabilities and capital accounts over the monetary base in percentage. Sources: Financial statements of national central banks, December 1998; gold and foreign exchange reserves are valued at market prices..

&5

n to

I 2"

s 0q

Are European central banks over-capitalized?

73

Why did NCBs accumulate so much capital and other liabilities unrelated to monetary policy? In the 1970s and 1980s, most central banks were not independent. Inflation was high and monetary policy was conducted also through administrative measures, creating distortions for the national banking system. Prior to monetary union, NCBs in a number of member countries had special relations with banks and gave the public sector privileged financing. This was compensated for by large provisions to strengthen their capital base, thereby increasing the size of the central bank's balance sheet, compared to the monetary base. In the 1990s, the process came to an end, partly because of the liberalization of the banking system under the Internal Market program and ultimately because the Maastricht Treaty prohibited central bank financing of the public sector. NCBs have nevertheless kept the assets and liabilities they had accumulated. The accounting procedures for reserve provisioning and profit distribution have not been modified. The composition of the asset side: Foreign exchange assets A good way to examine the asset structure of a central bank balance sheet is to look at the ratio of foreign assets (and gold) to the monetary base.7 There is great variability among the NCBs within the euro area. The ratio of overall foreign assets to the monetary base ranges from a low of 39 per cent for Italy to over 277 per cent in the case of Portugal (Fig. 3.1). In general, as one would expect, the larger countries hold relatively less foreign assets than the smaller ones. In judging the low values of the United States and Japan, one has to keep in mind, however, that these countries hold part of their foreign exchange reserves at a separate foreign exchange office. While it was understandable why smaller countries held large amounts of foreign exchange reserves in the past, this reasoning no applies in the Eurozone. Hence there is no reason for great diversity in the distribution of foreign exchange reserves among NCBs.

Figure 3.1

Ratios of foreign assets and gold over monetary base 277.7

300

200 4-

161.

156.7 118.3

108.4 100 -k6.4 18.3

3

108.0 106.5

83.7

78.6

73.7 39.2 42.9' '29.4

7.7

14.9

33.2

37.2

I

9.6

1.6

23.3

GO

• Foreign Assets/ Monetary Base EJ Gold/ Monetary Base

Sources: Financial statements of national central banks (data for the Eurosystem are from 01.01.99; all others are end of December 1998 figures).

to

3"

Are European central banks over-capitalized?

75

The role of gold also varies enormously. Since gold is no longer actively used for monetary policy purposes, it should no longer be part of the foreign assets of central banks. Gold does not play an important role for some smaller countries, for example Austria and Ireland. The French preference for gold clearly emerges as the gold holdings of the Banque de France are worth close to 43 per cent of the French monetary base. The positions of national central banks at the International Monetary Fund (IMF), including holdings of Special Drawing Rights (SDRs), are also part of foreign exchange reserves, but these should be treated differently because they are not affected by the move to EMU. Indeed, countries, not currencies, are members of the IMF. The introduction of the euro does not affect the legal positions of national central banks vis-a-vis the IMF. From an economic point of view, however, all countries in the Eurozone should now be treated equally. When the IMF needs liquidity in euro, it should obtain it from NCBs in fixed proportions. In a longer-term perspective, it would make sense to use capital shares in the ECB as the key. The claims on the IMF that appear on NCBs' books should thus be proportional to their shares in the ECB. This would facilitate the complete pooling of foreign exchange reserves. This line of reasoning would also suggest (but of course does not imply as a logical corollary) that over time the relative quotas of member countries in the IMF should also be proportional to their capital shares in the Eurosystem. A more radical solution would be for the IMF to deal directly with the ECB so that the use of the euro by the IMF would lead to claims by the ECB on the IMF. This is explicitly foreseen in Article 30.5 of the Statute: The ECB may hold and manage IMF reserve positions and SDRs and provide for the pooling of such assets'. NCBs would still be the ultimate beneficiaries of these claims, because they own the shares of the ECB. This solution would also pave the way for a re-arrangement (and probably also a reduction) of the quotas of EMU countries in the IMF.8

76

Open Issues in European Central Banking

Balance sheet size: Does it matter? The question may be raised of whether the current balance sheet structure of NCBs and the Eurosystem represents a problem to be tackled. A first issue is to try to understand the factors behind the current structure. If the above-mentioned conjecture about the genesis of the excessively large balance sheets is correct, there should be a correlation between the latter and central bank independence and inflation. Figure 3.2 plots the average inflation rate during the pre-Maastricht period (1967-90) against the capitalization indicator.9 In countries with higher inflation, NCBs typically had balance sheets in which the monetary base was only a small part of total liabilities and capital accounts. The correlation coefficient between inflation and capitalization is about 0.7 and statistically significant at the standard confidence levels.10 The correlation between capitalization and central bank independence is also strong and negative. We use the central bank independence indicator developed by Grilli, Masciandaro and Tabellini (1991).11 Figure 3.3 shows that the more independence a central banks enjoys, the lower is its capitalization. The correlation coefficient is about -0.7 (and -0.8 using IMF data) and statistically significant. This is confirmed by inspecting the figure. NCBs' balance sheet structures have not changed with the move to EMU. They reflect the NCBs' past degree of independence and inflation performance. The question is whether the large amounts of assets and liabilities that some central banks still carry on their balance sheets constitute a nuisance in the Eurosystem. The factors that have led to the current structure are not relevant anymore. The Eurosystem has been granted independence to conduct a monetary policy aimed primarily at maintaining price stability. Monetary financing of the government has been prohibited. Furthermore, monetary policy has to be conducted in line with the principles of an open-market economy with free competition. Finally, central bank transactions have to be fully collateralized. All these constraints make it useless for NCBs to hold a large amount of capital.

Average inflation and capitalization indicator

Figure 3.2

400-| c o o _c c o

350-

N

100-

Q.

O

Portugal Finland •

300250150-

Italy •

A u s t r i a ^ 1 .Belgium^

200-

? Germany• i n Lux

•

^

France

,re,and

#

oSpam •

500C)

2

4

6

8

10

12

14

16

Average annual inflation rate (1967-90) in %

Sources: Annual reports of national central banks, 1997; IFS, 1997; and authors' calculations assuming a minimum reserve ratio of 2%.

oo

Figure 3.3

Indicator of central bank independence and lean capitalization 400 350-

o o c c o

Portugal

300 -

Italy

250 -

#

200150 -

Spain*

N

J5

100 -

o

50 n W "1 0

Belgium Austria 1 Ireland • # France Germany

Netherlands •

I

|

I

|

I

I

i

2

4

6

8

10

12

14

Indicator for central bank independence

Sources: Annual reports of national central banks, 1997; IFS, 1998; V. Grilli, D. Masciandaro and G. Tabellini, 1991; and authors' calculations assuming a minimum reserve ratio of 2%.

Are European central banks over-capitalized?

79

Another anomaly, underlined in the previous section, is the excess of foreign exchange assets held by the Eurosystem. These considerations suggest that the amount of capital and foreign exchange assets held by NCBs is excessive and useless. Does it also imply that such excess may be a nuisance? The excess of foreign reserves is certainly a nuisance since it implies that NCBs assume a risk. Central banks are not investment funds. There is no reason why they should take such a risk on behalf of society, which is ultimately its shareholder. Apparently NCBs think otherwise, since the net holding of foreign exchange reserves has not decreased after the move to EMU, in spite of the fact that DM-denominated assets have become euro-denominated assets, and thus domestic assets. Most central banks have substituted DM assets with dollar-denominated ones. It is difficult to understand the logic behind such a choice. One reason may be that NCBs want to maintain a large share of assets under their own control, rather than under the control of the ECB. This issue was discussed in Chapter 1. It is logic based on power rather than on economic considerations. NCBs prefer to take a larger risk rather than relinquishing the management (but not the income) of these assets. The interests of the bureaucracy are given higher weight than those of society. Portugal represents an extreme case (see also Stella, 1997, pp. 23-27). The Banco do Portugal is really an investment fund. It has leveraged its capital by a factor of almost five to invest mainly in foreign assets and is thus taking enormous risks. Until not so long ago, investing in foreign assets was a better investment than investing in domestic Portuguese assets because the currency was depreciating and domestic interest rates were controlled. But this rationale no longer holds in monetary union. With an amount of noneuro assets (presumably mostly US dollar) a multiple of monetary base, the Banco do Portugal runs the risk of making large losses if the US dollar suddenly depreciates. More generally, a substantial holding of assets by NCBs would be justified if the latter made better use of these assets than other institutions. The above example of foreign assets, however, suggests the opposite. Even with respect to the investment of

80

Open Issues in European Central Banking

domestic assets, there is little evidence that central banks make better use of these assets than alternative institutions. Some would even suggest that financing investment projects might use these resources better. If central banks obtained better-than-market performance, there would be the suspicion that this resulted from the use of privileged information. As mentioned in Chapter 1, the management of domestic assets can create conflicts or interfere with the conduct of the single monetary policy. The holding of these assets can thus be a source of nuisance. NCBs do not need these assets in any case, since their income is largely sufficient to pay for their costs. 3.3

Cleaning up the Eurosystem balance sheet

A lean balance sheet We showed above that the balance sheets of national central banks are much more complicated and opaque than necessary. The example of the US (see the annex at the end of this chapter) shows that even a federally structured central bank can be transparent in its accounting.12 There is therefore no reason why NCBs should not aim at producing a 'lean' balance sheet. This would imply that NCBs simply pay off all their liabilities, except the monetary base plus a minimum capital base. Under these assumptions, the balance sheet of the Eurosystem would become much shorter (and simpler), as illustrated in Table 3.3. The total assets and liabilities would shrink by about 200 billion euro, from about 700 to 500 billion euro. The liability side, which should determine the size of the overall balance sheet, would be composed of the euro area monetary base, for an amount of about 430 billion euro, and capital and revaluation reserves at 50 to 70 billion euro. Capital would be set at about 10-14 per cent of the monetary base, which is higher than the approximately 4 per cent ratio for the Bundesbank and the Federal Reserve. It is also several times the capital of the ECB, which is only 5 billion euro. The result would be a capitalization indicator of 116, in the same league as the other two major central banks.

Table 3.3

The Eurosystem's balance sheet (billion euro)

Assets

Liabilities

The current balance sheet Foreign assets - Gold

337 100

Monetary base - Banknotes in circulation - Deposits of financial institutions

430 340 90

360

Other liabilities and capital account

267

697

Total liabilities

697

Foreign assets

130

Monetary base

430

-Gold - Foreign exchange - IMF positions Domestic assets

0 100 30 370

Capital and reserves

Total assets

500

Total liabilities and capital accounts

Domestic assets Total assets The 'lean' balance sheet

- Banknotes in circulation - Deposits of financial institutions

I I ST

340 90 I

Sources: Consolidated opening financial statement of the Eurosystem, 01.01 1999, and own calculations.

70 500

82

Open Issues in European Central Banking

One obvious item to reduce on the liability side are the revaluation accounts. The purpose of revaluation reserves is to allow the central bank to provide a cushion for fluctuations in the value of its assets, essentially gold and foreign exchange reserves. One reason why the revaluation reserve could be reduced is that gold has been eliminated from the central bank balance sheet, as seen in the next section, and there is thus no need for holding the difference between the former book value and the market value of gold in the revaluation reserves. The appropriate amount of revaluation reserves would depend only on the foreign exchange reserves, which dominate the overall risk of the asset portfolio of the Eurosystem. Even if a more advanced method (for example, a value at risk approach; see Annex 3.2 for details) was chosen to calculate the need of the Eurosystem for a revaluation reserve, the absolute amount should not be as high as the present revaluation reserve.13 Moreover, the proposed capital base could be reduced. Our proposal would still leave the Eurosystem with a very generous capital base compared to the US or the Bundesbank. Further items on the liability side could also be reduced. For example, the various kind of provisions NCBs keep on their balance sheets but which are not explicitly shown in the consolidated balance sheet of the Eurosystem could be scrutinized in more detail for this purpose. Excess foreign exchange

reserves?

After NCBs' balance sheets have been cut down to a reasonable size, the question arises of the appropriate mix of assets. The main choice here is between domestic and foreign assets.14 The total foreign assets (gold plus foreign exchange reserves and IMF positions) of the 11 NCBs are almost equal to the monetary base of the system. It follows that some foreign exchange reserves should be disposed of in favor of domestic assets. The level of reserves in most industrial countries is primarily determined by the perceived need for reserves for very short-term exchange rate management. Wealth or portfolio considerations do not generally influence the level of foreign exchange reserves held by

Are European central banks over-capitalized?

83

the central bank, but rather their composition.15 It is thus difficult to estimate the optimum level of foreign exchange reserves for countries or areas such as the Eurozone that tend to have very open capital markets. They might need a certain level of reserves not for actually using it in interventions but rather to have a 'war chest' for potential speculative attacks. This demand is highly subjective and thus cannot easily be transformed into an operational figure to measure the level of optimum reserves. There is, in any case, little chance that the euro participates in a unilateral exchange rate peg, in which the Eurosystem would be forced to use large amounts of reserves to defend against speculative attacks (see Chapter 5). We would thus argue that foreign exchange reserves of about 100 billion euro should be sufficient. This would be approximately the average foreign exchange reserves held by the US and Japan and twice the minimum amount foreseen under the Maastricht Treaty. The issue of excess foreign exchange reserves arises, in any case, independently from any reshaping of the Eurosystem balance sheet. The general question that arises in this context is why should central banks, which are after all part of the public sector, hold large amounts of low-yielding assets when the government pays a higher rate of interest on its debt. A related argument applies to the case of gold. As gold does not play a role in the international monetary system, there is no reason why it should be held by central banks; the current holdings of national central banks could thus be transferred to national Ministries of Finance. The question of whether or not gold should be sold is a different matter and depends mainly on expectations about future gold prices. A common argument for not selling the gold reserves of central banks is that major central banks are monopsonic. Gold sales might depress the market price so that the receipts from a massive sale might be disappointing. However, the recent decisions by the IMF and a number of central banks in Europe (including the Swiss National Bank) to sell sizable quantities of gold will soon provide a test of this proposition. National central banks have been extremely reluctant to divest themselves of any assets. The main reason for this attitude is that

84

Open Issues in European Central Banking

any bureaucracy instinctively wants to hold on to any power and wealth it possesses. The reason adduced by central banks is different. They argue that transferring any assets to governments will only lead to more wasteful expenditure, as most governments and parliaments would probably not resist the temptation to use at least part of these assets to finance their deficits instead of just reducing public debt. Stripping central banks of their tasks as investment managers would allow them to concentrate on their role as guardians of price stability. This can only increase their independence. Otherwise, they would have to be held accountable not only for price stability, but also for their performance as managers of a large portfolio of foreign investments. The decisions taken so far by the ECB imply that national central banks will keep all of their foreign assets, but distinguish between their investment portfolios and the rest that they want to keep liquid in case there is a further call on reserves by the ECB. There is no reason why central banks should manage a sizable investment portfolio on behalf of the country. Even if their performance is better than it has been in the past, they are unlikely to outperform the market. Conclusions The new monetary constitution in Europe represents a good occasion to simplify and streamline NCBs' balance sheets, which in many cases contain items that are only of historical interest. Moreover, the opportunity should be taken to achieve greater transparency of central banks in the EU. As this area falls formally under the responsibility of national central banks, it is up to them to act and dispose of parts of their assets and liabilities until the remainder is just equal to the monetary base plus a small capital and an appropriate revaluation reserve. Of the about 200 billion euro worth of excess balance sheet items, the excess assets that represent net wealth amount probably to about 100 billion euro - mostly in the form of the excess foreign exchange

Are European central banks over-capitalized?

85

reserves. The counterparts on the liability side are revaluation reserves, capital and reserves and other liabilities that are not further disclosed. It would be necessary to identify the latter in order to find out the exact amount of the excess assets that belong to the governments and represent real wealth that was accumulated before EMU started. The excess should be used to reduce public debt. Reducing NCBs' balance sheets by about 100 billion euro of net wealth appears politically attractive. Some have proposed that this source of financing could be used for European infrastructure projects. The direct financial gain from this operation appears to be relatively modest. In terms of public debt, it entails a reduction of about 3 per cent of GDP, on average for the Eurozone, compared to total debt of about 70 per cent of GDP. Moreover, while the reduction in public debt should lead to lower interest expenditures, it will also imply lower transfers of profits from national central banks. The gain for governments lies in the difference between the low returns central banks usually earn on their assets and the cost of public debt. Even if this difference were equal to 2 percentage points, the total gain for public sector budgets would be only 0.03 per cent of GDP for the Eurozone. For some countries, for example Portugal or Spain, the gain might be much larger, but it would still remain modest in relation to overall public debt. In summary, the restructuring of NCBs' balance sheets would enable the Eurosystem to avoid possible conflicts of interest and thus to concentrate on price stability. The financial gains from disposing of the excess foreign assets in the Eurosystem are much smaller than often assumed, but any occasion to reduce public debt should be used.

86

Open Issues in European Central Banking Annex 3.1

An example of simplicity: The allocation of assets and returns in the Federal Reserve System Formally the US Federal Reserve System is much more federally structured than the Bundesbank in that it is based on 12 Federal Reserve Banks which are all legally separate and independent entities, each with its own balance sheet and president who is elected by local representatives. In most respects, however, the System is de facto completely unified, even more than the German System in which the LandesZentralbanken have no separate balance sheets because they are dejure only departments (Hauptverwaltungen) of the Bundesbank. The regional accounting is based on two keys: a capital key that is relevant for foreign assets, and a note issue key that is relevant for domestic assets. Foreign assets and the capital key. The capital of the Federal Reserve System is based on contributions from member commercial banks. Any federally chartered (coimnercial) bank in the US must buy a part of the capital of the Federal Reserve System (on which it obtains a fixed dividend yield of 6 per cent). The participation of each commercial bank is based on its own capital. The share of any Reserve Bank in the total capital of the Federal Reserve System is thus proportional to the sum of the capital of the coimnercial banks located in its district. (It is surprising that despite the concentration of banking activity in Manhattan, the New York Federal Reserve has only about 20 per cent of the aggregate capital of the system. The reason is that although many large banks are headquartered in New York, their subsidiaries in different states are formally incorporated as independent banks because of the restrictions on inter-state banking. This might now change as all the remaining restrictions on inter-state banking are lifted. One consequence of this action could be seen in 1997, when a bank with headquarters in the Federal Reserve District of Richmond, Virginia, acquired another large bank from a different area. As a result, the share of the Richmond Federal Reserve in the overall capital doubled in one year). The New York Fed has a special role in the system in that it manages all foreign exchange interventions for the system, with the

Are European central banks over-capitalized?

87

exception of its own account. All foreign exchange interventions are booked directly to an account of the system; the role of the New York Fed is only to provide the personnel that manages this account of the System. The New York Fed acts under instruction from the Board in Washington, D.C. (and under a special authorization that has to be formally renewed every year). The foreign exchange position of the Federal Reserve System that results from any interventions is then distributed on the accounts of all Federal Reserve Banks according to the capital key. As foreign exchange positions carry the risk of losses, they are distributed on the basis of the 'ability to bear losses' principle, which is supposed to be represented by capital. (Foreign exchange swaps remain on the books of the New York Federal Reserve and thus constitute the only significant exception to the rule that foreign exchange-related operations are distributed according to the capital key.) Domestic assets and the note issue key. All dollar bank notes have to carry an identifier, which marks them as the liability of a particular Reserve Bank. This was needed at the start of the Federal Reserve System when the dollar was convertible into gold and each of the constituent Reserve Banks had its own separate gold holdings. At the beginning of each year, the printing department decides how many notes to print for each of the 12 District Reserve Banks. (Once a note has the New York identifier, it is considered to be issued by the Federal Reserve of New York even if it is handed out to the public by the San Francisco Federal Reserve.) By law, the issue of currency has to be backed by assets. Each individual Reserve Bank has therefore to have on its book Treasury Bills, or other securities, equivalent to the amount of bank notes with its identifier. But individual Federal Reserve Banks do not engage directly in purchases or sales of securities. Instead, the New York Federal Reserve Bank executes open market transactions on behalf of the System and holds the resulting securities in the so-called System Open Market Account (SOMA). The balance on this account is then distributed over the balance sheets of the individual Reserve Banks according to the note issue key, ensuring that the domestic assets of the entire system are allocated across the individual Federal Reserve Banks according to the amounts of currency that they have issued. The same

88

Open Issues in European Central Banking

applies to the returns; there is thus an (almost) perfect correspondence between the allocation of note issue and the allocation of the income from domestic assets. As almost 40 per cent of all notes are issued by the New York Federal Reserve, it is also credited with a similar percentage of the total return on domestic assets, but this is independent of its role as the manager of the open market operations of the system.

Annex 3.2

Estimating the revaluation reserves using a VaR approach The amount needed for the revaluation reserve could be estimated by a value-at-risk (VaR) approach. As the volatility of the euro against other major reserve currencies is yet unknown, the sample calculation is based on a portfolio worth 100 billion DM using actual volatilities as of the end of 1997. The portfolio consists of US dollars and Japanese yen. The US dollar is the primary7 reserve currency and the Japanese yen the third most important reserve currency at present. The procedure chosen for computing the VaR is based on a standard variance-covariance model, a most simple model as it is assumed that all asset price changes can be modeled as normally distributed. The VaR of the portfolio (VaRP) consisting of asset 1 and 2 would be:

VaRp = VCVaRJ2 + (VaR 2 ) 2 +2VaR 1 VaR 2 p 12 where p is the correlation between the two assets and the individual VaR, of each position is calculated: VaRj = E 3 x b X G ] E is the exposure of the individual position, b is a parameter depending on the desired confidence interval and a} is the standard deviation of the individual position. The computation chosen for calculating the VaRP for the foreign assets of the model central bank only focuses on the foreign exchange risk and is non-diversified. Blejer and Schumacher (1998, p. 23) argue that the calculation of a diversified VaR could be favorable for stable conditions; but in scenarios of crisis or contagion, correlations tend to be very high. Thus, it is a more cautious option not to consider effects of diversification.

Are European central banks over-capitalized?

89

Data for calculating the VaR are taken from the Regulatory Data Set (J.P. Morgan, 1998a) and are adjusted to comply with the volatilities suggested by the Basle Committee on Banking Supervision of the Bank for International Settlements (BIS). The BIS rules require market-risk estimates to be calculated over a 10-day holding period and a 99 per cent confidence interval. The VaRP of the foreign exchange portfolio of 100 billion DM ranges from 4.3 billion to 6.6 billion DM (or 4.3 to 6.6 per cent) depending on the composition of the portfolio (a higher US dollar share implying a lower VaRP in this case). The interpretation of the VaRP is that, given a foreign exchange portfolio of 100 billion DM at the end of 1997, the forecasted amount that may be lost given an adverse market move ranges between about 4.3 billion to 6.6 billion DM in 99 per cent of all outcomes over the next ten days. The procedure chosen has several disadvantages, however; for example, it cannot be applied to non-linear positions and it requires the assumption of normal distributions. The distribution of changes in asset prices has been found to have 'thicker tails' than predicted by a normal distribution. This means that extreme movements seem to occur more frequently than predicted by a normal distribution (J.P. Morgan, 1996, p. 64 and p. 65). Thus, especially in times of pressure on the exchange rate, the risk might be underestimated by this method. For the Eurosystem a revaluation reserve of about 5 per cent would thus be amply sufficient, provided, of course, that the volatility of the euro exchange rate is similar to that of the DM. But even if the euro were to be twice as volatile against the US dollar as the DM, a 10 per cent reserve would still be adequate, and in line with our proposal.

