Beyond Inflation Targeting: Assessing the Impacts and Policy Alternatives

Beyond Inflation Targeting

Beyond Inflation Targeting Assessing the Impacts and Policy Alternatives

Edited by

Gerald A. Epstein Political Economy Research Institute (PERI), University of Massachusetts Amherst, USA

A. Erinç Yeldan Bilkent University, Ankara, Turkey and International Development Economics Associates (IDEAs), India

Edward Elgar Cheltenham, UK • Northampton, MA, USA

© Gerald A. Epstein and A. Erinç Yeldan 2009 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical or photocopying, recording, or otherwise without the prior permission of the publisher. Published by Edward Elgar Publishing Limited The Lypiatts 15 Lansdown Road Cheltenham Glos GL50 2JA UK Edward Elgar Publishing, Inc. William Pratt House 9 Dewey Court Northampton Massachusetts 01060 USA

A catalogue record for this book is available from the British Library Library of Congress Control Number: 2009933402

ISBN 978 1 84720 938 2 Printed and bound by MPG Books Group, UK

Contents List of contributors List of abbreviations Preface PART I

1

2

3

5

6

3

28

44

THEMATIC ISSUES: CLASS RELATIONS AND GENDER IMPACTS OF INFLATION TARGETING

Income, class and preferences towards anti-inflation and anti-unemployment policies Arjun Jayadev The gendered political economy of inflation targeting: assessing its impacts on employment Elissa Braunstein and James Heintz Inflation and economic growth: a cross-country non-linear analysis Robert Pollin and Andong Zhu

PART III

7

INTRODUCTION AND THEORETICAL FRAMEWORKS

Beyond inflation targeting: assessing the impacts and policy alternatives Gerald Epstein and A. Erinç Yeldan Real exchange rate, monetary policy and employment: economic development in a garden of forking paths Roberto Frenkel and Lance Taylor Inflation targeting and the real exchange rate in a small economy: a structuralist approach Jose Antonio Cordero

PART II

4

vii xii xiii

71

93

116

INFLATION TARGETING: CRITIQUES AND COUNTRY-SPECIFIC ALTERNATIVES

Inflation targeting in Brazil: 1999–2006 Nelson H. Barbosa-Filho v

139

vi

8 9

10 11

12 13 14

Beyond inflation targeting

Alternatives to inflation targeting in Mexico Luis Miguel Galindo and Jaime Ros Five years of competitive and stable real exchange rate in Argentina, 2002–07 Roberto Frenkel and Martín Rapetti A general equilibrium assessment of twin-targeting in Turkey Cagatay Telli, Ebru Voyvoda and A. Erinç Yeldan Employment targeting central bank policy: a policy proposal for South Africa Gerald Epstein Inflation targeting and the design of monetary policy in India Raghbendra Jha Towards an alternative monetary policy in the Philippines Joseph Anthony Lim Monetary policy in Vietnam: alternatives to inflation targeting Le Anh Tu Packard

Index

158

179 203

227 248 271 299

315

Contributors Nelson H. Barbosa-Filho is Professor of Macroeconomics and Public Finance at the Institute of Economics of the Federal University of Rio de Janeiro, Brazil and holds a PhD in economics from the New School for Social Research. Since 2003, Barbosa has been working in President Lula’s Administration and he is currently the Secretary of Economic Policy at the Brazilian Ministry of Finance. Elissa Braunstein is Assistant Professor of Economics at Colorado State University, Fort Collins, Colorado. She received her PhD in economics from the University of Massachusetts Amherst and a Master of Pacific International Affairs from the School of International Relations and Pacific Studies at the University of California San Diego. Prior to joining CSU, Braunstein was an assistant research professor at the Political Economy Research Institute (PERI) at the University of Massachusetts Amherst. Her research focuses on feminist analyses of international economics, including the role of foreign direct investment in development. Recent work involves analysing the impact of foreign investment on gender-based wage inequality in mainland China, the gender-differentiated employment impacts of deflationary monetary policy in developing countries, the linkages between trade liberalization and reproductive health and rights, and the impact of US state tax policy on women’s employment. Among her recent publications are Trading Women’s Health & Rights? Trade Liberalization and Reproductive Health in Developing Countries (Zed Books, 2006, co-edited with Caren Grown and Anju Malhotra). Jose Antonio Cordero is Professor of Economics at the Escuela de Economia, Universidad de Costa Rica, San Jose, Costa Rica, and Center for Economic and Policy Research (CEPR). Portions of his chapter were written during an academic year as visiting professor at Mount Holyoke College. Cordero received his PhD from the University of Notre Dame in 1995. His research focuses on the macroeconomics of growth and distribution in open-monetary economies, and on the effects of FDI on economic development. Gerald Epstein is Professor of Economics and founding Co-Director of the Political Economy Research Institute (PERI) at the University of Massachusetts Amherst. His major research and teaching areas are the vii

