Central Banking in Eastern Europe (Routledge International Studies in Money and Banking)

Central Banking in Eastern Europe There is a growing academic consensus that governments can achieve lower inﬂation at...

Author: Nigel Healey | Barry Harrison

98 downloads 1022 Views 1MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

Central Banking in Eastern Europe

There is a growing academic consensus that governments can achieve lower inﬂation at a reduced social cost by making their central banks independent. Nowhere is this debate more relevant than in the transition economies of Eastern Europe, where the newly established central banks’ attempts to stabilise prices have come into conﬂict with the social objectives of national governments. This book, written by a multinational team of experts, explores the changing face of central banking in Eastern Europe in the light of modern macroeconomic thinking, providing important and novel insights into the design of monetary policy institutions. With its authoritative content, this book will interest students and academics involved with money and banking, macroeconomics and Eastern European studies. Professionals working for ﬁnancial institutions will also ﬁnd plenty that will appeal within these pages. Nigel Healey is Dean of Manchester Metropolitan University Business School, UK. Barry Harrison is Senior Lecturer in Economics at Nottingham Trent University, UK.

Routledge international studies in money and banking

1 Private Banking in Europe Lynn Bicker 2 Bank Deregulation and Monetary Order George Selgin 3 Money in Islam A study in Islamic political economy Masudul Alam Choudhury 4 The Future of European Financial Centres Kirsten Bindemann 5 Payment Systems in Global Perspective Maxwell J. Fry, Isaak Kilato, Sandra Roger, Krzysztof Senderowicz, David Sheppard, Francisco Solis and John Trundle 6 What is Money? John Smithin 7 Finance A characteristics approach Edited by David Blake 8 Organisational Change and Retail Finance An ethnographic perspective Richard Harper, Dave Randall and Mark Rounceﬁeld 9 The History of the Bundesbank Lessons for the European Central Bank Jakob de Haan 10 The Euro A challenge and opportunity for ﬁnancial markets Published on behalf of Société Universitaire Européenne de Recherches Financières (SUERF) Edited by Michael Artis, Axel Weber and Elizabeth Hennessy

11 Central Banking in Eastern Europe Edited by Nigel Healey and Barry Harrison 12 Money, Credit and Prices Stability Edited by Paul Dalziel and Barry Harrison 13 Monetary Policy, Capital Flows and Exchange Rates Essays in memory of Maxwell Fry Edited by William Allen and David Dickinson 14 Adapting to Financial Globalisation Published on behalf of Société Universitaire Européenne de Recherches Financières (SUERF) Edited by Morten Balling, Eduard H. Hochreiter and Elizabeth Hennessy 15 Monetary Macroeconomics A new approach Alvaro Cencini 16 Monetary Stability in Europe Stefan Collignon 17 Technology and Finance Challenges for ﬁnancial markets, business strategies and policy makers Published on behalf of Société Universitaire Européenne de Recherches Financières (SUERF) Edited by Morten Balling, Frank Lierman and Andrew Mullineux 18 Monetary Unions Theory, history, public choice Edited by Forrest H. Capie and Geoffrey E. Wood 19 HRM and Occupational Health and Safety Carol Boyd 20 Central Banking Systems Compared The ECB, the pre-Euro Bundesbank and the Federal Reserve System Emmanuel Apel 21 A History of Monetary Unions John Chown 22 Dollarization Lessons from Europe and the Americas Edited by Louis-Philippe Rochon and Mario Seccareccia 23 Islamic Economics and Finance: A Glossary (2nd Edition) Muhammad Akram Khan

24 Financial Market Risk Measurement and analysis Cornelis A. Los 25 Financial Geography A banker’s view Risto Laulajainen 26 Money Doctors The experience of international ﬁnancial advising 1850–2000 Edited by Marc Flandreau 27 Exchange Rate Dynamics A new open economy macroeconomics perspective Edited by Jean-Oliver Hairault and Thepthida Sopraseuth 28 Fixing Financial Crises in the 21st Century Edited by Andrew G. Haldane

Central Banking in Eastern Europe

Edited by Nigel Healey and Barry Harrison

First published 2004 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Simultaneously published in the USA and Canada by Routledge 270 Madison Ave, New York, NY 10016 Routledge is an imprint of the Taylor & Francis Group

This edition published in the Taylor & Francis e-Library, 2005. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” © 2004 Nigel Healey and Barry Harrison, editorial matter and selection; the contributors, individual chapters All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data A catalog record for this book has been requested

ISBN 0-203-46524-5 Master e-book ISBN

ISBN 0-203-77348-9 (Adobe eReader Format) ISBN 0-415-22619-8 (Print Edition)

Contents

List of illustrations Notes on contributors Preface Acknowledgements PART I

1 Theoretical perspectives on central bank independence

ix xii xvi xvii 1

3

ALI AL-NOWAIHI AND PAUL LEVINE

2 The background to reform: central banking in a command economy

45

NIGEL M. HEALEY AND JANET ILIEVA

3 Central banking in transition economies

68

NIGEL M. HEALEY AND JANET ILIEVA

PART II

4 The changing role of central banks in market economies

91

93

JANET ILIEVA, BARRY HARRISON AND NIGEL M. HEALEY

5 Central bank independence and macroeconomic performance: a survey of the evidence

107

FRIEDRICH KIßMER AND HELMUT WAGNER

6 Central bank independence and inﬂation performance in transition economies JANET ILIEVA, ANDROS GREGORIOU AND NICK TSITSANNIS

141

viii Contents PART III

7 Central bank independence in the Czech Republic

165

167

P A V E L S O U K U P , A N I T A T A C I A N D R O M A N M A T O U Sˇ E K

8 Central bank independence in Poland

193

ROBERT HUTERSKI, RICHARD NICHOLLS AND Z E N O N W I S´ N I E W S K I

9 Central bank independence in Romania

224

SILVIU CERAN, LILIANA DONATH AND BOGDAN DIMA

10 Central bank independence in Bulgaria

245

GARABED MINASSIAN, JANET ILIEVA AND OLGA TENEVA

Index

271

Illustrations

Figures 1.1 The inﬂationary bias 1.2 The inﬂationary bias in the open economy: co-operation and non-co-operation 1.3 Stabilization policy with perfectly correlated shocks 1.4 Stabilization policy with perfectly uncorrelated shocks 1.5 The choice of conservatism b/bm for the delegation regimes: ﬁve countries 1.6 The welfare outcomes relative to social optimum: ﬁve countries 1.7 The lower employment variance gain and loss for a single independent CB relative to the n others 1.8 The lower employment variance gain, G, and loss for one independent CB, L (one), and n ⫹ 1 independent CBs, (all), relative to n ⫹ 1 government-dependent CBs 2.1 Firm’s credit account 5.1 Average inﬂation and central bank independence (1955–1988) 5.2 Variance inﬂation and central bank independence 5.3 Average growth and central bank independence 5.4 Average growth variability and central bank independence 6.1 Inﬂation in selected transition economies, 1993–2000 10.1 Foreign currency deposits (in USD) and dollarisation of the economy (%) 10.2 Seigniorage (monthly, net issue of BGL as per cent from broad money) 10.3 New credits 10.4 Interest rates 10.5 Composition of the BGL deposits 10.6 Reﬁnancing in Bulgaria

11 17 19 21 22 23 27

29 56 119 120 127 127 158 252 253 254 255 255 259

x

Illustrations

Tables 2.1 2.2 2.3 2.4 2.5 2.6 2A.1 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 4.1 4.2 4.3 4.4 4.5 5.1 5.2 5.3 5A.1 5A.2 5B.1 5C.1 6.1 6.2 6.3 6.4a 6.4b 6.5a 6.5b 6.6 7.1

Banking system in socialist countries 51 Waiting for housing: international comparison, 1980s 54 Waiting for cars: international comparison, 1989 54 Sources of ﬁnancing the capital inputs 55 Foreign trade of CMEA members by trade bloc in 1970 and 1983 59 Investment and consumption 61 Occupational groups in the USSR in 1970 66 An overview of the transition economies 71 Recession and recovery in transition economies 73 The inﬂationary price of transition 75 Share of the unofﬁcial economy in GDP 77 General government budge balances 78 Broad money growth 81 External debt and servicing (1998) 82 The transition picture in 1999 86–87 Summary of major studies using GMT index of CBI 99 Political independence of the Bank of England (1694 to 1998) using the GMT index 100 Political independence of the Bank of France (1800–1993) using the GMT index 101 Political independence of the Federal Reserve (1913 to 1992) using the GMT index 103 Political independence of the German Bank (1870 to 1998) using the GMT index 104 The GMT index of political independence 113 The GMT index of economic independence 114 Legal indices of central bank independence 116 Bade and Parkin’s index of ﬁnancial independence 132 Bade and Parkin’s index of ﬁnancial independence 132 Indicator of central bank independence 133 The Cukierman, Webb and Neyapti (LVAW) (1992) indicator for central bank independence 133–136 Recent amendments of central bank law in Bulgaria, Czech Republic, Poland and Romania 150 CBI in twenty-two transition economies (2000) 154 Central bank independence in transition countries 156 Fixed effects 160 Random effects 160 Panel results 161 Panel results 161 GLS results 162 Selected economic indicators, 1991 to 2001 172

Illustrations xi 7.2 7.3 7.4 7.5 7.6 7.7 8.1 8.2 8.3a 8.3b 8.4 8.5 8A.1 8A.2 8A.3 9.1 10.1 10.2 10.3 10.4 10.5 10.6 10.7 10A.1

Bade and Parkin’s index of central bank independence 182 The GMT index (1991) of central bank independence 182 Alesian and Grilli’s (1992) index of central bank independence 183 Variables, weights and numerical coding for legal central bank independence 184 Banking supervision arrangements 185 Average inﬂation and GDP growth in chosen countries 188 Money supply growth in Poland, 1989 to 2001 195 Inﬂation, 1989 to 2002 199 Index of legal independence 201 Index of the turnover rate of the NBP Governor 202 Variables, weights and numerical coding 214 Indices of Cukierman, Webb and Neyapti – comparison 215 Index of legal independence (1997) 218 Index of legal independence (1998) 219 Index of legal independence (1999) 220 Principal steps of the monetary policy construction (1991–2001) 230–231 Financial institutions in Bulgaria, 1965 to 1980 248 Macroeconomic indicators 251 Functions of the central bank and the currency Board in Bulgaria 261 Bade and Parkin’s (1988) index of BNB independence 262 GMT index (1991) of BNB independence 263 Alesina and Grilli’s index (1992) of BNB independence 264 Cukierman, Webb and Neyapti’s index (1992) of BNB independence 265 Central bank independence in transition economies 268–269

Contributors

Silviu Cerna is Professor in the Faculty of Economics at the University of the West Timisoara, Romania. He teaches money and banking and international monetary economics at undergraduate level, and monetary policy at Master’s level. His main research interests are monetary economics and ﬁscal policy. Since 1992 he has been a member of the Board of The National Bank of Romania. He is also a member of the Romanian Economists Association, the French Speaking International Association and the International Institute of Public Finance. Bogdan Dima is Lecturer in the Faculty of Economics at the University of the West Timisoara, Romania. He teaches money and banking and international monetary economics at undergraduate level, and monetary and ﬁscal policy in the euro zone at Master’s level. His main research interests are monetary economics and ﬁscal policy. He is a member of the Romanian Economists Association, the French Speaking International Association and the International Institute of Public Finance. Liliana Eva Donath is Professor in the Faculty of Economics at the University of the West Timisoara, Romania. She teaches money and banking and public ﬁnance at undergraduate level, and tax administration and foreign exchange management at Master’s level. Her main research interests are monetary economics and ﬁscal policy. She is a member of the Romanian Economists Association, the French Speaking International Association and the International Institute of Public Finance. Andros Gregoriou is currently Lecturer in accounting and ﬁnance at Brunel University. His research interests are in market microstructure, ﬁnancial accounting, ﬁnancial econometrics and transition economies. He has published in mainstream economics and ﬁnance journals such as Applied Financial Economics, Journal of Business, Finance and Accounting, Ekonomia and the European Journal of Finance. Barry Harrison is Senior Lecturer at Nottingham Trent University and

Contributors xiii holds a visiting post at the European University at St Petersburg. He has previously held a visiting post at the National Institute for Management of the Economy in Baku (Azerbaijan) and has also taught in France, Germany and Holland. He has published articles on ﬁnance, macroeconomics and European integration. Nigel Healey is Professor and Dean of the Manchester Metropolitan University Business School. His research interests are in macroeconomic developments in Central and Eastern Europe and the Commonwealth of Independent States, and company strategy and economic integration in the European Union. He has served as policy adviser to the governments of Russia and Belarus. Robert Huterski is Lecturer at the Nicholas Copernicus University in Torun, Poland, and is also Lecturer at the High Banking School in Torun. Between 1990 and 1995 he worked for the National Bank of Poland in the Economic and Banking Supervision departments. His research interests include central banking, banking supervision, public ﬁnance and international ﬁnance. Janet Ilieva is a research student at the Manchester Metropolitan University Business School. Her current research interests are monetary policies in Europe and economic transition of Central and Eastern European economies. Friedrich Kißmer is Senior Lecturer in Economics at the University of Hagen (Fern Universität). Paul Levine is Professor of Economics at the University of Surrey. One of his main research areas is macroeconomic policy with a particular emphasis on credibility and co-ordination issues. He has published numerous articles on these topics in the Economic Journal, the European Economic Review, the Journal of Economic Dynamics and Control, the Journal of Monetary Economics and Oxford Economic Papers. Roman Matous˘ek is Senior Lecturer in Financial Economics at the Business School, London Metropolitan University. Previously, he was a banking expert in the Czech National Bank. He has been involved in a number of international research projects that have focused on ﬁnancial sector reform in transition economies. He has been a Visiting Fellow at London Business School and at the Bank of England. In addition, he received a prestigious Pew Fellowship from Georgetown University, USA. Garabed Minassian is a member of the Managing Board of the Bulgarian National Bank and, since 1997, he has been a member of the Economic Policy Advisory Board of the President of the Republic of Bulgaria. His

xiv Contributors main research interests are currency board arrangements, monetary policies and inﬂation. Richard Nicholls is Lecturer at the Poznan University of Economics. His main research interests include: the management of banks; central bank independence; methods of measuring customer satisfaction; service quality management; and post-socialist transition economies. Ali al-Nowaihi is Lecturer in Economics at the University of Leicester. His main interests are in industrial organisation, macroeconomics and mathematics. His publications have appeared in major journals including the Journal of Economic Theory, the Journal of Monetary Economics and the Journal of Algebra. Pavel Soukup is an economic analyst at the Monetary and Statistics Department of the Czech National Bank. His main research interests are central bank independence, inﬂation targeting, banking regulation, trends in the banking sector and the impact of developments in public ﬁnance on monetary policy. Anita Taci works as an economist in the Chief Economist Ofﬁce at the European Bank for Reconstruction and Development (EBRD), London. During 1998 and 1999 she worked as a consultant for the World Bank. Simultaneously she held a post as a Research Fellow on banking issues at the Economic Institute of the Academy of Sciences in the Czech Republic. In 1996 she was awarded a scholarship from New York University Leonard N. Stern School of Business. Her main research interests are banking and ﬁnancial markets, and public budget management. Olga Teneva is Head of the Department of Economic Sciences at the Academy of the Ministry of the Interior. She has previously served as adviser to the President and the Governor of the Bulgarian National Bank. Nick Tsitsannis is currently a Research Fellow at the University of Kent. His research interests are in labour economics and he has published in mainstream economics journals such as Applied Financial Economics. Helmut Wagner is Professor of Economics and Director of the Institute of Macroeconomics at the University of Hagen (Fern Universitaet), Germany. Prior to that, he was Professor of Economics at HWP/University of Hamburg. He has held visiting positions at the University of California (1982 to 1983), MIT (1987), the Bank of Japan’s Institute for Monetary and Economic Studies (1988), Princeton University (1991 to 1992), AICGS/The Johns Hopkins University (1997), Harvard University (2000), and (several times) at the IMF. He has published extensively in the ﬁelds of macroeconomics, monetary and international economics.

Contributors xv Zenon Wis´niewski is Professor of Economics at the Nicolaus Copernicus University in Torun, Poland. He has been a visiting scholar at the Universities of Glasgow, Leicester, Bamberg and Munich, and at the Institute for Employment Research of the Federal Employment Services (IAB) in Nuremberg. His main research interests are labour markets and economic policy, and he has published extensively in leading economic journals.

Preface

For almost two decades now the delegation of monetary policy to an independent central bank has been put forward as a possible solution to the well-known inﬂationary bias of time-inconsistent policy. It is now recognised that, in addition to considerations of reputation, the solution to the time-inconsistency problem involves central banks as independent agents facing incentives that differ from those of the representative government (the principal). The basic idea is that a conservative central banker will implement time-consistent policy which will therefore deliver lower inﬂation costs in terms of output and employment. Until recent times, most of the literature on independent central banks has focussed on developed countries. Far less attention has been paid to the transition economies of Central and Eastern Europe. The collection of research papers presented in this book seeks to remedy this omission in the literature. Part I of the book reviews the background to central bank independence and provides a background for Part II assesses the experience of ﬁve transition economies and is the core of the book. Part III provides a wide ranging survey of independent central banks in twentythree transition economies and assesses their independence against the independence of the ECB. The conclusion of these papers is that substantial progress has been made by the transition economies reviewed in establishing new laws and regulations which provide for an independent central bank. What emerges is that a gap still exists between the legal degree of independence and the actual degree of independence possessed by these newly created central banks. In particular, instead of relying on direct regulations which distance the inﬂuence of government from the activities of their central bank, governments sometimes use political pressure to inﬂuence the decisions of central bankers. There is widespread recognition that the autonomy of central banks is fundamental to the reform process, providing a commitment to price and exchange rate stability. It is hoped that this book makes a contribution to debate on this issue.

Acknowledgements

The task of producing a book invariably involves accumulating huge debts. This is particularly true when the book is a collective endeavour. Our major debts are undoubtedly to our fellow contributors who brought the full extent of their knowledge and expertise to bear in this edited volume. They have all worked tirelessly, sometimes under great time pressure to complete their contributions. While we express our gratitude to all our fellow contributors, it would be invidious of us not to single out Janet Ilieva for special praise. At various times during the writing of this book, Janet has been our author, research assistant and secretary. But for the energy and effort she has devoted to putting this book together, it would never have been completed. We are also grateful to the staff at Routledge who were both patient and supportive and in particular we thank Terry Clague and Robert Langham. We also gratefully acknowledge the help of Gail Welsh and Ann King in guiding this book through to the production stage of publication. There are also debts closer to home and economists are very familiar with the concept of opportunity cost. Time spent writing and editing is time that could be spent doing other things. Each of the contributors to this book gratefully acknowledges the support of those closest to them. Families particularly tolerate the long working hours that characterise the lifestyle of academics the world over. We hope our readers ﬁnd the results worthwhile. Barry Harrison Nigel Healey

Part I

1

Theoretical perspectives on central bank independence Ali al-Nowaihi and Paul Levine

1 Introduction The focus of this chapter is the theoretical literature on central bank (CB) independence. We provide a survey of the seminal contributions and of more recent research which reassesses the earlier work.1 Section 2 describes the ‘credibility problem’. An expectations augmented Phillips curve (or Lucas supply curve) is derived from a micro-foundations model of labour market and ﬁrms’ behaviour. Having clearly set out the source of an ‘inﬂationary bias’ in the conduct of monetary policy, Section 3 describes a second-best solution to the credibility problem ﬁrst proposed by Rogoff (1985a). The solution is to delegate monetary policy to an independent central bank with an appointed board chosen to be ‘conservative’, in the sense that they assign a higher priority to low inﬂation than that of the representative government. This results in a trade-off between low inﬂation and effective monetary stabilization policy, and allows for both goal and instrument independence. The government now exercises inﬂuence by its choice of banker which could alternatively be interpreted as choosing a particular degree of CB independence. The Rogoff delegation game is of interest because it appears to correspond to the game now being played with more and more countries delegating monetary policy to independent and (presumably) conservative bankers. However, we highlight a number of problems with this solution. The first is that the government has to find a central banker or choose a degree of independence which results in exactly the right degree of inflation averseness. The second is more fundamental and is of particular relevance to transition economies: the solution assumes that commitment to the type of banker or degree of independence is possible whereas commitment to a monetary rule is not. The public must be reassured that once their expectations of inflation are formed, the government will not sack or overrule the banker and appoint a less conservative one. The third problem is that when open-economy aspects are introduced and a role for fiscal policy allowed, the resulting interactions may result in an equilibrium that is inefficient even when compared with the

4

Ali al-Nowaihi and Paul Levine

original equilibrium with representative (i.e. government-dependent) bankers. If, as we assert, Rogoff-type delegation is the game being played, we need to address a puzzle that has emerged in very robust form from the empirical literature on CBI and its effect on economic performance: governmentindependent central banks provide a ‘free lunch’, in that lower inﬂation and lower inﬂation variance is apparently achieved at no cost in terms of greater output or employment variance.2 This contrasts with the prediction of Rogoff’s model that delegating monetary policy to an independent and inﬂation-averse central bank would result in lower average inﬂation, but at a cost of an increase in the variances of output and employment. Following Alesina and Gatti (1995), we introduce political parties and political uncertainty into an open-economy model. Our synthesis of political and openeconomy effects provides a new perspective on how the empirical puzzle can be reconciled with Rogoff-style delegation. Section 4 turns to ﬁrst-best solutions to the elimination of the inﬂationary bias. These consist of, ﬁrst, an equilibrium sustained in a repeated game by a private sector trigger strategy (Barro and Gordon, 1983) and second, a contract approach pioneered by Walsh (1995). In the contract approach the CB is instrument-independent but not goal-independent in that the inﬂation rule is chosen by the government and made incentive-compatible for an independent CB by an appropriate incentive mechanism such as an income transfer. The main problem with both solutions is identiﬁed as relying on equilibria concepts which are not renegotiation-proof, though in rather different senses. Trigger strategies are not renegotiation-proof in the sense suggested by Farrell and Maskin (1989), while Walsh contracts fail the test of renegotiation-proofness employed in agency theory. However, the strength of the Walsh contract, which we demonstrate, is that it can make the conduct of monetary policy open and accountable, and thus facilitate a reputational solution to the credibility problem in which zero (or low) inﬂation is combined with effective stabilization policy. Sections 2, 3 and 4 assume that governments are benevolent in the sense that they maximize either a social welfare function for society as a whole or, in the case of the partisan political business cycle model, governments are led by different political parties each of which is representative of a particular section of society. Section 5 assumes that governments value power for its own sake and derive utility from being in power as well as raising social welfare. We also introduce government competence. These two features lead to rational political monetary cycle models in which economic ﬂuctuations result from a government’s attempt to signal its competence to the electorate. However, we show that political monetary cycles can be avoided if the Walsh-type contracts of Section 4 between the government and the CB are available. Given the problems associated with Rogoff delegation, our overall conclusion and lesson for the transition economies is that mechanisms for pur-

Theoretical perspectives on CBI 5 suing the ﬁrst-best solution to the ‘credibility problem’ with ﬂexible inﬂation targets chosen by the government, and made incentive-compatible for instrument-independent (but not goal-independent) CBs, may well constitute the most promising approach to institutional structure with the monetary regime in place in New Zealand since 1989 as the model.

2 The credibility problem The expectations augmented Phillips curve (EAPC) Our starting point is the well-known expectations augmented Phillips curve (EAPC). Lucas (1972) provides early micro-foundations for this aggregate supply relationship based on the idea that producers only observe their own goods’ prices and not the aggregate price level. Here we provide a more ‘European’ perspective of the phenomenon in which labour market distortions take centre stage. Consider the determination of nominal wage contracts by wage-setters, who can either be regarded as monopoly unions or minimum wage legislators. They set one-period nominal wage contracts to minimize an expected welfare loss ␨t⫺1(Ut) where: Ut ⫽ v(wt ⫺ pt ⫺ w ˆ )2 ⫹ (1 ⫺ v)(lt ⫺ lˆ )2

(2.1)

All variables in (2.1) are expressed in logarithms, wt is the nominal wage, pt is the domestic price level and lt denotes employment. (2.1) says that wage-setters have bliss points w ˆ for the real wage and lˆ for employment and choose the nominal wage to get as close to (w ˆ , lˆ ) in real wage–employment space consistent with the model of employment. To derive the latter, assume a Cobb-Douglas production function: Yt ⫽ AKtbLt1⫺␤exp(⫺ut) where Yt, Kt and Lt denote output, capital stock (assumed exogenous throughout) and employment respectively in levels, and ut is a zero-mean negative productivity shock. Then, equating the real wage with the marginal product of labour we have: wt ⫺ pt ⫽ f(Kt) ⫺ ␤lt ⫺ ut

(2.2)

We allow for a state-contingent component of the wage contract through an indexing arrangement that links the ex-post nominal wage to the price level. Writing: w 苶t ⫽ ␨t⫺1(wt)

6

Ali al-Nowaihi and Paul Levine

then: wt ⫽ 苶 wt ⫹ wt ⫺ w 苶t the expected nominal wage, is found by minimizing (2.1) subject to (2.2). Performing this optimization, with indexing, the realized nominal wage is given by: wt ⫽ 苶 wt ⫹ ␬[p ⫺ ␨t⫺1(p)]

(2.3)

where: w ˆ ⫹ a(f(Kt) ⫺ ␤lˆ ) w 苶t ⫽ ␨t⫺1(pt) ⫹ ᎏᎏ ; (1 ⫹ a) a ⫽ (1 ⫺ v)/(v␤2)

(2.4)

Combining (2.3) and (2.4), the real wage is: w ˆ ⫹ a(f(Kt) ⫺ ␤llˆ ) wt ⫺ pt ⫽ (1 ⫺ k)(pt ⫺ ␨t⫺1(pt)) ⫹ ᎏᎏᎏ (1 ⫹ a)

(2.5)

Thus if v ⫽ 0, a → ⬁ the expected real wage will achieve the bliss employment level, l, speciﬁed in (2.1). At the other extreme, if v ⫽ 1, a ⫽ 0 the expected real consumption wage equals the bliss consumption wage target, w ˆ. This model of wage determination together with employment given by (2.2) and a Cobb-Douglas production function completes the supply side of the model for a given exogenous capital stock. Combining (2.2) and (2.5) we have: (1 ⫺ k) ut lt ⫽ 苶l ⫹ ᎏᎏ[pt ⫺ ␨t⫺1(pt)] ⫺ ᎏᎏ ⫽ l苶 ⫹ ␨[␲t ⫺ ␰t⫺1(␲t)] ⫺ ⑀t ␤ ␤

(2.6)

where l苶 is the equilibrium employment level deﬁned as: ␣␤l苶u ⫹ f(K) ⫺ w ˆ l苶 ⫽ ᎏᎏ ␤(1 ⫹ ␣)

(2.7)

In addition, we have deﬁned ␰ ⫽ (1 ⫺ ␬)/␤, ⑀t ⫽ ut/␤ and expressed the relationship in terms of inﬂation ␲t ⫽ pt ⫺ pt⫺1. (2.6) is our EAPC giving the familiar positive effect of surprise inﬂation on employment in the absence of full indexing (␬ ⫽ 1).

Theoretical perspectives on CBI 7 The monetary policy game Following the seminal article by Barro and Gordon (1983), the credibility problem involving the conduct of monetary policy is usually formalized in the literature in terms of a game between the private sector and the CB. The private sector consists of consumers, ﬁrms and wage-setters. We assume large numbers of these agents so the private sector is atomistic and does not act strategically. For the purposes of the game their behaviour is described comprehensively by the EAPC which then constitutes the economic environment or constraint facing the CB given the expectations of inﬂation. It remains to describe the moves and payoff of the CB. It is convenient to deﬁne the former as inﬂation rates set in each period. Of course CBs do not in fact ‘set’ inﬂation; but we can think of them as having inﬂation targets which they achieve successfully3 using the usual array of actual monetary instruments (the money supply, short-term interest rates, reserve ratios and so on). What is important is that the actual targets are credible in the sense described below. The CB objective is to achieve price stability; that is, zero inﬂation, but it is also aware that monetary policy has real effects in the short term and can therefore increase employment above its equilibrium level. Ideally, all policy-makers including the CB would prefer full employment to unemployment. If n is the full-employment level (still in logarithms and assuming a ﬁxed individual supply of labour by households) then the unemployment rate in equilibrium (the natural rate or NAIRU) is approximately n ⫺ lˆ ⫽ û, say. In employment–inﬂation space the bliss point of the CB is then (n, 0). These preferences may be captured by the social welfare function: Wt ⫽ b(lt ⫺ n)2 ⫹ ␲t2

(2.8)

This asserts that the variability of employment creates welfare costs. Whether this rationale for stabilizing the economy using monetary and ﬁscal policy is valid is a question that remains an unresolved controversy in macroeconomics. To explore this issue, ﬁrst suppose that the labour market clears and there is no unemployment. Any stabilization gains are then only those from stabilizing consumption about its mean. Lucas (1987) argues that these are insigniﬁcant. For a logarithmic utility function in consumption, this is easily seen by expanding log Ct as a Taylor series about its trend path log C 苶t and taking expectations to give the approximation ␨(log(Ct)) ⬇ log C 苶t ⫺ (␴c/C 苶t)2 where ␴c2 is the variance of consumption. Thus eliminating the variability 苶t)2 a generous estimate of of consumption raises expected utility by (␴c/C

8

Ali al-Nowaihi and Paul Levine

which is 0.5 ⫻ (0.02)2 ⫽ 0.0002. Compare this with raising mean consumption by 1 per cent which raises expected utility by an amount 0.01. Thus eliminating consumption variability of up to 2 per cent of mean consumption is equivalent to less than the welfare gain from a 0.02 per cent increase in consumption. Suppose now that unemployment is the only source of uncertainty facing the household who have one level of consumption CE when employed and CU when unemployed. The probability of the representative household being unemployed is 1 ⫺ u, where u is the unemployment rate. Then expected utility is (1 ⫺ u)log CE ⫹ u log CU and is independent of the variability of the unemployment rate. Again there seems to be no case for stabilization policy as implied by our loss function (2.8). Any potential gains from stabilization policy must be in the form of indirect beneﬁts not captured by the model. One possible source arises from the hysteresis effects of long-term unemployment. According to this view the long-term unemployed lose their skills and the work habit, and in effect withdraw from participation in the labour market. This raises the equilibrium level of unemployment. Another possible indirect beneﬁt from stabilization arises from the effect of output variability on investment. If the latter is irreversible then the ‘new investment theory’ (see e.g. Dixit and Pindyck, 1994) suggests that output uncertainty from supply shocks could reduce investment signiﬁcantly (treated as exogenous in our model). This could have substantial welfare consequences especially in an endogenous growth context. These linkages have yet to be fully explored even theoretically in the macroeconomic literature. Informally these arguments suggest a social welfare function of the type (2.8) with output replacing employment. A second ad hoc feature of these loss functions is the assertion of convex costs of inﬂation and an implied bliss point of zero inﬂation. This is another difﬁcult area in macroeconomics. It seems unlikely that the shoeleather costs associated with a positive nominal interest rate amount to signiﬁcant welfare costs. Costs associated with tax distortions (not present in our model anyway) could be easily overcome by indexation. The most plausible costs are associated with the link between high inﬂation, its variability and investment. It would be true especially in an endogenous growth context, but again these linkages are not well understood. In the absence of a well-articulated theory of the costs of output or unemployment variability and of the costs of inﬂation (other than the insigniﬁcant shoe-leather costs), we stick to a second-best research strategy of combining a micro-foundations model with a plausible, but ad hoc, loss function (2.8). The remaining ingredient in the game is the model of expectations. Before the ‘rational expectations’ revolution, for instance in the work of Milton Friedman who introduced the EAPC into macroeconomics, the usual scheme was the adaptive expectations rule:

Theoretical perspectives on CBI 9 ␨t⫺1(␲t) ⫽ ␨t⫺2(␲t⫺1) ⫹ ␭[␲t⫺1 ⫺ ␨t⫺2(␲t⫺1)]; ␭ 僆 [0, 1]

(2.9)

which says that this period’s forecast equals last period’s forecast plus some proportion of the observed forecast error. With this formulation there is no credibility problem. The credibility or ‘time-inconsistency’ problem, ﬁrst raised by Kydland and Prescott (1977), only emerges with rational expectations. In game-theoretic terms this amounts to assuming a complete information game, or a game in which the private sector knows the nature of the CB’s calculations and uses this knowledge to form its expectations. Pre-commitment and the time-inconsistency problem To see the nature of the credibility problem assume ﬁrst that the CB precommits to an inﬂation rule which takes the form of deterministic plus stochastic shock-contingent components ␲t ⫽ 苶 ␲ ⫹ ␮⑀t with 苶 ␲ and ␮ to be determined. The sequence of moves is: 1 2 3

The CB commits itself at the beginning of time t ⫽ 0 to apply the rule to all future periods t ⫽ 1, 2, . . . In period t ⫺ 1 the private sector forms an expectation ␨t⫺1(␲t); t ⫽ 1, 2, . . . In period t the CB observes the shock ⑀t and implements the rule ␲t ⫽ 苶 ␲ ⫹ ␮⑀t.

Since there are no structural dynamics in this set-up, the CB’s optimization problem is to minimize the expected welfare loss ␨0(Wt) given the EAPC, the sequence of events and the rational expectations assumption. From the latter the CB can put ␨t⫺1(␲t) ⫽ 苶 ␲. Hence from the EAPC (2.6) and the welfare loss function (2.8) we can write: 2 2 2 2 ␨0(Wt) ⫽ ␨0[(␲ 苶 ⫹ ␮⑀t) ⫹ b(␰␮⑀t ⫺ u 苶 ⫺ ⑀t) ] ⫽ ␲¯ ⫹ bu 苶 ⫹ 2 2 2 [␮ ⫹ b(1 ⫺ ␰␮) ␴⑀

(2.10)

where ␴⑀2 ⫽ var(⑀t). Minimizing (2.10) leads to the result: b␰ ␲ 苶 ⫽ 0; ␮ ⫽ ᎏ2 1 ⫹ b␰ i.e., the optimal rule is given by: b␰ ␲t ⫽ ᎏᎏ⑀ 1 ⫹ b␰2 t

(2.11)

10

Ali al-Nowaihi and Paul Levine

The optimal pre-commitment inﬂation rule then consists of zero average inﬂation plus a shock-contingent component which sees inﬂation raised (that is, monetary policy relaxed) in the face of a negative supply shock. Although the pre-commitment solution is optimal ex ante at stage 1 of the game, it ceases to be optimal ex post at stage 3 when the rule is to be implemented. If the CB were to re-optimize at this stage, then it would take expectations ␨t⫺1(␲t) ⫽ 0 as given. Observing the shock it would then choose the inﬂation rate to minimize: Wt ⫽ ␲t2 ⫹ b(l苶 ⫹ ␰␲t ⫺ ⑀t ⫺ n)2

(2.12)

leading to a revised higher inﬂation rate: b␰ ␲t ⫽ ᎏᎏ(u 苶 ⫹ ⑀t) 1 ⫹ b␰2

(2.13)

If no mechanism exists to enforce pre-commitment then the rational private sector will anticipate this act of reneging and the ex ante optimal rule with zero average inﬂation will lack credibility. The only credible rule is then that consistent with discretion or, in other words, period-by-period optimization by the CB. Discretion With discretionary policy the sequence of events is now: 1 2

In period t ⫺ 1 the private sector forms an expectation ␨t⫺1(␲t); t ⫽ 1, 2, . . . In period t the CB observes the shock ⑀t and chooses the inﬂation rate ␲t.

To solve this game we must proceed by backwards induction and start at stage 2. Given expectations ␨t⫺1(␲t) ⫽ ␲te say, the welfare loss (2.8) can be written: 2 Wt ⫽ ␲t2 ⫹ b[␰(␲t ⫺ ␲te) ⫺ ⑀tu 苶]

(2.14)

Then minimizing with respect to ␲t, the ﬁrst order condition is: ␲t ⫽ b 苶␰苶\苶1苶苶⫹ 苶苶b 苶␰苶2苶[u 苶 ⫹ ␰␨t⫺1(␲t) ⫹ ⑀t)

(2.15)

Proceeding to stage 1 the private sector uses (2.14) to form the expectation: ␨t⫺1(␲t) ⫽ b␰u 苶

(2.16)

Theoretical perspectives on CBI 11 Hence the discretionary policy takes the form: b␰ ␲t ⫽ b␰u 苶 ⫹ ᎏ2 ⑀t 1 ⫹ b␰

(2.17)

which has the same state-continent component as the ex ante optimal rule, but now includes a non-zero average inﬂation or inﬂationary bias equal to b␰u 苶 where u 苶 is the NAIRU. The credibility problem may be stated simply as how to eliminate the recalcitrant inﬂationary bias. The latter is illustrated in Figure 1.1. The ex ante optimal policy consistent with rational expectations (␲t ⫽ ␲te on the LRPC) is zero inﬂation at point P, but this requires pre-commitment to enforce. In the absence of such pre-commitment at time t expectations ␲te formed in the previous period are given and the policy-maker may choose a point on the SRPC to reach a utility curve closer to the bliss point B, at point C. This is a ‘cheating’ policy in which zero inﬂation is promised and believed, but non-zero inﬂation is delivered. However, in a rational expectations equilibrium the private sector can anticipate the calculations of the policy-maker. High inﬂation is anticipated and we end up on the LRPC

Figure 1.1 The inﬂationary bias.

12

Ali al-Nowaihi and Paul Levine

with employment back at its natural rate lt ⫽ l苶 and high inﬂation at point D. PD is the inﬂationary bias b␰u 苶. B is the bliss point at which lt ⫽ n (full employment) and ␲t ⫽ 0. PC is the SRPC with ␲te ⫽ 0. PD ⫽ the inﬂationary bias equal to b␰u 苶. Friedman’s ‘long, variable and uncertain lags’ The proposition that monetary policy should be ‘depoliticized’, by removing the power to set interest rates and monetary growth from the hands of elected governments, has a long history. Friedman (1968), for example, was an early advocate of a clear monetary rule (the so-called ‘k-per cent rule’). Friedman’s concern was primarily that, given the long, variable and uncertain lags involved, the discretionary use of monetary policy for stabilization purposes might destabilize the real economy and lead to higher inﬂation. His call for a constitutional rule for monetary growth was designed to remove the scope for destabilizing, discretionary policy. Friedman raises two separate issues here: ﬁrst, he points to the inability to forecast the effect of a particular monetary action on the price level and output, an issue of model uncertainty. The problem here is not long lags, which do not in themselves imply policy ineffectiveness. On the contrary the longer the lags the more active the stabilization response of monetary policy should be. However, variable and uncertain lags do imply a more cautious use of stabilization policy rather than its complete absence (Fischer and Cooper, 1973). The second issue concerns long lags and their implication for the timeinconsistency problem. Here Goodhart and Huang (1995) make a simple but important point: if monetary instruments operate with a lag that exceeds the length of the wage contract, the wage-setters will observe them and adjust their expectations accordingly. Hence both the SRPC and the time-inconsistency problem disappears! To assess this critique of the time-inconsistency literature, it is important to be precise about the meaning of the term ‘lag’. Estimates of the lag between changes in monetary instruments and changes in output or inﬂation usually refer to the mean lag. Supposing that one accepts a mean lag of a year or more between a change in an observable monetary instrument such as the nominal interest rate and a change in the price level. Then even with wage contracts of one year in duration there will be some change in prices within the contract period, perhaps as much as half of the long-run impact (if the mean is close to the median lag). The Goodhart and Huang critique implies that the inﬂationary bias with lags is less than that without lags; roughly speaking the parameter ␰ in the inﬂationary bias b␰u 苶 is reduced in size by the presence of lags. This of course leaves the possibility that the bias is still signiﬁcant if b and u 苶 are sufﬁciently large.

Theoretical perspectives on CBI 13

3 Rogoff delegation: a second-best solution to the credibility problem Rogoff (1985a) proposes a second-best solution to the credibility problem involving a trade-off between low average inﬂation and effective monetary stabilization policy. The solution is to delegate monetary policy to an independent central bank with an appointed board chosen to be ‘conservative’, in the sense that they assign a higher priority to low inﬂation than that of the representative government. An optimal choice of conservatism will then see bankers appointed who deliver low average inﬂation, but are not so over-conservative as to prevent monetary stabilization. The details of the delegation equilibrium are as follows: suppose that both the government and the appointed bankers have preferences represented by the welfare loss function of the form (2.8), but with a different weight on employment. For the government a weight b ⫽ bm is adopted representing the preferences of the median voter. The point of delegation is that the bankers have different preferences (b 苷 bm). One would expect bankers to be naturally conservative, in which case their weight on employment b is less than bm.4 Some bankers are more conservative than others and, in principle, by asking potential appointees how much higher inﬂation is worth sacriﬁcing for a 1 per cent reduction in unemployment, the government can employ a CB executive with a particular degree of conservatism. In what follows we assume the government does this in such a way as to minimize its own welfare loss function. The CB is subsequently goal- and instrument-independent, and pursues a discretionary policy corresponding to its own preferences. An alternative interpretation of choosing b ⬍ bm is that it represents the degree of central bank independence. A completely independent bank would eliminate the inﬂuence of the government over monetary policy. Then, assuming that bankers are naturally very conservative, the result would be a choice of the weight b close to zero. At the other extreme a low degree of independence would result in the CB adopting a weight on employment close to bm. The choice of b between these extremes is then one of choosing an institutional structure for the CB which allows for a particular degree of government inﬂuence over the conduct of monetary policy. The sequence of events in this ‘delegation game’ is as follows: 1

2 3

The degree of independence of the CB or the type of banker is chosen by the government determining the weight b for subsequent periods t ⫽ 0, 1, 2. In period t ⫺ 1 the private sector forms an expectation ␨t⫺1(␲t); t ⫽ 1, 2, . . . In period t the CB observes the shock ⑀t and chooses the inﬂation rate ␲t.

14

Ali al-Nowaihi and Paul Levine

For simplicity let us assume there is perfect control of inﬂation (assuming imperfect control in fact only results in trivial changes to the delegation game). Solving the game by backward induction, at stage 3 the inﬂation rate is chosen to be (2.17). Hence ␲te ⫽ b␰u 苶 at stage 2. At stage 1 the government chooses b to minimize E0(Wt) where: Wt ⫽ bm(lt ⫺ n)2 ⫹ ␲t2 ⫽ bm[␰(␲t ⫺ ␲te) ⫺ ⑀t ⫺ (n ⫺ lˆ )]2 ⫹ ␲t2

(3.1)

Substituting for inﬂation from (2.16) and (2.17) and putting 苶 u ⫽ n ⫺ 苶l and var(⑀t) ⫽ ␴2 we then have: ␴2 2 E0(Wt) ⫽ (bm ⫹ b2␰2) u 苶 ⫹ ᎏᎏ (1 ⫹ b␰2)2

冤

冥

(3.2)

which the government minimizes with respect to b. The ﬁrst order condition for a minimum is: (b ⫺ bm) 2 2 ⫽0 bu 苶 ⫹ ␴ ᎏᎏ (1 ⫹ b␰2)3

(3.3)

Since the ﬁrst term on the lhs of (3.3) is positive, the second term must be negative; i.e. b ⬍ bm and the optimally chosen banker is conservative.5 It is also apparent that if ␴ ⬎ 0 then b ⬎ 0 so that the optimal conservative banker does not completely eliminate the inﬂationary bias but, instead, achieves a compromise of a lower inﬂation rate than the representative banker and the retention of some degree of stabilization policy.6 We summarize these results as: Proposition 1 Under Rogoff delegation the optimal degree of conservatism b satisﬁes 0 ⬍ b ⬍ bm where bm is the weight on employment variability chosen by the government and median voter. Thus delegation achieves a compromise of a lower inﬂation rate than the representative banker and the retention of some degree of stabilization policy. There are a number of problems with Rogoff’s solution: at a practical level the government has to find a central banker or choose a degree of independence which results in exactly the right weight b. More fundamentally the solution assumes that commitment to the type of banker or degree of independence is possible whereas commitment to a monetary rule is not. The public must be reassured that once their expectations of inflation are formed, the government will not sack or overrule the banker and appoint a less conservative one. The third problem is that when open-economy aspects and fiscal interactions are introduced the delegation equilibrium may turn out to exhibit signific-

Theoretical perspectives on CBI 15 ant Nash inefficiencies. We now examine this last problem in some detail. Rogoff delegation in the open economy In open economies monetary spill-overs exist (i.e. a monetary expansion in one country affects employment and output in interrelated economies with which it trades). To model these spill-overs, consider a MundellFleming model in which n ⫹ 1 interdependent economies each specialize in the production of a distinctive composite good which are imperfect substitutes in consumption. Capital stock and investment are exogenously ﬁxed as before. All economies have identical economic structures, and are subject to a stochastic supply shock drawn from the same distribution. Label the countries with subscripts 0, 1, . . . , n. To ease the notation we drop the time subscript and drop the subscript for country 0. For country i let ␲i be CPI inﬂation and: yi ⫽ log(Yi/Y 苶i) ⬇ (Yi ⫺ Y 苶i)/Y 苶i be the proportional change in output relative to 苶 Yi, the steady-state natural rate path. Denote CPI inﬂation surprise by: ␲˜ i ⫽ ␲i ⫺ ␨(␲i) Then the reduced form of the model employed in the next two sub-sections is given by: n

1⫺␾ l ⫽ ␾(␰␲˜ ⫺ ⑀) ⫹ ᎏᎏ 冱 [␰␲˜ i ⫺ ⑀i] n i⫽1

(3.4)

For country 0, where employment is henceforth measured relative to the natural rate, 0 ⱕ ␾ ⱕ 1 is a parameter such that 1 ⫺ ␾ expresses the degree of openness of the economies and ␰ measures the impact of a domestic inﬂation surprise on domestic output. Levine and Pearlman (1997) derive (3.4) from micro-foundations,7 but a more traditional formulation would lead to the same basic structure. The advantage of our treatment is that the parameters ␾ and ␰ may be expressed in terms of the underlying fundamental parameters. From Levine and Pearlman (1997) we obtain: [␣ ⫹ ␥2(1 ⫺ ␤)/␤] ␾ ⫽ ᎏᎏᎏ ; ␰ ⫽ (1 ⫺ k)/␤ [␣ ⫹ (n ⫹ 1)␥2(1 ⫺ ␤)/␤]

(3.5)

where ␣ ⫽ (C 苶/Y 苶)(1 ⫺ G 苶/Y 苶), and 苶 G/Y 苶 are consumption and government spending expressed as proportions of baseline output respectively, ␤ is capital’s share of output, ␥2 is a consumer preference parameter for

16

Ali al-Nowaihi and Paul Levine

imported private (non-government) goods such that n␥2 is the share of these goods in total consumption and 0 ⱕ ␬ ⱕ 1 is the degree of indexing of nominal wage contracts. Thus if ␬ ⫽ 1 surprise inﬂation has no impact and there is no credibility problem, nor any role for stabilization policy. The mechanism through which the inter-country spill-over effects in (3.4) occur is through surprise real exchange effects. A relative inﬂation surprise in country 0 causes a surprise real depreciation for country 0 relative to i. For country i this is real appreciation which in turn lowers the cost of imported goods and causes its real product wage to fall. Thus a unilateral monetary expansion in the domestic country boosts foreign employment and output. This policy spill-over plays a central role in the monetary policy game that follows. In general we could investigate a number of important asymmetries between countries: differences in government priorities (reﬂected in differences in bm), different effects of surprise inﬂation on output (differences in ␰), differences in the NAIRU 苶 u and different shocks. We shall concentrate on the latter and consider the simplest form of asymmetry: random shocks ⑀i for the country i, with zero means and equal variances, but not necessarily highly correlated between countries. Turning to the open-economy delegation game, suppose that each monetary authority adopts a single-period loss function of the form: Wi ⫽ ␲i2 ⫹ bi(li ⫺ n)2

(3.6)

The sequence of events now becomes: 1

2 3

The degree of independence of the n ⫹ 1 CBs is chosen by the n ⫹ 1 governments determining the weight b, b1, b2, . . . , bn for subsequent periods t ⫽ 0, 1, 2, . . . There are two possible delegation ‘regimes’ to consider: non-co-operative delegation (NCD) in which the degree of independence is chosen independently by governments, and co-operative delegation (CD) where the choice is undertaken in a co-operative manner. In what follows we compare these regimes with a baseline non-delegation regime of government-dependent representative bankers (R) in which b ⫽ b1 ⫽ b2 ⫽ . . . ⫽ bn ⫽ bm. In period t ⫺ 1 the private sector in all countries forms an expectation ␨t⫺1(␲t); t ⫽ 1, 2, . . . In period t the CBs observe shocks ⑀, ⑀1, ⑀2, . . . ⑀n and independently and simultaneously choose the inﬂation rates ␲, ␲1, ␲2, . . . ␲n.

We solve for the equilibrium as before by backward induction. As for the closed economy case the inﬂation policies in the resulting equilibrium consist of an average (or deterministic) component and an additional unanticipated stabilization component which adjusts inﬂation in response to the current shock. Consider the former ﬁrst. Figure 1.2 shows the opti-

Theoretical perspectives on CBI 17

Figure 1.2 The inﬂationary bias in the open economy: co-operation and non-cooperation.

mization exercise facing each country for a particular degree of conservatism (i.e. in the game from event 2 onward). If countries choose inﬂation rates independently they take the other country’s decision as given. From (3.4) the slope of the SRPC is then 1/␰␾ which is more than 1/␰ as in the closed-economy case. Following the same reasoning as before, this results in an inﬂation rate PD in the absence of pre-commitment which is equal to bm␰␾u 苶 with representative bankers. If, however, countries co-operate in the choice of inﬂation rates then the slope of the SRPC is 1/␰, as for the closed economy, and we end up with a higher inﬂation rate PD. Thus in the absence of pre-commitment we have an example of counterproductive co-operation ﬁrst pointed out by Rogoff (1985b). Without co-operation, each country is inhibited in engaging in surprise inﬂation by the resulting real exchange rate depreciation. With co-operation this discipline disappears and the inﬂation bias in equilibrium rises. Now consider the unanticipated stabilization component of policy. For both the pre-commitment and non-pre-commitment case the policymakers can plan their stabilization policy in terms of a feedback rule such as (2.17) above which responds to shocks, but is contingent on their actual

18

Ali al-Nowaihi and Paul Levine

size and sign. A convenient property of the optimization problem is that the evaluation of the anticipated and unanticipated components now decomposes into independent calculations. We can therefore consider stabilization policy separately. Figure 1.3 demonstrates the outcome for the case of two countries (n ⫽ 1) using the famous Hamada diagram (Hamada, 1976) for the case of equal perfectly correlated shocks ⑀ ⫽ ⑀1. The utility curves are now expressed in terms of the stochastic adjustments to expected policy, ␲˜ and ␲˜ 1 for the two countries. The bliss point B for country one is where it makes no adjustment (␲˜ ⫽ 0) but country two obligingly inﬂates sufﬁciently to keep its output at its natural rate. B* is an analogous bliss point for country two. The utility curves are ellipses as before with conservative bankers represented by steeper curves (see Figure 1.3b). In the nonco-operative case each country takes the other country’s inﬂation rate as given which maps out reaction functions BN and B*N. The nonco-operative equilibrium is at the intersection N of these curves. Paretoefﬁcient points lie along the contract curve BCB*. If the monetary authorities were to co-operate over the joint choice of inﬂation rates they would bargain over outcomes along the contract curve Pareto-superior to N, the shaded region. Symmetry consideration suggests an outcome C which is the social optimum. Comparing C with N we see that there is a deﬂationary bias in the conduct of monetary stabilization policy in the face of a negative supply-side shock. Similarly inﬂation will be reduced by too little if positive supply-side shocks come alone. Compared with C, the nonco-operative stabilization rule is too muted because each country ignores the positive externality enjoyed by the other. What now happens if a more conservative banker is appointed? The sub-optimal character of N will worsen, as will be seen from Figure 1.3b. This applies only to stabilization policy and average inﬂation will be lower, but the overall effect of such an appointment is now ambiguous. More formally, suppose two bankers with output weight b in the welfare function are appointed. Then the equilibrium rule corresponding to (2.17) can be shown to be given by: b␰␾ ␲˜ ⫽ ␲˜ 1 ⫽ b␰␾u ⑀ ⫽ ␲˜ ⫹ ␲˜ 苶⫹ ᎏ 1 ⫹ b␰2␾

(3.7)

In the absence of policies that eliminate the underlying labour market distortion, the best outcome is reached if policy-makers can pre-commit both with respect to their promises to the public sector and to each other. The optimal rule is then bm␰ ␲ ⫽ ␲1 ⫽ ᎏ2 ⑀ 1 ⫹ bm␰

(3.8)

Theoretical perspectives on CBI 19

(a) Representative bankers

(b) Conservative bankers

Figure 1.3 Stabilization policy with perfectly correlated shocks.

which is the same as (2.11) for the closed economy with b ⫽ bm. Delegating monetary policy to conservative bankers lowers average inﬂation, but at the expense of worsening the efﬁcacy of stabilization policy. If the variance of the common shock is high compared with the output target, then delegation could actually result in a worse outcome in terms of the

20

Ali al-Nowaihi and Paul Levine

representative welfare measure than sticking with representative nonindependent bankers. However, there is a way of avoiding this outcome and that is for the two countries to co-ordinate over their choice of bankers, but thereafter allow them independence from government and from each other in the actual conduct of monetary policy. In regime CD at stage one of the game, an optimal choice of the weight bi ⫽ b can then be carefully selected, in effect choosing between different non-co-operative equilibria N in Figure 1.3b so as to achieve the best possible trade-off between low long-run inﬂation and effective stabilization policy. This will result in an inﬂation outcome somewhere along OC. Now consider uncorrelated shocks and, in particular, the other extreme of perfectly asymmetric shocks or ⑀ ⫽ ⫺⑀1. Figure 1.4 shows this case. The reasoning leading to the non-co-operative and co-operative equilibria, N and C respectively, is as before. C in Figure 1.4b is the social optimum, and the important difference now is that conservative bankers take the nonco-operative equilibrium closer to this outcome. Delegation of policy to a conservative banker now improves the efﬁcacy of stabilization policy. There are still gains from co-ordination in the choice of banker which will result in an outcome along OC in Figure 1.4b. Comparing this with the case of equal shocks we can see that the degree of conservatism should be much higher if perfectly anti-symmetric shocks are expected as opposed to equal (i.e. perfectly symmetric) shocks. Now suppose that the number of countries rises to ﬁve. Then the correlation coefﬁcient satisﬁes ⫺1/4 ⱕ ␳ ⱕ 1.8 Figure 1.5 shows the optimal choice of conservative banker (i.e. the choice of conservatism deﬁned as bm/b) for the three possible regimes we have discussed at event 1 of the game: non-cooperative delegation (NCD), co-operative delegation (CD) and representative bankers (R). In addition, we show results for a further possibility: a disequilibrium in which one country delegates unilaterally to an optimally chosen banker while the other country stays with a representative CB (regime UNIL). For our numerical results we choose the following calibration: 苶 u ⫽ 5 per cent, var(ut) ⫽ 3 per cent, where ut is the original supply shock in (2.2), bm ⫽ 1, and ␾ ⫽ 0.6. This results in an inﬂation rate of 5 per cent in the regime R. As we expect from the previous discussion, conservatism under CD decreases with ␳. In addition, when shocks are sufﬁciently correlated (␳ ⱖ ⫺0.9, in fact) NCD results in over-conservative bankers compared with the best delegation regime CD. Unilateral delegation results in even more conservative bankers for that country. Figure 1.6 shows the welfare losses for NCD, CD, R and UNIL relative to the social optimum which is achieved when all countries both cooperate and pre-commit. The welfare loss is expressed in terms of an equivalent per cent increase in employment. Points to note are: unilateral delegation (‘conservatism in one country’) works very well for that country, but when both countries delegate the gains are substantially less. There are small welfare gains from co-operative rather than

(a) Representative bankers

(b) Conservatative bankers

Figure 1.4 Stabilization policy with perfectly uncorrelated shocks.

22

Ali al-Nowaihi and Paul Levine

Figure 1.5 The choice of conservatism b/bm for the delegation regimes: ﬁve countries. Notes NCD, non-co-operative delegation; CD, co-operative delegation; UNIL, unilateral delegation.

non-co-operative delegation. For sufﬁciently symmetric shocks Figure 1.6 shows the result that non-co-operative delegation is counterproductive; that is, NCD is worse than R in welfare terms. This result becomes more pronounced as the number of countries increases further and/or the variance of the supply shocks increases (see Currie et al. (1994) for further details). Levine and Pearlman (1997) report more bad news for the Rogoff delegation game in open economies. When interactions between ﬁscal authorities engaging in stabilization and independent central banks are included, the inefﬁciency of regime NCD worsens signiﬁcantly to the extent that NCD is worse than R by as much as 5 per cent employment loss if shocks are perfectly correlated. To understand the intuition behind all these results recall that, for given inﬂation rates, ﬁscal expansion has real effects by appreciating the real exchange rate. Then for negative (positive) symmetric supply-side shocks ﬁscal expansion (contraction) implies a negative externality, an imbalance in the policy mix towards ﬁscal policy and resulting ‘beggar-thy-neighbour’ effects. The appointment of conservative bankers worsens this bias. It makes very good sense for one country alone to delegate monetary policy to a very conservative banker. In the face of a common negative shock the country can free-ride on other countries’ monetary expansion, engage in a

Theoretical perspectives on CBI 23

Figure 1.6 The welfare outcomes relative to social optimum: ﬁve countries. Notes R, representative bankers; welfare expressed as percentage equivalent employment increase.

beggar-thy-neighbour ﬁscal expansion that appreciates its exchange rate and, at the same time, enjoy very low inﬂation. In a world of many countries, when they attempt to act likewise the resulting non-co-operative equilibrium, NCD, is considerably worse than the original game with representative bankers, R, for all except extremely anti-symmetric shocks. This suggests that if shocks are sufﬁciently correlated, it may be better not to play the delegation game in the ﬁrst place. Delegation with political uncertainty We now introduce political parties representing different constituencies into the model along the lines of Alesina and Gatti (1995, AG). AG consider only the closed economy. The following is a synthesis of AG and the open-economy Rogoff delegation game of the previous sub-section. To keep the analysis manageable we focus solely on a common (i.e. perfectly symmetric) supply shock for all countries. In the analysis of this sub-section, country 0 is assumed to have an independent CB with a degree of conservatism b0 ⫽ b. The remaining n CBs are not independent and have preferences described by bi ⫽ bL if the ‘Left’ are in power and bi ⫽ bR if the ‘Right’ are in ofﬁce where b ⱕ bR ⱕ bL. Thus the Left are more (less) unemployment (inﬂation) averse than the

24

Ali al-Nowaihi and Paul Levine

Right. In each period there is an exogenous probability p of a Right government being elected and a probability (1 ⫺ p) of a Left government being elected. For large n these are also the approximate proportions of Right and Left governments. The sequencing of events is as follows: •

• • • •

Event 1. In the initial period t ⫽ 0, the CB in country 0 is appointed with a preference b0 ⫽ b. Government 0 is committed to this choice in all future periods. Event 2. In each country nominal wages are set based on expected inﬂation in the following period. Event 3. The governments determining monetary policy in countries 1 to n are elected. Event 4. The common supply shock occurs in each economy. Event 5. The CBs independently and simultaneously set inﬂation in response to observations of the shock.

We now solve the game from event 5 back to event 2; that is, treating the choice of conservatism at event 1 as exogenous. As before we decompose the inﬂation rates into two parts: an inﬂation bias ␲ 苶i which depends only on the type of monetary authority, and a term ␲˜ i which, in addition, is shock-contingent. Consider ﬁrst the equilibrium employment levels and inﬂation biases. For country 0 with an independent CB with degree of conservatism b, as in (3.7), we have: 苶l ⫽ 0; 苶 ␲ ⫽ b␾␰u 苶

(3.9)

Thus in the absence of an exogenous shock, for country 0 with an independent CB there is no political uncertainty. Employment always remains at its natural rate and the inﬂation bias is constant, neither being affected by the political uncertainty in the rest of the world. For countries 1 to n with government-dependent CBs monetary policy depends on the political complexity of the government, thus introducing political uncertainty. Solving the game as before we arrive at:9 2 2 ␲ 苶R ⫽ bR␾␰u 苶[1 ⫹ ␪bL␰ ]/⌬; ␲ 苶L ⫽ bL␾␰u 苶[1 ⫹ ␪bL␰ ]/⌬

(3.10)

2 l苶R ⫽ ⫺(1 ⫺ p)␪(bL ⫺ bR)␰2u 苶/⌬; l苶L ⫽ p␪(bL ⫺ bR)␰ u 苶/⌬

(3.11)

where ␪ ⫽ [(n ⫹ 1)␾ ⫺ 1]␾/n and ⌬ ⫽ 1 ⫹ ␪[(1 ⫺ p)bR ⫹ pbL]␰2. We can show that ␪ ⬎ 0. It is then straightforward to show that corresponding to bL ⬎ bR ⱖ b we must have 苶 ␲L ⬎ ␲R ⱖ ␲ 苶 and that 苶lR ⱕ 0 and l苶L ⱖ 0; that is the election of Left (Right) government results in employment rising above (below) its natural rate. For a single closed economy we put ␾ ⫽ 1 and all the above results reduce to those in the closed economy partisan political

Theoretical perspectives on CBI 25 business cycle literature. If elections are predictable at the time wages are set, then if p ⫽ 1, ␲R ⫽ bR␾␰u 苶 as obtained in (3.7) with b ⫽ bR with employment at its natural rate. Similarly if p ⫽ 0, and employment, again, is at its natural rate. The main results up to this point are summarized as: Proposition 2 Assuming weights bL ⬎ bR ⱖ b for a Right government-dependent CB, a Left government-dependent CB and an independent CB respectively, the corresponding inﬂation biases and employment, in the absence of shocks, satisfy 苶 ␲L ⬎ ␲R ⱖ 苶 ␲, l苶 ⫽ 0 and l苶 ⱖ 0. So far so good, but as yet we have produced no real insights over those of the partisan political business cycle literature for the closed economy. These are provided by examining how the open economy affects the results. Consider the source of the gain (G) from an independent CB, apart from a lower inﬂation bias, which is the elimination of the employment variance caused only by political uncertainty. G is given by G ⫽ p ⫺ ␨(li)]2 ⫹ (1 ⫺ p)[l苶L ⫺ ␨(li)]2 ⫽ p(1 ⫺ p)(l苶L ⫺ l苶R)2 ⫽ 2 2 p(1 ⫺ p)␰4␪2(bL ⫺ bR)2u 苶 /⌬

(3.12)

using the results (3.11) for l苶L and l苶R. How is the gain G from eliminating political uncertainty affected by the degree of global integration? The answer is provided by the following proposition: Proposition 3 Global integration reduces the output variance caused solely by political uncertainty. The intuition behind this result is that openness creates a disincentive to inﬂate because the boost to employment caused by surprise inﬂation is partially offset by a real depreciation. As a result the increase (decrease) in employment as a result of the election of a Left (Right) government falls and with it the employment variance G, given by (3.12), induced by political uncertainty. Now let us turn to the shock-contingent components of inﬂation rates and employment. Solving for the equilibrium we arrive at the stabilization rules ␲˜ ⫽ ⫺b␾␰(1 ⫹ ␪bL␰2)⑀/⌽; ␲˜ L ⫽ ⫺bL␾␰l˜L ⫽ bL␾␰(1 ⫹ ␪b␰2)⑀/⌽; ␲˜ R ⫽ ⫺bR␾␰l˜R ⫽ bR␾␰(1 ⫹ ␪b␰2)⌰⑀/⌽;

(3.13)

where ⌽ ⫽ (1 ⫹ b␰2␾2)(1 ⫹ ␪bL␰2) ⫹ (1 ⫹ ␪b␰2)(pbR⌰ ⫹ (1 ⫺ p)bL)␰2␾(1 ⫺ ␾) and, as before, ␪ ⫽ [(n ⫹ 1)␾ ⫺ 1]␾/n and ⌰ ⫽ (1 ⫹ ␪bL␰2)/(1 ⫹ ␪bR␰2) ⬎ 1. From (3.13) we can see the stabilization spill-over on employment in

26

Ali al-Nowaihi and Paul Levine

country 0 from the dependent CBs with more active monetary policy. Given this spill-over, the question now is whether country 0 can gain both in terms of a lower inﬂation bias and lower employment variance; that is, can CBI in country 0 alone lead to gain without pain? There are two possible ways of measuring the gain in terms of lower output variance from having an independent CB. The ﬁrst, which we examine in this section, is to compare the employment variance from having a single independent CB with that of the remaining n countries with government-dependent CBs. This is not strictly the gain, but it does indicate observed differences which feature in the empirical literature cited above. The actual gain, examined in the next sub-section, is the difference between the employment variance of a country before and after it unilaterally makes its own CB government-independent. This is an indication of the incentive for an individual country to make this institutional change. Gain without pain occurs in the ﬁrst sense if var(l) ⱕ var(li); i ⫽ 1, . . . , n. Since the two forms of uncertainty are independent, the employment variance for countries with government-dependent CBs decomposes into a political induced part G obtained above, and a shock-induced component given by p␨(l˜R)2 ⫹ (1 ⫺ p)␨(l˜L)2. Since var(l˜) ⫽ ␨(l˜)2 the condition for gain without pain becomes G ⬎ L where L is the stabilization loss given by L ⫽ [(1 ⫹ ␪bL␰2)2 ⫺ (1 ⫹ ␪b␰2)2(p⌰2 ⫹ 1 ⫺ p)]␴2⑀/⌽2

(3.14)

Since b ⱕ bR it follows that ⌰ ⫽ (1 ⫹ ␪bL␰2)/(1 ⫹ ␪bR␰2) ⱕ (1 ⫹ ␪bL␰2)/ (1 ⫹ ␪b␰2). Hence since bL ⬎ b we must have that L ⬎ 0. Thus we have: in the open economy L ⬎ 0; i.e. there is always some stabilization loss for the single independent CB. Thus the free-rider effect can never be sufﬁcient to totally eliminate the stabilization loss caused by a single independent CB. Unlike the political gain G we cannot sign L unambiguously as integration increases. This is because there are two opposite effects operating. As integration increases, the spill-over effect of a given monetary expansion (for a common negative supply-side shock) from the n government-dependent CBs on the single independent CB increases. However, integration also reduces the size of their expansion because they will be deterred by an increased exchange rate depreciation. We can see this ambiguity in some numerical results shown in Figures 1.7 and 1.8. Figure 1.7 plots G and L against the proportion n␥2 of imported goods entering the consumption basket which we refer to as the degree of integration. As economies move from autarky to complete integration, where there is no bias towards the domestic good, then this portion increases through the range n␥2⑀[0, n/(n ⫹ 1)]. As before, we examine a ﬁve-country trading bloc closed to the outside world (i.e. n ⫽ 4) and use the same calibration. The remaining parameters are b and bR. We set bR ⫽ bL/2 and examine two cases: b ⫽ bR and b ⫽ bR/2 corre-

Theoretical perspectives on CBI 27

b ⫽ bR ⫽bL/2

b ⫽ bR /2 ⫽ bL /4

Figure 1.7 The lower employment variance gain and loss for a single independent CB relative to the n others.

sponding to conservative and very conservative independent central bankers. With our calibration in the case of autarky at ␥2 ⫽ 0 (corresponding to AG) the gain G from independent CBs in terms of less politically induced employment variance by far outweighs the stabilization loss L. As economies begin to trade G falls as required by proposition 3. The

28

Ali al-Nowaihi and Paul Levine

stabilization loss remains quite ﬂat for the less conservative bankers, but falls if bankers are very conservative. In the former case we still see gain without pain at high levels of integration; in the latter case CB independence results in higher employment variance as economies begin to import around half their consumption needs. The main message from these results is that open economy effects can drastically alter the balance between the gain from a reduction in political uncertainty and the stabilization loss resulting from the appointment of an independent central banker in one country. We now address the question of what happens when all countries adopt independent CBs. We now measure the gains or otherwise relative to the equilibrium where all CBs are government-dependent. If we put bR ⫽ bL ⫽ b in the results of the previous sub-section, we in effect eliminate political uncertainty and arrive at the equilibrium with n ⫹ 1 independent CBs. Using the previous results we ﬁnd that 2 l˜ ⫽ 0; ␲˜ ⫽ b␾u 苶; ␲˜ ⫽ ⫺b␾␰l˜ ⫽ b␾␰⑀/(1 ⫹ b␰ ␾)

(3.15)

corresponding to (3.7). We can now show that var(l˜) increases as we proceed from one independent CB to n ⫹ 1 independent CBs; i.e. we have the proposition: Proposition 4 The stabilization loss associated with central bank independence increases as we proceed from one to n ⫹ 1 independent CBs. Finally let us compare the employment output variance for the two scenarios of proposition 3 with that for n ⫹ 1 government-dependent CBs. The stabilization loss from central bank independence is now measured as: L ⫽ var(l˜) ⫺ p var(l˜R) ⫺ (1 ⫺ p)var(l˜L)

(3.16)

where L ⫽ L (one) for the case of one independent CB when l˜ is given by (3.13) and L ⫽ L (all) for the case of n ⫹ 1 independent CBs in which case l˜ is given by (3.15). Figure 1.8 shows G (which is unchanged), L (one) and L (all) plotted against n␥2. From proposition 4 we know that L (all) ⬎ L (one) and this is conﬁrmed by our numerical results. Our results indicate that ‘gain without pain’ when one country unilaterally appoints an independent CB can give way to a signiﬁcant employment volatility loss when all countries join the Rogoff delegation game. The full delegation equilibrium for our identical economies requires that all countries do delegate and choose identical degrees of conservatism b at event 1 of the game as in the previous open economy Rogoff delegation game. A more comprehensive examination of the full delegation equilibrium with political uncertainty would clearly be of interest, but would take us beyond the scope of this survey.

Theoretical perspectives on CBI 29

b ⫽ bR ⫽bL /2

b ⫽ bR /2 ⫽ bL /4

Figure 1.8 The lower employment variance gain, G, and loss for one independent CB, L (one), and n ⫹ 1 independent CBs, (all), relative to n ⫹ 1 government-dependent CBs.

To conclude this sub-section, we ﬁnd that open economy effects alter signiﬁcantly the results of AG. While the positive conclusions are the same (the model is still consistent with the empirical ﬁndings of gain without pain) the normative conclusions to be drawn are quite different. Our results indicate that as more CBs become independent and economies

30

Ali al-Nowaihi and Paul Levine

become more integrated, then Rogoff’s trade-off between low inﬂation and high output variance re-emerges. The policy implications of these results are: either complement the ‘single market’ with more EMU-type arrangements in which case we move closer to the closed economy results of AG; or search for alternatives to Rogoff delegation that will enforce the ﬁrst-best (or near-ﬁrst-best) monetary policy rules. This takes us to Section 4.