90

Open Issues in European Central Banking References to Chapter 3

Andreatta, Beniamino (1998), 'Ecco il dividendo l'euro', // Sole, 24 Ore, 4 May. Blejer, Mario I. and Liliana Schumacher (1998/ Central Bank Vulnerability and the Credibility of Commitments: A Value-atRisk Approach to Currency Crises, International Monetary Fund Working Paper No. 65. Brookes, Martin (1997/ EMU's Excess Foreign Reserves, EMU Briefing Issue Number 6, Goldman Sachs. Eijffinger, Sylvester C.W. and de Haan, Jacob (1996), The Political Economy of Central Bank Independence, Special Papers in International Economics, International Finance Section, Special Papers in International Economics No. 19. ECB (1998), Guideline on the Legal Framework for Accounting and Reporting in the ESCB, ECB/1998/NP22). Federal Reserve Statistical Release H.4.1. Financial Statements of national central banks, Annual Reports, 1998. Bank of Japan (1998), Financial Statements, Tokyo. Grilli, Vittorio, Donato Masciandarro and Guido Tabellini (1991), 'Political and Monetary Institutions and Public Financial Policies in the Industrial Countries, Economic Policy, 13, pp. 341-392. Gros, Daniel (1998), Distributing Seigniorage under EMU, CEPS Working Document No. 118, Centre for European Policy Studies, Brussels. Gros, Daniel and Franziska Schobert (1999), "Excess Foreign Exchange Reserves and Overcapitalisation in the Eurosystem", ifo Schnelldienst, 52, 9 July. International Monetary Fund, International Financial Statistics, Yearbook 1997 and May 1998, Washington, D.C. J.P. Morgan (1996), RiskMethcs™ Technical Document, J.P. Morgan Bank, New York, 4th Edition. J.P. Morgan (1998a), RiskMetrics™ datasets, J.P. Morgan Bank, New York. J P. Morgan (1998b), EMU and Euro-Central Bank Assets, J.P. Morgan Securities Ltd., London, May. Pringle, Robert (1995), 'How the Reserve Bank Manages Australia's Reserves', Central Banking, Vol. VI, No. 2.

Are European central banks over-capitalized?

91

Roger, Scott (1993), The Management of Foreign Exchange Reserves, Bank for International Settlements, Economic Papers No. 38, Basle. Stella, Peter (1997), Do Central Banks Need Capital?, International Monetary Fund Working Paper No. 83, Washington, D.C. Vaez-Zadeh, Reza (1991), 'Implications and Remedies of Central Bank Losses', in Downes, Patrick and Reza Vaez-Zadeh, The Evolving Role of Central Banks, International Monetary Fund, Washington, D.C.

Notes to Chapter 3 1. The Maastricht Treaty leaves this matter to ad hoc decisions by the Governing Council of the ECB as foreseen in Art. 32.4 of the Statute of the ESCB and of the ECB: The Governing Council may decide that national central banks shall be indemnified for costs incurred in connection with the issuance of banknotes or in exceptional circumstances for specific losses arising from monetary policy operations undertaken for the ESCB. The indemnification shall be in the form deemed appropriate in the judgement of the Governing Council. 2. For a discussion of EMU's excess foreign reserves, see Brookes (1997); for a broader discussion of the management of the Eurosystem's assets under EMU, see J.P. Morgan (1998b). 3. The Italian Minister of Defense, Andreatta, proposed this idea in early 1998. See Andreatta (1998). 4. The most egregious case of accounting practices that hide the true value of assets concerns the 'lower of cost or market value' rule used until recently by the Bundesbank. Under this approach, the foreign exchange reserves are valued at the lowest value ever recorded. For example, at the end of 1996, the Bundesbank valued its US dollars in

92

Open Issues in European Central Banking

foreign exchange reserves at 1.36 (this rate was briefly reached during a period of exceptional dollar weakness in early 1995), whereas the endyear exchange rate was 1.55. At the end of 1997, the Bundesbank adjusted its accounting principles and valued its dollar reserves at 1.54 (compared with a market rate of 1.79), which was the main reason for its extraordinary high profit in 1997. 5. For more details of the legal framework for accounting and reporting in the ESCB, see ECB (1998). 6. It is hoped that NCBs no longer have substantial off-balance sheet liabilities for which we have no statistics. The Banque de France, for example, had large forward and/or swap positions outstanding whose size was never publicly revealed. 7. The ratio of foreign assets to total assets would give a similar picture, but would be distorted for countries with lean balance sheet indicators. 8. Quotas in the IMF are based, inter alia, on the importance of foreign trade; with the introduction of the euro, only the trade outside the EMU area counts as foreign trade. As the ratio of intra- to extra-EU trade is approximately 2 to 1 for most member countries, this implies that EMU will entail a considerable reduction in member countries' IMF quotas 9. The capitalization indicator is calculated directly from central bank balance sheet data at the end of 1997, assuming 2% minimum reserves. We also calculated them from data provided by the International Financial Statistics (IFS). Using IFS data gives similar results. In principle, IFS data has the advantage that the IMF provides a common presentational framework, but since some items are consolidated into 'other liabilities net', which at times is negative, the IMF data are less useful for our purposes. In any event, both sources yield capitalization indicators that are closely correlated with one another. 10. The correlation coefficient is -0.6 using IFS data; for the nine degrees of freedom this is also at the standard significance threshold.

Are European central banks over-capitalized?

93

11. The indicator for legal independence is the sum of the indices for political and economic independence by Grilli, Masciandaro and Tabellini (1991). The political independence indicator focuses on appointment procedures for board members, the length of members' terms in office and the existence of the statutory requirement to pursue monetary stability. The economic independence indicator considers the extent to which the central bank is free from government influence in implementing monetary policy (Eijffinger and de Haan, 1996, p. 23). 12. The Reserve Bank of Australia has set another example in transparency in disclosing not only how it manages its external reserves but also the return it makes on them. See Pringle (1995) and the annual reports of the Reserve Bank of Australia. 13. For an assessment of central bank risk, see Blejer and Schumacher (1998). 14. Ideally domestic assets should only consist of claims on the private sector. There is no reason why national central banks that are part of a system which has been barred from financing Governments either directly or indirectly, should keep any claim on Governments (foreign exchange reserves consist usually of government paper, but on governments outside the area). But as markets in private debt are usually not as deep or liquid as markets in Government debt, this proposal might not be feasible. It is also apparent that all nonmarketable assets should disappear. As far as possible, national central banks should also realize claims of a similar nature they have on private sector institutions. 15. See Roger (1993).

4. W H O GETS THE SEIGNIORAGE? Introduction The net income of central banks, called seigniorage, is the difference between the return received on assets and the sum paid on liabilities. The latter is generally small since a large part of liabilities comprises banknotes in circulation that bear no return. After operating costs have been deducted and provisions for reserves made, seigniorage is usually transferred to the government and constitutes an additional source of budgetary revenue. In the US, as in most other countries, the distribution of seigniorage is a non-issue. The net income of the Federal Reserve System, for instance, is transferred regularly to the US Treasury. Such a simple solution is not yet possible in Europe, however, basically because the EU does not have an important central budget and because there is no financial solidarity among the member states. The fiscal constitution of the EU is thus totally different from that of any existing state, even very federally structured ones. Moreover, because not all EU member states participate in EMU, it would also be difficult to transfer a revenue that is paid for by 11 members to the budget of the 15. The fiscal constitution of the EU is unlikely to change soon and it will take some time before all EU members have joined EMU. Thus, the question of how to distribute the seigniorage income of the Eurosystem to its components, ultimately the 11 member states that participate in EMU, needs to be addressed. Is the problem serious? The sums involved are not particularly large. On the basis of current data, the total amount of seigniorage for the Eurozone can be estimated at about 20 to 25 billion euro - or less than 0.3 per cent of GDP. The Maastricht Treaty contains detailed provisions on how seigniorage should be distributed. The treaty even allows for alternative methods, in case the proposed one cannot initially be fully implemented, and for a phasing-in transition. Despite these provisions, however, the ECB Governing Council has 94

Who gets the seigniorage?

95

taken the decision not to apply them until the year 2002, thus possibly breaching the treaty. National central banks will continue to collect and allocate income until 2002, as if EMU had not happened. The fact that it proved impossible to reach an agreement for the distribution of the Eurosystem's revenues shows that this issue contains a potential for conflict. This chapter tries to explain why. To provide a general background, it may be useful to start by briefly defining the concept of central bank profits or seigniorage. The provisions of the Maastricht Treaty concerning the distribution of seigniorage are then discussed, with a view to examining the potential gains and losses for each NCB. Since the problem hinges essentially on the distribution of currency in circulation, the case of large cross-border flows of euro cash after 2002 is also considered. An alternative approach for the distribution of seigniorage inside the Eurosystem is also analyzed. 4.1

What is seigniorage?

There is a large literature on seigniorage,1 and several definitions are in use. For the sake of simplicity, we have chosen a concept based on the savings in interest payments accrued to the government from the privilege of issuing currency. Indeed, currency can be assimilated to a zero interest rate bond. Taking into account interest payments on required reserves, the revenue from seigniorage (S) is given by the following equation: S = [i C + (i-r) RR] where C is currency in circulation, RR required reserves, i the interest rate on government debt and r the interest paid on reserves. The holding of required reserves by commercial banks with the central bank produces seigniorage if the interest rate on government debt exceeds the interest rate paid on reserves. In the euro area, required reserves are remunerated at close to market rates. The Eurosystem derives little seigniorage on that account. Most of its seigniorage revenues arise from the issuance of currency. The amount of income received by the government is lower than the theoretical amount calculated above, for two reasons. First,

96

Open Questions in European Central Banking

central banks deduct expenditures incurred in the running of their activities.2 Second, reserves are generally provisioned to protect central banks from fluctuations in the value of their assets. As shown in Chapter 3, the accumulation of reserves and capital from unrealized foreign exchange gains has been significant for some NCBs. In general, NCBs have considerable latitude in deciding how much of their income will effectively be transferred to the government budget. 4.2

The treaty's provisions

The Statute of the ESCB and of the ECB envisages that NCBs participating in the Eurosystem will earn 'monetary income' from their participation in the conduct of the single monetary policy. The sum of NCBs' income constitutes the total monetary income of the Eurosystem. This total should be distributed to the component NCBs according to their respective shares in the ECB capital. The reasoning underlying these provisions is that it is impossible in a monetary union to determine the 'nationality' of seigniorage. Currency and central bank deposits can flow freely from one country to another. The amount of monetary base recorded in NCBs' balance sheet does not necessarily reflect the amount of currency and bank reserves held by the residents of that country. It may be possible that the liabilities recorded in the balance sheet of one NCB can shrink substantially, and even disappear because of flows occurring within the Eurozone. This does not necessarily mean that the holding of currency in the country has decreased. Seitz (1995) shows that about 12-18 per cent of the total amount of DM notes in circulation is officially reported to migrate abroad every year, part of it coming back within the same period. Even if only half of these flows were to take place inside the euro area, they would have profound implications for the evolution of the measured monetary base in Germany and the rest of the euro area. Once euro notes and coins have been introduced, any banknote taken abroad by Germans would no longer need to be returned to the Bundesbank. The euro cash that Germans carry with them to spend on their southern European vacations might remain in those countries. The measured German monetary base might then increase by about 20

Who gets the seigniorage?

97

billion DM, or around 10 per cent, each year, while the monetary base of those countries hosting German tourists would fall by a corresponding amount. The experience of the US confirms that the regional demand for cash is impossible to predict. There are extreme disparities in cash demand across different Federal Reserve districts. Some districts have actually had large net redemption over the past decades. Porter and Judson (1996) report that the Federal Reserve District of New York alone accounted for 80 per cent of the roughly $300 billion increase in currency in circulation between 1975 and 1995. The predominant role of New York can be explained by the fact it is the primary supplier of cash for foreign users. Other examples (which are also more difficult to explain) are even more extreme: the Federal Reserve District of Atlanta over the same period had a net redemption of more than 34 per cent of the total. Given that currency in circulation in the entire US amounted to a mere $70 billion in 1974, the net redemption in this district alone (34 per cent of over $300 billion) must have been several times its base at the beginning of the period. If something similar were to happen in Europe, some countries might end up with a negative monetary base. Such an extreme result is unlikely. But the importance of both German tourism and Germany's strong ties with Eastern Europe, where the euro might soon replace the US dollar as parallel currency (DM holdings are automatically converted into euro in 2002), suggests that the Bundesbank might play the role of the New York Fed, so that the measured German share in the European monetary base would likely increase over time. To illustrate the importance of this issue let us assume, for example, that after 2002, German tourists continue to take cash on their European vacations (not an unreasonable assumption, since there will no longer be any costs associated with converting currencies), and that each year about 10 billion euro worth of cash is thereby exported. (This would be less than at present, and would also not be unreasonable given a German intra-EU yearly deficit on tourism of almost 20 billion euro that is growing continuously.) The German monetary base would thereby increase after ten years by

98

Open Questions in European Central Banking

100 billion euro (while the monetary base of the countries of destination would decrease by the same amount). The measured German monetary income would then increase by 6 billion euro, if interest rates are at 6 per cent, to almost 16 billion euro. If income was not pooled, Germany would receive net implicit transfers from its neighbors of about 6 to 8 billion euro each year. If no redistribution mechanism were designed, the countries whose NCB balance sheet would decrease would receive less income, even though their citizens indeed do hold cash and thus pay their seigniorage. Since NCB balance sheets in monetary union no longer reflect the currency and reserves held in the respective countries, how can the seigniorage cashed by the different NCBs be redistributed across countries? The treaty makes a simplifying assumption: citizens throughout the Eurozone hold the same amount of currency, depending on the level of income. Therefore, the redistribution across NCBs of the total amount of income earned by the Eurosystem should be made on the basis of GDP and population of the member states. A key for redistribution is calculated for each country on the basis of the percentage of that country's GDP and population over the total for the Eurosystem. This key is the same as the one used to calculate the capital shares of the ECB. The mechanism foreseen in the treaty does not involve transactions on the books of the ECB, but only entail transfers among national central banks. 4.3

The Problems

Implementation of the treaty's provisions raises two problems. The first problem concerns the definition of the concept of 'monetary income'. Article 32.2 of the statute stipulates that the monetary income should be equal to the income arising to NCBs from the holding of assets earmarked against the monetary base (currency in circulation plus deposit liabilities of commercial banks). If NCBs' balance sheets were similar to those in textbooks, this could be easily implemented. NCBs would earmark the assets held as counterpart to the monetary base. Two main difficulties arise, however. First, as mentioned in the previous chapter, the amount of assets held by NCBs is much larger than the monetary base. There is thus a

Who gets the seigniorage?

99

problem of selecting the assets to earmark. The second difficulty is that the assets held by NCBs are very different in nature, being the result of the central banks' past monetary policies. Since different assets yield different returns, NCBs would have an incentive to earmark the lowest yielding assets for defining the monetary income that must be allocated to the Eurosystem. Solutions could have been found to these problems, such as for instance earmarking assets at market value and progressively substituting old assets with new ones acquired through the conduct of monetary policy. Such solutions would have implied a cleaning up of NCBs balance sheets as suggested in the previous chapter. The second problem is that in the two years prior to EMU, the share of each country's seigniorage over the total differed from the key mentioned above. Based on the current data on holdings of national currency, Germany's monetary income amounts to around 40 per cent of the Eurozone's total, while its GDP and population amount to only 30 per cent. The implementation of the treaty's provisions from 1999 would have implied a net loss for Germany of about 2 billion euro per year (Table 4.1). The opposite would be true for France, which would receive a net transfer of about 1.7 billion euro. Why does France gain and Germany loses? The reason is that the monetary base of France is merely one-third that of Germany. The contribution of the Banque de France to the total to be distributed would be only one-third of that of Germany, 2.7 against 8.4 billion euro. But the shares of the two countries in the capital of the ECB (and hence in total monetary income) are comparable roughly 22 per cent for France and 30 per cent for Germany. Other central banks that are likely to gain are Portugal and Finland, both of which would receive net transfers of slightly less than 200 million euro.

o o

Table 4.1

Basic data for seigniorage distribution in EMU

Country

Base money

Capital share

Gain or loss

Monetary income (billion euro)

Billion euro

% of total

in %

Allocated (= debit)

Germany

131

40

31

8.4

France

42

13

21

Italy

58

18

1

Received (= credit)

2

Billion euro

% of GDP

6.5

-1.9

-0.10

2.7

4.5

1.7

0.15

19

3.8

3.9

0.2

0.05

2.6

2.4

-0.2

-0.18 -0.11

Spain

40

12

11

Austria

12

4

3

0.8

0.6

-0.2

Belgium

12

4

4

0.8

0.8

0.0

-0.01

Ireland

4

1

1

0.2

0.2

0.0

0.13

Netherlands

17

5

5

1.1

1.1

0.1

-0.01

Portugal

5

2

2

0.3

0.5

0.2

0.29

Finland

3

1

2

0.2

0.4

0.2

0.23

325

100

100

21

21

Total 1

National monetary base multiplied by a 6.34 per cent reference rate.

2

Capital share of total monetary base multiplied by a 6.34 per cent reference rate.

Who gets the seigniorage?

101

The likely pattern of bilateral transfers would be the following (the pairing is based on the magnitudes of the transfers): • • •

From Germany to France (approximately 1.5-2 billion euro), From Austria to Finland (about 160-180 million euro), and From Spain (about 250 million euro) to Italy and Portugal (about 160 million each). The size of the gains and the losses is proportional to the rate of return used to calculate the monetary income. This is clearly problematic, since those NCBs whose share in the capital is greater than their share in the area-wide monetary base would gain from a high rate of return (and vice versa). These transfers might appear modest from a macroeconomic point of view, e.g. if expressed as a percentage of GDP. Sinn and Feist (1997) arrive at higher numbers by concentrating on the size of the implicit transfer in stocks or assets mandated by the existing provisions. They argue that with EMU, Germany gives up control of 131 billion euro of its monetary base and receives in return only the income from about 97 billion euro, which is its share (31 per cent) of the overall 320 billion euro monetary base. From this point of view, Germany has a net loss of assets of about 34 billion euro. But the transfer of control over assets should be equivalent to the present value of future revenue losses. In the image used by Sinn (1997), the assets are the cows of the Bundesbank and the returns the milk. The value of a cow should be equal to the present value of the future streams of milk it is expected to yield. To make a concrete example: Given an interest rate of about 5%, an annual loss of 1.9 billion euro corresponds to a capital loss of about 38 billion euro, which is the order of magnitude mentioned by these authors as the potential loss for Germany. The present value calculations are valid only under the assumption that nothing will change for the indefinite future. While the distribution of the seigniorage base, i.e. cash, might not change quickly, the rapid evolution of national payment systems and habits suggests that large changes may take place over time. Experience shows that this has been the case in the past as well.

102

%ui

Open Questions in European Central Banking

Ok

O

o c o 3 u u 'u

C

o 05

u S OX)

s

Who gets the seigniorage?

103

Figure 4.1 describes the evolution of the ratio of currency in circulation/GDP for the larger EMU countries. These trends suggest that it is unlikely that the ratios will remain constant, since there have already been major changes over the last decade. The direction of future changes, however, is difficult to forecast. Even if national habits and preferences in the use of cash do not change immediately, it will no longer be possible to determine, or even to estimate the distribution of the stocks of currency in circulation in different countries after 2002. Thus far, it has proven impossible to identify the determinants of the demand for cash (over time or across countries), as is shown by the vast literature on the size of foreign holdings of US dollars and DMs.3 Given this ignorance, would it not be reasonable to assume that the status quo will persist? The answer is no, for the experience of large currency areas shows that there can be extensive inter-regional flows of cash that are impossible to predict. 4.4

The agreed non-solution

The only publicly available information on the stance of the ECB regarding the distribution of seigniorage is a short press communique of 3 November 1998, which read as follows: As regards the income arising from the issuance of national banknotes from the participating Member States, the Governing Council today decided that such income will not be redistributed prior to the introduction of the euro banknotes. The Governing Council also confirmed with respect to the allocation of monetary income of the national central banks of the participating Member States for the financial years 1999 to 2001, that the monetary income for the said period will be calculated by employing the socalled indirect method, by multiplying a defined liability base by a specified reference rate of interest. The sum of the monetary income of each national central bank will be allocated to the participating national central banks in proportion to their paid-up share in the ECB's capital. Owing to the temporary character of the current decision, the question of monetary income allocation

104

Open Questions in European Central Banking

will need to be revisited before the introduction of euro banknotes in 2002. This communique is another example of the reluctance of central banks to open their financial accounts to the public. No justification or legal basis is given for the decision nor any indication of the amounts involved. Moreover, the statement that the indirect method will be applied for the determination of monetary income during the period 1999-2001 is formally correct, but misleading in substance because it suggests that there will be some monetary income to be distributed. This will not be the case, since the monetary income on required reserves is directly offset by the interest paid on them. Hence there will be practically no monetary income to distribute until 2001. This decision of the ECB is inconsistent with both the letter and the spirit of the treaty, although two articles of the ECB Statute allow for some flexibility in this area. First, Article 32.3 allows the possibility that an alternative method for determining monetary income would be used for the first five years 'after the start of the third stage'. The article could cover the use of the indirect (or some other) method, but not the exclusion of the basis for the calculation of such income. Second, Article 51.1 of the statute allows the Governing Council to reduce the amount to be distributed, if 'the application of Article 32 results in significant changes in national central banks' relative income positions'. As this is certainly the case, it would have been entirely appropriate to use this article as the legal basis for any decision regarding the distribution of monetary income. But this article says that the reduction in the distribution of monetary income can be at most by 60 per cent of the total during the first year. The decision to disregard all currency in circulation for the determination of monetary income can thus not be based on this article either. The so-called 'indirect method' implies that monetary income is determined in a fictitious way: not by calculating the income earned on assets earmarked against the money base but by imputing a reference rate. Income is thus calculated by multiplying the NCBs' monetary base with a common reference. With the indirect method, the choice of the reference rate is crucial to determine the size of the

Who gets the seigniorage?

105

monetary income and hence the transfers among NCBs. It will most probably be very difficult to arrive at a consensus on the reference rate, since there are obvious conflicts of interest. There are no objective criteria allowing one to chose between, for example, shortor long-term interest rates or between actual market rates or 'normal' rates based on historical averages. 4.5

Alternative approaches to phasing-in

The treaty's provisions were drafted under the assumption that the holding of cash would converge across the euro area in a five-year period. There is no way to check, however, whether convergence takes place or not. The problem is that NCBs tend to look at the results obtained in previous years and do not want to be allocated a lower amount of income. The question is thus whether the provisions concerning the distribution of seigniorage should be modified to correspond to a more realistic scenario. An alternative approach would be to 'grandfather' existing stocks of currency held in the different countries and to only pool the income from newly created euro-cash. An easy way to implement this approach would be to consider all additional currency issued from 2002 onwards as 'European' and to account for it on the ECB's balance sheet. All required reserves would also be accounted for in the ECB's balance sheet and not just the increments recorded after 2002. The decision to impose reserve requirements on commercial banks is made by the Eurosystem. The financial consequences should thus be attributed to the European level, that is, the ECB. The income earned by the ECB on the assets it receives against the issuance of additional euro cash (or Central Bank money to commercial banks) would be redistributed according to the existing provisions for ECB profits (Article 33 of the ECB Statute). NCBs would keep the returns on the stocks they have accumulated up to 2002.4 What would be the implications of this proposal? From 1999 to 2002, there would practically be no (re-) distribution of seigniorage revenues (required reserves are

106

Open Questions in European Central Banking

remunerated at close to market rates (at the refinancing rate set by the ECB, and such remuneration has to be deducted from earned income). After 2002, the income on newly created currency would be redistributed. This proposal implies that NCBs' balance sheets would be frozen at their end-2001 level, whereas that of the ECB would grow over time, if the demand for cash grows, that is, if the velocity of circulation does not increase faster than the growth rate of nominal income, which would be unprecedented for any extended period in Europe. After 2002, the European component of seigniorage would start to increase gradually, and would be distributed to NCBs according to the capital key. Assuming a 1 per cent increase in the velocity of currency in circulation and a growth rate of nominal GDP of about 5 per cent per year, the balance sheet of the ECB would grow by about 12-15 billion euro annually. After 20 years, the ECB's balance sheet would be comparable to the combined balance sheets of the member NCBs; that is, about 240-300 billion euro. Seigniorage revenue at European level would reach about the same level as the sum of national revenues. The potential cross-border shifts in currency mentioned above would be irrelevant for the distribution of seigniorage. What if the demand for cash falls in absolute terms? This would require the velocity of circulation to fall by more than the growth rate of nominal GDP, or about 4 to 5 per cent each year. This has never happened over any substantial period of time, but cannot be altogether excluded if electronic forms of money become much more widespread. In this case, the entire seigniorage issue would gradually lose importance because the base for seigniorage revenues would be shrinking not only relative to GDP, but actually in absolute terms. If, for example, the demand for euro cash just remains constant in nominal terms after 2002, the approach proposed here would remain inoperative as the balance sheet of the ECB would remain negligible. However, the overall amount of monetary income to re-distribute to NCBs would also remain constant, in nominal terms (ceteris paribus, i.e. at unchanged interest rates). Seigniorage would thus decline as a percentage of Euroland GDP at about 4 to 5 per cent each year. After 20 years, the magnitude of seigniorage would thus effectively be halved. While

Who gets the seigniorage?