viii

Beyond inflation targeting

political economy of central banking, the political economy of international finance and egalitarian macroeconomic policy reform. He has published widely in these areas. He has also served as a consultant to the United Nations Development Programme (UNDP), International Labor Organization (ILO), United Nations Conference on Trade and Development (UNCTAD) and is a staff economist with the Center for Popular Economics (CPE) in Amherst, Massachusetts. Roberto Frenkel is Principal Research Associate at CEDES (since 1977) and Professor at the University of Buenos Aires, Argentina (since 1984). Presently he is also Director of the Graduate Program on Capital Markets, University of Buenos Aires and teaches graduate courses at the Di Tella and FLACSO-San Andrés Universities in Argentina and the University of Pavia in Italy. He is a member of the UNDP Advisors Group, member of the Board of the World Institute for Development Economic Research (WIDER), United Nations University and member of the Academic Council of CEFID-AR. He has published numerous books and articles in academic journals on macroeconomic theory and policy, money and finance, inflation and stabilization policies, and labor market and income distribution, with special focus on Argentina and Latin America. Luis Miguel Galindo is Professor of Economics at the Faculty of Economics at the National Autonomous University of Mexico in Mexico City. He received his PhD in economics from Newcastle University in Newcastle upon Tyne in the UK. His main research interests are monetary economics and environmental economics. His recent research is concentrated in the transmission mechanism of monetary policy in emerging economics and the economics of climate change. He has worked as an advisor to several financial institutions and international organizations. James Heintz, an Associate Research Professor and Associate Director of the Political Economy Research Institute (PERI at the University of Massachusetts, Amherst), holds a PhD from the University of Massachusetts and a Master’s degree from the University of Minnesota. He has written on a wide range of economic policy issues, including job creation, global labor standards, egalitarian macroeconomic strategies and investment behavior. He has worked as an international consultant on projects in Africa sponsored by the International Labor Organization (ILO) and the United Nations Development Program (UNDP), that focus on employment-oriented development policy. From 1996 to 1998 he worked as an economist at the National Labour and Economic Development Institute in Johannesburg , a policy think tank affiliated with the South African labor movement.

Contributors

ix

Arjun Jayadev is Assistant Professor of Economics at the University of Massachusetts Boston. He has also been a research fellow at Columbia University’s Committee on Global Thought. His work focuses on development and inequality. He received his PhD from the University of Massachusetts Amherst in 2005. Raghbendra Jha is Professor of Economics and Executive Director of the Australia South Asia Research Centre, Arndt-Corden Division of Economics, Australian National University, Canberra. He obtained his PhD in economics in 1978 from Columbia University, New York, and has previously taught at Columbia University and Williams College in the US, Queen’s University in Canada, the University of Warwick in the UK, and the Delhi School of Economics and Indira Gandhi Institute of Development Research in India. His work focuses on macroeconomics and public economics. Joseph Anthony Lim is currently a Professor in the Economics Department of the Ateneo de Manila University, the Philippines. He was also professor in economics at the University of the Philippines from 1978 to 2004. He was a Rockefeller scholar and his PhD and Master of Science graduate degrees were obtained from the University of Pennsylvania and Massachusetts Institute of Technology, respectively. His fields of expertise are macroeconomics and development economics. His research works include analyses of the Asian crises, growth analyses and diagnostics of the Philippine macroeconomy, the debt burden and its adverse impact on social and economic spending of developing countries, and the gender aspects of employment generation. He was a macroeconomics, finance and debt advisor to the Bureau for Development Policy (BDP) division of the United Nations Development Programme (UNDP) New York headquarters from 2002 to 2004. He is currently a board member of the Journal of the Asia Pacific Economy published by Francis and Taylor. Le Anh Tu Packard is Senior Economist at Moody’s Economy.com, a division of Moody’s Analytics. She is also a Research Fellow and Convener of the Research and Study Group on Vietnamese Social and Economic Reform, Center for Vietnamese Philosophy, Culture and Society, Temple University, Philadelphia, PA, US, and has spent many years helping the Vietnamese government on various aspects of economic reform. Her current research focus is on the international business cycle, trade and capital flows, and financial crises. Robert Pollin is Professor of Economics and founding Co-Director of the Political Economy Research Institute at the University of Massachusetts Amherst. He received a PhD in economics from the New School for Social