4 First-best solutions to the credibility problem Trigger strategies Barro and Gordon (1983) conﬁne themselves to the monetary policy game in the absence of shocks. In a repeated game, a zero inﬂation equilibrium can be sustained by the following trigger strategy on the part of the public: play ␲e ⫽ 0 in the ﬁrst period and thereafter if the history of play is ␲ ⫽ ␲e⫽0. Otherwise play ␲te ⫽ ␰bu 苶, the inﬂationary bias for some length of time. Then, provided that the punishment period is sufﬁciently long and the discount rate sufﬁciently low, the CB will not have an incentive to depart from zero inﬂation. The problems with this solution are widely acknowledged: what determines the length of the punishment period during which the private sector expects high inﬂation, and how can an atomistic private sector co-ordinate on one particular strategy? Another problem is that even if the private sector could co-ordinate, then the punishment is inefﬁcient and therefore not renegotiation-proof.10 This undermines the credibility of the trigger strategy even though it is subgame perfect (al-Nowaihi and Levine, 1994). Walsh contracts A recent new approach to the credibility problem pioneered by Walsh (1995)11 proposes a contract which links the income of the bank to economic performance in such a way that the bank will follow the optimal pre-commitment rule even when carrying out period-by-period optimization. What is required here is an incentive mechanism that will eliminate the inﬂation bias in the discretionary stabilization rule (2.18). Having posed the problem in this way, the solution is in fact remarkably simple. Consider the welfare loss (2.8) and let us convert this into monetary units by multiplying by a factor ␣. Let the income transfer from the government to the CB be given by Tt. Then the CB will maximize a monetary value: Ut ⫽ Tt ⫺ ␣[b(lt ⫺ n) ⫹ ␲t2]

(4.1)

Suppose that the income transfer takes the form: Tt ⫽ F ⫺ 2␣b␰u 苶␲t ⫽ F ⫺ 2␣␲*␲t

(4.2)

Theoretical perspectives on CBI 31 where we have written the inﬂationary bias under discretion (without any contract) as ␲* ⫽ b␰u 苶. (4.2) says that the CB is penalized for each per cent point of inﬂation by a reduction in its income by an amount 2␣␲*. The constant F can be chosen to be sufﬁciently large so as to allow the CB to meet its ﬁnancial commitments even if inﬂation turns out to be very large. Alternatively the transfer can be interpreted as a remuneration package for the CB executive, in which case F must be large enough to ensure an adequate supply of bankers. It is now straightforward to show that the maximization of (4.1) with the income transfer given by (4.2) subject to the EAPC (2.7) will eliminate the inﬂationary bias and result in the optimal rule. In terms of agency theory the principal (the government) has designed an incentive mechanism such that the agent (the CB) will pursue the policy she desires. Does the Walsh contract then solve the time-inconsistency problem? What the contract certainly achieves is the relocation of the problem as one of enforcing the contract (McCallum, 1995). The problem still remains that circumstances will arise when it will be in the interests of all concerned to renegotiate the contract rather than enforce it. This feature destroys the credibility of the Walsh contract in the absence of some other enforcement mechanism. To show the incentive to renegotiate, suppose that a positive supply shock occurs (⑀t ⬍ 0). Then the Walsh contract speciﬁes a deﬂationary response (␲t ⬍ 0 from (4.2)). However, all parties involved in the game (the government, the CB and the private sector) would be better off by adopting a neutral monetary stance (␲t ⫽ 0). The government values employment above the equilibrium rate and prefers zero inﬂation to deﬂation. The private sector prefers zero inﬂation to a deﬂation surprise. Thus both the government and private sector will have every incentive to renegotiate the Walsh contract and allow a neutral response, (␲t ⫽ 0, to a positive shock. Finally the CB can be induced to accept renegotiation by a suitable monetary reward which the government can ﬁnance from allowing output to rise by more than it would under the contract. The availability of an option that is better than the proposed equilibrium for everyone means that the latter equilibrium is not ‘renegotiation-proof’ in the terminology of agency theory and the credibility problem remains. The practical problem of ‘non-renegotiation-proofness’ is that it is difﬁcult to see how courts can uphold a contract if all parties wish to ignore it. In the legal literature the term ‘efﬁcient breach’ applies to this situation; that is, there may be circumstances where breach of contract is more efﬁcient than performance (Ulen, 1984). On the other hand, if one party refuses to sign or, in the case of the public, tolerate a new contract then the courts will uphold the original contract. Can we then overcome the problem of renegotiation by ensuring that at least one party always prefers the original contract? The following sub-section addresses this question by considering a reputational solution to the problem.

32

Ali al-Nowaihi and Paul Levine

Walsh contracts in a reputational equilibrium The great strength of the Walsh contract is not apparent in the present setup. Suppose we relax the assumption that inﬂation targets can be achieved exactly in each period, writing: ␲t ⫽ ␲tp ⫹ vt

(4.3)

where ␲tp is the planned inﬂation target for the period and vt is a random control zero-mean niid error which is only known ex post to the CB. In addition, assume that the CB has private information of the supply shock and the public and government only observe actual inﬂation and employment ex post (i.e. at the end of the period). From the EAPC all three parties can therefore evaluate the size of the supply shock ex post, but the public and government will never be able to disentangle the planned and random components of inﬂation in (4.3). It is straightforward to show that the ex ante optimal rule is of the same form as before with the planned inﬂation rate ␲tp replacing the actual inﬂation rate ␲t as the monetary instrument, and the Walsh contract with the transfer given by (4.2) will provide the necessary incentive for the CB to pursue this optimal rule. Note that the transfer is deﬁned in terms of the observed actual inﬂation rate and can therefore be implemented by the government despite the fact that the CB has private information of the shocks ⑀t and vt. In other words, the Walsh contract successfully solves the problem of asymmetric information. Although the Walsh contract solves the principal–agent problem of private information in the hands of the CB, the credibility problem in the shape of an incentive by the CB to renege on low inﬂation has not been solved, but has been relocated as a renegotiation problem. However, this is an essential step in the resolution of the problem. Whereas it is impossible for the public to monitor the rule owing to the random control error which they never observe,12 renegotiation can be made very visible. For example, this can be achieved by the publication of the minutes of meetings between the government and CB, as in the case of the UK. By making the conduct of monetary policy open and accountable we now outline a study which shows that the Walsh contract can create an incentive for a government to establish a reputation for not renegotiating the contract, hence sustaining the ex ante optimal rule (al-Nowaihi and Levine, 1996). This solution rests on the assumption that there in fact exist some governments who value honesty for its own sake (i.e. honesty enters into their utility function) and who will therefore never renegotiate, the remaining ‘normal’ governments will ﬁnd it optimal to mimic the ‘good’ governments. In the resulting pooling equilibrium Walsh contracts are not renegotiated and the credibility problem is solved.13 In al-Nowaihi and Levine a reputational equilibrium is constructed in

Theoretical perspectives on CBI 33 which the ‘normal’ government, say of type ␶ ⫽ 1, uses a prior reputation for not reneging and being of the honest type, say type ␶ ⫽ 0, to successfully implement its ex ante optimal rule. The key feature is that reneging should be visible. We consider here only the case of an inﬁnite time horizon for the board of the CB. This would apply if the reputation of one board could be passed on to its successor; i.e. the types of successive boards are perfectly correlated. In the last period of ofﬁce for a CB of type ␶ ⫽ 1 reneging would then not be inevitable and each CB would behave as if it were inﬁnitely lived.14 The following proposition describes perfect Bayesian equilibria (PBE) in the sense of Fudenberg and Tirole (1991a, b). We show that at each information set the strategy of each player is a best response to the strategies of the other players’ given beliefs, and that beliefs are updated using Bayes’ rule given strategies. Proposition 5 Let p be the prior expectation of the private sector that the government is of type ␶ ⫽ 0. Let ␦ be the government discount factor. Consider an inﬁnite time horizon and suppose that ␦ ⱖ 1/(2 ⫹ b). Then for any p ⬎ 0, there exists a PBE where the government of type ␶ ⫽ 1 never reneges and hence the ex ante optimal rule is implemented in each period. Proof See al-Nowaihi and Levine (1996). For the inﬁnite time horizon, since 1/(2 ⫹ b) ⬍ 1/2 it follows from the proposition that any non-zero prior p and ␦ ⱖ 1/2 are sufﬁcient to implement the optimal rule. Our unit of time is the length of a wage contract. If this is taken to be one year the latter condition on the discount factor ␦ is very weak. The ex ante optimal rule then becomes very easy to sustain if CBs act as if they are inﬁnitely lived. To summarize: the above result suggests that the optimal stabilization policy can be sustained in an inﬁnite horizon model under weak parameter restrictions (for our model, ␦ ⱖ 1/2, p ⬎ 0). As one would expect, the parameter restrictions become more stringent in the ﬁnite horizon model, but not excessively so. Moreover there are good reasons for regarding the latter results as overly pessimistic because they ignore the possibility that types of governments may be correlated, and that the loss of credibility might spill over into other areas such as ﬁscal policy and international relations.

34

Ali al-Nowaihi and Paul Levine

5 Non-benevolent governments and the political monetary cycle Previous sections have assumed that governments are benevolent in the sense that they maximize either a social welfare function for society as a whole or, in the case of the partisan political business cycle model, governments are led by different political parties each of which is representative of a particular section of society. We now assume that governments value power for its own sake and derive utility from being in power, as well as from raising social welfare. We also introduce government competence. These two features lead to rational political monetary cycle models in which economic ﬂuctuations result from a government’s attempt to signal its competence to the electorate. This section addresses an important question which arises out of the rational political business cycle literature: can political monetary cycles (PMC), in which policy-makers engage in a pre-election monetary expansion, be avoided? In a number of recent papers (Rogoff and Sibert, 1988; Rogoff, 1990; Persson and Tabellini, 1990) a surprisingly pessimistic view is put forward that it may not be desirable to even attempt to curtail the cycle because it provides an important signalling device that conveys information to the electorate regarding the incumbent government’s competence. According to this view it could well be that suppressing the cycle would be counterproductive, resulting in the more frequent election of incompetent governments, or the resort to more costly forms of signalling. But are there less costly forms of signalling? We argue that this is indeed the case. The signalling device we propose is a Walsh-type contract between the government and the CB. This forms part of a monetary regime which consists of the following essential elements: 1 2

3

An instrument-independent (but goal-dependent) CB. An inﬂation target set by the government which is credible in a precise sense to be spelt out. The CB pursues the target using the usual range of monetary instruments at its disposal. The CB is accountable to the government for its performance in a way that is transparent to the public. This is achieved through a public contract between the government and CB that speciﬁes the inﬂation target and the incentive mechanism to ensure that the CB pursues that target.

Among existing CBs, the arrangement closest to our proposal is the Policy Target Agreement in New Zealand.15 The Reserve Bank Act of 1989, which took effect in February 1990, speciﬁes that inﬂation targets should be determined jointly by the Governor and the Reserve Bank and explicit in assigning responsibility for achieving the targets; in particular it is the Governor’s duty ‘to ensure that the CB carries out the functions

Theoretical perspectives on CBI 35 imposed on it by this Act’ and failure to achieve the target can lead to the removal of the Governor. There is also provision for ﬂexibility in the face of unexpected shocks. Events that would justify deviations from the 0–2 per cent target range are aggregate supply shocks spelt out as terms-oftrade price shocks from external sources, changes in indirect tax rates, a natural disaster or major fall in livestock numbers owing to disease and price-level changes resulting from government levies. Thus, in effect the inﬂation target amounts to a monetary rule which combines low average inﬂation with a stabilization function. We now set out a PMC model in which either competent incumbent administrations attempt to signal their competence by a pre-election monetary expansion (the separating equilibrium), or incompetent administrations conceal their incompetence by the same means (the pooling equilibrium). Our treatment closely follows that of Persson and Tabellini (1990, Section 5.2).The set-up is a two-period signalling model in a closed economy. First note that the EAPC (2.6) in the absence of a supply shock can be written in terms of output: yt ⫽ y苶 ⫹ ␲t ⫺ ␲te, with a suitable normalization, where ␲te ⫽ ␨t⫺1(␲t). There are two types of policy-maker, one ‘competent’ and the other ‘incompetent’ in the sense explained below. Nature chooses the incompetent type with probability p⑀(0, 1). The type of policy-maker is known to itself, but not initially by the public. However, the probability p is public knowledge. A competent policy-maker will achieve a higher natural rate of output. Thus we can write the natural rate as 1 when the competent type is in power and as y* ⫹ X 苶 for the incompetent type, where X 苶 ⬎ X (output is expressed in logarithm form). The expected natural rate is then y* ⫹ pX ⫹ (1 ⫺ p)X 苶. For convenience we redeﬁne output yt in deviation form about this expected natural rate; deﬁne x苶 ⫽ 苶 X ⫺ (pX ⫹ (1 ⫺ p)X 苶) ⫽ p(X 苶 ⫺ X) in deviation form about the expected value pX ⫹ (1 ⫺ p)X 苶 and similarly deﬁne x ⫽ X ⫺ (pX ⫹ (1 ⫺ p)X 苶) X). We can then write a modiﬁed EAPC as: ⫽ (1 ⫺ p)(X ⫺ 苶 yt ⫽ ␲t ⫺ ␲te ⫹ xt

(5.1)

where ␲t and ␲te denote inﬂation and beginning-of-contract expectations of inﬂation, xt ⫽ x苶 ⬎ 0 if the competent type is in ofﬁce and xt ⫽ x ⬍ 0 if the 16 incompetent type is in power and E(xt) ⫽ px ⫹ (1 ⫺ p)x 苶 ⫽ 0. The electorate are assumed to have the same preferences over output and inﬂation given by the two-period social welfare function: 2

2

W ⫽ 冱␦t⫺1wt ⫽ 冱␦t⫺1[ayt ⫺ ␲t2]; a ⬎ 0 t⫽1

(5.2)

t⫽1

where wt is the single-period social welfare.17 Both policy-makers are drawn from the same electorate and therefore have the same preferences over output and inﬂation. However, both types derive the same extra

36

Ali al-Nowaihi and Paul Levine

utility v ⬎ 0 from being in ofﬁce. Corresponding to (5.2) we then have the following two-period utility function for each policy-maker: 2

2

U ⫽ 冱␦t⫺1ut ⫽ 冱␦t⫺1[ayt ⫺ ␲t2 ⫹ vt]; a ⬎ 0 t⫽1

(5.3)

t⫽1

where vt ⫽ v ⬎ 0 if the policy-maker is in ofﬁce, and vt ⫽ 0 otherwise. In order for the incompetent type with xt ⫽ x to actually stand for ofﬁce we need to assume that v is sufﬁciently large to ensure that the incompetent type prefers to be in ofﬁce than in opposition; that is, v ⬎ a(x苶 ⫺ x)

(5.4)

The model is completed by assuming that wage-setters’ aim is to correctly forecast inﬂation implying a single-period utility function zt ⫽ ⫺(␲t ⫺ ␲te)2. Now consider the timing of decisions. A central feature of political business cycle models of the type represented here is the information assumption regarding the policy instrument in the period before the election. In our model the monetary instrument is taken to be the inﬂation rate which we assume is observed only after the election. The expansionary effect of inﬂation on employment and output is observed before the election. These information assumptions are consistent with a world in which employment and output adjusts before prices. New Keynesian models of imperfect competition with menu costs would provide micro-foundations for such a world (see e.g. Romer, 1996). Different set-ups which capture the same basic idea are possible. For instance, in Rogoff (1990) an incompetent government has the incentive to divert government expenditure from less visible public investment to more visible public consumption.18 We now set out formally the sequencing of events in the game between the policy-maker, the wage-setters and the voters. Throughout this subsection monetary policy is conducted by the government in contrast with Section 3 which introduces an instrument-independent monetary authority. In period 1: 1.1 Nature chooses the type of government in ofﬁce; x1 ⫽ x with probability p and x1 ⫽ 苶x with probability 1 ⫺ p. The government knows its own type but this is not revealed to the public. However, the public knows p which we assume initially satisﬁes p ⱕ 1/2. The opposition party is of the alternative type. These types remain ﬁxed for the duration of the two-period game. The public adopts the prior p1 ⫽ p. 1.2 The public forms its inﬂation expectation ␲1e for the period and wagesetters sign wage contracts based on this expectation. Contracts last for one period. 1.3 The government chooses ␲1. 1.4 The public observes output y1 but not inﬂation ␲1.

Theoretical perspectives on CBI 37 1.5 Having observed y1 the public updates its belief about the type of government from p1 ⫽ p to p2. 1.6 An election is held in which the electorate elects the incumbent government or votes for the opposition. Events 1.1 to 1.3 occur at the beginning of the period and the remaining events occur towards the end of the period. If p2 ⫽ the electorate gives the incumbent the beneﬁt of the doubt. Period 2 is similar except that there is no election towards the end and the types of political parties are as chosen in period one. Speciﬁcally we have: 2.1 The type of government is the winner of the election at stage 1.6. 2.2 The public forms its inﬂation expectation ␲2e for the period and signs wage contracts. 2.3 The government chooses ␲2. Note that it does not matter what observational assumptions are made about ␲1, y2 and ␲2 as long as ␲1 is observed after the election and ␲2 is chosen after ␲2e is formed. The following proposition (and proposition 7) describes perfect Bayesian equilibria (PBE), again the appropriate equilibrium concept for our dynamic game of imperfect information. In the proofs provided to alNowaihi and Levine (1998a), we establish the conditions for a PBE, and, we show that at each information set the strategy of each player is a best response to the strategies of the other players’ given beliefs, and that beliefs are updated using Bayes’ rule given strategies. Proposition 6 Both pooling and separating equilibria exist. •

•

In a pooling equilibrium the incompetent administration conceals its type by engaging in a pre-election monetary expansion. The competent administration sets ␲1 ⫽ ␣ and the incompetent administration sets inﬂation at the higher rate ␲1 ⫽ a ⫹ 苶x ⫺ x. The outcome is the election of the incumbent if we assume that p ⱕ . In the separating equilibrium the competent government reveals its type to the electorate by the same means that its incompetent rival would conceal its type – a monetary expansion. The competent administration now sets inﬂation at the higher rate ␲1 ⫽ a ⫹ ␪ ⫺ (x苶 ⫺ x) ⬎ a compared with that of the incompetent administration, ␲1 ⫽ ␣. The outcome now is the election of the competent type.

38

Ali al-Nowaihi and Paul Levine

Efﬁcient signalling We now turn to ways of avoiding the PMC. The most obvious institutional arrangement would be the establishment of a genuinely goal and instrument-independent CB in which the government has no inﬂuence on the conduct of monetary policy. Then for the CB vt ⫽ 0 in (5.3) and the outcome is the discretionary inﬂation rate in both periods. The type of government is revealed to the electorate and the competent government elected. If, in addition, a Rogoff conservative banker is elected who cares only about inﬂation (i.e. a ⫽ 0 in (5.3) for the CB), then the discretionary inﬂation rate is zero and both the inﬂationary bias and the distortionary effect of the PMC is avoided. We have discussed a fundamental problem with this solution in Section 3: it assumes that some commitment mechanism is in place that prevents government intervention in the delegation of monetary policy when it is rational to do so. Constitutional guarantees of political and economic independence in the form of appropriate appointment procedures, a requirement of price stability and so on, which will earn a high mark for the various measures of central bank independence discussed earlier, may not be sufﬁcient. Indeed there is considerable evidence that governments inﬂuence monetary policy even in countries where central banks enjoy a substantial degree of legal independence (see e.g. the discussion in Cukierman (1992, ch. 17) and the introduction in Persson and Tabellini (1994)). As one might expect, political authorities appear to respond to the legal independence of the CB by ﬁnding more subtle ways of inﬂuencing monetary policy. For example, in the USA where, on the basis of various indices, the CB scores highly on the degree of legal and political independence, the precise channels of inﬂuence have been extensively documented by Havrilesky (1991, 1993). It is clearly of interest to explore mechanisms for avoiding the PMC which does not assume a commitment device is in place and which accepts that incumbent administrations can exert pressure on monetary authorities. The key to the solution we propose is to make the relationship between government and CB explicit and visible. The means by which this objective can be achieved is through a Walsh-type contract in which the government makes the implementation of its chosen monetary policy by the CB incentive compatible by an appropriate incentive mechanism such as an income transfer linked to the performance of the economy or provision for dismissal of the Governor. The Walsh contract was originally proposed as a device to eliminate the inﬂationary bias of discretionary monetary policy while allowing the CB to engage in stabilization policy in response to random shocks to the economy. Again its success at achieving this depends on some commitment mechanism: in this case the mechanism must enforce the contract and prevent its renegotiation. Legal constraints (implicitly assumed by

Theoretical perspectives on CBI 39 Walsh) would achieve this; but this would require the courts to uphold a contract even when both government and CB deem its renegotiation to be in both their own interests. An alternative mechanism for supporting a Walsh contract in the form of a reputational equilibrium has been examined in Section 4, but here we assume that this device is not in place. This enables us to focus clearly on two other ways in which the Walsh contract serves a useful purpose in two further respects. The ﬁrst role of the contract whose existence and content is made public is to solve the principal–agent problem in that it provides an incentive mechanism to induce a legally independent CB to choose the government’s preferred monetary policy when it may have different objectives and/or private information on matters relevant to the conduct of this policy. The second role is to provide an important signalling device that conveys information about the government’s type to the private sector and the electorate. In both our set-up and in cases of extreme informational asymmetries, a Walsh contract will provide an incentive device that results in the CB adopting the government’s desired monetary policy. We now show that its role as a signal can also solve the PMC problem. Faced with a legally independent CB the government has the following decisions to make: it must decide its desired monetary stance ␲ˆ t in period t, whether to formalize its relationship with the CB in the form of a Walsh contract or use some alternative means of inﬂuencing monetary policy, whether to publicize the nature of the contract if it exists, and ﬁnally whether to renegotiate a contract signed at some earlier time. In our two-period model the optimal solution is to sign a two-period Walsh contract, before inﬂationary expectations are formed in period 1, that implements zero inﬂation in both periods (i.e. ␲ˆ 1 ⫽ ␲ˆ 2 ⫽ 0). As we have discussed previously, the possibility of renegotiation after inﬂationary expectations are formed rules out this possibility. We therefore assume that the government choice of the inﬂation target and the signing of the Walsh contract take place at this point. Consider the decision problem facing the competent government. It is clearly in the interests of such a government to distinguish itself from its incompetent rival by signing a Walsh contract and publicizing the inﬂation target ␲ˆ 1. The rational electorate can work this out, and so an incompetent government will reveal its type by either not having a Walsh contract, or having one but not publicizing its contents. However, in doing either of these things the incompetent government will also reveal its type and fail to get elected. The same applies if the incompetent government chooses to adopt a public Walsh contract. At least if it chooses a Walsh contract it gets the CB to implement its chosen inﬂation rate; but it will be indifferent as to whether to publicize the contents. The announced Walsh contract between the competent government and the CB is then an efﬁcient signal to ensure the election of the competent type. Incompetent governments

40

Ali al-Nowaihi and Paul Levine

are revealed by either the absence of a public contract or low output, the latter happening in equilibrium. There is no need for any further signalling device; for both types of government the CB will pursue the discretionary inﬂation rate, eliminate the PMC and increase social welfare. The sequencing of events for period 1 is as follows. Period 2 is as before. 1.1 Nature chooses the type of government in ofﬁce; x1 ⫽ x with probability p and x1 ⫽ 苶x with probability 1 ⫺ p. The government knows its own type but this is not revealed to the public. The opposition party is of the alternative type. These types remain ﬁxed for the duration of the two-period game. The public adopts the prior p1 ⫽ p. 1.2 The public forms its inﬂation expectation ␲1e for the period and signs wage contracts based on this expectation. Contracts last for one period. 1.3 The government chooses ␲1. It chooses either to pursue this openly through a Walsh contract specifying the inﬂation target ␲ˆ ⫽ ␲1 to the CB and public alike, or it inﬂuences monetary policy covertly through other means which includes the possibility of a covert Walsh contract. 1.4 The public observes output y1 and either the inﬂation target ␲ˆ 1 or the failure of the government to implement a Walsh contract. 1.5 Using the observations of 1.4, the public updates its belief about the type of government from p1 ⫽ p to p2. 1.6 An election is held in which the public either elects the incumbent or the opposition. The equilibrium and outcome to this game is summarized as: Proposition 7 The game has an intuitive separating PBE for which the competent type (if in ofﬁce) signs a public Walsh contract specifying an inﬂation target ␲ˆ 1 ⫽ a. The incompetent government also signs a Walsh contract specifying the same inﬂation target ␲ˆ 1 ⫽ a. The contract may or may not be public. Either way the type of government is revealed and the competent type wins the election. Thus the availability of Walsh contracts eliminates the PBC and increases social welfare. Rogoff and Sibert (1988), Rogoff (1990) and Persson and Tabellini (1990) have set out rational political business cycle models where economic ﬂuctuations result from the government’s attempts to signal its competence to the electorate. Some empirical support for this genre of models is to be found in Alesina and Roubini (1990). In their overview of this literature Persson and Tabellini (1994, p. 4) say, ‘It is not clear, however how to get rid of this distortion. The reason is that the electoral cycle conveys information to the voters. Trying to suppress it, therefore may be counterproductive: the voters would choose an incompetent government

Theoretical perspectives on CBI 41 too often, or alternatively the incumbent could be forced to signal his type in costlier ways.’ A similar viewpoint is expressed in Rogoff (1990, p. 21). In contrast to these views, this section has shown how the political monetary cycle can be avoided if Walsh-type contracts between the incumbent administration and the CB are available to governments. Walsh originally proposed such contracts for the purpose of eliminating the inﬂationary bias of discretionary monetary policy while retaining its stabilization role. For this to work some mechanism is required to make Walsh contracts renegotiation-proof: for example, legal or institutional arrangements (implicitly assumed in Walsh (1995)) trigger strategies on behalf of the public (Herrendorf, 1995) or reputational effects examined in Section 4. In the model presented in this chapter we assume none of these devices exist. Walsh contracts therefore do not eliminate the inﬂationary bias. Rather they serve the different purpose of providing an efﬁcient signalling device to the electorate that eliminates the political monetary cycle, results in the election of competent governments and increases social welfare. Of course, if Walsh contracts could also eliminate the inﬂationary bias, as in Section 4, this would increase welfare further.

6 A brief conclusion This chapter has provided a survey of the literature on the credibility problem facing monetary authorities ﬁrst raised by Kydland and Prescott (1977). We have emphasized some more recent contributions that reassess the seminal papers. Many approaches offer some variant of Rogoff-style delegation. These are only second-best solutions and indeed we have shown that in the open economy context these can become severely second-best. By contrast the contract approach with ﬂexible inﬂation targets chosen by the government and made incentive-compatible for instrument-independent (but not goal-independent) CBs offers a ﬁrst-best or near-ﬁrst-best solution. Our overall conclusion and lesson for the transition economies is that this solution to the ‘credibility problem’ constitutes the most promising approach with the monetary regime in place in New Zealand since 1989 as the possible model.

Notes 1 For a more general survey of the literature on central bank independence which provides an excellent coverage of the empirical research, see Eijfﬁnger and De Haan (1996). 2 For example, Grilli et al. (1991), and Alesina and Summers (1993). 3 For the time being we assume that inﬂation targets are reached without error. In Section 4, the sub-section on reputational equilibria with Walsh contracts relaxes this assumption. 4 In fact Rogoff got the idea of delegating to conservative bankers from his experience working at the US Fed.

42

Ali al-Nowaihi and Paul Levine

5 The foc is a quartic in b so that there may be more than one choice of conservatism. In fact it is straightforward to show that there is only one positive solution to (3.6). 6 Lohmann (1992) shows how this compromise can be reduced, with a more favourable welfare outcome, using a rule with an escape clause specifying zero inﬂation for small supply shocks and discretionary behaviour with stabilization for large shocks. 7 The treatment draws on van der Ploeg (1993). 8 Notice that n

冤冱⑀ 冥 ⫽ (n ⫹ 1)(1 ⫹ n␳)␴

2

var

i

i⫽1

9 10 11

12

13 14 15

16 17

18

and therefore the correlation coefﬁcient ␳ varies between the extremes of ␳ ⫽ 1 for a common shock, to ␳ ⫽ ⫺1/n for perfectly symmetric shocks where shocks in the aggregate cancel each other out. Full details of these calculations are provided in al-Nowaihi and Levine (1998b). In the sense of Farrell and Maskin (1989). The alternative term ‘recoordination-proof’ has been proposed to avoid confusion with renegotiationproofness as used in agency theory. The idea ﬁrst appeared in published form in Persson and Tabellini (1994) who provide an insightful review of the general issues. However, they acknowledge Walsh’s pioneering contribution made in an earlier unpublished manuscript. See also Herrendorf (1995) and Svensson (1997) for developments of the contract approach. Canzoneri (1985) considers the possibility of sustaining the ex ante optimal rule in the context of imperfect monitoring using a private sector trigger strategy. However, this approach suffers from the drawbacks pointed out in the subsection on trigger strategies above. Technically the solution is similar to the pooling equilibria examine in Barro (1986) and Backus and Drifﬁll (1985a, b).These papers however do not consider the stabilization component of monetary policy. al-Nowaihi and Levine also consider the case of a ﬁnite time horizon. The new arrangement in the UK with the Labour Party administration incorporates elements (1) and (2). There is an implied accountability of the Governor and his monetary committee for its achievement but the mechanism is less direct and explicit compared with the New Zealand case. For example, there is no explicit penalty for missing the target, and incentive compatibility relies on political embarrassment and civil servants’ commitment to public duty. E(.) refers to expectations formed at the beginning of the game. Note that preferences are now expressed in terms of an output rather than an employment objective. In addition, since welfare is linear in output there is no role for stabilization policy in the face of a stochastic supply shock. Neither of these changes are fundamental for the results and the linear-in–output welfare function does simplify the analysis considerably. Note that our set-up is slightly simpler than that in Persson and Tabellini (1990). There the types of government and opposition are random variables and if a government is re-elected, then its type in period 2 is positively, but not perfectly, correlated with its type in period 1. Crucially, however, the information assumptions are the same and both set-ups give qualitatively similar results.

Theoretical perspectives on CBI 43

References Alesina, A. and Gatti, R. (1995) ‘Independent central banks: low inﬂation at no cost?’, American Economic Review, 85, 196–206. Alesina, A. and Roubini, N. (1990) ‘Political cycles in the OECD economies’, Review of Economic Studies, 59, 663–688. Alesina, A. and Summers, L.H. (1993) ‘Central bank independence and macroeconomic performance: some comparative evidence’, Journal of Money, Credit and Banking, 25, 151–162. al-Nowaihi, A. and Levine, P. (1994) ‘Can reputation resolve the monetary policy credibility problem?’, Journal of Monetary Economics, 33, 355–380. al-Nowaihi, A. and Levine, P. (1996) ‘Independent but accountable: Walsh contracts and the credibility problem’, CEPR DP No. 1387. al-Nowaihi, A. and Levine, P. (1998a) ‘Can political monetary cycles be avoided’, Journal of Monetary Economics, 42, 525–546. al-Nowaihi, A. and Levine, P. (1998b) ‘Central bank independence: gain without pain?’, presented to the RES Annual Conference, University of Warwick, March. Backus, D. and Drifﬁll, J. (1985a) ‘Inﬂation and reputation’, American Economic Review, 75, 530–538. Backus, D. and Drifﬁll, J. (1985b) ‘Rational expectations and policy credibility following a change of regime’, Review of Economic Studies, 52, 211–221. Barro, R. (1986) ‘Reputation in a model of monetary policy with incomplete information’, Journal of Monetary Economics, 17, 1–20. Barro, R.J. and Gordon, D.B. (1983) ‘Rules, discretion and reputation in a model of monetary policy’, Journal of Monetary Economics, 12, 101–121. Canzoneri, M.B. (1985) ‘Monetary policy games and the role of private information’, American Economic Review, 75, 1056–1070. Cukierman, A. (1992) Central Bank Strategy, Credibility and Independence: Theory and Evidence, Cambridge, MA: MIT Press. Currie, D.A., Levine, P. and Pearlman, J. (1995) ‘Can delegation be counterproductive?: the choice of conservative bankers in open economies’, CEPR No. 1148. Currie, D.A., Levine, P. and Pearlman, J. (1996) ‘The choice of conservative bankers in open economies: monetary regime options for Europe’, Economic Journal, 106, 345–358. Debelle, G. and Fischer, S. (1995) ‘How independent should a Central Bank be?’, in Fuhrer, J.C. (ed.), Goals, Guidelines and Constraints Facing Monetary Policymakers, Federal Reserve Bank of Boston, Conference Series No. 38, 195–221. Dixit, A.K. and Pindyck, R.S. (1994) ‘Investment Under Uncertainty’, Princeton, NJ: Princeton University Press. Eijfﬁnger, S.C.W. and De Haan, J. (1996) ‘The political economy of central bank independence’, Special Papers in International Economics, No. 19, Princeton University. Farrell, J. and Maskin, E. (1989) ‘Renegotiation in repeated games’, Games and Economic Behavior, 1, 327–360. Fischer, S. and Cooper, J.P. (1973) ‘Stabilization policy and lags’, Journal of Political Economy, 81, 847–877. Friedman, M. (1968) ‘The role of monetary policy’, American Economic Review, 58, 1–17.

44

Ali al-Nowaihi and Paul Levine

Fudenberg, D. and Tirole, J. (1991a) ‘Perfect Bayesian equilibrium and sequential equilibrium’, Journal of Economic Theory, 53, 236–260. Fudenberg, D. and Tirole, J. (1991b) Game Theory, Cambridge, MA: MIT Press. Goodhart, C.A.E. and Huang, H. (1995) ‘What is the central bank’s game?, LSE Financial Markets Group, DP No. 222. Grilli, V., Masciandaro, D. and Tabellini, G. (1991) ‘Political and monetary institutions and public ﬁnancial policies in the industrial countries’, Economic Policy, 15, 341–392. Hamada, J. (1976) ‘A strategic analysis of monetary interdependence’, Journal of Political Economy, 84, 677–700. Havrilesky, T.M. (1991) ‘The frequency of monetary policy signalling from the administration to the Federal Reserve’, Journal of Money, Credit and Banking, 23, 423–428. Havrilesky, T.M. (1993) The Pressures on Monetary Policy, Norwell, MA: Kluwer Academic. Herrendorf, B. (1995) ‘A feasible contract for a conservative central banker’, mimeo, European University Institute, Florence. Kydland, F. and Prescott, E.C. (1977) ‘Rules rather than discretion: the inconsistency of optimal plans’, Journal of Political Economy, 85, 473–492. Levine, P, and Pearlman, J. (1997) ‘Central bank independence and ﬁscal policy: More bad news for the delegation game’, Global Economic Institutions Working Paper No. 25. Lohmann, S. (1992) ‘Optimal commitment in monetary policy: credibility versus ﬂexibility’, American Economic Review, 82, 273–286. Lucas, R.E. (1972) ‘Expectations and the neutrality of money’, Journal of Economic Theory, 4, 103–124. Lucas, R.E. (1987) Models of Business Cycles, Oxford: Basil Blackwell. McCallum, B.T. (1995) ‘Two fallacies concerning central-bank independence’, American Economic Review, AEA papers and proceedings, May, 207–211. Persson, T. and Tabellini, G. (1990) ‘Macroeconomic Policy, Credibility and Politics’, Harwood Academic Publishers. Persson, T. and Tabellini, G. (1994) Monetary and Fiscal Policy, Vol. 2, Cambridge, MA: MIT Press. Ploeg, F. van der (1993) ‘Channels of international policy transmission’, Journal of International Economics, 34, 245–267. Rogoff, K. (1985a) ‘The optimal degree of commitment to an intermediate monetary target’, Quarterly Journal of Economics, 100, 1169–1190. Rogoff, K. (1985b) ‘Can international monetary policy co-ordination be counterproductive?’, Journal of International Economics, 18, 199–217. Rogoff, K. (1990) ‘Equilibrium political budget cycles’, American Economic Review, March, 21–36. Rogoff, K. and Sibert, A. (1988) ‘Elections and macroeconomic policy cycles’, Review of Economic Studies, January, 1–16. Romer, D. (1996) Advanced Macroeconomics, New York: McGraw-Hill. Svensson, L.E.O. (1997) ‘Optimal inﬂation targets, “Conservative” central banks and linear inﬂation contracts’, American Economic Review, forthcoming. Ulen, T. (1984) ‘The efﬁciency of speciﬁc performance: towards a uniﬁed theory of contract remedies’, Michigan Law Review, 84, 358. Walsh, C. (1995) ‘Optimal contracts for central bankers’, American Economic Review, 85, 150–167.

2

The background to reform Central banking in a command economy Nigel M. Healey and Janet Ilieva

1 Introduction The aim of this chapter is to provide an overview of the role of money and banking in the former socialist countries in Europe. The model was inspired by Marxist theory, as developed and promulgated by Lenin. In a centrally planned economy, the entire banking sector is nationalised and provides more than simply the state monopolisation of money and credit. It also provides a mechanism for gathering and processing the whole spectrum of ﬁnancial information (reﬂecting production and employment at enterprise level) necessary to control the economy. In the former centrally planned economies of Central and Eastern Europe, the banking system, described as a monobank system, tended to be a part of the state bureaucratic mechanism and was totally submissive to the government (Communist Party). The state set policy and the central bank implemented it. A characteristic feature of the socialist banking system was the huge buildup of savings, since that was the only legal way the population could undertake any kind of investment. Industrialisation was the priority of the socialist state and resources were allocated towards encouraging this without regard to the production of consumption output. Shortages of consumption goods inevitably led to a buildup of unspendable savings by the population, since no private sector investment opportunities were available. This chapter begins with a discussion on the theoretical background of the socialist banking system based on Marxist theory. It is followed by Lenin’s studies and interpretations of Marxist philosophy that gave birth to the banking model implemented in the socialist economies. The section closes with an analysis of the role of the planning process in socialist economies. The major plans, the plan for people’s income and expenditure, the cash plan and the credit plan that anticipated the central bank’s activity, are all described. Section 3 looks at the performance of the banking system in a socialist economy and the bank services provided by the banking sector to both private individuals and business organisations. The organisational pattern

46

Nigel M. Healey and Janet Ilieva

of the banking system is discussed and it is shown how this provided information to monitor the economy performance. The most common services offered to both households and ﬁrms are explored. Section 4 focuses on the monetary policy, and identiﬁes inconsistencies between monetary policy and the state of the macroeconomy. We explore the inconsistencies between the money supply and the rate of interest with particular emphasis on monetary equilibrium. Possible ﬁnancial system leaks are traced. This section also describes the way in which interest rates and exchange rate arrangements were set within the Council for Mutual Economic Assistance (CMEA). We discuss the suppressed nature of inﬂation and the consequent emergence of shortages in centrally planned economies as a shortage phenomenon. This does not conﬂict with the monetary nature of inﬂation; it simply explains that excess demand manifests itself in shortages when prices are prevented from rising. Monetary overhangs, typical of the centrally planned system, are grouped into two major categories: supply-side overhangs and demand-side overhangs. We discuss the major factors causing these distortions of the ﬁnancial system. The chapter closes with a summary of the ﬁndings.

2 Theoretical background The theory of the banking system in most of the former socialist countries of Central and Eastern Europe was based on Marxist theory interpreted by Lenin. The importance of the banking system made it essential to introduce strict and reliable controls, and this was one of the top priorities of the newly established socialist governments in the mid-1940s. The wholesale nationalisation of the banking sector was one of the ﬁrst steps undertaken during the economic reforms of the 1940s. The centralisation of banking and the establishment of a single credit system centralised ﬁnancial information that was necessary for effective state control over money ﬂows and total output. The structure of the banking system in a plandriven economy was designed under Lenin’s directives that found real implementation with the introduction of the monobank system; that is, a central bank with a branch network throughout the country. In other words, a single bank performed the role of central bank and simultaneously provided banking services to the entire community. Lenin’s studies on banking: a brief overview Lenin’s focus on the banking system implied centralisation and the creation of a monobank as a means of implementing state control over the entire economy. To achieve this, the entire economy was nationalised, and public property (collective ownership) was introduced and replaced all other existing forms of ownership. The ﬁrst ‘principal measure’ stated by Lenin in exercising control is:

Central banking in a command economy 47 ‘Amalgamation of all banks into a single bank, and state control over its operations or nationalisation of the banks.’1 The banks themselves are described as ‘principal nerve centres of the whole capitalistic economic system’.2 With respect to terminating the capitalist system, Lenin stated that ‘everything that is vicious in it’ should be cut off. He further argued that because of their inﬂuence on the level of economic activity, ‘without “laying hands” on the banks nothing of any value, nothing “revolutionarydemocratic”, can be accomplished’.3 Lenin also commented on the signiﬁcance of nationalisation of the banks. The necessity of state control was explained in the following way: ‘Only by nationalising the banks can the state put itself in a position to know where and how, whence and when, millions and billions of rubles ﬂow.’4 State monopoly of the banking system simultaneously gave the state monopoly control over the investment policy of the country. Investment was concentrated in developing heavy industry, and all disposable resources were transferred from agriculture to industrial development. The industrialisation that took place in Russia during Stalinist times could also be seen in the other former socialist states of Central and Eastern Europe after the mid-1940s. The motivation behind the economic and social dislocation that adjustment entailed was the creation of a self-sufﬁcient economy, and industrialisation was viewed as a means of achieving economic independence from the capitalist countries. The process was one of forced change, though it was supported by the creation of substantial wage differentials between the average industrial wage and the average agricultural wage. The importance of banking concentration and self-sufﬁciency could be traced back to Lenin’s studies on the banks and their role. He examined four countries: Germany, Britain, France and the USA, and found that ‘This transformation of numerous modest middlemen into a handful of monopolists is one of the fundamental processes in the growth of capitalism into capitalist imperialism; for this reason we must ﬁrst of all examine the concentration of banking’.5 The tendency towards capitalisation of the banks was supported by empirical data and Lenin noted that: Together, these four countries own 479,000 million francs, that is, nearly 80 per cent of the world’s ﬁnance capital. In one way or another, nearly the whole of the rest of the world is more or less the debtor to and tributary of these international banker countries, these four ‘pillars’ of world ﬁnance capital. . . . The export of capital inﬂuences and greatly accelerates the development of capitalism in those countries to which it is exported.6 To avoid falling under the capitalist yoke, it was therefore considered paramount to ensure that the newly created Soviet state was self-sufﬁcient.

48

Nigel M. Healey and Janet Ilieva

The formation of COMECOM after the Second World War and sectoral specialisation of the member countries were both motivated by the drive to achieve self-sufﬁciency. The banking system was given a fundamentally different role from the normal one associated with conducting monetary policy and the provision of a sound monetary environment. The key issue in banks’ activities was based on Marx’s view that the banking system possesses ‘the form of universal book-keeping and distribution of the means of production on a social scale, but solely the form’.7 Indeed, the banking system was given the role of universal bookkeeper of the country and a general supervisor of industrial activities – but nothing else! This role is underlined by Lenin’s perception of communist society: ‘Accounting and control – that is mainly what is needed for the “smooth working”, for the proper functioning, of the ﬁrst phase of communist society.’8 This description of a banking system envisaged no banking activities. Instead the framework corresponds perfectly to the needs of the bureaucratic machine necessary to monitor the entire economy and to control it. The basic functions of the central bank consisted of a series of bookkeeping, exercises made easier because prices were set in accordance with the overall plan for the economy. Thus prices did not reﬂect the relative scarcities and consequently failed to perform their information role. Resources were simply allocated in accordance with the directives of the plan. The role of the planning process State planning completely replaced the market as the allocative mechanism. Economic plans were designed with a ﬁve-year time horizon and economic targets were assigned to industry in accordance with this. These plans were presented and approved at the congresses of the Communist Party held each ﬁfth year. By rule, the congresses of most of the former socialist countries were held a couple of months after that of the former Soviet Union. Annual plans regulated the performance of the economy in accordance with the overall ﬁve-year plan. The annual plans had operational signiﬁcance, since they embodied the tools and means of achieving the targets in the ﬁve-year plan. The objectives of the ﬁve-year plans were associated with socialist stability and the establishment of a new order: the communist society. Economic performance was put in a framework that emphasised the quantitative side of total output. The qualitative side was given less prominence both because it is difﬁcult to measure qualitative changes in output and because it was felt that resources could be better used for the overfulﬁlment of the planned quantity of production than for improvements in quality. Planning the circulation of money was an integral part of the state plan where all the possibilities of money ﬂowing from the issuing institution to the population and vice versa were fully anticipated. The most common

Central banking in a command economy 49 way that money reached the population was through the payment of wages. All the necessary data – the labour force (the average employment in the early ﬁve-year plans was about 95 per cent), the proportion of national income to be spent on consumption and its money equivalent and so on – were fed into the national plan. As noted above, the aim of the plan was self-sufﬁciency, and this implies that the quantity of domestic goods and services should equal the income of the population. It also implies that the quantity of money in circulation was an integral part of the plan and was determined with respect to the input–output calculations of the different sectors of the economy. In fact, the plan for the circulation of money was based on: 1 2 3

the plan for the income and expenditure balances of private individuals; the cash plan; the credit plan.

The income–expenditure balance of private individuals represents that part of the national income which indicates the level of earnings in the economy and the level of spending. The structure of this balance is detailed in Appendix A at the end of this chapter. The cash plan outlines the banks’ turnovers providing the resources of cash income to the central bank and indicating the directions of money transfers. It is an operational tool for determining the scope of money circulation and controlling money emission in the economy. The cash plan is in strict accordance with the credit plan and, in fact, serves as the resource for the provision of credit. The money issue should therefore be in line with the planned cash residual. Money income and payments are indicated in the state plan along with the state budget. The cash plan thus serves as a control device for total output. It indicates the deviations from the preliminary planned parameters in the economic plan and the budget, and hence it reveals the performance of the economy in terms of the plan. The cash plan is therefore crucially important as a means of monitoring whether the economy is delivering the targets set in the state plan and the budget. The structure of the cash plan is detailed in Appendix B. The objective of the credit plan is the gradual formation and use of the socialist credit resource. The credit plan, together with the cash plan, is drawn up by the central bank. According to the credit plan, the socialist central bank had annual consultations with the sectoral banks, after which the central bank set the following limits on the provision of loans to enterprises: 1 2

the decrease/increase in the net debt held in socialist and non-socialist currency; the decrease/increase in the amount of currency released into circulation by the sectoral bank.

50

Nigel M. Healey and Janet Ilieva

The sectoral bank credit plan was harmonised with respect to the above and restrictions were established within agreed limits in accordance with the economic plan. The credit plan is prepared as a conventional balance sheet showing assets and liabilities. The liabilities are grouped in the following way according to their origin: (1) own resources, (2) resources of industrial and public organisations, budget resources, (3) National Insurance Institution’s resources, (4) National Savings Institution’s resources and household savings, (5) money in circulation (money issue). The assets are grouped in terms of the industry branch and intended recipients. The state credit plan consisted of two sub-plans: a plan for ﬁnancing the turnovers of enterprises and a plan for ﬁnancing the capital investment of enterprises. The ﬁrst plan covered short-term ﬁnancing decisions while the latter covered long-term ﬁnancing decisions. Both plans are discussed in detail in Section 3 of this chapter. Since the credit plan also revealed the currency ﬂows within the economy, it was viewed by some economists as the plan for money turnover within the economy.9

3 Banking system performance in a centrally planned economy The monobank system was therefore the most appropriate way both to centralise information about the performance of the economy, and simultaneously facilitated control over the business activity of the country. Naturally the entire banking sector, like all other sectors in the economy, was state owned. The central bank was effectively the chief accountant of the country and kept a record of all transactions that took place in the economy over time. The central bank was also the sole issuer of currency in the economy and performed the functions of a commercial bank. However, alongside the central bank, several sectoral banks specialised in providing banking services to particular industries. The provision of money and the supply of credit were both embodied in the economic plan. The central bank was independent to the extent that it had freedom to voluntarily obey the government and to be part of the administratively managed bureaucratic machinery of the country. Banking system organisation The monobank system was implemented in most of the socialist countries, along with state monopolisation of the ﬁnancial institutions. Parallel with the central bank there existed a National Savings Institution serving the population (accepting deposits and providing loans to the native population only) and a Bank for Foreign Trade which dealt with transactions outside the country – mainly the Comecom countries. Table 2.1 shows the structure of the banking system that was typical of socialist countries in the mid-1970s. In the 1980s there was a tendency to separate the commercial dealings of the central bank from its other activities. To facilitate this, sector

National bank (34) National bank (100) National bank (135)

National bank (– )

National bank (100) National bank (15,980)

National bank (40)

National bank (82,000) National bank (7)

Albania Bulgaria Czechoslovakia

German DR

Hungary

Poland

Romania

USSR

Yugoslavia

About 10 investment banks, 40 investment commerce banks

Bank of Agriculture and Food Industry; Investment bank Construction Bank

State Development Bank Bank of the Food Economy

Bank of Agriculture and Food

– – –

Sectoral banks or investment bank

Savings branches of the USSR State Bank Post Ofﬁce Savings Bank

Branches of the National Bank of Poland State Savings Bank

State Savings Bank

State Savings Bank

– State Savings Bank10 Sate Savings Bank

Savings Bank

USSR Foreign Trade Bank Jugobanka

Romanian Bank for Foreign Trade

Hungarian Foreign Trade Bank Commercial Bank in Warsaw

German Foreign Trade Bank

– BG Foreign Trade Bank Czechoslovak Commercial Bank

Bank of foreign trade

About 15 commercial banks and 380 social accounting ofﬁces

–

– – Zivnostenska Bank credit and loan cooperatives11 Berlin People’s Bank,12 Berlin Municipal Bank,13 co-op banks for farmers and tradesmen General Securities Trading Bank14 Polish Guardian Bank,15 savings and credit cooperatives –

Other banks

Source: J. Wilczynski (1978) Comparative Monetary Economics. Capitalist and Socialist Monetary Systems and Their Interrelations in the Changing International Scene, Basingstoke: Macmillan.

State bank (no. of branches in brackets)

Country

Table 2.1 Banking system in socialist countries

52

Nigel M. Healey and Janet Ilieva

commercial banks were formed to provide banking services to different sectors of the economy. The banks themselves could participate in joint activities with the enterprises with which they were associated. For instance, the Electronic Bank was supposed to provide investment facilities to the electronic industry, and the Agricultural and Co-operative Bank specialised in providing loan and deposit facilities mainly to the agricultural co-operatives. Special arrangements were made for branches of industry not covered by these facilities. For example, in Bulgaria the Economic Bank provided banking facilities to enterprises that were not otherwise catered for by specialised banking arrangements. The peculiarity of this system was that the sectoral banks did not have any branches, and the branches of the central bank provided all of the above facilities. Initially, the sectoral banks were created to support their respective industries, but all their transactions were carried out via the central bank branches, since the sectoral banks only had a head ofﬁce usually based in the capital city of a country. This was in accordance with Lenin’s philosophy that the state should be in a ‘position to know where and how, whence and when’ the money ﬂows. In 1989 the commercial banks were given the legal right to open branches throughout each country. Prior to that time only the central bank and the National Savings Institution were authorised to run their own branch network. All enterprises were obliged to keep accounts with a branch of the central bank and could not dispose freely of their own accounts, since funds placed with the central bank were earmarked according to their ﬁnal purpose (Kornai, 1992).15 For example, funds designated for investment were placed in a special account and could be used only for investment purposes, funds marked for operational inputs could be used ad hoc but not for any purpose other than operational inputs and so on. The operation of these arrangements deprived the national currency of a medium-ofexchange function and in consequence severely limited its domestic convertibility. Each enterprise could hold cash only up to a strictly deﬁned limit and all enterprises were obliged to hold these balances in their own account at the central bank. It was therefore not difﬁcult for the state to keep account of and control cash ﬂow throughout the economy. In reality the process of auditing business activity was simpliﬁed anyway because all enterprises were state property or, as in the case of most agricultural organisations, were managed as a co-operative. Interest rate policy In the centrally planned economies, the state determined the limits within which the interest rate could ﬂuctuate and the central bank was required to conduct the interest policy within these prescribed limits. The sectoral banks could set their own interest rate when dealing with enterprises, but they could not exceed by more than 2 per cent the basic interest rate set by

Central banking in a command economy 53 the central bank. Naturally, the level of the interest charged on loans was very low. Interest rates performed a different role in socialist economies than in market economies. In market economies, commercial banks earn interest income on their borrowing and lending arrangements, whereas in command economies, interest rates were not supposed to include any element of ‘surplus value’. Neither did they include any kind of risk premium, and throughout the centrally planned economies nominal interest rates were set at what, by Western standards, were unrealistically low levels. The result was that there were frequent shortages of capital, with demand exceeding the available supply and, since increases in the price of capital were impossible, the provision of capital to enterprises was regulated through the credit plans of the central bank. To relieve capital shortages, ﬁrms began circumventing the system and used barter deals among themselves. These arrangements involved exchanging goods against goods, and so-called inter-ﬁrm loans were introduced to replace the need for bank credit. Interest rates set on deposits also tended to be relatively low throughout the planned economies. In the late 1980s, rates on individual deposits of a short-term nature averaged about 1 per cent, and rates on long-term savings deposits average about 2 per cent. Taking account of inﬂation, both rates implied a substantially negative real return to depositors. Differential rates of interest were paid on foreign currency deposits, depending on whether the currency was from a socialist or a non-socialist country. Interest paid on deposits denominated in ‘socialist’ currency was 1 per cent and 0.75 per cent for ‘non-socialist’ currency deposits. Differential rates were also paid on deposits made by industrial enterprises and varied from 1.5 per cent to 7 per cent. Saving and the scarcity economy As noted above, the state had monopoly power over investment and its investment policy was implemented throughout state-owned businesses. Co-operative ownership of property was permitted, particularly in the agricultural sector, where a range of different types of co-operatives functioned. With the exception of agricultural smallholdings, defined by law in terms of acreage and restricted to meet individual household needs, private businesses were not permitted in most of the socialist economies. Under the central planning system the development of heavy industry was prioritised and investment was concentrated in this sector. The production of consumer goods was given less prominence and demand for these goods permanently exceeded supply – despite the fact that consumer goods were highly standardised, typically of low quality and failed to meet individual preferences and tastes! Throughout the socialist economies

54

Nigel M. Healey and Janet Ilieva

Table 2.2 Waiting for housing: international comparison County

Average length of time on a waiting list in the 1980s (years)

Bulgaria Czechoslovakia East Germany Hungary Poland Soviet Union

5–20 6–8 3–4 4–6 15–30 10–15

Source: Kornai (1992).

households were compelled, through lack of choice, to save – even when real interest rates became negative. The accumulation of involuntary deposits left most of the population holding unspendable deposits and created a deep feeling of injustice when inﬂation began to erode their real value. Tables 2.2 and 2.3 indicate the extent of the problem in terms of how long it would take to satisfy the demand for goods such as housing and cars in a selection of Central and Eastern European countries. People were compelled to wait for goods not because they lacked the funds with which to make purchases, but because of the unavailability of the goods which they demanded. The government deﬁned the priority of different families, and each had to wait according to the priority status they were accorded. Normally a higher priority was given to families with young children (the greater the number of dependent children, the higher the priority granted) or with distinguished party leaders as well as war veterans or participants in the socialist movement before the actual start of the socialist system (e.g. partisans). Family size was particularly important in the allocation of housing. The size of house a family was allocated was determined according to the size of the family, and occupation of more than one residence was permitted in exceptional cases only. Table 2.3 Waiting for cars: international comparison (1989) Country

Bulgaria Czechoslovakia East Germany Hungary Poland Romania Source: Kornai (1992).

Waiting period (years) Lada

Skoda

Moskvich Wartburg Trabant

Dacia

10–12 3–4 17 4–6 5–6 –

5 – 16 6 6–8 –

2 – – – – –

1 – 15 – – 4–6

2 – 14–16 1 3–4 –

1 – 14 0 2–3 –

Central banking in a command economy 55 Methods for bank credit provision Under the socialist system the provision of bank credit was speciﬁed in the state credit plan. However, the provision of loans was only one of the possible ways of ﬁnancing investment in the plan-driven economies. Alternative methods of ﬁnancing available to enterprises were: 1 2 3

budgetary ﬁnancing; self-ﬁnancing; combined ﬁnancing between own and other resources.

Budgetary ﬁnancing was one of the major sources of ﬁnancing public sector enterprises, and its importance was particularly crucial during the early ages of socialism when industrialisation was the major priority of the state. When economic reform started in most socialist countries in the 1960s, budgetary ﬁnance became less important and, as Table 2.4 illustrates, most industries began to rely more heavily on ‘own resources’ and bank loans as sources of investment funding. Financing the residuals Bank loans were also provided to support the turnover of enterprises, depending on whether their need for additional working capital was temporary or permanent. A temporary ﬁnancing problem arose when the output of an enterprise exceeded the amount stated in the plan. Temporary ﬁnance was needed in this case so that work in progress formed during the production process could be transformed into ﬁnished output. In this case ﬁnance was destined for the speciﬁc department in need and could not be used to meet the ﬁnancing needs of any other department, even within the same ﬁrm! Departments using this type of credit were required to have separate accounts where the loan was held. The maturity of the loan depended on the time needed to turn residuals into ﬁnal output.

Table 2.4 Sources of ﬁnancing the capital inputs 5-year periods

Total

Own resources

Bank loan

Budget resources

1st 5 years (1949–1952) 2nd 5 years (1953–1957) 5th 5 years (1966–1970) 7th 5 years (1976–1980) 8th 5 years (1980–1985) 9th 5 years (1986–1990)

100 100 100 100 100 100

19.0 28.0 32.0 28.0 38.1 35.3

8.0 7.0 30.0 49.0 30.7 31.7

73.0 65.0 38.0 23.0 31.2 33.0

Source: ‘Finansi i kredit na sotsializma’ Finance and Credit of the Socialism, Georgi Bakalov, Varna, 1989.

56

Nigel M. Healey and Janet Ilieva

Turnover credit provision Turnover credits were provided to meet the more permanent needs of public organisations which, for one reason or another, had been allocated insufﬁcient funds to fulﬁl their ﬁnancial obligations under the plan. Those organisations receiving longer term turnover funding had ﬂexibility to decide how the funding would be allocated between different departments within the organisation. Here again the maturity of the loan was dependent on the time required to initiate changes designed to remove the problem. The use of the turnover credit provision is illustrated in Figure 2.1. Financing capital investment The capital investment plan was one of the major components of the national plan and set the level of technical equipment to be produced during the coming ﬁve- to ten-year period. It revealed the tendencies expected in the production process and the priorities of the state. For instance, a peculiarity of the capital investment programme was that it tended to focus on a small number of projects, each of great importance. Priority was given to creating the means of production (equipment, machinery) and installation of new equipment in enterprises. Investment plans were usually drawn up with a ﬁve-year time horizon, though in some cases one-year plans where drawn up. In such cases state policy was speciﬁed in detail. Each investment project was awarded to an organisation that had won the preliminary competition among stateowned enterprises to fund a speciﬁc project. The transparency of the socialist planning process consisted in providing the framework of the national plan to the enterprises in a way that would assist them in preparing their individual plans. These individual plans were discussed with the Ministry of Industry and Planning with participation of the sectoral banks. Issues of debate were put before the Ministry Council for a ﬁnal decision.

Firm’s Credit Account

Usage of the Credit Money —— Goods

…

Pay back Production …

Figure 2.1 Firm’s credit account.

Goods’—— Money’

Central banking in a command economy 57

4 Monetary policy Money supply and money demand Money issue was perhaps the only ‘normal’ function of the central bank in a planned economy – or at least as normal as money issue could be in a centrally planned economy! The scope of money issue was stated in the cash plan of the central bank and this, in turn, was one of the components of the state plan. The cash plan was developed alongside the credit plan and, together with the plans for people’s income and expenditure balances, formed the plan for money circulation. Like any control device, the planning mechanism left the expectations of some groups unfulﬁlled. In centrally planned economies these unfulﬁlled expectations led to an unofﬁcial reallocation of funds from the channels designated in the plan. Funds were reallocated within the credit sector and within the money in circulation element of the plan. Unofﬁcial reallocations within the credit sector represent a deviation from the state’s credit plan. Excess demand for credit among state-owned ﬁrms increased competition between them, though not in terms of business performance or proﬁtability. The administrative stratum of the plandriven economy gave priority to party ofﬁcials (see Appendix C) and hence competition took the form of lobbying within the Communist Party. Party ofﬁcials were subject to constant pressure for loan provision from lobbying enterprises and, in successful cases, the central bank did not refuse the loan (see Section 5). This resulted in inefﬁcient capital allocation as funds dried up in highly indebted enterprises. The criteria for credit provision was thus political, and credit was allocated in ‘accordance with non-commercial, bureaucratic criteria. Nor is it just a matter of clear, opened criteria declared in advance; other factors – lobbying, vertical bargaining, personal connections, and corruption – are at work behind the scene’.16 Sometimes the lobby came from the government itself, and in these circumstances the provision of credit could not be refused. When the transformation of Central and Eastern Europe started at the beginning of the 1990s, household accounts were transferred to the emerging market-oriented banking system. However, initially at least, the state sector continued to dominate the economy, and the major clients of the newly reformed banking system were the large state ﬁrms. This precipitated a crisis within the banking sector throughout the Central and Eastern European countries. For example, in Bulgaria in 1996, sixteen commercial banks collapsed because of the default of state ﬁrms. The collapse of the CMEA in 1991 revealed the full extent of the monetary overhang that resulted from years of involuntary saving by the population of Central and Eastern Europe. As a result of the collapse, inter-ﬁrm indebtedness increased rapidly, since most ﬁrms lost their guaranteed markets and hence could not sell their output. The ‘habit’ of spend-

58

Nigel M. Healey and Janet Ilieva

ing easy money developed in the days of planning, continued in the rapidly changing economic reality of the transition process, and plunged ﬁrms into even deeper debt. In planned economies, money in circulation was carefully controlled, and any payment exceeding approximately one average wage had to be transacted through the monobank system. Inter-ﬁrm loans were therefore used to bridge the gap at ﬁrm level between the demand for and the supply of money at the existing price level. Similarly inter-ﬁrm loans were used to relieve capital shortages or simply as an expedient way of solving liquidity problems, enabling ﬁrms to meet their current liabilities. In any case, liabilities increased as inter-ﬁrm indebtedness grew. This ‘money creation’ encouraged ﬁrms to spend funds they did not have and contributed to an increase in the amount of money in circulation. Since inter-ﬁrm indebtedness was not envisaged in the economic plan; its emergence caused a discrepancy between the planned and actual levels of money circulation. As well as being a contributory cause of inﬂation, the increased quantity of money in circulation helped precipitate the collapse of the CMEA when the guaranteed market for ﬁrms’ output disappeared. The exchange rate system in a centrally planned economy The core application of the exchange rate system in centrally planned economies was expressed in ﬁnancial accommodation of the international relations within the Council of Mutual Economic Assistance17 (known as CMEA or Comecom in the Western countries). CMEA was based on division of labour and mutual collaboration. Theoretically its basic characteristics were equalisation of countries’ growth and development, sectoral specialisation aimed at increasing efﬁciency, and reducing the number of hours worked in member countries without diminishing output. In practice, administrative control and bureaucracy were the dominating features of this integration, and often contracts between countries were not in the best interests of either party. Trade was highly concentrated among the CMEA countries, as shown in Table 2.5.18 Member countries accepted that for purposes of trade, prices were to be based on the average of world prices during the most recent ﬁve-year period subject to the condition that the domestic price of goods was above that used for international trade. The aim was to encourage trade among CMEA countries.19 A peculiarity arising out of this system is that within CMEA, similar goods were traded for the same prices, no matter what the country of origin, and, by implication, prices had no real resource information content. Furthermore, once a contract was agreed, prices did not change for at least a further ﬁve years. Deriving and administering prices in this way was thought to provide stability against the impact of speculation on the prices of basic traded goods. However, following the oil crisis of the 1970s, annual price adjustments became possible.

Central banking in a command economy 59 Table 2.5 Foreign trade of CMEA members by trade bloc in 1970 and 1983 (in %)a Destination

CMEA countries

Developed capitalist countries

Country

1970

1983

1970

1983

CMEA Bulgaria Hungary Vietnam GDR Cuba Mongolia Poland Romania USSR Czechoslovakia

60.8 74.4 62.4 n.a. 67.3 n.a. n.a. 63.1 49.3 55.6 64.2

58.6 77.1 49.6 79.2b 62.5 83.1 97.2 68.5 47.7 51.2 71.9

23.7 17.0 24.8 n.a. 25.9 n.a. n.a. 27.1 35.8 21.2 22.4

26.0 12.1 33.3 13.8 29.4 9.4 1.2 21.6 23.8 30.1 16.6

Source: O. Bogomolov (1986) Strany sotsialisma v mezhdunarodnom razdelenii truda, Moscow, p. 292. Notes a The exchange rates used to aggregate the trade ﬂows probably lead to an exaggeration of the share of CMEA trade in total trade. b 1982.

Transactions among the CMEA countries were facilitated through the intermediation of the International Bank for Economic Collaboration which was established in October 1963 with its headquarters in Moscow. On 1 January of the following year, the collective currency of member countries, the transferable rouble, was introduced. Although the transferable rouble had no physical existence, it performed the usual functions of money: measure of value, store of value and medium of exchange. Member countries were allocated shares in the bank’s capital according to the ratio of a country’s exports to total turnover of member countries. Later, in 1971, a sister institution, the International Investment Bank, was set up. Whereas the main function of the International Bank for Economic Collaboration was to provide short-term ﬁnance to ensure sufﬁcient liquidity for the continuation of trade, the basic function of the International Investment Bank was the provision of medium-term (up to ﬁve years) and long-term (up to ten years) loans for investment purposes. Although loans were mainly provided for CMEA member states, some loans were also granted to certain less developed countries. All contracts between CMEA members were signed in transferable roubles at mutually agreed rates of exchange. Payments in any other currency among member states were not acceptable due to the ‘instability’ of the capitalist exchange rate system. It was argued that domestic disturbances could not affect the exchange rate with respect to the transferable

60

Nigel M. Healey and Janet Ilieva

rouble because only the prices on international contracts were taken into consideration. Domestic prices were excluded, and therefore had no inﬂuence on the value of the collective currency. The transferable rouble exchange rate was deﬁned as a ratio expressing socially necessary labour in domestic currency against internationally necessary labour. Obviously, the socially necessary labour was different in different countries, and exchange rates were set administratively by the contracting countries, and market forces had a very subordinate inﬂuence. The exchange rate arrangements among CMEA members were agreed on 8 February 1963 in Prague, and each country’s exchange rate vis-à-vis other countries was based on domestic purchasing power parities. However, the exchange rate in the socialist bloc performed an unusual role, since it was used only for non-commercial contracts with the transferable rouble being used for commercial contracts. (Transferable roubles were also used in some non-commercial transactions such as research teams’ exchange and technical support.) The Foreign Trade Bank of each country fulfilled all contracts involving exchange rates and opened accounts in the Foreign Trade Bank of each trading partner. When the Foreign Trade Bank had surplus on the non-commercial operations, this surplus was transferred to the country’s account at the International Bank for Economic Collaboration, and that country could spend it on buying goods at contract prices. Because of this, a special coefficient20 was applied when money from non-commercial contracts was spent for commercial purposes. The coefficients demonstrated the domestic price-level deviation from the international price level. The same exchange rates were also applied when a ‘socialist’ currency was converted into a ‘non-socialist’ currency. However, tough foreign currency controls strictly limited the availability of foreign currency from outside the CMEA and, for many traders, the black market provided the solution to the shortage of hard currencies. Inﬂation Inﬂation was an ambiguous concept in centrally planned economies. Ofﬁcially it referred to the administratively determined change in the average price level, but as the 1970s turned into the 1980s the divergence between ofﬁcial measures of inﬂation and the experience of most people was growing. Inﬂation is usually associated with shortages in the plan-driven economies,21 and some of the reasons for this are explored below. Inﬂation: shortage phenomenon? The investment ratio in most of the centrally planned economies was relatively high, sometimes much higher than in the developed capitalist coun-

Central banking in a command economy 61 Table 2.6 Investment and consumption Five-year periods

Investment

Consumption

1953–1957 1958–1960 1961–1965 1966–1970

19.8 26.1 27.4 31.3

80.2 73.9 72.6 68.7

Source: T. Zhivkov (1983) Za haraktera na sobstvenata materialno-tehnicheska baza na zrelia sotsialism, Partizdat, Soﬁa.

tries. As an illustration, Table 2.6 shows the ratio of investment and consumption to national income22 in Bulgaria23 for the period 1953 to 1970. Table 2.6 shows that the second projected ﬁve-year plan in Bulgaria appears as only a two-year time span. This is because the planned ﬁve-year targets were actually achieved within two years! It is inconceivable that the rate of investment undertaken in Bulgaria would have been in any other but a centrally planned system. Shortages of goods persuaded consumers to switch their plans according to the situation. Many were happy to pay even higher prices if that would guarantee them the goods they wanted to buy. At the same time prices were set administratively and could not send signals to producers to reﬂect consumer preferences. Even if this were possible, producers had limited resources and the targets were stated in the plan, not by consumer demand. Obviously ﬁrms were aware that most people were happy to pay higher prices than the administratively ﬁxed price. Consequently producers routinely received bribes to make goods available to consumers. If it were possible to catalogue all the bribes paid over a given period of time we would have a record of the actual prices paid by consumers. This could then be used to compute the actual rate of inﬂation experienced by most people. Informative though this would be, the criminal nature of accepting bribes ensures that no such record exists. Monetary overhangs Two types of monetary overhang may be distinguished in socialist economies: supply-side overhangs and demand-side overhangs. SUPPLY-SIDE OVERHANGS

Supply-side monetary overhangs were routed in inter-ﬁrm indebtedness. Effectively the ﬁrm sector was semi-monetised (Kornai, 1992, p. 145) and money did not really play any crucial role. For some ﬁrms with a strong political lobby, bank loans also continued to be readily available and were

62

Nigel M. Healey and Janet Ilieva

used to meet their current liabilities. The ready availability of bank loans was also due partly to the growth in the number of newly formed banks. Many of these were under-capitalised and were staffed by untrained personnel. Thus the stock of money was rising at a time when realised output was declining due to the collapse of formerly guaranteed markets for output. DEMAND-SIDE OVERHANGS

For years the growth of effective demand in the socialist economies had been restricted by involuntary savings. Households therefore possessed considerable spending potential and, in parallel with the ofﬁcial market in planned economies, there was a black market providing most of the goods that the ofﬁcial market failed to offer. Prices in the black market were several times greater than ofﬁcial prices but arbitrage ensured that after the payment of bribes, the prices paid were often similar! Nevertheless, as output collapsed, the black market could only partially absorb the savings that were unleashed upon it. It was certainly not possible to provide all the output, especially durable output (e.g. houses, cars) that consumers demanded. The demand-side effect of the monetary overhang was mitigated to some extent by the speed of the reform process. Along with domestic price liberalisation that took place in 1991, the current account and exchange rate in most of the former centrally planned economies were also liberalised. Imported goods began competing with domestic goods, as the former provided some relief to the shortages of domestic output. Consumers were well used to shortages, and their reaction to the increased availability of goods was to accumulate stockpiles. The result was that shortages quickly reappeared. In other words, the population had developed a ‘shortage psychology’ which persuaded them to spend their accumulated savings and store goods for ‘just in case’. Thus the excess in demand forced prices inexorably upwards. Other instability sources Inﬂation may also have had a political foundation in some transition countries. The desire of some politicians to hold on to their political appointments, mixed with their desire to further the privatisation process, led to rising budget usually ﬁnanced by printing money. Rising inﬂation was the inevitable result and, when the prospects of re-election for those responsible for the malaise were poor, it would be left to others to deal with the problem. The situation differs substantially from many Western democracies which have experienced fairly lengthy political cycles. For example, the Conservative government elected in the UK in 1979 retained power for eighteen years, and in Germany Christian Democrats were in power

Central banking in a command economy 63 for sixteen years. This compares with a political cycle in many transition economies of between one to two and ﬁve years.

5 Conclusion The central bank in a socialist economy had the role of a chief bookkeeper of the country. This was very convenient for purposes of control and audit of the entire economy by the government (the Communist Party). In applying its monetary policy, the central bank aimed to ensure that interest rates had no element of surplus value. Hence the price of capital failed to reﬂect its true opportunity cost and the result was excess demand from the business sector. In a market economy the price of capital would be forced upwards, but this was not part of the central plan, and in consequence resources were allocated inefﬁciently, and no corrective mechanism existed to encourage an efﬁcient allocation. The services provided by the banking system resembled a deposit box: people could easily deposit funds there and in return receive a symbolic rate of interest (about 1 per cent) which, after allowing for inﬂation, transposed into a negative real return on their savings. At the same time ﬁrms could borrow these funds at minimal rates of interest. However, the ‘magic box’ could show at any moment what, where and how was happening in the economy since monetary ﬂows were strictly monitored. Surprises could rarely happen, since everything was framed in a ﬁve-year plan and deviations were not encouraged. The two tier bank system, introduced in the late 1980s, was a unique mixture between centrally set directives and regulations, and competing state ‘commercial’ banks controlled by a central bank that was incorporated within the bureaucratic machine. This chaotic system had a short life and was swept away as the process of transition gathered momentum.