107

this would not solve the problem of how best to distribute seigniorage, the issue would become less important. The opposite case of a sharp increase in the demand for euro cash can also not be excluded. Even if the demand for cash from within Euroland increases only moderately following the historical patterns mentioned above, there could be a considerable increase in the use of the euro in the rest of the world, which would go far beyond the use of the DM in some parts of Central and Eastern Europe. The euro would be a true global currency and might be in demand in Latin America and Asia as well. Based on the estimates of dollar cash held abroad, it is conceivable that over 100 billion euro banknotes would be held abroad. This phenomenon would a be consequence of the economies of scale provided by a common currency and the seigniorage gain to Europe (the interest earnings on this sum) should thus accrue directly at the European level, namely the ECB. Therefore, booking all increases in euro cash at the ECB would be natural in case the demand for euro increases; and would do no harm (because it would remain inoperative) if the demand for euro does not increase. The 'grandfathering' approach is based on the assumption that the convergence in the demand for cash takes place in a much longer time period than the five years assumed in the provisions of the treaty. If cash holdings converged more rapidly towards a common European pattern, this solution would produce the opposite effect than the one illustrated in Table 4.1. Germany would become a net gainer; France a net loser. For any permanent solution, one would thus have to look at the present value of the flows over time instead only at the flows at the beginning or in some far away future. The difference in the present value of what Germany would retain under total grandfathering and what it should retain if payment patterns converged over time is a fraction of the initial difference between Germany's share in the EMU monetary base and its capital share. It can be shown that this fraction is less than one-half, if the speed of adjustment parameter is less than the reference rate (and vice versa). For example, with a reference rate of 6 per cent, total

108

Open Questions in European Central Banking

grandfathering would be closer to the ideal if the adjustment in payment habits were so slow that each year less than 5 per cent of the initial difference between capital shares and shares in the monetary base would be eliminated. Unfortunately, it is impossible to predict local demand for currency. National cash-to-GDP ratios have diverged over the past 20 years, despite the integration of financial markets and a generalized diffusion of technology (automated cash dispensers and point-of-sales terminals). In the case of Germany, this is often explained by pointing to the substantial foreign holdings of DM. But would that also apply to Austria? And why would the cash-to-GDP ratio be even higher in the case of Spain? The usual explanation in the latter case is the black-market economy. It is unclear whether monetary union will have an impact on the local habits that determine the demand for cash. Thus, it remains extremely difficult to decide on the appropriateness of one solution over another. If one assumes that the reference rate is likely to be about the same size (between 4 and 6 per cent) as the speed of adjustment parameter (half-life of adjustment between 25 and 16 years), total grandfathering might end up being about as far away from the appropriate solution as zero grandfathering. Given that one just does not know how long it will take for payment habits to converge, it would appear to be sensible to place oneself in the middle. An easy way to achieve this would be to adopt the proposal discussed so far but with the additional provision that national central banks must transfer each year a pre-specified fraction of their assets (for example 1/20 of their monetary base to the ECB). At the end of the 20 years, national central banks would still be left with the capital that they had accumulated before EMU, but all monetary bases would be accounted for in the ECB balance sheet. Conclusions The treaty's provisions for the distribution of seigniorage in monetary union are appropriate for an integrated currency area in which there is either fiscal solidarity or at least a convergence in payment habits to such an extent that the use of cash can be presumed to be uniform. These provisions were negotiated almost a

Who gets the seigniorage?

109

decade before the start of monetary union. It was assumed at that time that considerable convergence would take place in the meantime. This did not happen and the issue was rediscovered just before the start of monetary union. An immediate application of the treaty's provisions would have lead to transfers between national central banks of such a magnitude as to be politically sensitive. Since it proved impossible to achieve a satisfactory compromise between the presumed losers and winners from the redistribution of seigniorage revenues, it was finally decided, contrary to the text and the spirit of Maastricht, that the application of the treaty's provisions would be deferred until 2002. After that date, an alternative method for distributing seigniorage would be devised. The unsatisfactory solution to the seigniorage issue is a reflection of a larger unresolved issue, namely, the absence of financial solidarity between member states and therefore also among NCBs. The broader issue is also reflected in the imbalance in the relative roles of the center (the ECB) and the member national central banks within the Eurosystem, discussed in Chapter 1. The provisions for seigniorage reflect an unhappy mixture of centralization (all revenues from the monetary base are redistributed) and decentralization (no assets are attributed to the ECB, not even those acquired by the Eurosystem after monetary union). The proposal examined in this chapter would avoid both extremes by distinguishing between the seigniorage created under monetary union and the revenues from assets accumulated beforehand. The former should be controlled and distributed by the ECB, whereas the latter would remain in national hands. Even under the best of circumstances, the existing treaty provisions will provide a constant potential for distribution conflicts, which are always divisive since they constitute a zero-sum game. The legal obstacles to a change are not insurmountable (see the annex to this chapter), since the treaty already foresees a simplified procedure for modifying most of Article 32 of the statute. Hence, it would be sufficient to have a Council decision (with a qualified

110

Open Questions in European Central Banking

majority, provided there is a proposal from the ECB) to revise the relevant articles. Since this volume is concerned with the monetary construction of Europe, we have taken it for granted in this chapter that the fiscal constitution of the EU will not change any time soon and that seigniorage has to be redistributed to member states. In the long run, however, one should consider whether it would not be appropriate to transfer the seigniorage revenues of the Eurosystem to the EU budget. Annex 2 briefly discusses the pros and cons of this idea and comes to the conclusion that in the very long run, when the fiscal constitution of the EU is likely to change, it would be possible to assign the euro seigniorage directly to the EU budget. This would require a change in the treaty and would obviate the need for special provisions concerning the distribution of seigniorage among member states. As long as the EU budget remains relatively small, however, it would be necessary to take special precautions to insulate the revenues of the EU from the direct impact of interest rate decisions of the ECB. While it might be appropriate in a common currency area to assign seigniorage, which is the tax on this common currency, to the common budget, it would be dangerous if the EU budget would depend on this source of income for revenues. After all, the tendency of modern central banking is to de-link the management of money from fiscal considerations.

Who gets the seigniorage?

Ill

Annex 4.1 Legal Issues The alternative discussed here is not compatible with the provisions of Article 32.5 of the Statute of the ESCB and the ECB and would thus require a decision to alter them. Closer inspection of the legal texts, however, shows that it would not be too difficult to engineer a solution, especially for the five-year transition period allowed for in Article 32.3. We note that this article begins with 'If, after the start of the third stage,...' This seems to imply quite clearly that the five-year transitional period, during which time the indirect method could be used, started in 1999. It is difficult to see why the ECB has decided to ignore this clear wording and behaves as if the transitional period starts only in 2002. For this transitional period one could approximate the essence of the proposal made here for the period from 1999 through 2004 by choosing a sufficiently low reference rate. For example, the Governing Council of the ECB could just decide to apply a reference rate of about 1.5 per cent. But as this roundabout solution would not be consistent with the spirit of the statute, it would be better to initiate the procedure necessary for changing it decision to use the indirect approach requires a qualified majority of the Governing Board of the ECB (Article 10.3), where a qualified majority is defined as two-thirds of the members weighted by capital shares representing at least half of the shareholders (the members of the Executive Board do not count in financial matters). Since the votes are weighted by capital shares (that is, by the weights at the receiving end), there is a built-in tendency to vote for high reference rates. (Moreover, most countries are net recipients.) However, the only two substantial net payers under the provisions foreseen in the treaty (Germany and Spain) could together already muster a blocking minority, as they account for over 40 per cent of the shares in an EMU11. In any event, given that the transitory regime is limited to a period of five years, the provisions for the distribution of seigniorage

112

Open Questions in European Central Banking

should be changed. This change would be facilitated by the fact that the treaty (Article 106.5) already foresees a simplified amendment procedure for part of the statute, which (under Article 41 of the statute includes Article 32.2-32.6, but not Article 32.1. Under this procedure, the Council (ECOFIN) could enact modifications by a qualified majority provided there is a unanimous recommendation from the ECB. Annex 4.2 The distribution of seigniorage in the context of the fiscal constitution of the EU A key element of the constitution of the EU is that there is no financial solidarity among the member states and that the budget of the EU is small in relation to the financial resources of the member states. Progress towards EMU only became possible once the notion which had been prevalent in the 1960s and 1970s that further integration required a sizeable budget for the EU was abandoned. According to the Mastricht Treaty, seigniorage revenues of the Eurosystem are distributed to the participating national central banks according to the capital key. The underlying reason for this is that seigniorage represents essentially a tax on the holdings of cash and that cash holdings should depend mainly on income. However, the capital key, which determines the contributions of NCBs to the capital of the Eurosystem and their share in its monetary income, is based on two parameters: income (GDP) and population. It thus represents a compromise between the 'one man - one vote' approach used in societies in which there is financial solidarity and the idea that there is no solidarity in financial terms in the Eurosystem. In the latter view, the Eurosystem is considered as a mutual fund in which the share of the profits should be strictly proportional to the contribution to the capital. The treaty's provisions concerning seigniorage effectively imply that the EU has a common currency, but that the revenues from the normal tax on this common currency remain national. This would correspond to a narrow interpretation of EMU, namely that the latter is only a way to achieve exchange rate stability - a view that has also pervaded the institutional set-up. This narrow approach

Who gets the seigniorage?

113

might have been necessary to achieve agreement on EMU at Maastricht, but one might ask whether this choice is appropriate in the long run. But even apart from this question, which goes beyond the scope of the present volume, one could argue that the distribution of seigniorage leads to unnecessary two-way transfers. Distributing the income of the Eurosystem at the national level leads to a situation in which each member country receives its respective share of the monetary income of the Eurosystem each year and then is obliged to to make a transfer back to the EU level (the EU budget). This raises the question of whether it would not be more straightforward to assign the revenue from the common currency directly to the EU budget. The marginal source of revenue for the EU is already now largely based upon GNP shares. Except in the case of the poorest member countries, there is a close correspondence between the national shares in the GNP-based contribution to the EU budget and the shares in the monetary income of the Eurosystem. One could thus argue that the distribution of euro seigniorage to member countries leads to unnecessary round-trips of transfers. For poorer member countries, the distribution of seigniorage according to the capital key could be an advantage because their shares in the ECB (partially based on population weights) are higher than their shares in the GNP-based contributions (see Figure 4.A.1). One could therefore argue that the treaty's provisions regarding seigniorage already imply some re-distribution. Whether or not this is the case in reality depends on the actual distribution of cash holdings. If the demand for cash were strictly proportional to GDP, the key used by the Eurosystem (based on population and GDP with equal weights) would imply a redistribution of wealth to the poor. But this is unlikely to occur even in the very long run, when financial markets have been unified across the euro-area. In richer parts of the EU, electronic money might be used more intensively so that cash holdings fall, relative to income. But one can also observe that the demand for cash proportionally rises more with income (holding cash is a luxury good) because

Figure 4.A.1

The difference between GNP ratios and the ECB capital key (percentage points)

si «•«•.

3"

• 03 Difference between GNP ratio and ECB capital key

to

§

a: 3

Who gets the seigniorage?

115

wealthier individuals do not face credit constraints and do not need to manage their cash holdings tightly. It is thus impossible to predict how actual cash-holding patterns will evolve. And, therefore, it is impossible to say with any certainty whether the distribution of seigniorage implies any re-distribution from richer to poorer member states. What would be the implications of assigning the monetary income of the Eurosystem directly to the EU budget? Two issues emerge in this context: a) At present, only 11 member countries participate in EMU and it is unlikely that most of the outs will join soon. With the next enlargement, the number of countries that are members of the EU but not members of EMU might increase again. b) The seigniorage revenues of the Eurosystem are likely to remain negligible in relation to the GDP of the euro-area. As long as price stability is maintained and interest rates remain low, the amount to be distributed within the Eurosystem is likely to remain in the range of 0.2 to 0.4 % of GDP. For most member countries, this is also likely to amount to less than 1 per cent of all public sector revenues and will thus also be negligible in terms of fiscal policy as well. Nevertheless, the revenues of the Eurosystem would not be negligible in relation to the budget of the EU, which only amounts to approximately 1.2 per cent of the EU's GDP. A sum that is only 0.2 to 0.4 per cent of the GDP still represents between one-sixth and one-third of the EU's budget. Those responsible for the EU's budget would thus have a strong financial interest in monetary policy. An increase in interest rates of 1 percentage point would yield about 3 to 4 billion euro more per annum for the EU budget. This would be a considerable sum for in the context of the financing of the EU in which even large member countries make a question of principle of much smaller sums. Moreover, this source would be rather variable and difficult to predict because interest rates can be rather variable, as experience has shown. An unexpected drop in interest rates might lead to a shortfall in the contribution from the Eurosystem which

116

Open Questions in European Central Banking

could be made up via the GNP source, but this could lead to serious political problems as it would risk unbalancing (in the eyes of some member states) the budget framework which is so laboriously negotiated every five years. The problems posed by the short-term variability of monetary income could be overcome by smoothing the transfers from the Eurosystem, for example by basing them on an average of the monetary income over the preceding five years. However, this would not change the fact that seigniorage would become a major source of revenues for the EU budget.

References to Chapter 4 Central Banking (1996), The Winners and Losers from EMU, November, pp. 9-10. Gros, Daniel (1993), 'Seigniorage in the EC: The Effects of Financial Market Integration and of Monetary Unification', European Journal of Political Economy, 9, pp. 581-601. Gros, Daniel (1998), Distributing Seigniorage under EMU, Centre for European Policy Studies, CEPS Working Document No. 118, Brussels, May. Porter, Richard D. and Ruth A. Judson (1996), 'The Location of US Currency: How Much is Abroad?', Federal Reserve Bulletin, October, pp. 883-899. Rogoff, Kenneth (1998), 'Foreign and Underground Demand for Euro Notes: Blessing or a Curse?', forthcoming Economic Policy. Schobert, Franziska (1998), 'Seigniorage: Potential Measurement and Distribution under EMU', Manuscript, University of Frankfurt. Seitz, Franz (1995), The Circulation of the Deutsche Mark Abroad, Discussion Paper 1/1995, Economic Research Group of the Deutsche Bundesbank. Sinn, Hans-Werner (1997), 'Die Kuhe des Bauern Buba', Wirtschaftswoche, 49, 27 November. Sinn, Hans-Werner and Holger Feist (1997), 'Eurowinners and Eurolosers: The Distribution of Seigniorage Wealth in EMU', European Journal of Political Economy, 13.

Who gets the seigniorage?

117

Notes to Chapter 4 1. See Gros (1993). 2. In this chapter, we only discuss the income side of the profit and loss accounts of the NCBs in the Eurosystem, because the treaty does not contain any provisions for the distribution of the costs of running national central banks. These costs, mainly personnel and buildings (about 2-3 billion euro per annum) are small compared to the interest income. Nevertheless, it is likely that central banks with rather large staffs (for example, the Banque de France or the Bundesbank both of which employ close to 18,000 persons) will soon come under considerable pressure at the national level to reduce their personnel costs. The incentive to do so remains because any cost savings increases the transfers that national treasuries can expect from their NCBs. 3. See Rogoff (1998) and Seitz (1995). 4. Sinn and Feist (1997) and Sinn (1997) make a similar proposal, but do not spell out its consequences in detail.

5. How CAN THE EUROSYSTEM BECOME (AND REMAIN) INDEPENDENT? . . . Introduction A large body of literature has emerged in recent years on the importance of central bank independence.1 This literature has influenced greatly the design of the European Central Bank and has given rise to institutional changes in several countries aimed at making their central banks independent. The argument for giving independence to central banks rests on the benefits that such independence can bring to the conduct of monetary policy and thus to the credibility of the institution. Credibility can be defined as the extent to which market participants believe that the central bank will achieve its goals. If a central bank is independent, it can conduct monetary policy with the aim of achieving its primary objective. If it is not independent, it can be subject to pressures to pursue other goals. Indeed, monetary policy can be directed at different and at times conflicting objectives, such as price stability and economic growth. The effectiveness of monetary policy depends on agents' expectations. Monetary policy is most effective in the area of inflation when agents perfectly anticipate it; it is most effective on real economic variables, such as output or employment, when agents do not fully anticipate the intentions of the authorities. If monetary policy is aimed at controlling inflation, the central bank should act in a way that allows agents to anticipate its actions. If, on the contrary, monetary policy aims at stimulating output and employment, the central bank should 'surprise' economic agents and renege on preannounced commitments. This dilemma is well understood by economic agents. If agents believe that the central bank pursues, or is pressed to pursue, a growth and employment objective, rather than price stability, they will suspect the central bank of conducting a 118

How can the Eurosystem become independent?

119

more expansionary, and thus more inflationary policy than announced. Inflation expectations will be revised accordingly, with potentially negative effects on real economic activity. One way to resolve the time-consistency dilemma is to prohibit the central bank from pursuing a growth and employment objective. Monetary policy should be directed only at price stability and conducted separately from other economic policies. To achieve its goal, the central bank has to be entrusted with the necessary independence. Incentive-compatible contractual arrangements can be designed to provide the central bank with independence of action and accountability of results.2 The pressures exerted on the ECB during its early days, in particular by the German government, to cut interest rates provides an excellent illustration of this problem. Had the ECB not been given an independent status, it would have been forced to adopt a more expansionary monetary policy with a view to stimulating growth and employment, even at the cost of higher inflation. Long-term interest rates could have increased substantially, mirroring a rise in inflationary expectations. Empirical research has confirmed that independent central banks on average deliver a better performance in terms of price stability, without incurring any additional costs in terms of output variability. The Maastricht Treaty has been inspired by this analytical framework and the Statute of the ESCB and of the ECB has been drafted with a view to entrusting the latter with the highest possible degree of central bank independence. Independence does not depend only on statutes and laws, however, but on a series of factors that are internal and external to the central bank. Furthermore, as indicated in Chapter 1, the Eurosystem has a rather peculiar construction. It is made up of the ECB and the NCBs, both of which are in principle independent. However, NCBs can conduct a series of activities 'outside' the Eurosystem, for which independence is not required.

120

Open Issues in European Central Banking

Why should the ECB and the NCBs be independent for the performance of monetary policy, but not for the performance of other functions? This dualism might constitute a potential source for conflict that may undermine the independence of the Eurosystem in the conduct of monetary policy. This chapter examines the extent to which the Eurosystem can be expected to be independent. It starts from the analysis of the Maastricht Treaty's provisions relating to independence. Some standard criteria and indices are used for assessing the degree of independence. Section 3 discusses the limits and shortcomings of these indices, and Section 4 examines the possible threats to the independence of the Eurosystem. 5.1

The independence provided by the treaty

This section examines the main features of the Eurosystem's independence as provided for in the Statute of the ESCB and of the ECB. Four main areas of independence are considered: i) statutory, ii) functional, iii) economic and iv) personal. Statutory independence The constitutional status of the ECB is one of main sources of independence. The ECB is the only central bank for which the requirement of independence and the primary objective to 'maintain price stability' have a constitutional origin. The Maastricht Treaty has the same legal ranking as that of a constitution. Its adoption by the member states of the European Union has, in some countries, implied a constitutional amendment. A modification to the statute would require a new treaty between the member states, which would have to be agreed and ratified by all.3 The explicit reference to price stability as the main objective of the ECB - rather than a generic reference to monetary stability or to the safeguarding of the currency as is found in other central bank statutes - is a further guarantee of independence for the ECB in the event of conflict with other organs of government.4

How can the Eurosystem become independent?

121

If money is broadly defined as any one of various forms of financial wealth, the stability of its value does not necessarily bring about price stability. In fact, under certain conditions, price stability and the overall stability of monetary and financial wealth may even be in conflict. For instance, a restrictive monetary policy aimed at achieving price stability can induce a temporary reduction in the value of financial assets. If the central bank has to give priority to the latter, its ability to conduct a restrictive monetary policy with a view to price stability may be impaired. Even the concept of monetary stability is not necessarily always consistent with the objective of price stability. Indeed, the value of money can be defined in at least three ways. The first is the internal value, or the purchasing power of money measured in terms of the general level of prices. Under this definition, monetary stability and price stability do coincide. They do not coincide, however, in the context of a definition of monetary stability that refers to the external value of the currency, expressed in terms of the exchange rate. Exchange rate stability is not necessarily always consistent with price stability, especially in the short and medium term, since exchange rate developments can depart from relative inflation performance. This inconsistency was at the origin of the conflict between the Deutsche Bundesbank and the Federal Government in the late 1960s over the appropriate DMdollar parity in the Bretton Woods system. The final decision to revalue the Deutsche Mark, long requested by the Bundesbank, struck the balance in favor of the internal definition of monetary stability. The third concept of monetary stability is one measured in terms of the opportunity or the inter-temporal value of money, that is, the rate of interest. Under this definition, monetary stability implies interest rate stability. This concept was clearly in the minds of the drafters of the Federal Reserve Act, which specifies that the task of the Federal Reserve is to 'furnish an elastic currency'. The 1946 Employment Act further specifies that the Federal Reserve should set its monetary targets 'so as to promote effectively the goals

122

Open Issues in European Central Banking

of maximum employment, stable prices and moderate interest rates'. Under this definition, monetary and price stability do not necessarily coincide, as a stable rate of interest may give rise to inflation in the face of nominal shocks. The Statute of the ESCB and of the ECB does not give a specific definition of the concept of price stability. Therefore, unlike some central banks that have recently been given independence such as the Central Bank of New Zealand - the ECB, like the Bundesbank, may itself choose the specific inflation path it seeks to achieve. The ECB is allowed to exercise independence not only with respect to its instruments, as will be described below, but also with respect to its short-term goals, as long as they are consistent with the concept of price stability. This does not imply, however, that the ECB can easily change the definition of price stability without diminishing its credibility. The ECB has defined price stability as an increase in the price level less than 2 per cent. It has also clarified that a reduction in the price level would not be consistent with price stability, to make sure that deflation is viewed as being as much an evil as inflation. The ECB has the necessary independence to achieve price stability, on average, over the medium term. A change in the definition, if this were decided, would thus be interpreted as a threat to the independence of the ECB. Functional

independence

The ECB is provided with a wide range of instruments and procedures to perform its functions with a high degree of independence. In practice, there are no limitations, to the extent that 'the principle of an open market economy with free competition' (Article 2 of the statute) is respected. This principle, which is also mentioned in Article 3 a of the Maastricht Treaty, gives constitutional status to the concept of free competition. This latter article was meant to prevent the use of administrative measures by the ECB. It also protects the ECB from possible pressures to

How can the Eurosystem become independent?