x

Beyond inflation targeting

Research in New York City in 1982. His research centers on macroeconomics, conditions for low-wage workers in the US and globally, and the analysis of financial markets. Has also written and consulted in the area of labor market policies, including the viability of ‘living wage’ policies in the US. He has directed United Nations Development Programme sponsored projects on employment targeted macroeconomic policies for South Africa and Kenya, and most recently has focused his research on the analysis of job creation in the US in the transition to a renewable energy based economy. Martín Rapetti is Assistant Researcher at CEDES and a PhD student at the University of Massachusetts Amherst. His field of specialization is macroeconomics for developing countries and his current research focuses on the effects of competitive real exchange rates on economic development. Jaime Ros is Professor of Economics at the University of Notre Dame, Notre Dame, Indiana and Fellow of the Kellogg Institute for International Studies. Ros specializes in development economics with special reference to Latin America. His most recent book is Development Theory and the Economics of Growth (University of Michigan Press, 2000). His articles have appeared in the Cambridge Journal of Economics, World Development, Journal of Development Studies, The Manchester School of Economics and Social Studies, El Trimestre Economico, Desarrollo Economico and other scholarly journals and edited books. Current projects include an edited handbook with Amitava Dutt on development economics as well as a book on the history of Mexico’s economic development with Juan Carlos Moreno Brid. Lance Taylor is the Arnhold Professor of International Cooperation and Development at the New School University, New York. He received a PhD in economics from Harvard University in 1968. He has been Professor in the economics departments of Harvard and the Massachusetts Institute of Technology, as well as a visiting professor at the University of Minnesota, the Universidade da Brasilia, Delhi University and the Stockholm School of Economics. He moved to the New School for Social Research in 1993. Taylor has published widely in the areas of macroeconomics, development economics and economic theory. He has served as a visiting scholar or policy advisor in over 25 countries, including Chile, Brazil, Mexico, Nicaragua, Cuba, Russia, Egypt, Tanzania, Zimbabwe, South Africa, Pakistan, India and Thailand. Cagatay Telli is a Researcher at the State Planning Organization, Ankara, Turkey. He is a graduate of Bilkent University and his areas of expertise are on general equilibrium modelling and computational methods.

Contributors

xi

Ebru Voyvoda is Assistant Professor of Economics at the Department of Economics, Middle East Technical University, Ankara, Turkey. She holds a PhD from Bilkent University. Her areas of specialization include general equilibrium modeling, economics of productivity growth and technical change, and environmental economics. A. Erinç Yeldan is Professor of Economics at Bilkent University, Ankara, Turkey. He was a Fulbright Scholar and Visiting Professor at the University of Massachusetts Amherst and Amherst College while this book was being written. He is also one of the directors of the International Development Economics Associates (IDEAs), New Delhi. Yeldan received his PhD from the University of Minnesota in 1988. His recent work focuses on development macroeconomics and on applied general equilibrium models with emphasis on the Turkish economy. Andong Zhu is Assistant Professor of Economics at Tsinghua University, Beijing, China. He is a Research Associate of PERI at the University of Massachusetts Amherst. Zhu received his PhD from the University of Massachussetts Amherst in 2005. His recent work focuses on political economy and the imbalances in the world economy and China.