Appendix A Population income–expenditure balance I Money income A INCOMES FROM STATE AND CO-OPERATIVE INSTITUTIONS

1 2 3 4 5 6 7

Wages Incomes from sale of agricultural products Pensions Financial funding Stipends Insurance Loans

64

Nigel M. Healey and Janet Ilieva

8 9

Interests Others

B MONEY INCOMES FROM SALE OF GOODS AND SERVICES AMONG THE POPULATION

1 2

Incomes from sale of goods Incomes from sale of services

C OTHER INCOMES AND EXPENDITURES

1 2 3

Money transfers inside and outside the country Money transfers among the regions in the country Decrease in households’ smallholdings

II Money expenditures A MONEY EXPENDITURES PERFORMED BY STATE AND CO-OPERATIVE INSTITUTIONS AND SAVINGS

1

2 3 4 5 6 7 8 9

Purchase of goods • Purchase of goods and services in state and co-operative trade • Purchase of goods by agricultural co-operatives • Purchase of houses Purchase of services Taxes Social/pension insurance and other insurance Payments to public and co-operative organisations Debt service Interest payments and ﬁne Others Deposit increase

B MONEY EXPENDITURES ON PURCHASE OF GOODS AND SERVICES AMONG POPULATION

1 2

Expenditures on purchase of goods Expenditures on service provision

C OTHER INCOMES AND EXPENDITURES

1 2 3

Money transfers inside and outside the country Money transfers among the regions in the country Decrease in households’ smallholdings

Central banking in a command economy 65

Appendix B Cash plan structure A Income 1 2 3 4 5 6 7 8 9 10 11

Turnover Transport Public services Culture, health care and sports activities Taxes, ﬁnes and insurance Post Ofﬁces24 National Savings Institution Wages paid back to the budget Sold foreign currency Other incomes Issue result: release of money on circulation

B Payments 1 2 3 4 5 6 7 8 9 10 11 12

Current year wages Previous year wages Funding, insurance, compensation and stipends Purchase of agricultural products Post Ofﬁces National Saving Institution Purchase of foreign currency Business mobility Awards Industrial, co-operative and other expenditures Other payments Issue result: drain of money in circulation

66

Nigel M. Healey and Janet Ilieva

Appendix C25 Table 2A.1 Elite occupational groups in the USSR in 1970a Thousands Party ofﬁcials State, Komsomol and trade union ofﬁcials The intelligentsiab Enterprise managers The military, police, diplomatic service Total

%

95 60 43 22 30

38 24 17 9 12

250

100

Source: M. Mathews (1975) ‘Top Incomes in the USSR: Towards a Deﬁnition of the Soviet Elite’, Survey, 21, p. 13. Notes a Persons earning 450 roubles a month or more and having access to substantial non-cash beneﬁts. (Average wages in the USSR in 1970 were 122 roubles per month.) b i.e. academics, heads of higher educational institutions, institutes, faculties and laboratories; head doctors; senior legal ofﬁcials; editors and senior journalists; leaders in arts and artistic bureaucracy.

Notes 1 Lenin, V.I. (1917) Selected Works, Moscow: Progress Publishers, 1967, p. 222. The following principles listed after banks’ nationalisation are as follows: 2 Nationalisation of syndicates. 3 Abolition of commercial secrecy. 4 Compulsory syndication (i.e. compulsory amalgamation into associations) of industrialists, merchants and employers generally. 5 Compulsory organisation of the population into consumers’ societies, or encouragement of such organisation and the exercise of control over it. 2 3 4 5 6 7 8 9 10 11 12 13

Ibid. Ibid. Ibid. Lenin, V.I. (1916) Imperialism, the Highest Stage of Capitalism, Moscow: Progress Publishers, 1982, p. 30. Ibid., pp. 59, 62. Marx, Karl [1885] Capital, Volume III, Part II, Moscow: Progress Publishers, 1967. Lenin, V.I. (1917) Selected Works, Moscow: Progress Publishers, 1967, p. 344. ‘Finansi i kredit na sotsializma’, Georgi Bakalov, Varna, 1989 (Finance and Credit of the Socialism). In the original table this was labelled ‘Popular Savings Banks’ but they were nationalised in the 1950s and gave rise to the National Savings Bank. Handles non-commercial transfers of foreign exchange and other assets. Handles transactions between GDR and FRG. Engaged in savings banking, insurance, and exercises ﬁnancial control over cooperatives.

Central banking in a command economy 67 14 There were about twenty other banks engaged in foreign exchange transactions. 15 Kornai, J. (1992) The Socialist System, Oxford: Clarendon Press, p. 132. 16 Kornai, J. (1992) The Socialist System, Oxford: Clarendon Press, p. 545. 17 The Council of Mutual Economic Assistance was set up in 1949 and lasted until 1991. The basic objective of the integration was the promotion of socialist collaboration between the country members and to oppose economies of the capitalist countries. Originally it was formed among the European socialist countries (Bulgaria, Czechoslovakia, German Democratic Republic, Hungary, Poland, Romania and Soviet Union; Yugoslavia and Albania had been members for a certain period of time). Later on Cuba, Mongolia and Vietnam joined. 18 Elman, M. (1989) Socialist Planning, Cambridge: Cambridge University Press, p. 277. 19 This method was approved at the ninth session of CMEA in 1958. 20 The coefﬁcient in 1962 was estimated to be 3.4 non-commercial roubles for 1 transferable rouble, in 1971 it was 2.3, in the late 1970s it was 1.9, and in the late 1980s 1.70, revealing an increase in the domestic price levels of CMEA countries. (Source: , 1989./Finance and Credit of the Socialism/.) 21 Kornai, J. (1992) The Socialist System, Oxford: Clarendon Press, p. 262. 22 The role of total public output (part of which, the national income) equals that of GDP, but at the same time they are quite different, as the technique measuring them is different. For instance, total public output does not include the provided services, and that is why they were excluded; hence the comparison between periods before and after the start of the reforms would be inconsistent. 23 Zhivkov, T. (1983) Lecture to students from the Academy for Public Sciences and Social Management in Varna. 24 Post Ofﬁce branches in the rural areas of some of the former socialist countries played the role of both: a Post Ofﬁce and National Savings Institution branch for simple operations such as deposit collection, money transfers and so on. 25 Ellman, M. (1989) Socialist Planning, Cambridge: Cambridge University Press, p. 218.

References Ellman, M. (1989) Socialist Planning, Cambridge: Cambridge University Press, 218. Kornai, J. (1992) ‘The Socialist System’, Oxford: Clarendon Press. Lenin, V.I. (1967) Selected Works, Moscow: Progress Publishers (First published in 1917). Lenin, V.I. (1982) Imperialism, the Highest Stage of Capitalism, Moscow: Progress Publishers (First published in 1916). Marx, K. (1967) Capital, Moscow: Progress Publishers (First published in 1885). . ‘Finansi i kredit na sotsializma’, Varna: Georgi Bakalov (Finance and Credit of the Socialism, Collected Works). Zhivkov, T. (1983) Lecture to students from the Academy for Public Sciences and Social Management in Varna.

3

Central banking in transition economies Nigel M. Healey and Janet Ilieva

1 Introduction Over the 1990s, numerous countries have provided their central banks with greater legal independence from government. This trend to increased central bank independence has been witnessed both in industrialised and developing countries (in particular, Latin America). Within the former socialist countries of Central and Eastern Europe (CEE), economic transition has also been accompanied by substantial central bank reforms granting autonomy in the setting and execution of monetary policy. As Debelle and Fischer (1995) note, there are three broad reasons for the establishment of more independent central banks. First, central bank independence is justiﬁed theoretically by the existence of incentive constraints for politicians, which force dependent central banks to realise sub-optimally high rates of inﬂation. Since the seminal study by Kydland and Prescott (1997), this inﬂation bias has been discussed primarily in terms of the ‘credibility’ problems of monetary policy caused by dynamic inconsistency (e.g. Barro and Gordon, 1983). In addition, the political independence of the central bank is aimed at protecting against partisan or electoral cycles in monetary policy which, in combination with ex-ante uncertain election outcomes, may in turn lead to suboptimally high output volatility and contribute to a higher inﬂation bias (e.g. Alesina, 1989; Grilli et al., 1991). Second, a number of empirical studies claim that, in industrial countries, legal measures of central bank independence are inversely related to the average inﬂation (e.g. Bade and Parkin, 1988; Alesina and Summers, 1993). In contrast, there appears to be no systematic relationship between central bank independence and economic growth (Alesina and Gatti, 1995; see also Rogoff, 1985). In the case of developing countries, it appears that (in contrast to developed countries) legal independence of central banks is not a good measure of actual central bank independence. The negative correlation between central bank independence and inﬂation is conﬁrmed, however, if so-called ‘behavioural’ indices, such as the actual turnover rates of central bank governors or the political vulnerability of

Central banking in transition economies 69 the central bank governor, are used to measure the degree of central bank independence (Cukierman et al., 1992; Cukierman and Webb, 1995; Cukierman, 1996). Third, the inﬂationary track record of the Deutsche Bundesbank, which is regarded as a particularly independent central bank, is widely used as an exemplar of the beneﬁts of giving central banks autonomy from governments. These profound changes in attitudes to the role and functioning of central banks have coincided with the demise of central planning in Central and Eastern Europe. Since 1989, the marketisation and structural transformation of these economies has placed new demands on their central banks and led to experimentation with a variety of institutional frameworks for monetary control. The Baltic States, for example, have established currency boards, successfully stabilising their exchange rates and bringing inﬂation under control. In the four major states of CEE – Poland, the Czech Republic, Romania and Bulgaria (which has also adopted a currency board since mid-1997) – the central bank has gained sufﬁcient autonomy to achieve a substantial reduction in inﬂation. In contrast, in many countries in the Commonwealth of Independent States (CIS), the central banks continue to monetise large state deﬁcits, fuelling inﬂation (see Loungani and Sheets (1997) for an overview). This book addresses this central issue in contemporary monetary policy facing the transition economies of Central and Eastern Europe. Chapter 5 considers the changing role of central banks in market economies over time to provide a longer term perspective. Chapters 3 and 4 review the theoretical developments in central bank independence, and the empirical link between independence and macroeconomic performance. Chapter 6 provides an overview of banking in a centrally planned economy, while Chapters 7–10 are case studies of central bank reform, and the moves to greater central bank autonomy, in the Czech Republic, Poland, Romania and Bulgaria respectively. The reform of central banking in CEE has taken place against the background of major structural reform and economic dislocation. This introduction sets the changes in central banking in a broader macroeconomic and institutional context. It provides an overview of the wider challenges facing the transition economies and the factors which have inﬂuenced the pace and shape of monetary policy-making across the region. After a brief review of the size and economic structure of the twenty-eight ‘transition economies’, this introduction outlines the basic operation of centrally planned economies, highlighting the scale of the structural changes entailed by ‘marketisation’. It documents the early phase of the transition process, when all the transition economies experienced a slump in output and hyperinﬂation following the breakdown of central planning and the liberalisation of prices. The chapter then considers the main macroeconomic challenges involved in restoring stability and the microeconomic (or

70

Nigel M. Healey and Janet Ilieva

‘supply-side’) challenges of privatisation, market liberalisation and legal reform. It ﬁnally assesses the progress of different countries over the ten years since the process began in 1989 and considers the outlook for monetary policy-making in the region.

2 Central and Eastern Europe versus the Commonwealth of Independent States The term ‘communist bloc’ was formerly used in the West to describe the countries that constituted the Soviet sphere of inﬂuence, implying a high degree of homogeneity between the twenty-eight independent states that have subsequently emerged. In reality, the ‘transition’ states vary considerably, in terms of living standards (as measured by, for example, per capita GDP), economic structure, population size, history, religion and culture. Table 3.1 gives some idea of the differences between the transition states.1 It shows the contrast in size, from Russia with a population of 147m to Estonia with 1.5m. It highlights the vast gulf in relative living standards, from Slovenia with a per capita income (calculated in terms of purchasing power parity) of $9826 to Tajikistan with just $200. The importance of agriculture ranges from as little as 4 to 6 per cent of GDP in Slovenia, the Czech and Slovak Republics, Hungary and Poland, to 43 per cent in Kyrgyzstan. Table 3.1 also shows that despite the differences between individual states, it is possible to draw a broad distinction between the CEE states, here deﬁned to include the three Baltic States of the former Soviet Union, and the twelve states of the Commonwealth of Independent States (CIS). Taken as a group, the CEE states generally have a higher per capita GDP, a lesser reliance on agriculture and, reﬂecting the labour shedding which is an inevitable consequence of enterprise restructuring, much higher rates of unemployment.

3 The operation of the command economy Despite the differences between the CEE and the CIS states, both groups started the transition process with an all-pervasive central planning system, often termed a ‘command economy’. The command economy dominated every aspect of life, instructing factories where to buy their inputs, how much to pay their workers, how much to produce and where to sell their output; individuals were trained in specialist schools and universities, and were directed to work at speciﬁc factories, which provided their wages, houses and health care; the national plan informed the banking system how much to lend to which factories, how much cash to print to pay wages and so on. As a theoretical concept, central planning was very elegant. Using ‘input–output’ analysis (a planning framework which calculated the inputs required for each factory in order for it to deliver its planned outputs to

Central banking in transition economies 71 Table 3.1 An overview of the transition economies Population, Per capita millions GDP, (1998) $PPP (1998) CEE countries: Albania Bulgaria Croatia Czech Republic Estonia FYR Macedonia Hungary Latvia Lithuania Poland Romania Slovak Republic Slovenia CIS countries: Armenia Azerbaijan Belarus Georgia Kazakhstan Kyrgyzstan Moldova Russia Tajikistan Turkmenistan Ukraine Uzbekistan

Agriculture Industry as Unemployas % GDP % GDP ment (%) (1998) (1998) (1998)

3.2 8.3 4.53 10.3 1.45 2.0 10.1 2.4 3.7 38.7 22.3 5.4 2.0

929.6 1315 4820 5479 3593 1548 4730 2622 2890 3887 1695 3793 9779

62.6* 18.7 8.9 5.0 5.6 10.8* 5.8* 4.5 10.1 5.7* 16.0 4.4 3.8

12.2** 25.5 25.4 36.6 18.1 19* 25.0* 24.3 23.6 28.1 31.7 26.7 27.5

17.7 12.2 17.2 7.5 9.6 34.5 9.1 13.8 6.4 10.4 10.3 11.9 7.9

3.7 7.62 10.2 5.4 15.0 4.8 4.3 146.7 6.1 4.8 50.1 24.0

510 540 1396 967 1493 354.8 432.4 1867 213 421 846 591

34.2 20.0* 12.7 26.1 8.8 43.4* 30.0* 6.5 8.8 19.8* 11.4 29.0*

24.0 24.8* 38.4 9.6 22.5 15.5* 29.0* 42.7 45.7 44.3* 24.6 19.0*

9.3 19.3* 2.3 14.0 3.7 3.2* 1.6* 12.4 3.1 n.a. 3.7 0.6

Source: EBRD (1999). Notes * ⫽ 1997 ﬁgure, ** ⫽ 1996 ﬁgure.

the next stage in the production process), the planning ministry could calculate precisely how much labour, capital and raw materials each enterprise required to achieve its production targets. The various production targets for raw materials and intermediate and ﬁnal products all ﬁtted together to ensure a perfectly balanced expansion of the economy. Input and output prices were set precisely to ensure that all ﬁrms could pay their wage bills and repay loans from the national bank, while at the same time pricing consumer products to encourage consumption of socially desirable goods (e.g. books, ballet, theatre, public transport) and discourage consumption of politically unfavoured goods (e.g. international telephone calls, cars, luxury goods).

72

Nigel M. Healey and Janet Ilieva

The overall national plan was thus internally consistent. If each of the enterprises achieved its production targets, there could not be, by deﬁnition, shortages or bottlenecks in the economy. There would be full employment, with everyone working in an enterprise for which he/she had been speciﬁcally trained at school and university. The total wage bill for the economy, which was paid in cash, would be sufﬁcient to buy all the consumer goods produced. There would be zero inﬂation, and all the country’s citizens would have access to housing, education and health care. While central planning ultimately failed in CEE and the CIS, the immanent causes remain a matter of controversy. For present purposes it is sufﬁcient to note that, whatever the triggering factor, the precipitous collapse of the Soviet planning system in the period between 1989 and 1991 left the transition economies ill-prepared to manage in an environment of rapid structural change, social upheaval and reintegration into the global economy.

4 The early stages of the transition process The decades of central planning meant that the transition economies began the reform process without functioning markets for labour, consumer goods or capital. All three had been allocated by the central planning agency in accordance with an internally coherent economic plan. Because all the means of production had previously been in state hands, there was, moreover, no legislative framework for the enforcement of property rights, the valuation and disposal of assets or the liquidation of unproﬁtable enterprises. Nor, in a system of directed labour, was there any ofﬁcial unemployment and, hence, no need for a social security system. As noted above, social welfare was provided by enterprises, with the central planners setting their aggregate employment equal to the size of the labour force, and services such as education, health care and housing being provided by the enterprises to their employees. Banks in the central planning system were bookkeeping operations, allocating credit according to the demands of the plan rather than objective risk assessments, and capital markets were non-existent. The challenge for the transition states at the end of the 1980s was to build a modern market economy on these foundations. The early 1990s ushered in a period of economic crisis in the transition economies of CEE and the CIS. The breakdown of central plan discipline and the collapse of COMECON trade, taken together with the highly concentrated, vertically integrated nature of production (particularly in the CIS), led to an early collapse in production across the region (Table 3.2). Recovery began ﬁrst in CEE, with all countries except Bulgaria showing positive growth in 1996. Nevertheless, only in Poland has recent growth been sufﬁcient to regain the losses of the early years: in all other states, output in 1998 remained at or below the level in 1989. Recovery in the CIS

Source: EBRD (1999).

CEE countries: Albania Bulgaria Croatia Czech Republic Estonia FYR Macedonia Hungary Latvia Lithuania Poland Romania Slovak Republic Slovenia All CEE CIS countries: Armenia Azerbaijan Belarus Georgia Kazakhstan Kyrgyzstan Moldova Russia Tajikistan Turkmenistan Ukraine Uzbekistan All CI* ⫺7.2 ⫺7.3 ⫺11.7 ⫺3.3 ⫺14.2 ⫺8.0 ⫺3.1 ⫺34.9 ⫺21.3 2.6 ⫺8.8 ⫺6.5 ⫺5.5 ⫺3.6 ⫺52.6 ⫺22.6 ⫺9.6 ⫺44.8 ⫺2.9 ⫺19.0 ⫺29.1 ⫺14.5 ⫺29.0 ⫺5.3 ⫺13.7 ⫺11.1 ⫺14.1

⫺17.1 ⫺0.7 ⫺1.2 ⫺20.6 ⫺13.0 ⫺5.0 ⫺17.5 ⫺5.0 ⫺7.1 ⫺4.7 ⫺11.6 ⫺0.5 ⫺12.0

1992

⫺28.0 ⫺12.0 ⫺21.0 ⫺11.5 ⫺13.6 ⫺7.0 ⫺11.9 ⫺10.4 ⫺6.2 ⫺7.0 ⫺12.9 ⫺14.6 ⫺8.9 ⫺10.7

1991

⫺14.8 ⫺23.1 ⫺7.6 ⫺25.4 ⫺9.2 ⫺16.0 ⫺1.2 ⫺8.7 ⫺11.0 ⫺10.0 ⫺14.2 ⫺2.3 ⫺9.3

9.6 ⫺1.5 ⫺8.0 0.6 ⫺9.0 ⫺9.1 ⫺0.6 ⫺14.9 ⫺16.0 3.8 1.5 ⫺3.7 2.8 0.4

1993

5.4 ⫺19.7 ⫺12.6 ⫺11.4 ⫺12.6 ⫺20.0 ⫺31.2 ⫺12.7 ⫺18.9 ⫺18.8 ⫺23.0 ⫺4.2 ⫺14.0

9.4 1.8 5.9 3.2 ⫺2.0 ⫺1.8 2.9 0.6 ⫺9.5 5.2 3.9 4.9 5.3 3.9

1994

6.9 ⫺11.8 ⫺10.4 2.4 ⫺8.2 ⫺5.4 ⫺3.0 ⫺4.1 ⫺12.5 ⫺8.2 ⫺12.2 ⫺0.9 ⫺5.1

8.9 2.1 6.8 6.4 4.3 ⫺1.2 1.5 ⫺0.8 3.5 7.0 7.1 6.9 4.1 5.5

1995

5.8 1.3 2.8 10.5 0.5 7.1 ⫺8.0 ⫺3.5 ⫺4.4 ⫺8.0 ⫺10.0 1.6 ⫺3.5

9.1 ⫺10.1 6.0 3.8 3.9 0.8 1.3 3.3 4.9 6.1 4.1 6.6 3.5 4.0

1996

Table 3.2 Recession and recovery in transition economies (% annual change in GDP)

3.1 5.8 10.4 11.0 2.0 9.9 1.3 0.8 1.7 ⫺26.1 ⫺3.2 2.4 0.9

⫺7.0 ⫺7.0 6.5 0.3 10.6 1.5 4.6 8.6 7.4 6.9 ⫺6.9 6.5 4.6 3.6

1997

7.2 10.1 8.3 2.9 ⫺2.5 1.8 ⫺8.6 ⫺4.6 5.3 4.2 ⫺1.7 3.3 ⫺3.5

8.0 3.5 2.3 ⫺2.3 4.0 2.9 5.1 3.6 5.2 4.8 ⫺7.3 4.4 3.9 2.4

1998

4.0 3.7 1.5 3.0 ⫺1.7 0.0 ⫺5.0 0.0 5.0 17.0 ⫺2.5 3.0

8.0 0.0 ⫺0.5 0.0 0.0 0.0 3.0 1.5 0.0 3.5 ⫺4.0 1.8 3.5 1.6

1999

41 44 78 33 61 60 32 55 42 44 37 90 55

86 66 78 95 76 72 95 59 65 117 76 100 104 95

1998 1989 ⫽ 100

74

Nigel M. Healey and Janet Ilieva

countries started later and has been more hesitant. For the CIS countries as a group, output in 1998 remained at only 55 per cent of its 1989 level. With national and regional governments heavily reliant on ‘proﬁt’ and ‘turnover’ taxes on enterprises for state revenues, the early recession led to a sharp contraction of the tax base. In the absence of developed capital markets, which in market economies offer governments the opportunity of selling securities to ﬁnance themselves in a non-inﬂationary way, the resulting budget deﬁcits were ﬁnanced by printing cash through the central bank. Inﬂationary pressures were compounded further in many transition economies by the state’s attempts to avoid wholesale job losses by granting heavy subsidies to loss-making enterprises. Whether these were provided by the state directly, thereby adding to its budget deﬁcit, or in the form of ‘soft’ loans from the central bank, the result in both cases was to fuel further the rate of growth of the money supply. The slump in output was thus accompanied by hyperinﬂation in many transition states, as price liberalisation brought repressed inﬂation out into the open. Table 3.3 shows the rates of inﬂation experienced by countries in the region in recent years. Clearly, the challenges facing the transition economies at the start of the process were profound, and had both macroeconomic and microeconomic dimensions. At the macroeconomic level, the main priority was to construct new instruments of ﬁscal and monetary policy and thereby stabilise economies suffering from collapsing output and spiralling prices. Under the former system, the central planners could set both production levels and ﬁnal producer prices directly through the national plan. In the new emerging market economies, they needed to construct the tools to achieve these goals at the macroeconomic level through indirect means, by changing taxes and government spending and by altering interest rates and managing the exchange rate. At the microeconomic level, the priority was to transform an economy of large, state-owned enterprises (SOEs) into competitive, private companies capable of trading globally and to generate an enterprise culture where new ﬁrms would emerge to exploit proﬁtable opportunities.

5 Macroeconomic reform Governments in market economies use two main sets of policy instruments to indirectly inﬂuence economic activity: ﬁscal policy – the use of taxes and public spending; and monetary policy – the control of interest rates (and the exchange rate). Neither set of policy instruments exists in the same form in a command economy. In the central planning system, taxes were levied on SOEs, normally in the form of ‘turnover taxes’ (i.e. taxes on the enterprises’ sales revenue or turnover), and on individuals in the form of income tax. These taxes ensured that the government could, as in a market economy, ﬁnance the army and other executive functions (e.g.

25.0 126.0 93.0 131.0 136.8 170.0 151.0 161.0 204.0 155.0 161.0 169.0

CIS countries: Armenia Azerbaijan Belarus Georgia Kazakhstan Kyrgystan Moldova Russia Tajikistan Turkmenistan Ukraine Uzbekistan

Source: EBRD (1999).

104.0 338.9 249.0 52.0 303.8 229.0 32.2 262.4 345.0 60.0 222.8 58.3 247.0

CEE countries: Albania Bulgaria Croatia Czech Republic Estonia FYR Macedonia Hungary Latvia Lithuania Poland Romania Slovak Republic Slovenia

1991

1,341.0 1,395.0 1,559.0 1,176.9 2,984.1 1,259.0 2,198.0 2,506.1 1,364.0 644.0 2,730.0 910.0

236.6 79.4 938.2 12.7 953.5 1,935.0 21.6 959.0 1,161.1 44.3 199.2 9.1 92.9

1992

10,896.0 1,294.0 1,996.0 7,487.9 2,169.0 1,363.0 837.0 840.0 7,343.7 9,750.0 10,155.0 885.0

30.9 63.8 1,149.0 18.2 35.6 241.8 21.1 35.0 188.8 37.6 295.5 25.1 22.8

1993

1,885.0 1,788.0 1,960.0 6,474.4 1,160.0 95.7 116.0 204.4 1.1 1,328.0 401.0 1,281.0

15.8 121.9 ⫺3.0 9.7 42.0 55.0 21.2 26.0 45.0 29.4 61.7 11.7 19.5

1994

31.9 84.5 244.0 57.4 60.4 31.9 23.8 128.6 2,133.3 1,262.0 181.0 117.0

6.0 32.9 3.8 7.9 29.0 9.0 28.3 23.1 35.5 21.6 27.8 7.2 9.0

1995

Table 3.3 The inﬂationary price of transition (% annual change in retail prices, end year)

5.8 6.5 39.3 14.3 28.6 35.0 15.1 21.8 40.5 446.0 39.7 64.0

17.4 310.8 3.4 8.6 15.0 ⫺0.6 19.8 13.1 13.1 18.5 56.9 5.4 9.0

1996

21.8 0.3 63.4 7.2 11.3 14.7 11.2 10.9 163.6 21.5 10.1 50.0

42.1 578.6 3.8 10.0 12.5 2.6 18.4 7.0 8.5 13.2 151.4 6.4 8.8

1997

⫺1.3 ⫺7.6 181.7 10.7 1.9 18.3 18.2 84.5 2.7 19.8 20.0 26.0

8.7 1.0 5.4 6.8 4.4 ⫺3.1 10.3 2.8 2.4 8.6 40.6 5.6 6.5

1998

8.0 2.0 155.0 9.5 19.6 40.0 30.0 45.0 55.0 40.0 17.0 42.0

3.5 3.1 2.0 8.0 2.1 2.5 6.5 40.0 14.5 6.5

2.0 2.0

1999*

76

Nigel M. Healey and Janet Ilieva

law and order) that did not fund their activities by generating their own sales revenues. These taxes and associated government spending, like the conﬁguration of input and output prices, were an integral part of the national plan and were set to ensure its internal consistency. Signiﬁcantly, many social services that are provided by the state in market economies such as public housing, public transport, pre-school education, vocational training and health care were often provided by SOEs to their employees and families, and ﬁnanced out of the enterprises’ operating surpluses. As a consequence, the tax base at the start of the process in most transition economies was small relative to market economies (since so many social services were ﬁnanced directly by SOEs), and tax and spending decisions were part of the broader process of national planning. In contrast, in a market economy, governments need to raise taxes to ﬁnance the whole range of public goods and social services, at the same time setting the balance between taxes and public spending (i.e. the budget deﬁcit or surplus) in order to stabilise aggregate demand and output. The ﬁnancial system in command economies is also different from the system that operates in a market economy. Typically, command economies have a uniﬁed ﬁnancial system, with different branches of the national bank acting as a bookkeeper for transactions between companies in different sectors and providing deposit, but not lending, facilities (through a national savings branch) to the general population. Credit was allocated by the national bank’s sectoral branches according to the requirements of the national plan. Interest rates played no part in the allocation of, or demand for, credit. Lending facilities for retail customers were unknown. Domestic currencies were not freely convertible, but could (with ofﬁcial approval) be exchanged for foreign currencies through the national bank at artiﬁcial exchange rates depending upon the purpose (a favourable rate for exporters, a higher, penal rate for importers). There were no ﬁnancial markets of the sort recognisable in market economies. Because the creation of ﬁnancial markets which can price risk and allocate savings to projects and companies with the best, risk-adjusted returns is essential to the transition process (see below), microeconomic reform fundamentally alters the role of the national bank. In the presence of ﬁnancial markets made by private banks and ﬁnancial institutions and used to channel savings to private companies, the interest rate (and the exchange rate) become powerful tools for inﬂuencing saving, borrowing and spending in the economy as a whole. Fiscal policy reform Because the system of taxes and government spending had been constructed around the needs of central planning, the ﬁscal policy instruments at the disposal of governments in the early stages of the transition process were ill suited to dealing with the emerging market economies. Fiscal

Central banking in transition economies 77 policy was beset by two overriding problems. First, the tax base was small and not set up to deal with the tax affairs of small, private companies. As the central plan collapsed and SOEs output and turnover slumped, the revenue from turnover taxes fell sharply. In the economic chaos that followed, many SOEs switched to bartering their outputs for inputs. Other proﬁtable SOEs took advantage of the confusion and simply stopped paying taxes altogether. Attempts to bolster tax revenues by introducing new tax regimes for de novo private companies often foundered because tax rates were set so high that companies evaded them. The old tax systems in many countries proved unable to deal with the changing structure of economic activity, and there was a slump in tax revenues as more and more economic activity shifted into the ‘informal’ or ‘hidden’ economy (see Table 3.4). At the same time, the pressure on government spending increased. Many SOEs quickly found themselves in genuine ﬁnancial distress, unable to pay their workforces or continue to fund the social services they had traditionally provided. Many of the associated ‘social assets’ (public housing, schools, hospitals, public transport systems) were taken into local government control. With widespread popular resistance to attempts at levying user charges which more closely approximated the true opportunity costs of providing these services, this ‘nationalisation’ of social services, particularly in the former Soviet Union, added to government spending. Table 3.4 Share of the unofﬁcial economy in GDP 1989

1990

1991

1992

1993

1994

1995

CEE countries: Bulgaria Czech Republic Estonia Hungary Latvia Lithuania Poland Romania Slovak Republic

22.8 6.0 12.0 27.0 12.0 12.0 15.7 22.3 6.0

25.1 6.7 19.9 28.0 12.8 11.3 19.6 13.7 7.7

23.9 12.9 26.2 32.9 19.0 21.8 23.5 15.7 15.1

25.0 16.9 25.4 30.6 34.3 39.2 19.7 18.0 17.6

29.9 16.9 24.1 28.5 31.0 31.7 18.5 16.4 16.2

29.1 17.6 25.1 27.7 34.2 28.7 15.2 17.4 14.6

36.2 11.3 11.8 29.0 35.3 21.6 12.6 19.1 5.8

CIS countries: Azerbaijan Belarus Georgia Kazakhstan Moldova Russia Ukraine Uzbekistan

12.2 12.0 12.0 12.0 12.0 12.0 12.0 12.0

21.9 15.4 24.9 17.0 18.1 14.7 16.3 11.4

22.7 16.6 36.0 19.7 27.1 23.5 25.6 7.8

39.2 13.2 52.3 24.9 37.3 32.8 33.6 11.7

51.2 11.0 61.0 27.2 34.0 36.7 38.0 10.1

58.0 18.9 63.5 34.1 39.7 40.3 45.7 9.5

60.6 19.3 62.6 34.3 35.7 41.6 48.9 6.5

Source: Johnson et al. (1997).

78

Nigel M. Healey and Janet Ilieva

Even more signiﬁcantly, there was intense political pressure on governments to ‘bail out’ failing SOEs, by extending to them state subsidies to allow them to pay wages. The fear of social unrest, prompted by episodic strikes and demonstrations by powerful groups of workers (e.g. miners, shipyard workers), encouraged many transition governments to give large subsidies to their SOEs. In the late 1990s, as Table 3.5 shows, the ﬁscal situation in most countries was starting to stabilise as new, market-oriented taxes begin to take effect. Budget deﬁcits have gradually declined, particularly in CEE, to levels common in advanced market economies. Extending the tax base has not been easy, however, as companies used to trading in the informal economy have sought to resist the introduction of new company taxes and systems of value-added taxes (VAT). Table 3.5 General government budget balances (% of GDP) 1992

1993

1994

1995

1996

1997

1998

1999*

CEE countries: Albania Bulgaria Croatia Czech Republic Estonia FYR Macedonia Hungary Latvia Lithuania Poland Romania Slovak Republic Slovenia CEE and Baltic States

⫺20.3 ⫺14.4 ⫺12.4 ⫺10.3 ⫺12.1 ⫺12.6 ⫺10.4 ⫺13.8 ⫺5.2 ⫺10.9 ⫺5.8 ⫺6.4 ⫺10.4 ⫺3.0 1.0 ⫺2.8 ⫺3.9 ⫺0.8 1.6 ⫺0.9 ⫺0.4 ⫺1.3 0.6 ⫺3.6 ⫺3.1 0.5 ⫺1.1 ⫺1.8 ⫺1.1 ⫺2.1 ⫺2.6 ⫺5.0 ⫺0.3 ⫺0.7 1.3 ⫺1.3 ⫺1.9 2.2 ⫺0.3 ⫺3.0 ⫺9.6 ⫺13.8 ⫺2.9 ⫺1.2 ⫺0.5 ⫺0.4 ⫺1.7 ⫺7.8 ⫺7.2 ⫺6.6 ⫺8.4 ⫺6.4 ⫺3.0 ⫺4.8 ⫺4.8 ⫺4.5 ⫺0.8 0.6 ⫺4.0 ⫺3.9 ⫺1.7 0.1 ⫺0.8 ⫺3.8 0.5 ⫺3.3 ⫺5.5 ⫺4.5 ⫺4.5 ⫺1.8 ⫺5.8 ⫺7.0 ⫺6.7 ⫺3.1 ⫺3.1 ⫺2.8 ⫺3.3 ⫺3.1 ⫺3.0 ⫺3.0 ⫺4.6 ⫺0.4 ⫺1.9 ⫺2.6 ⫺4.0 ⫺3.6 ⫺3.3 ⫺2.7 N/a ⫺7.0 ⫺1.3 0.2 ⫺1.9 ⫺4.4 ⫺5.8 ⫺3.2 0.2 0.1 ⫺0.3 ⫺0.5 ⫺0.2 ⫺1.7 ⫺1.4 ⫺1.0 ⫺5.1 ⫺4.6 ⫺3.4 ⫺3.3 ⫺3.5 ⫺2.8 ⫺3.0 ⫺4.7

CIS countries: Armenia Azerbaijan Belarus Georgia Kazakhstan Kyrgystan Moldova Russia Tajikistan Turkmenistan Ukraine Uzbekistan CIS

⫺13.9 2.8 0.0 ⫺25.4 ⫺7.3 ⫺17.4 ⫺26.2 ⫺42.6 ⫺30.5 13.2 ⫺25.4 ⫺18.4 ⫺17.6

Source: EBRD (1999).

⫺54.7 ⫺10.5 ⫺11.0 ⫺15.3 ⫺12.1 ⫺4.9 ⫺1.9 ⫺2.5 ⫺1.9 ⫺26.2 ⫺7.4 ⫺4.5 ⫺4.1 ⫺7.5 ⫺2.7 ⫺13.5 ⫺11.6 ⫺17.3 ⫺7.4 ⫺8.7 ⫺5.7 ⫺15.9 ⫺9.7 ⫺5.9 ⫺23.4 ⫺5.4 ⫺11.9 ⫺0.5 ⫺1.4 ⫺1.6 ⫺16.2 ⫺9.1 ⫺4.9 ⫺10.4 ⫺6.1 ⫺4.1 ⫺15.8 ⫺7.7 ⫺6.4

⫺9.3 ⫺2.8 ⫺1.6 ⫺4.4 ⫺4.7 ⫺9.5 ⫺6.7 ⫺9.1 ⫺5.8 ⫺0.2 ⫺3.2 ⫺7.3 ⫺5.4

⫺5.9 ⫺1.7 ⫺0.7 ⫺3.8 ⫺6.8 ⫺9.0 ⫺7.5 ⫺8.1 ⫺3.3 0.0 ⫺5.6 ⫺2.3 ⫺4.6

⫺5.2 ⫺4.2 ⫺0.3 ⫺4.4 ⫺8.0 ⫺9.9 ⫺8.1 ⫺5.4 ⫺3.8 ⫺2.7 ⫺2.7 ⫺3.8 ⫺4.9

⫺5.5 ⫺4.0 ⫺3.0 ⫺3.7 ⫺7.0 ⫺7.0 ⫺4.5 ⫺6.0 ⫺3.0 ⫺4.0 ⫺1.5 ⫺3.8 ⫺4.4

Central banking in transition economies 79 Monetary policy reform Chapters 7 to 10 provide a rich insight into the reforms of the banking system in general, and the construction of central banks in particular in CEE, but it is useful at this point to provide a broad overview of the general process. In a command economy, the national bank performs many of the functions of the central bank and the commercial banking system in a market economy, controlling the money supply, taking deposits from the general public and lending money to enterprises. The essential difference is that these activities are all part of a wider national economic plan: money emission, in the form of bank notes, is set via the treasury plan to ensure that enterprises can pay their workers; and a separate credit plan allocates loans to enterprises to support their investment and ﬁnance stocks and work-in-progress. In the former communist countries, the national bank was normally structured into functional areas: the national bank which performed a co-ordinating role and liaised directly with the planning ministry, several sectoral banks (e.g. agriculture, industry) which managed the accounts of enterprises within their designated sector, an investment bank which made loans to enterprises for capital investment, and a national savings bank, which held the deposits of the general public. Under this arrangement there were, in effect, two parallel monetary systems, one cash-based, one credit-based. In the cash economy, the national bank, as part of the planning process, set the volume of cash to be printed within a given time period. The production plan for each enterprise speciﬁed the amount of cash they could draw from their sectoral bank to pay wages; all wages and salaries were paid in cash. The general public then used their cash wages to buy goods and services, and the state shops redeposited their cash takings in the national bank; household savings were deposited in the national savings bank. In the credit economy, the national plan set limits on the amount each enterprise could borrow from its sectoral or investment bank to ﬁnance investment and stocks. Inter-enterprise payments were all made by transfers between their accounts in the national banking system. Credits were not convertible into cash, so, although separate, the two systems were integrated through the national plan. This feature of the command economy gave rise to curious monetary problems. It was possible, for instance, for the two monetary systems to become misaligned. For example, if the national bank increased the cash issue to allow enterprises to pay higher wages, there could be increased demand for goods by households. At the same time there might be restrictions on credit availability so that enterprises could not expand production. Excess demand for consumer goods could co-exist with excess supply of producer goods. The immediate priority for the post-communist countries was to integrate the cash and credit systems and create a ‘two-tier’ ﬁnancial system, in

80

Nigel M. Healey and Janet Ilieva

which the national bank could assume the responsibilities of a conventional central bank, setting interest rates and ensuring the viability of the system as a whole, and a lower tier of commercial banks could take deposits and make loans on market terms. In the former Soviet republics there was an additional decision to be taken: namely whether to remain in monetary union with Russia after independence, or to secede and establish their own currency. In the event, the political difﬁculties of maintaining monetary ties with Russia during the turbulence of the transition process forced the issue, and even those states which might have preferred to remain within the rouble zone introduced their own national currencies. The restructuring of the national bank system was achieved fairly quickly with most countries choosing to privatise the sectoral, investment and savings divisions of the national bank and allowing new joint-stock banks to be created. The core of the national bank which remained in state control was normally reconstituted as a central bank and, in most countries, given a high degree of legal autonomy from the government, along with a mandate to manage monetary conditions in the pursuit of price stability. Indeed, the constitution of the Bundesbank was widely used as a model for designing the new central banks of CEE and the CIS. As Chapters 6 to 10 clearly demonstrate, the constitutions of the new central banks in the major CEE countries were, as measured by legal indices, at least as independent as that of the Bundesbank. However, as Table 3.6 clearly shows, establishing monetary stability has been made more difﬁcult for the new central banks by two main factors: budget deﬁcits; and the dollarisation of the economies. Budget deﬁcits and the money supply In a cash-based economy where the government has the monopoly over currency issue, printing cash provides a straightforward alternative to raising taxes as a means of ﬁnancing government spending. In the early years of the reform process, money emission was the primary way in which governments ﬁnanced the growing gap between government spending and tax revenues. There are essentially three ways of ﬁnancing government spending: 1 2 3

taxation; printing currency or borrowing from the banking system – which increases the money supply; borrowing from the general public or abroad – which increases government debt.

Transition governments have sought to rebuild their tax bases with varying degrees of success and, by Western standards, budgets deﬁcits in CEE and the CIS states are generally modest. Nevertheless, monetary ﬁnancing of the residual budget deﬁcits remains a problem. The main reason is that, in

Central banking in transition economies 81 Table 3.6 Broad money growth (% change p.a.) 1991

1992

1993

1994

CEE countries: Albania Bulgaria Croatia* Czech Republic Estonia FYR Macedonia Hungary Latvia Lithuania** Poland Romania Slovak Republic** Slovenia

104.4 110.0 n.a. 26.8 n.a. n.a. 35.7 153.0 143.0 37.0 101.2 n.a. n.a.

152.7 53.6 n.a. 20.7 59.0 n.a. 27.6 169.9 245.3 57.5 79.6 n.a. 131.6

CIS countries: Armenia Azerbaijan Belarus Georgia Kazakhstan Kyrgystan Moldova Russia Tajikistan Turkmenistan Ukraine Uzbekistan

n.a. n.a. n.a. n.a. 210.7 84.0 n.a. 125.9 68.0 n.a. n.a. n.a.

n.a. 1,077.0 729.0 n.a. 812.2 1,114.0 510 953.7 1,851.7 464.0 4,319.0 2,229.0 391.0 692.0 576.0 428.0 180.0 125.0 361.7 320.2 115.7 642.6 416.1 166.4 579.0 1,429.0 159.0 n.a. n.a. 984.0 n.a. 758.0 540.0 468.0 784.0 680.3

75.0 47.6 n.a. 22.5 93.0 n.a. 15.7 84.1 100.2 36.0 141.0 16.8 64.2

1995

40.6 51.4 78.6 39.6 111.9 24.6 20.8 19.4 40.1 34.5 n.a. 6.4 13.0 20.1 47.7 ⫺23.1 63.0 28.9 38.2 34.9 138.1 71.6 20.1 19.2 50.8 32.2 62.5 24.0 164 135.2 106.1 76.7 65.2 125.8 471.0 454.0 116.0 144.0

1996

1997

1998

40.0 124.5 37.9 7.8 35.6 0.1 22.5 19.9 ⫺3.5 29.4 66.0 16.5 19.4

28.5 359.3 20.9 8.7 42.3 17.6 19.4 38.7 34.1 29.1 75.9 9.1 23.8

20.6 9.6 ⫺1.5 5.2 0.1 14.9 15.5 5.9 14.5 25.0 48.9 2.6 20.9

35.1 18.9 52.4 41.9 13.8 22.9 15.3 30.6 93.2 413.0 35.0 113.0

29.1 33.6 111.4 45.5 27.6 24.7 34.0 28.4 110.7 82.0 34.0 36.0

37.0 ⫺9.7 276.0 ⫺1.2 ⫺13.1 17.0 ⫺8.7 19.9 9.5 50.0 25.0 28.0

Source: EBRD (1999). Notes * ⫽ M1, ** ⫽ M2.

Western countries, budget deﬁcits are normally ﬁnanced by borrowing from the general public through selling bonds on the stock market – and stock markets remain generally underdeveloped in many transition economies. The other alternative, borrowing from abroad, also has its drawbacks. While the IMF and the World Bank have been prepared to extend large loans to transition governments to assist them through the early stages, these loans are highly conditional upon the governments adhering to pre-agreed policies, restricting those governments’ subsequent room for policy manoeuvre. Unlike sales of bonds to the general public, which are normally in the local currency, governments borrow from the international agencies in foreign currency (usually US dollars) and this must be repaid, imposing a burden in terms of interest and capital payments (Table 3.7).

82

Nigel M. Healey and Janet Ilieva

Table 3.7 External debt and servicing (1998) External debt stock ($m) CEE countries: Albania Bulgaria Croatia Czech Republic Estonia FYR Macedonia Hungary Latvia Lithuania Poland Romania Slovak Republic Slovenia CIS countries: Armenia Azerbaijan Belarus Georgia Kazakhstan Kyrgyzstan Moldova Russia Tajikistan Turkmenistan Ukraine Uzbekistan

External debt/GDP

External debt/exports

874 10,101 8,197 2,3600 2,900 1,400 26,700 3,043 3,726 45,000 9,611 11,902 4,959

29.4 92.6 37.5 41.7 55.8 39.5 55.9 47.6 34.8 29.9 25.2 58.5 25.4

424.3 235.3 178.0 89.3 107.8 105.9 128.9 93.0 94.0 148.5 115.8 111.6 54.5

801 684 2,518 1,686 7,860 1,123 1,347 150,900 1,319 1,749 11,700 3,223

42.5 16.6 17.6 32.4 35.1 65.9 71.9 55.1 101.4 85.8 27.6 36.4

363.3 100.9 35.6 352.7 136.1 202.7 207.2 201.9 207.0 248.9 85.4 111.6

Debt service (% exports)

6.1 20.0 11.1 14.5 5.2 9.5 28.4 14.3 21.8 5.9* 23.7 11.6 13.2 12.3 6.7 1.6 14.8 ⫺7.4 8.8 27.6 na 13.3 ⫺82.8 19.4 13.9

Source: EBRD (1998). Note *1997 ﬁgure.

Dollarisation In a market economy, money performs the functions of a medium of exchange, a unit of account and a store of value. The ability of a currency to perform these functions is undermined by high rates of inﬂation. Typically, in many transition economies, US dollars ﬁrst began to circulate as an alternative store of value: individuals would switch their wages from the local currency into dollars and only switch back when they had goods to purchase, preserving the value of their wages in the interim. At higher rates of inﬂation, the ‘menu costs’ of continually altering the prices of goods and services becomes intolerable, and dollars have been used as the

Central banking in transition economies 83 unit of account. Prices in a shop, for example, are shown in dollars and, at the checkout, the till operator converts the prices into local currency at the prevailing exchange rate for payment. Finally, when there is hyperinﬂation, dollarisation often becomes complete, with goods and services being priced and paid for in dollars. It is impossible to estimate accurately the volume of dollars (and other hard currencies, notably the Deutschmark) in circulation in CEE and the CIS, but the sums are huge. The Federal Reserve Bank, for example, estimates that there are more dollars being used in the region than in the US. This phenomenon presents major problems for monetary control, since a signiﬁcant proportion of the ‘money supply’ is unrecorded and outside the central bank’s control. Transition economies which have succeeded in stabilising inﬂation have, however, been able to reverse the dollarisation of their economies. The normal mechanism is that, as inﬂation comes under control and real interest rates for deposits in local currency become positive, it becomes more attractive to hold bank deposits in local currency than to hold dollar bills. Individuals begin to sell their dollars to the central bank and there is a vicious circle as stabilisation leads to an increase in foreign exchange reserves.

The ‘new currency boards’: the case of Estonia In June 1992, Estonia became the ﬁrst of the former Soviet republics to leave the rouble zone and introduce its own national currency, the Kroon. At the same time, Estonia set up a currency board, ﬁxing the Kroon to the Deutschmark. The government’s decision was heavily inﬂuenced by the return of eleven tonnes of gold reserves from abroad, which had been deposited by the Estonian government shortly before the Soviet occupation in 1940. The early repayment of 4.8 tonnes of gold by the Bank of England in March 1992, which had a market value of DM80m, provided the capital for the initial currency issue. Like the old colonial currency boards, the note issue was 100 per cent backed by gold and hard currencies, and the Bank of Estonia was legally obliged to exchange notes for Deutschmarks, and vice versa, at the ofﬁcial rate. The advantage of the system is that, provided the government can obtain the necessary capital, it is simple to set up and operate. Latvia and Lithuania subsequently followed Estonia’s example and set up currency boards of their own, also using gold reserves reclaimed from foreign central banks. In analytical terms, a currency board is equivalent to using a foreign currency as the means of exchange and unit of account; that is, there is no difference in analytical terms between using dollars for domestic transactions and using local currency which is 100 per cent backed by dollars. The obvious advantage is that the currency board provides a credible guarantee of monetary stability, thereby reducing inﬂationary expectations and restoring faith in the domestic currency. For transition economies which have

84

Nigel M. Healey and Janet Ilieva experienced very rapid inﬂation rates and the resultant ‘dollarisation’ of the economy, as conﬁdence in the domestic currency collapses, a currency board can provide a powerful mechanism for bringing about a sustainable disinﬂation. There are two major risks to a currency board. One is related to the central bank’s role in preserving the viability of the commercial banking system. In most countries, the central bank acts as ‘lender of last resort’, lending reserves to commercial banks with liquidity difﬁculties and thereby maintaining conﬁdence in the banking system as a whole. Given the widespread problems of commercial banks in the transition economies, many of which have non-performing loans to enterprises which are bankrupt, there is a danger that political and social pressure to rescue failing banks will force the central bank to issue local currency in excess of its holdings of foreign exchange reserves. The cost of avoiding a wholesale collapse of the commercial banking system may be the breakdown of the currency board. The second risk, which applies to all forms of ﬁxed exchange rate systems since all involve a loss of a monetary sovereignty, stems from the government’s budget deﬁcit. Introducing a currency board clearly closes off the option of ﬁnancing the budget deﬁcit through the creation of money. Indeed, it is precisely because a currency board forces the government to ﬁnance itself in non-monetary ways that this form of ﬁxed exchange rate is so attractive to countries bedevilled by hyperinﬂation. However, if the government is politically incapable of reducing its budget deﬁcit, it may be that resort to the ‘inﬂation tax’ (i.e. ﬁnancing expenditure by printing money) is preferable to the social unrest that might follow from the nonpayment of state workers, pensioners and so on. Unless achieving a more balanced budget is politically feasible, a currency board will fail. The disadvantages of a currency board in economic terms are basically those of a ﬁxed exchange system, namely the loss of monetary sovereignty. In the special case of transition economies, this can cause adjustment difﬁculties. Because high-inﬂation countries tend to be extensively dollarised, the introduction of a credible currency board will encourage residents to switch out of dollars into local currency. Under the currency board arrangements, the central bank’s foreign exchange reserves and the stock of issued banknotes will grow. Because this portfolio reallocation out of dollars into local currency simply changes the form in which money is held, there need be no inﬂationary consequences from the subsequent rapid growth in the money supply. However, experience in the Baltic States and eastern Germany suggests that, as people switch from hoarded dollars to domestic currency, the velocity of circulation may rise. Because the exchange rate is ﬁxed, any temporary increase in prices leads to a rise in the real exchange rate and may cause problems for exporters.

Central banking in transition economies 85

6 Microeconomic reform A central planning system subsumes the activities performed in a market economy by: 1 2 3

dynamic, proﬁt-seeking ﬁrms, which seek out proﬁtable investment and production opportunities; open competitive markets, in which domestic and foreign consumers and producers come together to set prices, sales and output; the ﬁnancial system, which channels resources from savers to investors who can earn the highest, risk-adjusted returns.

Building a fully functioning market economy thus involves: transforming state enterprises into outward-looking, proﬁt-oriented ﬁrms; opening up domestic markets to internal and external competition and establishing a banking system and capital market in which funds are allocated on the basis of objective risk assessments and prospective returns, rather than in accordance with the demands of a central plan. Finally, transition requires the establishment of a functioning legal system, to deﬁne and protect property rights, clarify ownership and, most importantly, to provide a legal framework which facilitates investment. Table 3.8 sets out the performances of the transition economies in each of these key areas. The indices are ranked from 1 (⫽ little or no progress) to 4 (⫽ complete or near complete reform). The key below sets out the meaning of the indices in more detail. Table 3.8 shows that the CEE countries are far more advanced in the transition process than are their CIS rivals. Strikingly, of the twelve CIS countries, Russia is in the vanguard of reform, with the private sector now accounting for 70 per cent of GDP following a successful and extensive privatisation programme. Overall, Table 3.8 shows that most of the transition economies have completed (or nearly completed) the comprehensive liberalisation of prices, foreign trade and exchange rates; in almost all countries, small-scale enterprises have been largely privatised. The difference between the CEE and CIS countries is more stark in the ﬁeld of large-scale privatisation of SOEs and enterprise restructuring and ﬁnancial sector reform, where the latter lag markedly. It is also worth noting that the pace of structural change is beginning to slow as compared with the early 1990s. This is because most of the ‘easier’ reforms have already been achieved, so that the ‘hard core’ of tasks that remain, notably the restructuring and privatisation of large, loss-making SOEs and the creation of an efﬁcient ﬁnancial system, are those that promise to be the most socially or politically painful. It is also, in part, because the governments that were in power in the immediate aftermath of the collapse of communism enjoyed a popular mandate to undertake radical political change; as the social pain of the transition process has mounted, the electorate’s willingness to tolerate the further sacriﬁces

60 45 20 60 55 60 45 70 30 25 55 45

CIS countries: Armenia Azerbaijan Belarus Georgia Kazakhstan Kyrgystan Moldova Russia Tajikistan Turkmenistan Ukraine Uzbekistan

Source: EBRD (1997).

75 60 60 80 75 55 80 65 70 65 60 75 55

CEE countries: Albania Bulgaria Croatia Czech Republic Estonia FYR Macedonia Hungary Latvia Lithuania Poland Romania Slovak Republic Slovenia 3 2⫺ 1 3⫹ 3 3 3 3⫹ 2⫹ 2⫺ 2⫹ 3⫺

2 3 3 4 4 3 4 3 3 3⫹ 3⫺ 4 3⫹

Private LargeSector scale (% GDP) privat.

Firms

Table 3.8 The transition picture in 1999

3⫹ 3 2 4 4 4 3⫹ 4 3 2 3⫹ 3

4 3⫹ 4⫹ 4⫹ 4⫹ 4 4⫹ 4 4⫹ 4⫹ 4⫺ 4⫹ 4⫹

Smallscale privat.

2 3 1 2 2 2 2 2⫺ 2⫺ 2⫺ 2 2

2 2⫹ 3⫺ 3 3 2 3⫹ 3⫺ 3⫺ 3 2 3 3⫺

Ent. restruct.

■

3 3 2⫺ 3 3 3 3 3⫺ 3 2 3 2

3 3 3 3 3 3 3⫹ 3 3 3⫹ 3 3 3

Price liberal.

Markets

4 3⫹ 1 4 3 4 4 2⫹ 3⫺ 1 3 1

4 4⫹ 4 4⫹ 4 4 4⫹ 4⫹ 4 4⫹ 4 4⫹ 4⫹

Trade and forex

2 1 2 2 2 2 2 2⫹ 1 1 2 2

2 2 2 3 3⫺ 1 3 3⫺ 2⫹ 3 2 3 2

Comp. Policy

■

2⫹ 2 1 2⫹ 2⫹ 2⫹ 2⫹ 2⫺ 1 1 2 2⫺

2 3⫺ 3 3⫹ 4⫺ 3 4 3 3 3⫹ 3⫺ 3⫺ 3⫹

Bank reform

Finance

2 2⫺ 2 1 2 2 2 2⫺ 1 1 2 2

2⫺ 2 2⫹ 3 3 2⫺ 3⫹ 2⫹ 3⫺ 3⫹ 2 2⫹ 3

Capital markets

■

3⫺ 3⫺ 2 2 3⫹ 3 3⫹ 3 n.a. n.a. 2 3⫺

2 4⫺ 3⫹ 3 4⫺ 4⫺ 4⫺ 3 3⫹ 3⫹ 3⫹ 3 4

Index of legal reform

Law

Key 4

3

2

1

Large-scale privatisation

more than 50% large enterprise privatised

more than 25% large enterprise privatised

privatisation little or scheme for nothing done large enterprises about to be introduced

Small-scale privatisation

all small enterprises (e.g. shops, restaurants) privatised

most small enterprises privatised

signiﬁcant proportion of small enterprises privatised

little or nothing done

Enterprise restructuring

corporatisation of state enterprises completed

corporatisation in progress, but incomplete

legislative reforms passed, but not fully implemented

few reforms passed

Price liberalisation

complete price liberalisation and antimonopoly controls

substantial price liberalisation

price controls remain for signiﬁcant number of products

most prices still controlled

Trade and foreign exchange system

no import/ export quotas or controls, convertible exchange rate

few quotas, near full convertibility of exchange

few quotas, but exchange controls still in force

widespread import/export quotas and/or exchange controls

Competition policy

signiﬁcation anti-monopoly enforcement

some enforcement to control abuse of market power

basic competition policy rules and institutions set up

no competition policy rules or institutions

Banking reform

wellfunctioning banking system

substantial progress towards competitive banking system

interest rates main means of allocating credit

little or no progress

Capital markets

functioning, liquid capital market, with legal framework approaching international standards

substantial issues of shares by private ﬁrms on capital market

establishment of capital market and limited trading

little or no progress

Legal reform index: rules on pledge, bankruptcy and company law

comprehensive legal rules, independently and consistently enforced

legal rules exist, but enforcement patchy

legal rules are limited or contradictory, hard to enforce

legal rules very limited in scope, unclear and not consistently enforced

88

Nigel M. Healey and Janet Ilieva

necessary to maintain the rate of progress has waned. The blunting of voters’ appetite for economic transformation, even in the CEE where progress has been more swift and the beneﬁcial results more tangible and evenly spread, is best illustrated by their growing propensity to elect leftwing administrations pledged to slow the pace of reform in the interest of promoting social harmony.

7 Conclusions Recent years have witnessed a wave of change in central bank legislation around the world, designed to increase the independence of these institutions. These changes have partly been driven by theoretical advances showing that in a world of forward-looking rational agents, the control of monetary policy by governments which have multiple objectives (e.g. for inﬂation, growth, interest rates) leads to an inherent ‘inﬂationary bias’ which can be eliminated by delegating monetary control to an independent central bank mandated to pursue price stability. The perceived need for structural change in the division of monetary policy-making powers has been reinforced by cross-sectional empirical studies showing a negative correlation between the degree of central bank independence and inﬂation, with demonstrably independent banks like the Bundesbank providing an exemplar. Change has also been hastened by the failure of many governments to credibly pre-commit to non-inﬂationary policies in other ways (for example, by adopting a broad money growth target or an exchange rate peg); in an increasingly deregulated global ﬁnancial market, governments have accordingly used central bank independence as a way of pledging their anti-inﬂation resolve and thereby promoting ﬁnancial stability. This introduction has outlined the broad macroeconomic context within which the reform of central banking is taking place in transition economies generally, and the CEE states in particular. In contrast to the relatively modest institutional reforms being undertaken in many developed and developing countries, the construction of independent central banks in transition economies is part of a wider, fundamental restructuring of the economy as a whole. Central banks have had to be carved out of the previously uniﬁed national banking system and, in addition to establishing monetary stability, have been required to take over the supervision of commercial banking systems undergoing privatisation and, in many cases, burdened with non-performing loans to former state enterprises. Moreover, these reforms to the banking system take place in economies struggling with the heavy costs of economic transition. The collapse of former trading relations, deep slumps in economic activity, widespread poverty and popular hostility to restructuring all militate against the ability of the new central banks to establish anti-inﬂation credentials. Indeed, in economies accustomed historically to the absence of inﬂation and unem-

Central banking in transition economies 89 ployment, the new central banks are widely perceived as the agents of recession and economic dislocation. In seeking to sever their ties with government and to end automatic ﬁnancing of budget deﬁcits and soft credits to enterprises, they face levels of public and political pressure unknown in developed countries. The chapters that follow explore the theoretical modelling of central bank independence and the empirical links between independence and macroeconomic performance for developed and developing countries. Using four case studies of the major CEE states (Poland, the Czech Republic, Romania and Bulgaria) the book illustrates some of the special features of transition economies. Using these lessons, Chapter 10 then estimates new indices of central bank independence for a range of transition economies and draws some broad policy conclusions.

Note 1 The table includes Slovenia, Croatia and Macedonia, but excludes the other states of former Yugoslavia (Bosnia, Serbia and Montenegro) where war and economic sanctions have so far prevented economic transformation.

References Alesina, A. (1989), ‘Politics and business cycles in industrial democracies’, Economic Policy, 8, 55–98. Alesina, A. and Gatti, R. (1995) ‘Independent central banks: low inﬂation at no cost?’, The American Economic Review, 85, 196–211. Alesina, A. and Summers, L. (1993) ‘Central bank independence and macroeconomic performance: some comparative evidence’, Journal of Money, Credit and Banking, 25, 151–162. Bade, R. and Parkin, M. (1988) ‘Central bank laws and monetary policy’, Working Paper, University of Western Ontario. Barro, R.J. and Gordon, D.B. (1983), ‘Rules, discretion and reputation in a model of monetary policy’, Journal of Monetary Economics, 12, 101–121. Cukierman, A. (1996) Central Bank Strategy, Credibility and Independence: Theory and Evidence, Cambridge, MA: MIT Press. Cukierman, A. and Webb, S. (1995) ‘Political inﬂuence on the central bank: international evidence’, The World Bank Economic Review, 9, 397–398. Cukierman, A., Webb, S. and Neyapti, B. (1992) ‘Measuring the independence of central banks and its effect on policy outcomes, The World Bank Economic Review, 6, 353–398. Debelle, G. and Fischer S. (1994) ‘How independent should a central bank be?’, in Fuhrer, J.C. (ed.) ‘Goals, guidelines and constraints facing monetary policymakers’, Federal Reserve Bank of Boston, Conference Series 38, 195–221. EBRD (1999) Transition Report 1999, London: European Bank for Reconstruction and Development. EBRD (2003) Transition Report 2003, London: European Bank for Reconstruction and Development.

90

Nigel M. Healey and Janet Ilieva

Grilli, V., Masciandro, D. and Tabellini, G. (1991) ‘Political and monetary institutions and public ﬁnance policies in the industrial economies’, Economic Policy, 13, 341–392. Kydland, F. and Prescott, E.C. (1977) ‘Rules rather than discretion: the inconsistency of optimal plans’, Journal of Political Economy, 85, 473–492. Loungani, K.P. and Sheets, N. (1997) ‘Central bank independence, inﬂation and growth in transition economies’, Journal of Money, Credit and Banking, 29, 381–399. Rogoff, K. (1985) ‘The optimal degree of commitment to an intermediate monetary target’, Quarterly Journal of Economics, 100, 1169–1689.

Part II

4

The changing role of central banks in market economies Janet Ilieva, Barry Harrison and Nigel M. Healey

1 Introduction Among the main reasons for the emergence of central banks in Europe were the wars that ravaged the Continent from the seventeenth century onwards and the consequent pressure this exerted on government ﬁnance. In brief, governments granted monopoly power over the note issue to a commercial bank and in return were given privileged borrowing facilities. This marked the beginning of the ‘special relationship’ between governments and their central bank. However, in most cases, recent years have witnessed enormous changes in the nature of this relationship. In particular, since the beginning of the 1990s, many governments have become convinced that the way to ensure price stability is to sever the institutional links between government and the central bank, leaving the latter to manage monetary policy free from political interference. The focus of this chapter is on the historical developments, which have underpinned this new monetary orthodoxy. Following this introduction, Section 2 of this chapter examines the evolution of central banks. The development of central bank–government relations is traced with reference to the political and economic independence of the central bank from the government of the day. In this context, ‘political independence’ refers to the autonomy of the central bank to set its own objectives, as well as the independent appointment of governor and board. ‘Economic independence’ (sometimes referred to as ‘operational independence’) refers to the central bank’s freedom to choose the policy instruments to achieve its objectives. Section 3 discusses the changing nature of the objectives of central banks. It identiﬁes common trends in the evolution of central bank independence and studies their development in three major periods: the nineteenth and twentieth centuries up to the mid-1940s, the mid-1940s to the mid-1970s, and the mid-1970s to the present day. Section 4 provides a study of four central banks: the Bank of England, the Bank of France, the Bundesbank and the Federal Reserve System. The political index of central bank independence constructed by Grilli and

94

Janet Ilieva et al.

colleagues (1991) is used to demonstrate how the level of independence granted to some central banks changed at different stages of their development. Section 5 concludes this chapter and draws some common trends in the development of central bank independence. One important ﬁnding that we demonstrate is that in countries where the central bank was founded in the nineteenth century or earlier as a private institution, it possessed almost absolute independence. This state of affairs was reversed after 1945 when most central banks were effectively turned into institutions of government responsible for implementing the government’s monetary policy decisions. More recently there has been a tendency to reverse earlier post-war trends and many central banks have regained some of the independence that characterised earlier years of their existence.

2 The nature of central bank independence The extent of central bank independence is assessed against two criteria: political independence and economic independence. Political independence, as deﬁned by Grilli et al. (1991, p. 366), embraces three aspects of monetary policy: (i) the procedure for appointing members of central bank governing bodies; (ii) the relationship between these bodies and government; and (iii) the formal responsibilities of the central bank. . . . This is why we identify independence with autonomy to pursue the goal of low inﬂation. Deﬁned in this way, political independence was greater in the earlier history of central banks than in the present day. The fact that most of the central banks were established as private institutions gave them autonomy to make their own appointments, set their own regulations and pursue their own objectives. Economic independence, on the other hand, is deﬁned by Grilli et al. (1991, p. 368) as the freedom of a central bank to choose the instruments of monetary policy with regard to: (i) the inﬂuence of the government in determining how much to borrow from the central bank; and (ii) the nature of the monetary instruments under the control of the central bank. Central banks were established mainly to provide ﬁnance for governments to prosecute wars. Consequently, despite the high degree of political independence accorded central banks, until recently they were granted far less economic independence. The major problem confronting central banks throughout the early years of their existence was that their obliga-

Role of central banks in market economies 95 tions conﬂicted. On the one hand they were required to ﬁnance government wartime expenditures, and on the other they were required to maintain the full convertibility of gold at the ﬁxed rate. In reality, this conﬂict of objectives was more apparent than real, and the overriding objective of central banks was to maintain the Gold Standard. Furthermore, in times of peace no conﬂict arose, since the prevailing orthodoxy was one of laissezfaire.

3 Independence and the changing objectives of central banks No consensus has emerged in the literature over the historical development of central banks. Toniolo (1988) has referred to their development as ‘the free offspring of parents who were not born free’. Despite the lack of any consensus, following Goodhart (1994) this section identiﬁes three distinct phases in the historical development of central banks: the nineteenth and twentieth centuries up to the mid-1940s, the mid-1940s to the mid1970s and the mid-1970s to the present day. Period up to the mid-1940s Goodhart and colleagues (1994, p. 51) has argued that in the nineteenth and early twentieth centuries, central banks had considerably more independence than they currently possess. Elgie and Thompson (1998) offer three reasons for this. First, the laissez-faire economies of the nineteenth century provided no role for the state and left the problem of resource allocation to the market. Correspondingly, no active role existed for central banks in inﬂuencing the performance of the macroeconomy which was regarded as self-regulating. Second, the operation of the Gold Standard implied central bank independence, since their major objective was to maintain a stable economic environment consistent with ensuring convertibility with the national currency within the limits set by the ‘gold points’. Third, the equity of central banks was privately owned and this gave them considerable a priori independence. Goodhart (1988) has also stressed the role of the Gold Standard, and has argued that the objective of the early central banks was to ‘unify what had become in cases, e.g., in Germany, Switzerland and Italy, a somewhat chaotic system of note issue, to centralise, manage, and protect the metallic reserve of the country, and to facilitate and improve the payments system’. The operation of the Gold Standard provided a means to achieve at least some of these objectives and, during the late nineteenth and early twentieth centuries, central banks were charged with responsibility for maintaining the convertibility of national currencies. Central banks also provided ﬁnance for governments in times of war when tax revenues were insufﬁcient to meet government expenditures. An obvious conﬂict exists

96

Janet Ilieva et al.

between these objectives, but until the mid-1940s it was generally accepted that central banks would have no obligation to ﬁnance government expenditures in times of peace. The mid-1940s to the mid-1970s A second period stretching from the 1940s to the mid-1970s can be identiﬁed. During this period governments became increasingly active in managing the economy, and Goodhart (1995, p. 112) has noted that following the end of the Gold Standard ‘the links between central banks and governments in the conduct of the macro-policy became much closer’. The economy was no longer thought to be self-regulating and, among other things, central banks were now charged with responsibility for ensuring that the central government’s budget deﬁcit was ﬁnanced in accordance with planned changes in aggregate demand which governments felt would deliver their economic objectives. The multiple and inconsistent goals of central government (inﬂation, employment, growth and the balance of payments) were a source of conﬂict with their central bank because these goals had no clearly deﬁned hierarchy and their importance often changed in response to economic mismanagement, or as governments moved through the political cycle. The rate of interest became the main operational tool of monetary policy and, in order to ensure central bank compliance with required changes in the rate of interest, many governments nationalised their central banks. For example, during this period the central banks in Canada, Denmark, the Netherlands, England, France, Norway and New Zealand were all brought into public ownership (Elgie and Thompson 1998, p. 17). This removed at a stroke any independence central banks possessed, but the situation was different in Germany where, after the currency reform of 1948, the Bundesbank was constitutionally authorised to preserve the internal value of the currency. German experience of hyperinﬂation in the 1920s signiﬁcantly increased the country’s determination to maintain price stability, and this was accorded priority, even in times when most countries were targeting maximum employment! Being banker to the central government, central banks have gradually increased the degree of centralisation of commercial banks’ reserves. Consequently Goodhart (1988) has identiﬁed two dimensions of central bank monetary policy: a macro dimension and a micro dimension. The macro dimension involves setting monetary conditions for the macroeconomy, while the micro dimension involves ensuring the efﬁcient functioning of the individual entities that make up the banking system. The interrelationship between the central bank’s macro and micro functions resulted in the evolution of a supervisory function for central banks ultimately involving the provision of lender-of-last-resort facilities. This role was performed in different ways in different countries. In some, such as Germany and

Role of central banks in market economies 97 Switzerland where the central bank was publicly funded, banking supervision was entrusted to a separate body and the central bank was not empowered with lender-of-last resort facilities. In other countries, such as England, France and Italy where the central bank was initially funded by private shareholders, it was charged with responsibility for providing lender-of-last-resort facilities and was also empowered with a supervisory role over the commercial banks. Post-1970s The revival of independent central banks marks the third stage of their development as identiﬁed by Goodhart (1994). The policy of granting a greater independence to central banks became particularly popular during the 1990s when countries world-wide started providing their central banks with greater autonomy. Cukierman (1995) has argued that there are several reasons behind this tendency. First, the experience with ﬁxed exchanged rates, in particular the Bretton Woods System and later the European Monetary System, persuaded countries to design institutions increasing their commitment to price stability. In most countries until about the mid-1970s, economic policy was based on the assumed existence of a stable trade-off between inﬂation and unemployment, and decisions by the central bank were motivated by the particular combination of inﬂation and unemployment that satisﬁed the government’s objectives at each point in time. As the relationship between unemployment and inﬂation deteriorated during the 1970s a consensus emerged at the International Monetary Fund conference in Kingston, Jamaica in 1976, that the primary objective of central banks should be price stability. Increasingly, price stability has become the major objective of the majority of central banks world-wide with other goals, such as promoting stable employment, accorded far less prominence in the hierarchy of central bank objectives. The single policy objective greatly enhanced the independent status of central banks, and Goodhart (1994) has argued that central banks with a single objective are more likely to be less subservient to central governments than central banks with a plurality of vague objectives. A single objective for monetary policy also facilitates greater accountability, since it is abundantly clear whether an institution has achieved its objectives or not. Goodhart (1994) has further argued that targeting a single objective might reduce any dispute between central bank ofﬁcials and academic economists over operational techniques since, for most central banks, interest rate adjustment is the only instrument of policy available. The second reason identiﬁed by Cukierman (1994) for the emerging trend towards central bank independence was its establishment as one of the requirements for joining the single currency bloc. As a consequence, the central banks of European Union (EU) countries were granted increasing independence in the 1990s as a prelude to the creation of the

98

Janet Ilieva et al.

single currency. More recently, the EU accession countries have granted independence to their central banks, and more generally this is now an established global feature of central bank development. The third reason for emerging independence among central banks identiﬁed by Cukierman (1994) was the performance of the Bundesbank with its proven track record of delivering consistently low inﬂation in the postwar period. Progress was also made in providing the theoretical explanation for the Bundesbank’s success. In particular, Kydland and Prescott (1977) showed that when a central bank is not independent, policy announcements are subject to time inconsistency. Barro and Gordon (1983) extended this work and showed that in the absence of binding rules on central bank behaviour, an inﬂationary bias existed. Reputational considerations might reduce this inﬂationary bias, but it was felt that an independent central bank would act as a pre-commitment device, which would enhance credibility by transferring responsibility for monetary policy to a non-political body. This study, as well as an earlier study by Rogoff (1985), provided the rationale for what the Germans and the Swiss had known for decades: that price stability would be achieved more easily if central banks were granted greater independence from central government. The theoretical predictions of Rogoff (1979) and Barro and Gordon (1983) were conﬁrmed empirically by Cukierman (1992), Cukierman et al. (1992), and Grilli et al. (1991) who showed that independent central banks facilitate lower inﬂation.