123

introduce such measures that would distort the allocation of financial resources in favor of the government. Several prohibitions further reinforce this independence. The prohibition on monetary financing by the government, in particular through overdraft facilities or transactions in government securities in the primary market, represents an unprecedented severance of monetary links between the central bank and the government. This has implied modifying central bank legislation in all EU countries to bring it into line with that of the ECB. The norm is clearly not sufficient in itself to ensure independence, but it establishes an important principle that all countries and governments must respect. Economic

independence

The economic independence of the ECB is founded on a set of important factors. We mention just a few here. The ECB can decide autonomously on the type of assets, and their valuation method, that banks can use as collateral to refinance themselves with the Eurosystem. The foreign reserve assets of the member countries are owned by NCBs and part has been transferred to the ECB. The ECB Governing Council decides whether and in what form further foreign reserve assets may be called upon. The accounting rules for implementing the financial accounts of the ECB are established by the ECB Governing Council. The rules for establishing, paying up and raising the ECB capital are the responsibility of the ECB Governing Council. The external auditors are to be 'approved by the Council' on a recommendation by the ECB Governing Council. And finally, the provision for the allocation of profits and losses within the Eurosystem leaves the latter wide margins of maneuver, as examined in Chapter 4. It is difficult to assess the overall impact these provisions have on the ability of the ECB to perform its tasks in an independent manner. It is recognized, however, that the ECB enjoys a very high degree of financial independence compared to most other central banks.5

124 Personal

Open Issues in European Central Banking independence

The provisions of the statute grant the members of the ECB's governing bodies wide personal independence. Without seeking to be exhaustive, the following aspects are worth underlining. A first important notion is the principle of 'one man-one vote', already mentioned in Chapter 1. Irrespective of his or her country of origin, each member of the ECB Governing Council has the same vote.6 The Governor of Luxembourg's NCB has the same vote as the President of the Bundesbank, despite the fact that Germany is more than 150 times the size of Luxembourg in terms of income and population. This principle, which was adopted at the time the first draft of the ECB Statute was being prepared by the Committee of Governors in mid-1991, creates the presumption that a member of the governing body will take decisions in the interest of the whole Union, rather than solely on behalf of his country. A second aspect is the security of tenure for the members of the Executive Board (8 years),7 who must perform their functions on a full-time basis. Their salaries are set by the ECB Governing Council. The members of the Executive Board are nominated through a complex procedure, requiring unanimity of the Heads of State or of Government, on a recommendation from the Council. The ECB Governing Council and the European Parliament have to be consulted. The members of the Executive Board are explicitly required by the statute to be 'persons of recognized standing and professional experience in monetary or banking matters'. Although the nomination of the first ECB President has raised some doubts on the willingness of political authorities to respect the independence of the ECB, subsequent developments have been reassuring with respect to the personal independence of the Executive Board. A third aspect is that the ECB Governing Council is fully responsible for the decisions concerning the ECB's tasks and functions. A representative from the Commission and the President of the Council may participate in the meetings of the ECB Governing Council, but neither has the right to vote nor to suspend a decision,

How can the Eurosystem become independent?

125

which was a prerogative for example of the German and Dutch Ministers of Finance before monetary union.8 The ECB's independence may be undermined as a result of the potential size of its Governing Council, as mentioned in Chapter 1. If all the present members of the European Union were to participate in monetary union, the Governing Council would number 21 persons. When the next round of enlargement is completed, the number could increase to 30 before the end of the first decade of the 21 st century. Experts in management organization consider fifteen to be the maximum number for efficient decision-making. Beyond this number, efficiency may be compromised and commitment to a common goal weakened. In addition, as the number of countries increases, the number of central bank Governors will further outnumber the members of the Executive Board. This has been cited as a possible danger to the independence of the ECB9 and is recognized by many as a highly sensitive political problem which, as mentioned in Chapter 1, will have to be dealt with at a certain stage. 5.2

Measuring the ECB's independence against that of other central banks

In recent years, a growing literature has tried to analyze the degree of central bank independence (CBI) on the basis of quantitative indicators. Measures of CBI have generally been based on a set of objective criteria, referring to the legal and economic status of the central bank. A survey of such measures is provided by Eijffinger and de Haan (1996). Cukierman (1992) provides one of the most in-depth analyses on this subject. On the basis of sixteen criteria, Cukierman makes a quantitative ranking in four main areas (appointment of the bard, policy formulation, policy objectives and financing of the government), and devises an index of legal independence applied to the central banks of several industrialized countries and LDCs. The Deutsche Bundesbank and the Swiss National Bank emerged as the two central banks with the highest degree of legal independence,

126

Open Issues in European Central Banking

followed, among others, by the Federal Reserve. This ranking was similar to that of other measures presented in the literature.10 Cukierman (1992) ran regressions aimed at testing whether central bank independence had an impact on inflation. For industrial countries, legal independence of the central bank was found to have a negative effect on average inflation over the period 1950-89. These results are in line with those obtained in other research in this field, although the degree of significance of the relationship between CBI and inflation varied widely.11 The methodology followed by Cukierman to measure CBI can be implemented as a first step in analyzing the independence of the ECB. Table 5.1 reports, in the first three columns, Cukierman's results for the Bundesbank, the Swiss National Bank and the Federal Reserve. The results for the ECB, obtained on the basis of the ECB Statute, are reported in the fourth column. The results show that the ECB scores highest in terms of legal independence, higher than the other three central banks considered by Cukierman to be among the most independent. The ECB has also a higher score for the formulation of monetary policy, given the protection from interference granted by its statute and the prohibition of monetary financing of the government. It has a lower score than the Bundesbank, however, for the shorter average term of office of the ECB Governing Council.12 The ECB and the Bundesbank are granted an equal independence with respect to the objective of price stability and the powers to resolve potential conflicts with political authorities on monetary policy matters. The straightforward application of Cukierman's criteria suggests that the ECB should be among the most, if not the most, independent central bank in the world.13 Extending Cukierman's empirical results on the relationship between central bank independence and inflation performance, one would expect, ceteris paribus, that the inflation performance of the ECB should be as good as if not better than that of the Bundesbank or the Swiss National Bank.14

Table 5.1

Ranking of central bank independence (max=l; min=0) Weight

Bundesbank

Swiss National Bank

Federal Reserve

ECB

Bundesbank after 1994

1. Chief Executive Officer a; Term in office b. Appointment of CEO c. Dismissal d. CEO is allowed to hold other offices

0.05 0.05 0.05 0.05

1 0.075 1 0

0.75 0.25 n.a. 1

0.25 0.5 0 1

0.75 0.75 0.83 1

1 0.75 1 0

2. Policy formulation a. Who formulates monetary policy b. Who has final word in resolution of conflicts c. Role in Government's budgetary process

0.05 0.05 0.05

0.67 1 0

n.a. 1 0

n.a. 0.2 0

1 1 0

0.67 1 0

Variable

Ci

O

3 S'

I * 5S

to <1

Table 5.1, cont.

Ranking of central bank independence

Variable 3. Objectives (price stability)

Weight

Bundesbank

Swiss National Bank

0.15

Federal Reserve

ECB

Bundesbank after 1994

0.4 fX5

4. Limitations on lending to the government a. Advances b. Securitized lending c. Terms of lending d. Potential borrowers e. Definition of limits f. Maturity of loans g. Interest rates h. Buying in primary markets

0.15 0.1 0.1 0.05 0.025 0.025 0.025 0.025

Total

Source: Elaboration on Cuckierman (1992).

0.67 0.67 0.67 0.3 1 1 0.25 0

0.67 0.33 1 1 n.a. 1 0.25 0

1 0.33 0.33 1 n.a. 1 0.25 0

1 1 1 1 1 1 1 1

1 1 1 1 1 1 1 1

0.67

0.68

0.51

0.91

0.87

Si

How can the Eurosystem become independent?

5.3

129

The limits of measurement

The application of the criteria developed in the literature to assess the independence of the ECB shows how difficult it is to evaluate CBI. It is indeed difficult to define a complete list of criteria that can provide a precise picture of the legal status of a central bank. Even in a wide list, such as the one suggested by Cukierman (1992), important elements are omitted. One notable omission concerns the economic independence of a central bank, which has implications for the ability of the central bank to conduct monetary policy operations. Rules on the use of collateral and the procedure for allocating profits and losses may have an impact on the room for maneuver available to the central bank. These aspects are practically ignored in the literature. Another important element affecting central bank independence, which is also often overlooked, is the participation in decisions concerning the exchange rate regime and policy. In a small country, an independent central bank may have little control over monetary policy if it does not participate actively in the definition of the exchange rate policy.15 A further difficulty derives from the interpretation of legal provisions. The ranking used by Cukierman (1992) can be challenged on several grounds. For instance, he suggests that the German central bank participated in the formulation of monetary policy but did not have full control over it; on the other hand, the bank was considered to have the final word in the resolution of conflicts with the government, at least on issues defined by law as being directly related to its objectives. A close reading of the Bundesbank's Statutes and the historical experience of Germany's monetary policy suggests a slightly different interpretation. The Bundesbank has always had a high degree of autonomy in the formulation of monetary policy. Its statutes clearly assign to it the task of defining Germany's monetary policy with the aim of safeguarding the currency. On this criterion, the Bundesbank should therefore have been given a higher score by Cukierman. On the other hand, concerning the resolution of conflicts with the government on

130

Open Issues in European Central Banking

issues of common interest, the Bundesbank Act did not clearly specify who had the last word. The government has the right to suspend a decision of the Bundesbank Council for two weeks, although this has never happened. Furthermore, the government had the final word on exchange rate policy. On the basis of this latter criterion, therefore, a strict interpretation of the law suggests that the Bundesbank should have received a lower ranking than that assigned it by Cukierman. An additional problem with the creation of a CBI index is the interpretation to be given to certain criteria. According to Cukierman, for example, a nomination procedure that envisages that the Executive Board is nominated by the executive branch of the government reduces the degree of CBI, with respect to a more complex procedure involving other public bodies. This judgement is based on the assumption that the first procedure leads to the appointment of persons that are more favorably inclined towards the government. This assumption is not necessarily correct. If the nomination procedure is too complex, nobody will be held responsible in case the nominee proves to be unfit or insufficiently independent. Another example is the involvement of the central bank in the government's budgetary process. According to Cukierman (1992), a central bank that participates in the government's budgetary process is more independent. The opposite view seems more defensible. To preserve independence, a central bank should abstain from participating in the budgetary process so that it can more freely express a public judgement on it and retain autonomous margins of maneuver for its monetary policy. Overall, it is rather difficult to assess independence only on the basis of legal texts. A country's historical experience and national customs can have as much or even more influence than the law in defining the boundaries of central bank action. A central bank may de facto be more independent than it appears on the basis of legal texts, because its publicly recognized status enables it to act in a way that is more independent.

How can the Eurosystem become independent?

131

Let us once again take the Bundesbank as an example. Prior to 1992, its statutes contained few limitations on lending to the government and did not prohibit interventions in the primary market for government securities, thereby suggesting a low degree of independence with respect to the monetary financing of the government. On this criterion, the Bundesbank was given a relatively low marking in Cukierman's index (Table 5.1). Experience has shown, however, that, although there was no explicit prohibition on lending to the government, monetary financing remained historically very low. The Bundesbank Act, in particular Article 20, was revised in 1993, following ratification of the Maastricht Treaty, and monetary financing was formally forbidden. On the basis of Cukierman's classification, this revision has led to an increase of the CBI index by nearly 30 per cent. It is difficult to believe that the independence of the Bundesbank has effectively changed as a result of the new law. In summary, the independence of a central bank cannot be assessed outside its historical context, the general economic environment and the political experience of a country.16 Applying the legal status of a central bank in one country to that in another does not necessarily bestow the same degree of independence. In practice, central bank independence is established through a sequence of specific events, possibly entailing conflicts with other organs of government. The way these conflicts are resolved marks the boundaries of central bank independence and defines the limits of the central bank's scope for action. Conflicts, or potential conflicts, with other government bodies are intrinsic to central bank activity - even in the case where the central bank's primary objective is to maintain price stability. The reason is that price stability cannot be achieved at each and every point in time. The literature on CBI, which starts from BarroGordon's simplifying assumption that the central bank has full control over the inflation rate at each point in time, ignores this fact. If the assumption is removed, especially for the short term, some margins of discretion emerge in central bank action, which creates

132

Open Issues in European Central Banking

the scope for conflict with other organs of government. The way in which conflicts are resolved, or are expected to be resolved, is the true indicator of CBI. Concerning the ECB, it is difficult to attempt making an assessment based on much more than the legal texts. The experience of the first few months of monetary union suggests that the independence of the ECB has been strengthened from the first conflicts that have emerged in the public arena with some finance ministers, concerning in particular interest rate and exchange rate policy. In spite of the pressures, the ECB has kept a firm hand. Nevertheless, other, more subtle threats should not be ignored. 5.4

Potential threats to the ECB's independence

The exchange rate regime Several commentators have indicated that the allocation of responsibilities, in particular between the ECB and the Council of EU Ministers, concerning the conduct of the foreign exchange policy of the euro may restrict the independence of the ECB.17 Article 109 of the Maastricht Treaty, which defines the responsibilities in the field of the foreign exchange policy of the Union, is rather complex and required lengthy negotiations. Two cases are considered. First, the Council of Ministers may, acting unanimously, conclude formal agreements on an exchange-rate system for the euro vis-a-vis non-EU currencies, although it is obliged to have consulted the ECB 'in an endeavor to reach a consensus consistent with the objective of price stability'. The Council of Ministers may then, by a qualified majority, 'adopt, adjust or abandon the central rates of the euro within the exchange rate system'. This is the extreme, and at present, highly unlikely case. Under the more likely scenario that there is no formal agreement with non-EU countries or international institutions, the Council of Ministers may, by a qualified majority, 'formulate general orientations for exchange-rate policy...without prejudice to the primary objective.. .to maintain price stability'.

How can the Eurosystem become independent?

133

The provisions of the treaty suggest that the ECB is unlikely for a long time to be obliged to intervene in exchange markets against the dollar or the yen since unanimity in the Council to enter into explicit commitments in a Bretton Woods-like agreement with the United States, Japan and other nations, is rather difficult to reach. Furthermore, any move in this direction would require the assent of the other global players, which seems highly unlikely under current circumstances. A much less formal (possibly non-publicized) understanding of the type exemplified by the 1986 Louvre Accord would hardly be enough to pose a serious challenge to the emphasis given to price stability as the primary objective. Over the wide range of policies that stop short of 'formal agreements', the ECB is likely to retain the decisive influence over the external aspects of monetary policy in EMU. The treaty goes further towards limiting political influence over the central bank than analogous institutional arrangements in the United States and Japan. In the two other main industrialized countries, the authority for exchange-rate policy is effectively shared between the political authorities and the central bank. Both in the United States and Japan, conflicts have occasionally arisen over the desirability of influencing the exchange rate as well as over the appropriate means. This was especially the case during the 1980s. There is little doubt that in such conflicts the political authorities represented by the US Secretary of the Treasury or Japanese Minister of Finance have tended to have the upper hand in setting exchange-rate policy, even during the long span since 1973, when formal obligations in the international monetary system have been at a minimum.18 The political constraints on conducting foreign exchange interventions are also highly visible in the United States and Japan, since international reserves are owned by the respective Treasuries, which are accountable to Congress and the Diet. Some authors have used the term 'mutual veto', that is, that the central bank and the Ministry of Finance can both exercise a veto on such operations.19 This has typically implied fewer interventions (or other efforts at

134

Open Issues in European Central Banking

international coordination) than central banks would have preferred, because political authorities have tended to put a relatively greater weight on purely domestic objectives. In Europe, the situation seems to be quite different. The relative preferences seem to be in the opposite direction. The ECB is likely to be more single-minded in the pursuit of price stability than the Federal Reserve, whereas the participants in the Council of EU Ministers - some of whom are quite sympathetic to more active management of the relationship to the dollar - will occasionally want more attention paid to exchange-rate stability. Some important finance ministers have already made their preferences known. In contrast to the US, the Council majority cannot simply impose a decision on a reluctant ECB Council. The Council needs to persuade the central bankers in a consultative procedure. It cannot be excluded that, in case of conflict, the ministers would on some occasions prevail, thus threatening the ability of the ECB to pursue its mandate at all times. Overall, the institutional provisions make it unlikely that the Eurosystem will be forced into an exchange rate regime that limits its capacity to maintain price stability. The economics of exchange rate regimes provide further reassurance that we need not become too concerned about this threat. The key point is that the constraint exerted on monetary policy depends on the nature of the exchange rate regime. The constraint can be very strong in the case of a unilateral peg. This applies to a small country, which accepts the burden of adjustment to ensure exchange rate stability and therefore accepts severe constraints on its monetary policy. Such a system can hardly be foreseen for the Eurozone. It is unlikely that the latter would agree to participate in an exchange rate regime where the euro is unilaterally pegged to a foreign currency, the Eurosystem bearing the burden of adjustment. While this was the case under the Bretton Woods system, in which European currencies were unilaterally and separately pegged to the US dollar, such a situation cannot be imagined for the euro. This would hardly be rational from an economic or political point of view.

How can the Eurosystem become independent?

135

In a multilateral exchange-rate system, the burden of adjustment falls on the weakest currency, that is, the currency of the country with persistently higher inflation. In a system of fluctuation bands, with intervention obligations for all countries, the limited availability of reserves means that the country whose currency experiences downward pressure sooner or later has to adjust or devalue. Even when the possibility of financing interventions by borrowing reserves exists, reserves have in any case to be reimbursed some day. This means that in practice there can be no unlimited financing, except in a monetary union. A country whose currency is strong is not affected by other countries' interventions, if the effects of capital inflows on the money base are fiilly sterilized. This depends on the amount of assets available to the central bank. The experience of the EMS has shown that the Bundesbank was always able to sterilize the impact of foreign interventions on its operational target, although the German GDP accounted only for about one-third of that of the EMS. Since the Eurozone is as large as the US and several times larger than the economies of the other European countries that might be linked to the euro, there are few risks for the Eurosystem to participate in an exchange rate system. The key corollary of this exchange-rate system asymmetry is that it is highly unlikely that the ECB could ever be forced to loosen monetary policy through the participation in a multilateral monetary system. Any such system could at most discipline its weakest members, and this is unlikely to be the Eurozone. The new European Monetary System, that links pre-in currencies to the euro, is not discussed at length here. The asymmetry mentioned above fiilly applies in this case, since the Eurozone would always be the predominant economy. The euro is the anchor of the system. There is thus no reason to fear that the EMS2 could ever influence the monetary policy chosen by the ECB. At present, the EMS2 embraces only Denmark and Greece. The question would thus arise only if and when the United Kingdom will join for a trial period preceding full membership.

136

Open Issues in European Central Banking

NCB activities 'outside9 the ESCB As mentioned in Chapter 1, NCBs are allowed to continue performing functions that are not directly related to the tasks of the Eurosystem. Such functions, however, must not interfere with the conduct of the single monetary policy. A determination that interference has occurred must be established by a two-thirdsmajority of the ECB Governing Council (Article 14.4 of the Statute). NCBs continue to perform a series of functions that they performed before monetary union. The assessment of the potential for interference is very difficult. Interference implies that the action conducted by the NCB 'outside' the Eurosystem has an impact on the conduct of monetary policy, in particular through the creation of central bank money or the signaling of a future ECB decision to financial markets. The potential for interference may be limited in the Eurosystem. A single action by an NBC is unlikely to have a significant impact on the overall monetary policy of the Eurosystem. However, precisely because the interference potentially has so little significance, there are few incentives to address the conflict of interest. This gives rise to an externality. The aggregate result may be that several NCBs will behave in the same way on the assumption that their individual action will not constitute an interference. Instead, however, the overall collective result may create a significant interference. An illustration may help to clarify the issue. For relatively small NCB, such as for example the Banque de Belgique, a transaction of 1 billion euro conducted in the management of its portfolio would strongly interfere with the Belgian money market and, prior to monetary union, would have been undertaken with great caution. In the euro money market, however, a transaction of such a size would have little impact on the overall monetary conditions and could be easily sterilized, even at short notice. Such a transaction should thus be authorized. Nevertheless, if several NCBs take the opportunity of conducting transactions of a similar size, the overall result may become significant for the euro money market.

How can the Eurosystem become independent?

137

NCBs maintain the freedom to manage domestic assets that are not part of the Eurosystem refinancing operations and foreign exchange reserves that they have retained after the transfer to the ECB. As suggested in Chapter 3, however, the overall amount of these assets in the Eurosystem is as large as the money base. NCBs have no reason to hold these assets other than the search for profit, since the outright purchase or sale of securities is not one of their main monetary policy instruments. The question to ask is whether the management of these assets, in the search of a competitive return, can interfere with the conduct of monetary policy. To what extent can decisions taken in the performance of the former function interfere with the latter, and vice-versa? Will the ECB Governing Council take interest rate decisions, for example, in a completely independent way with respect to the repercussions that these will have on NCBs' profit and loss accounts? The same question may be asked with respect to a decision to intervene, or not to intervene, in the foreign exchange market, with a view to affecting the exchange rate of the euro. Within a single central bank, such a conflict of interest would normally not arise so acutely because the domestic and foreign assets are held mainly with a view to monetary and foreign exchange operations. The management of domestic and foreign assets is closely associated with monetary policy criteria, in particular with a view to insuring the liquidity of investments. This may be a reason for accepting lower rates of return on these assets. In the Eurosystem, however, assets held by NCBs have no relation to the conduct of the single monetary policy. NCBs may thus be criticized, as suggested in Chapter 3, if their asset management turns out to under-perform market benchmarks. NCBs are thus naturally led to use all the available information, including that derived from their contribution to the single monetary policy, to maximize profits. They are also led to try to undertake transactions motivated by the latter goal, even if they interfere with the single monetary policy. Other activities interfering with the conduct of the single monetary policy have been examined in previous chapters, the most

138

Open Issues in European Central Banking

notable being prudential supervision. Chapter 2 surveys the literature that examines this type of interference, and we therefore do not revisit the arguments here.20 In the Eurozone, several NCBs are closely involved or associated with the responsibility for prudential supervision. This may lead to interference with the single monetary policy decision-making, even though the Eurosystem itself has no direct responsibility over prudential supervision. The performance of the lender-of-last-resort function may also impinge on the single monetary policy. Chapter 2 describes the state of uncertainty underlying the way in which this function is supposed to be performed in EMU. If this function is conducted outside the Eurosystem, the potential for interference increases, as a result of the externalities mentioned above. Finally, the performance of other functions related to the supervision of the money and financial markets, including the promotion of domestic markets, which is actively pursued by some NCBs inside and outside the Eurozone, is likely to have an impact on the conduct of the single monetary policy. Chapter 1 provides some examples of the way in which these considerations have affected the design of the instruments and procedures related to the single monetary policy. Conclusions The Statute of the ESCB and of the ECB has been drafted with a view to granting the European Central Bank a high degree of independence in the performance of its functions. It is generally agreed that the ECB is indeed one of the most independent central banks in the world. The treaty provides adequate protection against the classic threats to central bank independence, such as the participation in an exchange rate regime. The Stability and Growth Pact reinforces the protection against pressures for excessive public spending or financial market turmoil resulting from doubts about the ability of governments to honor their obligations.

How can the Eurosystem become independent?

139

The main danger to the ECB's independence originates from within the Eurosystem itself. It comes from potential conflicts of interest that may arise between the single monetary policy and the other policies and functions that the NCBs are allowed to continue to perform allegedly 'outside' the Eurosystem. The carrying out of these functions, about which little is disclosed to the public, may jeopardize the integrity of the single monetary policy and may lead to conflicts of interest. These conflicts could be minimized by limiting the financial activities of NCBs along the lines put forward in Chapter 3, with a view to protecting the independence of the Eurosystem. The most insidious threat is also the most difficult one to evaluate. As Posen (1993) argues, central bank independence does not just happen by chance. It is the expression of a political system in which independence is generally agreed to be in the interest of society. The high inflation rates of the 1970s and early 1980s convinced society, especially opinion leaders and financial markets, that a tight, anti-inflationary monetary policy was needed and could only be ensured by an independent central bank. If the euro area continues to perform unsatisfactorily in terms of growth and unemployment and if the ECB were perceived to be partly responsible because of an excessively tight monetary policy, its independence would be questioned in public opinion. After all, such changes of opinion are not new in history. The movement away from the gold standard (and the central bank independence it implied) was due to the fact central bankers had failed to combat, if not contributed through wrong policies, to the depression of the 1930s. The legal protection for the independence of the ECB is theoretically very strong. It would turn out to be useless, however, if the ECB were perceived as not contributing to what society wants, which is ultimately growth and prosperity. Therefore, price stability in a stagnating economy poses as great a threat to central bank independence as do inflationary pressures.