Abbreviations BCB CBRT CGE CMN CPE CPI ET FX GEAR ILO IMF ISSP IT MDG MLE PERI PPP RER RBI SAPs SBV SCRER SME SOCB SOE UNCTAD UNDP VAR WTO

Brazilian Central Bank Central Bank of Turkey computable general equilibrium National Monetary Council Center for Popular Economics consumer price index employment targeting foreign exchange growth, employment and redistribution International Labour Organization International Monetary Fund International Social Survey Program inflation targeting Millennium Development Goals medium to large enterprise Political Economy Research Institute purchasing power parity real exchange rate Reserve Bank of India structural adjustment programs State Bank of Vietnam stable and competitive real exchange rate small- to medium-sized enterprise state-owned commercial bank state-owned enterprise United Nations Conference on Trade and Development United Nations Development Programme vector auto-regression World Trade Organization

xii

Preface The chapters in this edited volume report on the results of a three year international research project designed to evaluate the impacts of the inflation targeting approach to central banking which has swept the global economic landscape in the last decade or so. Moving beyond critique, though, our main motivation for this project has been to develop socially useful alternatives to inflation targeting. This task has become only more crucial over the years since we began the project, because, as we describe more fully in the following pages, we believe inflation targeting has largely had negative consequences for economic development, equality and poverty reduction, especially in the developing world, even as more and more countries adopt this flawed monetary policy regime. This has become especially apparent as a global financial crisis engulfs much of the world while many central banks have been slow to respond, partly due to their obsession with keeping commodity inflation in the low single digits. In structuring the project, we have been anxious to avoid the one size fits all approach that underlies not only inflation targeting itself but also much of the policy approach of the neoliberal ‘Washington Consensus’ of which inflation targeting has become a key part. To that end, the authors of the country study chapters were tasked with designing concrete, countryspecific alternatives to inflation targeting. In addition, other authors have written studies of long neglected themes in central banking, including the gender impacts and class aspects of monetary policy. While most of the chapters are historical and empirical in nature, a few lay out a basic macroeconomic modeling framework to help us understand the key issues at stake. Initiated and directed by Gerald Epstein of the Political Economy Research Institute (PERI) and Economics Department of the University of Massachusetts and Erinç Yeldan of Bilkent University in Turkey, this ‘Alternatives to Inflation Targeting project’ has benefited from the numerous presentations and discussions we have had over these several years in various parts of the world with many economists, activists and practitioners. More generally, a multi-author, multi-country project such as ours requires many other things to be successful, including, first and foremost, creative, skilled and dedicated researchers, which we are fortunate to have found. The work of many of them appears in these pages.1 In xiii

xiv

Beyond inflation targeting

addition it takes financial resources to mount a project such as this, and we thank the Ford Foundation, the Rockefeller Brothers Fund, the United Nations Department of Economic and Social Affairs (UN-DESA) and PERI for financial support. The project has also benefited from the help and encouragement of many people and we especially thank K.S. Jomo of UN-DESA, Manuel Montes and Michael Conroy previously of the Ford Foundation and Rockefeller Brothers Fund respectively, and Jo-Marie Greisgraber and Jamie Baker of New Rules for Global Finance. We are also grateful to Roberto Frenkel and CEDES for hosting an important project conference in Buenos Aires. Thanks are also due to our graduate students, Hasan Comert and Luis Rosero, who have contributed first rate research assistance, and to Judy Fogg of PERI who has provided untold, key support along the way. Finally, we are indebted to Alan Sturmer, Acquisitions Editor at Edward Elgar Publishing, for his help, support and encouragement along the way.

NOTE 1. Many of these chapters appeared in shorter form in a Special Issue of the International Review of Applied Economics (IRAE), March 2008: ‘Inflation targeting, employment creation and economic development: assessing the impacts and policy alternatives’. Thanks very much to Malcolm Sawyer, editor of the IRAE, for his help with that issue and his cooperation with this book project as well.

PART I

Introduction and theoretical frameworks

1.