4 Central bank independence trends: country comparisons This section provides some inter-country comparisons of trends in CBI since their formation up to the present day. The central banks included in the comparison are the Bank of England, the Bank of France, the Federal Reserve and the Bundesbank. This chapter measures the degree of political independence of these central banks using the index designed by Grilli and colleagues (1991) from their formation to the present day. In addition, their index (referred to hereafter as GMT index) has become increasingly popular in the economic literature after its introduction in 1991. Table 4.1 summarises the studies on CBI that have used GMT index. The Bank of England From its very earliest days the Bank of England could appoint and nominate its own personnel. The Governor, Deputy Governors and directors were chosen every year between March and April (Elgie and Thompson 1998, p. 36). Independence was guaranteed because the Committee of the Treasury was created comprising the Governor, Deputy Governor and the most senior of the directors with responsibility for preparing proposals for the election of Governors and Directors. For 1931, a lower score on the

Role of central banks in market economies 99 Table 4.1 Summary of major studies using the GMT index of CBI Empirical studies

Indices used

Grilli, Masciandro and Tabellini (1991)* Dvorsky (2000) Maliszewki (2000)* Alesina and Summers (1993)*

GMT Cukierman (1992), GMT GMT GMT, Bade and Parkin (1988), Alesina (1988) GMT GMT, Alesina (1989), Eijfﬁnger and Schaling (1992) Cukierman (1992) Alesina (1988), Eijfﬁnger and Schaling (1993), GMT, Cukierman (1992)

Alesina and Grilli (1992)* De Haan and Sturm (1992)* De Haan and Siermann (1994) Eijfﬁnger and Schaling (1995)* Note *those which meet the independence criteria.

overall index of independence is recorded because the degree of economic independence of the Bank fell as a result of its responsibility for selecting the instruments of monetary policy being withdrawn. An attempt to measure the political independence of the Bank of England has been made using the GMT index (1991) summarised in Table 4.2. Table 4.2 shows how the degree of independence of the Bank of England changed over the period since its formation in 1694 until 1998. Both tables (4.1 and 4.2) conﬁrm that in 1946, when the Bank was nationalised, there was a dramatic fall in the overall level of independence. Prior to this, all appointments were made independently of government, but after nationalisation all positions were government appointments, and the anchor for monetary stability, the Gold Standard, was replaced by a plurality of competing objectives to be achieved through government intervention. The amendments enshrined in the Banking Act of 1998 granted the Bank greater independence and constituted the main focus of the Act. In line with greater independence, price stability was established as the major objective of the Bank and supervision of the banking system was transferred from the Bank to the FSA. As a result, the index score of political independence increased from one to three during 1998. The Bank of France The Bank of France was founded at the very beginning of the nineteenth century with private shareholder capital. It therefore possessed a high degree of political independence, since nominations and appointments to the General Council (the governing council) were made independently of government. The General Council consisted of ﬁfteen members who appointed the Central Committee which was charged with responsibility

100

Janet Ilieva et al.

Table 4.2 Political independence of the Bank of England (1694 to 1998) using the GMT indexa Question

1694

1931

1946

1992b 1998

1 Governor not appointed by the government 2 Governor appointed for more than ﬁve yearsc 3 All the Board not appointed by the government 4 Board appointed for more than ﬁve years 5 No mandatory participation of government representative on the Board 6 No government approval of monetary policy required 7 Statutory requirements that central bank pursues monetary stability among its goals 8 Legal provision strengthening the central bank’s position in conﬂict with the government Overall index of political independence

*

*

–

–

–

–

*

*

–

–

*

*

–

–

–

* *

* *

– –

– *

– *

*

–

–

–

*

*

–

–

–

*

n.a.

na.

–

–

n.a.

6

5

1

1

3

Notes a The authors derive these estimates from reference to the bank’s statute. An asterisk indicates that the criterion is satisﬁed and a dash indicates that the criterion is not satisﬁed. b These results are taken from Alesina and Grilli (1992, p. 49). c The results for the period 1694 to 1946 are reported by Elgie and Thompson, according to their index of term of ofﬁce. During this period the term of ofﬁce was between ﬁve and eight years.

for supervision of the Bank’s activities. The Regents (members of the General Council) were elected by the General Assembly of the shareholders. The tenure of the Governor and Sub-governors was completely free of any outside interference. Monetary policy was conducted by the Bank and decisions of the Board were taken independently of any instructions from the government of the day. Using GMT index, the political independence of the Bank of France is measured and our results using this index are reported in Table 4.3. Comparing Table 4.2 with Table 4.3 reveals a common trend between the Bank of France and the Bank of England. Their activities were heavily controlled by their respective governments during the period 1945 to 1992, but both were highly independent during the Gold Standard period, as they are again in the 1990s. The Governor and Sub-governors now have six-year terms of ofﬁce, secure tenure and are prohibited from accepting any instructions from the central government. Monetary policy was entrusted solely to the Bank. With respect to economic independence, the major change affected lending to government and the Bank was prohibited from ‘authorising credit or granting any form of debt facility to the

Role of central banks in market economies 101 Table 4.3 Political independence of the Bank of France (1800 to 1993) using the GMT index Question

1800

1808

1945

1992a

1993

1 Governor not appointed by the government 2 Governor appointed for more than ﬁve yearsb 3 All the Board not appointed by the government 4 Board appointed for more than ﬁve years 5 No mandatory participation of government representative on the Board 6 No government approval of monetary policy requiredb 7 Statutory requirements that central bank pursues monetary stability among its goals 8 Legal provision strengthening the central bank’s position in conﬂict with the government Overall index of political independence

*

*

–

–

–

–

*

*

*

*

*

–

–

–

–

*

*

*

*

*

*

*

–

–

–

*

–a

–

–

*

–

–

–

–

*

n.a.

n.a.

–

–

*

5

4

2

2

5

Notes a These results are taken from Alesina and Grilli (1992, p. 49). b The Board does not accept instructions from the government but there are government representatives with the right of veto, and thus we assume that government approval is necessary for policy formulation.

Treasury. The direct acquisition of government debt is also prohibited’ (Elgie and Thompson 1998, p. 133). The Federal Reserve In contrast to most of the European countries, the USA did not have a central bank until 1913. Instead the US Treasury performed the role of central bank. Sylla (1988, p. 20) has described the system thus: The entire system was the victim of a kind of irregular and vicious centralisation. . . . The money power of the country passed into the hands of a few ﬁnanciers and big bankers, and the treasury itself, through politics and manipulation, acted in sympathy with them. Clifford (1965, p. 50) argues that: A few years ago, when the US Treasury was burdened with excessive revenues and the money market depended on the whim of the Secretary

102

Janet Ilieva et al. of the Treasury, practically all public men of whatever shade or political belief, were agreed that the government ought to be taken out of the banking system.

The corporate elite became the driving force in the process of separating the central bank from the Treasury and creating an independent Federal Reserve. A central banking system, with twelve regional Federal Reserve banks instead of a single central bank, was created with the Federal Reserve Act in 1913. The rationale behind this was to prevent a single bank (New York) from dominating the nation’s administrative and ﬁnancial centre (Sylla 1988). The Federal Reserve banks appoint six directors and the Board in Washington appoints three other directors, making a total of nine directors in all. The nine-member Board appoints ofﬁcers of the respective regional reserve banks along with regional governors. With regard to these appointments, the Federal Reserve System is completely independent, since none of the appointees come from institutions outside the Federal Reserve. The Federal Reserve Board, consisting of ﬁve members, is appointed by the President of the USA for a period of ten years, and in addition two exofﬁcio members are appointed. These are the Treasurer and his(her) subordinate who acts as the Comptroller of the Currency. The Federal Reserve Board is thus highly independent of government and its main role is to stand between the latter institution and the Reserve banks, to conduct a uniﬁed monetary policy and to supervise the Reserve banks’ operations. The longer term of ofﬁce of the Board of Governors makes it difﬁcult for a president to inﬂuence the Board’s decisions. Despite this, the President retains power over the Board and is allowed, via the Treasurer, to be involved in open market operations with or without the approval of the Federal Reserve. Moreover, in times of emergency these ofﬁces are also free to intervene in the central bank activities whenever and however they decide appropriate (Sylla 1988). Table 4.4 shows the changing nature of Federal Reserve Bank independence between 1912 and 1992. With respect to the degree of economic independence envisaged in the Banking Act of 1913, the regional reserve banks, acting as lender of last resort and ﬁscal agents, were partially allowed to issue banknotes. The inﬂuence of government became more explicit in 1917 when the USA entered the First World War. The Federal Reserve objected to the low interest rates set by the government on loans and securities – but these objections had no effect on policy and interest rates remained as set by the government! The Banking Act of 1935 conferred greater independence on the Federal Reserve. The major changes regarding economic independence are: •

the Board of Governors (hereafter called the Board) could alter the legal reserve requirements of member banks;

Role of central banks in market economies 103 Table 4.4 Political independence of the Federal Reserve (1913 to 1992) using the GMT index Question

1913

1935

1992

1 2 3 4 5

– * – * –

* – * *

– – – * *

– –

– –

* *

–

–

*

2

3

5

Governor not appointed by the government Governor appointed for more than ﬁve yearsb All the Board not appointed by the government Board appointed for more than ﬁve years No mandatory participation of government representative on the Board 6 No government approval of monetary policy required 7 Statutory requirements that central bank pursues monetary stability among its goals 8 Legal provision strengthening the central bank’s position in conﬂict with the government Overall index of political independence

• • •

the Board could set maximum interest rates on time deposits that banks could pay; the Board could set margin requirements on loans to purchase securities; the Federal Reserve Open Market Committee was established with responsibility for carrying out open market operations. This marked a departure from the past when the Reserve banks had the authority to implement their own open market operations.

The above amendments increased the power of the Federal Reserve over the member banks and brought about a more efﬁcient and uniﬁed monetary policy across the country. The Federal Reserve is now regarded as one of the most independent central banks in the world. The GMT index shows a relatively high degree of political independence, despite the Governor and the Board being appointed by the President of the USA. The Bundesbank The foundation of the German central bank took place on 22 January 1870, but the Reichbank de facto started to operate in 1876. Most of the founders were private shareholders. The primary objective of the Reichbank was to unify the note issue, but its other tasks as central bank were to improve and organise the payment system in the country. Lexis has noted that ‘the nature of its [Reichbank’s] task is that it shall maintain the value of monetary unit as stable as possible’ (Quoted in Goodhart 1988, p. 108). The Reichbank’s administrative functions were performed by the Administrative Board and Management. The board was the ‘holder of all powers of attorney on the company’s behalf’. (Quoted in Gall 1995, p. 13).

104

Janet Ilieva et al.

Table 4.5 Political independence of the German Bank (1870 to 1998) using the GMT index Question

1880

1939

1997

1 2 3 4 5

* * * *

n.a. n.a. n.a. n.a.

– – – *

– *

n.a. n.a.

* *

*

n.a.

*

* 7

n.a. n.a.

* 6

Governor not appointed by the government Governor appointed for more than ﬁve yearsb All the Board not appointed by the government Board appointed for more than ﬁve years No mandatory participation of government representative on the Board 6 No government approval of monetary policy required 7 Statutory requirements that central bank pursues monetary stability among its goals 8 Legal provision strengthening the central bank’s position in conﬂict with the government Overall index of political independence

The managers of the Reichbank were required to operate in ‘accordance with instructions given by the Administrative Board (subsequently the Supervisory Board)’. The Chairman of the Board was appointed by election. Instructions to the Reichbank came from its shareholders pursuing their own interest, rather than from government. This is illustrated by the resignation of one of the two members of the Board with political afﬁliations because of the ‘business activity that has so powerfully imposed itself since. I wished to protect my parliamentary position on major economic issues against any possibility of attack’ (Gall 1995, p. 13). Table 4.5 provides the degree of independence of the German central bank using the GMT index of political independence. The statute of the Reichbank changed dramatically at the end of January 1933. The newly appointed State Secretary, Gottfried, concluded that ‘Of course the banks need to be directed by the State. . . . One cannot accuse the Government of a lack of initiative’ (quoted in Gall et al. 1995, p. 284). This period has been described thus: ‘The Bank, especially after September 1938, became part of the machine of the German imperialism, and its employees the agents of a brutal political process’ (James 1995, p. 352). The issues of independence came to the fore in Germany with the creation of the Bundesbank which established a standard of independence against which other central banks were judged. Similarly, its track record of delivering low inﬂation became the standard against which other policy makers were judged.

5 Conclusions The cases detailed in this chapter highlight some common trends in the development of central bank independence. In countries where the central

Role of central banks in market economies 105 bank was founded in the nineteenth century or earlier as a private institution, the central bank started with almost absolute political independence (according to the modern interpretation of the term ‘political independence’) along with a comparatively high degree of economic independence that lasted until the beginning of the twentieth century. Later, especially in the 1940s, many central banks lost their independence and in effect became government institutions subservient to the government of the day. The recent developments in central bank independence show a tendency for central banks to regain the independence they possessed in the nineteenth and early twentieth centuries. The catalyst for this trend was a reaction to the rising levels of inﬂation that characterised the 1960s and 1970s in many countries which focused the attention of policy makers on the anti-inﬂationary track of Germany and Switzerland during the postwar period. Both countries were well known for having highly independent central banks and these served as models for others to emulate. The increasing emphasis on price stability as the primary objective of policy encouraged interest in central bank independence not simply in the developed world, but also among emerging economies. With the collapse of the centrally planned system, many countries in Central and Eastern Europe began granting their central banks greater independence with a view to achieving low and stable rates of inﬂation and, in the case of several East European countries, as a prelude to their application for their accession to the European Union.

References Barro, R.J. and Gordon, D.B. (1983) ‘Rules, Discretion, and Reputation in a Model of Monetary Policy’, Journal of Monetary Economics 12, pp. 101–120. Clifford, A. (1965) The Independence of the Federal Reserve System, Philadelphia, PA: University of Pennsylvania Press. Cukierman, A. (1992) Central Bank Strategy, Credibility, and Independence: Theory and Evidence, Cambridge, MA: MIT Press. Cukierman, A. (1995) ‘The Economics of Central Banking’, paper presented at the Eleventh World Congress of the International Economic Association, Tunis, December. Cukierman, A., Rodriguez, P. and Webb, B. (1992) ‘Central Bank Autonomy and Exchange Rate Regimes – Their Effects on Monetary Accommodation and Activism’, in Eijfﬁnger, E. and Huizinga, H. (eds), Positive Political Economy: Theory and Evidence, Cambridge: Cambridge University Press, pp. 78–120. Elgi, R. and Thompson, H. (1998) The Politics of Central Banks, London: Routledge. Gall, L. (1995) ‘The Deutsche Bank from its Foundation to the Great War 1970–1914’, in Gall, L, Feldman, G.D., James, H., Holtfrerich, C.L. and Büschgen, H. (eds), The Deutsche Bank 1870–1995, London: Weidenfeld & Nicolson. Gall, L., Feldman, G.D., James, H., Holtfrerich, C.L. and Büschgen, H. (1995) The Deutsche Bank 1870–1995, London: Weidenfeld & Nicolson.

106

Janet Ilieva et al.

Goodhart, C.A.E. (1988) The Evolution of Central Banks, Cambridge, MA: MIT Press. Goodhart, C.A.E. (1994) ‘What Should Central Banks Do? What Should be their Macroeconomic Objectives and Operations?’, The Economic Journal 104 (November), pp. 1425–1436. Goodhart, C.A.E. (1995) The Central Bank and the Financial System, Basingstoke: Macmillan. Goodhart, C., Cappie, F. and Schnadt, N. (1994), The Development of Central Banking’, in Capie, F., Goodhart, C., Fischer, S. and Schnadt, N. (eds), The Political Economy of Integration: States, Markets and Institutions, Cambridge: Cambridge University Press. Grilli, V., Masciandro, D. and Tabellini G. (1991) ‘Political and Monetary Institutions and Public Financial Policies in Industrialized Countries’, Economic Policy 13, pp. 342–392. Kydland, F.E. and Prescott, E.C. (1977) ‘Rules Rather than Discretion: The Inconsistency of Optimal Plans’, Journal of Political Economy 85, pp. 473–491. Sylla, R. (1988) ‘The Autonomy of Monetary Authorities: The Case of the U.S. Federal Reserve System’, in Toniolo, G. (ed.), Central Banks’ Independence in Historical Perspective, Berlin: Walter de Gruyter and Co. Toniolo, G. (ed.) (1988) Central Banks’ Independence in Historical Perspective, Berlin: Walter de Gruyter and Co.

5

Central bank independence and macroeconomic performance A survey of the evidence Friedrich Kißmer and Helmut Wagner

1 Introduction In the 1990s numerous countries provided their central banks with greater legal independence from government. This trend towards increased central bank independence is not only apparent in highly industrialised countries, but also in the transition countries of Central and Eastern Europe (CEE) and in many other emerging economies, particularly in Latin America. Cukierman (1996) lists twenty-ﬁve countries in various parts of the world which strengthened the independence of their central banks between 1989 and the beginning of 1996. Already this list is incomplete and changes in the statutes of many other central banks, increasing their independence, have taken place since 1996. For example, in summer 1997 the Bank of England became instrument-independent, though not goal independent since the government has set a direct inﬂation target. The reasons for institutional changes to central bank structures vary from country to country. In the member states of the European Union, some of the amendments to national central bank statutes stem from the requirements of legal convergence imposed by the Maastricht Treaty and the Statute of the European System of Central Banks (ESCB). An example of this is the amendments to the German Central Bank Act of 22 December 1997, which, among other things, increased the minimum term of ofﬁce of members of the decision-making bodies of the Bundesbank from two to ﬁve years and abolished the government’s right of suspensory veto. In addition, price stability was laid down as the primary objective of monetary policy (Deutsche Bundesbank 1998, pp. 25f.). With the establishment of a European Central Bank, monetary policy in the eleven participating countries is now delegated to a supranational institution with a high degree of legal independence from the interests of national governments (Alesina and Grilli 1992; De Haan 1997). The transition process in the former socialist countries of CEE has also been accompanied by substantial central bank reforms. Some countries (e.g. the Czech Republic, Hungary) have taken the statutes governing independent central banks in industrialised countries in Western Europe

108

Friedrich Kißmer and Helmut Wagner

as a model for the reform of their own central bank statutes. Many of these countries aspire to EU membership and this will necessitate further reform of the relevant statutes governing their central banks (Radzyner and Riesinger 1997, p. 57). Some developing countries have also initiated institutional reform of the relationship between government and central bank. For example, in the Western Hemisphere, Argentina, Chile, Colombia, Mexico and Venezuela have all conferred greater legal independence as part of their stabilisation programme (Cukierman 1996; Siklos 1995). The institutional reforms to establish independent central banks vary from country to country in essential details. Apart from creating independent central banks by law, some countries enforced additional speciﬁc mechanisms to make central banks explicitly accountable. Countries like Australia, Finland, New Zealand, Sweden, Spain and the United Kingdom have all adopted an inﬂation-targeting framework for monetary policy. The announcement of an inﬂation target provides a nominal anchor for monetary policy and is an alternative to monetary targeting or pegging the exchange rate. The purpose of a nominal anchor is to restrict the central bank’s ability to initiate discretionary changes in policy and to increase the transparency of monetary policy. If the inﬂation targets are set by the government for a deﬁned period, or agreed between the government and the central bank, inﬂation targeting forms an alternative institutional design for an (instrument) independent central bank which stresses the accountability of the central bank. In this sense, inﬂation targeting is aimed at preventing a democratic deﬁcit which can emerge when political decisions are taken by an authority which is not directly legitimated by elections (Briault et al. 1997; Nolan and Schaling 1996). The empirical evidence on the performance of independent central banks does not suggest that they have always been an unqualiﬁed success. In some cases, the consistency of indices based on interpretations of central bank statutes used to measure the degree of independence is disputed in general, and in particular concerning some concrete indices (Eijfﬁnger and Schaling 1995; Mangano 1997; Neumann 1996). In addition, the early correlations between central bank independence and the performance of macroeconomic variables are not always conﬁrmed (Campillo and Miron 1996; Fuhrer 1997), the causality between central bank independence and inﬂation is disputed (Posen 1993), and higher disinﬂation costs as a result of a higher sacriﬁce ratio with central bank independence is also claimed (Debelle and Fischer 1994; Fischer 1996; Hutchison and Walsh 1997; Jordan 1997). The remainder of this chapter is organised as follows. In Section 2 the criteria used to measure central bank independence in the empirical literature are presented. In Sections 3 and 4 an overview of the macroeconomic consequences of central bank independence focuses on the empirical evidence regarding inﬂation, output (economic growth) and the costs of disinﬂation. This procedure can be justiﬁed by the small number

CBI and macroeconomic performance 109 of empirical studies dealing with the relationship between central bank independence and other macroeconomic variables such as interest rates, budget deﬁcits and so on (see Eijfﬁnger and De Haan 1996, p. 38). Furthermore, many economists consider inﬂation and output as the main determinants and indicators of the level of social welfare. The empirical literature on this is evaluated against the background of Grilli et al.’s ‘freelunch’ hypothesis discussed in Section 2. This hypothesis states that central bank independence entails social beneﬁts in terms of low inﬂation rates, but no apparent costs in terms of real macroeconomic performance (e.g. increased output volatility or reduced economic growth). Section 5 presents some conclusions and suggests topics for further research.

2 Measures of central bank independence The empirical substantiation of central bank independence is based on studies which see a ‘free lunch’ in the existence of an independent central bank (Grilli et al. 1991, p. 375). On average, countries with more independent central banks realise comparatively low inﬂation rates without real economic costs in terms of lower economic growth or higher output volatility. In fact, many empirical studies on the relationship between central bank independence and inﬂation conﬁrm a negative correlation between inﬂation and central bank independence. However, the derivation of these ﬁndings is based on the use of different measures of central bank independence. For industrial countries, legal indices are mainly used as proxies for central bank independence whereas for developing countries, more behaviourally oriented indicators are used (Cukierman 1996, p. 15; Eijfﬁnger and De Haan 1996, p. 33). In addition, there appears to be hardly any correlation between central bank independence and long-run growth or output volatility. Some studies even report a positive correlation between disinﬂation costs and central bank independence. However, it is questionable whether these empirical ﬁndings between central bank independence and inﬂation, output and disinﬂation costs can be seen as stylised facts. More recent studies have questioned the consistency of the indicators used to measure central bank independence, the robustness of the statistical associations and interpretations of causality ascribed to them. To examine the links between central bank independence and economic performance it is necessary to devise appropriate criteria for measuring central bank independence. In the industrialised countries, measures of central bank independence are based mainly on the interpretation of central bank laws and therefore concern legal independence only. The construction of so-called legal indices is done by structuring those criteria regarded as relevant and assigning them a value on the same numerical scale of independence. The following sub-section is concerned with the indices used most frequently in empirical studies in market economies.

110

Friedrich Kißmer and Helmut Wagner

Notwithstanding this, actual central bank independence might differ substantially from legal independence. For example, in some transition economies there is inadequate enforcement of the laws governing central bank independence, and a discrepancy exists between the legal perception of central bank independence and the actual degree of central bank independence (Radzyner and Riesinger 1997, p. 75; Wagner 1998, pp. 15f.). The personal characteristics and abilities of central bankers, as well as informal relations between the government and the central bank and subtle methods of political inﬂuence, may also lead to a deviation of actual, from legal independence (Cukierman 1992, pp. 393ff.; Eijfﬁnger and De Haan 1996, p. 22). Furthermore, the extent with which legal independence becomes actual independence may also be characterised by historical experiences which lead to speciﬁc ‘stability cultures’ and a general readiness to comply with valid laws or to respect for laws in general (Cukierman 1996, p. 37; Radzyner and Riesinger 1997, p. 76). In addition, despite wide legal independence from government, there may in fact be political dependence on national parliaments if central bank laws can be amended, abrogated or suspended by simple majorities. (See Cukierman 1996, p. 40) on the Central Bank of Russia and Radzyner and Riesinger (1997, pp. 82ff.) on other examples in Central and Eastern European countries.) Because of the problems of interpreting legal indices of central bank independence, additional, behaviourally oriented indices have also been developed, in particular by Cukierman (1992) and Cukierman et al. (1992). These indices include a questionnaire-based index for identifying deviations of the legal position from actual practice. In addition, the actual turnover rate of central bank governors and the political vulnerability of the central bank governor are used as proxies for the actual independence of central banks, particularly in developing countries. Legal indices of central bank independence Because the legal foundations of monetary policy rarely change in most countries, legal indices possess practically no explanatory power for the behaviour of a country’s macroeconomic aggregates. Most of the empirical studies on the relationship between central bank independence and macroeconomic performance are therefore cross-sectional (Cukierman 1996, p. 8; Eijfﬁnger and De Haan 1996, p. 28). Evaluation of central bank laws is done by assigning a series of index values to different variables. These values are then used to rank central banks according to their degree of independence. Many econometric studies go beyond a purely ordinal aspect because they use the numerical values of indices as explanatory variables in econometric regressions. It is therefore particularly important to know how legal measures of central bank independence differ with respect to their index values. Primarily there are three problem areas which have led to different index values in the literature. The index values

CBI and macroeconomic performance 111 are dependent (Eijfﬁnger and Schaling 1993, p. 50; Mangano 1997, p. 6) on: 1 2 3

the criteria contained in the index, the interpretation and evaluation of the law with regard to each individual criterion, the way in which evaluations are aggregated into an overall index, including the weighting of the criteria.

These problems make it clear that the construction of indices for measuring central bank independence inevitably contain subjective and arbitrary assessments (Cukierman et al. 1992, p. 356; Eijfﬁnger and Schaling 1993, p. 51; Mangano 1997, p. 6). In the following section we introduce and compare some frequently used indices of central bank independence. A starting point for further research was, without doubt, the construction of two indices by Bade and Parkin (1988). The authors differentiate between an index of political independence and an index of ﬁnancial independence. We begin by examining Bade and Parkin’s index of political independence only because the differentiation between political and ﬁnancial independence in the deﬁnition of ﬁnancial independence selected by Bade and Parkin has not been followed in the academic literature on this subject (see Appendix A). To describe the extent of political independence, Bade and Parkin examined the legal provisions of twelve countries for the period 1972 to 1986 with regard to the following three criteria: 1 2 3

Is the central bank the ﬁnal policy authority? Are more than half of the policy board appointments made independently of the government? Is there a government ofﬁcial (with or without voting power) on the policy board?

Since Bade and Parkin permit only yes/no decisions, eight (23) policy types may be distinguished. However, the authors ﬁnd that just four types occur in reality. Accordingly, the index values run in integers from 1 to 4, where a higher value implies greater political independence of the respective central bank (Eijfﬁnger and Schaling 1993, p. 53; Appendix A below). Akhtar (1995) remarks that Bade and Parkin appear to classify an explicit legal speciﬁcation of monetary policy objectives as being unimportant with regard to the degree of central bank independence since legal speciﬁcations of policy goals in central bank laws are not weighted (p. 432). This procedure contrasts with the indices in Grilli et al. (1991) and Cukierman (1992), which also take into account whether price stability or low inﬂation are explicit goals of monetary policy in central bank legislation. Alesina (1988, 1989) has extended the Bade and Parkin index by

112 Friedrich Kißmer and Helmut Wagner considering additional industrial countries, while Eijfﬁnger and Schaling (1993) place greater emphasis on the signiﬁcance of policy independence by modifying question (⫹) in the Bade and Parkin index (see Appendix B). This takes into account, that both the central bank and the government may have some, albeit incomplete, policy authority. Grilli et al. (1991) have developed legal indices that, in comparison to Bade and Parkin, permit a more extensive consideration of relevant details. The authors distinguish between economic and political independence. Economic independence is deﬁned as the ability of the central bank to determine the use and the choice of its monetary policy instruments autonomously and without interference from government.1 Economic independence may be adversely affected by the central bank’s obligations to ﬁnance the government budget, to supervise commercial banks and by a lack of freedom to set interest rates (Grilli et al. 1991, p. 368; Alesina and Grilli 1992, p. 56). In addition, the question of who (the government or the central bank) is in charge of exchange rate policy has to be addressed in this context. Political independence is deﬁned as the ability of the central bank to choose monetary policy goals autonomously and without interference from government. The basic determinants of this ability are found in personal independence (e.g. procedures for appointing and dismissing central bankers’ terms of ofﬁce), in the government’s rights to give instructions to the central bank as well as the right to veto, to suspend or to defer decisions of the central bank, and its right to representation on the central bank board. Furthermore, the degree of political independence depends on the formal responsibilities of the central bank. In this context, different interpretations are placed on the role of statutory requirements that the central bank pursues price stability. Grilli et al. (1991), Alesina and Grilli (1992) and Cukierman (1992) see an explicit mandate for the central bank to restrain inﬂation as a strengthening of central bank independence. In contrast, Debelle and Fischer (1994) and Fischer (1995a, b) characterise this as a reduction of the central bank’s ‘goal independence’. The reason for these deviating interpretations is found in the different emphasis placed on setting central bank goals. Debelle and Fischer (1994) and Fischer (1995a, b) emphasise that the central bank’s discretionary powers may be restricted, while Grilli et al. (1991), Alesina and Grilli (1992) and Cukierman (1992) mainly stress the partial commitment of government. The following sections deal with the meaning of these institutional details. Table 5.1, shows the criteria for assessing the degree of political independence of the central bank. The Grilli, Masciandaro and Tabellini index of political independence corresponds to the number of all fulﬁlled criteria from (P1) to (P8). Criteria (P1) and (P3) assume that central bankers behave independently if they are not appointed directly by the government. Appropriate rules should prevent stable political majorities from leading to political one-sidedness of the monetary policy decision-

CBI and macroeconomic performance 113 Table 5.1 The GMT index of political independence Political independence (P1) (P2) (P3) (P4) (P5)

Governor not appointed by government * Governor appointed for more than 5 years * All the Board not appointed by government * Board appointed for more than 5 years * No mandatory participation of government representative on the * Board (P6) No government approval of monetary policy formulation required * (P7) Statutory requirements that central bank pursues monetary stability * amongst its goals (P8) Legal provisions that strengthen the central bank’s position in * conﬂicts with the government are present Overall index of political independence, constructed as the sum of the asterisks

making body in the course of time. Neumann (1996) emphasises that these criteria can be misleading if the nomination committees are formally, but not actually, independent of government (Neumann 1996, p. 5). Long terms of ofﬁce (P2) and (P4) are aimed at preventing the government from carrying out short-term changes in the composition of the decision-making body. Criteria (P5), (P6) and (P8) refer to the relationship between the government and the central bank with regard to the formulation and implementation of policy. Monetary policy decisions are not independent of government if government representatives have voting rights, or if the government must approve monetary policy decisions before implementation. However, even when fulﬁlling these criteria ((P1) to (P6) and (P8)), the political independence of central bankers need not coincide with political neutrality. Vaubel (1997), for instance, stresses the view that independent central bankers have partisan preferences of their own and may wish to support their preferred parties. According to his ‘party preference hypothesis’, an independent central bank tries to improve the government’s electoral chances if the majority in the central bank council shares the government’s partisan views. Conversely, it tries to prevent the re-election of incumbent politicians if the opposition parties command the majority on the central bank’s board. (See the discussion between Vaubel (1997) and Berger and Woitek (1997) on the question whether the German Bundesbank has partisan preferences.) Criterion (P7) deﬁnes political independence as being the ability of the central bank ‘to pursue the goal of low inﬂation’ (Grilli et al. 1991, p. 367). This deﬁnition is wide-ranging and covers more than independence of the central bank from government. It is true that a legal mandate for monetary stability is supposed to reduce the possibility of short-term political considerations leading to changes in the objectives of monetary policy. But

114

Friedrich Kißmer and Helmut Wagner

it is also true that a legal mandate restricts the discretionary power of the central bank to set the goals of monetary policy. However, the concrete formulation of criterion (P7) chosen by Grilli, Masciandaro and Tabellini weights only price stability as the sole goal of monetary policy and places no weight on other possible goals (see also Akhtar 1995). The ability of the central bank to choose its own monetary policy instruments without interference from government is measured by Grilli, Masciandaro and Tabellini with the help of the following criteria (Table 5.2), which form the index of economic independence. Criteria (E1) to (E4) refer to different characteristics of the direct lines of credit the central bank extends to the government. With respect to direct ﬁnancing of budget deﬁcits, close limits (E4) on credit facilities, or a complete prohibition of direct central bank credits to the government, in particular appear to be effective. Criterion (E5) refers to the purchase of government bonds in the primary market, but not to sale and purchase in the secondary market. Therefore even if (E5) is fulﬁlled, the central bank could ﬁnance budget deﬁcits (indirectly) through purchases of government bonds in the secondary market. Criteria (E6) and (E7) concern the independence of monetary policy from an administrative ﬁxing of the discount rate (E6), and from inﬂuences which can arise from bank supervision (E7). (For a criticism of the economic criteria see e.g. Neumann 1996, p. 5; Eijfﬁnger and De Haan 1996, p. 57.) As with other indices (e.g. Cukierman, Cukierman et al., Alesina or Eijfﬁnger and Schaling), the way central bank independence is inﬂuenced by the exchange rate system is not taken into account explicitly (an exception is an extension of the Cukierman, Webb and Neyapti indicator by Siklos (1995) for ﬁve countries in Latin America). This procedure is justiﬁed by the fact that in nearly all countries the government is responsible for choosing the exchange rate regime (as well as for determining exchange rate parities or upper and lower limits in systems of ﬁxed exchange rates), whereas the central bank simply manages the exchange rate within the limits of the system adopted. The total index of central Table 5.2 The GMT index of economic independence Economic independence (E1) (E2) (E3) (E4) (E5) (E6) (E7)

Direct credit facility: not automatic Direct credit facility: market interest rate Direct credit facility: temporary Direct credit facility: limited amount Central bank does not participate in primary market for public debt Discount rate set by central bank Banking supervision not entrusted to the central bank (**) or not entrusted to the central bank alone (*) Overall index of economic independence (being the sum of the asterisks)

* * * * * * (**) (*)

CBI and macroeconomic performance 115 bank independence results from the sum of the index values for political and economic independence. The most detailed indices for measuring the legal independence of central banks are those developed by Cukierman (1992) and Cukierman et al. (1992). Indices LVAU (unweighted index of legal independence) and LVAW (weighted index of legal independence; see Appendix C) take into account sixteen characteristics of central bank independence which are divided into four clusters. LVAU and LVAW differ from each other through their different weighting of some components. The main groups cover characteristics of the personal independence of the chief executive ofﬁcer (CEO) (term of ofﬁce, appointment, reasons for dismissal), aspects of the central bank’s policy independence (policy formulation, resolution of conﬂicts – PF), the ﬁnal objectives (OBJ) laid down in the central bank statute and the legal restrictions on the ability of the government to borrow from the central bank (LL) (see Cukierman 1992, p. 372). In contrast to Grilli et al. (1991), detailed questions make graded evaluations possible. For example, there are six possible ratings for the question concerning the objective of the central bank, each determined by the weight given to price stability compared with other stated objectives. Following Grilli et al. (1991), a statutory mandate to pursue the goal of price stability is seen against the background that delegating monetary policy to an independent central bank is aimed at mitigating the credibility problem (p. 367). In Rogoff’s (1985) terminology, ‘The objectives variable measures the strength of the “conservative bias” of the bank’s charter’ (Cukierman et al. 1992, p. 357). Table 5.3 is taken from Eijfﬁnger and De Haan (1996, p. 23) and shows the numerical index values for twenty-two industrial countries. The indicators have similar outcomes (the Swiss National Bank and the German Bundesbank are consistently classiﬁed as comparatively independent), but along with differences of detail, there are also more substantive differences (see the classiﬁcation of the Bank of Japan; cf. Pollard 1993, p. 29). As has already been noted on p. 000, the selection and weighting of criteria and the interpretation of statutes are responsible for deviations. It may be said that the indices from Alesina (1988), and Eijfﬁnger and Schaling (1993) place a smaller weight on the central bank’s economic independence than do the indices from Grilli et al. (1991) and Cukierman (1996) (see Eijfﬁnger and De Haan (1996), in particular Table 4, p. 24). Mangano (1997) compares the indices of Grilli, Masciandaro and Tabellini, and Cukierman with each other and concludes that 40 per cent of the criteria in the Grilli, Masciandaro and Tabellini index are not regarded as relevant in the Cukierman index (vice versa the level is 45 per cent). Selection and weighting differences are, of course, important in identifying those aspects of central bank independence which are particularly correlated with macroeconomic performance. However, indices used to measure independence are also subject to ‘interpretation bias’; that is,

116

Friedrich Kißmer and Helmut Wagner

Table 5.3 Legal indices of central bank independence2 Country

Alesina

Grilli, Masciandaro, and Tabellini

EifﬁngerSchaling

Cukierman (LVAU)

Australia Austria Belgium Canada Denmark Finland France Germany Greece Iceland Ireland Italy Japan Netherlands New Zealand Norway Portugal Spain Sweden Switzerland United Kingdom United States

1 – 2 2 2 2 2 4 – – – 1.5 3 2 1 2 – 1 2 4 2 3

9 (3) 9 (3) 7 (1) 11 (4) 8 (3) – 7 (2) 13 (6) 4 (2) – 7 (3) 5 (4) 6 (1) 10 (6) 3 (0) – 3 (1) 5 (2) – 12 (5) 6 (1) 12 (5)

1 3* 3 1 4* 3* 2 5 – – – 2 3 4 3* 2* 2* 3* 2 5 2 3

0.31 0.58 0.19 0.46 0.47 0.27 0.28 0.66 0.51 0.36 0.39 0.22 0.16 0.42 0.27 0.14 – 0.21 0.27 0.68 0.31 0.51

Source: Eijfﬁnger and De Haan (1996, p. 23). Note *Extensions are based on Eijfﬁnger and Van Keulen (1995). Except for Denmark, the ranking of these seven countries refers to central bank laws adjusted during the past ten years.

identical legal foundations are often interpreted in different ways. Mangano (1997) has observed that ‘in the 17 countries included in both Cukierman and Grilli, Maciandaro and Tabellini samples, virtually a third of the values attributed to their nine common criteria are subject to nonnegligible interpretation problems’ (Mangano 1997, p. 7). Mangano (1997) further stresses that for ten out of seventeen countries Cukierman (1992) and Grilli et al. (1991) come to a contradictory conclusion when deciding whether or not the central bank is legally allowed to purchase government bonds in the primary market. Furthermore, when examining the laws of Denmark, France, Greece and Japan, four out of nine criteria are interpreted completely differently and in just one out of seventeen cases (Italy) the interpretation bias disappears (Mangano 1997, pp. 7, 28). The subjectivity of the indices has led, among other things, to different indices being combined. Alesina and Summers (1993) adopted this approach and formed an average index from the Alesina (1988) index and

CBI and macroeconomic performance 117 the Grilli et al. (1991) index. We consider this approach further in Sections 3 and 4. Indicators based on actual behaviour To identify differences between actual and legal central bank independence Cukierman (1992) and Cukierman et al. (1992) calculated the actual turnover rates for central bank governors for the period 1950 to 1989. This indicator is based on the assumption that a higher frequency of change of central bank governors indicates a lower level of central bank independence. A longer term of ofﬁce does not necessarily imply greater independence, but actual terms of ofﬁce that are shorter than the electoral cycle lead to the suspicion of personal dependence. The turnover rate is deﬁned as the average change of central bank governors per year. Because the electoral cycle is four to ﬁve years in most countries, the threshold turnover rate is between 0.2 and 0.25 (Cukierman et al. 1992, p. 402). For twenty-one industrial countries surveyed over the reference period, there was a maximum turnover rate of 0.2 (Japan and Spain), while the turnover rates for twenty-seven of ﬁfty developing countries were over 0.2 with the maximum of 0.93 for Argentina (Cukierman et al. 1992, p. 364). The authors conclude that the turnover rate is a reasonable, but noisy, proxy for the lack of central bank independence in developing countries. Cukierman and Webb (1995) develop an indicator of political vulnerability, deﬁned for each country as the fraction of political transitions that are followed promptly (within six months or within one month) by the replacement of the central bank Governor (Cukierman and Webb 1995, p. 406). For the period 1950 to 1989 they ﬁnd that for a sample of sixtyseven countries (twenty-four industrialised and forty-three developing countries) the average turnover rate is signiﬁcantly higher shortly after political transitions than in any other period. Once again there are considerable differences between groups of countries and also within groups, for different forms of political transitions. In this context, the authors identify four types of political transition: (1) change of regime (from democratic to authoritarian and vice versa); (2) change from one authoritarian government to the next; (3) change of party without a change in regime; and (4) change of head of government (Cukierman and Webb 1995, p. 409). In the case of developing countries vulnerability (within six months) is on average 0.35 in comparison to 0.1 for industrial countries (Cukierman and Webb 1995, p. 406). Developing countries which alternate between democratic and authoritarian regimes (‘mixed countries’) have the greatest degree of vulnerability (0.39). In addition, vulnerability is also dependent on the type of political instability. For example, in the context of developing countries Cukierman and Webb (1995) give a value of 0.61 for vulnerability with type (1) (change of regime), while a lower level of political instability type (4) (change of head of government) receives a value of

118

Friedrich Kißmer and Helmut Wagner

0.26. Cukierman (1996, p. 13) concludes that the political vulnerability of the central bank governor is a noisy indicator of the actual independence of the central bank which may be used, in principle, for an analysis of developing and industrial countries. The quality of this measure is comparatively better for countries and periods with many political changes.

3 Central bank independence and inﬂation The theoretical foundations (see Section 2) of central bank independence lead to the expectation that measures of central bank independence are inversely related to both the mean rate of inﬂation and the variance of inﬂation. Eijfﬁnger and De Haan (1996, Table 3) list twenty-one empirical studies of the relationship between central bank independence and inﬂation which were published between 1988 and 1996. If details are ignored, twenty studies conﬁrm the inverse relationship between central bank independence and the average rate of inﬂation. For Eijfﬁnger and De Haan (1996), the main conclusion is that the measures of central bank independence and inﬂation both in developed and developing countries are inversely related if central bank independence is measured in developed countries with legal indices, and in developing countries with behavioural indices (the turnover rate of central bank governors or the political vulnerability of the central bank) (Eijfﬁnger and De Haan, 1996, p. 33).3 Cukierman (1996) ﬁnds similar results to Eijfﬁnger and De Haan. However, as well as conﬁrming a lower average level of inﬂation when the central bank is independent, he also ﬁnds a lower variance for the rate of inﬂation (Cukierman 1996, p. 15). The following sub-section explains the so-called stylised facts for the statistical association between central bank independence and inﬂation for industrial countries. The next sub-section surveys the evidence on whether the ﬁndings of the previous sub-section are sensitive to changes in the legal indices used to measure central bank independence. The third sub-section focuses on the relationship between central bank independence and inﬂation in developing and transition countries. Finally, the fourth sub-section considers the sensitivity of the results to the degree of central bank independence and assesses the importance of other factors that might explain the differences in inﬂation performance between countries. The so-called stylised facts for industrial countries One of the ﬁrst empirical studies on central bank independence was carried out by Bade and Parkin (1988) who concluded that there existed a signiﬁcant negative correlation between the degree of political independence of central banks, and the average rate of inﬂation in twelve industrial countries in the period 1972 to 1986 (see Section 2 and Appendix A for the Bade and Parkin index). However, they found no correlation between

CBI and macroeconomic performance 119

Average inflation

inﬂation and the degree of ﬁnancial independence of central banks. Neither did this early study establish any relationship between the variance of inﬂation and the degree of central bank independence. Subsequent authors have built upon Bade and Parkin’s (1988) methodology and have developed various indices of central bank independence which have facilitated pure cross-sectional comparisons of economic outcomes. Grilli et al. (1991) examined data from eighteen OECD countries in the period 1950 to 1989, subdivided into decades. They concluded that their indices for measuring economic and political independence of central banks yielded the expected negative correlations with respect to the average rate of inﬂation in all sample periods. However, the indicator for political independence alone (see p. 00) was only signiﬁcant for the 1970s, while the measure for economic central bank independence alone (see p. 00) was signiﬁcant except in the ﬁrst two sub-periods (1950 to 1959; 1960 to 1969) (Grilli et al. 1991, p. 372). Some of the best known scatter plots within the studies of the relationship between central bank independence and economic outcomes are those published by Alesina and Summers (1993). Figure 5.1 shows combinations of average inﬂation rates in the period 1955 to 1988 and the indicator values for central bank independence for sixteen industrial countries. Figure 5.2 shows the relationship between central bank independence and the average variance of inﬂation. The Alesina and Summers (1993) index is based on averaging the Alesina (1988) index and the Grilli et al. (1991) index. The overall index takes values between 1 and 4 (Alesina and Summers 1993, pp. 154ff.).

Index of central bank independence

Figure 5.1 Average inﬂation and central bank independence (1955–1988).

Friedrich Kißmer and Helmut Wagner

Variance inflation

120

Index of central bank independence

Figure 5.2 Variance inﬂation and central bank independence.

The ﬁgures not only show a strong negative relationship between inﬂation and central bank independence, but also an inverse relationship between the variance of inﬂation and the degree of central bank independence on the Alesina and Summers (1993) index.4 A series of studies examined the question whether the statistical associations claimed by Alesina and Summers (1993) were robust if different indicators for central bank independence, different periods and countries, and additional determinants for differences in inﬂation between the countries (e.g. political instability, degree of openness, labour market institutions) are considered. Various measures of central bank independence Fuhrer (1997) and Jenkins (1996) stress among other things that the signiﬁcance of the Alesina and Summers (1993) ﬁndings is lower if, instead of the Alesina and Summers index, the LVAW index from Cukierman (1992) or from Cukierman et al. (1992) is used for the Alesina and Summers sample of countries.5 When this alternative measure of central bank independence is used, the negative correlation between central bank independence and inﬂation remains, but the correlation is comparatively weak and less signiﬁcant (Fuhrer 1997, p. 28; Jenkins 1996, pp. 255ff.). Cargill (1995), using the LVAW index, conﬁrms that the statistical association between central bank independence and inﬂation is not robust even for the group of industrial countries when different countries are sampled or a different time period is considered. Cargill (1995), Jenkins (1996) and Fuhrer (1997) consider only the LVAW index, while Schaling (1995) and

CBI and macroeconomic performance 121 Eijfﬁnger et al. (1997) analyse the sensitivity of the ﬁndings to different legal indices of independence and use different sample periods. We focus here on the more recent of these studies because Schaling only looks at the twelve countries surveyed by Bade and Parkin and uses only the index of political independence established by Grilli et al. in their study.6 Eijfﬁnger et al. (1997) examine the relationship between the degree of central bank independence and macroeconomic performance for twenty countries and include the indices of Alesina (1989), Grilli et al. (1991), Eijfﬁnger and Schaling (1993) (see Appendix B) and Cukierman (1994) (LVAU). The period 1972 to 1992 was divided into two sub-periods (1972 to 1982 and 1983 to 1992) in order to distinguish between EMS and nonEMS countries (Eijfﬁnger et al. 1997, p. 4). The authors conﬁrm the inverse correlation between inﬂation and the degree of central bank independence for all indices, but this correlation is much more signiﬁcant in the ﬁrst sub-period than in the second. Eijfﬁnger et al. (1997) emphasise that in the second sub-period, monetary policy in the EMS countries (except Germany) becomes endogenous because the central banks in these countries were forced to pursue exchange rate targets. The ﬁndings of Grilli et al. (1991), De Haan and Sturm (1992) and Jonsson (1995) support the view that the relation between central bank independence and inﬂation is less straightforward if central banks have to pursue exchange rate targets. For example, Jonsson (1995) studies data for eighteen OECD countries between 1961 to 1989 and concludes that the dampening effect of central bank independence on inﬂation is much stronger under a ﬂoating exchange rate regime than under a ﬁxed exchange rate regime. With regard to the correlation between inﬂation variance and the degree of central bank independence, the analysis by Eijfﬁnger et al. (1997) shows wider differences between the various indices. The view that more central bank independence is also related to less variability of inﬂation is supported here by the LVAU index and by the overall index from Grilli et al. (1991). The low sensitivity of the link between the mean rate of inﬂation and central bank independence to the selection of the index, as shown in the analysis by Eijfﬁnger et al. (1997), does not necessarily imply that the various aspects of central bank independence dealt with in Section 2 are unimportant. On the contrary, the studies by Debelle and Fischer (1994) and De Haan and Kooi (1997) lead to the expectation that the degree of economic independence of the central bank, or its instrument independence, should be regarded as a particularly important feature of the bank’s legal independence. Debelle and Fischer (1994) decompose the indices used by Grilli et al. (1991) into three different indices (see pp. 17–19 for the following characteristics of the Grilli, Masciandaro and Tabellini index): INFOBJ is the value for question (P7) whether there are statutory requirements that the bank should pursue monetary stability among its goals. In the terminology of Debelle and Fischer (1994) this measures the

122

Friedrich Kißmer and Helmut Wagner

lack of goal independence. POL7 contains (P1) to (P6) and (P8) from the Grilli et al. (1991) index as a measure of the central bank’s political independence. EC6 corresponds to the index for economic independence, minus the bank supervision criterion (E7), and its aim is to measure the degree of instrument independence. In a regression for seventeen countries, Debelle and Fischer (1994) conclude that legal provisions on appointment procedures for board ofﬁcials and the length of their term of ofﬁce, as recorded in POL7, are not signiﬁcantly correlated with the inﬂation rate, while INFOBJ and EC6 have a stronger impact on inﬂation. The authors interpret their ﬁndings as giving support to the view that a central bank should be given instrument independence, but not goal independence. However, Debelle and Fischer (1994) also recognise that their interpretation of the ﬁndings is quite daring, because simply replacing the (0, 1) indicator INFOBJ with the more differentiating variable (OBJ) from the Cukierman et al. (1992) index (see Appendix C), cluster objectives (OBJ can take values between 0 and 1) leads to the result that only the extent of instrument independence is signiﬁcantly correlated with inﬂation (Debelle and Fischer 1994, pp. 218ff.). De Haan and Kooi (1997) go one step further, and in an analysis of twenty-one OECD countries for the periods 1972 to 1979 and 1980 to 1989 report that only the extent of instrument independence matters for inﬂation performance. Contrary to what might be expected, they also ﬁnd that the ‘conservativeness’ (i.e. the lack of goal independence in the terminology of Debelle and Fischer (1994)) and other aspects of central bank independence have much less, or no, impact on inﬂation (De Haan and Kooi 1997, p. 24). To test the robustness of this result the authors modify both the Grilli et al. (1991) index and also the LVAU index from Cukierman (1992) and arrive at similar results in both cases. Developing and transition countries The above focused on the relationship between the legal independence of central banks and the distribution of inﬂation in industrial countries. This is in keeping with the vast majority of the literature on the macroeconomic effects of central bank independence. In the following we look at the empirical evidence relating to developing and transition countries. Developing countries We know from the work by Cukierman (1992) and Cukierman et al. (1992) that for developing countries, there is no correlation between legal measures of central bank independence (LVAU or LVAW index) and inﬂation. These ﬁndings are conﬁrmed by De Haan and Siermann (1996) who analysed forty-three developing countries over the period 1950 to 1989, and by Cukierman (1992) and Cukierman et al. (1992) who analyse a

CBI and macroeconomic performance 123 sample of seventy-one countries (twenty-one developed countries and ﬁfty developing countries) for the period 1950 to 1989. These studies not only conﬁrm an absence of any correlation between the legal independence of central banks and inﬂation in developing countries, but they also show that no correlation exists between actual central bank independence and inﬂation. However, the studies by Cukierman (1992) and Cukierman et al. (1992) in particular show that the mean and variance of the inﬂation rate in developing countries are correlated with actual central bank independence if the turnover rate of central bank governors (see above) is used as a proxy for central bank independence. In contrast to this, there is no relation between turnover rate and inﬂation in the industrial countries analysed by Cukierman et al. (1992, p. 373). We return to the question whether aspects other than the extent of central bank independence are responsible for differences in inﬂation between countries in the following sub-section. However, we ﬁrst examine the role of political instabilities because, as stressed in particular by Cukierman (1992), the signiﬁcance of non-legal indices for measuring central bank independence in developing countries leads directly to the question whether political instabilities are responsible for the inﬂation differentials between countries. In some studies it was found that indicators of political instability are positively linked to inﬂation (e.g. Cukierman et al. 1992). In a study of sixty-seven countries, of which forty-three are developing countries, Cukierman and Webb (1995) examine whether instabilities are directly or indirectly linked with inﬂation through political inﬂuence on the central bank. To do this they examine the vulnerability of central bank governors, non-political turnover, and several measures of political instability (see Section 2). The authors conclude that political instabilities mainly have an indirect effect through the vulnerability of the central bank governor on the mean and standard deviation of inﬂation. In addition, there is also a signiﬁcant correlation between non-political turnover and the distribution of inﬂation (Cukierman and Webb 1995, p. 410; Cukierman 1996, p. 14). Transition countries The transition process in CEE countries has been accompanied by reform of their central banks. Previous cross-sectional analyses are less suitable for recording the short-term macroeconomic effects of a change in central bank independence in these countries because transition fundamentally alters the nature and functions performed by the central bank. Consequently, there have been few empirical studies on the relationship between central bank independence and inﬂation in transition economies. Loungani and Sheets (1997) examine the inﬂuence of central banks’ legal independence on inﬂation rates in 1993 for a group of twelve transition countries. They conclude that there is a highly negative correlation

124

Friedrich Kißmer and Helmut Wagner

between central bank independence and inﬂation, even after controlling for other factors such as ﬁscal balance, reform index and the average tenure of central bank chairman (Loungani and Sheets 1997, p. 391). The degree of central bank independence is measured here with the help of a so-called SIB index, which was ‘obtained by assessing the similarity between the characteristics of a given central bank and the characteristics of the German Bundesbank’ (Loungani and Sheets 1997, p. 391). However, the focus on the rate of inﬂation for only a single year, and the fact that many of the central bank statutes which were examined had only recently been enacted casts doubt on how robust these ﬁndings will prove to be. For example, Eijfﬁnger and van Keulen (1995) ﬁnd no signiﬁcant relationship between central bank independence and inﬂation for a total sample of eleven countries including three transition countries (Hungary, Poland and Czech Republic). However, they do ﬁnd that a negative relationship between central bank independence (using the Grilli et al. (1991) index and the Eijfﬁnger and Schaling (1993) index) and inﬂation for those six countries (without any transition country), where the central bank law has been in force for more than ﬁve years. Other determinants of inﬂation It is possible that the correlations between central bank independence and inﬂation are based on an inaccurate speciﬁcation of the role played by other relevant macroeconomic variables. For example, Romer (1993) has argued that inﬂation differences between countries might be explained by the degree of openness (share of imports in GDP) of the economy, because in open economies the incentive to create surprise inﬂation is less than in economies which are less open. Romer’s empirical study of 114 countries shows that for the period after 1973 the degree of openness and the level of inﬂation are inversely linked, but this correlation is less clear in countries which are politically stable (measured against the average number of revolutions and coups per year) or which have established independent central banks (LVAW index). There is no link between inﬂation and the degree of openness for the group of the nineteen wealthiest (other than oil producers) countries (Romer 1993, p. 883). Other possible reasons for differences in inﬂation between countries might, for example, be political instabilities (Cukierman et al. 1992) or aspects of the wage-bargaining structure such as corporatism and centralisation (Bleaney 1996). Various control variables have therefore been included in a series of studies. For example, Grilli et al. (1991) include proxies for political instability, but these have no qualitative effect on the link between central bank independence and inﬂation. Other studies, which take labour market structures into account (Havrilesky and Granato 1993; Bleaney 1996) or which include indicators of the degree of openness of an economy (Al-Marhubi and Willet 1995; Eijfﬁnger et al.

CBI and macroeconomic performance 125 1997), also support the view that the relationship between the indicators for central bank independence and inﬂation is robust (see De Haan 1997, p. 409; De Haan and Kooi 1997, p. 27; Eijfﬁnger and De Haan 1996, p. 31). However, studies by Jenkins (1996), Fuhrer (1997), and Campillo and Miron (1996) reach different conclusions. Jenkins (1996) ﬁnds that the index of central bank independence is no longer signiﬁcant when an index for the corporatist structure of the labour market is included. Using panel data, Fuhrer (1997) ﬁnds that the inﬂuence of central bank independence on the mean inﬂation rate (variance of inﬂation) is only signiﬁcant for a bivariate pure cross-section analysis with the sample of countries analysed by Alesina and Summers (1993). In another study using a sample of sixtytwo countries, Campillo and Miron (1996) show that the mean inﬂation rate from 1973 to 1994 is explained by variables such as the degree of openness of the economy, political instability, debt and inﬂation history (measured against the mean inﬂation rate from 1948 to 1972). These results are important, but one criticism of all three studies is that they use the LVAW index which, in previous studies, gave the lowest correlation with inﬂation. The LVAW index of central bank independence is not signiﬁcant (see also De Haan and Kooi 1997, p. 27).7

4 Central bank independence and the real economy In contrast to the number of studies on the link between central bank independence and inﬂation, the empirical literature on the effect of central bank independence on the real economy is comparatively thin. Nevertheless, two broad approaches have been adopted to address this issue. The ﬁrst asks whether there are direct links between the extent of central bank independence and output (in levels or growth rates), or output variability, and the second asks whether central bank independence is related to the costs of disinﬂation. Central bank independence and economic growth Economic theory suggests several reasons why central bank independence might impact upon the real economy. Volatility and persistence If central bank independence is interpreted in the sense of a Rogoff-type ‘conservative’ banker, the greater inﬂation aversion of an independent central bank should lead to greater output volatility. Insofar as supply shocks produce persistent output effects, an insufﬁcient shock absorption can lead to a reduction of average macroeconomic performance (e.g. Svensson 1997). On the other hand, Alesina and Gatti (1995) show that if central bank independence is associated with a dampening of political

126

Friedrich Kißmer and Helmut Wagner

monetary cycles, the establishment of an independent central bank does not necessarily lead to an increase in output volatility (see pp. 000–000). Inﬂation and economic growth If central bank independence is linked to lower average inﬂation or reduced inﬂation variance, central bank independence can have subsequent effects on the average rate of economic growth. Losses may occur with regard to the level of long-run growth if a lower rate of inﬂation is associated with a smaller capital stock per man. Tobin (1965) shows that by decreasing the inﬂation rate, that is, increasing the real rate of return on money holding, the central bank shifts the structure of the private sector’s portfolio towards money. Given the level of real wealth in the economy, this portfolio shift reduces the capital stock. However, this Tobin effect is based on the assumption that money and capital are substitutes. In the Tobin model, money serves as a store of value. In contrast, cash-in-advance models emphasise the transaction demand for money. If some market activities, such as the purchase of investment goods, are subject to cash-in-advance restrictions, low inﬂation means that these market activities are subject to lower inﬂation tax in comparison to the level that would exist at higher rates of inﬂation. This complementarity of money and capital may then lead to an inverse relationship between inﬂation and economic growth (Stockman 1981). However, in the most recent literature the growth effects of inﬂation are analysed not only with respect to the long-term growth level, but also with respect to the long-term rate of growth. Using endogenous growth models, some authors (e.g. De Gregorio 1993; Marquis and Referet 1991) show that inﬂation reduces the longterm growth rate. More recent empirical studies conﬁrm qualitatively the inverse (but not necessarily the causal) relationship between inﬂation and economic growth. However, the extent of growth losses remains controversial (Barro 1995; Bruno and Easterly 1998; Fischer 1993; Grimes 1991). Most empirical studies have concluded that for industrial countries, neither long-term growth nor the variability of growth correlate with the degree of central bank independence (Eijfﬁnger and De Haan 1996, p. 34). Grilli et al. (1991) ﬁnd no evidence that their indicators for central bank independence correlate with the average growth of the period 1950 to 1987. Figures 5.3 and 5.4 are taken from Alesina and Summers (1993), and show the poor correlation both between average growth and central bank independence (Figure 5.3), and between central bank independence and the average variability of growth (Figure 5.4). Similar results are reported in more recent studies (De Haan and Kooi 1997). Another recent study by Eijfﬁnger et al. (1997) ﬁnds no correlation between central bank independence and the variance of growth in the period 1972 to 1992 (including the sub-periods 1972 to 1982 and 1983 to 1992) for any of the indices considered (Alesina 1989; Grilli et al. (1991);

Average real GNP growth

CBI and macroeconomic performance 127

Index of central bank independence

Variance real GNP growth

Figure 5.3 Average growth and central bank independence.

Index of central bank independence

Figure 5.4 Average growth variability and central bank independence.

Eijfﬁnger and Schaling 1993; Cukierman (LVAU) 1994). A positive correlation with average growth was found solely for the Grilli et al. (1991) index for the sub-period 1983 to 1992. In addition, the coefﬁcients of some indices change sign in the periods observed. An exception to these general ﬁndings is reported by De Long and Summers (1992) who ﬁnd a positive correlation between central bank independence and growth for the period 1955 and 1990.

128

Friedrich Kißmer and Helmut Wagner

There are relatively few studies available for developing countries which suggest a positive correlation between economic growth and central bank independence if central bank independence is measured by means of indices such as the turnover rate of central bank governors or the vulnerability of central bank governors (Cukierman et al. (1993); Cukierman and Webb 1995). These ﬁndings appear to remain robust even if other growth determinants like the initial GDP, the initial education enrolment ratios and indicators for political instabilities are taken into consideration. Central bank independence and disinﬂation costs If the statistical associations between measures of central bank independence and inﬂation or output are accepted as they are shown in most of the empirical studies, having an independent central bank seems to be a ‘free lunch’, at least in the long run. In this context, the observation that apparently the (short-term) real-economic costs of disinﬂation are linked positively with central bank independence is puzzling. The time-inconsistency hypothesis suggests that relatively independent central banks receive a ‘credibility bonus’ since an announced disinﬂation policy has greater public credibility. This implies an inverse relationship between the sacriﬁce ratio (i.e. the sum of output losses divided by the change in inﬂation during a disinﬂation phase) and central bank independence. Furthermore, Ball (1993) has shown that disinﬂation costs are lower with a cold turkey policy (high speed of disinﬂation) than with a longer lasting gradual disinﬂation policy. One explanation of this is that central banks that are independent of political inﬂuences are more able and willing than dependent banks to realise rapid disinﬂation. Despite this, there are also some arguments to suggest the existence of a higher sacriﬁce ratio with greater central bank independence. For given inﬂation expectations, the amount of output loss associated with disinﬂation is inﬂuenced by the slope of the Phillips curve, which can basically be linked via three channels with the degree of central bank independence (Hutchison and Walsh 1997, p. 8.): 1

2

Insofar as central bank independence is linked with lower average inﬂation, central bank independence could be associated with a higher sacriﬁce ratio because of menu costs. In the presence of menu costs and monopolistic competition, ﬁrms balance the costs of adjusting prices against the costs of not doing so. Lower inﬂation rates therefore lead to less frequent price adjustments and, through this, are associated with a larger extent of nominal rigidity and thus with a ﬂatter Phillips curve (Ball et al. 1988). Insofar as more central bank independence is linked with a lower variance of the aggregate inﬂation rate, the degree of central bank independence affects the signal extraction problem emphasised by

CBI and macroeconomic performance 129

3

Lucas (1973). Lucas assumes that economic decisions are based on uncertain perceptions regarding current aggregate prices. Individual suppliers use the information content of their own product prices (‘local prices’) to form rational (conditional) expectations of the aggregate inﬂation rate. Here it can be shown that a reduction in the variance of the aggregate inﬂation leads to a ﬂatter Phillips curve. A similar argument focuses on nominal wage setting behaviour. As Gray (1978) shows, the optimal degree of wage indexation is substantially inﬂuenced by the ratio of the variances of real shocks to monetary shocks. As far as central bank independence is linked with relatively low stabilisation of supply shocks, the optimal degree of wage indexation is reduced when the wage-setters aim for real wages that lead to the lowest possible squared-deviations of employment from the desired level. A lower degree of nominal wage indexation implies a relatively ﬂat Phillips curve. (See Hutchison and Walsh (1997) on a derivation of optimal indexation, if a wage-setting monopoly union, when choosing the optimal degree of indexation, takes into account that the extent of wage indexation inﬂuences the incentive of the central bank to generate a surprise inﬂation.)

A series of empirical studies have concluded that greater central bank independence is linked to higher disinﬂation costs. For example, Debelle and Fischer (1994) show that the sacriﬁce ratio in Germany is generally higher than in the USA, although, according to conventional indices for measuring central bank independence, the German Bundesbank is classiﬁed as more independent than the Federal Reserve. In addition, the authors show that output losses in periods of recession since 1962 are linked positively with the Grilli et al. (1991) index for central bank independence. Walsh (1995) shows, for the member states of the European Community, that a higher degree of central bank independence is associated with a greater real output effect for changes in nominal income growth. Fischer (1996) examines the correlation between central bank independence and the sacriﬁce ratio for a sample of fourteen industrial countries for the period 1960 to 1988 (sub-period 1973 to 1988) using three different measures of central bank independence. He conﬁrms that the sacriﬁce ratio is higher when independent central banks pursued a low speed disinﬂation policy. In contrast to this, in a study of nineteen industrial countries for the period 1960 to 1992, Jordan (1997) concludes that the higher sacriﬁce ratio associated with greater central bank independence cannot be explained by the speed of disinﬂation. He ﬁnds only very weak evidence that more independent central banks disinﬂate more gradually. In addition, his index for nominal wage rigidity is not signiﬁcant. In Jordan’s (1997) analysis the argument put forward by Ball et al. (1988) becomes crucial, because independent central banks have relatively low initial inﬂation rates at the beginning of a disinﬂationary period.

130

Friedrich Kißmer and Helmut Wagner

Furthermore, Jordan (1997) ﬁnds that countries with a higher degree of central bank independence have a ﬂatter sloping Phillips curve during disinﬂations, but not during periods of accelerating inﬂation. In a study of New Zealand, Hutchison and Walsh (1997) examine whether the short-term output-inﬂation trade-off was altered by the New Zealand Reserve Bank Act 1989. Rather than a cross-sectional analysis, this is a study of the consequences of institutional change within a country. The authors conclude that the output-inﬂation trade-off increased after 1989, although they ﬁnd some evidence that the Reserve Bank of New Zealand received a ‘credibility bonus’ after 1989. This ‘credibility bonus’ is evidenced by a reduction of the difference between a survey measure of expected year-ahead inﬂation and the inﬂation forecast published by the Reserve Bank of New Zealand (Hutchison and Walsh 1997, p. 23).

5 Conclusion and some suggestions for further research The empirically observed trend towards more central bank independence is supported to a large extent by modern theory of monetary policy. The main idea of different approaches is that central banks which are independent of governments are able to reduce the inﬂation bias without having to implement completely inﬂexible rules. However, monetary theory does not conclude that establishing an independent central bank alone leads to an improvement in social welfare. To achieve this, either the preferences of the independent central bank must differ from those of the government in an appropriate way, and/or the independently acting central bank must be provided with policy targets or incentive structures. The vast majority of those empirical studies dealing with the relationship between central bank independence and macroeconomic performance conﬁrm a negative correlation between measures of central bank independence and inﬂation. The extent of central bank independence is measured with the help of legal indices for industrialised countries, while for developing countries variables like the actual turnover rate of the central bank Governor or the vulnerability of the central bank Governor are used as proxies for actual independence. The empirical studies do not support the view that increasing central bank independence entails longrun harmful growth effects or higher (long-run) output volatility. However, increased central bank independence might be combined indirectly with higher costs in terms of real economic performance and some studies conclude that disinﬂation costs positively correlate with the degree of central bank independence. These empirical results have been criticised in the literature. An obvious drawback lies in the subjectivity and possible inconsistency of the measures of central bank independence. There is no objective method which can determine whether one index measures central bank independence better than other indices. Furthermore, some authors doubt the

CBI and macroeconomic performance 131 signiﬁcance and robustness of the empirical results. In particular, it is pointed out that differences in inﬂation between countries correlate with other macroeconomic variables and that the importance of central bank independence is overestimated. Apart from the statistical relationship between the degree of central bank independence and macroeconomic variables in particular, the assumed causal relationship is also being questioned. Recent empirical studies try to evaluate whether the behaviour of independent central banks differs from that of dependent central banks by estimating monetary reaction functions. Siklos (1995) surveys ﬁve Latin American countries and concludes that central banks in these countries do not behave signiﬁcantly differently after institutional reforms granting greater central bank independence were enacted, than those in countries which did not enforce central bank reforms. Solveen (1998) estimates reaction functions of central banks in the G7-states and, among other things, does not ﬁnd support for the hypothesis that independent central banks react to changes in the inﬂation rate or the macroeconomic utilisation rate systematically differently from dependent central banks (Solveen 1998, p. 69). Up until now only a few studies dealing with the relationship between central bank independence and monetary reaction functions exist and this is one area where further research might bear fruitful results. In some recent studies the degree of central bank independence is explicitly endogenised and the determinants of central bank independence are searched for (see Eijfﬁnger and De Haan (1996, p. 41); Cukierman (1996, p. 18) for an overview). Posen (1993) claims that the observed relationship between central bank independence and inﬂation can ultimately be explained by a third factor: the effective ﬁnancial opposition to inﬂation (FOI). He argues that monetary policy is guided by political interests; that is, both the extent of actual central bank independence and the inﬂation rate are directly inﬂuenced by the extent of inﬂation aversion and the antiinﬂationary stance embedded in these political interests. Posen (1993) supposes a rather strong anti-inﬂationary stance for ﬁnancial intermediaries in particular and claims that the stronger and more effective the opposition of the ﬁnancial sector against inﬂation, the greater central bank independence will be. Posen (1993) focuses on ‘interests and not institutions’ so that his arguments also become a criticism of the theoretical foundations of central bank independence which sees partial, as well as complete, solutions of the credibility problem primarily in the institutional design of monetary policy (see Section 2).

132

Friedrich Kißmer and Helmut Wagner

Appendix A The Bade and Parkin (1988) indices of central bank independence Table 5A.1 Bade and Parkin’s index of political independence Central bank is highest monetary policy authority

No government members on the central bank council

Some members of the central bank council are appointed independent from the governmenta

Degree of political independence

– – * *

– * * *

– – – *

1 2 3 4

Source: Hamburg (1997, p. 31). Note a Operationalised as proportion of members not appointed directly or indirectly by the government.