140

Open Issues in European Central Banking

References to Chapter 5 Alesina, A. (1988), Macroeconomics and Politics, NBER Macroeconomics Annual, Vol. 3, pp. 13-52. Alesina, A. and V. Grilli (1992), 'The European Central Bank: Reshaping Monetary Policies in Europe', in M. Canzoneri, V. Grilli and P. Masson (eds.), Establishing a Central Bank: Issues in Europe and Lessons from the US, Cambridge University Press, Cambridge. Bade, R. and M. Parkin (1988), Central Bank Loan and Monetary Policy, Department of Economics, University of Western Ontario. Barro, R. and D. Gordon (1983), 'Rules, Discretion and Reputation in a Model of Monetary Policy', Journal of Monetary Economics, Vol. 12, pp. 101-122. Begg, D. and others, (1992), 'European Monetary Union: The Macro Issues, in Monetary European Integration: The Making of Monetary Union, CEPR. Ciocca, P. (1992), 11 principio d'autonomia nel Central Banking', Quaderni di economia efinanza, dicembre, No. 1, pp. 5-22. Cukierman, A., S. Webb and B. Neyapti (1992), 'Measuring the Independence of Central Banks and Its Effect on Policy Outcomes', World Bank Economic Review, Vol. 6, No. 3, pp. 353-398. Cukierman, A. (1992), Central Bank Strategy, Credibility and Independence: Theory and Evidence, MIT University Press, Cambridge; Mass. Eijffinger, S. and E. Schaling, (1993), 'Central Bank Independence in Twelve Industrial Countries', BNL Quarterly Review, No. 184, pp. 64-68. Eijffinger, S. and J. de Haan (1996), The Political Economy of Central Bank Independence, Special Papers in International Finance, No. 19, Princeton University Press Eizenga, W. (1987), The Independence of the Deutsche Bundesbank and the Nederlandsche Bank with regard to Monetary Policy: a Comparative Study, SUERF Papers on Monetary Policy and Financial Systems, No. 2, Tilbuy.

How can the Eurosystem become independent?

141

Fischer, S. (1995a), Modern Approaches to Central Banking, NBER Working Paper No. 5064, Cambridge. Fischer, S. (1995b), 'Central Bank Independence Revisited', American Economic Review, Papers and Proceedings, Vol. 85, No.2, May, pp. 201-206. Funabachi, Yoichi (1988), Managing the Dollar: From the Plaza to the Louvre, Institute for International Economics, Washington, D.C. Giovannini, A. (1992), Central Banking in a Monetary Union: Reflections on the Proposed Statutes of the ECB, CEPR Occasional Paper No. 9. Goodhart, C. (1992), The Draft Statutes of the European Central Bank, London School of Economics, Special Paper No. 37. Grilli, V., D. Masciandaro and S. Tabellini (1991), 'Political and Monetary Institutions and Public Financial Policies in the Industrial Countries', Economic Policy, Vol. 13, pp. 341-92. Kenen, P. (1992), EMU after Maastricht, Group of Thirty, Washington, D.C. Marsch, D. (1992), The Bundesbank, Heinemann, London. Posen, A. (1993), 'Why central bank independence does not cause low inflation: There is no institutional fix for polities', Finance and the International Economy (The Amex Bank Review), Oxford University Press, Oxford. Tabellini, G. and T. Persson (1994), Monetary and Fiscal Policy, Vol.1: Credibility, M.I.T. University Press, Cambridge, Mass. Volcker, P., (1994), 'L'indipendenza delle banche centrali: valori e limiti, II diritto e l'economia,' No. 2, pp. 279-294.

Notes to Chapter 5 1. See, for instance, Alesina and Grilli (1992), Cukierman (1992), Eijffinger and de Haan (1996) and Fischer (1995a and 1995b). 2. See Tabellini and Persson (1994). 3. For a restricted number of articles that do not concern the main tasks and functions of the ECB or the principle of independence, a more

142

Open Issues in European Central Banking

flexible procedure is possible. This requires a qualified majority of the Council when the proposal is made by the ECB itself (subject to unanimity of the Governing Council of the ECB) or unanimity when the proposal is made by the Commission (see Article 41 of the Statute of the ESCB and of the ECB). 4. See Ciocca (1992). 5. See Goodhart (1992). 6. Except for certain provisions relating to financial resources, which are mentioned in Chapter 1. 7. Except for the initial appointments, which envisage a different length of the term of office for the Board members, to provide for a staggered replacement. The term of C. Noyer expires in June 2002, in June 2003 for D. Solans, in June 2004 for S. Hamalainen, in June 2005 for T. Padoa-Schioppa and in June 2006 for O. Issning. 8. A decision by the Bundesbank Council could have been suspended for two weeks. See Eizenga (1987). This possibility was never resorted to in practice but, according to some commentators (Marsch, 1992), it has at times acted as an implicit deterrent. 9. See Goodhart (1992). 10. See, among others, Bade and Parkin (1980), Alesina (1988) and Grilli, Masciandaro and Tabellini (1991). 11. See Alesina (1988), Grilli, Masciandaro and Tabellini (1991), Eijffinger and Schaling (1993), Fischer (1995 a) and Eijffinger and de Haan (1996). 12. The term of office for the members of the Executive Board is eight years, but the term of the other members of the ECB Governing Council is determined by their respective national legislation, with a minimum of five years, as opposed to eight years in the Bundesbank. 13. This is in line with the analysis of Alesina and Grilli (1992) using another indicator of CBI. 14. During the 1980s, Germany and Switzerland recorded an average yearly inflation rate of around 3 per cent. 15. This point has been made by several authors, including Fischer (1995). 16. See Volcker (1994).

How can the Eurosystem become independent?

143

17. See Begg et al. (1991), Giovannini (1992) and Kenen (1992). 18. As demonstrated in the case of the US by Destler and Henning (1989), and for Japan by Funabashi (1988). 19. See Destler and Henning (1989), p. 88. 20. Some authors have constructed CBI indicators in which the conduct of prudential supervision is considered a factor that reduces independence. See Grilli, Masciandaro and Tabellini (1991).

6.... AND ACCOUNTABLE? Introduction The previous chapter suggested that, in spite of several threats, the European Central Bank might be one of the most independent central banks in the world. Such an unprecedented degree of independence is considered by some, however, as a cause for concern unless it is accompanied by adequate accountability. This chapter examines the extent to which the ECB is an accountable institution. The first two sections explore the concept of accountability. The third tries to define the acts and procedures through which central banks perform their accountability. The fourth section compares the accountability of the ECB with that of three other central banks: the Federal Reserve Bank of the US, the Bank of Japan and the Bank of England. The aim is not to rank central banks according to a predetermined index of accountability. Accountability can be performed in different ways and through different practices and procedures, depending on the interaction between the central bank and the other institutions of society, their respective roles and the environment within which they interact, in particular the structure of financial markets and the instruments used by the central bank in the conduct of monetary policy. There is thus no one single model of accountability that can be applied to all central banks in all cases and circumstances. It is nevertheless of importance to assess the extent to which the ECB gives an adequate account of its activities 6.1

Democratic control, independence and accountability

As in the previous chapter, this one builds on the literature on central bank independence and legitimacy which has been widely developed in recent years. All caveats, related in particular to the limitations 144

... and accountable?

145

and simplifications of the theoretical underpinnings of this line of research, must be kept in mind in reading both of these chapters. A couple of clarifications may be worth making as a preliminary step towards the analysis of central bank accountability: the first concerns the concept of accountability itself; the second, its relationship with independence. The concepts of democratic control and democratic accountability1 are often confused. Democratic control refers to the following three constraints on the exercise of government:2 •

Ex-ante control defines the rules, standards and principles laid down in advance by a democratically elected body, to be followed by the accountable body in the exercise of its functions.

•

Accountability is the act of listening to criticism and responding to questions about the past and future behavior that may be put forward by a democratically elected body.

•

Popular mandate refers to the attribution of power through democratic procedures.

The above criteria, which define the way democratic control is generally exercised, cannot be fully applied to central banks. The third criterion, in particular, is not relevant in that independent central banks do not receive popular mandates nor do they choose their own policy objectives. This would imply that the central bank had become a new separate branch of democratic power. Instead, central banks receive their mandate from the government or through a legal act, to perform specific tasks in view of specific objectives. The first two criteria (ex-ante control and accountability) however, do apply to central banks. The way in which accountability is exercised is closely related to the concept of independence of the central bank. The exante control over a central bank which is not independent can be effected in the context of the general conduct of the government's economic policy. The rules and principles underlying such control may be constantly revised, depending on the circumstances faced by the executive branch of government and on its policy goals. For instance, monetary policy may be geared in turn towards the objective of price stability or towards stimulating economic growth,

146

Open Issues in European Central Banking

depending on the wishes of the government. There is no need in this case to precisely specify in advance the tasks and objectives of monetary policy. Before its recent reform, the Bank of England was not subject to precise rules or guidelines, as suggested by Roll et al. (1993): The so-called Parliamentary accountability for monetary policy connotes no more that the presence of the topic in a general and continuing Parliamentary debate about the government's economic performance, a debate whose real constitutional function is to furnish information relevant to the quinquennial popular control by elections.3 For an independent central bank, the way in which accountability is exercised needs to be precisely defined. Independence can be granted only if clear rules and principles are laid down to define the boundaries of central bank action. Ex-ante control is a necessary condition for delegating power. If the central bank could choose its own goals and policies, it would become a political body that could not be independent from government. The issue is thus to define the 'optimal contract' for the central bank in a satisfactory way.4 Such a contract must ensure that the central bank is given the appropriate incentives to pursue its statutory goals. The rules and principles for central bank action can thus not be continuously changed by the government according to specific and changeable interests. Accountability aims at verifying that the rules and principles laid down for the central bank are respected. The way in which accountability is performed is linked to the way ex-ante control is specified. If ex-ante control is defined with a low degree of precision, for instance including escape clauses, accountability can only be of poor quality. If the objective of a central bank is not clearly specified, the reporting will only be vague and will not allow any association between monetary policy decisions and the bank's final performance. When escape clauses are set, which allow the central bank under unspecified circumstances to deviate from its targets, bad performance can be easily justified ex-post. The public is not in a position to assess the effective contribution of the central bank to the achievement of the specific objective. This undermines the

... and accountable?

UI

accountability of the central bank. When the mandate of a central bank is clearly and precisely defined, the public can focus on its performance. Accountability can be exercised on the basis of precise performance indicators. In sum, contrary to the views expressed in some quarters, there is no a priori trade-off between independence and accountability, if the two concepts are appropriately defined. Accountability can be seen as a complement, if not a necessary requirement for independence. A central bank cannot be made fully independent if its objectives are not clearly and precisely defined; it cannot remain independent if it does not give a public account of its actions. Another issue that has been discussed in the literature is whether accountability can be distinguished from transparency. Accountability refers to the legal obligation to give an account of a bank's actions and performance, which derives from the delegation of power.5 Transparency goes beyond the fulfillment of a given reporting requirement and refers to 'more subtle forms of accountability' related to the 'way of doing business'. This distinction is to some extent artificial, at least in the case of the ECB, since its objectives and tasks are clearly defined in the Maastricht Treaty. Although the ECB has precise reporting obligations, its accountability can be understood in a very general sense, to be exercised not only to the Council or the Parliament, but to the public at large. This interpretation is consistent with Article 2 of the ECB Statute, which specifies that 'the ESCB shall act in accordance with the principle of an open market economy with free competition, favoring an efficient allocation of resources...' A broad interpretation of the concept of accountability, not restricted to legal requirements but extended to the broader concept of transparency and openness of monetary policy, seems thus appropriate for the European Central Bank. 6.2

The economics of central bank accountability

Central bank performance can be assessed by observing the results of its policies. If the primary objective of monetary policy is price stability, the assessment of central bank performance can be based

148

Open Issues in European Central Banking

on the regular observation of inflation statistics. Given the time lag between the setting of monetary policy and its actual effect on price levels, however, observation of the latter enables us to assess actions taken about two years earlier. This of course is of little interest to market participants and the public at large. Two further elements complicate the exercise of accountability. The first is that inflation is a monetary phenomenon in the long run, but not at each and every point in time. Inflation is affected by several variables, such as labor costs, import prices and taxes, which are beyond the control of the central bank and are difficult to forecast. The central bank can thus not be held accountable for temporary deviations from price stability which are not due to its own behavior. On the other hand, how can market participants know if such deviations are temporary or not, and whether they are due to monetary factors or to some other cause? These questions, which lie at the root of the problem of central bank accountability, can be regrouped into three main broad areas. The first is the well known issue of time inconsistency of the optimal monetary policy, already mentioned in the previous chapter.6 Although a price stability-oriented monetary policy is optimal in the long run, it is not optimal in the short term in the absence of unexpected shocks Welfare may temporarily increase if the central bank deviates from its long-term objective and conducts an unexpected monetary expansion that would bring about higher growth and lower unemployment, even at the expense of higher inflation. The incentive to deviate from the optimal long-term policy is known to economic agents, which anticipate this temptation. The central bank is thus led to conduct a more expansionary policy than desired, with a sub-optimal inflation result for society. The stronger is the reputation enjoyed by a central bank, the greater may be the gains obtained from temporarily deviating from its long-term objective. The problem is made more acute by the fact that in the short term the rate of inflation may be affected by other factors than just monetary policy, such as wage shocks and changes in public expenditure or import prices. The simple observation of the price level does not enable market participants to understand whether the

... and accountable?

149

deviation from price stability, if and when it occurs, is temporary or permanent and whether it is due to a non-monetary shock or to the intention of the central bank to effect a surprise monetary expansion with a view to stimulating the economy. In this context, the literature has analyzed the issue of central bank 'secrecy' or 'ambiguity'.7 If only unanticipated monetary policy is effective in influencing output and employment, disclosing partial information may be interpreted by market participants as the wish of the central bank to keep some information private, so as to retain the possibility of springing inflationary surprises.8 This line of reasoning has been used, in particular by the Federal Reserve, to justify the fact that information about the FOMC meetings should not be disclosed as requested by some market participants.9 This view may be justified for a central bank such as the Fed, which is not fully independent and whose primary objective is not price stability but also to support economic growth. A central bank that is obliged to focus on price stability as its primary objective, and thus cannot conduct an active stabilization policy, should benefit from disclosing information that might be used by market participants to interpret current developments. Accountability can be seen as a part of a commitment technology by which the central bank provides economic agents with symmetric information, and thus deprives itself of the possibility to follow a different policy from the one it has announced, with a view to enhancing the credibility of its action. It is noteworthy that the issue of accountability has been promoted in particular by central banks whose statutes have changed and that have no track record. To summarize the above arguments, transparency increases the credibility of the central bank, especially when the latter does not enjoy a particularly high reputation. It thus enables the central bank to conduct a less restrictive monetary policy than would otherwise be required.10 Transparency and accountability increase the overall welfare of the economy. The second argument for accountability is linked to the previous one and arises from the distributive effects produced by monetary policy decisions, in particular unanticipated ones. Unexpected interest rate changes can produce a significant

150

Open Issues in European Central Banking

redistribution of wealth between debtors and creditors. The small increase in the fed fund rate decided by the Federal Reserve in early 1994 was sufficient to wipe out most of the profits of the large investment banks for that year. Such redistributive effects may take place not only across sectors and agents, but also across regions. They may be exacerbated if different groups of society have different information about the intentions and the behavior of the central bank. If the central bank is dependent from the government, the distributive effects may be linked to explicit political decisions, and possibly compensated through other policy instruments. In this context, the secrecy surrounding central bank decisions can be justified as a way to achieve the desired income distribution objective. An independent central bank, which has no income distribution objective, must prevent a situation in which private information concerning its actions is made available to certain groups of society and not to others. This aspect is particularly relevant in a highly decentralized economy such as the European one, where the decision-making process may be seen as distant and may not be fully understood by some sectors of society. Without the necessary transparency, the suspicion may arise that the interests of some parties are given more attention than others.11 This would clearly be to the detriment of European Central Bank's credibility. To avoid this suspicion, information should be disseminated with similar intensity, frequency and depth across the economy. This can be more easily done in an atmosphere of disclosure and accountability than one of secrecy. The third reason for accountability is that it facilitates cooperative behavior between economic agents, thereby achieving Pareto-superior social welfare. In the presence of uncertainty, the availability of information from the central bank enables economic agents to extract information not only on aggregate developments but also on their own behavior, relative to the rest of society. Agents can thus modify their behavior or try to influence that of the other components of society. This argument applies in particular to decentralized economies, where agents have difficulty in coordinating their actions and expectations. For instance, if the ECB anticipated that the government of one of the large EU states was in

... and accountable?

151

the process of launching a major public expenditure plan, with potential inflationary pressures for the whole Union, it could consider increasing interest rates, possibly in a pre-emptive way. In this case, it would be desirable to publicly spell out such policy intentions, possibly in advance of the decision. This would give the government in question the opportunity to reconsider its policy action. The other countries would also have the opportunity to try to influence the undisciplined country, since the effects of the latter's policies would be borne by all. Indeed, the increase of the Union's interest rate would have restrictive consequences on all countries, including those 'that are fiscally disciplined'. The outcome of a reconsideration of the initial policy intention would lead to a more efficient solution than the one contemplated in the case of an ex-post reaction by the central bank.12 The same reasoning can be applied to other shocks, such as for instance asymmetric wage and price developments. Central bank transparency contributes to cooperative equilibrium between economic agents, which is a Pareto-improvement. Central banks are aware of this fact, and often use their communication channels with market participants to 'guide' them on possible policy developments. Central banks have at times even 'threatened' to increase interest rates if wage negotiations were concluded that were inconsistent with price stability or if budget projections were overshot. 6.3.

How to assess central bank accountability?

The easy way to assess central bank accountability is to create an indicator as in the literature on central bank independence. Briault et al. (1996) have followed this route and created an accountability index based on four criteria: a) whether the central bank is subject to external monitoring by Parliament; b) whether the minutes are published of the meetings in which monetary policy is decided; c) whether the central bank publishes an inflation or monetary policy report of some kind, in addition to the standard central bank bulletins; and d) whether there is a clause that allows the government to override a decision of the central bank. In our view, these criteria are unsatisfactory. They oversimplify the process and thereby give an incomplete picture of

152

Open Issues in European Central Banking

the framework within which accountability is exercised. In particular, the last criterion seems more appropriate for evaluating central bank independence than accountability. The possibility of overriding the central bank is unrelated to ex-ante control and thus to accountability. Cukierman (1992) includes this criterion in the index of legal independence. The third criterion is also dubious; it is not clear why an inflation report would necessarily ensure any more accountability than would be achieved through the regular monthly reports, if the latter contained the necessary information. We examine a different set of criteria, derived from the concepts of ex-ante control and accountability discussed in the first section. Our aim is not to create a numerical index, but rather to examine the ways and opportunities that central banks use to interact with public opinion, market participants and the other institutions in society. We consider 15 criteria of central bank accountability, divided into three main groups (see Table 6.1): • Ex-ante accountability, • Ex-post accountability and • Procedures. The first criterion in the first grouping consists of a clear definition of the objective of price stability. This concerns in particular the way in which price stability is defined, the type of price index to be used as a final target, the precision of the target (a point estimate as opposed to a range) and the horizon of the target. The more precise is the target, the more accountable a central bank can be held. This criterion has to be viewed against the well known and documented difficulties to correctly measure inflation. There is clearly a trade-off between the simplicity of the public announcement, which enhances transparency, and the complexity of the inflation phenomenon.

Table 6.1

Accountability criteria Federal Reserve

Ex-Ante 1. Clear definition of the objective of price stability 2.

Announcement of the operational target

3.

Announcement of intermediate target

4.

Announcement of indicators for assessing monetary policy

Bank of England

ECB

no

no

Government (2.5% CPIX inflation for 1998)

Treaty/ECB less than 2% CPI inflation

fed fund rate

overnight lending rate

repo rate

overnight rate

direct inflation target

no

no specific one

reference ranges for monetary aggregates

no monitoring ranges for money and credit aggregates

no specific one

implicitly

implicitly

implicitly

implicitly

money and credit aggregates

no

yes, inflation report

no

Publication of inflation forecast and deviation from target

twice a year (for the HH reports)

no

yes, inflation report

no

Explanation of main policy measures (or absence thereof) and underlying reasons

yes

yes

yes

yes

5.

Explanation of how monetary policy targets affect other policies & objectives Ex-post 6. Publication of data on intermediate target or explanation of possible deviation 7.

Bank of Japan

Explanation of how these measures affect other policies

fib

4^

Table 6.1, cont. Accountabilitity criteria Federal Reserve

Bank of Japan

Bank of England

ECB to

Procedures 10. Regular (monthly, quarterly, yearly) public reports covering issues 1-8 above 11.

Hearings in parliament with Q&A

12.

Participation of government representative at meeting of the decisionmaking bodies (as observers)

13.

Publication of summary minutes

14.

Publication of detailed minutes

15.

Publication of the votes of the members of the decision-making bodies

monthly

monthly

quarterly inflation report

monthly

to to

at least twice a year

at least twice a year

yes, regular

quarterly

no

Ministry of Finance and of Economic Planning Agency

Treasury representative

President of ECOFIN and Commission

7 weeks later

one month later

5 years later

to be decided

7 weeks later

one month later

6 weeks later

| 6 weeks later

... and accountable?

155

Nevertheless, complexity cannot be used as an excuse for not defining the target, as this would give rise to the suspicion by market participants that other policy targets are being followed. This would lead to a loss of credibility for the central bank. The second criterion is the announcement of an operational target. A clear understanding by market participants of which operational target the central bank uses to conduct its monetary policy operations enables a frequent monitoring of a central bank's actions. Market participants should be able to clearly distinguish between the instruments of monetary policy, such as the 'official' or 'policy' interest rate, and the operational target that is affected by a central bank's actions but is ultimately determined by market conditions. If such a distinction cannot be made, market participants have less information about a central bank's actions. The third criterion is the announcement of an intermediate target. Given the time lag, estimated to be around 18 to 24 months, between implementation of monetary policy and its effects on price levels, the inflation rate provides information only on past monetary policy. To assess current policy, one needs to compare it against inflation performance one or two years in the future, a statistic that is not available. An intermediate target is generally used to try to express future inflation developments in terms of contemporaneously observable variables. Several variables can be used for this purpose, such as the exchange rate, monetary or credit aggregates, or other indicators of future inflation, depending on their effective relationship to the price level. The announcement of the target enables the public to monitor how the central bank is reacting to inflationary pressures, as reflected in these indicators. It can thus check whether the central bank is sticking to its price stability objective. The setting of an intermediate target requires that there are variables that enable, with some degree of precision, to forecast future inflation. This is an empirical question. If there is no variable closely related to future price developments, the central bank may use its own inflation forecast as a target, as is done under a direct inflation targeting strategy. The fourth criterion is the announcement of indicators for assessing the appropriateness of monetary policy. Given that it is

156

Open Issues in European Central Banking

generally difficult to find one single variable that enables the targeting of future inflation, central banks use indicators to interpret possible deviations from the intermediate target and to assess whether a policy reaction is granted. Disclosing information on these indicators helps market participants to understand how the central bank interprets inflation forecasts and deviations from the intermediate target. An excessive amount of indicators may however confuse the picture and give the impression that the central bank wants to pick the indicator it prefers, depending on circumstances, to justify its own behavior. The statement that central banks 'look at everything' is not very informative for market participants. The fifth criterion is the explanation of how monetary policy targets affect other policies and objectives. Even if the primary objective is price stability, central banks may pursue secondary objectives, such as supporting the economic policies of the government. Providing a forecast and an assessment of inflationary trends may help the public to clearly distinguish between the responsibilities of the various policy-making institutions and to create transparency in the dialogue between them. The second group of criteria defines the ex-post accountability of the central bank with respect to its pre-announced targets and indicators. They include the following: • Publication of data on intermediate variables and explanation of possible deviation from target; • Publication of an inflation forecast and possible deviation from the target; • Explanation of the main policy measures, or absence thereof, and underlying reasons; and • Explanation of how these measures affect other policies. The publication of these data and the explanation of the main developments related to the target variables and the analysis leading to the policy decisions is a way to enhance the transparency of the decisions and thus the accountability of the central bank. The third group of accountability criteria refers to the procedures followed by the central bank to meet the above criteria. The procedure may include the issuing of public reports in which the central bank provides data on its targets and indicators and explains

... and accountable?