Beyond inflation targeting: assessing the impacts and policy alternatives Gerald Epstein and A. Erinç Yeldan1

1.1

INTRODUCTION

Inflation targeting (IT) is the new orthodoxy of mainstream macroeconomic thought. The approach has now been adopted by 24 central banks, and many more, including those in developing countries, are expressing serious interest in following suit. Initially adopted by New Zealand in 1990, the norms surrounding the IT regime have been so powerful that the central banks of both the industrialized and the developing economies alike have declared that maintaining price stability at the lowest possible rate of inflation is their only mandate. It was generally believed that price stability is a pre-condition for sustained growth and employment, and that ‘high’ inflation is damaging the economy in the long run. In broad terms, the IT policy framework involves ‘the public announcement of inflation targets, coupled with a credible and accountable commitment on the part of government policy authorities to the achievement of these targets’ (Setterfield, 2006, p. 653). As advocated, ‘full fledged’ inflation targeting consists of five components: absence of other nominal anchors, such as exchange rates or nominal GDP; an institutional commitment to price stability; absence of fiscal dominance; policy (instrument) independence; and policy transparency and accountability (Bernanke et al., 1999; Mishkin and Schmidt-Hebbel, 2001, p. 3). In practice, while few central banks reach the ‘ideal’ of being ‘full fledged’ inflation targeters, many others still focus on fighting inflation to the virtual exclusion of other goals. For its proponents, the appropriate inflation target is typically prescribed as maintaining price stability, though there is less agreement on the meaning of this term and on its precise measurement. Many practitioners simply adopt the widely cited definition of Alan Greenspan, the former Governor of the US Fed, issued at the July 1996 meeting of the Federal Open Market Committee, as ‘a rate of inflation that is sufficiently low that households and businesses do not have to take it into account in 3

4

Beyond inflation targeting

making every day decisions’. For Feldstein (1997), however, price stability meant a long-run inflation rate of zero. In addition, IT is usually associated with appropriate changes in the central bank law that enhances the independence of the institution (Bernanke et al., 1999, p. 102; Mishkin and Schmidt-Hebbel, 2001, p. 8; see also Buiter, 2006 for an evaluation). Note that this promotion of central bank independence often is inconsistent with the above mentioned commitment to accountability, if by accountability, one means democratic accountability. Ironically, employment creation has dropped off the direct agenda of most central banks just as the problems of global unemployment, underemployment and poverty are taking center stage as critical world issues (Heintz, 2006). The International Labour Organization (ILO) estimates that in 2003 approximately 186 million people were jobless, the highest level ever recorded (ILO, 2004a). The employment to population ratio – a measure of unemployment – has fallen in the last decade, from 63.3 percent to 62.5 percent (ILO, 2004). And as the quantity of jobs relative to need has fallen, there is also a significant global problem with respect to the quality of jobs. The ILO estimates that 22 percent of the developing world’s workers earn less than $1 a day and 1.4 billion (or 57 percent of the developing world’s workers) earn less than $2 a day. To reach the Millennium Development Goal of halving the share of working poor by 2015, sustained, robust economic growth will be required. The ILO estimates that on average, real GDP growth has to be maintained at 4.7 percent per year to reduce the share of $1 a day poverty by half by 2015, and significantly more than that to reduce the share of $2 a day poverty by half. Moreover, China’s and India’s opening up to the global markets and the collapse of the Soviet system together have added 1.5 billion new workers to the world’s economically active population (Freeman, 2004, 2005; Akyuz, 2006). This means almost a doubling of the global labor force and a reduction of the global capital-labor ratio by half. Concomitant with the emergence of the developing countries in the global manufacturing trade, about 90 percent of the labor employed in world merchandise trade is low skilled and unskilled, suffering from marginalization and all too frequent violation of basic worker rights in informalized markets (see, for example, Akyuz, 2003, 2006; Akyuz et al., 2006). Under these conditions, a large number of developing countries have suffered de-industrialization, serious informalization and consequent worsening of the position of wage-labor, resulting in a deterioration of income distribution and increased poverty. Many of these phenomena have occurred in tandem with the onset of neoliberal conditionalities2 imposing rapid liberalization of trade and premature deregulation of the indigenous financial markets.