Table 5A.2 Bade and Parkin’s index of ﬁnancial independence Budgetary independence

Salaries of central bank members are determined by the central bank

Allocation of Degree of proﬁts is determined ﬁnancial by the central bank independence

– * * *

– – * *

– – – *

Source: Hamburg (1997, p. 31). Notes *Statement is true. – Statement is not true.

1 2 3 4

CBI and macroeconomic performance 133

Appendix B The Eijfﬁnger and Schaling (1993) indicator of central bank independence Question (⫹) of the Bade/Parkin index is replaced by: (⫹ES): Is the bank the sole ﬁnal policy authority (b), is this authority not entrusted to the central bank alone (b/g), or is it entrusted completely to the government (g)? Table 5B.1 Indicator of central bank independence Bank is ﬁnal policy authority

No government ofﬁcial on bank Board

Some Board appointments of government

Policy type (degree of independence)

– (g) – (g) * (b/g) ** (b) ** (b)

– * * * *

– – – – *

1 2 3 4 5

Source: Eijfﬁnger and Schaling (1993, p. 65).

Appendix C Table 5C.1 The Cukierman, Webb and Neyapti (LVAW) (1992) indicator for central bank independence Variable number 1

Description of variable

Weight

Chief executive ofﬁcer (CEO) 0.20 (a) Term of ofﬁce Over eight years Six to eight years Five years Four years Under four years or at the discretion of appointer (b) Who appoints CEO? Board of central bank A council of the central bank board, executive branch, and legislative branch Legislature Executive collectively (e. g council of ministers) One or two members of the executive branch

Numerical coding

1.00 0.75 0.50 0.25 0.00 1.00 0.75 0.50 0.25 0.00

134

Friedrich Kißmer and Helmut Wagner

Table 5C.1 continued Variable number

Description of variable

Weight

(c)

2

3

Dismissal No provision for dismissal Only for reasons not related to policy At the discretion of central bank board At legislature’s discretion Unconditional dismissal possible by legislature At executive’s discretion Unconditional dismissal possible by executive (d) May CEO hold other ofﬁces in government? No Only with permission of the executive branch No rule against CEO holding another ofﬁce Policy formulation 0.15 (a) Who formulates monetary policy? Bank alone Bank participates, but has little inﬂuence Bank only advises government Bank has no say (b) Who has ﬁnal word in resolution of conﬂict?a The bank, on issues clearly deﬁned in the law as its objectives Government, on policy issues not clearly deﬁned as the bank’s goals or in case of conﬂict within the bank A council of the central bank, executive branch, and legislative branch The legislature, on policy issues The executive branch on policy issues, subject to due process and possible protest by the bank The executive branch has unconditional priority (c) Role in the government’s budgetary process. Central bank active Central bank has no inﬂuence Objectives 0.15 Price stability is the major or only objective in the charter, and the central bank has the ﬁnal word in case of conﬂict with other government objectives Price stability is the only objective Price stability in one goal, with other compatible objectives, such as a stable banking system Price stability is one goal, with potentially conﬂicting objectives, such as full employment No objectives stated in the bank charter Stated objectives do not include price stability

Numerical coding 1.00 0.83 0.67 0.50 0.33 0.17 0.00 1.00 0.50 0.00 1.00 0.67 0.33 0.00 1.00 0.80 0.60 0.40 0.20 0.00 1.00 0.00 1.00

0.80 0.60 0.40 0.20 0.00

CBI and macroeconomic performance 135

Variable number 4

Description of variable

Limitations on lending to the government (a) Advances (limitation on non-securitised lending) No advances permitted Advances permitted, but with strict limits (e.g. up to 15 per cent of government revenue) Advances permitted, and the limits are loose (e.g. over 15 per cent of government revenue) No legal limits on lending (b) Securitised lending Not permitted Permitted, but with strict limits (e.g. up to 15 per cent of government revenue) Permitted, and the limits are loose (e.g. over 15 per cent of government revenue) No legal limits on lending (c) Terms of lending (maturity, interest, amount) Controlled by the bank Speciﬁed by the bank charter Agreed between the central bank and executive Decided by the executive branch alone (d) Potential borrowers from the bank Only the central government All levels of government (state as well as central) Those mentioned above and public enterprises Public and private sector (e) Limits on central bank lending deﬁned in Currency amounts Shares of central bank demand liabilities or capital Shares of government revenue Shares of government expenditures (f) Maturity of loans Within six months Within one year More than one year No mention of maturity in the law (g) Interest rates on loans must be Above minimum rates At market rates Below maximum rates Interest rate is not mentioned

Weight

Numerical coding

0.15 1.00 0.67 0.33 0.00 0.10 1.00 0.67 0.33 0.00 0.10 1.00 0.67 0.33 0.00 0.05 1.00 0.67 0.33 0.00 0.025 1.00 0.67 0.33 0.00 0.025 1.00 0.67 0.33 0.00 0.025 1.00 0.75 0.50 0.25

136

Friedrich Kißmer and Helmut Wagner

Table 5C.1 continued Variable number

Description of variable No interest on government borrowing from the central bank Central bank prohibited from buying or selling government securities in the primary market? Yes No

Weight

Numerical coding 0.00

0.025 1.00 0.00

Source: Cukierman et al. (1992, pp. 358ff.) See pp. 360ff. for a description of the way they aggregated the individual components of legal independence. Notes The ranking under each criterion indicates the degree of independence of central banks – the higher the code, the more independent the central bank. a Often the law does not contain a separate provision on the resolution of conﬂict. In those cases, the variable was coded on the basis of the impression from reading the law in its entirety. If the law gives the impression that the government formulates policy guidelines that the bank simply follows, then the ranking is low.

Notes 1 This deﬁnition of economic independence is very similar to the meaning of instrument independence introduced by Debelle and Fischer (1994), who distinguish between instrument independence and goal independence. For more details see pp. 00ff. of this chapter and Debelle and Fischer (1994, p. 197). 2 The Alesina index runs from 1 to 4, the Eijfﬁnger and Schaling index from 1 to 5, the Grilli, Masciandaro and Tabellini overall index from 0 to 16 (index for political independence alone is shown in parentheses) and the Cukierman (LVAU) index from 0 to 1. Higher index values signify greater central bank independence. 3 If a negative or positive relationship between central bank independence and macroeconomic variables is referred to in the context of turnover rates or vulnerability it should be noted that what is meant here by greater central bank independence is lower turnover rates or vulnerability. 4 Of course, laws and constitutions are subject to change. Alesina and Summers stress that central bank reforms in the 1980s were too recent to be relevant for their sample (Alesina and Summers 1993, p. 153). 5 Fuhrer examines the period from 1950 to 1989, while Jenkins analyses the period 1973 to 1988 and other sub-periods. Alesina and Summers dealt with the period 1955 to 1988, but stated that their ﬁndings qualitatively remain the same if the period 1973 to 1988 is selected (Alesina and Summers 1993, p. 154). 6 The weak correlation between inﬂation and the political index found by Schaling (1995) corresponds to the ﬁndings of Grilli et al. (1991). 7 Fuhrer stresses, however, that his ﬁndings apply to the Alesina and Summer index as well (Fuhrer 1997, p. 33).

CBI and macroeconomic performance 137

References Akhtar, M.A. (1995) Monetary Policy Goals and Central Bank Independence, Banca Nazionale del Lavoro Quarterly Review 195, 423–439. Alesina, A. (1988) Macroeconomics and Politics, NBER Macroeconomics Annual, Cambridge, MA, MIT Press, 13–62. Alesina, A. (1989) Politics and Business Cycles in Industrial Democracies, Economic Policy 8, 55–98. Alesina, A. and Gatti, R. (1995) Independent Central Banks: Low Inﬂation at No Cost?, American Economic Review 85, 196–200. Alesina, A. and Grilli, V. (1992) The European Central Bank: Reshaping Monetary Politics in Europe, in Canzoneri, M.B., Grilli, V. and Masson, P.R. (eds), Establishing a Central Bank: Issues in Europe and Lessons from the US, Cambridge and New York, Cambridge University Press, 49–77. Alesina, A. and Summers, L.H. (1993) Central Bank Independence and Macroeconomic Performance: Some Comparative Evidence, Journal of Money, Credit, and Banking 25, 151–162. Al-Marhubi, F. and Willett, T.D. (1995) The Anti-Inﬂationary Inﬂuence of Corporatist Structures and Central Bank Independence: The Importance of the Hump Shaped Hypothesis, Public Choice 84, 158–162. Bade, R. and Parkin, M. (1988) Central Bank Laws and Monetary Policy, mimeo, University of Western Ontario. Ball, L.N. (1993) What Determines the Sacriﬁce Ratio?, in Mankiw, N.G. (ed.), Monetary Policy, Chicago, IL, Chicago University Press, 155–182. Ball, L.N., Mankiw, G. and Romer, D. (1988) The New Keynesian Economics and the Output-Inﬂation Trade-off, Brookings Papers on Economic Activity 1, 1–65. Barro, R.J. (1995) Inﬂation and Economic Growth, Bank of England Quarterly Bulletin 35, 166–176. Berger, H. and Woitek, U. (1997) How Opportunistic are Partisan German Central Bankers: Evidence on the Vaubel Hypothesis, European Journal of Political Economy 13, 807–821. Bleaney, M. (1996) Central Bank Independence, Wage Bargaining Structure, and Macroeconomic Performance in OECD Countries, Oxford Economic Papers 48, 20–28. Briault, C., Haldane A. and King, M. (1997) Independence and Accountability, in Kuroda, I. (ed.), Towards More Effective Monetary Policy, New York, St. Martin’s Press. Bruno, M. and Easterly, W. (1998) Inﬂation Crises and Long-Run Growth, Journal of Monetary Economics 41, 3–26. Campillo, M. and Miron, J.A. (1996) Why Does Inﬂation Differ Across Countries?, NBER Working Paper No. 5540, Cambridge, MA. Cargill, T.F. (1995) The Statistical Association between Central Bank Independence and Inﬂation, Banca Nazionale del Lavoro Quarterly Review 193, 159–172. Cukierman, A. (1992) Central Bank Strategy, Credibility and Independence: Theory and Evidence, Cambridge, MA, MIT Press. Cukierman, A. (1994) Central Bank Independence and Monetary Control, The Economic Journal 104, 1437–1448. Cukierman, A. (1996) The Economics of Central Banking, Discussion Paper Center for Economic Research, No. 9631.

138

Friedrich Kißmer and Helmut Wagner

Cukierman, A. and Webb, S.B. (1995) Political Inﬂuence on the Central Bank: International Evidence, The World Bank Economic Review 9, 397–423. Cukierman, A., Edwards, S. and Tabellini, G. (1992) Seignorage and Political Instability, American Economic Review 82, 537–555. Cukierman, A., Webb, S.B. and Neyapti, B. (1992) Measuring the Independence of Central Banks and Its Effect on Policy Outcomes, The World Bank Economic Review 6, 353–398. Cukierman, A., Kalaitzidakis, P., Summers L.H. and Webb, S.B. (1993) Central Bank Independence, Growth, Investment, and Real Rates, Carnegie-Rochester Conference Series on Public Policy 39, 95–140. Debelle, G. and Fischer, S. (1994) How Independent Should a Central Bank Be?, in Fuhrer, J.C. (ed.), Goals, Guidelines and Constraints Facing Monetary Policymakers, Boston, MA, Federal Reserve Bank of Boston, Conference Series No. 38, 195–221. De Gregorio, J. (1993) Inﬂation, Taxation, and Long-run Growth, Journal of Monetary Economics 31, 271–298. De Haan, J. (1997) The European Central Bank: Independence, Accountability and Strategy: A Review, Public Choice 93, 395–426. De Haan, J. and Kooi, W. (1997) What Really Matters: Conservativeness or Independence?, Banca Nazionale del Lavoro Quarterly Review 200, 23–38. De Haan, J. and Siermann, C.L.J. (1996) Central Bank Independence, Inﬂation and Political Instability in Developing Countries, Policy Reform 1, 135–147. De Haan, J. and Sturm, J.-E.E. (1992) The Case for Central Bank Independence, Banca Nazionale del Lavoro Quarterly Review 182, 305–327. De Long, J.B. and Summers, L.H. (1992) Macroeconomic Policy and Long-Run Growth, Federal Reserve Bank of Kansas City Economic Review 5–29. Deutsche Bundesbank (1998) Monatsbericht Januar 1998, Frankfurt am Main. Eijfﬁnger, S.C.W. and De Haan, J. (1996) The Political Economy of Central-Bank Independence, Special Papers in International Economics No. 19, Princeton, NJ. Eijfﬁnger, S.C.W. and Schaling, E. (1995) Central Bank Independence in Twelve Industrial Countries, Banca Nazionale del Lavoro Quarterly Review 184, 49–89. Eijfﬁnger, S.C.W. and Van Keulen, M. (1995) Central Bank Independence in Another Eleven Countries, Banca Nazionale del Lavaro Quarterly Review 192, 39–83. Eijfﬁnger, S.C.W., Schaling, E. and Hoeberichts, M. (1997) Central Bank Independence: A Sensitivity Analysis, Discussion Paper Center for Economic Research, No. 9710. Fischer, A.M. (1996) Central Bank Independence and Sacriﬁce Ratios, Open Economies Review 7, 5–18. Fischer, S. (1993) The Role of Macroeconomic Factors in Economic Growth, Journal of Monetary Economic 32, 485–512. Fischer, S. (1995a) Central Bank Independence Revisited, American Economic Review 85, Papers and Proceedings, 201–206. Fischer, S. (1995b) The Unending Search for Monetary Salvation, in Bernanke, B.S. and Rotemberg, J.J. (eds), NBER Macroeconomics Annual 1995, Cambridge, MA, 274–286. Fuhrer, J.C. (1997) Central Bank Independence and Inﬂation Targeting; Monetary Policy Paradigms for the Next Millennium?, New England Economic Review 1/2, 19–36.

CBI and macroeconomic performance 139 Gray, J.A. (1978) On Indexation and Contract Length, Journal of Political Economy 86, 1–18. Grilli, V., Masciandaro, D. and Tabellini, G. (1991) Political and Monetary Institutions and Public Financial Policies in Industrialized Countries, Economic Policy 13, 341–392. Grimes, A. (1991) The Effects of Inﬂation on Growth: Some International Evidence, Weltwirtschaftliches Archiv 127, 631–644. Hamburg, B. (1997) Central Bank Independence and the European Monetary Union, IGS Dicussion Papers, Series Number 97/15. Havrilesky, T. and Granato, J. (1993) Determinants of Inﬂationary Performance: Corporatist Structures vs Central Bank Autonomy, Public Choice 76, 249–261. Hutchison, M.M. and Walsh, C.E. (1997) Central Bank Institutional Design and the Output Cost of Disinﬂation: Did the 1989 New Zealand Reserve Bank Act Affect the Output-Inﬂation Tradeoff?, Working Paper No. PB97–02, Center for Paciﬁc Basin Monetary and Economic Studies, Economic Research Department, Federal Reserve Bank of San Francisco. Jenkins, M.A. (1996) Central Bank Independence and Inﬂation Performance: Panacea or Placebo?, Banca Nazionale del Lavoro Quarterly Review 197, 241–270. Jonsson, G. (1995) Institutions and Macroeconomic Outcomes – The Empirical Evidence, Swedish Economic Policy Review 2, 181–212. Jordan T.J. (1997) Disinﬂation Costs, Accelerating Inﬂation Gains, and Central Bank Independence, Weltwirtschaftliches Archiv 133, 1–21. Loungani, K.P. and Sheets, N. (1997) Central Bank Independence, Inﬂation, and Growth in Transition Economies, Journal of Money, Credit, and Banking 29, 381–399. Lucas, R.E. (1973) Some International Evidence on Output-Inﬂation Tradeoffs, Journal of Economic Theory 4, 326–334. McCallum, B.T. (1995) Two Fallacies Concerning Central Bank Independence, American Economic Review 85, Papers and Proceedings, 207–211. Mangano, G. (1997) Measuring Central Bank Independence: A Tale of Subjectivity and of Its Consequences, Cahiers de recherches economiques 9704, University de Lausanne. Marquis, M.H. and Referet, K.L. (1991) Real Interest Rates and Endogenous Growth in a Monetary Economy, Economic Letters 105–109. Neumann, M.J.M. (1996) Problems in Measuring Central-Bank Independence, Discussion Paper B 352, Sonderforschungsbereich 303, Universität Bonn. Nolan, C. and Schaling, E. (1996) Monetary Policy Uncertainty and Central Bank Accountability, Working Paper Series No. 54, Bank of England, London. Pollard, P.S. (1993) Central Bank Independence and Economic Performance, The Federal Reserve Bank of St. Louis Review 75, 21–36. Posen, A. (1993) Why Central Bank Independence Does Not Cause Low Inﬂation: There Is No Institutional Fix for Politics, in O’Brien, R. (ed.), Finance and the International Economy 7, Oxford. Radzyner, O. and Riesinger, S. (1997) Central Bank Independence in Transition: Legislation and Reality in Central and Eastern Europe, Vienna Focus on Transition 2, 57–90. Rogoff, K. (1985) The Optimal Degree of Commitment to an Intermediate Monetary Target, Quarterly Journal of Economics 100, 1169–1689.

140

Friedrich Kißmer and Helmut Wagner

Romer, D. (1993) Openness and Inﬂation: Theory and Evidence, Quarterly Journal of Economics 4, 869–903. Schaling, E. (1995) Institutions and Monetary Policy: Credibility, Flexibility and Central Bank Independence, Aldershot, Edward Elgar. Schaling, E., Hoeberichts, M. and Eijfﬁnger, S. (1998) Incentive Contracts for Central Bankers under Uncertainty: Walsh-Svensson Non-Equivalence Revisited, Discussion Paper Center for Economic Research, No. 9811. Siklos, P.L. (1995) Establishing Central Bank Independence: Recent Experiences in Developing Countries, Journal of International Trade & Economic Development 4, 351–384. Solveen, R. (1998) Der Einﬂuß der Unabhängigkeit auf die Politik der Zentralbanken, Kieler Studien 288, Mohr Siebeck, Tübingen. Stockman, A.C. (1981) Anticipated Inﬂation and the Capital Stock in a Cash-inadvance Economy, Journal of Monetary Economics 387–393. Svensson, L.E.O. (1997) Optimal Inﬂation Targets, ‘Conservative’ Central Banks and Linear Inﬂation Contracts, American Economic Review 87, 98–114. Tobin, J. (1965) Money and Economic Growth, Econometrica 33, 671–684. Vaubel, R. (1997) The Bureaucratic and Partisan Behavior of Independent Central Banks: German and International Evidence, European Journal of Political Economy 13, 201–224. Wagner, H. (1990) Demokratie und Inﬂation: Eine ‘rationale’ wahlpolitische Theorie eines Inﬂationbias, Jahrbücher für Nationalökonomie und Statistik 207, 356–373. Wagner, H. (1998) Central Banking in Transition Countries, IMF Working Paper 98, Washington, DC. Walsh, C.E. (1995) Central Bank Independence and the Costs of Disinﬂation in the EC, in Eichengreen, B., Frieden J., and von Hagen, J. (eds), Monetary and Fiscal Policy in an Integrated Europe, Berlin, Heidelberg, and New York, Springer-Verlag, 12–37.

6

Central bank independence and inﬂation performance in transition economies Janet Ilieva, Andros Gregoriou and Nick Tsitsannis

1 Introduction The newly established central banks in transition economies followed the monetary design of the European Central Bank (referred to hereafter as the ECB). This has become particularly evident in the mid- and late 1990s when most of the legislative reforms resulted in increased independence of the central banks from the central governments. The rationale behind this trend of increased CBI is multi-fold: for some transition economies it stems from their desire to join the European Union and the acquis communautaire that should be adopted by the applicant countries. Another incentive comes from the International ﬁnancial institutions (e.g. the World Bank and the International Monetary Fund), where loans are conditional unless the necessary criteria are met. The national governments, on the other hand, to better demonstrate the progress of the conducted reforms in the country, are likely to use the CBI to signal foreign investors the nation’s creditworthiness to potential investors. Following Maxﬁeld (1997, p. 4) the above stems from: (1) the greater their country’s need for balance-of-payments support, (2) the higher the expected effectiveness of signalling, (3) the more secure the tenure of the politicians, and (4) the fewer their country’s restrictions on ﬁnancial transactions. This chapter contains three main sections: the ﬁrst looks at the prerequisites for CBI in transition economies and compares the legislative design of the central banks with the Statute of the ECB. It shows that following the legislative approach towards the CBI, all the transition economies in this sample, with a few exceptions, have highly independent central banks. It provides a discussion of the institutional reforms in twenty-three transition economies. The sample covers all the Central and Eastern European countries (referred to hereafter as CEE). The Czech Republic, Hungary, Poland and Slovak Republic became OECD members in the mid-1990s (respectively 1995 for the Czech Republic, 1996 for Hungary and Poland, 2001 for Slovak Republic). To avoid double entry they are referred to as transition economies.

142

Janet Ilieva et al.

The second section discusses the CBI in transition economies as measured by the index constructed in this study. Comparisons between different CBI indices used in transition economies are made. A triangulation technique is used to test whether central bankers and the public (academics were used to triangulate the data within this study) perceive the independence of the central bank in the same way and whether there are some patterns in the understanding of CBI by these two groups in CEE and CIS countries. The third section looks at the implication of CBI on inﬂation performance. It studies the impact of the components of the CBI index on average inﬂation aiming to detect the relationship between the former and the latter. This is done with the use of panel regression. The ﬁnal section summarises the major ﬁndings.

2 Institutional developments in transition economies The constitution of the Bundesbank, which inﬂuenced the design of the ECB, was widely used as a model for designing the new central banks in CEE and the Commonwealth of Independent States. The Bundesbank provided the practical example of an independent central bank that maintained low average inﬂation during the post-war period.1 Another rationale of having an independent central bank, as mentioned above, emanates from the desire of the applicant country for an earlier accession to the EU. The following sub-section compares the legislative framework of the central banks in transition economies with the Statute of the ECB aiming to capture the legislative similarities between them. It thus attempts to outline the causes behind the deﬁciency of the indices based on the legislative approach per se when applied to measure the CBI in transition economies.

Central bank independence as a legislative provision in transition economies The literature on CBI is related to the work of Barro and Gordon (1983) and Rogoff (1985), arguing that central governments are tempted to ‘sacriﬁce’ long-term welfare for the sake of short-term political gains. Therefore establishing a politically independent central bank is a prerequisite for lower average inﬂation. In line with these studies, Alesina and Grilli (1992) show that, following Rogoff’s delegation (1985), the European median voter has an incentive to appoint central bankers who are more ‘inﬂation-averse’ than are the voters per se. In line with the above, the Maastricht Treaty granted the ECB a degree of independence widely regarded as comparable to the Bundesbank. Article 108 of the Treaty declares that:

CBI and inﬂation in transition economies 143 neither 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 government of a Member State or from any other body. In line with the above, the acquis communautaire of Stage 2 of the European Monetary Union which must be adopted by new member states required: Central Bank independence, co-ordination of economic policies, and adherence to the relevant provisions of the ‘Stability and Growth Pact’. New member states will have to forego any direct central bank ﬁnancing of public sector deﬁcits as well as privileged access of public authorities to ﬁnancial institutions.2 The independence of the central bank is stated in the legislative provision in almost all the transition countries participating in this survey. Countries with more recent accession ‘plans’ therefore have greater incentives to legislatively grant their central banks greater independence. Speciﬁc aspects of the legal independence are explored in the following sub-sections. Autonomy of the monetary policy formulation Following Grilli and colleagues (1991), the political independence of the central bank relates to its ability to choose autonomously objectives and to make decisions on monetary policy. The Statute of ECB only allows a member of the European Commission to participate, without having voting or veto rights, in meetings of the General Council.3 Furthermore, the Statute does not envisage a formal approval of monetary policy. Seventeen transition economies’ banks (out of twenty-three) in this study report that there has not been any interference from central government in formulation of monetary policy within the past ﬁve years and only eleven (less than 50 per cent) state that there is no governmental or political control of the central bank’s budget. Attaining monetary policy autonomy depends on the term of ofﬁce of central bankers, which must be sufﬁciently long to avoid the patronage involved in renewal of the term of ofﬁce. The Executive Committee of ECB is appointed for an eight-year period of time. In our opinion the actual term of ofﬁce does not matter provided it exceeds the maximum term of ofﬁce of the central government. The term of ofﬁce of the central bank governor exceeds the longest of the central government in nineteen countries (except for Armenia, Azerbaijan, Bosnia and Herzegovina,4 and Kazakhstan). Regarding the term of ofﬁce of the bank’s Board, this principle is kept in sixteen countries. The term of ofﬁce in transition countries in our sample varies between ﬁve and eight years.5

144

Janet Ilieva et al.

Another important ‘behavioural’ aspect of the index is whether there are non-completions of the term of ofﬁce of the central governor. Cukierman (1992, p. 383) uses the turnover rate of central bank governors as a proxy for actual independence. His presumption is that a higher turnover of central bank governors indicates lower CBI. Thirteen transition countries (65 per cent) in this sample report one or more non-completions of the term of ofﬁce of their central bank governors. However, this question allows different interpretations, and the practices of different countries’ banks provide interesting examples. Quite unique is the case of the Czech National Bank (referred to hereafter as CNB).6 Ofﬁcially the Bank’s governors have completed their mandates as governors, and as such no one ended the term of ofﬁce in advance or on being released by the government or Parliament. Drawing upon the CNB Act it is only the President who is empowered to dismiss the Governor and CNB Board members. However, the reality is rather different: in fact, there has been one Governor only, Mr Josef Tosovsky, since 1990. From January 1990 to December 1992 he headed the SBCS (central bank of former Czechoslovakia), so he ﬁnished his ofﬁce in advance (the term of ofﬁce was six years) because of the split of Czechoslovakia at the end of 1992. He was then reappointed for another six years by the President as Governor of the newly established CNB. In December 1997 he resigned due to the political turbulences in the Czech Republic to take the post of Prime Minister of the so-called ‘non-political’ or ‘bureaucratic’ government, replacing Vaclav Klaus in his ofﬁce, who had to resign because of a corruption scandal in his party. The CNB was without a Governor from January 1998 until July 1998, a period during which the Vice-Governor Mr Kysilka was responsible for managing the Bank. Following the early elections in June 1998, when social democrats headed by Mr Zeman came to power, Tosovsky has been appointed again as Governor (since July 1998). Objectives The rationale of a single policy objective instead of a plurality of objectives is to eliminate the scope for central governments to interfere with demand for short-term objectives aiming at short-term output before election period. The primary objective of the European System of Central Banks (referred to hereafter as ESCB), as stated in Article 105 of the Treaty, is to maintain price stability. Article 14 of the Treaty obliges the member states to ensure that the statutes of its national bank are compatible with the provisions of the Treaty. This resulted in reforms that started in central banks’ statutes across the European Union (EU) member states as well as in the associate members from Central and Eastern Europe. This tendency is revealed by the legislative amendments to the central banks’ charters in transition economies such as Hungary (1996), the Czech

CBI and inﬂation in transition economies 145 Republic (with the 1995 amendment), Poland (1997), Bulgaria (1997) and others. Nearly all CEE countries have closely modelled their legislation on the ECB and there are many legislative similarities. Regarding the primary objective, with a few exceptions (Armenia, Azerbaijan and Estonia), the transition economies in this sample chose the maintenance of price stability. Least is the bank’s independence with regard to the autonomy of the Bank’s budget, a subject of political or governmental control, which is the case for half of the banks in our sample. Financing the central government This sub-section and the following concern the economic independence of the central bank (as formulated by Grilli et al. (1991), i.e. the ability of the central bank to choose the monetary instruments in pursuing its goals without government interference. The most signiﬁcant constraint regarding the use of monetary policy instruments is related to ﬁnancing the public deﬁcit. Article 21.1 explicitly prohibits the ECB from participating in the primary market for public debt as well as any type of credit facility ‘in favour of Community institutions or bodies, central governments, regional, local or other public authorities, other bodies governed by public law, or public undertakings of Member States’. However, only eleven countries’ banks in this sample do not provide advances for ﬁnancing the central government and thirteen central banks do not participate in the primary market for public debt. Banking supervision According to its Statutes the ECB is not expected to play an important supervisory role. Only two central banks in our sample (Hungary and Bosnia and Herzegovina) have separate agencies for prudential supervision of the ﬁnancial institutions. However there is no consensus in the economic literature as to whether the central bank should be involved in the commercial banks’ supervision. Some economists believe that, as long as the central bank is acting as a lender of last resort, supervision should be entrusted to the central bank. Landau and Garber (1992, p. 102) argue that if the central bank is lending to credit institutions, it should be responsible for the commercial banks’ supervision. The rationale for this is that by delegating to the central bank a supervisory and regulatory role, it can assess accurately the solvency of eventual borrowers, thereby defending its own interests. However, this issue remains unresolved in the economic literature in terms of CBI, although some authors believe it is of no particular importance (Fischer, 1995). Interference in the lender-of-last-resort function of the central bank is an important aspect of CBI, especially in countries engaged in the privatisation of the banking sector or where the banking sector is dominated by

146

Janet Ilieva et al.

the state. In the face of a liquidity crisis, many transition governments are unwilling to allow state banks to go bankrupt, particularly if those banks are on the privatisation list. In countries retaining a major state banking sector, there is even less chance that a bankruptcy could take place. Most of the transition countries experienced such interference during the banking crises in the mid- and late 1990s (the early 1990s for Poland and the Baltic States, 1997 for Bulgaria and so on). The central bank then had to reﬁnance the banks with liquidity problems at the cost of its antiinﬂationary stance. More than one-third of the countries in our sample indicated interference in the lender-of-last-resort function. Summarising the above sub-sections, the transition economies have fully embraced the current Western approach and provided their central banks with considerable legislative independence from the central government. Many governments in transition economies are trying to fulﬁl the requirements imposed by the international institutions in order to obtain debt rescheduling, further loan facilities or/and quicker accession to the EU. Some governments in transition economies used to refer to decrees to oblige their central bank to absorb excess supply of government securities, although this may be in an explicit conﬂict with the central bank constitution. To demonstrate how successfully the economic transformation is progressing, the legislative system of the country perfectly reﬂects the recommendations given by international institutions, but in practice Parliament (or government) in the respective country used to issue ‘other law’ or a decree which was applied when the existing legislation was not appropriate for the goals of the government. In conclusion, the transition economies in this sample have central banks with very high legislative independence. On the other hand, the evidence from these countries suggests that their central banks are often pressed to follow instructions given by the government although they conﬂict with the existing legislation. Therefore, in some countries, the legislative independence, taken on its own, does not reveal the true picture.

3 Measuring central bank independence in transition economies The assumption made when constructing the CBI index was that the legislative and actual independence might diverge when there are traditions in the country with the power of legislation, or existence of other legislation, which may conﬂict with the central bank charter. These two aspects of CBI were taken into account when attempting to measure it. Furthermore, the understanding of the respondents is checked whenever possible, and this is done in this study by employing a triangulation technique. Researchers from two institutions independent of each other in each country were approached to participate in this survey: experts from the central bank and an academic institution, who were qualiﬁed to answer

CBI and inﬂation in transition economies 147 the relevant questions. The rationale behind doing so was to test whether the central bankers overstated their independence or vice versa, as well as detecting patterns, if any, in the responses of the two respondents’ groups. Most of the research on CBI in transition economies has been done during recent years. Dvorsky (2000) in her study of ﬁve CEE countries (the Czech Republic, Hungary, Poland, Slovakia and Slovenia) updates the governors’ turnover rate estimates of Radzyner and Riesinger (1997), aiming to measure the actual independence. Given the short period of observation (eleven years) the author warns that the estimates should be treated with caution. The countries in this sample show extremely low turnover rate (e.g. Slovenia is below the lower threshold, which is 0.25). The author concludes that the governors’ turnover rate (introduced as a proxy for actual independence by Cukierman (1992, p. 383)) does not reﬂect the actual CBI in transition economies. Earlier studies have been done by Loungani and Sheets (1997), who found an inverse relationship between their CBI index and the average inﬂation in 1993. More light has been shed by the recent publications of Cukierman et al. (2001), Maliszewski (2000) and Hochreiter and Kowalski (2000). The latter authors look at the central banks’ statutes in ten CEE countries (Bulgaria, the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, Slovenia and Slovak Republic) which, except for Romania, have adapted their legislation to resemble that of the European Central Bank (ECB). Furthermore, the analysis of the CEE countries’ central banks in question shows that the democratic accountability is comparable to that of the ECB. In line with Hochreiter and Kowalski (2000), Cukierman et al. (2001) extend the CBI study on twenty-six transition economies and conclude that ‘the levels of their legal independence are substantially higher, on average, than those of developed economies during the eighties’. Second, the indices used in the literature were designed for OECD countries and do not capture the picture in transition and developing economies. Referring to Section 2 of this chapter, transition economies have fully applied the legislation of the EU countries whereas the central bank practice there suggests that the rule of law is arguably observed. The following sub-section introduces a new index of CBI, which aims to deal with limitations of previous indices when applied in transition economies. New index of central bank independence The index provided in this section is structured into ﬁve categories: (1) objectives and policy formulation; (2) appointments of the governor and board of the central bank; (3) legislation and tradition; (4) the banking sector; and (5) ﬁnancing the government. The questions provided in each category approach the issues of independence from both a legislative and

148

Janet Ilieva et al.

behavioural perspective. There is an attempt by the new indices to embrace the recommendation of Elgie and Thompson (1998) to allow comparisons to be made ‘across time and across countries’. Objectives and policy formulation 1

2

The objectives of each central bank are of key importance in its overall performance. Price stability as a sole objective of a bank is given priority by most indices (GMT, CWN) and is the major objective of the European Central Bank (ECB). This objective is also embedded in the legislative amendments of many central banks’ charters in transition economies such as Hungary, the Czech Republic (with the 1995 amendment), Poland (1997), Bulgaria (1997), Romania (1995) and others. The reason for having one clearly identiﬁed objective instead of a plurality of objectives is to eliminate the possibility of governments intervening to exploit short-run trade-offs between inﬂation and growth (or unemployment) before elections. Goodhart (1994) suggests that having price stability as a single objective may also reduce conﬂict between central banks’ ofﬁcials and monetarists over the most appropriate operational techniques. Autonomy of the policy formulation is another key feature of independent central banks. The ﬁve-year period chosen in this index is more or less the average term of ofﬁce of governor or board. The rationale of selecting a certain period is that the maximum terms of ofﬁce in different countries vary from country to country and is usually between three and eight years. The last question in this cluster traces presence of governmental or political or other forms of control over the budget of the central bank. This issue has proved to be crucial for the Czech National Bank, where the independent budget of the Bank has been debated continuously in Parliament. However, the suggested amendments in the Bank’s charter, envisaging the government’s control over the Bank’s budget, were ﬁnally overruled by the Supreme Court of the Czech Republic in June 2001 as not being in line with the Constitution of the country. The difference the second point makes is that it traces for evidence of actual interference in the policy formulation coming from the government. Thus the score must be based on what has actually happened rather than on what is stated in the legislation.

Appointments 1

Issues related to appointments have been well outlined in GMT and CWN indices. Provided the term of ofﬁce is longer than ﬁve years (as required in the GMT index) or eight years (CWN index), it does not

CBI and inﬂation in transition economies 149

2

3

4

necessarily mean that it is longer than the maximum term of the government. The more important issue is not the actual term of ofﬁce of the Governor and central bank Board, but whether it exceeds the maximum term of ofﬁce of the government, and therefore the new indices highlight this aspect of the terms of ofﬁce. There is a possibility that some of the important issues related to the central bank are stated not in the central bank charter, but in the Constitution of the country or elsewhere, and this may constrain the freedom granted initially to the central bank. This concern becomes very relevant in cases when dismissal of the Bank’s governor is in question (as described by Cukierman (1992) with the Governor of the Bank of Argentina). The bodies making the appointments for the positions of Governor and Board members must be clearly identiﬁed. Legally the appointments may come from Parliament/Congress (when not from the central bank itself), but at the same time it could be the government, as the only body entitled to nominate names that Parliament or the bank later approve. For instance, the Governor in Romania, although legally elected by Parliament over the past few years, in fact, it was the Prime Minister who nominated candidates for this post. Independence is clearly maximised when the appointments of the Governor and central bank Board are made within the central bank, which on the other hand has been widely criticised as an undemocratic act, allowing concentration of too much power in a handful of non-elected bureaucrats. The Governor should not hold an ofﬁce outside the central bank (governmental or political) that might affect his/her decisions with regard to the management of the bank. The ﬁnal issue covered in this cluster looks at whether there are government representatives present on the Bank’s board, irrespective of whether they are with or without voting and/or veto rights.

Legislation and tradition There are several ways through which the central bank law may be overruled, the most common among them being the power of traditions, which conﬂict with the existing legislation; alternatively, the same aim may be achieved by a parallel law dealing with the same matters as the bank’s charter. Another way is by a change in the legislation per se, when the existing charter is no longer convenient for the Chief Executive. And ﬁnally – not as common as the above – direct breaches of the Bank’s legislation by the Chief Executive. Elster (1979, p. 93) argues that the ‘rule of law’, following Max Weber and others, and a ‘stable institutional framework’ are essential for a growing economy, mainly because of the ability to predict. There are two

150

Janet Ilieva et al.

sources of uncertainty about future decisions stemming mainly from two sources: ambiguous law or constantly changing law. The latter outlines the need for constant law as far more essential than the need for just a law. 1

2

The changeable nature of the legislation or the ‘rule of law’ in transition economies is an important issue, which has not been captured by the existing indices. The more changeable the legislation, the less the scope for the central bank to behave independently in monetary policy-making (see Table 6.1 for details). The recent amendments in the legislation of the central banks in Bulgaria, the Czech Republic, Poland and Romania are indicated in the table, suggesting that the constancy of law in the studies of transition economies is far below the desired one. It is likely that traditions in some transition and developing economies play as important a role as the legislation per se, but, even so, they are not part of the legislation, and therefore they have not been captured by previous indices. Sometimes they may conﬂict with the existing legislation ad hoc affecting the legislatively based independence of the central bank. This appears to be among the main causes for the gaps between formal and actual independence. The well-known case of the Central Bank of Argentina, described by Cukierman (1992), provides a good illustration of how a central bank charter can be overruled by

Table 6.1 Recent amendments of central bank law in Bulgaria, the Czech Republic, Poland and Romania Country

Amendments

Under review

Bulgaria

State Gazette 50/91 State Gazette 32/96 State Gazette 46/97 Act No. 130/1989 Act No. 22/1992 Act No. 6/1993 1995 Act No. 9, 30/06/95 Act No. 47, 25/02/92 04/04/97 Law No. 34, 1991 Rule No. 3, 07/95 1999

No (until the expiry of the Currency Board Arrangement) No

The Czech Republic

Poland Romania

Yes Yes

Sources: Bulgarian State Gazette 46, 1997; S. Cerna, L. Donath and B. Dima (1999) ‘Central Banking in Transition Economies: The Case of Romania’; Huterski, R., Nicholls, R. and Z. Wisniewski (1999), ‘Central Bank Independence in Poland’; Soukup, P., Taci, A. and R. Matusek (1999), ‘Central Bank Independence and Macroeconomic Performance in the Czech Republic’ in ‘Central Banking in Transition Economies’ (Eds: Healey, N. and Z. Wisniewski), Torunska Szkola Zarzadzania, Torun.

CBI and inﬂation in transition economies 151

3

the power of traditions, which was the case of the countries studied in the pilot survey. The existence of other laws in the country, interfering with the central bank charter, are broadly associated with the economic transition of Eastern Europe and the CIS (e.g. the annual budget law). The Bulgarian central bank provides a unique case during the period preceding the introduction of a currency board arrangement. Although independence was guaranteed by the bank’s charter, the government used to refer to the annual law for the budget, and, supported by Parliament, it obliged the Bulgarian central bank to absorb the excess supply of long-run securities. This problem is present when the government wants to show to the public and international institutions lower inﬂation commitment, and equally, it is not ready to give up the old ‘habits’ which assured easy money access. Alternatively, parallel legislation, treating the same issues as that of the central bank’s, may result from the pressure for change coming from international institutions, for which the government is not prepared. Many governments in transition economies are trying to fulﬁl the requirements imposed by the World Bank, International Monetary Fund and European Bank for Reconstruction and Development in order to obtain debt rescheduling or further loan facilities. Thus to demonstrate how successfully the economic transformation is progressing, the legislative system of the country is adjusted according to the recommendations of the international institutions. In reality, there used to be in power ‘another’ law that was applied when the existing legislation was not suitable for the goals of the government.

The banking sector 1

2

The relationship between the central bank and the commercial banks is important in terms of independence. The latter is higher when the central bank is not involved in supervision of the commercial banks. Historically, it is linked with the lender-of-last-resort function of the central bank, which guarantees the well-being of the individual commercial banks acting as a lender of last resort. This role was performed in different ways in the different countries. In some countries, such as Germany and Switzerland where the bank was publicly funded, the banking supervision was entrusted to a separate body outside the CB and hence it was not entitled to be the lender of last resort. In other countries, such as England, France, Italy and others, where the bank was funded through private shareholders’ capital, the bank was entitled to the role of lender of last resort and respectively to the commercial banks’ supervision. Government interference in the bank’s lender-of-last-resort function is mostly associated with transition countries privatising the banking

152

Janet Ilieva et al. sector or state enterprises. In the face of a liquidity crisis, many transition governments are unwilling to allow state banks to go bankrupt, particularly if the bank is on the privatisation list. In countries retaining a major state-owned banking sector, there is even less chance that a bankruptcy would take place. Most of the CEE countries experienced such interference during the banking crisis in 1993 for some countries and 1996 for others. The central bank then had to reﬁnance the banks with liquidity problems, at the cost of its anti-inﬂationary stance.

Financing the government 1

This section mainly looks at facilities that the central bank provides when ﬁnancing the central government. Independence is best served when automatic advances to the government are not permitted, the terms of lending are under the control of the bank and it does not participate in the primary market for public debt.

Summarising the above, the new index put forward in this study is presented below. The coding technique employed in this study differs from those employed in the previous CBI indices aiming at a better clarity of the index. Each question was allocated same weight (i.e. all questions contribute equally to the overall score). The reason behind not allocating weights to the questions was partly due to the critiques raised by Eijfﬁnger and Schaling (1993) and Mangano (1998) regarding the weighting spread, which reﬂects the extent to which the ﬁnal value of the index is affected by the weight attributed to each individual criterion. Furthermore, this index is based on the assumption that the more criteria are met by the respective central bank, the more independent it is according to this index, whereas the criteria included in the weighted indices result in different scores. The questions covered in the index are ticked in a manner to demonstrate absolute independence.

Central bank independence index and coding technique A 1 2 3 B 4

Objectives and policy formulation Is price stability the major and/or only objective of the central bank? Has there been any government interference in formulation of the monetary policy within the past ﬁve years? Is there any government or political control of the central bank’s budget? Appointments Is the central bank Governor appointed by the government?

嘺 □ □ □

CBI and inﬂation in transition economies 153 5 6 7 8 9 10 11

12 C 13 14 15

16

17 D 18 19 20

E 21 22 23

Is there legislative provision for dismissal of the Governor? Can the central bank Governor hold governmental or political ofﬁce? Does the central bank Governor’s term of ofﬁce exceed the maximum term of the government? Does the central bank’s Board ofﬁce exceed the maximum term of the government? Are there any appointments to the central bank Board made by the government? Are there any government representatives attending central bank Board meetings with/without voting/veto rights? Does the government consult (e.g. nominates candidates) the bodies appointing the Governor and the central bank Board? Have there been any non-completions of the central bank Governor(s) term of ofﬁce within the past ten years? Legislation and tradition Has the central bank legislation been changed within the past ﬁve years? (if yes, which year) Has the central bank legislation been breached within the past ﬁve years? Are there any ‘other laws’ (in the country) that conﬂict with the central bank Constitution (evidence of breaching the central bank Charter within the past ﬁve years)? Are there any traditions that interfere with the central bank Charter (e.g. central bank Governor and/or Board members resigning with change of government)? Are there provisions which strengthen the central bank’s position in case of conﬂict with the government? The banking sector Is the banking supervision entrusted to the central bank? Is the banking supervision entrusted to the central bank alone? Has there been government or political interference in fulﬁlling/not fulﬁlling the ‘lender-of-last-resort’ function of the central bank for the past ﬁve years? Financing the government Are advances permitted when ﬁnancing the government? Are terms of lending controlled by the central bank or bank Charter when ﬁnancing the government? Does the central bank participate in the primary market for public debt?

Overall central bank independence

□ □ 嘺嘺 □ □ □ □

□ □ □ □ 嘺 □ □ □ □ 嘺 □

(23) 1.00

154

Janet Ilieva et al.

Table 6.2 compares three CBI indices for transition economies. Although the LVAW index and our index report the same average CBI, there are signiﬁcant differences for some countries. A possible explanation could be the changeable nature of the legislation in transition economies (one of the criteria included in the new index) and different authors have been referring to the legislation in different years (i.e. not the last amendment to the bank’s law). Such is the case with the Bulgarian National Bank (LVAW based on the law from 1991, the GMT and new index on the one from 1997, when Currency Board Arrangement was introduced), Kyrgyzstan (LVAW from 1992, whereas GMT and the new index based on the amendments from 1997), Macedonia (LVAW 1995, GMT based on the amendments in 1997 and the new index reﬂects the changes in 1998), Romania (LVAW and GMT on the law from 1991, the new index on the law from 1998, which was changed again in 19997) and so on. However, the most feasible explanation stems from the selection of different criteria, chosen by different authors (i.e. the criteria spread) which were included in the index.

Table 6.2 CBI in twenty-two transition economies (2000) No.

Country

LVAWa

Grilli et al.b

CBI

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

Albania Armenia Azerbaijan Belarus Bulgaria Croatia Czech Republic Estonia Georgia Hungary Kazakhstan Kyrgyzstan Latvia Lithuania FYROM Moldova Mongolia Poland Romania Russia Slovak Republic Slovenia Average CBI

0.51 0.85 0.25 0.73 0.55 0.44 0.73 0.78 0.73 0.67 0.44 0.52 0.49 0.78 0.41 0.73 0.55 0.89 0.34 0.49 0.62 0.63 0.6

12 14 n.a. 11 15 11 13 13 13 10 n.a. 15 12 15 14 13 n.a. 14 7 11 11 11 12.37

0.48 0.52 0.39 0.43 0.61 0.52 0.65 0.65 0.7 0.7 0.48 0.7 0.65 0.7 0.74 0.74 0.43 0.72 0.61 0.52 0.57 0.7 0.6

Notes a Data obtained from Cukierman et al. (2001). b Data obtained from Maliszewski (2000).

(0.71) (0.82) n.a. (0.65) (0.88) (0.65) (0.76) (0.76) (0.76) (0.59) n.a. (0.88) (0.71) (0.88) (0.82) (0.76) (n.a.) (0.82) (0.41) (0.65) (0.65) (0.65) (0.73)

CBI and inﬂation in transition economies 155 The highest is the average CBI reported by the GMT index 12.37, which is difﬁcult to compare with the other indices due to the different metric system employed. However, this result is consistent with the results from our pilot study, showing that the legislative approach per se overstates the actual CBI in transition economies (valid also for developing countries in the 1980s (Cukierman et al., 1992 and Cukierman, 1992). Using the index introduced in this study, the average CBI in transition economies is comparatively high (0.6). One rationale behind this is that the CBI tendency, mainly evident in the CEE countries, stems from the acquis communautaire, which must be adopted by the new member states aiming for membership to the EMU. However, some patterns within the transition economies can be detected. Highly dependent are the banks from the Commonwealth of Independent States (referred to hereafter as CIS, e.g. Azerbaijan, Belarus and Kazakhstan) participating in our survey. On the other hand, those Central European countries planning quicker accession to the European Union have highly independent central banks. Some CIS countries show very high CBI (e.g. Georgia (0.7), Kyrgyzstan (0.7) and Moldova (0.74)). Triangulating the CBI index Having approached the central bankers, the survey then aims to test the credibility of the data collected by employing a triangulation technique. The latter involves approaching independent institutions (e.g. academic institutions) in the respective country, using the same questionnaire approach to measure CBI. The rationale behind this approach is to test whether the central banks overstate their independence or vice versa. The research thus aims to improve the quality of the gathered data by counterbalancing the subjectivity of the respondents. Universities in CEE countries planning early accession to the EU rank CBI higher than the central bankers per se (e.g. Hungary 83 per cent independence scored by universities and 70 per cent by the central bank, Poland 83 per cent by universities and 72 per cent by the central bank). A possible explanation is that the bank has a good reputation and credibility with the public, and therefore the public scores it higher. The Czech Republic and Bulgaria are given the same score by the academic researchers and central bankers. In contrast, Macedonian central bank has overstated its independence (0.74) compared with the result obtained from the university (0.7). CIS countries’ banks reveal the opposite pattern compared to those of the CEE countries. Universities score CBI only half as high as the central bank (e.g. Georgia 70 per cent independence according to the central bank, while the university scores it at 35 per cent; Kyrgyzstan provided the same example: 70 per cent CBI by the central bank and 35 per cent by the university). Table 6.3 provides the results from the triangulation for a

156

Janet Ilieva et al.

Table 6.3 Central bank independence in transition countries No.

Country

Central bank

University

Eastern Europe and Baltic States 1 Czech Republic 2 Hungary 3 Latvia 4 Poland 5 Bosnia and Herzegovina 6 Bulgaria 7 Macedonia

0.65 0.7 0.65 0.72 0.70 0.61 0.74

0.61 0.83 0.74 0.83 0.70 0.61 0.7

CIS countries 8 Belarus 9 Georgia 10 Kyrgyzstan 11 Russia

0.43 0.7 0.7 0.52

0.26 0.35 0.35 0.43

0.64

0.57

Average

small number of countries, in which the academics approached participated in our survey. Table 6.3 suggests that the biggest is the gap between CBI, as reported by the central bankers, and the CBI as reported by independent experts, in CIS countries. On the other hand, the ‘successful’ transition economies reveal the opposite pattern: the independent experts overstated the CBI as scored by central bankers, such as Hungary with CBI 0.7 (CB) and 0.83 (IE), Latvia with 0.65 (CB) and 0.74 (IE), and Poland 0.72 (CB) and 0.83 (IE). Triangulating some of the results with independent experts based mainly at academic institutions has been carried out, and they provide a different picture. The triangulation results received so far deviate signiﬁcantly from the already reported CBI for some countries The banks in transition economies appear most dependent, having to supervise the commercial banks and being a lender of last resort (67 per cent of all the answers indicated that the central bank is responsible for supervising the commercial banks and that government interfered in the lender-of-last-resort function of the bank). However, the issue of whether the central bank should or should not be in charge of the banking supervision has been highly debated in the economic literature. Another ‘sensitive’ issue for the banks in transition is the changeable nature of their legislative provision and existence of traditions, inherited from the past, which are in conﬂict with the bank’s law. About 56 per cent of the surveyed banks reported recent changes in their legislation, as well as other laws and/or traditions, not consistent with the bank’s legislation. Compared to OECD and developing countries, the above issues appear to be most relevant for transition economies, and they have the lowest CBI in this section.

CBI and inﬂation in transition economies 157

4 Central bank independence and inﬂation in transition economies The average annual inﬂation in transition economies has marked a signiﬁcant decrease in the mid-1990s compared to the start of the reforms in the early 1990s. Meanwhile the CBI, arguably, has been the sole cause of the inﬂation developments. This section explores this issue and studies whether CBI had any impact on the macroeconomic performance of the studied transition countries over the review period of time. Inﬂation developments in transition economies As previously mentioned, the empirical evidence for OECD countries suggests a negative relationship between the CBI and the average inﬂation. The nature of inﬂation in transition economies is far more complex, and many factors, besides CBI, can signiﬁcantly inﬂuence it, which makes the investigation of the relationship a difﬁcult task. Maliszewski (2000) outlines that a main cause of the high inﬂation in Eastern and Central Europe is the monetary expansion driven by political factors. Figure 6.1 illustrates the slowing down of inﬂation in the countries from this sample over the period 1993 to 2001. The chart shows that most of the countries fell in the bandwidth of 100 to 1000 per cent average inﬂation at the start of their reforms, whereas in the period after 1996, with a few exceptions (except for Belarus through the following years and Bulgaria’s and Azerbaijan’s hyperinﬂation in 1997), the transition economies are experiencing inﬂation of less than 100 per cent. Most of the countries had annual inﬂation of less than 10 per cent in the late 1990s except for a few CIS and Romania. Central bank independence and average inﬂation in transition economies – empirical analysis We study the relationship between average inﬂation and CBI by estimating the following model: Yit ⫽ ␣ ⫹ ␤Xit ⫹ ui ⫹ eit

[1]

In this model ui ⫹ eit is the residual. The former is the unit-speciﬁc residual; it differs between units but, for any particular unit, its value is constant. The latter is the ‘usual’ residual with the usual properties (mean 0, uncorrelated with itself, uncorrelated with Xit, uncorrelated with ui and homoscedastic). Whatever the properties of ui and eit, if equation [1] is true it must be true that Y*i ⫽ ␣ ⫹ ␤X*i ⫹ ui ⫹ e*i

[2]

158

Janet Ilieva et al. 10000

1000

100

10

1 1993

1994

1995

1996

1997

1998

1999

2000

0.1 Albania

Czech R.

Macedonia

Armenia

Belarus Bulgaria

Hungary Kazakhstan

Poland Romania

Azerbaijan Russia

Figure 6.1 Inﬂation in selected transition economies, 1993–2000.

where Y*i ⫽ ⌺tYit/Ti, X*i ⫽ ⌺tXit/Ti and e*i ⫽ ⌺teit/Ti. Subtracting [2] from [1], it must be equally true that (Yit ⫺ Y*i) ⫽ ␤(Xit ⫺ X*i) ⫹ (eit ⫺ e*i)

[3]

These three equations provide the basis for estimating b. In particular, ﬁxed effects (also known as the within estimator) amount to using OLS to estimate [3]. Between estimator and amounts to using OLS to estimate [2]. Random effects is a (matrix) weighted average of the estimated produced by the between estimator and within estimators. In particular, the random effects turns out to be equivalent to estimation of (Yit ⫺ ␪Y*i) ⫽ (1 ⫺ ␪)␣ ⫹ b(Xit ⫺ ␪X*i) ⫹ [(1 ⫺ )ui ⫹ (eit ⫺ ␪e*i)

[4]

where ␪ is a function of ␴u2 and ␴2e. If ␴2u ⫽ 0 meaning ui is always 0, ␪ ⫽ 0 and equation [1] can be estimated by OLS directly. Alternatively, if ␴2e ⫽ 0

CBI and inﬂation in transition economies 159 meaning eit is always 0, ␪ ⫽ 1 and the within estimator returns all the information available (which will, in fact, be an R2 ⫽ 1 regression). Few assumptions are required to justify the ﬁxed effects estimator of [3]. The estimates are, however, conditional on the sample in that ui are not assumed to have a distribution but are instead treated as ﬁxed and estimable. The random effects of [4] require no-correlation assumption. In comparison with the between estimator, the random effects estimator produces more efﬁcient results, albeit ones with unknown small-sample properties. The between estimator is less efﬁcient because it discards the over-time information in the data in favour of simple means. The random effects estimator uses both the within and the between information. Descriptive statistics Variable | Mean Std. Dev. Min Max INFLAT ⬃ N overall | 99.33709 257.2648 between | 73.62389 12.86364 221.3364 within | 247.0178 ⫺119.9993 1556.355

| Observations ⫺24.5 1664 | N ⫽ 151 | n ⫽ 14 | T-bar ⫽ 10.7857

The average inﬂation is 99.33. The standard deviation between groups is smaller than the standard deviation across (within) years. Therefore, the inﬂation is more time dependent than country-speciﬁc dependent. This is expected since the CPI is a dummy variable through time. Econometric application INFLATIONit ⫽ ␣it ⫹ ␤CBIit ⫹ uit where: INFLATIONit is the average of country i at time period t. CPIit is a dummy that is equal to 1 when the Central Bank became Independent and equal to 0 otherwise. ␣it is a constant for country i at time period t. uit is a random error term for country i at time period t which is normally distributed with a mean of 0 and a variance of ␴2. Estimation Tables 6.4a and 6.4b report a panel regression model. In the panel estimation we have to choose whether to estimate a ﬁxed or a random effects model. In this case, we base our decision on the Hausman test (1978). The

160

Janet Ilieva et al.

Table 6.4a Fixed effects Inﬂationit ⫽

␣it

⫹␤CPIit

(Stand. errors) T-ratios R-Squared No. of observations No. of countries

231.49 (31.20) 7.42 0.1445 151 14

⫺229.38 42.724 ⫺5.37

Inﬂationit ⫽

␣it

⫹␤CPIit

(Stand. errors) T-ratios R-Bar-Squared No. of observations No. of countries

214.51 (30.53) 7.03 0.1445 151 14

⫺200.02 (39.45) ⫺5.07

⫹uit

Table 6.4b Random effects ⫹uit

Hausman test follows a chi-square distribution and is equal to 3.21 with a corresponding p-value of 0.0733. This means that the random effects model cannot be rejected at 5 per cent level of signiﬁcance but can be rejected at 10 per cent level of signiﬁcance. Based on this result we use either the random effects or the ﬁxed effects model. The coefﬁcients have the same signs and signiﬁcant levels in both models. In addition, the Rsquared are exactly the same in both models. The results from Tables 6.4a and 6.4b reveal a negative and signiﬁcant CPI variable. This means that when the central bank becomes independent the average inﬂation is signiﬁcantly reduced. Possible reasons for this are the following. First, it is possible that when independence occurs, the level of uncertainty in the economy is reduced thereby reducing inﬂation. Second, the central bank may have an independent or ﬂexible monetary policy. For example, in Germany, the Bundesbank keeps inﬂation low irrespective of government policy. In order to obtain robust results we also estimate cross-sectional regressions to look at the effect of individual years. In particular, we split the sample size into two time periods, namely 1991 to 1995 and 1996 to 2001. The reason for this is that the effect of central bank independence is generally felt around or after the mid-1990s. SAMPLE PERIOD: 1991 TO 1995

From Table 6.5a and the corresponding diagnostic results, we can see that the results of the panel still stand. The CBI coefﬁcient is negative and

CBI and inﬂation in transition economies 161 Table 6.5a Panel results Inﬂationit ⫽

␣it

⫹␤CPIit

(Stand. errors) T-ratios R-Squared No. of observations No. of countries

238.13 (45.78) 5.20 0.0702 70 14

⫺199.22 (87.87) ⫺2.26

⫹uit

Diagnostics Serial correlation (Breusch-Godfrey)

Heteroscedasticity (White test)

F-statistic 1.356

F-statistic 2.577

Probability 0.26

Probability 0.11

signiﬁcant (and the parameter stability of the coefﬁcient remains relatively intact) and, since there are no diagnostic problems, the model is relatively well speciﬁed.8 SAMPLE PERIOD: 1996 TO 2001

From Table 6.5b and the corresponding diagnostic results, we can see that the CBI is negative and signiﬁcant as we found previously. Interestingly enough, the coefﬁcient on the CBI variable is signiﬁcantly reduced. This means that for the sample period (1991 to 1995), the CBI parameter is Table 6.5b Panel results Inﬂationit ⫽

␣it

(Stand. errors) T-ratios R-Squared No. of observations No. of countries

107.83 (30.76) 3.50 0.096 82 14

⫹␤CPIit

⫹uit

⫺98.57 (33.78) ⫺2.91

Diagnostics Serial correlation (Breusch-Godfrey)

Heteroscedasticity (White test)

F-statistic 0.050

F-statistic 6.349

Probability 0.95

Probability 0.01

162

Janet Ilieva et al.

Table 6.6 GLS results Inﬂationit ⫽

␣it

(Stand. errors) T-ratios R-Squared No. of observations No. of countries

107.83 (74.15) 1.45 0.096 82 14

⫹␤CBIit

⫹uit

⫺98.57 (74.17) ⫺1.32

stable whereas for the sample period (1996 to 2001), the parameter changes considerably. From the diagnostic results, we ﬁnd that the model is mis-speciﬁed because even though there are no serial correlation problems there are heteroscedasticity problems. The latter implies that the reported standard errors are relatively ‘weak’. In order to overcome this ﬂaw in our econometric analysis, we correct the standard errors by using generalised least squares (GLS). This econometric technique calculates standard errors that are robust to heteroscedasticity. The GLS results are reported in Table 6.6. From Table 6.5, we can see that out variable of interest (CBI) becomes insigniﬁcant. This is a striking result because it implies that once central bank independence has occurred, inﬂation achieves a new lower equilibrium. This result further supports the claim that central bank independence reduces inﬂation to a new lowered equilibrium level. In other words, our empirical analysis is in line with the observable descriptive analysis.

5 Conclusion A close inspection of the legislation of transition economies shows that they have modelled their central banks on the ECB, so that the legislative independence of their central banks is very high. There are some clear patterns appearing in the CBI of transition economies. Countries planning early accession to the EU have higher CBI compared to the average of all transition countries. A comparison between transition countries and OECD and developing countries shows that transition economies have comparatively high overall CBI, still lower than the CBI in OECD countries from our sample, but higher compared to the sample of developing countries. A triangulation technique, using a survey of independent academic experts to provide an alternative estimate of the independence of their respective national banks, revealed an interesting result. In transition countries planning early accession to the EU, universities rated their central banks as being somewhat more independent than the central banks believed themselves to be; in contrast, in the CIS, independent experts

CBI and inﬂation in transition economies 163 viewed their central banks as much less independent than the central banks themselves had reported. These results presumably reﬂect the gap between legislative and actual independence in the least advanced transition economies (i.e. that the public in the CIS have little conﬁdence in the ability or willingness of their central banks to act according to their statutes), while in the pre-accession CEE states, the determination of the new central banks (supported by their governments) to prove their antiinﬂation credentials has created an impression that they are more independent than their legislative status would otherwise suggest. Studying the CBI for implication on the inﬂation performance in transition economies revealed a strong inverse relationship between the CBI and average inﬂation. However, when we split the sample into sub-periods (1991 to 1995, 1996 to 2001), we ﬁnd that the relationship persists only for the ﬁrst sample period. This implies that after 1996 there is no relationship between CBI and average inﬂation. A possible reason is that most of the transition economies achieved CBI by the mid-1990s.

Notes 1 Broadly referenced is the Swiss Bank, which similar to the Bundesbank is highly independent and has maintained very low average inﬂation. 2 European Commission, Agenda 2000, Supplement 13/97, p. 44. 3 General Council is constituted as a third decision-making body of the ECB, comprising the President, Vice-president and the Governors of the national central banks (Chapter IX, Art. 45 of the Protocol on the Statute of European System of Central Banks and the European Central Bank). 4 The Central Bank of Bosnia and Herzegovina provides a very unusual case. It operates as Currency Board Arrangement and, according to the charter, the Governor for the ﬁrst six years must be a foreigner. The appointment is by the Managing Director of the International Monetary Fund with the approval of the Presidency. 5 It used to be eight years’ term of ofﬁce in Romania, but with the recent amendments it was reduced to six years. 6 We are indebted to Pavel Soukup and Roman Matusek for their interesting comments on the independence of the Czech National Bank. 7 The term of ofﬁce of the Governor and the Bank’s board was reduced from eight to six years with this change in the law, which does not affect the CBI according to our index. 8 The diagnostic tests for normality and non-linearity are not reported. The normality test is not reported, since the CBI dummy ensures that normality is rejected. The non-linearity RESET test is not reported due to lack of observations. The model may be non-linear but this is not the focus of this chapter.

References Dvorsky, S. (2000) ‘Measuring central bank independence in selected transition countries and disinﬂation process’, Bank of Finland Institute for Economies in Transition, Discussion paper No. 13. Goodhart, C.A.E. (1994) ‘What shall central banks do? What shall be their macro-

164

Janet Ilieva et al.

economic objectives and operations?’, The Economic Journal 104 (247), 1425–1436. Hochreiter, E. and Kowalski, T. (2000) ‘Central banks in European emerging market economies in the 1990s’, Oesterreichsche Nationalbank, Working Paper 40. Loungani, K. and Sheets, N. (1997) ‘Central bank independence, inﬂation and growth in transition economies. Journal of Money, Credit and Banking, 29, 381–399. Maliszewski, W.S. (2000) ‘Central bank independence in transition economies’, Economics of Transition, 8 (3), 749–789. Mangano, G. (1998) ‘Measuring central bank independence: a tale of subjectivity and of its consequences’, Oxford Economic Papers, 50 (3), 468–493. Radzyner, O. and Riesinger, S. (1997) ‘Central bank independence in transition: legislation and reality in Central and Eastern Europe’, Vienna Focus on Transition, 2, 57–90.

Part III

7

Central bank independence in the Czech Republic Pavel Soukup, Anita Taci and Roman Matousˇek

1 Introduction In the strictest sense, the history of Czech central banking started on 1 January 1993 when the independent Czech Republic emerged. However, we can consider the development of central banking in the former Czechoslovakia up to this date as the direct predecessor of an independent central bank in the Czech Republic. Czechoslovakia originated after the dissolution of the AustroHungarian Empire in 1918, at the end of World War I. The urgent task of the new government was to restore order to the monetary system because of the threatened continuation of the wartime depreciation of the currency. The functions of the central bank were temporarily carried out by the main ofﬁce in Prague and regional branches of the Austro-Hungarian Bank which came under the control of the new Czechoslovak state authorities. One of the steps of the monetary reform (together with the currency restriction and separation from the successor states of the former AustroHungarian Empire) was the setting up of the Bank Ofﬁce of the Ministry of Finance in March 1919. Bank Ofﬁce of the Ministry of Finance, 1919 to 1925 The Bank Ofﬁce of the Ministry of Finance functioned provisionally over a relatively long period as the central bank of the Czechoslovak State. The Ofﬁce was a department of the Ministry of Finance. Its management was in the hands of the Bank Committee, chaired by the Minister of Finance. The activities of the Ofﬁce were terminated in March 1926 when the conditions for establishing the central bank were favourable. In the second half of the 1920s the value of the Crown stabilised, the current account of the balance of payments was positive and a sound currency reserve system was created.

168

Pavel Soukup et al.

Czechoslovak National Bank, 1926 to 1950 The monetary tasks of the Bank Ofﬁce were transferred to the Czechoslovak National Bank (NBCS) in 1926. The Bank was a joint stock company with one-third government share, though the power of the government was more signiﬁcant than this implies. The main monetary instruments were the prime rate and discount policies. One of the most important developments was the introduction of full liberalisation of the foreign currency regime in 1929 and, at this time, the Czech Crown was backed by gold. However, the Great Depression in the 1930s brought strong deﬂationary pressures which upset both the balance of payments and the state budget. This resulted in devaluation of the currency in 1934 and 1936 after which limited convertibility was restored. During World War II, the Czechoslovak National Bank was brought under the control of the German occupants. After the liberation, in the period from 1945 to 1948, the activities of the Bank were restored to a limited extent. The development of both the economy and policy in the former Czechoslovakia increasingly came under the inﬂuence of the Soviet Union. In the banking sector this resulted in the uniﬁcation of the commercial banks and central bank into one bank based on the model of (and with the ‘help’ of) the Soviet Gosbank in 1950. The banking system under the command economy, 1950 to 1989 The central-planning system of the Soviet Union negatively affected the banking system and the role of the central bank in the economy. In 1950, Law No. 31/1950 established the Czechoslovak State Bank (SBCS) which assumed all commitments of the former Czechoslovak National Bank, and commercial banks – the Trade Bank, the Slovak Tatra Bank and the Post Savings Bank. The merger of the central bank with the commercial banks into one unit (SBCS) created a one-tier banking system ‘suitable’ for the management of the fully nationalised economy. This single bank, often referred to as the ‘monobank’, was in charge of both crediting and issuing functions; that is, apart from holding the responsibilities of the central bank, it was also involved in a great majority of both deposit and credit transactions in relation to state enterprises. The role of the SBCS was supplemented by the activities of the Czech State Savings Bank and the Slovak State Savings Bank in the Slovak part of Czechoslovakia, which were focused on individual deposits and personal loans; by the Czechoslovak Trade Bank (CSOB) which specialised in foreign trade operations, and later by the activities of the Trade Bank which dealt with forex payments for individual customers and travel agencies. Effectively the SBCS was downgraded to the status of a purely administrative institution. This role of the SBCS was accompanied by changes in its structure and management so that the budgetary organisa-

CBI in the Czech Republic 169 tion of the bank resembled that of other ministries. An unsuitable organisational structure and professional ‘nomenclature’ were forced upon it, just as they were on other ‘central authorities’. Moreover, the SBCS was subordinated to the Ministry of Finance and was in fact only one of its departments. The regional branch ofﬁces of the SBCS were highly dependent on the Communist Party authorities. The branches represented the main source of information both on a ‘district-wide’ basis and on individual enterprises for the district Party committees. The system of monetary policy implementation was divided into three separate sectors: cash management, credit and foreign exchange relations. These three sectors were controlled through currency, credit and foreign exchange plans. At the end of the 1950s the ﬁrst reformation efforts resulted in the detachment of the SBCS from the Ministry of Finance. The role of the Bank and its monetary policy as ‘independent’ of the state executive authority had been strengthened a little! The social ferment at the beginning of 1968 resulted in partial reform of the Czechoslovak banking sector. Parliament passed a new Czechoslovak State Bank Act, effective on 1 January 1971. The Law introduced a monetary plan; that is, an aggregation of currency, credit and foreign exchange plans. This constituted something of a separation from the centrally planned activities of the Commission of the State Plan. Under the Law, the monetary plan became a part of a triad of ‘supreme plans’: the state plan, the state budget and the monetary plan. However, the SBCS continued to be fully subordinated to the government, particularly to the Ministry of Finance, and to the dominant role of the central state plan. The principal task of the monetary plan was the supervision of the ﬁnancial balance of the central plan through the implementation of the credit and currency plans. The credits provided by the SBCS were limited to ﬁnancial provisions covering the needs of enterprise and those of government. In principle, only administrative instruments for control of the money supply were available including interest rates on credits supplied. However, the discussion on the legal structure of the Czechoslovak State, primarily the ‘central bank part’ of the SBCS, soon inﬂuenced the former efforts to substantially reform the banking sector. In the event, the only visible result was the establishment of SBCS Main Ofﬁces in place of the former asymmetrical structure, so there were two Main Ofﬁces in the capitals of the newly established republics within the federal state in 1970. The main changes to the banking system within the framework of marginal reform of the socialist planned economy were embodied in Act no. 130/1989 of the Czechoslovak State Bank, and Act no. 158/1989 of the banks and savings banks. The two-tier system was reoriginated with the central bank accorded responsibility for the stability of the currency and monetary policy, while commercial activities were to be carried out by commercial banks. Although the new law on the SBCS deﬁned its role as that of central bank, it remained entirely a product of the system of central

170

Pavel Soukup et al.

planning. For example, according to the Law, the Bank, in co-operation with the relevant state administrative authorities, was obliged to formulate and submit for deliberation by the government ﬁve-year and annual monetary plans. It was further required to adjust these monetary plans and monetary policy in accordance with developments in the economy. The Act did not separate state ﬁnances from the central bank credit, and the role of the bank was merely that of agent for the implementation of the state budget. According to the Law, the Bank was responsible for securing the treasury requirements of the government’s budget which included ﬁnancing any disparity between revenues and expenditures during the course of the year.