157

its policies, their underlying reasons and how well it has performed in view of the pre-announced targets. Reporting also can be made through testimony presented in hearings before Parliament by members of the decision-making body. These presentations are generally less technical but allow, through the question and answer sessions, for a more articulated assessment. The participation of the government in meetings of the decision-making body is another opportunity through which institutions exchange information and views on each other's policies. Although market participants are not necessarily informed about such exchanges, it may be reflected in the respective policies. Finally, the background analysis and reasoning for the central bank's decision-making can be made available through the publication of the minutes of the meeting of the decision-making body, in either an abridged or full version. This information can be disaggregated up to the level of each member of the decision-making body, by making public the votes of the individual members. This raises the question of whether accountability is required for the central bank as a whole or for each member of the decision-making body; it relates to the issue of collective vs. individual accountability, which is addressed in the next section. As has already been mentioned, this list of criteria should not be used to construct an index of central bank accountability. Indeed, some of these criteria are complementary and cover overlapping issues. Furthermore, the list is not exhaustive nor has it been drawn up on the basis of a normative analysis of accountability but rather on the observation of certain practices in some central banks and the discussion that has taken place in academic and policy fora. 6.4

How accountable is the ECB?

There is as yet little literature that attempts to assess the accountability of the ECB.13 There are at least two difficulties with such an exercise. The first difficulty is that, as mentioned above, the assessment can only be of a qualitative, rather than quantitative nature. The criteria outlined in the previous sections are to be considered as references to help understand the main issues at stake

158

Open Issues in European Central Banking

rather than pointers to rank central banks. Furthermore, these criteria need to be examined collectively, and in the broad institutional context within which the central bank operates. The second difficulty is that the ECB has not been operating for a very long time. In light of these caveats, we try to examine the possible accountability of the ECB, in comparison with the experience of three other major central banks - the Federal Reserve in the US, the Bank of Japan and the Bank of England - the latter two having recently experienced a change in their statutes, in particular with a view to increasing both independence and accountability. Again, the comparison is not aimed at ranking the central banks, but at assessing the accountability of the ECB in the light of its counterparts' experiences. Table 6.1 indicated the extent to which these three major central banks and the ECB meet the various criteria and in what length of time. Concerning the first criterion, the ECB has publicly announced its definition of price stability, with a view to enhancing transparency and credibility: a rate of inflation lower than 2 per cent. The harmonized consumer price index, as calculated by Eurostat, is the measured variable. The reference level takes into account measurement problems encountered in measuring inflation. The ECB has also explained that the lack of a lower bound for inflation does not imply that a reduction in the price level is deemed to be consistent with price stability. Deflation is not price stability. The ECB makes no explicit reference to time in its definition of price stability. It has only stated that price stability is an objective to be achieved over the medium run - a term that is clearly lacking in specificity with respect to the explicit objective. For instance, if a shock pushes inflation above the 2 per cent level, thus breaching the definition of price stability, what is the time frame in which the ECB is expected to restore price stability? The lack of an inflation forecast, as indicated below, leaves wide margins of discretion to the ECB to decide, without disclosing it to the public, the pace of disinflation.

... and accountable?

159

The Federal Reserve and the Bank of Japan do not announce precise objectives for inflation rate. The Fed has developed a nonquantitative definition of the concept of price stability, being a rate of change of the price level which does not induce economic agents to modify their behavior. This concept does not make it easy for market participants to understand the policy intentions of the Fed. It should be recalled that the statutory objective of the Fed is not only to achieve price stability, but also to promote economic activity and employment.14 The Bank of England has a clear target, in terms of price inflation, which is set by the government with the budget law, and can be modified year-by-year. The problem of time inconsistency of monetary policy has thus been shifted away from the central bank to the government. The time horizon of the policy objective is limited and can thus conflict with the constraints arising from political business cycle considerations. Concerning the second and third criteria, related to the operational and intermediate targets, the ECB has set the overnight money market rate as its operational target. The instruments of the Eurosystem are: the overnight lending and deposit facilities, whose rates will determine the ceiling and the floor for the overnight rate; the weekly repurchase agreement, with bi-weekly maturity; and finetuning operations, largely based on repos. All operations of the ESCB have to be conducted against collateral. Concerning the intermediate target, the ECB has not followed the advice given by its predecessor, the EMI. The latter suggested that among the indispensable elements of any strategy adopted by the ECB should be 'the public announcement of a specific target (or targets) against which performance can be assessed on an ongoing basis by the general public, thereby enhancing accountability'.15 The ECB instead has chosen to refer to a reference growth rate for M3. No clear explanation has been given about the operational role that such a monetary reference should play in the ECB's strategy. It is obviously not an intermediate target, not even in the way the Bundesbank has intended it to serve in the past, that is, as a main explanatory factor for policy decisions.

160

Open Issues in European Central Banking

The Fed and the Bank of Japan have no intermediate targets: the Bank of England publishes an inflation forecast, which de facto represents its intermediate target. As for other indicators of the ECB monetary policy, they encompass all relevant financial variables (in particular, the money market yield curve, money and credit aggregates, credit market conditions, bond yields, exchange rates and other asset prices) as well as various non-financial variables (price and cost variables, indicators of aggregate demand and supply conditions, etc.). None of these indicators has been given particular importance, with a view to establishing a hierarchy of information variables for market participants. The impression that the ECB will look at everything, which is equivalent to not disclosing any information, has not being dispelled. The Fed uses monitoring ranges for money and credit aggregates, although their importance has been strongly downgraded in recent years. The Bank of England also uses a set of indicators to assess its own inflation forecasts, but no specific one is given particular importance. Finally, the ECB has been requested, in particular by the European Parliament, to explain how the conduct of its policy interacts with other policies and affects economic developments in the European Community. This practice is undertaken more or less explicitly by all central banks, in the context of cooperation with other economic policy authorities and with a view to ensuring the consistency of the respective policy plans. The appropriateness of the policy-mix and its consequences on the economy as a whole, are often a matter of public debate. In the European Community, the interaction between monetary and other policies will be an input to the regular multilateral surveillance exercise. Once a year, broad guidelines are defined by the Council of EU Finance Ministers for the conduct of the economic policies of the member states. The Council then monitors the policies of the member states, with a view to ensuring their consistency with the above guidelines. The monetary policy conducted by the Eurosystem represents an important reference for the definition of policy guidelines. Turning to the ex-post criteria, the ECB regularly publishes data on the relevant monetary aggregate. Given the unclear role that

... and accountable?

161

the latter plays in the monetary strategy, however, the explanation of its behavior is of doubtful relevance to market participants. Concerning criterion 7, the ECB does not intend for the moment to publish its own inflation forecast. The justification offered is that: There are conceptual difficulties in formulating forecasts conditioned on unchanged policies since some financial market indicators which provide input to these forecasts are typically influenced by the markets' anticipated stance of monetary policy over the forecasting horizon. Furthermore, there is a risk that publishing inflation forecasts may, at times, have adverse effects on financial markets and wage and price setting and that the credibility of the ESCB could be damaged in the medium-term if the conditional nature of the forecasts is not well explained. In this respect, it is seen as crucial that the ESCB be fully independent in its decisions on policy actions and not in apparent need of a published inflation forecast to convince the public about the appropriateness of its decisions. In view of such considerations, most EU central banks, including most of those targeting inflation directly, do not, at present, provide quantitative forecasts for future inflation rates to the public. Yet, even if quantitative inflation forecasts are not revealed it will be desirable to publish some form of information on inflation prospects, and this should include a discussion of the perceived risks around the central inflation projection' (EMI, 1997, p. 16). This justification is not convincing. Producing an inflation forecast is undoubtedly a difficult exercise, especially at the start of monetary union in view of the wide uncertainties. Nevertheless, the ECB will have in any case to make its own forecasts for the purpose of taking policy decisions. Not revealing to the public the background analysis that has led to policy decisions suggests that the ECB wants to hide information. This is unjustified from any relevant viewpoint. As suggested in the previous sections, the public at large, social partners and the governments of the euro zone can only benefit from the information provided from the ECB. For instance, in conducting wage negotiations, social partners need to make reference to an inflation forecast. Such a forecast, prior to monetary union, was based on the national government targets or projections. In the

162

Open Issues in European Central Banking

Eurozone, national governments have lost competence on inflation. Without some indication from the ECB of the inflation forecast for the contract period, social partners have less information than in the past with which to conduct negotiations. This reduces efficiency in the European economy. The explanation of policy decisions (criterion 8) is an important element of the accountability of a central bank. As shown in Table 6.1, all central banks conduct this exercise. This task may be simpler and better understood by the public, the more transparent are the targets and the operational functions of the central bank. The frequency and content of the decision-making process are also factors that may affect accountability. For instance, in the case of the Fed, given that the FOMC meets only every six weeks, the explanations to be provided to the public have to take into account the fact that the object of the decision is a policy guideline for the Fed fund rate for the following 6 weeks, a relatively long period of time. Such a guideline needs to take into account contingencies for unforeseen possible developments and has therefore to be rather complex in nature. This complexity is one of the main reasons why the Fed requests that the guidelines be kept secret for at least one month.16 Since the ECB Governing Council meets frequently, the decisions regarding the main refinancing operations and the interest rate corridor (rates on marginal lending and overnight deposit facilities) leave little scope for discretion in the interval between meetings. The announcement of the key rates thus contains nearly all the relevant information concerning the ongoing policy. It is certainly more transparent than a broad guideline for a market-determined rate to be implemented in the following month, as in the case of the Fed. In summary, the frequency with which the decision-making body meets and the operational framework adopted by the ECB ensures a relatively high degree of transparency on decisions, compared to other central banks. Turning to the procedures for making information available to the public or the other policy authorities, there is a broadly similar situation across central banks with respect to regular public reports or hearings in Parliament or Congress. The quality of the reporting

... and accountable?

163

may vary, depending also on the requirements at the receiving end, that is, the needs of the public at large, the academic and market practitioner or the Parliament. A striking feature of the first ECB monthly reports is that not only do they not contain forecasts, but the analysis is mainly backward-looking. Little or no indication is provided as to future developments, which are those most relevant for monetary policy decisions. This is not a good sign for accountability and transparency. It has also been pointed out that the accountability of the ECB may be impaired by the fact that the European Parliament is not sufficiently representative of the European constituency. Others have suggested that the ECB is not sufficiently challenged in the course of its hearings before the European Parliament, compared to the experience, for example, of the Federal Reserve with the US Congress. Only experience will whether how the European Parliament will live up to expectations. Concerning criterion 12, the participation, but not the right to vote, of a member or a representative of the executive branch of government in the meetings of the central bank's decision-making body is foreseen for the Bank of Japan and the Bank of England but not for the Federal Reserve. For the ECB, the invitation is extended not only to the President of the Council of the EU Finance Ministers but also to the European Commission. This raises two problems. The first is that the Presidency rotates every six months, which poses problems of continuity and of transmission of information across countries and across time. In addition, in periods when the Presidency is held by a country not participating in the euro area, a problem of confidentiality may arise. Furthermore, the choice of whether to attend is left to the Presidency of the Council, and may thus vary. When the President of the Council attends, he should keep the Finance Ministers of the other member states informed about the discussion in the ECB Governing Council, with a view to ensuring a level playing field. The second issue is linked to the publication of the detailed minutes of the meeting of the ECB Governing Council, and in particular the voting record. Such publication is foreseen in the case of the Fed (7 weeks later for the voting, 5 years later for the detailed

164

Open Issues in European Central Banking

minutes), the Bank of Japan (one month later) and the Bank of England (6 weeks later). The Maastricht Treaty forbids the ECB from publishing the detailed minutes and votes, as Article 10.4 of the Statute states: 'The proceedings of the meetings shall be confidential. The Governing Council may decide to make the outcome of its deliberations public'. Only the outcome of the decision, not the procedure, may be made public. It is the ECB as a whole which is held accountable, not each of the individual components of its decision-making body. Two questions arise. The first is whether the publication of the minutes really increases accountability. It certainly cannot reduce it, but what is really important here is that the discussions leading to a policy decision are as open as possible. If the impression were created that some arguments, if made public, could be used against the members of the ECB Governing Council or create distorted perceptions of financial markets, discussions would become constrained and not result in efficient decision-making. Ultimately, the true discussions would take place beforehand, in informal meetings, while the formal discussion would be largely 'sanitized'. The minutes of the FOMC meetings provide scant details of dissenting arguments. The second question concerns the reason for collective, rather than individual accountability. This choice derives from several factors. The first is that the members of the ECB Governing Council are nominated through different procedures: the Governors of the NCBs through national procedures; the members of the Executive Board through a European procedure, specified in the treaty (Art. 109a2). This is a rather different situation than that of any other central bank. A second aspect is that the ECB's primary objective is that of maintaining price stability in the euro area as a whole. The ECB is not responsible for national price developments. Each member of the ECB Governing Council should thus in principle have the same "European" objective. This is confirmed not only by the provisions related to the personal and institutional independence of the ECB governing body, but also by the voting system: one man, one vote.

... and accountable?

165

Considering the two aspects jointly, a tension can be observed between the expectation that the ECB has a 'European' objective (price stability in the euro area) and fact that the members of its decision-making body are not nominated on the basis of a fully 'European' procedure. A system of individual accountability, by which each member of the ECB Governing Council was individually accountable, would raise the problem of the body to which they should be accountable. It would not be consistent for an NCB Governor to account for his actions, in particular his vote, taken with a view to a 'European interest', to a body such as a national Parliament or a national government, which represents national interests. A national political body has no authority to judge how the accountee has performed in his 'European' task. The accountability of the ECB is not the sum of the national accountabilities of the members of its Governing Council. It could be envisaged that the members of the ECB Governing Council are individually accountable only to the European Parliament or to the European Council. The European political bodies (Parliament and Council) however, play no role in the appointment of NCB Governors. Their membership in the ECB Governing Council is automatic, not subject to an act of confirmation by a European body (as in the case of the Court of Justice or the Commission). The European Parliament does not organize hearings to confirm the appointments of the Governors of the NCBs. It would thus be peculiar to request the NCB Governors to be accountable to the European Parliament. Only the members of the Executive Board could be individually accountable to the European bodies that have nominated them. But it would not make much sense to publish only the votes of the ECB Executive Board members, while the others are kept secret; individual accountability would then be required of only some members. In summary, given the peculiar nature of the ECB and its decision-making bodies, its members cannot be individually accountable. The solution that has been chosen is to have a collegiate accountability for the whole ECB Governing Council to the European political bodies. Such collegiality makes confidentiality of proceedings necessary, as called for in the Maastricht Treaty.

166

Open Issues in European Central Banking

The interplay between the fact that the President of the Council participates in the meetings of the ECB Governing Council and the requirement of confidentiality of the voting system in the latter may create the peculiarity that confidentiality on ECB deliberations is ensured only with respect to the public at large, and not with the Council of Ministers. Unless the requirement of confidentiality were extended to the President of the Council of EU Ministers, vis-a-vis the other EU Finance Ministers, there would be an asymmetry of information. That is, the governments of the member states, but not the public nor the market participants, would know how the various members of the Governing Council have voted. This might not be fully consistent with collegiate accountability and transparency, and would encourage the search for private information about the behavior of the various members of the ECB Governing Council. Conclusions On the basis of the treaty's requirements, one would think that the ECB is more accountable than most other central banks. For example, it must make regular reports to the European Parliament. In its early operations, however, the ECB has shown less accountability than one would have expected. Much effort is needed to improve on the present situation which looks unsatisfactory. Contrary to some suggestions, the way to improve the accountability of the ECB is not to ask it to publish detailed minutes of the ECB Governing Council meetings. This would only result in shifting the true debate to informal meetings of the Governing Council, so that the formal meetings would only record pre-packaged consensus with no or little discussion. The minutes would become irrelevant. One way to make the ECB more accountable is to engage it in substantive discussions where it is pressed to provide the relevant information about the background analysis that have led to policy decisions. For example, the ECB should be pressed to make its inflation forecast public. Accountability cannot be ensured by the ECB alone. An important role has to be played by its counterparts, such as the

... and accountable?

167

European Parliament, the Council of EU Finance Ministers and the public at large. The hearings for the nomination of the ECB Executive Board held in May 1998 suggest that the European Parliament intends to give serious attention to this task. However, much more is required to counteract the impression of inadequate accountability, which would ultimately lead to constraints on the independence of the ECB.

References to Chapter 6 Barro, Robert and David Gordon (1983a), 'Rules, Discretion and Reputation in a Model of Monetary Policy', Journal ofMonetary Economics, 12, pp. 101-22,. Barro, R. and David Gordon (1983b), 'A Positive Theory of Monetary Policy in a Natural Rate Model', Journal of Political Economy, 91, pp. 589-610,. Bini Smaghi, Lorenzo (1996), How can the ECB be credible?, EUI Working Paper No. 96/24, Florence. Briault, C, A. Haldane and Mervyn King (1996), Independence and Accountability, Working Paper No. 49, Bank of England. Canzoneri, Matthew (1985), 'Monetary Policy Games and the Role of Private Information', American Economic Review, 75, pp. 10561070. Cukierman, Alex (1992), Central Bank Strategy, Credibility and Independence: Theory and Evidence, MIT Press, Cambridge, Massachusetts. Cukierman, Alex and Alain Metzler (1986), 'A theory of ambiguity, credibility, and inflation under discretion and asymmetric information', Econometrica, 54, 5, pp. 1099-1128. De Haan, Jacob (1997), 'The European Central Bank: Independence, Accountability and Strategy: A Review', Public Choice, No. 93, pp. 395-426. Demertzis, M., Andrew Hughes Hallett and N. Viegi (1998), Independently blue? Accountability and Independence in the New European Central Bank, CEPR Working Paper No. 1842, 1998. Dornbusch, Rudiger, Carlo Favero and Francesco Giavazzi (1998), 'Immediate challenges for the ECB', in EMU: Prospects and

168

Open Issues in European Central Banking

Challenges for the Euro, David Begg et al. (eds.), Blackwell, London. Eijffinger, Sylvester and Jacob De Haan (1996), The Political Economy of Central Bank Independence, Special Paper in International Economics, No. 19, Princeton University. European Monetary Institute (1997), The Single Monetary Policy in Stage Three: Elements of the Monetary Policy Strategy of the ESCB, Frankfurt, February. European Parliament (1998), 'Democratic Accountability in Stage Three of EMU', Public hearings and resolution of the Subcommittee on Monetary Affairs, Brussels. Faust, J. and Lars Svensson (1998), Transparency and Credibility: Monetary Policy with Unobservable Goals, CEPR Working Paper No. 1852. Faust, S. (1995), 'Central Bank Independence Revisited', American Economic Review, No. 85, pp. 201-6. Garfinkel, M., and S. Oh (1995), 'When and how much to talk: Credibility and flexibility in monetary policy with private information', Journal of Monetary Economics, 35, pp. 341-357. Gormley, L. and Jacob De Haan (1996), 'The democratic deficit of the European Central Bank', European Law Review, No. 21, pp. 95112. Goodfriend, M. (1986), 'Monetary Mystique: Secrecy and Central Banking', Journal ojMonetary Economics, 17, pp. 63-92. Kydland, F. and Edward Prescott (1977), 'Rules rather than discretion: The inconsistency of optimal plans', Journal of Political Economy, 85, pp. 473-491. McCallum, Bennett (1995), 'Two Fallacies concerning Central Bank Independence', American Economic Review, Papers and Proceedings, pp. 207-11. Persson, R. and Guido Tabellini (1993), Designing institutions for monetary stability, Carnegie-Rochester Conference Series on Public Policy, No. 39, pp. 53-84. Roll et al.(1993), Independent and Accountable: A New Mandate for the Bank of England, CEPR, London. Svensson, Lars (1997), 'Inflation forecast targeting: Implementing and monitoring inflation targets', European Economic Review, 41, pp. 1111-1146.

... and

accountable?

169

Walsh, Carl (1995), 'Optimal contracts for central bankers', American Economic Review, No. 85, pp. 150-67.

Notes to Chapter 6 1. It is interesting to note that during the debate in the European Parliament on 'Democratic accountability in stage three of EMU', the French, Italian and Portuguese versions of the draft report initially used the term controle democratique, controllo democratico and controlo democrdtico, respectively, while the Spanish version referred instead to responsabilidad democrdtica. The texts were later harmonized with the use of the word responsibility. 2. See Roll et al. (1993). 3. CEPR (1993), pp. 48-49. 4. See, in particular, Persson and Tabellini (1993), Walsh (1995) and McCallum(1995). 5. See Briault etal. (1996). 6. See Kydland and Prescott (1997) and Barro and Gordon (1983a and 1983b). 7. See Cukierman and Metzler (1986) and Garfinkel and Oh (1995). 8. See, for example, Canzoneri (1985). 9. Goodfriend (1986) presents in a critical way the arguments of the Fed in its case against Merrill. 10. See Faust and Svensson (1998) for a formal proof of these propositions. 11. The issue is already debated in the literature; see Dornbusch, Favero and Giavazzi (1998). 12. A formal derivation of this reasoning can be found in Demertzis et al. (1998). 13. See, in particular, Gorwley and De Haan (1996) and De Haan (1997).

170

Open Issues in European Central Banking

14. As defined in the act establishing the Federal Reserve System, the tasks of the Fed is to 'furnish an elastic currency...so as to promote effectively the goals of maximum employment, stable prices and moderate interest rates'. 15. European Monetary Institute (1997), p. 10. 16. See Goodfriend (1996).