Impacts and policy alternatives

5

The key problem is that the ongoing ‘financial globalization’ appears primarily to redistribute shrinking investment funds and limited jobs across countries, rather than to accelerate capital accumulation across a global scale (Adelman and Yeldan, 2000; Akyuz, 2006). Simply put, the world economy is growing too slowly to generate sufficient jobs and it is allocating a smaller proportion of its income to fixed capital formation. In addition, asset price bubbles and crashes, with their attendant financial fall-out are plaguing the system. Under these conditions, it ought to be clear that price stability, on its own, will not suffice to maintain macroeconomic stability, as it cannot suffice to secure financial stability and employment growth. In the words of Akyuz (2006, p. 46), ‘the source of macroeconomic instability now is not instability in product markets but asset markets, and the main challenge for policy makers is not inflation, but unemployment and financial instability’ (emphasis added). Yet, surprisingly, despite a disappointing record, this almost single minded focus on commodity inflation is gaining a more secure foothold in monetary policy circles and the circles are widening to include an increasing number of developing countries. According to a recent report by the International Monetary Fund (IMF), an increasing number of central banks in emerging markets are planning to adopt IT as their operating framework (Batini et al., 2006; Table 1.1). An IMF staff survey of 88 non-industrial countries found that more than half expressed a desire to move to explicit or implicit quantitative inflation targets. More relevant to our concerns, nearly threequarters of these countries expressed an interest in moving to ‘full-fledged’ IT by 2010. To support and encourage this movement, the IMF is providing technical assistance to many of these countries and is willing to provide more (Table 1.1 and further discussion below). In addition, the IMF is considering altering its conditionality and monitoring structures to include inflation targets. In short, despite little evidence concerning the success of IT in its promotion of economic growth, employment creation and poverty reduction, and mixed evidence at best that it actually reduces inflation itself, a substantial momentum is building up for full-fledged IT in developing countries. Promotion efforts by the IMF and Western-trained economists are at least partly responsible for this increasing popularity. While it might seem obvious that stabilization focused monetary policy represents the only proper role for central banks, in fact looking at history casts serious doubt on this claim. Far from being the historical norm, in many of the successful currently developed countries, as well as in many developing countries in the post-Second World War period, pursuing development objectives was seen as a crucial part of the central banks’ tasks (Epstein, 2007). Now, by contrast, development has dropped off the policy agenda of central banks in most developing countries.

6

Table 1.1

Beyond inflation targeting

Inflation targeting countries: initial conditions and modalities

Developing countries (in order of adoption)

IT adoption date

Inflation rate at start (% per annum)

Current inflation target (% per annum)

Israel

1997Q2

8.5

1–3

Czech Rep. Poland

1998Q1 1998Q4

13.1 9.9

3 (1/–1) 2.5 (1/–1)

Brazil Chile Colombia South Africa Thailand Korea Mexico Hungary Peru The Philippines Slovak Republic Indonesia Romania Turkeya Turkeyb

1999Q2 1999Q3 1999Q3 2000Q1 2000Q2 2001Q1 2001Q1 2001Q2 2002Q1 2002Q1 2005Q1 2005Q3 2005Q3 2006Q1 2001Q2

3.3 2.9 9.3 2.3 1.7 3.2 8.1 10.5 –0.8 3.8 3.2 7.8 8.8 7.8 82.0

4.5 (1/–2) 2–4 5 (1/–0.5) 3–6 0–3.5 2.5–3.5 3 (1/–1) 3.5 (1/–1) 2.5 (1/–1) 5–6 3.5 (1/–1) 5.5 (1/–1) 7.5 (1/–1) 5 (1/–2) n.a

Industrial Countries New Zealand Canada

1990Q1 1991Q1

7.0 6.2

1–3 1–3

United Kingdom Sweden Australia Iceland Norway

1992Q4 1993Q1 1993Q2 2001Q1 2001Q1

3.6 4.8 1.9 3.9 3.7

2 2 (1/–1) 2–3 2.5 2.5

Candidate Countries Costa Rica, Egypt, Ukraine Albania, Armenia, Botswana, Dominican