2 Post-1989 reform The period from 1990 to 1996 may be characterised as a transformation of the whole of society from a centrally planned economy under the Communist Party, to that of a market-driven economy with a pluralistic democratic political system. The development of the Czech economy from the end of World War II until 1989 was heavily dependent on the former Soviet Union both from political and economic perspectives. The liquidation of private ownership in communist Czechoslovakia was given the highest priority and ‘nationalisation’ reached its maximum limit. This caused high concentration of industry, agriculture and trade in large specialised enterprises, followed by monopolisation that promoted unrealistic prices, low productivity and production of non-competitive goods. The consistent ‘nationalisation’ enterprises went hand in hand with the implementation of the strict central planning model, and eventually resulted in the development of a number of problems and distortions in the economy. However, compared to the other post-communist countries, the starting conditions of the Czechoslovak economy were relatively favourable, especially due to its low foreign debt, relatively high consumption in the household sector and balanced consumer market. The government budgets were relatively sound, the total government debt did not exceed 1 per cent of the GDP and statistics show that the annual rate of inﬂation did not exceed 1 to 2 per cent. (With hidden inﬂation it was estimated that the annual rate was, in fact, 5 per cent.) Overview of main policy reforms As the 1980s drew to a close the drive to reform gathered momentum and social and political changes were introduced in late 1989. In autumn 1990, the Czechoslovak government presented to Parliament an ‘economic and social reform scenario’ which set out the basic framework for the transformation of the Czechoslovak economy. The main objective of the radical

CBI in the Czech Republic 171 economic reform was a transition from a command economy to a market economy. The concept was based on a sweeping departure from central planning and refused to make marginal improvements to the current status quo or reconsider previous reform attempts as a ‘third way’. The reform objectives were to be implemented as quickly as possible and it was anticipated that the more radical and speedy the reforms, the lower the costs of transition. The ﬁnal version of the transformation programme, as started in January 1991, included ﬁve main pillars: 1 2 3

4

5

Liberalisation of prices: after forty years of central price ﬁxing, the bulk of prices were to be freed. Liberalisation of external trade. Introduction of partial convertibility (for resident ﬁrms’ current account transactions). that required a devaluation towards an exchange rate seen as viable with respect to the balance of payments. Macroeconomic stabilisation involving restrictive ﬁscal and monetary policies to remove the monetary overhang and to reintroduce macroeconomic equilibrium. Massive and rapid privatisation, mainly through the voucher scheme.

The experience from elsewhere showed that free trade and partial convertibility should not be attempted without ﬁrst establishing macroeconomic stability. The ﬁrst year of transformation (1990) was devoted particularly to the goal of stabilisation; further transformation steps taken in 1991 were accompanied by continuing stabilisation policies. Measures seeking external stability were also an indispensable part of the programme. It was understood that the exchange rate would have to be undervalued in relation to purchasing power parity to sustain the newly liberalised trade and to maintain the partly convertible currency. Compared to the liberalisation of prices, freeing of external trade and the introduction of partial convertibility, it was anticipated that privatisation would be a gradual, more time-consuming process. In the meantime, steps were also taken to withdraw state subsidies to ﬁrms (as well as to prices), and to let ﬁrms face a hard budget constraint including a stricter credit policy. Deceleration of the inﬂation process was given a higher priority than that of economic growth and employment. The core of the reform process therefore included a plan to create a social security network to help maintain the basic standard of living for people suffering from the effects of transition. Table 7.1 gives an indication of changes in the economy that have occurred over the decade 1991 to 2001.

1994

1995

1996

1997

1998

1999

2000

Source: Transition Report Update, EBRD (1999), and CNB publications (1990–2002).

52 12.7 18.2 9.7 7.9 8.6 10.0 6.8 2.5 4.0 ⫺11.6 ⫺0.5 0.1 2.2 5.9 4.3 ⫺0.8 ⫺1.0 0.5 3.3 3,580 3,560 3,560 3,640 3,850 4,020 4,000 3,960 3,980 4,120 4,440 4,620 4,170 4,330 5,040 5,100 4,540 4,440 4,240 3,810 n.a. n.a. 19.2 17.7 15.3 13.2 12.9 13.1 14.4 16.7 26.4 23.7 28.7 29.0 33.1 36.9 44.6 39.5 43.2 41.2 4.1 2.6 3.5 3.2 2.9 3.3 4.9 7.5 9.4 8.8 n.a. 40.2 34.9 33.6 33.3 35.3 34.1 32.6 31.8 32.8 6.0 6.1 6.5 4.9 4.7 4.8 4.4 4.7 3.9 3.9

1993

CPI (av. y-o-y change in %) GDP (y-o-y change in %) GDP per capita in euros GDP per capita in US dollars Public debt/GDP (%) External debt/GDP (%) Unemployment (%) Share of industry in GDP (%) Share of agriculture in GDP (%)

1992

1991

Selected indicators

Table 7.1 Selected economic indicators, 1991 to 2001

4.1 3.3 4,260 3,820 23.4 36.5 8.9 33.9 3.9

2001

CBI in the Czech Republic 173 Creation of a two-tier banking system A two-tier banking system has been reoriginated according to Act no. 130/1989 on the Czechoslovak State Bank which, paradoxically, was passed by Parliament two days before the ‘Velvet Revolution’ started, and Act no. 158/1989 on banks and savings banks. Commercial transactions were removed from the SBCS by the newly established commercial banks (the Commercial Bank General Credit Bank in Slovakia respectively) and the activities of the Investment Bank were re-established. Some ‘commercial’ banks were retained from the past (the Czechoslovak Trade Bank and the Czech Slovak State Savings Banks) and only now started to play a commercial role. Several newly created private banks, both of domestic and foreign origin, also appeared. Altogether, twenty-three commercial banks were registered in December 1990. Finally, although the new law on the SBCS deﬁned its role as the central bank of the state, it was biased by the ‘centrally-planned’ approach of its authors (see Section 1). Establishment of a central bank The fundamental social and political changes at the end of 1989 and preparations for economic reform in 1990 created an urgent need for a new law on the central bank. This law was drafted during 1990 and especially during 1991 and took effect on 1 February 1992. However, until it took effect, the previous law from the pre-reform year 1989 remained in force, so that the bank was entirely subordinated to the government. The new Act on the Czechoslovak State Bank (No. 22/1992) enabled the central bank to play one of the pivotal roles in the transition process. The central bank was established as an independent institution whose primary objective was to ensure the stability of the Czech currency. It is responsible for the conduct of monetary policy and is independent of government. Apart from the above mentioned targets, the activities of the central bank include supervision over the commercial banks, operating as a clearing centre, maintaining foreign currency reserves, issuing banknotes and managing the money supply. A signiﬁcant shift therefore took place, both in the operational autonomy of the bank and in the strict separation from the state budget. Furthermore, the bank’s management, auditing methods and the obligation of submitting reports for the approval of Parliament were set up.

3 The present role and function of the Czech National Bank (CNB) After the breakup of Czechoslovakia on 1 January 1993, the Czech National Bank (CNB) became the successor institution to the former SBCS. There were no signiﬁcant changes in the new Central Bank Act

174

Pavel Soukup et al.

(No. 6/1993) in comparison with the former Act. The new Act reﬂected changes mainly arising from the dissolution of the Federation. The political independence of the bank was further strengthened by the stipulation that the Governor and other members of the bank Board are appointed and revoked by the President of the Republic only. However, in 1995, owing to some turbulence in the banking sector, several important amendments concerning the banking supervision and deposit insurance system were adopted (both in the CNB Act and in the Banking Act (No. 21/1992)). During 1999, close attention was paid to further improving the legislative framework towards harmonisation with the rules in force in EU countries. Nevertheless, the ﬁrst wording of the ‘harmonisation amendment’ to the Act on the CNB had paradoxically anti-harmonisation features, signiﬁcantly weakening central bank independence. These features of the Act reﬂected political pressures stemming from the economic recession and the political cycle. After a two-year ‘battle for independence’ between several political factions in Parliament (the President, the Constitutional Court, banking specialists) and also under pressure from international ﬁnancial and banking institutions, the last harmonisation amendment (in force as of May 2002) is fully compatible with the EU requirements in terms of central bank autonomy. The structure of the CNB The supreme governing body of the bank is the bank Board, which is composed of seven members, comprising the Governor, two Vice-Governors and four other members selected from CNB executive managers. The structure of the bank consists of its headquarters, regional branch ofﬁces and special-purpose operational units. The headquarters of the CNB was created by combining the former federal headquarters and the Head Ofﬁce of the Czech Republic. On 1 January 2002, a new management system and organisational structure took effect at the CNB. The organisational measures adopted are a natural part of the process of preparation for the CNB’s entry into the European System of Central Banks. The organisational sections of the headquarters and branches of the CNB are managed by the bank Board as a collective body. Wider powers have also been given to the executive directors of each of the departments. They are now responsible to the bank Board for managing and conducting the activities delegated to them in line with the medium-term plans approved by the Board for speciﬁc areas of the central bank’s activities. In an effort to optimise the CNB’s organisational scheme, the number of departments has been reduced from sixteen to eleven. The departments are as follows: •

monetary and statistics department,

CBI in the Czech Republic 175 • • • • • •

banking regulation department, banking supervision department, ﬁnancial markets department, risk management and transaction support department, cash and payment systems department, ﬁve ‘services’ departments – general secretariat, human resources, budget and accounting, information systems, and administration.

There are seven regional branch ofﬁces and several specialised operational units; for example, the economic research department, internal audit, and control and crisis management. Statutory duties and operating procedures The main objectives and tasks of the central bank are laid down in the Constitution of the Czech Republic and the Czech National Bank Act (No. 6/1993 Collection of Laws, amendment in force as from 1 May 2002). These charge the bank with maintaining price stability as the primary objective. Without prejudice to this objective, the CNB shall support the general economic policies of the government leading to sustainable economic growth. In accordance with this objective the CNB: • • • • •

sets monetary policy, issues banknotes and coins, controls the circulation of currency, administers payments and clearing between banks, supervises banks, foreign bank branches and consolidated groups containing banks, performs other activities (e.g. announces the exchange rate of the Czech currency, manages monetary reserves, issues securities).

The CNB is obliged to submit to Parliament a report on monetary developments at least twice a year. Similarly it is obliged to inform the general public of monetary developments at least once every three months. The CNB also provides, every ten days, a report to the general public on its ﬁnancial stance. Furthermore, the CNB acts in an advisory capacity to the government in respect of matters relevant to the conduct of monetary policy and banking. It is empowered to comment on proposals presented to the government for consideration that concern the ﬁelds of competence of the CNB. It submits to the government draft legislation on the currency and circulation of money, the money market, the payments system, and other amendments concerning the competencies and position of the central bank. The CNB, together with the Ministry of Finance, shall submit to the government draft legislation on foreign exchange

176

Pavel Soukup et al.

management and on banking issues. The CNB has the exclusive right to issue banknotes and coins. It manages the stock of banknotes and coins and organises supplies of them, and withdraws them from circulation. The CNB, after discussion with the government, stipulates the exchange rate regime and declares the exchange rate of the Czech currency vis-à-vis foreign currencies, with the proviso that the primary objective of the CNB is not jeopardised. According to the CNB Act and other special legislative Acts, the bank sets the interest rate, structures, maturities and other terms and conditions for the transactions it performs (e.g. repo rates, discount and lombard rates, minimum reserves requirements). It sets forth prudential rules for banks, foreign bank branches and other money market participants along with the terms and conditions of money market transactions. The CNB may require banks to hold a prescribed part of their funds (required minimum reserves) on accounts with the CNB. The reserves may not exceed 30 per cent of a bank’s total liabilities. Where a bank fails to maintain the minimum reserves, the CNB may charge it interest at double the effective discount rate on the amount of reserves which that bank fails to provide. The CNB keeps the accounts of banks, accepts their deposits and makes transactions with banks. It may grant to them for a maximum of three months credits guaranteed by special securities or government bonds. In order to regulate the money market, the CNB can buy and sell negotiable securities. The CNB may issue short-term securities maturing within six months and trade in them. In order to ensure uniform payments and settlements of accounts in the Czech Republic, the CNB sets forth the manner of execution of the payments system among banks, the clearing of accounts of the banks, and the manner of applying payment instruments. The CNB performs supervision of banking activities to ensure the safe, sound and efﬁcient operation of the payment system. Supervision includes assessment of licence and permit applications pursuant to special legislative Acts, supervision of adherence to the conditions stipulated in licences and permits, and to other laws, decrees and provisions in the competence of the CNB. The CNB grants and revokes banking licences in agreement with the Ministry of Finance. It may impose remedial measures and penalties where shortcomings are detected, or a conservatorship when the ﬁnancial situation or liquidity of a bank has been markedly or repeatedly at variance with the requirements in force. Open market operations and other instruments In 1991, the overall monetary objective was expressed as a percentage of the permitted growth of credits and loans. The initial target rate was ⫺2 to ⫹1 per cent compared to the credit volume in 1989, but this was later

CBI in the Czech Republic 177 modiﬁed to ⫹1 to ⫹2.6 per cent. (The actual increase was 0.8 per cent.) The following instruments of monetary policy were used to achieve the target in this period: • • • • •

credit limits, minimum reserves requirements, a system of credit reﬁnancing, discount interest rate and interest rate ceilings, exchange rate policy.

In the early period the emphasis was on direct controls (credit limits and credit ceilings) and these were successful in establishing a platform for currency stability. In 1992, the emphasis of control shifted from direct to indirect instruments. After the abolition of both interest and credit ceilings, the SBCS regulated monetary development ﬂexibly with respect to the prevailing monetary situation. From 1992, the CNB was able to operate with the entire spectrum of main instruments, and open market operations are used for steering interest rates in the economy, managing liquidity on the money market and signalling the CNB’s monetary policy stance. They are usually executed in the form of repo operations. With regard to their aim and regularity, the CNB’s open market operations may be divided into the following categories: •

The main monetary policy instrument takes the form of repo tenders. The CNB accepts surplus liquidity from banks and in return transfers eligible securities to them as collateral. The two parties agree to reverse the transaction at a future point in time when the CNB, as borrower, repays the principal of the loan plus interest, and the creditor bank returns the collateral to the CNB. The basic maturity of these operations is fourteen days, although repos with shorter maturities are executed from time to time depending on the forecasts of banking sector liquidity. Owing to the systemic liquidity surplus in the Czech banking sector, repo tenders are currently used exclusively for absorbing liquidity. The CNB conducts variable rate tenders, which means that the declared repo rate serves as the maximum limit rate at which banks’ bids can be satisﬁed in the tender. The bids are ranked using the American auction procedure; that is, those with the lowest interest rate are satisﬁed as having priority and those with successively higher rates are accepted until the total predicted liquidity surplus for the day is exhausted. If the volume ordered by the banks exceeds the predicted surplus, the CNB either completely refuses the bids at the highest rate or reduces them pro rata. Repo tenders are usually announced every day at around 9.30 a.m. Banks may submit their orders; that is, the amounts of money and the interest rates at which they want to enter into transactions with the CNB, within a prescribed time. The minimum acceptable volume is CZK 300 million. Bids

178

•

•

•

•

•

Pavel Soukup et al. exceeding the minimum must be expressed as multiples of CZK 100 million. The supplementary monetary instrument is the three-month repo tender. Here, the CNB accepts liquidity for a three-month period. The three-month repo tender again uses the American auction procedure (see the two-week repo tender). In these operations, the CNB does not intend to send signals to the market and therefore the threemonth repo rate used for this tender is not the CNB’s rate, but the money market rate in effect at the time of calling the tender. At present this instrument is not used. The last three-month tender was called in January 2001. Fine-tuning instruments (foreign exchange operations and securities operations) are used ad hoc, mainly to smooth the effects on interest rates caused by unexpected liquidity ﬂuctuations in the market. These instruments are rarely used. Automatic facilities are used for providing and depositing liquidity overnight. To the banks, these represent standing facilities for depositing or borrowing money, the interest rates applied to them form the corridor for short-term money market rates (as well as for the twoweek repo rate). The deposit facility is a non-collateralised standing facility which banks may use to make overnight deposits of surplus liquidity with the CNB. The deadline for submitting requests is ﬁfteen minutes before the end of the clearing day. The minimum volume is CZK 100 million. Deposits exceeding the minimum must be multiples of CZK 50 million. The deposits are remunerated at the discount rate, which generally provides a ﬂoor for short-term interest rates on the money market. The marginal lending facility is a standing facility which banks may use to obtain overnight liquidity from the CNB in the form of repo operations. The deadline for submitting requests is twenty-ﬁve minutes before the end of the clearing day and there is no minimum volume prescribed. The interest rate applied to this facility is the Lombard rate. Owing to the persistent liquidity surplus, banks make minimal use of this facility. The Lombard rate provides a ceiling for short-term interest rates on the money market.

Every bank, building society and foreign bank branch having a banking licence in the Czech Republic is required to hold a pre-speciﬁed volume of liquidity, known as minimum reserves, on its clearing account with the CNB. The reserve requirement is 2 per cent of the eligible liabilities for calculating the reserve requirement. With effect from 12 July 2001, eligible liabilities form the bulk of the volume of a bank’s primary liabilities (chieﬂy deposits from non-banks) whose maturity does not exceed two years. Each bank is required to maintain an average end-of-day balance on

CBI in the Czech Republic 179 its clearing account equal to, or greater than, the reserve requirement over a one-month maintenance period. Starting from 12 July 2001, the funds in this account are remunerated at the CNB two-week repo rate up to the pre-speciﬁed volume of minimum reserve requirements (these were not previously remunerated). The reserve requirement is currently of little signiﬁcance as a monetary policy instrument, but the money held in these accounts fulﬁl another important role: they serve as cushions for the smooth functioning of the interbank payment system at CNB clearing. In 1999, the CNB completed the process of gradually lowering its reserve ratio to 2 per cent, which is equal to that set for the Eurosystem by the European Central Bank. To maintain the smooth functioning of the interbank payment system, even following the lowering of the reserve requirement to its present level, a collateralised (i.e. extended to banks in exchange for securities) intraday credit facility was introduced after the reserve requirement was lowered. Within this facility, the CNB, as the operator of the payment system and the short-term bond settlement system, provides short-term intra-day credit to banks to enable them to make payments even if they do not have sufﬁcient funds on their accounts with the CNB. No interest is charged on intra-day credit and there is automatic spill-over into the marginal lending facility in the event of non-repayment.

4 The constitutional relationship between the CNB and the government According to the constitution of the Czech Republic, any interference with central bank activities has to have some basis in law. In accordance with the Central Bank Act, neither the CNB nor the bank Board shall seek or take instructions from the President, from Parliament, from the government, from administrative authorities or from any other body when pursuing its primary objective, that of price stability. The Governor is not a member of the government but he(she), or a vice-governor nominated by him(her), may attend the meetings of the government in an advisory capacity. An appointed member of the government (usually the Minister of Finance) may attend the bank Board’s sessions in an advisory capacity and may submit motions for discussion. Budgetary independence The CNB is a legal entity governed by public law and acting as an administrative authority to the extent deﬁned in the Central Bank Act and other special Acts. It manages, independently, assets entrusted to it by the state. The bank manages its ﬁnances in compliance with a budget, broken down so as to show clearly its operating and investment expenditures. The CNB is not paid from the state budget and defrays the necessary costs of its

180

Pavel Soukup et al.

operations from its income. The proﬁt it generates is used to replenish its reserve fund and other funds created from proﬁts. The remaining proﬁt is transferred to the state budget. Within three months after the end of a calendar year, the CNB submits its annual proﬁt and loss statement to Parliament (lower house) for review. If Parliament rejects the ﬁnancial report of the CNB, it is obliged to submit a revised report within six weeks that complies with the requirements. Political independence Political independence is deﬁned as the ability of a central bank to choose its policy objectives autonomously, without constraints or inﬂuence from the government (Alesina and Grilli, 1992, p. 253). The primary objective of monetary policy, namely to maintain price stability, is enshrined in both the Constitution of the Czech Republic and the CNB Act. The bank Board sets monetary policy, the instruments for its implementation, and the principles for CNB activities and transactions with no interference from government. Each member of the bank Board, including the Governor and vice-governors, is appointed (or recalled) by the President of the Republic only, for a term of six years. Regulations are in place to prevent anyone from holding the position of bank Board member more than twice. Membership of the bank Board is incompatible with membership of any legislative body or government, and membership of the governing, supervisory or inspection bodies of other banks or commercial undertakings. Neither is it permissible for a member of the bank Board to engage in any independent gainful occupation (with several exceptions such as pedagogical or research activities). Recalling of any member of the bank Board is possible only in very speciﬁc circumstances (in the event he(she) no longer fulﬁls the conditions required for the performance of his(her) duties, has been guilty of serious misconduct or fails to perform his(her) duties for a period exceeding six months). Any decision to relieve the governor from his(her) ofﬁce may be referred to the European Court of Justice by the Governor concerned or the Governing Council of the ECB on grounds of infringement of the Treaty establishing the European Community. There is no requirement to gain formal approval of the government to implement monetary policy. The relevant statute simply allows a member of the government (appointed by the government) to attend bank Board sessions, but only in an advisory capacity and without voting rights. Economic independence Alesina and Grilli (1992, p. 255) deﬁne economic independence of the central bank as the ability to use, without restrictions, monetary policy

CBI in the Czech Republic 181 instruments to pursue its goals, and speciﬁcally not to be obliged to ﬁnance public deﬁcits. The CNB is prohibited from providing the government with direct credits. However, until 2001, the CNB was allowed to grant to the government short-term credit by way of purchasing state T-bills maturing within three months from the day of their purchase. The total amount was limited to 5 per cent of the revenues of the state budget of the previous year. In practice, there was usually sufﬁcient demand for state bills on the primary market. According to the new wording of the CNB Act (since 2001), the bank may not provide returnable funds or any other ﬁnancial support to the Czech Republic or its bodies, or to regional authorities, bodies governed by public law or legal entities under the control of the state, a regional authority or a body governed by public law, with the exception of banks, not even through the purchase of debt securities from such entities. On the other hand, the CNB is an active participant in the primary market for public debt. It offers for sale government bonds and, in agreement with the Ministry of Finance, performs on its behalf, for an agreed remuneration, activities related to the management, redemption and transfer of government bonds. The evaluation of the CNB independence Economic research regarding the performance of independent central banks has become very topical. Bade and Parkin (1988), Alesina (1988, 1989), Grilli et al. (1991), Cukierman (1992) and Eifﬁnger and Schaling (1992, 1993) have made major contributions to the literature. Each of these studies adopts the same methodology involving the identiﬁcation of indices for comparing the degree of central bank independence. However, one must be aware of the limits of using such an index. In many cases, there are differences between the legal and actual independence of central banks. More recently, attempts have been made to measure the independence of the CNB in other ways (see e.g. Eijfﬁnger and van Keulen, 1995; Radzyner and Riesinger, 1997; Sergi, 1994). However, there are some inexactitudes in these studies. For example, Sergi’s assertion that the direct credit facility is automatic and unlimited is, in our opinion, not correct (see pp below). We analyse the degree of central bank independence using four methods: the index of Bade and Parkin (1988), along with those of Grilli et al., Alesina and Grilli, and Cukierman. These indexes are all based on the interpretation of various elements of central bank legislation. We start with the Bade and Parkin index shown in Table 7.2. According to these criteria the Czech National Bank is completely independent. The second method of measuring the degree of central bank independence is through the index of Grilli, Masciandaro and Tabellini (1991), which consists of two parts: an index of political independence and an

182

Pavel Soukup et al.

Table 7.2 Bade and Parkin index of central bank independence Question

Score

1 Is the Central Bank the ﬁnal authority? 2 Are more than half of the policy Board appointments made independently of the government? 3 There is no government ofﬁcial (with or without voting power) on the central bank Board. Total score

* * * 4 (***)

index of economic independence. The criteria for political independence are those identiﬁed by Alesina and Grilli and are considered later. Here we consider only economic independence (Table 7.3).

Table 7.3 Grilli, Masciandro and Tabellini index (1991) of central bank independence Index of economic independence 1 2 3 4 5

The direct credit facility is not automatic. The direct credit facility is based on the market interest rate. The direct credit facility is temporary. The direct credit facility is for a limited amount. The central bank does not participate on the primary market for public debt. 6 The discount rate is determined by the central bank. 7 The banking supervision is not entrusted to the central bank. 8 The banking supervision is not entrusted to the central bank alone. Overall index of economic independence

* * * * – * – – 5

In our third evaluation we assess the independence of the CNB by using the index of Alesina and Grilli (1992). This approach identiﬁes two forms of independence: political and economic. The responses are drawn from the legal constraints embodied in the central bank statute – the CNB Act No. 6/1993 (see Table 7.4): The overall degree of independence is the sum of the asterisks. A higher index value implies a greater degree of central bank independence. Using the Alesina and Grilli (1992) index, the CNB has a very high degree of legal autonomy as declared explicitly in the CNB Act. Our ﬁnal evaluation of the independence of the CNB is based on the index constructed by Cukierman et al. (1992). This index is based on the interpretation of central bank laws (see Table 7.5).

CBI in the Czech Republic 183 Table 7.4 Alesina and Grilli (1992) index of central bank independence Index of political independence 1 2 3 4 5

Governor not appointed by the government. Governor appointed for more than ﬁve years. Board not appointed by the government. Board appointed for more than ﬁve years. There is no mandatory participation of government’s representative on the Board. 6 There is no government approval of monetary policy required. 7 Statutory requirements that CNB pursues monetary stability. 8 Explicit conﬂicts with the government are not possible. Overall index of political independence.

* * * * * * * – 7

Index of economic independence 1 2 3 4 5

Direct credit facility – not automatic. Direct credit facility – on market interest rate. Direct credit facility – temporary. Direct credit facility – limited amount. The CNB does not participate in primary market for public debt. 6 Discount rate set by CNB. 7 No portfolio constraints since 1993. 8 No bank loan ceilings since 1993. Overall index of economic independence. Total index

* * * * – * * * 7 14

This result, like the previous results, suggests that the CNB possesses a high degree of independence. However, a question arises as to whether the high degree of legal independence also implies a high degree of actual independence. In some theoretical papers (e.g. Eijfﬁnger and van Keulen, 1995), one learns that the CNB is legally very independent, but not necessarily independent in practice. The institutional framework of supervision and regulation There are three main possibilities for creating an institutional framework with responsibility for supervising the banking sector: within a central bank, a separate government agency, or within the Ministry of Finance. Table 7.6 provides a review of banking supervision arrangements across the European member countries of the IMF. The United Kingdom, France, Greece, Italy, Luxembourg, The Netherlands, Portugal and Spain have entrusted central banks with the role of supervising the banking sector. However, in Austria, Belgium, Denmark, Germany,1 Ireland and Switzerland, bank supervision is carried out by a separate government

184

Pavel Soukup et al.

Table 7.5 Variables, weights and numerical coding for legal central bank independence Description of variables

Weight

1 Governor (a) Six-year term of ofﬁce (b) Appointed by the legislative power (c) Recalling possible only for reasons not related to policy (d) Holding another government ofﬁce forbidden 2 Policy formulation (a) CNB alone formulates monetary policy (b) CNB’s responsibility clearly deﬁned in the law as its objectives (c) CNB is active in the budgetary process (advice) 3 Objectives Price stability is one of the CNB objectives with other objectives, such as a stable banking system 4 Limitation on lending to the government (a) No advances permitted (b) Securitised lending permitted with the limit up to 5 per cent of government revenue (c) Maturity of loans provided by the CNB is speciﬁed in the CNB’s Act (three months) (d) Potential borrowers from the CNB only the central government (e) Limits on central bank lending are given as a share of government revenue (f) Maturity of loans is three months (g) Interest rates on loans is not mentioned (h) The CNB can sell and buy Treasury bills on the primary market Overall Index of Central Bank Independence

0.20

Num. cod 0.75 0.50 0.83 1.00

0.15 1.00 1.00 0.50 0.15 0.60 0.15 0.10

1.00 0.67

0.10

0.67

0.05

1.00

0.025

0.33

0.025 0.025 0.025

1.00 0.25 0.00 0.768

Source: Cukierman et al. (1992).

agency. It is worth noting that the Maastricht Treaty does not presume to include banking supervision within the remit of the European Central Bank. The nature of supervision varies between countries in Central and Eastern Europe. For example, in Hungary banking supervision is performed by a separate supervisory agency, whereas in Poland and Romania banking supervision is entrusted to the central bank. In Bulgaria and the Czech Republic, banking supervision is currently entrusted to the central bank, although possible changes are currently being explored. Table 7.6 shows the nature of supervisory arrangements currently in place in different countries. When the central bank is simultaneously charged with responsibility for implementing monetary policy and supervision of the banking sector, an

CBI in the Czech Republic 185 Table 7.6 Banking supervision arrangements Country

Central bank

Albania Austria Belgium Czech Republic Denmark Finland France Germany Greece Hungary Ireland Italy Luxembourg Netherlands Norway Poland Portugal Romania Slovak Republic Spain Sweden Switzerland United Kingdom

X

Ministry of Finance Other agency X X

X X X X X X X X X X X X X X X X X X X X

Source: Tuya and Zamalloa (1994).

important question arises about whether there is any conﬂict of interest. The primary target of monetary policy is stability of the currency, whereas the key tasks of banking supervision are to ensure that the banking system displays the elements of efﬁciency, stability and credibility. In fact, a potential conﬂict of interest would arise if a central bank implemented a tough monetary policy in order to slow down inﬂationary pressure in the economy. This would inevitably entail an interest rate increase, but increases in the rate of interest increase the probability of loan defaults for banks. Conversely, banking supervision decisions also have a direct impact on monetary policy. For example, the limitation of bank risks could cause a reduction in the growth of new loan business at a time when the authorities were easing monetary policy with the aim of encouraging economic expansion. Nevertheless, we believe that the inclusion of banking supervision within the remit of the central bank has a number of advantages. First of all, it is quite clear that if banking supervision is encompassed within a central bank, negotiations and co-ordination of monetary policy and bank supervision policy should be easier. At the same time, it is required, or it should be required, that bank supervision is independent of political

186

Pavel Soukup et al.

pressures. It follows that if the central bank is highly independent, it is beneﬁcial to place banking supervision within the remit of the central bank or, if the evidence on the independence of the CNB is accepted, it must be beneﬁcial to retain the supervisory role of the CNB. An alternative option is to place banking supervision under the auspices of the Ministry of Finance. However, this might create a moral hazard problem since the Ministry would be in a position to pressure the department with responsibility for bank supervision to soften its requirement on banks with respect to ﬁnancing the central government budget deﬁcit. In recent years in the Czech Republic, the position of the Ministry of Finance has been adversely affected because banks have been able to gain a tax concession from building up their reserves! The third possibility is to establish an agency independent of the state, as exists in Hungary. Although the results of this supervisory framework have not displayed any weaknesses up until now, state bureaucracy can threaten the independence of such an institution. It seems, therefore, that the main issue in deciding whether or not to place bank supervision within the terms of reference of the central bank is the impact of this decision on the credibility of a central bank. The banking situation in the Czech Republic is complicated by the difﬁculty of establishing a market-based banking system in the course of only a few years. It is our view that the supervisory arrangements currently in place are adequate and do not impinge adversely upon the credibility of the CNB. Nevertheless, it is essential to make clear at all times the objectives and strategies of banking supervision.

5 From monetary targeting to inﬂation targeting From the beginning of the transition process, the monetary policy of the central banks in Central and Eastern Europe has been restrictive, primarily reﬂecting anxieties concerning inﬂation that followed the liberalisation of prices. In Czechoslovakia, monetary policy in the ‘pre-transition’ year reﬂected a complicated situation. On the one hand, the economy was subject to the inﬂuence of direct instruments, and on the other, a space gradually opened up for the effect of indirect instruments which were tested during 1990. Departing from the totally regulated system of a centrally planned economy, the central bank installed credit volume and interest rate ceilings to assist the implementation of their monetary and credit policies. The central bank set individual credit volume limits for all major banks. In 1991 the overall monetary objective was expressed as a percentage of the permitted growth of credit and loans (from ⫹1 to ⫹2.6 per cent to the year 1989). The main instruments of policy were limits, minimum reserve requirements, credit reﬁnancing, discount and interest rate ceilings, and exchange rate policy. Combining these instruments provided a restriction

CBI in the Czech Republic 187 on the growth of credits and interest rates which established a platform for currency stability thereafter. In 1992 the application of monetary policy switched from direct to indirect instruments. After the abolition of both interest and credit ceilings, the central bank regulated monetary development on the basis of analyses of the monetary situation. From 1992, the central bank was able to operate with all the main instruments used in standard economies, such as changes in discount and Lombard rates and in minimum reserve requirements. The role of intermediate target was performed by the money supply measured by the M2 aggregate. The operational target was the money market interest rate 1W PRIBOR (one Week Prague Inter-Bank Offered Rate), which is mainly under central bank control through the use of open market operations. Open market operations are carried out largely on the basis of repurchase agreements (repos). However, since mid-1994, the ability of the central bank to exercise control over growth of the money supply deteriorated signiﬁcantly in the Czech Republic. The money stock M2 increased by 21.5 per cent in 1994 which was an increase on the previous year and moved M2 out of its planned ‘corridor’. Adequate control of monetary aggregates has become more difﬁcult, particularly due to short-term capital inﬂows, and M2 now functions as a monetary indicator only. As far as exchange rate policy is concerned, the stable exchange rate of the Czech koruna and the ﬁxed exchange rate system were a signiﬁcant part of economic policy. The nominal exchange rate of the CSK (CZK after the currency split in 1993) was adjusted in four steps in the course of 1990 and the resulting cumulative devaluation amounted to 95.6 per cent. Until February 1996, the CZK moved within a rather narrow band of ⫾0.5 per cent of its central parity. The band was widened on 28 February 1996 to ⫾7.5 per cent without any change in the central parity in response to monetary difﬁculties. The most important change in exchange rate policy occurred in May 1997, during the ‘monetary shake-up’. The CNB was forced to allow the exchange rate to ﬂoat after heavy intervention and sharp interest rate hikes because it failed to fend off the speculative attack on the Czech currency. While allowing the Czech koruna to ﬂoat, monetary policy has eased somewhat, but needs to remain tight enough to keep the currency stable. At the end of 1997, the CNB announced a change in its monetary policy regime. It adopted direct inﬂation targeting as its new strategy with the aim of achieving its inﬂation target of 3.5–5.5 per cent by the year 2000. Rationale for the inﬂation targeting A number of countries have adopted the strategy of monetary targeting, arguing that the central bank should be held accountable only for the

188

Pavel Soukup et al.

monetary impulses to inﬂation which it can control more directly than developments in inﬂation. An alternative approach is direct inﬂation targeting, where more emphasis is put on transparency, because inﬂation forecasts are more widely understood than monetary aggregates. Moreover, a formal target is a direct reﬂection of the ultimate goal of price stability. In transition economies monetary targeting is more difﬁcult than in developed economies for two main reasons. First, velocity of circulation is neither stable nor predictable, and second, the estimation of potential output suffers from many uncertainties. This partly reﬂects the fact that transition involves signiﬁcant institutional change – privatisation, ﬁnancial and capital market restructuring and so on. In the Czech Republic the loss of a nominal anchor in the form of a ﬁxed nominal exchange rate, or a set ﬂuctuation band, together with the limited capability of monetary aggregates to fulﬁl the role of an intermediate target, have necessitated the introduction of a disciplinary element into the economy designed to encourage rational inﬂation expectations. A system of direct inﬂation targeting, involving a clearly deﬁned and quantiﬁed inﬂation target, has been introduced to perform this function. There is also some empirical evidence to suggest that inﬂation targeting is a more efﬁcient method of decreasing inﬂation than other strategies of monetary policy. Hrncir and Smidkova (1999) have shown (Table 7.7) that countries which have adopted a strategy of inﬂation targeting achieve more desirable outcomes in terms of lower inﬂation and higher GDP

Table 7.7 Average inﬂation and GDP growth in chosen countries Countries with inﬂation targeting

Decade before IT adoption inﬂation (GDP)

Decade after IT adoption inﬂation (GDP)

Australia Canada Finland Sweden New Zealand Great Britain Average

6.2 5.8 5.2 6.6 11.6 5.2 6.8

2.7 2.0 1.1 2.3 2.5 2.8 2.2

Other countries

1980s

France Germany Italy Japan USA Average

7.4 2.9 11.3 2.5 5.6 5.9

3.2 2.8 1.4 1.6 1.8 2.4 2.2

4.2 1.9 3.2 1.9 2.4 3.0 2.8

1990s 2.3 1.8 2.4 3.8 2.8 2.6

2.3 3.1 5.2 1.4 3.4 3.1

1.4 2.1 1.2 2.2 2.1 1.8

CBI in the Czech Republic 189 growth than other countries. These results are very encouraging – and not only for the Czech Republic! Inﬂation targeting in practice The main features of inﬂation targeting are its medium-term focus, the use of an inﬂation forecast and the explicit public announcement of an inﬂation target or sequence of targets. In its monetary policy decision-making the CNB Board assesses the latest CNB forecast and evaluates the risks of non-fulﬁlment of this forecast. Based on these considerations the bank Board then votes on whether and how to change the settings of monetary policy instruments. By changing these instruments, the central bank seeks to offset excessive inﬂationary or disinﬂationary pressures which cause deviations in forecast inﬂation from the target band. For example, an increase in the repo rate generally leads, via the transmission mechanism, to a weakening of aggregate demand, which in turn causes inﬂation to fall. Lowering the repo rate generally has the opposite effect. If the central bank expects that future inﬂation will deviate above the inﬂation target, this is a signal that monetary policy should be tighter; that is, that the repo rate should be increased. At the beginning of the new monetary policy regime, the inflation target was set at the operational level of 6 ⫾ 0.5 percentage points of the ‘net inflation’ for the end of 1998 and a final level of 4.5 ⫾ 1 percentage points for the end of 2000. Net inflation was especially defined for the purpose of inflation targeting. It represents the movement in unregulated prices; that is, the CPI excluding changes in so-called regulated prices adjusted for the influence of indirect taxes or subsidies. In 1998, the CPI consisted of 754 items, of which ninety-one items were for regulated prices leaving 663 items contributing to net inflation. Taking into account the weights of particular goods and services in the consumer basket, the ratio of the two price groups was 82:18 in favour of net inflation for 1998. The CNB Monetary Strategy document published in April 1999 formulated the long-term objective of price stability in terms of net inﬂation within the 1 to 3 per cent range for the end of 2005. In April 2000, a target was set for the end of 2001, and in April 2001 a decision was made to switch to targeting headline inﬂation; that is, the growth in the total consumer price index, and to expressing the target trajectory for headline inﬂation by means of a continuous band. A band was announced starting in January 2002 at 3 to 5 per cent and ending in December 2005 at 2 to 4 per cent. Once announced, all inﬂation targets remain unchanged. Changes in inﬂation, or expected inﬂation, result from shocks, whose inﬂationary or disinﬂationary effects unwind over a period of time. Any attempt to entirely offset these effects by changing the settings of monetary policy instruments would needlessly cause short-term ﬂuctuations in

190

Pavel Soukup et al.

the economy. If inﬂation is moving for a certain period outside the target band due to exogenous factors (particularly demand-side shocks), it is usually within the inﬂation targeting regime to regard this deviation as an exception to the central bank’s inﬂation target. The CNB, like many other inﬂation-targeting central banks, has adopted this practice and, in particular circumstances, works with a set of ‘caveats’; that is, exceptions to its obligation to hit the inﬂation target. These exogenous factors were ﬁrst deﬁned at the end of 1998 and extended in April 2001 to include: • • • • • •

major deviations in world prices of raw materials, energy-producing materials and other commodities; major deviations of the koruna’s exchange rate that are not connected with domestic economic fundamentals or domestic monetary policy; major changes in the conditions for agricultural production having an impact on agricultural producer prices; natural disasters and other extraordinary events having cost and demand impacts on prices; changes in regulated prices whose effects on headline inﬂation would exceed 1 to 1.5 percentage points; step changes in indirect taxes.

As the CNB stated in its Monetary Strategy, in connection with the Czech Republic’s accession to the European Union, ‘the set of escape clauses is also likely to include one involving one-time, difﬁcult-to-predict price shocks associated with the adoption of EU standards’. When evaluating inﬂation developments, the CNB uses the statistical data published by the CSO only. It is expected that such arrangements (using data published by an institution independent of the CNB) can improve the transparency and credibility of inﬂation targeting and inﬂuence the formation of inﬂationary expectations.

6 Conclusions It is obvious from a comparison of the CNB legislation with the requirements set out in the Maastricht Treaty (or the Protocol on the Statute of the ESCB and ECB) that there are no signiﬁcant discrepancies in the legal nature of central bank independence. Moreover, the actual independence of the CNB is also probably quite close to the level in EU countries. It is important to be clear that a high degree of central bank independence does not guarantee per se an anti-inﬂationary consensus in a country. In order to achieve the long-term goal of price stability at the lowest possible cost, it is desirable to ensure a broad political consensus, supported by the main players in the economy, including not only government, but also the parties involved in the wage-setting process.

CBI in the Czech Republic 191

Note 1 In Germany, however, the Federal Banking Supervisory Ofﬁce mainly provides licences. Off-site supervision is carried out by the Deutsche Bundesbank.

References Alesina, A. and Grilli, V. (1992) The European Central Bank: Reshaping Monetary Politics in Europe, in Canzoneri, Grilli and Masson (eds), Establishing a Central Bank, Issues in Europe and Lessons from the US, Cambridge: Cambridge University Press. Bulletin ’90 SBCS, State Bank of Czechoslovakia 1990. Collection of Lectures from the Conference in Honour of the 70th Anniversary of Central Banking in the Czech Republic, CNB, 1996. Cukierman, A., Webb, S.B. and Neyapti, B. (1992) Measuring the Independence of Central Banks and Its Effects on Policy Outcomes, The World Bank Economic Review, 6. Czech National Bank, Indicators of the Monetary and Economic Development, 1996 to 2001. Czech National Bank, Inﬂation Report, 1998 to 2002. Czech National Bank, State Bank of Czechoslovakia, Annual Reports 1990 to 2001. Dedek, O. et al. (1996) The Break-up of Czechoslovakia: An In-depth Economic Analysis, Avebury. Eijfﬁnger, S. and van Keulen, M. (1995) Central Bank Independence in Another Eleven Countries, BNL Quarterly Review, 192 (March), pp. 39–83. Grilli, V. (1991) The European Central Bank: Reshaping Monetary Policies in Europe, CEPR Discussion Paper, No. 563. Grilli, V., Masciandaro, D. and Tabellini, G. (1991) Political and Monetary Institutions and Public Financial Policies in Industrial Countries, Economic Policy, 13, pp. 341–392. Hrncir, M. and Smidkora, K. (1999) The Czech Approach to Inﬂation Targeting, in Workshop on Inﬂation Targeting, Proceedings of Czech National Bank, Prague. Matousek, R. (1996) Recent Monetary Developments in the Czech Republic, ACE Workshop, 31 May to 1 June, Budapest. Pulpanova, S. and Pulpan, K. (1994) Bank Ofﬁce of the Ministry of Finance, Finance a úver, 8, pp. 419–428. Radzyner, O. and Riesinger, S. (1997) Central Bank Independence in Transition: Legislation and Reality in Central and Eastern Europe, Focus on Transition, 1, Oesterreichische Nationalbank, pp. 57–90. Report on the establishment of the Czech National Bank, CNB 1993. Sergi, B.S. (1994) The Autonomy of the Czech National Bank, Prague Economics Papers, 2 (June), pp. 165–174. Tietmayer, H. (1996) Independence, Please, WSJ, 22 April. Tosovsky, J.(1995) Managing Financial Turbulence: Czech Experience, CNB draft. Tosovsky, J. (1996) Losses of our Banking Sector are the Price for the Transformation of the Czech Economy, interview for the ‘Hospodarske noviny’ Newspapers, 9 September.

192

Pavel Soukup et al.

Tuya, J. and Zamalloa, L. (1994) Issues on Placing Central Bank Supervision in the Central Bank, in T. Balini and C. Cottarelli (eds), Framework for Monetary Stability, Washington, DC: IMF. Vencovsky, F. (1996) Solving Monetary Policy Issues during the First Republic, Finance a úver, 4, pp. 189–203.

8

Central bank independence in Poland Robert Huterski, Richard Nicholls and Zenon Wis´niewski

1 Introduction This chapter provides a detailed case study of central bank reform in Poland. Section 2 considers the new policy environment in the 1990s and the institutional reforms introduced over this period. Section 3 outlines the present constitutional status of the National Bank of Poland (NBP), including its structure and management, concluding with an assessment of its political and economic independence using the standard index-based measures.

2 The 1990s: a new environment and a new independence Assumptions and directions of state policy The ﬁrst non-communist government in Poland, led by Tadeusz Mazowiecki, commenced its activities in September 1989 and initiated the execution of a new economic strategy at the beginning of 1990. The strategy aimed to bring about a fundamental transformation in the Polish economy and involved two radical steps: rapid stabilisation and the immediate liberalisation of prices, exchange rates and international trade. Taken together, these steps were intended to end ﬁnancial chaos and to introduce competition from abroad so as to create an appropriate environment for an efﬁcient allocation of resources. At the same time, the strategy included a series of legislative and executive measures aimed at creating the foundations of a market economy. These measures included privatisation, a competition programme to break up and prevent monopolies, removal of entry restrictions on new enterprises, banking system modernisation, tax reform and so on. As a consequence of the policy launched in 1990, the system of automatic ﬁnancing was abandoned. The policies of the National Bank of Poland (NBP) had therefore to be fundamentally changed. According to the amendment updating the banking law and NBP Acts, central bank policy was to focus on money rather than adopting the previous approach which

194

Robert Huterski et al.

focused on money and credit. This apparently slight alteration in terminology reﬂects a desire to change the nature of operations carried out by the NBP with regard to both its objectives and the methods used to fulﬁl them. Since 1990 the main objective of NBP policy has been control of the money supply. This anti-inﬂationary stance is based on two foundations. First, the NBP aims at ensuring a positive real rate of interest. Second, a ﬁxed zloty rate of exchange to the dollar was introduced. The positive rate of interest was meant to encourage savings in the national currency, as well as to lower the demand for loans. The ﬁxed zloty exchange rate was designed to stabilise inﬂation by providing a nominal anchor for monetary policy. It was also expected to impact directly on domestic prices in two ways. First, by ﬁxing the rate of exchange, the domestic prices of imported goods would be stabilised and this source of rising prices would therefore be eliminated. Second, it was envisaged that the ﬁxed rate of exchange would limit the growth of inﬂationary expectations which were widely regarded as a major factor in generating rising inﬂation (Bruno 1991). The programme introduced, although designed by Polish economists, was a typical International Monetary Fund (IMF) adjustment programme. Both its assumptions and accomplishments were monitored and assessed by the IMF. Many critics accused the government of undertaking macroeconomic policies that were too restrictive, and even blamed it for the recession that followed. However, in reality the basic causes of this recession were the rapid breakdown of the centrally planned economy and the radical system-transforming reforms that followed. It seems to us that both recession and inﬂation are the costs which each post-socialist economy has to pay for its transition into a market system. Unfortunately there is no way to avoid this. Both are the result of inefﬁciencies and distortions created by the planning mechanism. Monetary policy: tools and results Since 1990, the NBP has followed a clear anti-inﬂationary policy. However, the actual instruments used have evolved during this period. Such changes in the application of monetary policy are largely a reﬂection of the changing nature of the ﬁnancial system and the emergence of new economic problems. Lags in behavioural changes by both lenders and borrowers were also inﬂuential. Despite varying emphasis on different techniques within the period, from 1990 onwards the primary goal of NBP monetary policy has been the control of inﬂation with money growth as the intermediate target. Period 1990 to 1991 As noted above, the stabilisation programme introduced in 1990 aimed at restricting the growth of credit. In January 1990 the NBP increased its dis-

CBI in Poland 195 count rate from 10 per cent monthly to 36 per cent monthly. Other measures designed to tighten monetary policy were also introduced. In particular, reserve requirements were increased and open-market operations based on Treasury bills issued by the NBP were introduced in July 1990. However, contrary to expectations, the rate of credit expansion continued unrestrained, and in October 1990 the Governor of the NBP introduced credit ceilings. The direct limitation of credits was designed to control the growth of net domestic assets. The experience of 1990 suggests that at this stage the NBP was ineffective in its control of the money supply. Table 8.1 shows that in the period 1990 to 1991 the money supply continued to grow at a relatively high rate. The 1990 to 1991 credit expansion has to be viewed in the context of inertia in the early stages of the transition process. At this time, most credits continued to be supplied by large state banks to large state ﬁrms. Behaviour remained substantially unchanged: state ﬁrms continued to request new credits, despite paying higher interest rates, and the banks frequently obliged – usually without much analysis of borrowers’ creditworthiness! Indeed, much of the demand for credits originated from the so-called ‘payment jams’; that is, multilateral delays in payment due to lack of liquidity. This constrained policy actions by the NBP because any further increase in interest rates would have added to the cost of servicing existing credits thus extending the ‘payment jams’. A second important inﬂuence on the money supply came from foreign trade. In 1990 Polish exports rose dramatically (15.1 per cent) while imports fell substantially (10.2 per cent), causing a surplus in the balance of payments current account. While reserves of foreign currency had been Table 8.1 Money supply growth in Poland, 1989 to 2001a Year

Money supply (%)

1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

190.5 396.3 64.8 57.2 28.8 38.7 50.2 30.3 29.1 25.6 19.3 11.7 13.7

Source: Polish Central Statistical Ofﬁce. Note a Domestic (M2 minus foreign currency accounts).

196

Robert Huterski et al.

expected to fall in 1990, in fact they increased by US$4.5 billion, giving an upward impetus to money growth. Period 1992 to 1993 At the end of 1992, it was decided that credit ceilings were no longer necessary and during 1993 they were phased out. This reﬂected the reality that large state banks were by now reacting to the combination of a deepening recession and an escalation in the number of uncreditworthy debtors. Not only did the credits previously granted frequently prove to be bad loans, but the collapse of the Council for Mutual Economic Assistance (Comecon) led to the bankruptcy of many ﬁrms. This worsened the bad loans crisis, which reached its peak in 1992 to 1993. Against such a background, banks increasingly favoured ﬁnancing the growing budget deﬁcit (in 1992 the budget deﬁcit was 6 per cent of GDP) rather than the much more risky business of loans to ﬁrms. Consequently credit ceilings on ﬁrms and households became irrelevant in the light of banks’ slow-down in credit activities, and led to increased purchases of treasury bills. Indeed, the actual credit expansion of state banks was below the levels imposed by credit ceilings. Accordingly the NBP switched to relying entirely on the use of interest rate policy to control net domestic assets. However, NBP policy was not based merely on discount rate policy, but also involved a crucial role for open market operations. Indeed, the increasing role of Treasury bills in the portfolios of banks (at the end of 1993 Treasury securities accounted for 9.3 per cent of commercial banks’ assets) has enabled the NBP to engage regularly in open market operations as its main tool to control interest rates and the money supply since the beginning of 1993. Period 1994 to 1995 In 1994 the NBP changed both its operational and intermediate targets and switched the focus of policy to the control of interest rates. One reason for this change was that banks were becoming less and less sensitive to the level of central bank rates. Banks began lowering their interest rates on long-term deposits, particularly in January to February 1995. This was extraordinary for two reasons. First, the monthly rate of inﬂation was rising, and second, the NBP had kept its rates unchanged since May 1994. A number of factors contributed to the banks’ behaviour. First, the 1994 budget deﬁcit, at 58 billion zlotys, was lower than expected due to buoyant budgetary revenues and, given the NBP’s obligation under the 1994 Budget Act to ﬁnance the deﬁcit by no less than 30 billion zlotys, the NBP actually ﬁnanced over 50 per cent of the deﬁcit. Consequently banks ended up competing for a reduced availability of Treasury bills, causing the proﬁtability of these bills to decrease. In 1994

CBI in Poland 197 the proﬁtability of Treasury bills fell by 7 points and the proﬁtability of one-year Treasury bills fell by 12 points (from 38 per cent to 26 per cent). Second, the growing foreign exchange reserves (from mid-1994) contributed to the appearance of structural over-liquidity in the banking system. Indeed, in 1995 and 1996 foreign exchange reserves were the main contributor to money growth and, in 1995 alone, gross foreign reserves increased by some 250 per cent, from $6 billion to almost $15 billion. Third, banks were more inclined to expand their credit activities. Important factors inﬂuencing this change include: (1) a declining volume of bad loans from 1994 due partly to new legal measures enabling banks to settle bad loan problems with their corporate clients, and partly to tighter banking supervision; (2) reduced economic uncertainty; (3) the development of consumer credit. However, given their credit policies, the banks had limited investment possibilities for the deposits which were ﬂowing in. Besides, against the background of privatisation of state banks and the entrance of foreign banks to the Polish market, banks were increasingly forced to compete with each other by offering lower interest rates to creditors. In consequence, banks concentrated their lending activities on the purchase of governmental papers and on operations with the NBP. The NBP’s operations (so-called conditional operations with a maturity of up to fourteen days and ﬁxing of one-day operations’ interest rate at 26 per cent) enabled it to prevent too deep a decrease in eight and twentysix-week Treasury bonds, but it had an imperceptible inﬂuence on the proﬁtability of ﬁfty-two-week Treasury bonds. In this latter case, the NBP attempted to counteract their decrease by outright (unconditional) operations and by issuing its own ‘monetary bonds of the NBP’. When these actions failed to bring the expected results, in February 1995 the NBP increased its discount rate. This move suppressed the reduction of deposit interest rates offered by banks to their customers. At the same time (midMay 1995) the NBP allowed the exchange rate to ﬂoat within a wider band of ﬂuctuation (from ⫾2 per cent to ⫾7 per cent), relieving expectations of appreciation and reducing for a few months the monthly level of foreign currency inﬂow to the NBP. These actions assisted in dampening inﬂationary expectations and the annual inﬂation rate fell from 30 per cent in June 1995, to 20 per cent in December 1995. Period 1996 onward In 1996 and the ﬁrst half of 1997 there was a relatively large credit boom in Poland. During the period 1995 to 1997 nominal credits to business grew by 37 per cent and to households by 107.80 per cent. The growth in the volume of credit was signiﬁcantly inﬂuenced by the fall in nominal interest rates, even though real interest rates in 1996 and 1997 averaged 12 to 15 per cent. The credit boom allowed economic growth to accelerate. In 1996

198

Robert Huterski et al.

GDP grew 6.1 per cent and in 1997 it grew 6.9 per cent. This rapid growth rate meant that imports grew faster than exports and the current account deﬁcit rose from 1 per cent of GDP in 1996, to 3.1 per cent in 1997. At the same time the saving rate declined from 13.5 per cent of GDP in 1995, to 10.5 per cent in 1996. Accordingly, in the second half of 1997, the NBP sought to limit credit creation and raised interest rates from 2 to 2.5 per cent, thus increasing the cost of reserve requirements. The increase in interest rates brought to an end what had come to be regarded as a dangerous credit boom. The Polish economy managed to resist the Asian and Russian crises of 1998, but the rate of GDP growth declined. This is testimony to the resilience of the Polish economy given that until 1998 Russia was Poland’s third largest trading partner, and in 1999 Polish exports to Russia were 74 per cent down on 1998. The general opinion is that the Polish economy is not in recession, but is suffering from a temporary deterioration in economic conditions. This slowdown is predicted to be followed by an economic revival before the end of 1999. Given the unfavourable external situation, the current account deﬁcit for 1999 may reach 6 per cent, and this causes pressure for directing short-term monetary policy at improving the current account deﬁcit situation. Economists have also been concerned about the high budget deﬁcit in 1999 which, after only the ﬁrst four months, reached 72.6 per cent of the ﬁgure envisaged for the entire year. However, the high budget deﬁcit in the ﬁrst half of the year is linked to reform-related expenditure, which should diminish. Indeed, the Polish government’s macroeconomic assumptions for the 2000 budget assumed a 5 per cent rise in GDP and year-onyear inﬂation falling to 5.7 per cent. The medium-term inﬂation target is 4 per cent by 2003. In January 1999, the NBP cut interest rates. The key Lombard rate was reduced from 20 per cent to 17 per cent. It was not a surprise that rates were cut, but many were shocked by the size of the cut (the average Lombard rate for 1998 was 24.7 per cent). There seem to be a number of considerations behind the move. The large cut was intended as a clear signal to commercial banks that they should reduce their credit rates. The cut also made Poland a less attractive location for hot money, and a weaker zloty helps exports. On the other hand, lower interest rates tend to increase domestic demand and accelerate the growth of imports. It is unfortunate that the optimism associated with the interest rate cut was soon drowned out by the pessimism associated with the extensive social unrest, in particular the road blockades carried out by Polish farmers desiring increased price subsidies. The 3 per cent interest rate cut was useful in helping to reduce inﬂationary expectations, but it also means that subsequent cuts will not seem so spectacular. The NBP had allowed the Polish zloty to move within a wider band of ﬂuctuation since March 1999. The new band was ⫾15 per cent, and this

CBI in Poland 199 was seen as a move towards liberalising the zloty. This is an important policy goal because allowing the market to freely determine the value of the zloty is an essential condition for entry into the euro. Since April 2000, the zloty exchange rate has been a ﬂoating rate not subject to any restrictions. The central bank does not set a target for the predetermined zloty exchange rates against other currencies. The NBP is not obliged to intervene on the foreign exchange market, though it has the right to engage in such interventions if necessary to achieve the inﬂation target. The zloty was seen increasingly as a more stable currency, but at the beginning of 2003 the Ministry of Finance announced its plans of interventions depreciating the Polish currency in order to boost economic growth. Clash with the central bank seems to be inevitable. In assessing the success of monetary policy in reducing inﬂation, it should be noted that inﬂation has been a signiﬁcant feature of Polish society for two decades. Accordingly, the current monetary policies are burdened with widespread inﬂationary expectations which are the legacy of the former system. Reference to Table 8.1 and Table 8.2 shows that despite the problems of monetary control, inﬂation in Poland has declined substantially throughout the 1990s. The medium-term target of monetary policy strategy for the years 1999 to 2003 to lower the inﬂation rate to below 4 per cent by the end of 2003 has been achieved. Setting the target at the level of 3 per cent with the ﬂuctuation range ⫹/⫺1 percentage point means the end of the inﬂation lowering phase and the beginning of the phase of stabilising inﬂation at a low level.

Table 8.2 Inﬂation, 1989 to 2002 Year

1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

Inﬂation (%) Dec. to Dec.

Yearly average

640.3 249.3 60.4 44.3 37.6 29.5 21.6 18.5 13.2 8.6 9.8 8.5 3.6 0.8

351.1 685.8 170.3 143.0 135.3 132.2 127.8 199.9 114.9 111.8 107.3 110.1 105.5 101.9

Source: Polish Central Statistical Ofﬁce.

200

Robert Huterski et al.

There has also been a signiﬁcant increase in the real money demand for Polish currency. Inﬂuential factors bringing this about are: (1) the economic revival which generated additional demand for transactions balances, and (2) the NBP policy of positive real interest rates which have encouraged saving in zloty denominated deposits.

3 The National Bank of Poland today The year 1997 was last when former Polish central banking regulations were in force. The new Act of Parliament passed in August 1997 (Ustawa z dnia 29 sierpnia 1997) radically changed the organisation of the NBP and introduced two new collegial bodies responsible for the formulation of monetary policy and banking supervision. The majority of these changes came into force at the beginning of 1998. However, some of them (e.g. the prohibition of the central bank lending to the government) were introduced with a one-year delay. Because the description of the general economic environment and monetary policy transformation in Poland in the ﬁrst two parts of this chapter provides only a background to the legal framework of central bank activity created over the past decade, we present a more detailed analysis below. It is not possible to assess their inﬂuence on the real independence of the NBP, especially since not all important supplementary acts and decisions are yet issued, but the amount of information available is sufﬁcient to make a comparison of the old and new legal frameworks (see Table 8.3). General provisions According to the 1997 NBP Act, the National Bank of Poland is the central bank of the state. The NBP is the bank of issue of the state, the central credit and settlement institution, and the central banking foreign exchange institution. The NBP has the status of a legal entity, although it is not registered as a state-owned enterprise. Such a description is controversial, and many people consider that the central bank should be named precisely as a part of the system of state authority institutions. The major economic aim of the NPB is the maintenance of price stability, while at the same time acting in support of government economic policies insofar as this does not constrain pursuit of the basic objective of the NBP. The reformulated catalogue of detailed responsibilities of the NBP includes: 1 2 3 4

organising monetary settlements, managing the ofﬁcial foreign exchange reserves, conducting foreign exchange operations within the bounds stipulated by statute, providing banking services to central government,

CBI in Poland 201 Table 8.3a Index of legal independence – comparison of results for 1997, 1998 and 1999 No.

Description of variable

Numerical coding Weight

1

2

3 4

Chief executive ofﬁcer (CEO): (a) Term of ofﬁce (b) Who appoints CEO (c) Dismissal (d) May CEO hold other ofﬁces in government? Policy formulation: (a) Who formulates monetary policy? (b) Who has ﬁnal word in resolution of conﬂict? (c) Role in the government’s budgetary process Objectives: Limitations on lending to the government:a (a) Advances (b) Securitised lending (c) Terms of lending (maturity, interest, amount) (d) Potential borrowers from the bank (e) Limits on central bank lending deﬁned in (f) Maturity of loans (g) Interest rates on loans must be (h) Central bank prohibited from buying or selling government securities in the primary market The NBP legal independence index:

1997

1998

1999

0.75 0.50 0.83 1.00

0.75 0.50 0.83 1.00

0.75 0.50 0.83 1.00

0.15 (0.25) (0.50)

0.75 0.40

1.00 1.00

1.00 1.00

(0.25)

1.00

1.00

1.00

0.15

0.60

1.00

1.00

0.15 0.10 0.10

1.00 0.67 0.67

1.00 0.67 0.67

1.00 1.00 1.00

0.05 1.00 (0.025) 1.00

1.00 1.00

1.00 1.00

(0.025) 1.00 (0.025) 0.75 (0.025) 0.00

1.00 1.00 0.00

1.00 1.00 1.00

0.75

0.86

0.95

0.20

Notes Weights in brackets (0.25), (0.50) and (0.25) are used in the ﬁrst round of aggregation. a Since the beginning of 1999 all forms of central bank credit facility to the government have been forbidden and, as a result, the ﬁrst four questions from the index of economic independence become irrelevant. Positive answers are given to comply with the internal logic of this index despite the fact that there should be no answers at all according to the formal construction of these questions.

5 6 7

8

regulating the liquidity of the banks and providing them with reﬁnancing facilities, establishing the necessary conditions for the development of the banking system, drawing up an account of the national balance of payments for reporting purposes, together with balances of the foreign assets and liabilities of central government, performing other responsibilities as speciﬁed by statute.

202

Robert Huterski et al.

Table 8.3b Index of the turnover rate of the NBP Governor (average number of changes a year) Period

Number of Governors during ﬁve years

Term of ofﬁce (on average)

Turnover rate of the NBP Governor

September 1989 to September 1994 August 1994 to August 1999 January 1998 to January 2003

4

1.25 year

0.8

1

5 years

0.2

2

2.5 years

0.4a

Note a Former Governor of the NBP, Mrs Hanna Gronkiewicz-Waltz, stepped down from the ofﬁce in December 2000 because she became Vice-Governor of the European Bank for Reconstruction and Development.

Within the sphere of its activity, the NBP also co-operates in the formation of economic policy of the state and its realisation, initiates the directions and forms of monetary policy, and represents the interests of the state in co-operation with international banking institutions and foreign banks. The main shift in the description of central bank activities lies in assignment of greater weight to its primary objective of maintaining price stability. This is a narrower task than its former objective of strengthening the Polish currency. Support for government economic policy became a conditional, no longer parallel, co-objective. In addition, the inﬂuence of Parliament is visibly reduced giving the NBP greater goal and instrument independence from the state. Present structure of the National Bank of Poland 1 The structure of the NBP comprises its head ofﬁce in Warsaw (divided into functional departments) and thirteen regional branches. The head ofﬁce departments can be generally divided into staff (subordinated directly to the Governor of the NBP), administrative and technical servicing of the bank (subordinated to the Vice-Governor) and operational (subordinated to both Prime Deputy and Vice-Governor). These departments carry out the different roles of the central bank in the economy. The General Inspectorate of Banking Supervision is a speciﬁc agency with the rank of a department subordinated directly to the Governor of the NBP. The issue of combining central banking and banking supervision in the same institution is further described on pp. 000–000. Regional branches support the centre by implementing and co-ordinating the central bank tasks at a local level. The reduction in the administrative division of Poland from forty-nine to sixteen provinces introduced in

CBI in Poland 203 1999 will lead to a ﬁnal structure with one regional branch in each province to be accomplished in 2003. In 2001 the NBP employed about 6300 personnel. The structure and employment of the NBP is rather modest compared to many other European countries. Major changes are expected to follow the introduction of electronic techniques and ﬁnancial instruments that might reduce the need for such a dense network of branches and high number of employees. Current issues and the unfolding future Questions concerning the activities of the National Bank of Poland One of the key points concerning changes in the organisational structure of the NBP is whether to leave banking supervision in the hands of the central bank, or to move this activity to a separate institution. At present, the supervision of banks is carried out by the General Inspectorate of Banking Supervision (GINB) within the head ofﬁce of the NBP, and by thirteen supervision departments in the regional branches of the NBP. Despite the organisational link of GINB with NBP, supervisory policy is formulated by a special Commission for Banking Supervision, the status of which is speciﬁc. The Commission is organisationally part of the NBP, but is functionally independent. Similarly, the GINB executes its own decisions. The GINB operates in the head ofﬁces of commercial banks, but also undertakes some local branch controls and co-ordinates the work of the regional supervision departments which are responsible for supervision of the branches of the commercial banks in their district. Until 1995 regional supervision departments were mainly responsible for co-operative banks, but the newly created regional co-operative banks have now taken over the majority of these tasks. Although only about 4.9 per cent of the total assets of the banking sector belong to the co-operative bank sector, its ﬁnancial situation was, and to some extent still is, very unstable, since it consists of more than 600 separate banks. To cope with that number of separate banks, the NBP currently employs more than 300 inspectors. In June 1995, a group of Members of Parliament submitted to the Diet a bill proposing the creation of State Banking Supervision (PNB).2 The PNB would be a central organ of state administration in the ﬁeld of banking control and supervision, subordinated directly to the Prime Minister. As their main reasons for the change, the proposers cite: 1 2 3

poor performance of the NBP as banking supervisor, conﬂicts of roles and interests inside the central bank as monetary policy-maker, the owner of several banks and supervisor of banks, examples from abroad of efﬁciently functioning banking supervision agencies separated from the central bank.

204

Robert Huterski et al.

Despite this, there is no clear evidence that systems with supervision outside the central bank perform better than those with the central bank carrying on that function. This might explain why it is more common to ﬁnd responsibility for supervision performed individually by the central bank (62.3 per cent from 114 countries surveyed by the IMF) or with the co-operation of other institutions (18.4 per cent of countries), than it is to ﬁnd supervision performed by an independent agent. Apart from some peripheral arguments against creating a separate institution of banking supervision such as the relatively high operating costs, possible loss of well-qualiﬁed staff or temporary decrease in supervision efﬁciency and institutional authority, there are also other, more substantive arguments for entrusting supervision with the central bank which are particularly relevant to the Polish situation. In transition economies the role of the central bank and other supervisory bodies of banking are not well understood either by politicians or by society in general. The central bank provides the banking supervision agency with a kind of protection against pressures from commercial banks and politicians to soften prudential requirements. The short-term interest of the banks in softening of supervision is obvious. It allows them to extend their credit activity and increase their operating income. The government is still the owner of several important Polish banks and maintains control over many others (the Treasury has about 25 per cent share in the assets of the total banking system). Until they are fully privatised, a moral hazard problem exists because government and commercial banks have a common interest, since a larger budgetary income from banks depends on bank proﬁtability. Also of great importance is the question of central bank efﬁciency in performing its basic tasks. Of course, banking supervision should not be a tool of monetary policy. However, the institutional link between those two areas allows the central bank to be better informed about the condition and problems of the banking system and to ﬁnd solutions that are more effective and less damaging for monetary policy in the case of a banking crisis. It also protects the central bank against attempts by commercial banks or the government to alter its monetary policy via control over banking supervision and its prudential requirements. Regarding the institutional location of Polish banking supervision, the International Monetary Fund3 supports the view that although in principle location is not as important as co-ordination between the agencies involved, in this particular case removal of the supervisory function from the central bank could do more harm than good. Recommendations given by the IMF to Hungary are similar, although in Hungary the removal supervisory role has already taken place. The idea of entrusting the central bank with the role of institutional guardian of an independent banking supervision agency seems to be a universal prescription for CEE countries. This mainly reﬂects their weak banking systems and the strong position in banking of their governments.

CBI in Poland 205 As a result of political compromise at the beginning of 1998 (Ustawa z dnia 29 sierpnia 1997), the new collegial body, called the Commission for Banking Supervision (CBS), took over formulation of banking supervision strategy from the Governor of the NBP. The decisions and tasks set by the CBS are executed and co-ordinated by the existing General Inspectorate of Banking Supervision (GINB) which is organisationally separate from, but still remaining within the structure of, the NBP. The main tasks of the Commission for Banking Supervision are in particular: 1 2 3

4

setting rules of activity for the banks so as to guarantee the safety of ﬁnancial resources collected from clients by these banks, supervision and observance of the law and of the ﬁnancial norms binding the banks, periodic assessment of the economic standing of the banks and presentation of this to the Monetary Policy Council, as well as the assessment of the inﬂuence of monetary, ﬁscal and supervisory policies on the banks, formulating opinions about the organisation of banking supervision and setting procedures for their realisation.

The Commission for Banking Supervision consists of: the Chairman (the Governor of the NBP); the Vice-Chairman (the Minister of Finance or delegated Secretary or Under-Secretary of State in the Ministry of Finance); a representative of the President of Poland; the President of the Board of the Banking Deposits Guarantee Fund; the President of the Commission of Securities and Stock Markets or his deputy; a representative of the Ministry of Finance and the General Inspector of Banking Supervision. A representative of the Union of Polish Banks can participate in the meetings of the Commission, with an advisory voice, but only when the regulation of banking supervision is on the agenda or the safety of ﬁnancial resources entrusted to the banks is being discussed. The CBS can empower the Chairman to undertake certain decisions between its meetings with the exception of: granting a licence for a new bank; suspending a bank’s activity and petitioning for its bankruptcy; decisions about penalties and ﬁnes. The Governor of the NBP informs the CBS of any decisions taken by the NBP during the next meeting of the CBS. This solution should be judged negatively in terms of central bank independence because it makes co-ordination of banking supervision with monetary policy and other central bank activities more difﬁcult. The inﬂuence of the government on banking supervision via its representatives in the Commission makes supervision vulnerable to political inﬂuence, especially when the government performs an ownership role over a large part of the banking sector. Another vital issue relating to Polish central bank activity is its role as

206

Robert Huterski et al.

owner of commercial banks. As previously noted, such a situation can negatively affect the performance of the central bank in almost all aspects of its role. It is inevitable that, even if the central bank does not misuse any privileges of such a situation, it creates a degree of uncertainty about the future and destroys the authority of that institution. Until the end of 1996, the NBP was the owner of three banks under recovery (Prosper Bank SA, Pierwszy Komercyjny Bank SA and BudBank SA) and the Polish Investment Bank (the PBI). Objections to this dual role were raised several times in 1995 during arguments between the NBP and the Ministry of Finance. The NBP claimed that it was ready to sell off the banks if a suitable buyer could be found. In 1995 there was a very good opportunity to solve the problem when the Banking Guarantee Fund (BFG – deposits guarantee institution) was created. In one of the bills, the BFG was to be granted the right to take over failing banks and bring about their recovery. According to the Act in force, the BFG has very limited possibilities to carry out the recovery of banks and is not allowed to become their owner. Finally, the Polish Investment Bank was sold in a package4 together with the recovering Prosper Bank SA to private Kredyt Bank SA on 3 April 1997, and the NBP sold the last small bank under recovery (BudBank SA) in 2000. Under current regulations the NBP is not allowed to own any other bank. The National Bank of Poland in the New Constitution of the Republic of Poland and new Acts of Parliament The former Constitution of the Republic of Poland was based on regulations established under the old Constitution of the Polish People’s Republic from 1952. The original Act was supplemented by the Constitutional Act of Parliament of 1992, known as the ‘Small Constitution’, which regulated mutual relations between legislative and executive powers as well as local governments. Neither Act included any regulation concerning the central bank. Until the end of 1997 the position and functioning of the NBP were regulated by the National Bank of Poland Act passed by Parliament in 19895 (together with the Banking Law Act). The Constitution is not, of course, the place to regulate in detail central banking issues, and some countries do not even mention it in their Acts at all. However, the majority of European constitutions were written when central banking was not considered such an important aspect of the national ﬁnancial system as it is now. Giving constitutional support to the regulation of the Polish central bank is important to overcome a historically rooted consciousness which still exists from when the NBP was functionally a part of the Ministry of Finance. This experience co-exists with the present lack of understanding and acceptance by a large part of society and by many politicians of the role of the modern central bank.