7 . 3 X 2 CONCRETE PROPOSALS FOR STRENGTHENING THE EUROSYSTEM Introduction The first six chapters of this book discussed the strengths and weaknesses of the Eurosystem. The overall picture that emerges is that of an institution whose foundations are still weak. Our analysis suggests that reforms are required in several areas. We summarize them here under six concrete proposals. These recommendations should be seen as a stimulus for further discussion. We do not pretend to offer a solution to all the problems. Nevertheless, we believe that the issues that we have identified are of concern even for the achievement of the ECB's primary objective. It is imperative therefore that a solution be found. 7.1

Transform the Eurosystem into a true European Central Bank

The first two proposals are largely based on the analysis developed in Chapter 1. Revise the voting system in the ECB Governing Council The Governing Council will over time become too large to function efficiently. It will be unbalanced in the relative representation of national and European elements. In particular, following the eastern enlargement of the EU, it might comprise over 30 members - 24-25 NCB Governors plus the six members of the Executive Board. As in the FOMC, the right of vote should thus rotate among NCB Governors, with a view to creating a body that is small enough to take quick decisions and to re-establish a balance of votes between Governors and Executive Board members. No more than nine NCB Governors should be allowed to vote at the same time. The others would be allowed to attend meetings of the Governing Council, but 171

172

Open Issues in European Central Banking

without the right to vote or to influence the agenda. The principle of 'one man, one vote' should be kept. Furthermore, the members of the Executive Board should be allowed to vote on financial matters to which the 'one man, one vote' principle should be extended. Implementing this proposal would require a change in the Maastricht Treaty, but this should not constitute an insurmountable obstacle. It is generally agreed that the EU needs to adapt its institutions before the next enlargement. The re-organization of the Governing Council of the ECB is just one of the issues that should be put on the agenda of the next Intergovernmental Conference that will prepare the EU for enlargement. 'Europeanize' the nomination of NCB Governors NCB Governors now perform a European function. Their nomination should thus be subject to the approval of the European Parliament and/or of the Council of Ministers of the EU, in analogy to the nomination of the Executive Board members. This would enhance the accountability of the Eurosystem The spirit of this proposal can be implemented without any revision to the treaty if national parliaments and/or governments voluntarily decide to modify the procedure for nominating their respective NCB Governor, with a view to asking for an opinion of the relevant committee of the European Parliament and ECOFIN. 7.2

Clarify who is responsible for financial stability

The following two proposals are largely based on the analysis developed in Chapter 2. Create a European Financial Supervisory

Agency

The independence of the ECB and the NCBs is based on the understanding that they are primarily responsible for price stability. They should thus concentrate on the conduct of monetary policy to achieve this goal. Other activities, in particular banking supervision, should be devolved to a separate agency in the few member countries where this is not yet the case. National agencies should merge into a European Financial Supervisory Agency (EFSA). This institution could start by focusing its efforts on strengthening coordination between national agencies,

3x2 Concrete Proposals

173

as the European Monetary Institute (EMI) has done initially in the contest of monetary policy coordination. Over time, the EFSA would also become responsible for implementing supervision for the largest pan-European banking institutions. As financial markets in Europe integrate further, EFSA should become fully in charge of all supervisory responsibilities. The execution of supervision could be undertaken in a decentralized way, to the extent possible and feasible. An appropriate flow of micro- and macro-prudential information should be implemented between the ECB and the EFSA. Early warning signals should be put in place to detect systemic financial market problems. The first element of this proposal requires legislative acts in the countries where banking supervision is still performed by the NCB. The second element, the creation of the European Financial Supervisory Agency, would require a legislative act by the EU. Create a European Liquidity Consortium At present, it is not clear who is responsible for the provision of emergency liquidity to banks that experience financial difficulties (lender-of-last-resort function, LOLR). This uncertainty is undesirable because it might lead to the worst-case scenario in which banks engage in risky business because they expect to be bailed out by their respective NCB, possibly with preferential national treatment. It should be made clear from the outset that only the Eurosystem is responsible for providing emergency liquidity in the Eurozone. The European banking sector should be encouraged to create a European Liquidity Consortium (ELC). This private sector institution, to consist of a pool of banks that provide mutual guarantees, should be able to take care of small- and medium-sized bank failures. Applying the self-insurance principle should reduce the unavoidable moral hazard problem inherent in the interplay between the lender of last resort function and banking supervision. This proposal does not require any treaty revision; a memorandum of understanding within the Eurosystem and collective action by the major European banks would suffice.

174 7.3

Open Issues in European Central Banking Clarify financial relationships within the Eurosystem

The following two proposals are largely based on the analysis developed in Chapters 3 and 4. Clean up the NCBs9 balance sheets NCBs should clear their balance sheets of the dead wood they have accumulated over the last decades. By getting rid of all securities that NCBs still manage outside the Eurosystem, potential conflicts of interest and threats to independence may be eliminated. After the balance sheet clean-up operation, NCBs will be left with about 200 billion euro of foreign assets (foreign exchange reserves and gold). These assets should be dismissed. As some of these assets have as counterpart liabilities with the banking system, the net transfer to national treasuries would be about 100 billion euro. National treasuries should use this capital transfer either to reduce their public debt or to create an investment fund outside the NCB, which would be responsible for investments with long-term objectives. This would require collaboration between NCBs and national treasuries, but no legislative act at the EU level. Implement a fair redistribution scheme for seigniorage If balance sheets have been cleaned up, the treaty's provisions on this point are appropriate in the very long run. An interim solution would be to suspend all formal re-distribution of seigniorage among NCBs and to ensure at the same time that seigniorage revenues arise directly at the European level. This can be achieved by booking directly in the ECB's balance sheet any increment to the balance sheet of the Eurosystem after 2002 (instead of booking them on the balance sheets of the NCBs as foreseen). This would freeze the balance sheets of NCBs at their 2002 situation and would 'grandfather' the revenues that the NCBs obtain from the stocks they accumulated prior to EMU. ECB profits, which would be negligible at the beginning but would grow over time, would be distributed according to the capital key as foreseen in the treaty. In the very long run, when all member countries have joined EMU and there is greater financial solidarity within the EU, one could consider

3x2 Concrete Proposals

175

transferring the monetary income of the Eurosystem to the EU budget. Precautions would have to be taken to render these transfers independent of the day-to-day conduct of monetary policy. The interim solution would require a qualified majority decision in the Council of EU Ministers (on a proposal from the ECB, which in turn requires a decision by its Governing Council) and would thus not require a change in the treaty. Transferring seigniorage directly to the EU budget would require not only a change in the treaty, but would make sense primarily if the fiscal constitution of the EU is changed in the sense of instituting some degree of financial solidarity among member countries.

ANNEXES

Annex I

Chapters 2 and 3 of Title VI on Economic and Monetary Policy of the Treaty on European Union CHAPTER 2 MONETARY POLICY

Article 105 1. 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 the achievement of the objectives of the Community as laid down in Article 2. The ESCB shall act in accordance with the principle of an open market economy with free competition, favoring an efficient allocation of resources, and in compliance with the principles set out in Article 3a. 2.

The basic tasks to be carried out through the ESCB shall be:

•

to define and implement the monetary policy of the Community;

•

to conduct foreign exchange operations consistent with the provisions of Article 109;

•

to hold and manage the official foreign reserves of the Member States;

•

to promote the smooth operation of payment systems.

3. The third indent of paragraph 2 shall be without prejudice to the holding and management by the governments of Member States of foreign exchange working balances. 4.

The ECB shall be consulted: 177

178

Open Issues in European Central Banking

•

on any proposed Community act in its fields of competence;

•

by national authorities regarding any draft legislative provision in its fields of competence, but within the limits and under the conditions set out by the Council in accordance with the procedure laid down in Article 106(6).

The ECB may submit opinions to the appropriate Community institutions or bodies or to national authorities on matters within its field of competence. 5. The ESCB shall contribute to the smooth conduct of policies pursued by the competent authorities relating to the prudential supervision of credit institutions and the stability of the financial system. 6. The Council may, acting unanimously on a proposal from the Commission and after consulting the ECB and after receiving the assent of the European Parliament, confer upon the ECB specific tasks concerning policies relating to the prudential supervision of credit institutions and other financial institutions with the exception of insurance undertakings. Article 105 a 1. The ECB shall have the exclusive right to authorize the issue of bank notes within the Community. The ECB and the national central banks may issue such notes. The band notes issued by the ECB and the national central banks shall be the only such notes to have the status of legal tender within the Community. 2. Member States may issue coins subject to approval by the ECB of the volume of the issue. The Council may, acting in accordance with the procedure referred to in Article 189c and after consulting the ECB, adopt measures to harmonize the denominations and technical specifications of all coins intended for circulation to the extent necessary to permit their smooth circulation within the Community.

Selected Articles from the Treaty on European Union

179

Article 106 1. The ESCB shall be composed of the ECB and of the national central banks. 2.

The ECB shall have legal personality.

3. The ESCB shall be governed by the decision-making bodies of the ECB which shall be the Governing Council and the Executive Board. 4. The Statute of the ESCB is laid down in a Protocol annexed to this Treaty. 5. Articles 5.1, 5.2, 5.3, 17, 18, 19.1, 22, 23, 24, 26, 32.2, 32.3, 32.4, 32.6, 33.1(a) and 36 of the Statute of the ESCB may be amended by the Council, acting by a qualified majority on a recommendation from the ECB and after consulting the Commission or unanimously on a proposal from the Commission and after consulting the ECB. In either case, the assent of the European Parliament shall be required. 6. The Council, acting by a qualified majority either on a proposal from the Commission and after consulting the European Parliament and the ECB, or on a recommendation from the ECB and after consulting the European Parliament and the Commission, shall adopt the provisions referred to in Articles 4, 5.4, 19.2, 20, 28.1, 29.2, 30.4 and 34.3 of the Statute of the ESCB. Article 107 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 a national central bank, nor any member of 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.

180

Open Issues in European Central Banking Article 108

Each Member State shall ensure, at the latest at the date of the establishment of the ESCB, that its national legislation including the statutes of its national central bank is compatible with this Treaty and the Statute of the ESCB. Article 108a 1. In order to carry out the tasks entrusted to the ESCB, the ECB shall, in accordance with the provisions of this Treaty and under the conditions laid down in the Statute of the ESCB: •

make regulations to the extent necessary to implement the tasks defined in Article 3.1, first indent, Articles 19.1, 22 or 25.2 of the Statute of the ESCB and in cases which shall be laid down in the acts of the Council referred to in Article 106(6);

•

take decisions necessary for carrying out the tasks entrusted to the ESCB under this Treaty and the Statute of the ESCB;

•

make recommendations and deliver opinions.

2. A regulation shall have general application. It shall be binding in its entirety and directly applicable in all Member States. Recommendations and opinions shall have no binding force. A decision shall be binding in its entirety upon those to whom it is addressed. Articles 190 to 192 shall apply to regulations and decisions adopted by the ECB. The ECB may decide to publish its decisions, recommendations and opinions. 3. Within the limits and under the conditions adopted by the Council under the procedure laid down in Article 106(6), the ECB shall be entitled to impose fines or periodic penalty payments on undertakings for failure to comply with obligations under its regulations and decisions.

Selected Articles from the Treaty on European Union

181

Article 109 1. By way of derogation from Article 228, the Council may, acting unanimously on a recommendation from the ECB or from the Commission, and after consulting the ECB in an endeavor to reach a consensus consistent with the objective of price stability, after consulting the European Parliament, in accordance with the procedure in paragraph 3 for determining the arrangements, conclude formal agreements on an exchange rate system for the ECU in relation to non-Community currencies. The Council may, acting by a qualified majority on a recommendation from the ECB or from the Commission, and after consulting the ECB in an endeavor to reach a consensus consistent with the objective of price stability, adopt, adjust or abandon the central rates of the ECU within the exchange rate system. The President of the Council shall inform the European Parliament of the adoption, adjustment or abandonment of the ECU central rates. 2. In the absence of an exchange rate system in relation to one or more non-Community currencies as referred to in paragraph 1, the Council, acting by a qualified majority either on a recommendation from the Commission and after consulting the ECB, or on a recommendation from the ECB, may formulate general orientations for exchange rate policy in relation to these currencies. These general orientations shall be without prejudice to the primary objective of the ESCB to maintain price stability. 3. By way of derogation from Article 228, where agreements concerning monetary of foreign exchange regime matters need to be negotiated by the Community with one or more States or international organizations, the Council, acting by a qualified majority on a recommendation from the Commission and after consulting the ECB, shall decide the arrangements for the negotiation and for the conclusion of such agreements. These arrangements shall ensure that the Community expresses a single position. The Commission shall be fully associated with the negotiations. Agreements concluded in accordance with this paragraph shall be binding on the institutions of the Community, on the ECB and on Member States.

182

Open Issues in European Central Banking

4. Subject to paragraph, the Council shall, on a proposal from the Commission and after consulting the ECB, acting by a qualified majority decide on the position of the Community at international level as regards issues of particular relevance to economic and monetary union and, acting unanimously, decide its representation in compliance with the allocation of powers laid down in Articles 103 and 105. 5. Without prejudice to Community competence and Community agreements as regards Economic and Monetary Union, Member States may negotiate in international bodies and conclude international agreements. CHAPTER 3 INSTITUTIONAL PROVISIONS

Article 109 a 1. The Governing Council of the ECB shall comprise the members of the Executive Board of the ECB and the Governors of the national central banks. 2. (a) The Executive Board shall comprise the President, the VicePresident and four other members. (b) The President, the Vice-President and the other members of the Executive Board shall be appointed from among persons of recognized standing and professional experience in monetary or banking matters by common accord of the Governments of the Member States at the level of Heads of State or of Government, on a recommendation from the Council, after it has consulted the European Parliament and the Governing Council of the ECB. Their term of office shall be eight years and shall not be renewable. Only nationals of Member States may be members of the Executive Board.

Selected Articles from the Treaty on European Union

183

Article 109b 1. The President of the Council and a member of the Commission may participate, without having the right to vote, in meetings of the Governing Council of the ECB. The President of the Council may submit a motion for deliberation to the Governing Council of the ECB. 2. The President of the ECB shall be invited to participate in Council meetings when the Council is discussing matters relating to the objectives and tasks of the ESCB. 3. The ECB shall address an annual report on the activities of the ESCB and on the monetary policy of both the previous and current year to the European Parliament, the Council and the Commission, and also to the European Council. The President of the ECB shall present this report to the Council and to the European Parliament, which may hold a general debate on that basis. The President of the ECB and the other members of the Executive Board may, at the request of the European Parliament or on their own initiative, be heard by the competent Committees of the European Parliament. Article 109c 1. In order to promote coordination of the policies of Member States to the full extent needed for the functioning of the internal market, a Monetary Committee with advisory status is hereby set up. It shall have the following tasks: •

to keep under review the monetary and financial situation of the Member States and of the Community and the general payments system of the Member States and to report regularly thereon to the Council and to the Commission;

•

to deliver opinions at the request of the Council or of Commission or on its own initiative for submission to those institutions;

184

Open Issues in European Central Banking

•

without prejudice to Article 151, to contribute to the preparation of the work of the Council referred to in Articles 73f, 73g, 103(2), (3), (4) and 5, 103a, 104a, 104b, 104c, 109e(2), 109f(6), 109h, 109i, 109j(2)and 109k(l);

•

to examine, at least once a year, the situation regarding the movement of capital and the freedom of payments, as they result from the application of this Treaty and of measures of the Council; the examination shall cover all measures relating to capital movements and payments; the Committee shall report to the Commission and to the Council on the outcome of this examination.

The Member States and the Commission shall each appoint two members of the Monetary Committee. 2. At the start of the third stage, an Economic and Financial Committee shall be set up. The Monetary Committee provided for in paragraph 1 of this Article shall be dissolved. The Economic and Financial Committee shall have the following tasks: •

to deliver opinions, at the request of the Council, or of the Commission, or on its own initiative for submission to those institutions;

•

to keep under review the economic and financial situation of the Member States and of the Community and to report regularly thereon to the Council and to the Commission, in particular on financial relations with third countries and international institutions;

•

without prejudice to Article 151, to contribute to the preparation of the work of the Council referred to in Articles 73f, 73g, 103(2), (3), (4) and (5), 103a, 104a, 104b, 104c, 105(6), 105a(2), 106(5) and (6), 109, 109h, 109i(2) and (3), 109k(2), 1091(4) and (5), and to carry out other advisory and preparatory tasks assigned to it by the Council;

•

to examine, at least once a year, the situation regarding the movement of capital and the freedom of payments, as they

Selected Articles from the Treaty on European Union

185

result from the application of this Treaty and of measures adopted by the Council; the examination shall cover all measures relating to capital movements and payments; the Committee shall report to the Commission and to the Council on the outcome of this examination. The Member States, the Commission and the ECB shall each appoint no more than two members of the Committee. 3. The Council shall, acting by a qualified majority on a proposal from the Commission and after consulting the ECB and the Committee referred to in this Article, lay down detailed provisions concerning the composition of the Economic and Financial Committee. The President of the Council shall inform the European Parliament of such a decision. 4. In addition to the tasks set out in paragraph 2, if and as long as there are Member States with a derogation as referred to in Article 109k and 1091, the Committee shall keep under review the monetary and financial situation and the general payments system of those Member States and report regularly thereon to the Council and to the Commission. Article 109d For matters within the scope of Articles 103(4), 104c with the exception of paragraph 14, 109, 109j, 109k and 1091(4) and (5), the Council or a Member State may request the Commission to make a recommendation or a proposal, as appropriate. The Commission shall examine this request and submit its conclusions to the Council without delay.

Annex II

Protocol (No. 3) on the Statute of the European System of Central Banks and of the European Central Bank THE HIGH CONTRACTING PARTIES,

DESIRING to lay down the Statute of the European System of Central Banks and of the European Central Bank provided for in Article 4 a of the Treaty establishing the European Community, HAVE AGREED upon the following provisions, which shall be annexed the Treaty establishing the European Community: CHAPTER I CONSTITUTION OF THE

ESCB

Article I The European System of Central Banks 1.1. The European System of Central Banks (ESCB) and the European Central Bank (ECB), shall be established in accordance with Article 4 a of this Treaty; they shall perform their functions and carry on their activities in accordance with the provisions of this Treaty and of this Statute. 1.2. In accordance with Article 106(1) of this Treaty, the ESCB shall be composed of the ECB and of the central banks of the Member States Cnational central banks'). The Institut monetaire luxembourgeois will be the central bank of Luxembourg.

186

Statute of the ESCB and of the ECB

187

CHAPTER II OBJECTIVES AND TASKS OF THE ESCB

Article 2 Objectives In accordance with Article 105(1) of this Treaty, the primary objective of the ESCB shall be to maintain price stability. Without prejudice to the objective of price stability, it shall support the general economic policies in the Community with a view to contributing to the achievement of the objectives of the Community as laid down in Article 2 of this Treaty. The ESCB shall act in accordance with the principle of an open market economy with free competition, favoring an efficient allocation of resources, and in compliance with the principles set out in Article 3 a of this Treaty. Article 3 Tasks 3.1.

In accordance with Article 105 (2) of this Treaty the basic tasks to be carried out through the ESCB shall be:

•

to define and implement the monetary policy of the Community;

•

to conduct foreign exchange operations consistent with the provisions of Article 109 of this Treaty;

•

to hold and manage the official foreign reserves of the Member States;

•

to promote the smooth operation of payment systems.

3.2. In accordance with Article 105 (3) of this Treaty, the third indent of Article 3.1 shall be without prejudice to the holding and management by the governments of Member States of foreign exchange working balances. 3.3. In accordance with Article 105 (5) of this Treaty, the ESCB shall contribute to the smooth conduct of policies pursued by the competent authorities relating to the prudential supervision of credit institutions and the stability of the financial system.

188

Open Issues in European Central Banking Article 4 Advisory functions

4.1.

In accordance with Article 105 (4) of this Treaty:

(a) the ECB shall be consulted: •

on any proposed Community act its fields of competence;

•

by national authorities regarding any draft legislative provision within its fields of competence but within the limits and under the conditions set out by the Council in accordance with the procedure of Article 42;

(b) the ECB may submit opinions to the appropriate Community institutions or bodies or to national authorities on matters its fields of competence. Article 5 Collection of statistical information 5.1. In order to undertake the tasks of the ESCB, the ECB, assisted by the national central banks, shall collect the necessary statistical information either from the competent national authorities or directly from economic agents. For these purposes it shall cooperate with the Community institutions or bodies and with the competent authorities of the Member States or third countries and with international organizations. 5.2. The national central banks shall carry out, to the extent possible, the tasks described in Article 5.1. 5.3. The ECB shall contribute to the harmonization, where necessary, of the rules and practices governing the collection, compilation and distribution of statistics in the areas within its field of competence. 5.4. The Council, in accordance with the procedure laid down in Article 42, shall define the natural and legal persons subject to reporting requirements, the confidentiality regime and the appropriate provisions for enforcement.

Statute of the ESCB and of the ECB

189

Article 6 International cooperation 6.1. In the field of international cooperation involving the tasks entrusted to the ESCB, the ECB shall decide how the ESCB shall be represented. 6.2. The ECB and, subject to its approval, the national central banks may participate in international monetary institutions. 6.3. Articles 6.1 and 6.2 shall be without prejudice to Article 109 (4) of this Treaty. CHAPTER III ORGANIZATION OF THE

ESCB

Article 7 Independence In accordance with Article 107 of this Treaty, when exercising the powers and carrying out the tasks and duties conferred upon them by this Treaty and this Statute, neither the ECB, nor a national central bank, nor any member of 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 and of the national central banks in the performance of their tasks. Article 8 General principle The ESCB shall be governed by the decision-making bodies of the ECB. Article 9 The European Central Bank 9.1. The ECB which in accordance with Article 106 (2) of this Treaty shall have legal personality, shall enjoy in each of the

190

Open Issues in European Central Banking

Member States the most extensive legal capacity accorded to legal persons under their laws; it may, in particular, acquire or dispose of movable and immovable property and may be a party to legal proceedings. 9.2. The ECB shall ensure that the tasks conferred upon the ESCB under Article 3 are implemented either by the ECB's activities pursuant to this Statute or through the national central banks pursuant to Articles 12.1 and 14. 9.3. In accordance with Article 106 (3) of this Treaty, the decision-making bodies of the ECB are the Governing Council and the Executive Board. Article 10 The Governing Council 10.1. In accordance with Article 109a(l) of this Treaty, the Governing Council shall comprise the members of the Executive Board and the Governors of the national central banks. 10.2. Subject to Article 10.3, only members of the Governing Council present in person shall have the right to vote. By way of derogation from this principle the Rules of Procedure referred to in Article 12.3 may lay down that members of the Governing Council may cast their vote by means of teleconferencing. These rules shall also provide that a member of the Governing Council who is prevented from voting for a prolonged period may appoint an alternate as a member of the Governing Council. Subject to Articles 10.3 and 11.3, each member of the Governing Council shall have one vote. Save as otherwise provided for in this Statute, the Governing Council shall act by a simple majority. In the event of a tie, the President shall have the casting vote. In order for the Governing Council to vote, there shall be a quorum of two-thirds of the members. If the quorum is not met, the President may convoke an extraordinary meeting at which decisions may be taken without regard to the quorum.

Statute of the ESCB and of the ECB

191

10.3. For any decisions to be taken under Articles 28, 29, 30, 32, 33 and 51, the votes in the Governing Council shall be weighted according to the national central banks' shares in the subscribed capital of the ECB. The weights of the votes of the members of the Executive Board shall be zero. A decision requiring a qualified majority shall be approved if the votes cast in favor represent at least two thirds of the subscribed capital of the ECB and represent at least half of the shareholders. If a Governor is unable to be present, he may nominate an alternate to cast his weighted vote. 10.4. The proceedings of the meetings shall be confidential. The Governing Council may decide to make the outcome of its deliberations public. 10.5. The Governing Council shall meet at least ten times a year. Article 11 The Executive Board 11.1. In accordance with Article 109a(2)(b) of this Treaty, the Executive Board shall comprise the President, the Vice-President and four other members. The members shall perform their duties on a full-time basis. No member shall engage in any occupation, whether gainful or not, unless exemption is exceptionally granted by the Governing Council. 11.2. In accordance with Article 109a(2)(b) of this Treaty, the President, the Vice-President and the other Members of the Executive Board shall be appointed from among persons of recognized standing and professional experience in monetary or banking matters by common accord of the governments of the Member States at the level of the Heads of State or of Government, on a recommendation from the Council after it has consulted the European Parliament and the Governing Council. Their term of office shall be 8 years and shall not be renewable. Only nationals of Member States may be members of the Executive Board.

192

Open Issues in European Central Banking

11.3. The terms and conditions of employment of the members of the Executive Board, in particular their salaries, pensions and other social security benefits shall be the subject of contracts with the ECB and shall be fixed by the Governing Council on a proposal of a Committee comprising three members appointed by the Governing Council and three members appointed by the Council. The members of the Executive Board shall not have the right to vote on matters referred to in this paragraph. 11.4. If a member of the Executive Board no longer fulfills the conditions required for the performance of his duties or if he has been guilty of serious misconduct, the Court of Justice may, on application by the Governing Council or the Executive Board, compulsorily retire him. 11.5. Each member of the Executive Board present in person shall have the right to vote and shall have, for that purpose, one vote. Save as otherwise provided, the Executive Board shall act by a simple majority of the votes cast. In the event of a tie the President shall have the casting vote. The voting arrangements will be specified in the Rules of Procedure referred to in Article 12.3. 11.6. The Executive Board shall be responsible for the current business of the ECB. 11.7. Any vacancy on the Executive Board shall be filled by the appointment of a new member in accordance with Article 11.2. Article 12 Responsibilities of the decision-making bodies 12.1. The Governing Council shall adopt the guidelines and take the decisions necessary to ensure the performance of the tasks entrusted to the ESCB under this Treaty and this Statute. The Governing Council shall formulate the monetary policy of the Community including, as appropriate, decisions relating to intermediate monetary objectives, key interest rates and the supply of reserves in the ESCB, and shall establish the necessary guidelines for their implementation. The Executive Board shall implement monetary policy in accordance with the guidelines and decisions laid down by the Governing

Statute of the ESCB and of the ECB

193

Council. In doing so the Executive Board shall give the necessary instructions to national central banks. In addition the Executive Board may have certain powers delegated to it where the Governing Council so decides. To the extent deemed possible and appropriate and without prejudice to the provision of this Article, the ECB shall have recourse to the national central banks to carry out operations which form part of the tasks of the ESCB. 12.2. The Executive Board shall have responsibility for the preparation of Governing Council meetings. 12.3. The Governing Council shall adopt Rules of Procedure which determine the internal organization of the ECB and its decision-making bodies. 12.4 The Governing Council shall exercise the advisory functions referred to in Article 4. 12.5. The Governing Council shall take the decisions referred to in Article 6. Article 13 The President 13.1. The President or, in his absence, the Vice-President shall chair the Governing Council and the Executive Board of the ECB. 13.2. Without prejudice to Article 39, the President or his nominee shall represent the ECB externally. Article 14 National central banks 14.1. In accordance with Article 108 of this Treaty, each Member State shall ensure, at the latest at the date of the establishment of the ESCB, that its national legislation, including the statutes of its national central bank, is compatible with this Treaty and this Statute. 14.2. The statutes of the national central banks shall, in particular, provide that the term of office of a Governor of a national central bank shall be no less than 5 years.

194

Open Issues in European Central Banking

A Governor may be relieved from office only if he no longer fulfills the conditions required for the performance of his duties or if he has been guilty of serious misconduct. A decision to this effect may be referred to the Court of Justice by the Governor concerned or the Governing Council on grounds of infringement of this Treaty or of any rule of law relating to its application. Such proceedings shall be instituted within two months of the publication of the decision or of its notification to the plaintiff or, in the absence thereof, of the day on which it came to the knowledge of the latter, as the case may be. 14.3. The national central banks are an integral part of the ESCB and shall act in accordance with the guidelines and instructions of the ECB. The Governing Council shall take the necessary steps to ensure compliance with the guidelines and instructions of the ECB, and shall require that any necessary information be given to it. 14.4. National central banks may perform functions other than those specified in this Statute unless the Governing Council finds, by a majority of two thirds of the votes cast, that these interfere with the objectives and tasks of the ESCB. Such functions shall be performed on the responsibility and liability of national central banks and shall not be regarded as being part of the functions of the ESCB. Article 15 Reporting commitments 15.1. The ECB shall draw up and publish reports on the activities of the ESCB at least quarterly. 15.2 A consolidated financial statement of the ESCB shall be published each week. 15.3. In accordance with Article 109b(3) of this Treaty, the ECB shall address an annual report on the activities of the ESCB and on the monetary policy of both the previous and the current year to the European Parliament, the Council and the Commission, and also to the European Council. 15.4. The reports and statements referred to above shall be made available to interested parties free of charge.

Statute of the ESCB and of the ECB

195

Article 16 Banknotes In accordance with Article 105a(l) of this Treaty, the Governing Council shall have the exclusive right to authorize the issues of banknotes within the Community. The ECB and the national central banks may issue such notes. The bank notes issued by the ECB and the national central banks shall be the only such notes to have the status of legal tender within the Community. The ECB shall respect as far as possible existing practices regarding the issuing and design of banknotes. CHAPTER IV MONETARY FUNCTIONS AND OPERATIONS OF THE

ESCB

Article 17 Accounts with the ECB and the national central banks In order to conduct their operations, the ECB and the national central banks may open accounts for credit institutions; public entities and other market participants and accept assets including book-entry securities as collateral. Article 18 Open market and credit operations 18.1. In order to achieve the objectives of the ESCB and to carry out its tasks, the ECB and the national central banks may: •

operate in the financial markets by buying and selling outright (spot and forward) or under repurchase agreement, and by lending or borrowing claims and marketable instruments, whether in Community or in non-Community currencies, as well as precious metals;

•

conduct credit operations with credit institutions and other market participants, with lending being based on adequate collateral.

196

Open Issues in European Central Banking

18.2. The ECB shall establish general principles for open market and credit operations carried out by itself or the national central banks including the announcement of conditions under which they stand ready to enter into such transactions. Article 19 Minimum reserves 19.1. Subject to Article 2, the ECB may require credit institutions established in Member States to hold minimum reserves on accounts with the ECB and national central banks in pursuance of monetary policy objectives. Regulations concerning the calculation and determination of the required minimum reserves may be established by the Governing Council. In cases of non-compliance the ECB shall be entitled to levy penalty interest and to impose other sanctions with comparable impact. 19.2. For the application of this Article, the Council shall, in accordance with the procedure laid down in Article 42, define the basis for minimum reserves and the maximum permissible ratios between those reserves and their basis, as well as the appropriate sanctions in cases of non-compliance. Article 20 Other instruments of monetary control The Governing Council may, by a majority of two-thirds of the votes cast, decide upon the use of such other operational methods of monetary control as it sees fit, respecting Article 2. The Council shall, in accordance with the procedure laid down in Article 42, define the scope of such methods if they impose obligations on third parties. Article 21 Operations with public entities 21.1. In accordance with Article 104 of this Treaty, overdrafts or any other type of credit facility by the ECB or by the national central banks to Community institutions or bodies, Central Governments,

Statute of the ESCB and of the ECB

197

regional or local authorities, public authorities, other bodies governed by public law, or public undertakings of Member States shall be prohibited, as shall the purchase directly from them by the ECB or national central banks of debt instruments. 21.2. The ECB and national central banks may act as fiscal agents for the entities referred to in Article 21.1. 21.3. The provisions of this Article shall not apply to publiclyowned credit institutions, which in the context of the supply of reserves by central banks shall be given the same treatment by national central banks and the ECB as private credit institutions. Article 22 Clearing and payment systems The ECB and national central banks may provide facilities, and the ECB may make regulations to ensure efficient and sound clearing and payment systems within the Community and with other countries. Article 23 External operations The ECB and the national central banks may: •

establish relations with central banks and financial institutions in other countries and, where appropriate, with international organizations;

•

acquire and sell spot and forward all types of foreign exchange assets and precious metals; the term "foreign exchange asset' shall include securities and all other assets in the currency of any country or units of account and in whatever form held;

•

hold and manage the assets defined above;

•

conduct all types of banking transactions in relations with third countries and international organizations, including borrowing and lending operations.

198

Open Issues in European Central Banking Article 24 Other operations

In addition to operations arising from their tasks, the ECB and the national central banks may enter into operations for their administrative purposes or for their staff. CHAPTER V PRUDENTIAL SUPERVISION

Article 25 Prudential supervision 25.1. The ECB may offer advice to and be consulted by the Council, the Commission and the competent authorities of the Member States on the scope and implementation of Community legislation relating to the prudential supervision of credit institutions and to the stability of the financial system. 25.2. In accordance with the Council decision under Article 105(6) of this Treaty, the ECB may perform specific tasks concerning policies relating to the prudential supervision of credit institutions and other financial institutions with the exception of insurance undertakings. CHAPTER VI FINANCIAL PROVISIONS OF THE

ESCB

Article 26 Financial accounts 26.1. The financial year of the ECB and the national central banks shall begin on the first day of January and end on the last day of December. 26.2. The annual accounts of the ECB shall be drawn up by the Executive Board in accordance with the principles established by the Governing Council. The accounts shall be approved by the Governing Council and shall thereafter be published.

Statute of the ESCB and of the ECB

199

26.3. For analytical and operational purposes, the Executive Board shall draw up a consolidated balance sheet of the ESCB, comprising the assets and liabilities of the national central banks that fall within the ESCB. 26.4. For the applications of this Article, the Governing Council shall establish the necessary rules for standardizing the accounting and reporting of operations undertaken by the national central banks. Article 27

Auditing 27.1. The accounts of the ECB and the national central banks shall be audited by independent external auditors recommended by the Governing Council and approved by the Council. The auditors shall have full power to examine all books and accounts of the ECB and national central banks, and to be fully informed about their transactions. 27.2. The provisions of Article 188b of this Treaty shall only apply to an examination of the operational efficiency of the management of the ECB. Article 28 Capital of the ECB 28.1. The capital of the ECB, which shall become operational upon its establishment, shall be ECU 5.000 million. The capital may be increased by such amounts as may be decided by the Governing Council acting by the qualified majority provided for in Article 10.3, within the limits and under the conditions set by the Council under the procedure laid down in Article 42. 28.2. The national central banks shall be the sole subscribers to and holders of the capital of the ECB. The subscription of capital shall be according to the key established in accordance with Article 29. 28.3. The Governing Council, acting by the qualified majority provided for in Article 10.3, shall determine the extent to which and the form in which the capital shall be paid up.

200

Open Issues in European Central Banking

28.4. Subject to Article 28.5, the shares of the national central banks in the subscribed capital of the ECB may not be transferred, pledged or attached. 28.5. If the key referred to in Article 29 is adjusted, the national central banks shall transfer among themselves capital shares to the extent necessary to ensure that the distribution of capital shares corresponds to the adjusted key. The Governing Council shall determine the terms and conditions of such transfers. Article 29 Key for capital subscription 29.1. When in accordance with the procedure mentioned in Article 1091(1) of this Treaty the ESCB and the ECB have been established, the key for subscription of the ECB's capital shall be established. Each national central bank shall be assigned a weighting in this key which shall be equal to the sum of: •

50% of the share of its respective Member State in the population of the Community in the penultimate year preceding the establishment of the ESCB;

•

50% of the share of its respective Member State in the gross domestic product at market prices of the Community as recorded in the last five years preceding the penultimate year before the establishment of the ESCB.

The percentages shall be rounded up to the nearest multiple of 0.05 % points. 29.2. The statistical data to be used for the application of this Article shall be provided by the Commission in accordance with the rules adopted by the Council under the procedure provided for in Article 42. 29.3. The weightings assigned to the national central banks shall be adjusted every five years after the establishment of the ESCB by analogy with the provisions laid down in Article 29.1. The adjusted key shall apply with effect from the first day of the following year.

Statute of the ESCB and of the ECB

201

29.4. The Governing Council shall take all other measures necessary for the application of this Article. Article 30 Transfer of foreign reserve assets to the ECB 30.1. Without prejudice to the provisions of Article 28, the ECB shall be provided by the national central banks with foreign reserve assets, other than Member States' currencies, ECUs, IMF reserve positions and SDRs, up to an amount equivalent to ECU 50,000 million. The Governing Council shall decide upon the proportion to be called up by the ECB following its establishment and the amounts called up at later dates. The ECB shall have the full right to hold and manage the foreign reserves that are transferred to it and to use them for the purposes set out in this Statute. 30.2. The contributions of each national central bank shall be fixed in proportion to its share in the subscribed capital of the ECB. 30.3. Each national central bank shall be credited by the ECB with a claim equivalent to its contribution. The Governing Council shall determine the denomination and remuneration of such claims. 30.4. Further calls of foreign reserve assets beyond the limit set in Article 30.1 may be effected by the ECB, in accordance with Article 30.2, within the limits and under the conditions set by the Council in accordance with the procedure laid down in Article 42. 30.5. The ECB may hold and manage IMF reserve positions and SDRs and provide for the pooling of such assets. 30.6. The Governing Council shall take all other measures necessary for the application of this Article. Article 31 Foreign reserve assets held by national central banks 31.1. The national central banks shall be allowed to perform transactions in fulfillment of the obligations towards international organizations in accordance with Article 23.

202

Open Issues in European Central Banking

31.2. All other operations in foreign reserve assets remaining with the national central banks after the transfers referred to in Article 30, and Member States' transactions with their foreign exchange working balances shall, above a certain limit to be established through Article 31.3, be subject to approval by the ECB in order to ensure consistency with the exchange rate and monetary policies of the Community. 31.3. The Governing Council shall issue guidelines with a view to facilitating such operations. Article 32 Allocation of monetary income of national central banks 32.1. The income accruing to the national central banks in the performance of the ESCB's monetary policy function (hereafter referred to as 'monetary income') shall be allocated at the end of each financial year in accordance with the provisions of this Article. 32.2. Subject to Article 32.3, the amount of each national central bank's monetary income shall be equal to its annual income derived from its assets held against notes in circulation and deposit liabilities to credit institutions. These assets shall be earmarked by national central banks in accordance with guidelines to be established by the Governing Council. 32.3. If, after the start of the third stage, the balance sheet structures of the national central banks do not, in the judgement of the Governing Council, permit the application of Article 32.2, the Governing Council, acting by a qualified majority, may decide that, by way of derogation to Article 32.2, monetary income shall be measured according to an alternative method for a period of not more than five years. 32.4. The amount of each national central bank's monetary income shall be reduced by an amount equivalent to any interest paid by that central bank on its deposit liabilities to credit institutions in accordance with Article 19. The Governing Council may decide that national central banks shall be indemnified for cost incurred in connection with the issuance of

Statute of the ESCB and of the ECB

203

bank notes or in exceptional circumstances for specific losses arising from monetary policy operations undertaken for the ESCB. The indemnification shall be in a form deemed appropriate in the judgement of the Governing Council; these amounts may be offset against the national central banks' monetary income. 32.5. The sum of the national banks monetary income shall be allocated to the national central banks in proportion to their paid up shares in the capital of the ECB, subject to any decision taken by the Governing Council pursuant to Article 33.2. 32.6. The clearing and settlement of the balances arising from the allocation of monetary income shall be carried out by the ECB in accordance with the guidelines established by the Governing Council. 32.7. The Governing Council shall take all other measures necessary for the application of this Article. Article 33 Allocation of net profits and losses of the ECB 33.1. The net profit of the ECB shall be transferred in the following order: (a) an amount to be determined by the Governing Council, which may not exceed 20 % of the net profit, shall be transferred to the general reserve fund subject to a limit equal to 100 % of the capital; (b) the remaining net profit shall be distributed to the shareholders of ECB in proportion to their paid-up shares. 33.2. In the event of a loss incurred by the ECB, the shortfall may be offset against the general reserve fund of the ECB and, if necessary, following a decision by the Governing Council, against the monetary income of the relevant financial year concerned in proportion and up to the amounts allocated to the national central banks in accordance with Article 32.5.

This page intentionally left blank

INDEX D Democratic control Depression (the great)

Accountability 3,4, 11, 149, 152-169 -Index of 151-152 - Assessment of 152-157 Asian financial market crisis 29

145, 169 29

E EC Treaty 64 Economic and Monetary Union (EMU) 1,2 Employment 4 EMS crises 57 Euro 5, 107 European Central Bank (ECB) .. 1, 2, 4, 5, 6, 7, 20, 64, 90, 92 - Accountability of 144-146, 153-154, 157-167 - Balance sheet 69 -Budget 7, 8 - Decision-making 16, 17 25, 50, 163, 165 - Executive Board . 6, 7, 8, 9, 11, 16,28,48, 124, 130, 142, 163 - Governing Council 6,7,8, 9, 13,14, 16, 17,24,25,48, 123125, 147, 162, 165, 171-172 - Independence of 122-132, 138-139, 141-142 -Staff 10,24 -Threats to 132-135 European Commission 7, 27, 28,64 European Financial Supervisory Agency (EFSA) 172-173 European Liquidity Consortium (ELC) 173 European Monetary Institute 7, 18, 19,27,62,65, 159, 168, 170 European Parliament 168 European System of Central Banks (ESCB) 1,2,28 Eurosystem 1, 2, 3, 5, 28 - Balance sheet .. 3, 8, 43, 67, 68, 71,80,81

B Bail-outs 44,45,46,50,52 69 Banca d'ltalia Banco di Napoli 42 Banco do Portugal 70, 79 Bank of England ..43, 51, 52, 146, 153-154 Bank of Finland 69 Bank of Japan 70-71, 90, 153-154 Bank of New York 47, 49 Bank Reserve of Australia 93 Banque de France 21, 75 BCCI crisis 52 Bundesbank ....5,9, 10, 11, 16,28, 33,70,91-92, 122, 125, 126-128, 129-131 -Direktorium 12,28, 142 C Central bank independence ....3, 4, 54-55,60,76,78,93, 118-143, 146-147 - Literature on 125-132 - Legal independence 125 Central Bank of New Zealand 122 Chinese walls 92 Collateral 53 Constructive ambiguity 41 Contagion 47 Convergence 16 Council of Ministers 7, 13, 132 Credit Lyonnaise 42 Currency in circulation ....102, 103

205

206

Index

- Balance of powers ..9, 119-120 - Decision-making process ...5, 6, 9. 11, 16,51, 150 - Independence of 120 -Liabilities 78,80 - Monetary policy 76, 159 - Monetary policy operations .18, 46,48,51, 159 - Strengthening of 171-175 Eurozone 1, 2, 3, 5, 28 - Foreign exchange reserves ...83 F Fed Open Market Committee (FOMC) 9, 162 Federal Reserve ....5,9. 10, 11. 12, 16,43,47,70-71,86-88,90, 121122 -Governors 9, 10, 12 -Staff 10,21 Federal reserve banks 9 - Presidents 9 Financial fragility 2 Financial services 64 Financial stability 2, 8, 29, 43, 46,55,64, 121 - Prudential supervision ...29, 32, 35-39,41,50,51,52,53,54-55, 57,58-61,65,66, 138 -Regulation ....30,31,32,48,57 Foreign exchange interventions 133 Foreign exchange reserves 3, 8, 17 Foreign exchange policy 2, 5 Frankfurt 20 G Gold Group of Thirty Growth H Harmonisation I Independence

3,69,74,75,83 63. 64 4

33, 35, 49

3

Inflation ... 1,4, 15,54,60,76,77, 140, 148, 158 - Forecasts 161-162 Inter-bank deposits 33 Interest rates 1, 14, 121 Internal market 73 International Monetary Fund (IMF) 75, 90, 92 Ireland 15 Irish Central Bank 15 L LandesZentralbank 9, 12,28 Lender of last resort (LOLR) 2, 40,45,46,49,50,52, 173 Lender of penultimate resort 53 Liquidity crises 2, 40, 45, 46, 47, 49, 50 Loans 33, 53 M Maastricht Treaty ... 3, 4. 8, 11, 17, 29,42,44,68,73,91,94, 132 - Provision for seigniorage 96-98 Moral hazard 41, 45, 48, 52, 56 Monetary policy .... 1. 2, 3, 4, 5, 8, 15, 16, 17,25,43,67 - Centralisation 26, 51, 56 -Decentralisation ...2, 17, 19-21, 25-26,49,51 - Instruments and procedures 23, 43,47,53,56 - Transmission mechanism 16, 20 N National Central Banks (NCBs) 2, 3 , 4 , 5 , 6 , 8 , 11, 17,20,22,49 -Ambiguity 149 - Asset management 137 - Balance sheets ....68-69, 80, 84, 85,91,98-99, 106, 174 - Budget Committee 8 - Capitalisation .... 70, 72, 76, 77, 79,92 - Credibility of 118

Index - Foreign assets ....73, 74, 79, 82, 174 -Governors 6, 7, 8,9, 11, 13, 16,28, 172 -Stability 1,3, 11,29,43,84, 119, 121, 148-149, 158 -Staff 10,24, 117 (The) Netherlands 66 P Payment system -Target R Rescue operations Revaluation reserves

5, 8, 38 38

65 88-89

S Securities 33 Seigniorage 23, 94-117 - Definition of 94 - Distribution of 94, 96, 98, 99-101, 103-104, 174-175 - Germany's share of 101 Solvency crisis 40, 44, 48, 50 Stability and Growth Pact (SGP) 42, 138 Statute of the ESCB and the ECB 4,5,6,8,29,48,68,69,75,91, 96, 104, 138, 142, 147, 186-203 47 Stock market crash (1987) Swiss National Bank 125, 126-128 T Tender procedure 19 Tier-one assets 22, 25 Tier-two assets 22, 24, 25 Transparency ...147, 149, 151, 163 Treaty on European Union 177-185

NAMES INDEX Aksoy, Y Alesina, A Andreatta, B

27 140, 141, 142 90,91

B Bade, R Bagehot, W Barro, R Begg, D Begg, 1 Bini Smaghi, L Blejer, M Bordes, C Borio, C Briault, C Brookes, M

140, 142 47,48,60 140, 167, 168 140 62,65, 149 167 88,90,98 63,65 27,28 151, 167, 169 98

E Eichengreen, B 27, 28 Eijffinger, S 60,62,66,90, 93, 125, 140, 141, 142, 168 Eizenga, W 140, 142 : F Faust, J 168, 169 Favero, C 27, 167, 169 Feist, H 101, 116, 117 Fischer, S 141, 142, 168 Folkerts-Landau, D 62, 65 Friedman, B 62 Friedman, M 12, 13,27,66 Frenkel, J 63 Funauach, Y 141 G Garber, P 62,65 Garfinkel, M 168, 169 Gerlack, S 27,28 Giavazzi, F 27, 167, 169 Giovannini, A 62, 64, 65, 141, 143 Goldstern, M 63 Goodfriend, M 63,65, 168, 169, 170 Goodhart, C 52,54,58,59, 3,64,65,66, 141, 142 Gordon, D 167, 169 Gormley, L 168, 169 Green, D 62,65 Grilli, V 59,62,63,66, 76,78,90,93, 140, 141, 142, 143 Gros, D 27,28,90, 116, 117

Canzoneri, M 162, 167, 169 Castelo-Branco, M 64, 66 CEPR 169 Ciocca, P 140, 142 Cukierman, A 60, 62, 66, 125, 126, 129, 130, 140, 142, 167 D DeBandt, O 62,64 De Beaufort Wijnholds, J 62, 65,66 De Grauwe, P 27,28 deHaan, J 58,60,62,66,90, 93, 125, 140, 141, 142, 167, 168, 169 Demectzis, M 167, 169 Dewachter, H 27 Dornbusch, R 27, 28, 167, 169 Dowes, P 64

H Haldane, A 167 Hartmann, P 63 Heller, R 34,38,39,63,65,66 Hoogduin, L 62, 65, 66

208

Names Index Hughes Hallett, A Humphrey, T J Judson, R.A.,

167 63, 65, 66

97, 116

K Kaufman, H Keleher, R Kenen, P King, M King, R Kydland, F

63, 65 63,65,66 141, 143 167 63 168, 169

L Lannoo, K Llewellyn, L

27,28,63,65 63

M Marsch, D Masciandaro, D 78,90, 143 Masson, P Mayer, C McCallum, B Melitz, J Metzler, A Mishkin, F Morgan, J.P N Neyapti, B O Oh, S P Padoa-Schioppa, T Parkin, M Persson, T Porter, R.D Posen, A Prati, A Prescott, E Pringle, R

141, 142 59, 63, 66, 76, 62 64 168, 169 63,65 167 63 89,90,91

R Rojas-Suarez, L. Roger, S Rogoff, K Roll,

209 63 91,93 116, 117 146, 168, 169

S Schaling, E 60,62, 140, 142 Schinazi, G 64,65 Schobert, F 67,90, 116 Schoenmaker, D 52, 54, 58, 59, 63, 65 Schumacher, L 88, 90, 93 Schwartz, A.J 12, 13,27 Seitz, F 96, 116, 117 Sinn, H-W 101, 116, 117 Smets, F 27,28 Stella, P 91 Svensson, L 168, 169 Swinburne, M 64, 65 T Tabellini, G 59, 63, 66, 76, 78, 90,93, 141, 142, 143, 168, 169 V van'tHag, G Vaez-Zadeh, R Viegi,N Vives, X Volcker, P

58,62 64, 91 167 64 141, 142

140

168, 169

63, 66 140, 142 141, 168, 169 96, 116 139, 141 64,65,66 168, 169 90,93

W Walsh, C Webb, S Weisbrod, S

169 140 63