CBI in Poland 207 At the time of writing, the Constitution voted by the National Assembly on 2 April 1997 has already come into force. A translation of the sole article of the Constitution dedicated to central banking matters is given below.6 ARTICLE 227 OF THE CONSTITUTION

1

2

3 4

5

6

7

The central bank of the state is the National Bank of Poland. It has the exclusive right to issue currency and to determine and conduct monetary policy. The National Bank of Poland is responsible for the value of the Polish currency. The organs of the National Bank of Poland are: the Governor of the National Bank of Poland, the Monetary Policy Council and the Board of the National Bank of Poland. The Governor of the National Bank of Poland is appointed by the Diet, on the proposal of the President, for a six-year term. The Governor of the National Bank of Poland must not be a member of a political party or of a trade union, or engage in similar public activity irreconcilable with the dignity of the ofﬁce. The Monetary Policy Council consists of the Governor of the National Bank of Poland as the Chairman and of persons with distinguished knowledge in the ﬁeld of ﬁnance appointed for six-year terms in equal number by the President, the Diet and the Senate. The Monetary Policy Council determines annually the foundations of monetary policy and submits them to the Diet for information purposes simultaneously with the submission of the Budget Bill proposed by the Council of Ministers. The Monetary Policy Council, within ﬁve months after the end of the budgetary year, submits to the Diet a report about the accomplishment of the foundations of monetary policy. The organisation and rules of the functioning of the National Bank of Poland, as well as detailed rules of appointment and recalling of its organs, are deﬁned by a separate Act of Parliament.

Such regulation is very valuable in establishing the formal, legal independence of the NBP. Although it deﬁnes the main objective of the central bank’s function in a vague way (strengthening of the Polish currency), it eliminates the co-objective, given in the 1989 NBP Act, which underlined co-operation in the realisation of national economic policy. In practice the co-objective was often understood as an unconditional obligation for the NBP to co-operate with the government and, in particular, with the Ministry of Finance. Regulations about the structure and appointment of the governing organs of the NBP, as well as the six-year term of the Governor of the NBP and members of the Monetary Policy Council ofﬁce, give a solid

208

Robert Huterski et al.

foundation for one of the most important aspects of central bank independence. Empirical research in CEE countries suggests that stability of the governing body of the central bank is very strongly correlated to the level of inﬂation. The new Constitution also removes the necessity for the NBP to submit annual proposals on the foundations of monetary policy to be voted on by the Diet. This was criticised as useless because of the lack of professionalism in changes introduced by Members of Parliament and the lack of instruments to execute its accomplishment. Under the new regulations the foundations would be submitted by the Monetary Policy Council for information purposes only. The Constitution provides for detailed regulation of the NBP’s organisation and functioning in a separate Act of Parliament. The appropriate Act was passed by Parliament on 29 August 1997 and came into force at the beginning of 1998 (Ustawa z dnia 29 sierpnia 1997). Except for the introduction of the Commission for Banking Supervision, which was not provided for by the Constitution, the main innovation of this Act is the detailed regulation of the organisation and tasks of the Monetary Policy Council. The Council consists of the Chairman and nine members appointed in equal number by the President of the Republic of Poland, the Diet and the Senate (each institution appoints three members). The term of ofﬁce for members is six years and they can be dismissed only for the enumerated reasons not related to monetary policy such as resignation, illness, sentence for crime, or parallel political or trade union activity of the member. The meetings of the Council are organised by its Chairman and take place at least once a month. A representative of the Council of Ministers can participate in the meetings without the right to vote, but with the right to submit proposals. Decisions of the Council are undertaken by normal majority in the presence of no less than ﬁve members, including the Chairman. If votes are evenly split, the view of the Chairman is decisive. The opinions of the members of the Council are published no less than six weeks after the meeting, but no later than three months of the meeting. Acting according to the foundations of monetary policy, the Monetary Policy Council in particular: 1 2 3 4 5 6

sets the interest rates of the NBP, sets the rules and rates of the obligatory reserve for the banks, sets maximum limits of the NBP’s liabilities from foreign credits (from the banks and ﬁnancial institutions), approves the ﬁnancial plan of the NBP and the report of the NBP’s activity, approves the annual ﬁnancial report of the NBP, sets the rules of open market operations.

The Monetary Policy Council also evaluates the activity of the Board of

CBI in Poland 209 Management of the NBP in terms of implementation of monetary policy and votes on the rules of accounting for the NBP proposed by the Governor. The Monetary Policy Council approves the exchange rate policy and the rules setting the exchange rate for the zloty against foreign currencies determined by the Council of Ministers and implemented by the NBP. The proposed organisation and set of monetary policy powers of the Monetary Policy Council should be appropriate to ensure that monetary policy is pursued by a body independent from government. However, the role of the Monetary Policy Council seems to reach beyond the aspects of monetary policy itself, and this body has a signiﬁcant inﬂuence on the current management of the NBP. Managing the National Bank of Poland There are two main approaches to managing central banks. It may be done either by assigning the most important powers to one person, usually the Governor of the bank, or by establishing a special collective organ, in which case the Governor is only an executor of this organ’s decisions. In both cases, to ensure a real possibility to govern, the method of appointment and recalling, as well as the stability of the term of ofﬁce, must be guaranteed both legally and in practice. Appropriate regulations in Poland underline the role of the NBP’s Governor in governing the central bank and give a legal framework to ensure the stability of the post. According to the NBP Act (1997) currently in force, the activity of the NBP is governed by the Governor of the NBP appointed by the Diet on the proposal of the President of the Republic of Poland, for six years. The right to recall from the post also belongs to the Diet. The same person cannot govern the NBP for longer than two consecutive terms (twelve years). The Governor of the NBP can only be recalled before the end of the term in the case of renouncement of duty, inability to work due to sustained illness lasting at least three months, valid conviction for crime or sentence of the Tribunal of State banning him from high ranking posts. The resolution of the Diet about holding the Governor of the NBP to account, according to the Constitution, brings about the suspension of duties. The collective organ supporting management, the Board, consists of the Governor of the NBP as its Chairman, two Vice-Governors and six other members. Vice-Governors of the NBP and members of the Board are appointed and recalled by the President of the Republic of Poland on the recommendation of the Governor of the NBP. In case of absence of the Governor of the NBP, his or her function is performed by the First Deputy. Vice-Governors supervise and direct the parts of activity assigned to them by the Governor. The Board examines the main issues relating to NBP activity and undertakes the resolutions necessary to complete the NBP’s tasks.

210

Robert Huterski et al.

However, the Governor is the agent who issues commonly binding regulations in the form of dispositions, instructions, notices and so on published in the Polish Monitor. The Governor of the NBP also publishes the Ofﬁcial Gazette of the NBP, which includes his dispositions concerning exclusively the activities of banks, the balance sheets of the NBP and other banks as well as notices about the creation, liquidation or bankruptcy of a bank. The Governor of the NBP is the most senior of all staff in the NBP, and he or she nominates other staff to posts at his or her disposal. He or she can also dismiss them from such posts. The Governor can represent the NBP in property matters individually or can give power of attorney to some other person. Overall, the above regulations have weakened the position of the Governor of the NBP. The former set of the Governor’s prerogatives was criticised by some ofﬁcials because, in their opinion, the balance between effectiveness of governing and accountability was uneven. As with Marshall Pilsudzki in the 1930s, the Governor of the NBP during his term seemed to be ‘responsible only to God and history’. The NBP Act (1989) validates this opinion to some extent since it effectively placed a governing institution in the hands of one person, and this institution was responsible for decisions about the formation and implementation of monetary policy which has economy-wide implications. Of course, the Governor had members of the Board, Economic and Scientiﬁc Councils and staff to assist him in his work, but it was the Governor who decided whom to consult. In any case, from the legal perspective, whoever’s thoughts were behind the formation of monetary policy, its author was the Governor of the NBP. According to the new Act of Parliament, which came into force at the beginning of 1998 (Ustawa z dnia 29 sierpnia 1997), the position of the Governor of the NBP is weaker because many of the Governor’s powers are transferred to the newly created collegial bodies responsible for banking supervision and monetary policy. The Governor still maintains an important position because he or she is Chairman of the Board of Management, the Commission for Banking Supervision, the Monetary Policy Council and also represents the NBP in public. The Governor can still participate in the sessions of the Diet and Council of Ministers, and present to these organs proposals on monetary policy foundation and so on. However, the Governor’s position is now effectively primus inter pares because he or she has to represent the views of the collegial bodies rather than his or her own opinions. The decisions about the internal ﬁnance and accounting of the central bank and evaluation of the performance of the Board of Management in terms of monetary policy implementation are now also excluded from the Governor’s powers. These are now entrusted to the Monetary Policy Council. The set of unchanged powers relates purely to the managerial functions of the Governor, and cover areas such as the role of the employer.

CBI in Poland 211 It is too early to give a full assessment of the real position of the Governor of the NBP because his or her actual role in the collegial bodies responsible for strategic decisions will be noticeable only after some time of their operation. An assessment of the level of independence of the National Bank of Poland The National Bank of Poland within the structure of state institutions One of the more important aspects of a central bank’s independence is its position within the structure of state institutions. In Poland this refers to the relations of the NBP with the Diet (the Senate plays a rather passive role controlling the Diet’s decisions), the President of Poland and the government. As in many other countries, the most important government agency in this case is the Ministry of Finance. Other state institutions have a role that impinges signiﬁcantly on the central bank. As at mid-1999, the Polish central bank occupied a strong position in the structure of state institutions. Although the President of the Republic of Poland and the Diet have an inﬂuence on appointing the Governor of the NBP, they have no instruments to control the current activity of the bank. A six-year term of ofﬁce and the limited possibilities to recall the Governor of the NBP from the post are formal guarantees of the stability and independence of the central bank from political cycles. In addition, the Ministry of Finance cannot directly affect the functioning of the NBP. The Governor of the NBP is not a member of the government, but has the right to participate in government meetings as well as in parliamentary sessions. In forming and conducting monetary policy, the NBP has to present its foundations to the Ministry of Finance for its opinion and to the Diet for informational purposes. However, in practice there is no mechanism to force the NBP to follow either the will of Parliament or the Ministry of Finance. The NBP sets basic interest rates and exchange rates, and the extent of its independence in this area has been a cause of conﬂict with the Ministry of Finance. The main point of conﬂict was whether or not the NBP can change the rates without acceptance from the Ministry of Finance. Existing legal regulations are imprecise in this matter, and during the presidential campaign in 1995 (the Governor of the NBP, Mrs Hanna Gronkiewicz-Waltz, was one of the presidential candidates) it seemed to be a real problem. At the same time there was also a dispute concerning the revaluation of the foreign currency reserves of the NBP as a result of the zloty’s appreciation on the foreign exchanges. The problem arose because the exchange rate shift resulted in foreign currency being sold to the commercial banks

212

Robert Huterski et al.

at a higher price than was paid for it by the NBP. The Council of Ministers accused the NBP of reducing its own income, most of which goes to the budget, by excessive borrowing from the banks in order to reduce the money supply. To balance the budget, the Council of Ministers proposed to treat the revaluation reserves as a part of the budget income. So far, the change has not been introduced and this testiﬁes to the real strength of the NBP in its dealings with other authorities. The last two years of model cooperation between both ﬁnancial authorities shows that agreement is possible without legally binding rules. The Chief Board of Supervision was also unable to force the NBP to reveal all its data and certain information about commercial banks (at the end of 1995 and the beginning of 1996). Measuring the National Bank of Poland’s independence There are many ways of calculating a central bank’s independence. Wellknown indices are those devised by Bade and Parkin (1988), Alesina (1988, 1989), Grilli et al. (1991) and Eijfﬁnger and Schaling (1992, 1993). Despite differences in their construction, these indices were often recalculated to become comparable. This was done by Havrilesky and Granato (1993) and Eijifﬁnger et al. (1996) among others. A common problem with these measures when applied to the reality of former communist countries is that they were constructed for stable democratic, market-oriented economies and do not reﬂect some of the speciﬁc features of the economies undergoing transition. For example, in many indices the fact that the central bank performs banking supervision is treated as negative and its effect is to lower the overall independence index. In the case of countries such as Poland, inclusion of the banking supervision agency into the central bank was the only way to give real power to this agency and to protect it, and the central bank, from political attempts to use banking supervision to weaken central bank inﬂuence on the banking sector and on monetary policy. Another problem with using these indices in some transition economies concerns the reliability of the index of legal independence. In Western countries the index of legal independence is usually reliable because rules written into legal Acts are obeyed in practice. Thus it is sufﬁcient to measure reality through its legal background. However, in countries where the legal framework is copied from more developed countries, but where the political and economic agents are unwilling or unable to respect it (for example, in Bulgaria), indices of legal independence describe more what is desirable than what actually exists. Accordingly, for economies in transition a better measure of central bank independence appears to be the frequency of changes in the post of Governor of the central bank. This problem does not arise in the Polish case because the legal constitution of central banking is fully introduced and adhered to. The ability of the NBP to protect its position and to force literal application of rules is

CBI in Poland 213 not only proof of its strength, but also of its political will to ensure the application of rules. A more serious problem in the case of Poland concerns the use of an index of stability of the Governor’s post since there was no term of ofﬁce in Poland before 1992. The question is whether periods should be counted up to 1992 and then from 1993, or whether to calculate the index with periods chosen independently. Indices based on questionnaires depend on the quality of the sample of respondents chosen by the researcher and are biased by their subjectivity of evaluation. Nevertheless, they do provide additional information worth analysing. However, because it is a time-consuming method of gathering information, for the purpose of this chapter some values are adjusted for 1997 to account for what, in the view of the authors, values would have been recorded by accurate responses. Table 8.3 presents an attempt to calculate indices of central bank independence for the National Bank of Poland in 1999 when all new regulations came into force. For comparison, 1997 was the last year under the former legal framework and 1998 was the year when the main set of changes was introduced. The methodology followed is that of Cukierman et al. (1992). The ﬁrst attempt made in 1994 by Strzelec (1994) (see Table 8.4 for details) contain some methodological errors in the legal index. For example, the index contained an additional numerical coding added by the author, and one-step instead of two-step aggregation which made the results incomparable with those obtained by Cukierman et al. (1992). The questionnaire index follows Strzelec (1994) without any changes. However, there is some uncertainty about whether some of the values differ from the original numerical coding as a result of calculating the mean from several questionnaires, or whether such differences have some other explanation peculiar to the way in which they were originally calculated. Fortunately, the weight of those doubtful values in the ﬁnal result is insigniﬁcant. The detailed description of variables for each particular year is included in Appendix A on pp. 218–220. Numerical coding and the index values closer to 1.00 resemble greater central bank independence in every particular respect. Summarising, indices calculated for Poland using the methodology of Cukierman et al. (1992) show a high level of NBP independence. Recalculation for the circumstances of 1998 and 1999 yields even better results. Table 8.5 presents them together with values for some countries chosen from the original paper by Cukierman et al. (1992). Less signiﬁcant, but still of interest, are the indices of Bade and Parkin (1988) and Grilli et al. (1991) which may be found in Appendix B on pp. 221–223. From a formal perspective, all possibilities to increase the NBP’s independence suggested by these indices have already been implemented into Polish legislation. This is reﬂected in the growing values of the legal

214

Robert Huterski et al.

Table 8.4 Variables, weights and numerical coding in the questionnaire used by A. Strzelec to describe the independence of the NBPa No.

Description of variable

Weight

Numerical coding

1

The term of ofﬁce of the Governor overlaps only slightly the terms of political powers. Modest restrictions in lending to the government. Several examples of conﬂict with the government solved in favour of the NBP. Financial independence: (a) In principle, the NBP decides about its budget alone or together with other institutions.b (b) Remuneration of high-ranking ofﬁcials of the NBP and use of proﬁt is deﬁned by law or the bank decides itself. Intermediate targets of monetary policy: (a) Quantitative targets are not perfect approximation of future money quantity. (b) Interest rate is used neither as a formal nor an informal target. Price stability is an objective of NBP activity, but not the sole one.c The NBP used to grant loans, also at subsidised interest rates, but only up until 1993. At present the NBP does not carry out such an activity. The NBP independence index based on a questionnaire (1994):

0.10

1.00

0.20 0.10

0.66 1.00

2 3 4

5

6 7

0.10 0.66 1.00 0.15 0.66 1.00 0.15

0.58

0.20

1.00 0.83

Source: Based on Cukierman et al. (1992). Notes a Calculated by computing mean numerical coding from answers received in several different interviews. b In practice, the NBP decides about its budget. Numerous funds ﬁnanced from proﬁt deductions, according to the NBP Act prescriptions, are a guarantee of budget revenue stability (besides which the government has an interest in the high proﬁtability of the NBP as NBP proﬁts are an important source of the state budget revenue). The expenses side is always only an estimation due to, for example, unpredictable costs of open market interventions. The dispute over the so-called revaluation fund in 1995 and other unsuccessful attempts of the government to limit the ﬁnancial independence of the NBP suggest a probable change of the numerical coding in point 4(a) to 1.00. c Price stability, in the light of NBP activity during the past two years, is the NBP’s leading objective causing criticism of its methods or of the hard line it has followed, but never of its preference for other objectives. It allows a higher numerical coding equal to 1.00 to be used for point 6. The above two changes would probably increase the questionnaire index for 1997 to 0.91.

index over time, reaching almost the maximum possible value for 1999. However, these indices do not take into account many speciﬁc features of transition economies such as the government’s role as an owner of ﬁnancial institutions and much larger political vulnerability. Consequently the results may be less reliable than expected.

CBI in Poland 215 Table 8.5 Indices of Cukierman, Webb and Neyapti (1992) – comparison with other countries

Poland 1994 Poland 1997 Poland 1998 Poland 1999 Poland 2002 Germany 1992 United Kingdom 1992 United States 1992 Japan 1992

Index of legal independence

Turnover rates of governors

Questionnaire index of independence

0.72 0.75 0.86 0.95 0.95 0.69 0.27

0.8a 0.2 0.2 0.2 0.4 0.1 0.1

0.83 0.91b – – – 1.00 0.64

0.48 0.18

0.1 0.3

n.a.c n.a.c

Notes a Update by the authors for full ﬁve-year period. b Estimated update by the authors. c n.a. – not available according to Cukierman et al. (1992).

The existing indices are not ﬂexible enough to include these speciﬁc features just by wider interpretation of questions because it would destroy their comparability. The best solution, therefore, is to develop a new index of central bank independence for transition economies. However, this is beyond the scope of this chapter.

4 Conclusions This chapter has examined the role of central banking, and in particular the issue of central bank independence, within the Polish economy. It has discussed the role and the position of the National Bank of Poland both before and since 1989. An attempt has been made to calculate indices of independence for the NBP. Issues of current interest and their implications for the future level of NBP independence are also discussed. The typical monetary situation found in centrally planned economies, such as existed in Poland prior to 1989, is described. The role of money and its circulation in the classical socialist planned economy differed fundamentally from that in market economies. The crucial difference is the subordinate role that money played in the centrally planned economy since expenditure patterns had little direct inﬂuence on the allocation of resources. An important feature of monetary circulation was its disaggregated nature. Monetary ﬂows in the sphere of consumption were independent from those in the sphere of production and from ﬂows between enterprises and the state budget. The function of money as a medium of exchange was

216

Robert Huterski et al.

very restricted and its circulation did not integrate the whole national economy as it does in market economies. Instead, the centrally planned economy was integrated by a system of planning and bureaucratic co-ordination. However, the role assigned to money and its disaggregated circulation in the centrally planned economy had a great impact on the role and structure of the banking system. It was highly dependent on the planning commission and the branch ministries. The monobank concept, although not fully realised in practice, assumed that the banking system would take the form of one large national bank providing services to all economic units. The banking system played practically no role in the determination of economic policy. Since 1982 there have been various efforts to increase the degree of monetary control in Poland. The main reason for this was the desire to lower inﬂation. The methods used in the attempt to control inﬂation were ﬁscal policy, income/price policy and monetary policy. Despite this, inﬂation accelerated throughout the 1980s and in 1989 inﬂation remained high. It became increasingly apparent that attempts to control the money supply were incompatible with the regulations which governed the workings of the monetary system. The IMF-type adjustment programme launched in 1990, commonly known as the Balcerowicz Plan, was aimed at making Poland an open market economy. It was crucial that money should become a scarce good. The policies of the NBP had to undergo fundamental change. Since 1990 the NBP has followed a clear anti-inﬂationary policy. However, the actual instruments used have evolved during this period. Such alterations to monetary policy are largely a reﬂection of the changing nature of the ﬁnancial system and of the emergence of new economic problems. From 1990 onward money growth was the intermediate target of the NBP. Important policy instruments have been positive real interest rates and the exchange rate. Nevertheless, the NBP has shown a marked willingness to change its tactics in response to the situation before it. It is important to note that, despite problems in monetary control, inﬂation in Poland declined substantially throughout the 1990s. This is strongly connected with the assertion of independence by the NBP. This paper has attempted to assess the level of independence of the NBP. The Polish central bank occupies a strong position with regard to state institutions. Although the President of the Republic of Poland and the Diet have an inﬂuence on appointing the Governor of the NBP, they have no instruments to control the current activity of the bank. A six-year term of ofﬁce and the limited possibilities to recall the Governor from post are formal guarantees of the stability and independence of the NBP from political cycles. A variety of current issues connected with the NBP and its independence have also been discussed. These include: whether the central bank

CBI in Poland 217 should incorporate the banking supervision function; the constitutional position of the central bank; and the degree of power held by the Governor of the NBP. The authors attach importance to the constitutional support given to the central bank in Article 227 of the Constitution approved by the nation in the referendum of May 1997. They believe this will assist in promoting increased understanding and acceptance by both society and politicians of the role of the modern central bank. Given the recent past when the NBP was functionally a part of the Ministry of Finance, the new Constitution is most valuable in establishing and upholding the formal, legal independence of the NBP. The new Act of Parliament introduced from 1998 two new collegial bodies that took over many of the Governor’s powers related to monetary policy and banking supervision. This is inevitably a much more democratic and accountable way for the central bank to make vital decisions. However, questions arise about how far the Monetary Policy Council will become involved in the current management of the NBP and what inﬂuence will the Commission for Banking Supervision have on central bank independence and performance. The new regulations leave the NBP still responsible for microeconomic supervision with all the accompanying burdens associated with this. It also creates new possible ﬁelds of conﬂict concerning the co-ordination of monetary and supervisory policies. As long as the government remains an inﬂuential owner of commercial banks and has an impact on the policy of banking supervision, this particular solution of the new legislation may be used to stiﬂe the development of central bank independence. The actual measurement of a central bank’s independence using established indices can be problematic in the case of former communist countries. For example, many indices treat a central bank performing banking supervision as negative, but this may be inappropriate. Likewise, indices of legal independence need to be applied with caution in some new democracies where the legal framework is not necessarily followed in practice – though this is not the case in Poland. This chapter develops an attempt made in 1994 by A. Strzelec to calculate indices of central bank independence for the NBP. The methodology followed was that of Cukierman et al. (1992). Our ﬁndings suggest that all index values of NBP independence are extremely high and are usually close to the maximum. This is because all possibilities to increase the NBP’s independence suggested by these indices have already been incorporated into Polish legislation. However, these indices do not include many speciﬁc features of transition economies and therefore the results cannot be accepted without question. The new or adjusted measure of the central bank independence appropriate for countries such as Poland still awaits development.

218

Robert Huterski et al.

Appendix A Legal independence of the National Bank of Poland – description of variables Table 8A.1 Index of legal independence (1997) No.

Description of variable

Weight

1

The Governor: (a) six-year term of ofﬁce (b) appointed by the legislative power (c) recalling possible only for reasons not related to policy conducted (d) holding another government ofﬁce forbidden The procedure of elaborating monetary policy: (a) the NBP co-operates and has visible inﬂuence on creating monetary policy (b) the legislative power opinion (the Diet) has decisive impact in case of conﬂict concerning monetary policy (c) the NBP gives an opinion about draft budget act Objectives: Stability of money and prices is one of the NBP objectives apart from certain targets (e.g. stability of banking system) Restrictions in lending to the government: (a) ‘cash’ loans are forbidden (b) lending on security has rigid limitation (by the sum given in the budget act; or up to 2% of government expenses) (c) conditions (the term, interest rate, amount) of lending by the NBP are quoted in the budget act and the government, and the NBP can discuss only technical details (d) potential borrower from the bank is exclusively the central government (e) the maximum level of a central bank loan is given as a sum of money (f) the term of loans – up to one year (g) interest rate of the loan is close to the market interest rate (h) the NBP can buy and sell treasury bonds on the primary market

0.20

2

3

4

Numerical coding 0.75 0.50 0.83 1.00

0.15 (0.25)

0.75

(0.50)

0.40

(0.25)

1.00

0.15 0.60

0.15 0.10

1.00 0.67

0.10

0.67

0.05

1.00

(0.025)

1.00

(0.025) (0.025)

1.00 0.75

(0.025)

0.00

Notes Weights in brackets (0.25), (0.50) and (0.25) are used in the ﬁrst round of aggregation. The NBP legal independence index (1997): 0.75.

CBI in Poland 219 Table 8A.2 Index of legal independence (1998) No.

Description of variable

Weight

1

The Governor: (a) six-year term of ofﬁce (b) appointed by the legislative power (c) dismissal possible only for reasons not related to policy conducted (d) holding another government ofﬁce forbidden The procedure of monetary policy formulation: (a) the Council of the Monetary Policy of the NBP formulates monetary policy alone (b) the NBP has ﬁnal word in resolution of conﬂict on issues clearly deﬁned in the law as its objectives (c) the NBP gives an opinion about draft budget act Objectives: Price stability is the major objective in the charter and the NBP has the ﬁnal word in case of conﬂict with other government objectives Restrictions in lending to the government: (a) ‘cash’ loans are forbidden (b) lending on security has rigid limitation (by the sum given in the budget act; or up to 2% of government expenses) (c) conditions (the term, interest rate, amount) of lending by the NBP are quoted in the budget act and the government, and the NBP can discuss only technical details (d) potential borrower from the bank is exclusively the central government (e) the maximum level of a central bank loan is given as a sum of money (f) the term of loans – up to one year (g) interest rates on loans is above minimum rates (h) the NBP can buy and sell Treasury bonds on the primary market

0.20

2

3

4

Note The NBP legal independence index (1998): 0.86.

Numerical coding 0.75 0.50 0.83 1.00

0.15 (0.25)

1.00

(0.50)

1.00

(0.25)

1.00

0.15 1.00

0.15 0.10

1.00 0.67

0.10

0.67

0.05

1.00

(0.025)

1.00

(0.025) (0.025)

1.00 1.00

(0.025)

0.00

220

Robert Huterski et al.

Table 8A.3 Index of legal independence (1999) No.

Description of variable

Weight

1

The Governor: (a) six-year term of ofﬁce (b) appointed by the legislative power (c) dismissal possible only for reasons not related to policy conducted (d) holding another government ofﬁce forbidden The procedure of monetary policy formulation: (a) the Council of the Monetary Policy of the NBP formulates monetary policy alone (b) the NBP has ﬁnal word in resolution of conﬂict, on issues clearly deﬁned in the law as its objectives (c) the NBP gives an opinion about draft budget act Objectives: Price stability is the major objective in the charter and the NBP has the ﬁnal word in case of conﬂict with other government objectives Limitations on lending to the government:a (a) no advances permitted (b) securitised lending not permitted (c) terms of lending controlled by the bank (d) potential borrower from the bank is exclusively the central government (e) the maximum level of a central bank loan is given as a currency amount (f) within six months (g) interest rate on loans must be above minimum rates (h) the NBP prohibited from buying or selling government securities in the primary market

0.20

2

3

4

Numerical coding 0.75 0.50 0.83 1.00

0.15 (0.25)

1.00

(0.50)

1.00

(0.25)

1.00

0.15 1.00

0.15 0.10 0.10 0.05

1.00 1.00 1.00 1.00

(0.025)

1.00

(0.025) (0.025)

1.00 1.00

(0.025)

1.00

Notes Weights in brackets (0.25), (0.50) and (0.25) are used in the ﬁrst round of aggregation. a Since the beginning of 1999 all forms of central bank credit facility to the government have been forbidden and, as a result, the ﬁrst four questions from the index of economic independence become irrelevant. Positive answers are given to comply with the internal logic of this index although there should be no answers at all according to the formal construction of these questions.

CBI in Poland 221

Appendix B 1 Index of central bank independence for Poland according to Bade and Parkin (1988) This is the ﬁrst and the simplest index of central bank independence which was the direct base for indices of Alesina (1988, 1989) and Eijfﬁnger and Schaling (1993). The policy type in all these indices corresponds with index values calculated by adding the number of asterisks plus one. Therefore a bank with all negative answers receives an index value of 1. All those indices use their own codes for potential central bank type and policy type as combinations of answers for all their questions. If some types are described as non-existent it only means that in the sample of countries included in research there was no example of such a combination of features. Generally, most indices relate only to twelve industrial countries. Central bank-type coding only allows potential readers to ﬁnd the Polish central bank on the list of types of central banks in appropriate papers of quoted authors and does not represent any valuation.

The NBP legal independence index (1998): 0.95.

Questions

1997

1998

1999

1 Is the bank the ﬁnal policy authority? 2 Is there no government ofﬁcial (with or without voting power) on the Board? 3 Are more than half of the Board appointments made independent of government? Bade and Parkin central bank type: Bade and Parkin policy type:

* *

* *

* *

*

*

*

(h) 4 (***)

(h) 4 (***)

(h) 4 (***)

Note: Identical results for 1997, 1998 and 1999, despite the fact that from 1998 a government representative can participate as an observer without voting power in meetings of the Monetary Policy Council. A positive answer to question 2 is chosen because the government representative is not a member of the Council. The Governor of the Polish central bank can participate on the same basis, as an observer without voting power, in government meetings, and this solution is not regarded as a limitation of government independence from the central bank. The aim of this solution is to improve mutual exchange of information and it cannot be used as an instrument of intervention in the policy of the other institution. An obligation to publish current reports from meetings of the central bank Board (as happens in some other countries) would give similar results in terms of actual relations between central bank and government but would not be reﬂected by this index. In the case of the opposite interpretation of question 2 it gives for 1998 and 1999, according to Bade and Parkin, potential bank type (c) and non-existent policy type (*-*).

222

Robert Huterski et al.

2 Index of central bank independence for Poland according to Grilli et al. (1991) This index presents a new approach compared with those described above. It divides central bank independence into political and economic, thus giving many more questions to answer. Question 7 about economic independence allows three options of answer as to where supervisory powers are located. If the central bank has nothing in common with banking supervision, the answer receives double points (two stars). It raises the weight of that question compared to the rest and makes the maximum overall index of economic independence equal 8 despite only seven questions being included. Question Political independence: 1 Governor not appointed by the government 2 Governor appointed for more than ﬁve years 3 All the Board not appointed by the government 4 Board appointed for more than ﬁve years 5 No mandatory participation of government representative on the Board 6 No government approval of monetary policy formulation required 7 Statutory requirements that central bank pursues monetary stability among its goals 8 Legal provisions that strengthen the central bank’s position in conﬂicts with the government are present Overall index of political independence Economic independence: 1 Direct credit facility: not automatic 2 Direct credit facility: market interest rate 3 Direct credit facility: temporary 4 Direct credit facility: limited amount 5 Central bank does not participate in primary market for public debt 6 Discount rate set by central bank 7 Banking supervision not entrusted to the central bank (**) or not entrusted to the central bank alone (*) Overall index of economic independence Total index of political and economic independence

1997

1998

1999

* *

* *

* *

*

*

*

– *

* *

* *

*

*

*

–

*

*

–

*

*

5

8

8

* * * * –

* * * * –

* (!) * * * *

* –

* *

* *

4 9

6 14

7 15

CBI in Poland 223 Since the beginning of 1999 all forms of central bank credit facility to the government have been forbidden and, as a result, the ﬁrst four questions from the index of economic independence become irrelevant. Positive answers are given to comply with the internal logic of this index, although there should be no answers at all according to the formal construction of these questions.

Notes 1 See Regulamin organizacyjny Narodowego Banku Polskiego, Disposition no. C/13/I/95 of the Governor of the National Bank of Poland, 8 September 1995, and later amendments. 2 See Poselski projekt ustawy o Pan ´ stwowym Nadzorze Bankowym z 30.06.1995 r., Diet’s Print, no. 1538. 3 Nie rozdzielac´, Gazeta Bankowa, no. 15, 14 April 1996. 4 See PBI dla Kredyt Banku, Rzeczpospolita, no. 79, 4 April 1997. 5 See Ustawa z dnia 31 stycznia 1989 r. o. Narodowym Banku Polskim, uniform text in Journal of Law, 72 (1992), position 360 and later amendments. 6 See Konstytucja Rzeczypospolitej Polskiej z dnia 2 kwietnia 1997 r., Rzeczpospolita, 4 April 1997.

References Alesina, A. (1988) Central bank independence and growth: A sensitivity analysis, Canadian Journal of Economics, 31, 303–317. Alesina, A. (1989) Politics and business cycles in industrial democracies. Economic Policy, 8, 55–98. Bade, R. and Parkin, M. (1988) Central bank laws and monetary policy, mimeo, University of Western Ontario. Bruno, M. (1991) High Inﬂation and the Nominal Anchors of an Open Economy, Princeton University Press, Princeton. Cukierman, A., Webb, S.B. and Neyapti, B. (1992) Measuring the independence of central banks and its effect on policy outcomes, World Bank Economic Review, 6, 353–398. Eijfﬁnger, S.C.W. and Schaling, E. (1992) The political economy of central bank independence, Special Papers in International Economics, 19, Princeton. Eijfﬁnger, S.C.W. and Schaling, E. (1995) Central bank independence in twelve industrial countries, Banca Nazionale del Lavoro Quarterly Review, 184 (March), 49–89. Eijfﬁnger, S.C.W., Van Rooj, M. and Schaling, E. (1996) Central bank independence: A panel data approach, Tilburg University, Center Discussion Paper, Series No. 9493. Grilli, V., Masciandaro, D. and Tabellini, G. (1991) Political and monetary institutions and public ﬁnance policies in industrial economies, Economic Policy, 13, 341–392. Strzelec, A. (1994) Niezalezᠨnos´c´ Narodowego Banku Polskiego w latach 1989–1993. Studium empiryczne na podstawie metody Cukiermana, Webba, Neyapti, Bank i Kredyt, 7. Ustawa z dnia 29 sierpnia (1997) r.o. Narodowym Banku Polskim, Journal of Law, 140, position 938.

9

Central bank independence in Romania Silviu Cerna, Liliana Donath and Bogdan Dima

1 Historical overview: central banking pre-1989/1991 The National Bank of Romania (NBR) is a public institution established 117 years ago. The Law enacted on 11 and 23 April 1880 and published in the Ofﬁcial Register on 17 and 29 April 1880 stipulated the establishment of a discount and circulation bank named the National Bank of Romania that had the ‘exclusive privilege to issue the national currency of Romania’. The Bank also had the privilege to rediscount commercial paper, Treasury bills and Lombard state bonds; and to participate in the constitution of capital of certain institutions together with the state. Throughout its entire existence, though its economic role, the regime of its ownership and institutional dimensions has changed, sometimes radically, the NBR remained the sole institution authorised to issue the national currency of Romania. The traditional form of organisation of the NBR was that of a limited company whose equity was held by the state and private agents. The NBR was authorised to issue banknotes at bearer, convertible in gold or silver, and to make the speciﬁc operations of a central bank (discount, Lombard, credit granting). The banknotes issued by the NBR were guaranteed with commercial paper and resulted from merchandise transactions. When discounted, the value of banknotes issued could not exceed one-third of the bank’s gold and silver reserves. The ﬁrst banknotes in circulation were the former ‘state securities’ issued during the War of Independence (1877 to 1878), which were withdrawn and transformed into banknotes by stamping them with the seal of the NBR. Later, new banknotes were manufactured and circulated, and the older ‘state securities’ were liquidated. In general, the monetary circulation was extremely healthy at the end of the nineteenth century, partly because in 1890 the NBR adopted the gold standard and the principles of this restricted the note-issuing ability of the state. The First World War adversely affected the Romanian economy. The government took its refuge in Moldova and contracted new loans to enable the NBR to ﬁnance larger military expenditures. Therefore, the

CBI in Romania 225 newly issued banknotes of the NBR (whose Treasury was sent to Moscow to be safeguarded – without ever returning!) became entirely ﬁduciary and ceased to be convertible into gold. After the war, the monetary authorities tried to reintroduce the gold standard regime, but the problems of monetary uniﬁcation and monetary reform that Romania continued to face made this almost impossible. The reforms were completed by 1929 and it was agreed that the national currency of Romania should be named ‘leu’ (plural ‘lei’) having a legal deﬁnition in gold of 10 mg/‰ (i.e. purity of gold). Between 1929 and 1933 the gold exchange regime was instituted in a ‘mixed form’ together with the gold bullion and gold exchange. Despite the best efforts of the NBR to build on these reforms and strengthen the economy, the depression of 1929 to 1933 caused economic stagnation and the NBR issued a restrictive regulation in 1932 which effectively abolished the gold standard. A new phase in the evolution of the NBR began after the end of the Second World War. On 28 December 1946 the NBR became state-owned under Law no. 1056. A comprehensive set of monetary reforms were introduced on 15 August 1947 precipitated by the drastic fall in production and the necessity to absorb the immense monetary surplus. It was established under law that the new leu should be deﬁned by 6.6 mg gold/9‰ purity of gold without the obligation of the central bank to convert the existing banknotes in circulation into gold. In the meantime, the institutional nature and functions of the NBR were radically transformed by its inclusion in the planning mechanism of a command economy. In 1948, the NBR was renamed The Bank of The Peoples Republic of Romania, and renamed again in 1970 as The National Bank of The Socialist Republic of Romania. Under these circumstances the monetary reform launched in 1947 had a great class character being completed without foreign loans and without the leasing of economic activities. It succeeded in stopping inﬂation momentarily, conferred certain stability on the leu and led to a new distribution of national income along with a new wage/price ratio. After nationalisation of the main means of production and the banks in 1948 to 1949, which laid the foundations of the state-owned economy, economic growth increased, but not sufﬁciently to maintain the stability of the leu. During the period 1947 to 1952 inﬂation caused the leu to depreciate substantially on the foreign exchanges, and on 28 January 1952, a new series of monetary reforms were implemented. The new gold equivalent of the leu was ﬁxed at 0.079346 gr., and the exchange rate for the leu against other currencies was established according to this value. New banknotes were issued, along with notes of the Ministry of Finance and coins. After 1952, the economic, ﬁnancial and monetary situation of the country was progressively consolidated and no further monetary reforms of any signiﬁcance were implemented. A new redeﬁnition of the leu took place in 1954, but this had no substantial internal implications.

226

Silviu Cerna et al.

During the communist regime, the central bank was subordinated to the Council of Ministers and functioned through the central administration (having the tasks of co-ordination and control), county branches (with operative tasks) and subsidiaries in the larger towns. The new role of the NBR included monopoly of currency issue and gold operations; monopoly to gather state incomes and to make payments for the budget (the central bank being ‘the state’s general cashier’); foreign exchange monopoly together with the Romanian Bank of Foreign Exchange; the establishment of the ‘credit plan for the economy’; the issue of the ‘cash plan for the economy’; granting short-term loans to banks’ clients; establishment of the income and payment balance of the population and so on. The NBR therefore had considerable inﬂuence over the Romanian economy. When the communist regime was abolished, the Romanian economy entered a new phase of transition towards establishing a market economy. Extensive monetary reforms were initiated and the central bank faced a series of new problems while trying to regain its status as ‘bank of the banks’, and the monetary authority of the Romanian state.

2 Financial and banking reform after 1989/1991 The year 1989 marked the beginning of the transition of the Romanian economy from a highly centralised system to a market-based system. The transition proved to be more difﬁcult than expected and reform extended over political, economic, ﬁnancial, banking and social areas. In the political ﬁeld, Romania aimed primarily at achieving internal stability. The adoption of the new Constitution completed in 1991 was considered a priority because it promoted democracy, political pluralism and the development of a market economy. The organisation of the elections in 1990, 1992 and 1996 also underlined the desire of the people to create a market economy. In the economy, the goal of economic stabilisation was given priority. The government, elected in 1990, considered that gradual transition to a market economy would reduce the social costs of adjustment, and would maintain internal stability. Although signiﬁcant progress in macro stabilisation was achieved, the strategy chosen was not without problems. In particular, a lack of trust was apparent in the business community with respect to the ‘safe’ reforms introduced by what were regarded as Western-oriented politicians. Nevertheless, ‘shock therapy’ was not considered expedient for at least three reasons: 1

2

during the 1980s, Romania was already exposed to ‘shock therapy’ by making an immense effort to pay back its entire external debt and this severely reduced the standard of living of the population; the Romanian economy was characterised by an external rigidity,

CBI in Romania 227

3

having highly unbalanced productive capacities and a highly centralised economy; the economic environment was unfavourable for the beginning of the transition period because of the collapse of CMEA (accounting for more than 45 per cent of Romanian foreign trade) and the crises in Iraq and Yugoslavia. The effects of these events reverberated throughout the economy and resulted in substantial losses for several enterprises.

Given these circumstances the macro-stabilisation priorities were to reduce the rate of inﬂation, to end demonetisation of the economy and its dollarisation, and to improve the foreign exchange reserves of the banking system and exchange rate liberalisation. The ﬁrst of these priorities, to lower the rate of inﬂation, was achieved in 1994 when the average monthly rate of inﬂation fell from 12 per cent in 1993 to 4.3 per cent in 1994. Prior to 1994, given the hyperinﬂation, the population and the economic entities made most of their transactions through barter (which led to the demonetisations of the transactions) and, in order to preserve the purchasing power, expressed the transactions in hard currency (i.e. dollars). This process was called ‘demonetisation’ and ‘dollarisation’ of the economy. Demonetisation and dollarisation were stopped in 1994 when the growth of the money stock in the economy exceeded the growth in prices. The latter shows that the demonetisation process that started in 1990 immediately after the liberalisation of prices was completed, and the reverse process of regaining public trust in the national currency was launched. These developments were very important in restarting sustainable economic growth and the restructuring process. The growth of foreign reserves in the banking system reached $1,495 million and those of the NBR foreign exchange reserves reached $536 million. This was made possible because of improvements in the current account of the balance of payments, which recorded a deﬁcit of $428 million in 1994 compared to a deﬁcit of $1174 million in 1993. Also in 1994, beginning in April, the foreign exchange market was liberalised. This process was completed on 1 August 1994 with the creation of an interbank foreign exchange market. The completeness of the macro-stabilisation process in 1994 was possible due to the coherence of the programme, and because ﬁscal and monetary policies succeeded in controlling inﬂation. The NBR also used the interest rate lever to limit the amount of liquidity in the system. The most important interest rates for the reﬁnancing loans increased to 15 per cent per month for the auction credit (the Romanian equivalent of the openmarket operations), and to 21 per cent per month for the overdraft credit (the equivalent of the Lombard rate). The average interest rate of the NBR grew to 11 to 12 per cent per month and substantially surpassed the rate of inﬂation.

228

Silviu Cerna et al.

Macro stabilisation was also due partly to the response of the real sector of the economy. Stimulated by the real depreciation of the leu, which was equivalent to 15 per cent during the ﬁrst four months of 1994, and deregulated access to the foreign exchange market, Romanian ﬁrms were impelled towards the external market. As a result, a monthly increase in exports of $410 million was recorded in the ﬁrst quarter of 1994, rising to $590 million in the fourth quarter and to $636 million in the second quarter of 1995. However, in 1996, the economic situation deteriorated and inﬂation increased to a monthly rate of 10 per cent in December 1996, compared to 2 per cent in January 1996. The current account deﬁcit increased from $0.5 billion in 1994 to $1.7 billion in 1995 and to $2.3 billion in 1996. Consequently, foreign exchange reserves fell in 1995 and in 1996 the current account deﬁcit was ﬁnanced almost entirely by relatively shortterm loans. Debt service payments increased from an annual average of $900 million in 1994 to 1996 to $1.6 billion in 1999 to 2001. A deterioration of the state budget was also recorded and the budget deﬁcit increased from 1.9 per cent of GDP in 1994, to 2.6 per cent of GDP in 1995 and 5.7 per cent of GDP in 1996. This situation partly reﬂected the volume of payments from subsidies and pensions, and partly the growing arrears in tax payments as a means of supporting non-proﬁtable ﬁrms. The budget deﬁcits of the state and state-owned ﬁrms were ﬁnanced by credits granted by the NBR. Although the central bank lost control over the increase in the money stock, the NBR reduced interest rates in mid-1996. At the end of the year the leu traded on the unofﬁcial foreign exchange market at a rate 30 to 40 per cent above its ofﬁcial rate, and the current account deﬁcit reappeared. At the time, Romania was still clearly in the early stages of transition and its economic structures were still dominated by the state. Production and use of labour were maintained through subsidies, and in 1997 a new economic programme was negotiated with the IMF. Within the economic and ﬁnancial reform, those relating to the banking system have been particularly signiﬁcant. In 1991, important banking laws (i.e. Law no. 33/1991 concerning banking activity and Law no. 34/1991 concerning the statute of the central bank) were adopted. These two pieces of legislation were the foundation of the two-tier banking system; that is, the commercial banks separated from the NBR as the central bank. As a consequence, the banks were legal entities having as their main activity the attraction of funds from individual, corporate and state bodies, and granting credits to these same individuals and institutions. Their respective charters govern the organisational structure, management and administration of banking companies. The NBR, as central bank of the Romanian state, became ‘the bank of the banks’, being the sole institution that could issue the national currency and conduct monetary and credit policy within the framework of the country’s economic and ﬁnancial policy. The major goal of policy was to

CBI in Romania 229 preserve the stability of the national currency. The central bank also ensures liquidity of the banking system and supervises the activities of the banks. In the 1990s the banking system developed at a rapid pace. Prior to 1989 there had been four banks: the National Bank of the Socialist Republic of Romania and three specialised banks, namely the Bank of Agriculture and Food Industry, the Bank for Investment and the Bank for Foreign Trade. The number of banks rose gradually to forty-ﬁve in 1992. The economic and ﬁnancial development allowed the central bank to launch the most important changes in monetary policy. The most important reforms in the development of the banking sector and monetary policy in Romania are described in Table 9.1. In accordance with the agreement with the IMF, monetary policy focused on the monetary base and on the necessity to drain the excess of liquidity existing in the banking system. The objective was to gain trust in the anti-inﬂationary stance of the central bank. As agent of monetary base management, the NBR no longer granted direct loans to any sector of the economy and all new loans were made at the market rate of interest. The auction had become the main mechanism through which the NBR granted credit and the aim was to ensure increased accessibility for the banks by enabling them to compete on equal terms with other market participants. The development of an efﬁcient ﬁnancial market was regarded as a priority, and the ﬁrst step in achieving this was the development of a competitive auction system, with interest rates determined by the market. The NBR and the Ministry of Finance had collaborated in establishing a system of primary dealers (to purchase state securities) that would also assist in the development of the secondary market. The government intended to submit a new public debt law to Parliament in order to create the legal base for the management of state securities and of public debt. To encourage the market to accept these securities it would have been necessary to ensure that investors receive a positive real rate of interest. Finally, macro-stabilisation policy was conceived on the basis of a ﬂexible exchange rate system. Market forces on the foreign exchange markets determined the exchange rate. The market became free on 18 February 1997 when the NBR re-authorised as primary dealers all participating Romanian and foreign banks. The licensing of foreign banks was important because, as intermediaries on the foreign exchange market, they ensured an alternative for ﬁrms that traditionally dealt with the stateowned banks. The new foreign exchange arrangements eliminated the arrears in external payments, restarted activity on the inter-bank market and reduced the margin between the exchange rates quoted by the exchange houses and those quoted on the inter-bank market. The evolution of the banking system conﬁguration and functioning after the initial stage represented by Law nos 33 and 34/1991 determined several adjustments of the legal framework in order to increase central

230

Silviu Cerna et al.

Table 9.1 Principal steps of the monetary policy construction between 1991 and 2001 May 1991 August 1991 August 1991 September 1991 September 1991 November 1991 January 1992 March 1992 May 1992 June 1993 January 1994 March 1994 April 1994 August 1994 September 1994 April 1995

June 1995 August 1995 December 1995 1996 August 1996 1997

New banking laws introduced: Law no. 33/1991 concerning the banking activity and Law no. 34/1991 concerning the statute of the NBR The Board of Directors is appointed for eight-year period Interest rates liberalised in the banking system with the obligation to maintain differential rates First regulating norms for reﬁnancing adopted. Credit line, auction credit and Lombard (overdraft) introduced Credit ceilings abolished and interest rates completely liberalised New foreign exchange regime introduced and limited convertibility of the ‘leu’ declared. The foreign exchange rate established through daily auctions at the NBR New reﬁnancing regulations adopted Reserve ratio requirements regulations adopted First attempt to use interest rate as anti-inﬂationary lever NBR and the Ministry of Finance signed an agreement to open the States Treasury account at the central bank New rule of reserves’ requirements introduced The norms of organisation and functioning of foreign exchange auctions introduced; de facto liberalisation of the foreign exchange NBR adopted the technical norms concerning the cheque, the bills of exchange and promissory notes NBR issued new norms concerning the foreign exchange market; NBR published the reference exchange rate (the average spot transactions on the market) Romanian-owned banks licensed as dealers on the foreign exchange market The clearing house started its activity; the inter-banking monetary market became ofﬁcial; the NBR started publishing the daily level of the reference interest rate for deposits (BUBID) and for placements (BUBOR) Foreign banks licensed to act as dealers on the inter-bank foreign exchange market New reﬁnancing rules issued The access of banking companies for Lombard credit was limited at 75 per cent of their funds Banking system crisis; The ﬁrst two banks are bankrupt (Dacia Felix and Credit Bank); NBR must grant loans to cover the deposit withdrawal The bank deposit guarantee fund is created The legislation which enabled the NBR to grant loans with subsidised interest is abolished; the monetary policy starts to concentrate on inﬂation; the foreign exchange market is fully functional; the foreign exchange rates are uniﬁed de facto; two other banks suffer severe crisis (The Bancorex and the Agricultural Bank)

CBI in Romania 231 Table 9.1 Continued 1998

1999

2000

2001

Art no. 8 of the IMF statute accepted; Law no. 58 concerning the bank activity and Law no. 101 concerning the statute of the NBR adopted; Law no. 83 concerning the bank bankruptcy is adopted The situation of Bancorex and Agricultural Bank is solved. The bad loans are taken from their portfolio by the Banking Claims Resolution Agency (AVAB) in exchange for state securities; the NBR is reorganised (nineteen county branches are transformed in agencies) The strategy of accession to EU is adopted; Rule no. 1/2000 concerning the money market; Rule no. 2/2000 concerning the loans and placement classiﬁcation according to EU norms. The lending co-operatives enter under the jurisdiction of the NBR The Agricultural Bank is privatised (bought by Austrian Reifeissen Bank); The bank accounts are harmonised with Directive 86 of EEC and the IAS applicable to the lending institutions; collaboration protocol between the NBR and The Romanian National Securities Commission; preparation of the legislation to adopt the euro as legal tender

bank independence, to strengthen the NBR’s ruling powers and prudential supervision, to allow the diversiﬁcation of the monetary policy instruments that can be used as well as their efﬁciency. Thus Law no. 101/98 concerning the statute of the NBR and Law no. 58/98 concerning the banking activity were adopted.

3 The organisation of the NBR according to its statute The legal framework of the NBR deﬁnes the terms of reference and functional organisation of the NBR which are comparable with those of central banks in developed economies. The NBR has its headquarters in Bucharest and is permitted to establish branches and agencies in the capital and other towns. A Board consisting of nine members, appointed by Parliament for a sixyear renewable term, manages the NBR. The term can be renewed. The Board decides the strategic policy of the NBR in the monetary, foreign exchange, lending and payments ﬁelds as well as in the authorisation of the new banks, the prudential supervision, the internal organisation, the wages of personnel and so on. Under special circumstances, the Board can delegate some of its functional responsibilities to the executive management. The Governor, the prime Vice-Governor and two Vice-Governors provide executive management. The Governor undertakes the necessary measures to implement the requirements of the banking laws and the decisions of the Board. The Governor also appoints the NBR headquarters

232

Silviu Cerna et al.

personnel as well as the branch and agency directors. It is important to note that all the appointments are made without any interference from the central or local government, which grants political independence of the NBR staff. The Governor represents the NBR by signing directly or by delegation all relevant treaties and agreements. On behalf of the Board the Governor is accountable to Parliament by presenting the annual report. However, the NBR statutes do not stipulate either the approval of the report by Parliament or the possible measures the government might undertake based on the information contained in the report. Law no. 101/98 also stipulates that appointed auditors are to verify the way the banking law is applied in evaluating the NBR assets, the balance sheet and the securities existing in the NBR’s portfolio along with the state of the budget. Independent auditing ﬁrms selected by the Board by auction analyse all of the different accounts of the NBR. The capital of the NBR is 100 billion lei (about 5.5 million euros according to the exchange rate in November 1999) and is entirely subscribed by the state. The reserve funds are set up primarily at the level of 20 per cent of the gross proﬁt until it becomes equal to the capital, and, from this point forward the percentage is lowered to 10 per cent until the reserve fund level becomes twice the capital when the percentage is lowered to 5 per cent. The NBR headquarters is organised into departments reﬂecting the main operations of the NBR. During 1999 the NBR went through a major reorganising process, an optimal correlation among the organisation of the bank, activities, objectives and responsibilities being envisaged. The ﬁrst stage of reorganisation resulted in a reduction of positions and the transformation of the territorial entities in twenty-one branches and nineteen agencies.

4 The objectives of the NBR The fundamental objective The fundamental objective of the NBR is stated in art. 2, al. 1 of the statute. It reads: ‘The fundamental objective of the NBR is to ensure the stability of the national currency in order to contribute to the stability of prices.’ And further, ‘In order to achieve its fundamental objective the NBR conceives, applies and is accountable for the monetary, foreign exchange, lending and payment policy, and for the authorisation banking supervision within the general policy of the state.’ It is interesting to mention, by comparison, that the fundamental objective of the Bundesbank stated in art. 3 of its statute reads as follows:1 ‘The Bundesbank, as the central bank of the Federal Republic of Germany, is part of the European System of Central Banks (ESCB). It will participate in the fulﬁlment of the ESCB tasks that has as its main objective the main-

CBI in Romania 233 tenance of the price stability and will ensure the internal and international payments.’ We consider that this wording in the NBR statute is clearer than the wording in the former statute (i.e. Law no. 34/91 which in art. 1 stated that: ‘the NBR conceives and conducts the monetary and lending policy within the framework of the economic and ﬁnancial policy of the state aiming at the stability of the national currency’) because it explicitly mentions the stability of prices as the ﬁnal objective of the monetary policy. Obviously, a certain redundancy exists in mentioning both the stability of the currency and of prices, but nevertheless there is an advantage in this ‘security’ wording.2 The interpretation is that in case of conﬂict between the internal and external dimension of monetary stability, the internal one prevails. However, if there is no conﬂict in pursuing both the stability of prices and of the foreign exchange then the NBR is compelled to fulﬁl both aspects of monetary policy. However, another problem arises through using the phrase ‘to contribute’ instead of ‘maintaining’ (price stability). Far from being only a linguistic nuance, this wording raises at least two fundamental problems: (1) the nature of inﬂation in the transition economies, and (2) the ability of the NBR to effectively control the inﬂationary dynamics. Both problems are, of course, speciﬁc to the transition economies. Other objectives Besides the fundamental objective concerning the stability of prices, the NBR has a number of other objectives connected to its role as lender of last resort and its responsibility for ensuring the normal functioning of the ﬁnancial and banking system. As ‘bank of the banks’, the NBR had to face the dilemma whether to act as lender of last resort towards the banks that encountered difﬁculties or to limit itself to maintaining monetary stability through a non-differentiated and credible monetary policy. Thus the shortterm action as lender of last resort implies intervention to support the distressed banks, assuming at the same time the necessary risks involved by the injection of liquidity above the level required by the monetary equilibrium. On the other hand, in the medium term, the objective is to restructure the banking system without affecting the viability and credibility of its other activities. This objective is completed with the strengthening of prudential supervision, the restructuring of distressed banks including radical measures (withdrawal of authorisation, bankruptcy) to avoid the spread of shock waves through the entire banking system. From this point of view, we consider that omitting from Law. no. 101/98 the provisions of art. 27 of the former statute Law no. 34/91 (stating that the ‘NBR will act, under certain conditions, as lender of last resort towards the distressed banks’) is extremely positive and is meant to strengthen the ability of the NBR to act in order to ensure the viability of the entire banking system.3

234

Silviu Cerna et al.

5 The monetary policy instruments After abandoning the monobank system speciﬁc to the centrally planned economy and the adoption of a two-tier banking system, the NBR used, for a short time, direct monetary policy instruments such as ceilings on the lending volume and administrative interest rates. Gradually, as the legal framework was set up, the indirect monetary instruments were used. These instruments were used with different intensity from one period to another according to the actual economic environment, the assumed objectives and the legal framework evolution. Lending (reﬁnancement) facilities In 1991 the NBR abandoned the loan ceilings granted for each bank, preferring to lend through different types of loans (e.g. the credit line, auction loans, ﬁxed-term loans). Later, Norm no. 1/1994 added a new type of loan, namely overdraft facilities, to the lending facilities of the NBR. After four years, the accession of the banks to the resources offered by the NBR became more restrictive, as the new lending rules approved in August 1995 required the banks to guarantee the new borrowings. This rule maintains approximately the same types of loans: structural loans (the interest rate is the ofﬁcial discount rate), auction loans (as the main possibility to grant loans to the banks), special loans (granted to distressed banks) and Lombard loans (granted to cover the temporary deﬁcit of liquidity). The weight of each type of loan varied from one stage to the other. The structural loan,4 present along the entire period analysed, was one of the most important lending possibilities available to the banks. These loans were granted by the approval of Parliament or government decisions and were meant to support certain economic ﬁelds such as agriculture, mining or energy. It is extremely important to note that since 1997 the authorities chose a transparent subvention of the economy through the state budget and not by the loans granted by the NBR, and as a consequence the structural loan was abandoned. In the second part of 1997, the role of the central bank lending signiﬁcantly diminished as an active instrument of monetary policy. Moreover, while, for approximately ﬁve years, lending was the main channel through which the central bank injected liquidity into the banking sector, it became an instrument for the absorption of liquidity (by stopping further lending and providing for the repayment of old loans). The minimum reserve requirements This mechanism was used for the ﬁrst time in 1992. Since then it has undergone several changes, but its basic fundamentals have remained the same:

CBI in Romania 235 • • • •

•

The institutions involved are all banks operating in Romania whether Romanian juridical entities, or branches of foreign banks. The minimum required reserves are computed and maintained through the settlement period. The minimum required reserves are saved in a current account opened at the NBR. The interval speciﬁed to carry out the requirements is the current month and the observation interval is the last ﬁfteen days of the last month and the ﬁrst ﬁfteen days of the current month. The NBR pays interest on the actual level of the required reserves. Any excess reserves do not bear interest.

In 1998 a new rule was adopted (Rule no. 4/1998) to fulﬁl several requirements: to improve the mechanism, to improve the operation of Law no. 101/98, and to increase transparency of this mechanism. The main changes were: • • • • •

the suppression of the banks’ option concerning the currency in which the minimum reserves are denominated; the inclusion of the transit sums denominated in lei in the computing base; the suppression of cash to compute the actual level of the minimum reserves; the suppression of foreign exchange liquidity rate; the introduction of the minimum and maximum settled level for the reserves.

In 1999, further modiﬁcations were introduced and the ﬂuctuation limits of the banks’ accounts at the NBR were set at 0 per cent for the minimum level and 110 per cent for the maximum level. Finally, in 2002, the NBR abolished this norm and adopted Rule no. 6/2002, stating that the computing base for reserves is determined as the daily average of the credit institutions’ liabilities consisting of drawn deposits and other funds in lei and foreign exchange. The following are exempt from calculations of this base: • • •

the resources drawn from the NBR; the resources drawn from credit institutions which are subject to the minimum reserves regime; the equity of the credit institutions.

The minimum reserve mechanism introduced in 1992 became increasingly efﬁcient and was considered an extremely active monetary instrument until the second half of 1997.

236

Silviu Cerna et al.

Open market operations In 1994, the Ministry of Finance issued its ﬁrst securities, thus creating the conditions for open-market operations. However, for two years, the use of security transactions to shape the liquidity in the banking sector was almost impossible given the lack of clear rules, on the one hand, and the low yield of the securities, on the other.5 Under the circumstances that interest rates of the state securities were lower than the ones offered on bank deposits, the demand for such securities was quite low, and they were frequently taken up by the state-owned banks. Starting in 1996, the interest rates offered by the Ministry of Finance were close to those offered on the inter-bank market (and occasionally even above this level), and the placements in state securities rose considerably, thus allowing a more effective use of open-market operations to inﬂuence the liquidities in the banking system. Nevertheless, it should be noted that according to international practice and the terms of the Maastricht Treaty, the NBR cannot buy state securities on the primary market. In order to fulﬁl its monetary policy objectives the NBR can buy state securities on the secondary market either to inject liquidity or to create the necessary portfolio for the open-market operations. By the second half of 1998 these operations were gradually intensiﬁed and open-market operations became one of the most important channels through which the NBR injected liquidity into the banking system. The use of open-market operations was associated with the difﬁculties encountered by the state-owned banks and, with the growing need to ﬁnance the state budget deﬁcit, monetary stability might have been jeopardised. This process imposed a huge cost on the central bank to sterilise the excess of liquidity. The problem was exacerbated because some of the state securities were issued under unfavourable circumstances (i.e. issue of securities to absorb the bad loan portfolio of the banks to the public debt, their interest rate being set according to the last bid of securities of that quarter of the year) It is to be expected that this situation will change substantially under the impact of Norm no. 1/2000 concerning the money market operations of the NBR. According to this rule, the main operations of the NBR are: • • • • • •

reverse-repo buying and selling of eligible assets; granting loans guaranteed by eligible collaterals; buying/selling eligible assets; foreign exchange swap; deposits; permanent lending and deposit facilities.

The eligible participants are the Romanian banks and branches of foreign banks based in Romania if they comply with the requirements of the NBR. The eligible assets for transaction as collateral are state securities

CBI in Romania 237 and other negotiable assets decided by the Board of the NBR. The terms of the open-market operations should not exceed ninety days excepting the certiﬁcates of deposit issued by the NBR that have a maximum term of one year. Foreign exchange market operations Although usually subordinated to the stabilisation of the foreign exchange rate, the NBR operations on the foreign exchange market were undertaken, within the managed ﬂoating regime of the ROL, by taking into account their impact on the money supply. The most important recent event in the exchange rate ﬁeld was the ofﬁcial notiﬁcation by the IMF that Romania has accepted its obligations settled in art. VIII, al. 2–4 of the IMF statute (i.e. the suppression of the restrictions for the current account operations). Besides this important event, the orientation of monetary policy did not experience any essential changes between 1999 and 2000. Thus the foreign exchange rate has, to a large extent, been freely determined on the market, the NBR intervening to avoid relatively large ﬂuctuations of the exchange rate. However, one signiﬁcant development did take place in February 2003 when the NBR decided to use the euro as reference instead of the dollar. This, of course, is in line with Romania’s aspirations for accession to the EU.

6 Prudential supervision The main objectives of prudential supervision exercised by the NBR mainly embraced the following: • •

•

to evaluate the viability and performance of the banks as the banking system evolves; to signal the cases when prudential indices do not reach the proscribed levels, in order to take the necessary measures to prevent the deterioration of the ﬁnancial situation of the banks; to organise on-site inspections.

To promote an efﬁcient banking supervision, aligned with the international practices, the Banking Rating System and Early Notiﬁcation was set up during 1999. Its role is to evaluate the banks in order to identify those which are ﬁnancially and operationally inefﬁcient.6 The introduction of prudential supervision revealed several problems, including: • •

the economic recession that affected the banks’ debtors and their possibilities to respect the liabilities towards the banks; the frequent absence of an adequate policy for resource and placement management;

238 • • •

Silviu Cerna et al. the under-capitalisation of some segments of the banking system; inadequate internal audit; the absence of a liquid market in short-term assets.

When a bank breaches prudential regulations the NBR may impose sanctions including a ﬁne, dismissal of managerial staff and/or auditors (as well as any other person who caused the ﬁnancial deterioration) and ultimately instituting special supervision arrangements which, in the most severe cases, can result in the offending bank being taken into administration. The NBR, as central bank, launched several juridical actions to reorganise or liquidate banks in case of payment cessation and ﬁnally withdrew the licences of banks which could not be rehabilitated. It should be noted that government decision no. 51/98 set up the juridical framework to recover bad loans on the books of state-owned banks and to prepare them for privatisation. The Banking Claims Resolution Agency was founded, its main role being to take over the bad loans of the state-owned banks and their recovery from the debtors. The legal framework was constantly improved in order to deﬁne more clearly the nature of banking risks, and to ensure that risk assessment is in accordance with EU directives and the Basel procedures. In fact, the procedures concerning the control of banking risks comprise three steps: 1

2

3

Prudential supervision of the banking activity is made by the NBR based on the reports on banking prudence transmitted to the NBR (off-site supervision) or at the banks (on-site supervision). If prudential supervision reveals a constant or repeated failure to comply with NBR procedures of the NBR rules and regulations then the NBR can sanction the banks through written warnings, limitation of banking operations, ﬁnes, or withdrawal of their banking licence. Special supervision is instituted by the Board of the NBR in cases where there has been a blatant disregard of prudential regulations or where the actions of a bank might place the ﬁnancial system in a precarious ﬁnancial situation. Special supervision is exercised by a special commission (ﬁve to seven members) each having attributes established by the NBR Board. The term for the special supervision cannot exceed 120 days and, during that period, the bank management works with the NBR representatives who provide advice and make recommendations. However, they do not replace the bank’s management. During the supervision, the appointed Commission must inform the NBR on the situation of the supervised bank and, if the situation does not improve, the Board of the NBR can place the bank in administration. Special administration is undertaken by an individual or juridical entity appointed by the NBR. The special administrator undertakes the management of the bank and also acts as auditor. In order to protect the creditors’ interests the law proscribes:

CBI in Romania 239 • •

the suspension of the voting rights of the shareholders concerning the appointment of the administrator; the suspension of dividend rights of the shareholders in order to pay the bank’s liabilities.

Within forty-ﬁve days of being appointed, the special administrator must inform the Board of the NBR about the ﬁnancial situation of the bank and the prospects for rehabilitation. If the report shows that prudential parameters are obeyed, then the NBR can extend, for a limited period, the special administration arrangements. However, if the reports show that it is impossible to rehabilitate the ﬁnancial situation, then the authorisation may be withdrawn and the bank reorganised or liquidated.

7 Operations with the public sector According to art. 28 Law no. 101/98, the NBR keeps in its books the current account of the Public Treasury opened in the name of the Ministry of Finance. The operation of this account is governed by special agreements between the Ministry of Finance and the NBR. According to art. 29, the NBR receives incomes through the Treasury current account and makes expenditures within the limits of incomes received. For a positive net balance the NBR pays interest, but it does not require any fees for the operations it performs. During the annual financial exercise the NBR can grant loans to the Treasury within the limits established by the law. Finally, based on agreement with the Ministry of Finance, the NBR can act as agent on behalf of the state. Loans based on state securities can be granted according to the agreement, and encompass: • • • • •

the volume of the loan; the term of the loan; the interest rate payable; the cost of debt service; other information concerning the negotiable state securities issued for each loan.

The limits on the loans to the State Treasury are both quantitative and qualitative (i.e. 7 per cent of the state budget revenues of the previous year, for the loans granted during the year, and double the capital and reserve fund of the NBR for the total volume of loans).

240

Silviu Cerna et al.

8 The independence of the NBR: possibilities for quantiﬁcation The independence of the central bank can be estimated, base of the legal criteria, by using a variety of methodologies conceived for this purpose. The most important are the following: Bade and Parkin (BP) (1998) • • •

Is the central bank the highest authority in formulating the monetary policy? Yes Are more than half of the members of the Board independently appointed? Yes Is not there any representative of the government (with or without voting right) on the Board? Yes Total: 4 points (on BP scale)

Grilli, Masciandaro and Tabellini (GMT) (1991) (a) Political independence 1 2 3 4 5 6

7 8

Is the Governor not appointed by the government? Yes He is appointed by Parliament. Does the term of the Governor exceed ﬁve years? Yes The term is six years. Are the members of the Board not appointed by the government? Yes They are appointed by Parliament. Does the term of the Board exceed ﬁve years? Yes The term is six years. There is no representative of the government on the Board. Yes Is the Government approval for the monetary policy necessary? No The monetary, foreign exchange, lending and payment policy is the exclusive competence of the NBR. Must the central bank pursue the price stability objective? Yes This objective is stated explicitly in the statute. Are there any legal provisions strengthening the central bank’s position in case of conﬂict with the Government? No Political independence (GMT scale): 7 points

(b) Economic independence 1 2

Is direct lending to the government not automatic? Is direct lending at market interest rate?

Yes Yes

CBI in Romania 241 3 4 5 6 7 8

Is direct lending temporary? Yes Is direct lending limited? Yes Does the central bank not participate in the primary market of the public debt? Yes Does the central bank establish the discount rate? Yes The prudential supervision is not entrusted to the central bank? No The prudential supervision is not entirely entrusted to the central bank? No Economic independence (GMT scale): 6 points Total (GMT scale): 13 points

Cukierman (LVAU) (1992) 1

2

3

4

The description of the variables The Chief Executive Ofﬁcer (CEO) (a) The term of ofﬁce is six years (TOO). (b) Appointment by Parliament (APP). (c) Dismissal possible only for reasons not related to policy (DISS). (d) Cannot hold ofﬁce in the government (OFF). The monetary policy formulation (a) The NBR decides alone the monetary policy (MONPOL). (b) The NBR is the ﬁnal authority in case of conﬂict concerning the objectives deﬁned in the statute (CONF). (c) The NBR has no active inﬂuence in the budgetary process (ADV). Objectives (OBJ) The stability of prices is the fundamental objective of the NBR, and the collaboration with the public authority pursues the fulﬁlment of this objective. Limitation of lending to the government (a) Advances (limitations of non-securitised lending) limited to 7 per cent of the state budget revenues of the previous year, and the permanent balance cannot exceed double the equity and reserve fund (LLA). (b) Unlimited securitised lending (LLS). (c) Terms of lending agreed between the NBR and the Ministry of Finance (LDEC). (d) Only the central public authority can borrow from the NBR (LWIDTH). (e) The limits of government lending are deﬁned by

Numeric Code 0.75 0.50 0.83 1.00

1.00

1.00 0.00

1.00

0.67 0.00 0.33 1.00

242

Silviu Cerna et al. share in the capital (and reserve fund) of the NBR and by the budgetary revenues (LTYPE). (f) The maturity of loans is maximum 180 days (LMAT). (g) Interest rates are at market level (LINT). (h) The NBR cannot buy or sell state securities on the primary market (LPRM).

0.67 1.00 0.75 1.00

LVAU: 0.67 If the law does not explicitly formulate provisions concerning the resolution of conﬂict between the government and the central bank, then the choice is down to the judgement of the analyst. If the opinion of the analyst is that the government formulates the objectives and the central bank only fulﬁls them passively, then the points awarded are lower and vice versa. Some comments are necessary to clarify the way the LVAU index is calculated. First, it should be noted that compared to Law no. 34/91, the term of the Governor is reduced from eight years to six years. This reduction is compensated by the fact that the government cannot interfere (and is not even consulted as Law no. 34/91 permitted) in appointing the Governor. This fact is not clearly shown by the LVAU index alternatives. Neither is dismissal at the discretion of Parliament. The Governor can be dismissed only for non-political reasons mentioned expressly in the NBR statute. Second, it should be noted that Law no. 101/98 does not clearly mention the resolution of possible conﬂict between the government and the central bank. In appreciating this aspect we considered the provisions of art. 3, al. 1 which states that: ‘The NBR collaborates with the central and local public authorities in fulﬁlling its fundamental objectives.’ This article must be corroborated with al. 2 of the same article which states that: ‘to fulﬁl its fundamental objective the NBR will be asked its point of view concerning the rules and regulations of the public authority in respect with the monetary policy, the banking activity, the foreign exchange regime, and the public debt.’7 The NBR has a consultative role in establishing the terms of any loans granted to the public sector. It expresses its point of view in all the problems concerning public debt (art. 3), without actually determining the resolution of problems, and thus its role in the budgetary process is a passive one. The fundamental objective of the NBR, according to Law no. 101/98, is stability of the price level, and in case of conﬂict with other government objectives, the NBR has freedom to pursue its own objectives. According to art. 2, ‘in order to fulﬁl its fundamental objective, the NBR formulates, applies and is accountable for monetary, foreign exchange, lending policies, for the authorisation and prudential supervision of the banks, within the general policy of the state, pursuing the normal functioning of the

CBI in Romania 243 banking system and the development of a market based ﬁnancial system.’ This situation is different from the one stated in Law no. 34/91, where the obligation of the NBR, as lender of last resort, was accorded the same status as its obligation to ensure monetary stability. It should be noted, though, that the independence of the NBR is comparable, from a legal point of view, with the independence of other monetary authorities in developed and transition countries. The differences derive from the recent evolution of the laws and the ‘old’ and ‘new’ institutional framework.

9 Conclusions An essential characteristic of the monetary organisation of countries participating in the euro is independence of the central bank. In Romania this criteria is accomplished de jure and de facto. Thus, according to Law no. 101/98, the NBR is totally independent in the formation and execution of monetary policy and foreign exchange decisions, and is not accountable to any political entity. The same Law limits the loans the NBR can grant the state Treasury within a year to 7 per cent of the previous year’s state budget revenues and the permanent balance should not exceed double the Bank’s capital plus the reserve fund. After several crises at the end of the 1990s, the Romanian banking system recorded a visible improvement in terms of capitalisation and ﬁnancial performance of registered banks, partly as a result of the improved prudential supervision rules implemented by the NRB. In addition, the consolidation of the banking system is much improved due the following measures: • • • •

adoption of new rules concerning the minimum capital of the banks (320 bn lei in May 2003 and 370 bn lei in May 2004); harmonisation of the accounting legislation with the EU Directives and the International Standards of Accounting (IAS); harmonisation of the acquis communitaire concerning the legal framework for the lending co-operatives; improved prudential supervision by the NBR.

From now on the NBR will implement monetary policy with a view to furthering the prospects of accession to the EU. This will involve: •

• •

the alignment of the monetary policy with that pursued by the European central bank so as to avoid shocks due to a structural change in monetary policy on accession; lowering of the inﬂation to EU levels and direct targeting of inﬂation as an anchor for monetary policy in the near future; adopting a foreign exchange policy which allows for adjustment of the real exchange rate of the leu;

244 • •

Silviu Cerna et al. as of March 2003, the euro is adopted as reference currency to determine the exchange rate of the leu; adoption of the acquis in order to strengthen central bank independence and to make its statute compatible with the functioning of the European System of Central Banks.

Notes 1 As stated in the sixth amendment to the Bundesbank statute. 2 Without going into detail, we may say that by this way a certain dilemma of the monetary policy, coming from the double dimension of the currency stability, internal and external, is prevented. 3 The provisions of art. 2 of Law no. 101/98 according to which the NBR will act ‘pursuing the normal functioning of the banking system’ is compatible, for instance, with the provisions of art. 3 of the ESCB statute stating that this institution is meant to ensure the stable conduct of the prudential supervision of lending institutions undertaken by the competent authorities. 4 The possibility of the banks using these loans ceased before Law no. 101/98 came into force. 5 The norm concerning the possibility of the NBR acting as agent on behalf of the state and making transactions in state securities was issued at the beginning of 1997. 6 A change to this system consisted in the introduction of a new analytical component concerning the quality of the shareholders, considered essential in determining the risk proﬁle of the banks, the policies and the strategic development; moreover, special attention was given to the quality of the management and to a more adequate rating of the capital by using the real index of the capital growth (because the inﬂation is constantly eroding it). Consequently, the considered elements envisage: (1) the adequacy of the capital (C), (2) the quality of the shareholders (S), (3) the quality of the assets (A), (4) the quality of the management, (5) the proﬁtability (P), (6) the liquidity (L). Each of these components are evaluated on a 1(maximum) to 5 (minimum) scale of performance. 7 It is true that the point of view of the NBR cannot be forced on to the government. There is, obviously, a certain ambiguity of the law, which allows for more alternatives. Thus, a possible interpretation is that since Parliament approves the members of the government and appoints the Board of the NBR the legislative body should have the ﬁnal word in case of conﬂict. However, the statute does not expressly mention the recourse to Parliament to solve the conﬂicts, and thus we prefer to interpret this aspect as mentioned.

10 Central bank independence in Bulgaria Garabed Minassian, Janet Ilieva and Olga Teneva

1 Introduction This chapter examines the independence of the Bulgarian Central Bank (Bulgarian National Bank – BNB) and related macroeconomic performance for 1990 to 1998. It starts with a brief historical overview of the period 1879 to 1989, focusing particularly on theoretical issues during the period of the command economy from 1944 to 1989 (Section 2). The progress of banking sector reform since the collapse of the command economy, focusing on its performance and macroeconomic policy during the transition period, is discussed in Section 3. Special attention is paid to the Currency Board Arrangement (CBA) issues and the new status of the BNB with particular emphasis on political and economic independence from government. These issues are discussed in the ﬁnal part (Section 4) of this chapter.

2 Historical overview The BNB was established on 25 January 1879 after Bulgaria secured its independence from the Ottoman Empire as a sovereign state. Three main periods of development of the ﬁnancial and banking system can be distinguished: the Central Bank System of the Kingdom of Bulgaria, with three sub-periods (1879 to 1923, 1924 to 1939 and 1939 to 19471), the command economy phase (1948 to 1989), and the transition period (1990 to 1998). The BNB, 1879 to 1947 The ﬁrst period (1879 to 1947) records the formation and development of the BNB and its performance until the end of 1947, by which time half of Europe was incorporated into what became known as ‘the Soviet Bloc’, with centrally planned, socialist economies. Formerly, Bulgaria was a monarchy. In the sub-period 1879 to 1923, the BNB’s operations resembled those of a commercial bank dealing with customers and ﬁrms. The Bulgarian currency (the lev2) was ofﬁcially

246

Garabed Minassian et al.

introduced in 1880 but it was issued for the very ﬁrst time in 1881. Until then, foreign currency (mainly silver) was in circulation. The more difﬁcult times for the BNB were those of the war periods: the Balkan War (1912 to 1913), the Alliance War (1913) and the First World War (1914 to 1918) when the country had to pay large reparations to its neighbours and was thus burdened with huge war debts. In all these cases the government ﬁnanced current deﬁcits by resorting to the classic policy of loans, which led to signiﬁcant inﬂation. At the beginning of the Balkan War, the BNB started issuing banknotes not convertible into silver and gold coins. The government debt to the BNB rose from eight million levs in 1912 to ﬁve billion levs in 1922 – an increase of some 475-fold. The ﬁrst currency controls were introduced in 1919 aimed at restricting foreign currency payments and protecting domestic production. All exporters were obliged to exchange their foreign currency within ten days at the exchange rate determined by the BNB. This kind of control was amended in 1924 but it lasted until almost 1945. The period 1924 to 1939 might be considered the ‘recovery’ period. The post-war period up until 1922 was marked by high inﬂation rates. The ensuing sharp price rises culminated in the collapse of the Stock Market in 1923, and led to the BNB imposing restrictions on the national currency’s convertibility. Government debt also increased to 300 billion levs in 1924. Faced with the deteriorating economic situation, the government sought advice from the League of Nations and this resulted in the adoption of the gold standard. The aim was to create domestic monetary stability while, at the same time, stabilising foreign exchange rates and facilitating international ﬁnancial transactions. Contemporary observers see this as the ﬁrst attempt at establishing an organisation akin to the CBA set up in 1997. Another requirement, documented on 10 March 1928 and set by the League of Nations, was the total reorganisation of the BNB – it became an issuing body dealing with the government and ﬁnancial institutions only with its status enshrined in the Bank Act of 1928. The third sub-period includes the Second World War when BNB’s activities were affected by the government’s military spending, but less so than those of other combatant countries. After the war, it was compelled to conform to the economic policies of the new socialist government, based on the Soviet economic model. The central banking system and the command economy (1948 to 1989) The basic feature of this system was the high concentration of decisionmaking power in the economy and the central role of the Bulgarian Communist Party (BCP) in it. When private property was abolished by the Decree of Nationalisation, all ﬁnancial institutions were merged into the BNB; that is, the entire ﬁnancial system became state-owned.

CBI in Bulgaria 247 The new government’s aim was to establish a ‘single state bank with branches in every rural district, in every factory’ which ‘will constitute as much as nine-tenths of the socialist apparatus. This will be countrywide bookkeeping, countrywide accounting of the production and distribution of goods.’ The means to achieving this model was the complete nationalisation of the banking sector which took place in 1947 along with the creation of the Popular Banks in 1951 that gave birth to the State Savings Bank (SSB). The BCP justiﬁed this creation of state monopoly on the grounds that property gives rights to one or small groups of individuals, which was deemed to be against the interests of society as a whole. Control of the economy required the concentration of all property in the hands of the central government. A typical feature of the socialist banking system was that it had one tier where the BNB combined its traditional activities and those of commercial banks. For that reason it was called ‘Monobank’. This concentration of control in one institution meant that information on money ﬂows was readily available and thus readily subject to control when required. It would not be possible to talk about ﬁnancial interaction since this implied both money supply and money demand. The relationship between them was distorted because the demand for money was monopolised since investment was undertaken by the state itself. Under this system, all enterprises were required to hold accounts with the BNB and their ﬁnances were managed as part of ﬁve-year plans. Normally, production ﬁgures for the quantity of output were set, while quality was given lesser prominence. Managers were faced with the dilemma of quality versus quantity, the latter clearly set out in the plan. Production of goods was standardised since it was much easier to produce more of the same products rather than a range of different products that might require more investment which could have been used to increase quantity and achieve over-fulﬁlment of the plan. Moreover, enterprises could not freely switch money from agreed production targets, set under the plan, because the money was ‘earmarked’ for designated products and their quantities of production. As Kornai (1992, p. 132) described it, ‘money designated to cover the purchase of materials and semi-ﬁnished products cannot be used to pay wages; money for wages cannot be used for materials, and so forth’. He also raised the issue of convertibility, noting that, under this centralised system, the national currency ‘is not actually a universal means of exchange . . . and it is not convertible within the country’. Furthermore, it was not convertible with other currencies since these were held exclusively by the BNB. Bulgarian nationals returning from abroad were required to sell their foreign currency to the BNB at a ﬁxed exchange rate within seven days of entering the country. None the less, nationals working abroad were allowed to spend their earnings in foreign currency stores called ‘Corecom’. In the

248

Garabed Minassian et al.

ﬁnal years of the system, international passport holders were permitted to exchange currencies at prescribed, special rates and hold a maximum of $30. Table 10.1 shows the number of ﬁnancial institutions (state-owned), operating in the period 1965 to 1980 and controlled by the Bulgarian Foreign Trade Bank (BFTB). This bank specialised in foreign trade transactions with the other member states of COMECOM. Compared with the situation before 1945, the results were poor. People’s savings were concentrated in the SSB, where deposits and household loans were held. Interest rates on deposits and loans were set centrally, but were so low that the real rate was (mostly) negative. The SSB functioned under a law dating from the 1960s, which was backed by the state and operated in the national currency only. It had a tightly drawn network of departments and branches throughout the country, covering the whole population. The state exercised monopoly power over investment, and investment policies were implemented through state-owned businesses. Co-operative property holding was permitted, particularly in the agricultural sector, where a range of different types of co-operatives functioned. With the exception of smallholdings, deﬁned by law in terms of acreage and restricted to meeting individual household needs, private businesses were not permitted. Since, under the central planning system, the development of heavy industry was given priority, investment was concentrated there, while the production of consumer goods was given less prominence. The production of consumer goods was standardised, failing to meet individual preferences and tastes. Demand for these goods greatly exceeded supply, resulting in voluntary savings on the part of the population. Savings of this kind were common in the other socialist economies, and was one of the hallmarks of ‘scarcity economies’. The economy was characterised by rigid state planning and regulation, normally through ﬁve-year plans. Quantity of production took precedence over quality. State intervention and control led to distortions in the system: this was the case in Bulgaria from 1945 to 1989.

Table 10.1 Financial institutions in Bulgaria, 1965 to 1980

Central ﬁnancial institutions Branches and departments The State Savings Bank network Departments Departments’ branches Units

1965

1970

1975

1980

3 211

4 139

2 139

3 140

198 84 3,038

211 98 3,038

254 107 2,991

196 195 2,939

Source: Statistical Yearbook of PR Bulgaria (1981, pp. 454–456).

CBI in Bulgaria 249 In a centrally planned economy, a crisis or even collapse is likely to occur when sectors of the economy fail to fulﬁl their plan requirements. In a planned economy, where prices are centrally set, the absence of the market sending out signals about demand leads to mis-matches between supply and demand. Five-year plans were essentially about aggregates, a notional phenomenon. In carrying out the plan, discretion was allowed to ministries, industries and factory management. Enterprises were rewarded if they met prescribed targets. It was a system that encouraged the overestimation of inputs of labour and raw materials, while underestimating output capacity; thus there was misallocation of resources and distortions in the economy. The ﬁnancial system failed to provide a sound basis for economic development because of its inefﬁciency, and the consequence of this was the emergence of a parallel black market. As economic growth declined in the 1980s, pressure for change mounted. The BCP responded by setting up a number of forums to discuss economic restructuring. One result of this was an attempt to change the ﬁnancial system with progress towards establishing a two-tier banking system. The BNB had its commercial functions transferred to new, specialised banks to serve the needs of key sectors of the economy. These banks were the BFTB for international transactions; the Commercial Bank Electronics for the electrical and electronics industries such as computers, radio and television production; Biochim for the chemical industry; the Transport Equipment Commercial Bank; the Agrarian and Co-operative Bank; the Construction Industry Bank; the Bank for Economic Initiatives, and the Economic Bank. By the end of 1989, considerable progress had been made, with ﬁfty-nine commercial banks established and located in the former branches of the BNB. They were all very small and limited to operating only on a regional basis. The pressure for change accumulated over almost half a century and had its origins in complex economic, political and social factors. Over that time some changes were attempted, but with little success. The ‘scarcity economy’, highly bureaucratised, was plagued by chronic shortages, the overproduction of some goods and underproduction of others. In the absence of market incentives, there were price distortions and a serious misallocation of resources. The BNB in the early 1990s The period after 1990 led to dramatic changes in the entire economy. The headquarters of the BNB started performing as a real central bank from the beginning of 1990. Its commercial bank activities, such as bank ﬁnancing and loan provision to the economy that were mainly associated with the past, were cancelled. Most of its branches and ofﬁces in the country were transferred to the newly established commercial banks.

250

Garabed Minassian et al.

Specialised commercial banks opened branches and ofﬁces in the regions. Thus, together with the existing BFTB, SSB and Mineral Bank, there appeared another sixty-seven commercial banks in 1990 but only eleven met the requirements for all types of banking operations inside the country and abroad. Later that number was increased by the Bank for Agricultural Credit, specialising in loan provision, based on foreign capital, to farmers and the First Private Bank. All banks were based on private capital excluding the BNB and BFTB. The commercial banks, based on the former BNB branches, started with capital completely originating from the BNB, and BNB later announced the sale of its shares in these banks. Only state ﬁrms were allowed to buy shares at the beginning, but private capital was allowed to participate later. The market price of the shares was based on their nominal value according to the balance sheet of the commercial bank. The key year of economic and banking reform in Bulgaria was 1991. On 1 February most domestic prices, as well as the current account and the exchange rate, were liberalised. The ﬁrst stand-by agreement with the IMF was signed on 15 March. Parliament enacted a new law for the BNB in June, and a month later the new Constitution was ratiﬁed. In 1992 some private commercial banks emerged, competing with the state banks. Poor supervision on behalf of the BNB, combined with the deteriorating economic and legislative environment, led to the collapse of seventeen banks in 1996. The banks themselves could not claim back the loans they had made to loss-making enterprises due to oversights and lapses in the legislative system. Thus the debt burden of many state-owned enterprises was transferred to the taxpayer, and some of the above enterprises appeared later on the privatisation lists.

3 Economic performance, 1990 to 1998 Overview After a promising start at the beginning of the transition period when prices, exchange rates and interest rates were liberalised, the economic policies conducted in the country obstructed structural reform. The privatisation programme was postponed and most of the economy remained under state control. The banking system collapse of 1996 and the high inﬂation rates at the end of 1996 and the beginning of 1997 contributed to the deteriorating economic environment. Table 10.2 shows the development of the basic macroeconomic indicators for Bulgaria in the period 1990 to 1998. The recession that had persisted for several years was reversed in 1994 when, for the ﬁrst time since the end of the 1980s, positive economic growth was registered. A new socialist (former communist) government came to power at the beginning of the following year enjoying a majority in Parliament. Despite the

GDP (percentage change) GDP per capita (current USD) Export of goods (fob, m USD) Inﬂation (CPI avr, %) Exchange rate (BGL/USD, eoy) Broad money (% of GDP) Basic interest rate (avr, %) Consolidated central government balance (% of GDP) Unemployment (%) Current account/GDP (%) External debt (bl USD) FDI (m USD) Non-gold reserves (eoy, m USD)

Table 10.2 Macroeconomic indicators 1992

⫺8.4 ⫺7.3 947 1,014 3,737 3,956 338.4 82.0 21.9 24.5 82.5 79.0 45.4 48.7 ⫺4.5 ⫺4.9 11.1 15.3 ⫺0.9 ⫺4.2 12.2 13.8 56 42 325 902

1991 ⫺1.5 1,278 3,727 72.8 32.7 78.3 48.6 ⫺12.1 16.4 ⫺10.2 13.8 40 655

1993

1995

1.8 2.9 1,150 1,563 3,935 5,345 96.0 62.1 66.0 70.7 79.5 66.3 63.9 50.0 ⫺4.6 ⫺5.2 12.8 11.1 ⫺0.3 ⫺0.2 11.3 10.1 105 90 1,002 1,236

1994

⫺9.4 1,201 4,890 121.6 487.4 74.4 115.9 ⫺15.3 12.5 0.2 9.6 109 483

1996

1998 ⫺5.6 4.0 1,255 1,547 4,940 4,194 1,058 18.7 1,777 1,675 34.5 29.4 73.5 3.5 2.0 2.7 13.7 12.2 4.1 ⫺0.5 10.4 10.9 505 401 2,121 2,679

1997

252

Garabed Minassian et al.

%

Current bln USD

favourable conditions for economic stabilisation, a range of unsuccessful macroeconomic decisions led to something of an economic breakdown in 1996.3 Privatisation was slowed down and direct foreign investment decreased – exactly when economic boom had been expected. The conﬁdence of the population and the economic agents in the state institutions drastically decreased. A panic evasion from the domestic currency started. The BNB had to raise the interest rate during the second half of 1996 and the average monthly compound interest rate on deposits at the inter bank market attained 1066 per cent (October 1996, on yearly basis). However, the Bulgarian lev (BGL) evasion process was not put a stop to. At the end of 1995 the exchange rate was 70.7 BGL/USD and in mid-February 1997 it reached the level of 3000 BGL/USD. Domestic currency depreciation facilitated the disproportion between its nominal and the purchasing power parity. Foreign currency substitution contributed to BGL depreciation and sharp increase in the dollarisation of the economy (Figure 10.1). The inﬂation rates kept growing reaching hyperinﬂationary value at 242.7 per cent in February 1997. The sharp BGL depreciation led to an improvement in the current account but hyperinﬂation at the beginning of 1997 adversely affected the interests of deposit holders. The BNB’s revenues from seignorage increased in 1997 (Figure 10.2), and the commercial banks’ net revenues from spread differences covered almost 80 per cent of their ﬁnancial expenditures, excluding the spread differences. However, deposit holders lost about three-quarters of the real value of their deposits!

Foreign currency deposits (l.s.)

Dollarisation

Figure 10.1 Foreign currency deposits (in USD) and dollarisation of the economy (%).

%

CBI in Bulgaria 253

Figure 10.2 Seigniorage (monthly, net issue of BGL as per cent from broad money).

The CBA introduction in mid-1997 stabilised the country’s ﬁnances. Inﬂation rates decreased (the average monthly CPI for the ﬁrst twelve months after CBA introduction was 1 per cent while for the twelve months prior to CBA introduction it was 26 per cent). Restoring people’s trust and the trust of economic agents in the state institutions (Balino et al., 1997, p. 7) managed to return part of the withdrawn foreign currency to the commercial banks and the BNB. During the period July 1997 to December 1998 the BNB’s net foreign currency purchases were estimated at DM 1.8 billion (81 per cent net increase in currency board assets). The commercial banks became very cautious when granting loans. Their preparations for the CBA started in the spring of 1997, mostly associated with getting back the loans given. High awareness and caution in granting new ones resulted in highly restricted lending activities. The latter brought about a predominance of the short-term credits in the credit portfolio of commercial banks (Figure 10.3). the long-term loans provided by the commercial banks were BGL 170.6 billion or less than 1 per cent of GDP in 1997. Prevailing among them were the long-term loans given to the population by the SSB. Partly backed by the government were 23 per cent of the short-term loans allocated for assisting the wheat harvest. Both increased caution in loan provision and reduced issue of government securities for ﬁnancing the budget deﬁcit because of the restricted state expenditures led to high liquidity in the economy. Interest paid for

254

Garabed Minassian et al.

300

m BGN

200

Short-term 100

Long-term 0 12.96

12.97

12.98

Figure 10.3 New credits.

time deposits during the ﬁrst half of 1998 fell below 3 per cent. The interests at the interbank market, where revenues were dominated by the expenditures incurred through serving the time deposits (Figure 10.4), were lower. The interesting thing in this situation was that the commercial banks paid a higher price for attracting deposits than the one for interbank ﬁnancing. The only source of revenues to cover these expenditures were interests paid for bank lending and the result was the increasing spread between the interests for loans and deposits. This pattern of interest rates created an imbalance between the asset and liability structures of the commercial banks. Lack of incentives for time deposits resulted in an increase in demand deposits that restrained lending in the long term (Figure 10.5). The stabilisation achieved when the CBA commenced operations was obvious – though much still remained to be done. Monthly deviations in the basic ﬁnancial indices demonstrated a lack of sufﬁcient credibility. Events of January 1998 taught a good lesson. Because of the Christmas holidays, the BNB exchange services were off, as they usually were every year at that time, but it was enough for rumours to start that new arrangements for the ﬁxed exchange rate were expected and the BGL would depreciate. The BNB did not react to these rumours. However, speculation resulted in a depreciation of the lev, and in the course of one month

1000

%

100

10

Index of central bank independence 1 Jan ’95

Jan ’96

Jan ’97

Time deposits Government securities (up to one yr.)

Jan ’98 Inter-bank money market Long-term credits

Figure 10.4 Interest rates.

Demand deposits Saving deposits

Time deposits

Figure 10.5 Composition of the BGL deposits.

256

Garabed Minassian et al.

it traded at an ofﬁcial price far below the ofﬁcial ﬁxed rate. During the ensuing couple of months, the ﬁnancial situation returned to normal – but an important lesson had been learned. Money supply The major tool used in money supply regulation at the beginning of the 1990s was ‘credit ceilings’ but these were abandoned in July 1994. To be cautious, the reserve requirements were increased. They were initially ﬁxed at 7 per cent at the beginning of the 1990s but reached a level of 10 per cent in April 1994. Later, the BNB began using this instrument frequently, and in the thirty months until the end of 1996 it was changed ten times. At the same time the BNB gradually started open-market operations. Since the beginning of 1993 the BNB has performed limited operations in the open market. For the ﬁrst time an auction for repos contracts was organised in mid-1994. The BNB also started acting as a lender of last resort. Financial stabilisation in Bulgaria was heavily dependent on control of the monetary aggregates. The broad monetary aggregate increased by 50 per cent in 1992, which was very close to the nominal GDP growth. Monetary policy could therefore be described as restrictive – a tendency for some time. Currency outside the banks increased in 1992 by 52 per cent, showing people’s preference not to use banking services. Later, in 1996, it decreased by 50 per cent while demand deposits fell by 42 per cent. Interest rates The interest rate policy of the BNB prior to 1990 was symbolic. Interest paid on deposits was ﬁxed and during the second half of the 1980s stood at 1 per cent. The basic interest rate was ﬁxed at 4.5 per cent at the beginning of 1991 and at 45 per cent soon after. According to the law, all interest rates in the country were set with regard to the basic interest rate set by the BNB. All monetary deals between the commercial banks and BNB and the government securities issued by the Ministry of Finance were in accordance with the basic interest rate. The interest rate policy of the BNB was set so as to achieve certain aims such as control of inﬂation, economic growth and stabilisation. In reality, the ‘ﬁne-tuning’ policy caused frequent changes in interest rates, and using interest rates to support economic growth was downgraded in the hierarchy of central bank priorities. Exchange rate Prior to 1990, there were several exchange rates for different types of transactions. The unbalanced exchange rate was kept, but the BGL was relatively strong against other currencies (0.80 BGL/USD), which stimu-

CBI in Bulgaria 257 lated imports, but depressed exports from the country. However, when the exchange rate was liberalised the BGL depreciated. Gross external debt The Bulgarian external debt almost tripled in 1990 reaching USD 11.2 billion compared to 1985 when it was USD 4 billion. A debt service moratorium was imposed in March 1990 mainly because more than 40 per cent of the debt consisted of short-term liabilities which were due in 1990, and, since the main Bulgarian debtors – Libya and Iraq – postponed servicing their debts, Bulgaria joined the embargo against Iraq. Agreement with the Paris Club was easier as the ofﬁcial loan coming from governmental institutions was comparatively small (15 per cent of the gross external debt (GED)). The negotiations with the London Club were prolonged until mid-1994. The authorities did not publish the information about the GED prior to 1990. The BFTB had acted as an intermediary between the Bulgarian government and the external lenders, and therefore formally the GED was not part of government debt, but a debt of the BFTB. The international ﬁnancial institutions were the only ones that followed the development of the Bulgarian GED. In accordance with the London Club agreement, Bulgaria had to issue three types of Brady bonds: DISC, IAB and FLIRB. The Bulgarian Brady deal covered the total amount of USD 8.13 billion. The Russian Central Bank and the National Bank of Poland, which together held some USD 540 million of the Bulgarian debt, decided to settle their claims on a bilateral basis and did not participate in the restructuring agreement. The overall gross debt and debt-service reduction was 46 per cent in present value terms. The total debt and debt-service reduction operation costs included: • • •

257 million USD for the buyback of 1.02 billion USD of debt at a price of 25.19 per cent of the face value; 222 million USD for the purchase of thirty-year US treasury bonds that served as a collateral for the newly issued bonds; 170 million USD for the purchase of US Treasury bonds maturing in 1995 which covered the interest payment for 1995.

The Brady bonds became more attractive due to the possibility for participation in the privatisation programme. Relations with the state budget In practice the BNB was a department of the Ministry of Finance until the 1990s and thus it possessed no independence and was required to ﬁnance

258

Garabed Minassian et al.

the state budget. With the law of 1991, the BNB started reporting its activities to Parliament. According to the law the BNB would ﬁnance the state under the following conditions: 1 2

Loans for no longer than for three months and must be paid back no later than the end of the calendar year. The total sum of the loans that are not paid back at any moment must not exceed 5 per cent of the budget revenues.

The BNB data revealed changes in its assets structure.4 Deposits placed with and credits extended to commercial banks, and credits extended to the government, represented all the BNB assets at the beginning of the 1990s. Gradually the share of the state budget credits decreased but remained at around 10 per cent during the second half of the 1990s. Preparations for the introduction of the CBA started in the spring of 1997 and most of the liabilities of the commercial banks and the government were removed from the BNB portfolio. All the credits provided to the collapsed banks were in the BNB balance, but all government credits were covered by new credit from the IMF transfers under the ﬁfth stand-by agreement. Licence policy and bank supervision The ﬁrst acts of the BNB as an orthodox central bank were inaccurate and inadequate in both its licence policy and supervision of the banks. The license policy suffered from lack of sound control and banks’ capital cover, whereas the banks’ supervision activities allowed for the accumulation of bad loans and a high-risk exposure. Commercial banks’ reﬁnancing by the BNB as a lender of last resort was extremely high in 1995 and 1996 (Figure 10.6). It resulted from pressure coming from the government and political parties. In the spring of 1997 the decision in favour of adopting the CBA became clear and the BNB started adjusting towards the new conditions. For a very short period of time the BNB got rid of all Treasury bills, put a stop to its activities in the open money market and concentrated on providing deposits to the commercial banks. During the second half of 1997 the only ‘bad’ liabilities of the commercial banks were those of the bankrupt banks; that is, the ‘bad’ loans provided by them. At the beginning of the transition there were a large number of bank start-ups. However, as the consolidation process started, their number rapidly decreased. The real difﬁculties began in 1996 when a radical approach was the only way out of the banking system collapse. Seventeen commercial banks (including two big state-owned banks) were closed because of permanent liquidity problems. During the second half of the period three joint commercial banks were set up with foreign capital and six other foreign banks opened branches in Bulgaria. At the end of the

CBI in Bulgaria 259 200 180 160

Billion BGN

140 120 100 80 60 40 20 0 10.93

04.94

10.94

04.95

Turnover portfolio of BNB

10.95

04.96

10.96

04.97

10.97

Funds extended in BGN

Figure 10.6 Reﬁnancing in Bulgaria.

period the big banks held about three-quarters of the commercial banks’ assets. The BNB increased the capital adequacy requirements of the banks with a view to their further consolidation.

4 CBA in Bulgaria: recovering credibility The CBA was introduced in Bulgaria on 1 July 1997 at a time when conﬁdence in state institutions was low. Prior to the introduction of the CBA a number of laws providing for the CBA’s automatic performance and central bank’s independence were enacted. The BNB independence: personal independence The Law for the BNB was adopted by Parliament in mid-June 1997 and set the number of members of the Managing Board of the BNB at seven. Four of these were appointed by Parliament: the Governor and three managers of the major departments of the BNB (suggested by the Governor). The Departments are (1) Issue Department (de facto the Currency

260

Garabed Minassian et al.

Board), (2) Banking Department, and (3) Banking Supervision Department. The President appoints the other three members of the Managing Board. They are supposed to be specialists in banking, macroeconomics and ﬁnance who are not employed by any other bank or institution with executive power; that is, they must be external to the BNB and are expected to represent the interests of society in order to ensure the BNB’s independence. According to the Law, the tenure of the Managing Board members is six years, though the number of renewable tenures is not determined. To ensure that the Managing Board is replaced gradually, only the tenure of the Governor and the Head of the Issue Department was initially six years. The Head of the ‘Bank Supervision’ Department was initially given a four-year tenure and the Head of the ‘Banking Department’ was initially given a two-year tenure. The members appointed by the President started with tenures of ﬁve, three and one year respectively. The next tenure for each of these members is six years. BNB independence: legal independence The Law for the BNB determined the functions of the BNB under the conditions of the CBA (Table 10.3). The Law provided for almost total BNB independence under the CBA from governmental and political decisions. A prerequisite for this independence was the highly limited functions of the BNB. De facto it did not perform any monetary or foreign exchange policy actions. The only tool of monetary control the BNB had was the power to vary the reserve requirements imposed on the banks. The CBA in Bulgaria could provide budget deﬁcit ﬁnancing under strict conditions. The CBA was introduced under a bilateral agreement with the IMF and almost all conditions are to be agreed with them. The ﬁfth standby agreement was contracted in spring 1997 where the relations between the BNB and the state budget were stated. Financing was possible up to the scale of the stand-by agreements with the government. One week after receiving the money transfer from the IMF, the Managing Board should decide whether and how much loan to provide to the government up to the amount of the payment. This decision was ﬁnal and could not be changed later, even if conditions changed. The government is given ninety days to decide whether or not to use the loan. Its decision is also ﬁnal. The implication is that ﬁnancing was totally external and this makes statistical estimation of government debt difﬁcult. On the one hand, external debt was increasing with each transfer from the IMF, but on the other hand, it was simultaneously an increase in government debt at the BNB (the domestic debt). The same debt transaction was recorded twice: (1) as a debt of the BNB to the IMF, and (2), as a debt of the government to the BNB. This peculiarity provided better precision in analysing the domestic debt development.

CBI in Bulgaria 261 Table 10.3 Functions of the central bank and the currency board in Bulgaria Central bank

Currency Board

Monetary policy Open-market operations

None

Interest rate policy • Determines the basic interest rate (BIR) • Determines interest rates at the BNB operations Reserve requirement Regulates: • The reserve base • Percentage of the reserve holdings • The way of their keeping Budget deﬁcit ﬁnancing Provides loans for ﬁnancing the budget deﬁcit under the Parliament decision Issue policy Under the BNB’s decision. There is no direct correlation between money supply and foreign currency reserves Forex policy Interventions at the forex market under the BNB decision, forex regime and disposable forex reserves Bank supervision Monitoring and supervision of commercial banks – monitors whether the regulations are kept concerning capital adequacy, risk exposure, etc.

• Develops methodology for BIR determination but does not directly inﬂuence the BIR • Determines interest rates at the BNB operations

No change

Provides loans to the state budget up to the IMF transfers denominated in SDRs Money liabilities are in line with foreign exchange reserves. Each new money issue must be covered by foreign exchange reserves Keeps ﬁxed ER of the BGL to the DEM. Provides free market exchange between domestic and foreign currency for the population and all economic agents No change

During the ﬁrst year of operation of the CBA, the Managing Board of the BNB respected the budget needs. The stand-by agreement signed in April 1997 was completely fulﬁlled; that is, the IMF made all the transfers. The decisions of the BNB for the ﬁrst four transfers were almost automatic – they were directly transferred to the MoF. In the ﬁrst half of 1998, a surplus was run when the planned deﬁcit was 2 per cent. Bulgaria was negotiating with the IMF for a three-year EFF agreement which was signed in the autumn of 1998, and hence the BNB decided to retain, as a reserve, part of the two transfers coming from the ﬁfth stand-by agreement. The ﬁfth transfer was at the disposal of the government (SDR

262

Garabed Minassian et al.

62.5 m) and the sixth covered BNB reserves (SDR 62.5 m). This agreement demonstrated the extent of BNB independence in its collaboration with the government and the state budget. The Law provides an opportunity for the CBA to act as lender of last resort in certain circumstances. It also provides for reﬁnancing the commercial banks (as a lender of last resort) of up to the amount of the Banking Department deposit; that is, the foreign currency cover. The BNB approved regulations determining the reﬁnancing procedure. According to the regulations there should exist a systematic risk and the reﬁnancing itself depends on the scale of the commercial bank together with a 125 per cent guarantee. The term is up to three months and, if the guarantee falls below 105 per cent, the BNB could terminate the reﬁnancing and gain the guarantee by taking the commercial bank to court. Measuring BNB independence This section provides measurements of the extent of BNB independence. Considerable research has been done in this ﬁeld and among the most robust techniques are those developed by Alesina and Summers (1998, 1989), Grilli et al. (1991), Alesina and Grilli (1992), Cukierman (1992), Cukierman et al. (1992), Eijfﬁnger and Schaling (1992, 1993), Alesina and Summers (1993), Fratianni and Huang (1994), Debelle and Fischer (1994), Van Keulen (1995), Radzyner and Riesinger (1997) to name but a few. Bade and Parkin index of central bank independence The ﬁrst attempt to measure the political independence of a central bank was done by Bade and Parkin (1988). Their measurement is applied below (Table 10.4) to measure the independence of BNB. According to Bade and Parkin’s criteria, the BNB is given the maximum score of independence 4. The methodology employed in this index suggests one score for each positive answer and one score for applying the index. Thus the minimum score on the Bade and Parkin index is 1 and the maximum score is 4. Table 10.4 Bade and Parkin’s (1988) index of BNB independence Question

Score

1 2

Yes

3

Is the BNB the ﬁnal authority? Are more than half of the policy Board appointments made independently of the government? Is there no government ofﬁcial (with or without voting power) on the BNB policy Board?

Total score

Yes Yes 4

CBI in Bulgaria 263 Grilli, Masciandro and Tabellini index (1991) Another index, constructed by Grilli, Masciandro and Tabellini (GMT) (1991), provides a better measurement of central bank independence (CBI). The methodology is almost the same as that used by Bade and Parkin (1988). Each positive score brings a score of 1, and each negative response brings a score of 0. The higher the overall score, the higher the degree of independence. GMT assess the degree of independence in terms of political independence and economic independence (Table 10.5). The degree of political independence is assessed by considering such issues as appointments to the central bank Board, term of ofﬁce of the Board, autonomy in pursuing the monetary policy and setting the goals as well as resolution in case of conﬂict between the central bank and the government. Economic independence focuses mainly on the nature of credit facilities provided to the central government, the conditions for providing them and bank supervision. The results of applying this index to the BNB are reported in Table 10.5. Alesina and Grilli index Similarly, the index developed by Alesina and Grilli (1992) approaches CBI in terms of political and economic independence. Each positive Table 10.5 GMT index (1991) of BNB independence Index of political independence

1991

1997

1 2 3 4 5 6 7 8

Is the Governor not appointed by the government? Is the Governor appointed for more than ﬁve years? Are Board members not appointed by the government? Are Board members appointed for more than ﬁve years? Is there no government representative on the Board? No government approval of monetary policy required? Is the bank required to pursue monetary stability? Are there legal provisions that strengthen the bank’s position in case of conﬂict with the government? Overall index of political independence of BNB Index of economic independence 1 Is the direct credit facility not automatic? 2 Is the direct credit facility based on the market interest rate? 3 Is the direct credit facility temporary? 4 Is the direct credit facility of a limited amount? 5 Does the central bank not participate in the primary market for public debt? 6 Is the discount rate determined by the central bank? 7 Is banking supervision not entrusted to the central bank? 8 Is banking supervision not entrusted to the Central Bank alone? Overall index of economic independence

Yes No Yes No Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes Yes Yes

6

8

Yes Yes Yes Yes No

Yes Yes Yes Yes Yes

Yes No Yes

Yes No Yes

6

7

Total GMT index

12

15

264

Garabed Minassian et al.

Table 10.6 Alesina and Grilli’ index (1992) of BNB independence Index of political independence

1991

1997

1 2 3 4 5

Governor not appointed by the government Governor appointed for more than ﬁve years Board not appointed by the government Board appointed for more than ﬁve years No mandatory participation of government’s representative on the Board 6 No government approval of monetary policy required 7 Statutory requirements that BNB pursues monetary stability 8 Explicit conﬂicts with the government not possible Overall index of political independence Index of economic independence 1 Direct credit facility – not automatic 2 Direct credit facility – on market interest rate 3 Direct credit facility – temporary 4 Direct credit facility – limited amount 5 The BNB does not participate in primary market for public debt 6 Discount rate set by BNB 7 No portfolio constraints 8 No credit ceiling Overall index of economic independence

No Yes Yes No Yes

Yes Yes Yes Yes Yes

Yes Yes Yes 6

Yes Yes Yes 8

Yes Yes Yes Yes No

Yes Yes Yes Yes Yes

Yes Yes No 6

Yes Yes Yes 8

Total index

12

16

answer brings a score of 1 in the index and, the higher the overall score, the higher the degree of independence. The results of applying this index to the BNB are reported in Table 10.6. Cukierman, Webb and Neyapti index So far, the most precise measurement of the CBI has been provided by Cukierman et al. (1992). In this index, each response corresponds to a numerical code classiﬁed within the range of 0.00 to 1.00. The lower the score, the lower the degree of independence. The answers are grouped in clusters and each cluster is accorded a weight in the overall index (Table 10.7). The above indices rank the BNB among the most independent central banks in the world. The problem arising here is whether it is as independent as it appears by reference to these indices. The similarity between the indices employed above to measure the independence of the BNB is that they are all based on the legislative approach; that is, they consider the nature of the legislative provisions only. However, there is a discrepancy between the legislative provisions and the actual behaviour of the central banks in most of the transition and developing countries. A better

CBI in Bulgaria 265 Table 10.7 Cukierman, Webb and Neyapti’s (1992) index of BNB independence Description of variables

Weight

1

0.20

Governor (CEO): (a) Six-year term of ofﬁce (b) Appointed by the legislative power (c) Dismissal possible only for non-policy reasons (d) CEO prohibited by law from holding other government ofﬁce 2 Policy formulation (a) BNB formulates monetary policy (b) BNB given ﬁnal authority over issues clearly deﬁned in the law as BNB’s objectives (c) BNB organises and executes the cash performance of the state budget through commercial and other banks 3 Objectives of BNB: (a) The main task is to maintain the stability of the national currency along with ensuring the stability of the banking system 4 Limitation of lending: (a) Advances to government prohibited (b) Securitised lending only credit against purchases of SDR from IMF (c) Terms of BNB lending speciﬁed in the law, ninety days (d) Only CG can borrow from BNB (e) Type of limit – the state budget is credited annually with the remainder of the annual excess of the BNB’s revenue over expenditure (f) Maturity of BNB loans limited to three months (g) No explicit legal provisions regarding the interest rate (h) BNB prohibited from buying government securities on the primary market The BNB legal independence index

Number code 0.75 0.50 0.83 1.00

0.15 1.00 1.00 0.50 0.15 0.66 0.15 0.10

1.00 0.66

0.10

0.66

0.05 0.025

1.00 1.00

0.025

1.00

0.025

0.25

0.025

1.00 0.801

approach would be to test the banks’ independence according to its actual behaviour, rather than by reference to the BNB chart.

5 Conclusion, with a look beyond 1998 Since the second half of the 1940s, when the communist regime was established in Bulgaria, the system in operation had been a monobank. The BNB, through its branch network extending throughout the country, was the only bank in Bulgaria. Its functions were purely accounting – keeping

266

Garabed Minassian et al.

a track record of the ﬂows and composition of the domestic money aggregates without being involved in decisions on their distribution and redistribution. All such decisions were made at the headquarters of the Bulgarian Communist Party. The role of the BNB was simply to implement these decisions. As early as the start of the democratic changes in 1990, the banking system was ofﬁcially turned into a two-tier system. The BNB formally became a central bank and divested itself of the functions typical of a commercial bank. However, the complete emergence of a market-based banking system took several years to establish fully. The lack of know-how and experience, as well as the habits of obedience nurtured during the long years of totalitarian rule, tended to distort proportions and decisions. In fact, during the ﬁrst years of the 1990s, the BNB continued to act as a subordinate to the ruling majority, despite the independence proclaimed by BNB Law. This resulted in draining the country’s resources and the collapse of the ﬁnancial system. The country was faced with having to make a radical and dramatic decision, and to resort to extreme measures in order to restore conﬁdence and to impose tight ﬁnancial discipline without using political force. Public opinion ﬁrmly supported this course of action and these issues were settled by the introduction of the Currency Board. The subsequent dilemma was connected with the choice of a reserve currency. There were two options: the USD or the DEM. Over 70 per cent of the country’s forex payments were denominated in USD. Despite this, the lev is pegged to the DM (and subsequently the euro) due to aspirations to join the EU. The overwhelming view is that this has proved to be the right decision. The CBA has delivered an encouraging performance. Interest rates fell quickly – surprising even the specialists! Inﬂation also fell to unexpectedly low levels. The trust of economic agents throughout Bulgaria was gradually restored. The ﬁnancial system has adjusted smoothly and has created the prerequisites for economic growth. For ﬁve years in a row, following the introduction of the Currency Board, the Bulgarian economy has delivered positive growth at an annual average of 4 per cent – despite what has sometimes been an unfavourable international environment. The years after 1997 have demonstrated the BNB’s ability to successfully pursue independent policies despite the economic and political upheavals that have taken place during this time. In particular: 1

2

In the ﬁrst half of 1997 the crash occurred in the ﬁnancial markets in South-East Asia. However short, it had a worldwide impact and adversely affected investors in the emerging economies. A year later another ﬁnancial crash occurred – this time much closer geographically, in Russia. This had a more tangible impact on Bulgarian investors and resulted in increased reluctance to enter into longterm commitments.

CBI in Bulgaria 267 3

4

5

6

The war in Kossovo also had an effect by increasing the perceived risk from investing in the region which spilt over into Bulgaria and adversely affected investment. Since the introduction of the CBA, two commercial banks have been wound up due to insolvency (the Credit Bank in early 1999 and BUB (Balkanska Universalna Banka) a year later). Winding up a commercial bank is very risky irrespective of the stability of the ﬁnancial system. Particularly sensitive handling of this was necessary in Bulgaria because the large-scale banking crisis in 1996 to 1997 was still fresh in peoples minds. A re-denomination was carried out in mid-1999, which could also be described as another risky step to take in view of the situation in the country. Mass criticism of the re-denomination was anticipated and there was a widely held view that re-denomination would destabilise the currency. Since the beginning of 2000, a new Currency Act has come into force which almost completely liberalises both the current and the capital accounts of the balance of payments. Bulgarian nationals are now able to transfer unlimited amounts of currency of any type across borders. Such a step under the conditions of uncertain money equilibrium is equivalent to a psychological shock whose consequences can only be logically assessed a priori.

Despite all of these problems, the CBA has managed to withstand the attack of external and internal shocks, and ﬁnancial stability has been maintained and the conditions for economic expansion have been ﬁrmly established.

268

Garabed Minassian et al.

Appendix 10A.1 Central bank independence in transition economies No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Country

Q1

Albania 1 Armenia 0 Azerbaijan 0 Belarus 1 Bosnia and 1 Herzegovina Bulgaria 1 Croatia 1 Czech R. 1 Estonia 0 Georgia 1 Hungary 1 Kazakhstan 1 Kyrgyzstan 1 Latvia 1 Lithuania 0.5 FYROM 1 Moldova 1 Mongolia 1 Poland 1 Romania 1 Russia 1 Slovak R. 1 Slovenia 0.5

Q2

Q3

Q4

Q5

Q6

Q7

Q8

Q9

Q10 Q11

1 0 0 0 1

1 0 0 0 1

1 1 0 0 0

0 0 0 1 1

1 1 1 1 1

1 0 0 1 0

1 0 0 1 1

1 1 0 0 0

0 0 0 0 1

0 1 1 1 0

0 1 1 1 1 1 1 1 1 0 1 1 1 1 1 0 1 1

0 1 1 0 1 0 1 1 1 0.5 0 1 0 0.5 1 0 0 0

1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 0 1

1 1 0 0 0 0 1 0 0 0.5 1 0 0 0 1 1 1 1

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1

1 0 1 1 1 0 0 1 1 1 1 1 1 0.5 1 1 1 1

1 1 1 1 1 1 0 1 0 1 1 1 0 1 1 1 0 1

1 0 1 1 1 0 0 1 0 1 1 0 1 0 1 1 1 0.5

1 0 1 1 1 0 0 1 1 1 1 1 0 1 1 1 0 1

Note 1 Some central banks have given more than one response, and then the average weight of the given answers is shown.

Notes 1 120 Years Bulgarian National Bank 1879–1999, ed. R. Avramov, BNB, Soﬁa, 1999, p. 13. The year 1947 is taken as the end of the period, since the reforms in the banking system started at that time, even though Bulgaria became a communist country in September 1944. 2 The name of the Bulgarian currency ‘lev’ derives from the old version of the word ‘lev’ meaning means lion, which is the crest of the country. It was introduced with a special law for Money Issue published in the State Gazette 49 of 4 June 1880 and approved by Parliament on 27 May 1880 in Veliko Turnovo. The above law was based on the Latin Monetary Convention, observed by France, Italy, Spain, Switzerland and Belgium, to be joined later by Austria and Romania. 3 Reasons for the economic disaster in 1996–1997 are analysed in details in G. Minassian et al. (1998); 4 After CBA introduction BNB changed the balance sheet methodology, which is why the data lines stopped in May 1997.

CBI in Bulgaria 269

Q12 Q13 Q14 Q15 Q16 Q17 Q18 Q19 Q20 Q21 Q22 Q23 Total (%) 0 1 1 0 1

0 1 0 1 1

1 1 1 0 1

0 1 1 0 1

0 1 0 1 1

0 1 1 0 0

0 0 0 0 1

0 1 1 0 0

1 1 1 1 1

0 0 0 0 1

1 0 1 1 0

0 0 0 0 1

0.48 0.52 0.39 0.43 0.7

1 0 0 1 0 0 0 0 1 0 0 1 0 0.5 0 1 1 1

0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1

0 0 1 1 1 1 1 1 1 1 1 1 0 1 1 0 1 1

0 1 1 1 1 1 1 0 1 1 1 1 0 1 1 1 1 1

0 1 1 1 1 1 0 1 1 1 1 1 0 1 1 0 1 1

1 0 0 0 0 1 1 1 1 1 0 0 0 1 0 0 0 0

0 0 0 0 0 1 0 0 0 0 0 0 0 0.5 0 0 0 0

0 0 0 0 0 1 0 0 0 0 0 0 0 0.5 0 1 1 0

0 1 0 1 1 1 0 1 1 0.5 1 1 0 1 0 0 0 1

1 0 0 1 1 1 0 1 0 1 1 1 1 1 0 0 0 0

1 1 1 0 0 1 1 0 1 1 1 1 1 0 0 0 1 0.5

1 0 0 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0.5

0.61 0.52 0.65 0.65 0.7 0.7 0.48 0.7 0.65 0.7 0.74 0.74 0.43 0.72 0.61 0.52 0.57 0.7

References Alesina, A. and Grilli, V. (1992) ‘The European Central Bank: Reshaping Monetary Politics’, in Canzoneri, M., Grilli, V. and Marson, P. (eds), Establishing a Central Bank: Issues in Europe and Lessons from the US, Cambridge: Cambridge University Press, pp. 49–77. Bade, R. and Parkin, M. (1988) ‘Central Bank Laws and Monetary Policy’, mimeo, University of Western Ontario. Barro, R.J. and Gordon, D.B. (1983) ‘Rules, Discretion, and Reputation in a Model of Monetary Policy’, Journal of Monetary Economics, 12, pp. 101–120. Cukierman, A. (1992) Central Bank Strategy, Credibility, and Independence: Theory and Evidence, Cambridge, Mass., MIT Press. Cukierman, A. and Webb, S. (1995) ‘Political Inﬂuence on the Central Bank: International Evidence’, The World Bank Economic Review, 9, pp. 397–398.

270

Garabed Minassian et al.

Cukierman, A., Webb, S. and Neyapti, B. (1992) ‘Measuring the Independence of Central Banks and its Effects on Policy Outcomes’, The World Bank Economic Review, 6, pp. 353–398. Eijfﬁnger, S.C.W. and Schaling, E. (1993) ‘Central Bank Independence in Twelve Industrial Countries’, Banca Nazionale del Lavoro Quarterly Review, 184, pp. 49–89. Elgie, R. and Thompson, H. (1998) The Politics of Central Banks, London, Routledge. Elster, J. (1979) Ulysses and the Sirens: Studies in Rationality and Irrationality, Cambridge, Cambridge University Press. European Commission (1997) Agenda 2000, Supplement 13. Fischer, S. (1995) ‘Central Bank Independence Revisited’, The American Economic Review, Papers and Proceedings, 85 (2), pp. 201–206. Gordon, P. and Barro, D. (1983) ‘Rules, Discretion and Reputation in a Model of Monetary Policy’, Journal of Monetary Economics, 12, pp. 101–121. Grilli, V., Masciandro, D. and Tabellini, G. (1991) ‘Political and Monetary Institutions and Public Financial Policies in Industrialized Countries’, Economic Policy, 13, pp. 342–392. Hausman, J. (1978) ‘Speciﬁcation Tests in Econometrics’, Econometrica, 46, pp. 1251–1271. Healey, N. and Wisniewski, Z. (eds) (1999) Central Banking in Transition Economies, Torún, Poland. Ilieva, J. and Healey, N. (2000) ‘Central Bank Independence in Selected Transition Economies’, Slovo, 12, SEES, London, pp. 87–103. Kydland, F.E. and Prescott, E.C. (1977) ‘Rules Rather than Discretion: The Inconsistency of Optimal Plans’, Journal of Political Economy, 85, pp. 473–491. Landau, D.F. and Garber, P. (1992) The ECB: A Bank or a Monetary Policy Rule?’, in Canzoneri, M., Grilli, V. and Marson, P. (eds), Establishing a Central Bank: Issues in Europe and Lessons from the US, Cambridge, Cambridge University Press. Maxﬁeld, S. (1997) Gatekeepers of Growth, Princeton, NJ, Princeton University Press. Rogoff, K. (1985) The Optimal Degree of Commitment to an Intermediate Monetary Target’, Quarterly Journal of Economics, 100, pp. 1169–1190.

Index

acquis communautaire 141, 143, 155 adaptive expectations 5 aggregate supply 5, 35 Agricultural and Cooperative Bank 52 Akhtar, M.A. 111 Albania 51, 71, 73, 75, 78, 81, 82, 86, 154, 158, 185, 268 Al-Marhubi, F. and Willet, T.D. 124 al-Nowaihi, A. and Levine, P. 32, 33, 37 Alesina, A. 68, 99, 111, 113, 115, 116, 119, 121, 126, 181, 212, 221; and Gatti, R. 68, 40, 125; and Grilli, V. 4, 23, 99, 100, 101, 107, 112, 142, 181, 182, 183, 263, 264; and Roubini, N. 68; and Summers, L.H. 41, 68, 99, 116, 119, 120, 124, 126 Argentina 108, 117, 149, 150 Armenia 71, 73, 75, 78, 81, 82, 86, 143, 145, 154, 158, 268 Australia 96, 116, 188 Austria 116, 183, 185 Azerbaijan 71, 73, 75, 77, 78, 81, 82, 86, 143, 145, 154, 155, 157, 158, 268 Backus, D. and Drifﬂe, J. 42 Bade, R. and Parkin, M. 68, 99, 111, 112, 118, 119, 121, 131, 181,182, 212, 213, 240, 262 Balcerowicz Plan 216 Ball, L.N. et al. 128, 129 Baltic States 69, 70, 146, 156 Bank of England 93, 98, 99 Bank for Foreign Trade 50 Bank of Japan 115 Banque de France 93, 98, 99, 101 Barro, R.J. 126; and Gordon, D.B. 4, 7, 30, 68, 142 Bayesian equilibria 33, 37

Belarus 71, 73, 77, 78, 81, 82, 86, 154, 155, 156, 158, 268 Belgium 116, 183, 185 Berger, H. and Woitek, U. 113 Bleaney, M. 124 Bosnia and Herzegovina 143, 145, 156, 268 Brialt, C. et al. 108 Britain 47, 188 Bruno, M. and Easterly, W. 126 Bulgaria 51, 52, 54, 57, 59, 61, 69, 71–3, 75, 78, 81, 82, 86, 89, 145, 146, 148, 150, 154, 155, 156, 157, 158, 184, 212, 248, 250, 268 Bulgarian Central Bank 151 Bundesbank 69, 80, 93, 96, 98, 103, 104, 107, 115, 124, 129, 142, 160, 233 business cycles 25, 34 Campilo, M. and Miron, J.A. 108, 125 Canada 96, 116, 188 Canzoneroni, M.B. 42 Cargill, T.F. 120 cash plan 49, 57, 79, 80 Chile 108 Cobb-Douglas production function 5, 6 Columbia 108 Communist Party 45, 48, 57, 58, 63, 168, 246 Council for Mutual Economic Assistance (CMEA or Comecon) 46, 48, 50, 57, 59, 60, 67, 72, 227, 247 credibility 3, 5, 6, 9, 11, 13, 30, 32, 33, 41, 68, 128, 130, 131, 155, 186, 233 credit plan 49, 50, 57, 79, 80 Croatia 71, 73, 75, 78, 81, 82, 86, 154, 268 Cuba 59

272

Index

Cukierman, A. 38, 69, 97, 98, 99, 107–10, 111, 112, 113, 115, 116, 117, 118, 120, 121, 122, 123, 128, 131, 133–6, 144, 147, 149, 150, 154, 155, 181, 182, 213, 214, 215, 217, 241, 264, 265; and Webb, S. 69 117, 123, 128 currency board 69, 83, 84, 154, 253, 254, 258–62, 266, 267 Currie, D.A. et al. 22 Czech National Bank 144, 148, 167, 173–90 Czech Republic 69, 70, 71, 73, 75, 77, 78, 81, 82, 86, 89, 107, 124, 141, 147, 148, 150, 154, 155, 156, 158, 167, 175,180, 184, 185, 186, 188, 190, 268 Czechoslovakia 51, 54, 59, 144, 167, 186 De Gregorio, J. 126 De Haan, J. 107, 125; and Kooi, W. 121, 122, 125, 126; and Sierman, C.L.J. 99, 102; and Sturm, J.-E.E. 99, 121 Debelle, G. and Fischer, S. 68, 108, 112, 121, 122, 129 DeLong, J.B. and Summers, L.H. 127 demand for money 57, 126, 247 Denmark 96, 116, 183, 185 discretionary policy 10–11, 30, 38, 40, 41 dollarisation 8, 83, 227, 252 Dvorsky, S. 99, 147 Economic Bank 52 economic growth 199 economic independence 93, 94, 112, 122, 180, 181 Eijfﬁnger, S.C.W. 121, 124, 125, 212; and DeHann, J. 41, 109, 110, 114–16, 118, 125; and Van Keulen, M. 124, 181, 182; and Schaling, E. 99, 108, 111, 112, 115, 124, 127, 131, 152, 181, 212, 221 electronic bank 52 Elgi, R. and Thompson, H. 95, 96, 98, 100, 101, 148 Ellman, M. 67 employment 5, 6, 7, 13, 20, 22, 24–8, 32, 36, 96, 97, 129 Estonia 70, 71, 73, 75, 77, 78, 81, 82, 86, 145, 147, 154, 268 European Central bank (ECB) 141–3, 145, 148, 180, 190, 243 exchange rate depreciation 26 expectations 10; adaptive see adaptive expectations; rational see rational

expectations; augmented Phillips curve see Phillips curve (expectations augmented Phillips curve) Farrell, J. and Maskin, E. 4, 42 Federal Reserve 93, 98, 101–3, 129 feedback rule 17 Finland 108, 116, 185, 188 ﬁscal policy 3, 7, 76–8 Fischer, A.M. 108, 112, 126, 129, 145; and Cooper, J.P. 12 ﬁxed exchange rates 83, 84, 88, 97, 108, 121, 187, 194, 247 ﬂoating exchange rates 121, 237 Foreign Trade Bank 60 France 47, 96, 116, 183, 185 Friedman, M. 8, 11, 12 Fudenberg, D. and Tirole, J. 33 Fuhrer, J.C. 108, 120, 125 FYR Macedonia 71, 73, 75, 78, 81, 82, 86, 154, 156, 158 Gall, L. 103, 104 Germany 47, 51, 54, 59, 62, 160, 183, 185, 188 Georgia 71, 73, 75, 77, 78, 81, 82, 86, 154, 155, 156, 268 Gold Standard 95, 96, 99, 100 Goodhart, C.A.E. 95, 96, 97, 103, 148; and Huang, H. 12 Greece 116, 183, 185 Grilli, V. et al. 41, 68, 94, 98–101, 103, 104, 109, 111–16, 119, 121, 122, 126, 129, 143, 145, 148, 154, 155, 181, 182, 212, 213, 221, 240, 263 Hamada, J. 18 Havrilesky, T. and Granato, J. 212 Hamburg, B. 131 Herrondorf, B. 41, 42 Hochreiter, E. and Kowalski, T. 147 Housman Test 159, 160 Hungary 51, 54, 59, 70, 71, 73, 75, 77, 78, 81, 82, 86, 107, 124, 141, 144, 147, 148, 154, 155, 156 Hutchinson, M.M. and Walsh, C.E. 108, 128–30 IMF 81, 97, 141, 151, 183, 194, 204, 216, 228, 229, 260, 261 inﬂation 3, 4, 6, 9, 12–16, 18, 19, 25, 26, 30–2, 36, 53, 60–2, 68, 69, 74, 75, 82, 88, 96–8, 105, 108, 109, 113, 118–26,

Index 273 128, 130, 131, 147, 151, 157, 158, 159, 160, 161, 162, 188, 194, 199, 216, 225, 228, 229, 243, 246, 252, 253, 256 Hrncir, M and Smidkora, K. 188; bias 4, 11, 12, 24, 30, 31, 38, 68; expectations, 10, 36, 39, 40, 128, 199; rule 9; target 7, 32, 34, 35, 41, 108 instrument independence 108, 122 interest rates see rate of interest International Bank for Economic Collaboration 59, 60 International Investment Bank 59 Iraq 257 Ireland 116, 183, 185, 216 Italy 95, 116, 183, 185, 188 Japan 116, 117, 188 Jenkins, M.A. 120, 125 Jonsson, G. 121 Jordan, T.J. 108, 129, 130 Kazakhstan 71, 73, 75, 77, 78, 81, 82, 86, 143, 154, 155, 158, 268 Kornai, J. 52, 61, 67, 247 Kydland, F. and Prescott, E.C. 9, 41, 68, 98 Kyrgyzstan 70, 71, 73, 75, 78, 81, 82, 86, 154, 155, 156, 268 Latin America 107 Latvia 71, 73, 75, 77, 78, 81, 82, 86, 147, 154, 156, 286 Lenin 45–8, 52, 54, 66, 67 Levine, P. and Pearlman, J. 15, 22 Libya 257 Lithuania 71, 73, 75, 77, 78, 81, 82, 86, 147, 154, 258 Lohmann, S. 42 Loungani, K.P. and Sheets, N. 69, 123, 124, 147 Lucas, R.E. 3, 5, 7, 129 Luxembourg 183, 185 Maastricht Treaty 142, 144, 145, 184 McCallum, B.T. 31 Maliszewski, W.S. 99, 147, 157 Mangano, G. 108, 111, 115, 116, 151 marginal product of labour 5 Marquis, M.H. and Referet, K.L. 126 Marxist theory 45, 46, 48 Matthews, M. 66 Mexico 108

Miron, J.A. 125 Moldova 71, 73, 75, 77, 78, 81, 82, 86, 154, 155, 268 monetary overhang, 61, 62 monetary policy 3, 4, 6, 7, 12, 13, 20, 26, 30, 32, 36, 38–41, 46, 68, 74, 79, 93, 96, 98, 100, 102, 108, 110–15, 131, 143, 173, 175, 187, 194, 195, 199, 202, 207–10, 212, 216, 217, 229, 233, 234, 236, 242, 243, 256; Bulgarian National Bank Chapter 10; Czech National Bank 175–9, 186, 187; Romanian National Bank Chapter 9 monetary rule 3, 35 money supply 7, 57, 80–2, 88, 194, 237, 247, 256 Mongolia 59, 154, 268 Moscow 59, 225 NAIRU 7, 11, 16 Nash inefﬁciency 15 National Bank of Poland 193–223 natural rate: employment 12; output 35; see also unemployment Netherlands (the) 96, 116, 183, 185 Neumann, M.J.M. 108, 112, 113 New Keynesian 36 New Zealand 96, 108, 116, 130, 188 Nolan, C. and Schaling, E. 108 nominal wage rate 4–6, 129 non-cooperative delegation 16, 20, 22 Norway 96, 116, 185 Persson, T. and Tabellini, G. 34, 35, 38, 40, 42 Phillips curve 128–30; expectations augmented 5; long run 11; short run 12 planned economy see state planning Poland 51, 59, 69, 70–3, 75, 77, 78, 81, 82, 86, 89, 124, 141, 145, 147, 148, 150, 154, 155, 156, 158, 184, 185, 193, 198, 199, 212, 215, 216, 217, 268 policy ineffectiveness 12 Policy Target Agreement (New Zealand) 34 political independence 93, 94, 98, 99–105, 111, 112–13, 118, 119, 123, 180 political monetary cycle 34–41, 126, 174 Portugal 116, 183, 185 Posen, A. 108, 131 price level 5, 12, 35

274

Index

privatisation 85–7, 171, 197, 252 Radzyner, O. and Riesinger, S.108, 110, 147, 181 rate of interest 7, 46, 53, 75, 80, 88, 96, 109, 112, 194, 195, 197, 198, 200, 228, 236, 252, 256 rational expectations 8, 9, 11 real wage 5, 6 reserve ratios 7 Rogoff, K. 3, 13, 17, 23, 30, 34, 36, 38, 40, 41, 68, 98; delegation game of 3, 4, 13, 14, 30, 41, 142; and Sibert, A. 34 Romania 51, 59, 69, 71, 73, 75, 77, 78, 81, 82, 86, 89, 147–9, 150, 154, 158, 184, 185, 225, 227, 243, 268 Romer, D. 36, 124 Russia 47, 51, 70, 71, 73, 75, 77, 78, 80–2, 85, 86, 110, 154, 156, 158, 198, 266, 268 sacriﬁce ration 108, 128, 129, 141 Schaling, E. 120, 121 Sergi, B.S. 181 signalling 39–41, 128 Siklos, P.L. 108, 131 Slovak Republic 70, 71, 73, 75, 77, 78, 80–2, 85, 86, 141, 147, 154, 185, 268 Slovenia 70, 71, 73, 75, 77, 78, 80–2, 85, 86, 147, 154, 268 socialist banking system 45 Solveen, R. 131 Soviet Union 46–8, 54, 59, 70 Spain 108, 116, 117, 183, 185 stabilisation 3, 4, 7, 8, 13, 14, 16–20, 25, 27, 28, 30, 195, 196, 199, 227, 228, 237 Stability and Growth Pact 143 state planning 48–50, 53, 56, 58, 69, 70–2, 79–82, 194, 215–16, 247, 248 Stockman, A.C. 126 supply side 18

surplus value 53 Sweden 108, 116, 183, 185 Switzerland 95, 116, 183, 185 Svensson, L.E.O. 42 Sylla, R. 101, 102 Tajikistan 70, 71, 73, 75, 77, 78, 80–2, 85, 86 time inconsistency 12, 31, 128 Tobin, J. 126 trade-off 3, 13, 31 Turkmenistan 70, 71, 73, 75, 77, 78, 80–2, 85, 86 Ulen, T. 31 Ukraine 70, 71, 73, 75, 77, 78, 80–2, 85, 86 unemployment 6–8, 71, 72, 97, 148; natural rate see natural rate of unemployment United Kingdom 108, 116, 185 USA 47, 100–3, 116 utility 8, 11, 35 Uzbekistan 70, 71, 73, 75, 77, 78, 80–2, 85, 86 Vaubel, R. 113 Venezuela 108 Vietnam 59 wage see nominal wage and real wage wage contracts 5, 57 Walsh, C.E. 30, 129 Walsh contract 4, 12, 31, 32, 34, 38–41 welfare 4, 5, 9, 13, 20, 34, 35, 42, 142 World Bank 82, 141, 151 Yugoslavia 51 Zhivkor, T. 67

Central Banking in Eastern Europe (Routledge International Studies in Money and Banking)

Understanding Pensions (Routledge International Studies in Money and Banking)

Monetary Policy in Central Europe (Routledge International Stuidies in Money and Banking)

Challenges in Central Banking

Monetary Stability in Europe (Routledge International Studies in Money and Banking, 16)

Money and Payments in Theory and Practice (Routledge International Studies in Money & Banking)

Dollarization: Lessons from Europe for the America (Routledge International Studies in Money and Banking, 22)

Central Banking and Globalization

Money, Uncertainty and Time (Routledge International Studies in Money and Banking)

Money, Credit and Price Stability (Routledge International Studies in Money and Banking)

Monetary and Banking History: Essays in Honour of Forrest Capie (Routledge International Studies in Money and Banking)

Institutional Change in the Payments System and Monetary Policy (Routledge International Studies in Money and Banking)

Cooperative Banking in Europe: Case Studies (Palgrave Macmillan Studies in Banking and Financial Institutions)

Cooperative Banking in Europe: Case Studies (Palgrave Macmillan Studies in Banking and Financial Institutions)

Financial Policy and Central Banking in Japan

Organisational Change and Retail Finance: An Ethnographic Perspective (Routledge International Studies in Money and Banking)

Islamic Economics and Finance: A Glossary (Routledge International Studies in Money and Banking, 23)

Bank Deregulation and Monetary Order (Routledge International Studies in Money and Banking, 2)

The Future of Payment Systems (Routledge International Studies in Money and Banking)

Financial Geography: A Banker's View (Routledge International Studies in Money and Banking)

The Structure of Financial Regulation (Routledge International Studies in Money and Banking)

Monetary Macroeconomics: A New Approach (Routledge International Studies in Money and Banking, 15)

Finance: A Characteristics Approach (Routledge International Studies in Money and Banking, Number 7)

Central Banking in the Twentieth Century

Open Issues in European Central Banking

Monetary Operations (Handbooks in central banking)

Careers in commercial banking, corporate banking, investment banking

Trade, Investment and Competition in International Banking

Strategic Alliances in Eastern and Central Europe

Banking

Banking

Central Banking in Eastern Europe (Routledge International Studies in Money and Banking)

Understanding Pensions (Routledge International Studies in Money and Banking)

Monetary Policy in Central Europe (Routledge International Stuidies in Money and Banking)

Challenges in Central Banking

Monetary Stability in Europe (Routledge International Studies in Money and Banking, 16)

Money and Payments in Theory and Practice (Routledge International Studies in Money & Banking)

Dollarization: Lessons from Europe for the America (Routledge International Studies in Money and Banking, 22)

Central Banking and Globalization

Money, Uncertainty and Time (Routledge International Studies in Money and Banking)

Money, Credit and Price Stability (Routledge International Studies in Money and Banking)

Monetary and Banking History: Essays in Honour of Forrest Capie (Routledge International Studies in Money and Banking)

Institutional Change in the Payments System and Monetary Policy (Routledge International Studies in Money and Banking)

Cooperative Banking in Europe: Case Studies (Palgrave Macmillan Studies in Banking and Financial Institutions)

Cooperative Banking in Europe: Case Studies (Palgrave Macmillan Studies in Banking and Financial Institutions)

Financial Policy and Central Banking in Japan

Organisational Change and Retail Finance: An Ethnographic Perspective (Routledge International Studies in Money and Banking)

Islamic Economics and Finance: A Glossary (Routledge International Studies in Money and Banking, 23)

Bank Deregulation and Monetary Order (Routledge International Studies in Money and Banking, 2)

The Future of Payment Systems (Routledge International Studies in Money and Banking)

Financial Geography: A Banker's View (Routledge International Studies in Money and Banking)

The Structure of Financial Regulation (Routledge International Studies in Money and Banking)

Monetary Macroeconomics: A New Approach (Routledge International Studies in Money and Banking, 15)

Finance: A Characteristics Approach (Routledge International Studies in Money and Banking, Number 7)

Central Banking in the Twentieth Century

Open Issues in European Central Banking

Monetary Operations (Handbooks in central banking)

Careers in commercial banking, corporate banking, investment banking

Trade, Investment and Competition in International Banking

Strategic Alliances in Eastern and Central Europe

Banking

Banking

Recommend Documents