Near Term (1–2 years) Medium Term (3–5 years)

Officially declared policy instrument headline O/N rate 2 week repo 28 day intervention selic O/N rate O/N rate repo 14 day repo O/N call rate 91-day Cetes 2 week deposit reverse repo 1-month SBI CB O/N rate CB net domestic assets cash rate O/N funding rate repo repo cash rate

Impacts and policy alternatives

Table 1.1

7

(continued)

Rep., Gutemala, Mauritius, Uganda, Angola, Azerbaijan, Georgia, Moldova, Serbia, Sri Lanka, Vietnam, Zambia

Medium Term (3–5 years)

Belarus, China, Kenya, Kyrgyz Rep., Moldova, Serbia, Sri Lanka, Vietnam, Zambia Bolivia, Honduras, Nigeria, Papua New Guinea, Sudan, Tunisia, Uruguay, Venezuela

Long Term (. 5 years)

Notes: a. Official adoption date for Turkey. b. Turkish central bank declared ‘disguised inflation targeting’ in the aftermath of the 2001 February crisis. Source:

Batini et al. (2006).

The theme of this book is that modern central banking ought to have more policy space in balancing out various objectives and instruments. In particular, employment creation, poverty reduction and more rapid economic growth should join inflation stabilization and stabilization more generally as key goals of central bank policy. In introducing this volume, this chapters outlines why a shift away from IT, the increasingly fashionable, but problematic approach to central bank policies, and a move toward a more balanced approach is both desirable and feasible. The rest of the chapter is organized as follows. In the next section, we briefly survey the macroeconomic record of IT and its current structure. Section 1.3 focuses on the role of the exchange rate as one of the key macro prices, and discusses alternative theories of its determination. We also make remarks on the issue of IT in the context of the so-called ‘trilemma’ of monetary policy. In Section 1.4 we discuss various alternatives to inflation focused central banks, concentrating on the results of a multi-country research project undertaken with the support of UN-DESA, among other organizations. This section shows that there are viable, socially productive

8

Beyond inflation targeting

alternatives to IT, including those that focus on employment generation, and makes the case that these alternatives should be further developed. Section 1.5 concludes the chapter.

1.2

MACROECONOMIC RECORD OF INFLATION TARGETING

Much of the existing literature on the record of IT has focused mostly on whether systemic risks and accompanying volatility has been reduced in the IT economies, and whether inflation has come down actually in response to adoption of the framework itself or due to a set of ‘exogenously welcome’ factors. On the one side, there is a fair amount of agreement that IT has been associated with reductions in inflation. Furthermore, exchange rate pass-through effects were reportedly reduced and consumer prices have become less prone to shocks (Edwards, 2005; Mishkin and SchmidtHebbel, 2001). Yet existing evidence also suggests that IT has not yielded inflation below the levels attained by the industrial non-targeters that have adopted other monetary regimes (Ball and Sheridan, 2003; Mishkin and Scmidt-Hebbel, 2001; Roger and Stone, 2005). Moreover, even if domestic monetary policy has reduced inflation, the hoped for gains in economic growth and employment have, generally, not materialized. On the ‘qualitative’ policy front, it is generally argued that with the onset of central bank independence, communication and transparency have improved and that the central banks have become more ‘accountable’. Yet in practice, ‘central bank independence’ means that central banks have become less accountable to their governments, and, arguably, more accountable to financial elites and international organizations such as the IMF. Moreover, little is known about the true costs of IT on potential output growth, employment, and on incidence of poverty and income distribution. Bernanke et al. (1999) and Epstein (2007), for instance, report evidence that IT central banks do not reduce inflation at any lower cost than other countries’ central banks in terms of foregone output. That is, IT does not appear to increase the credibility of central bank policy and therefore does not appear to reduce the sacrifice ratio. Per contra, based on an econometric study of a large sample of inflation targeters and nontargeters, Corbo et al. (2001) concluded that sacrifice ratios have declined in the emerging market economies after adoption of IT. They also report that output volatility has fallen in both emerging and industrialized economies after adopting IT. This position is recently complemented by a study of the IMF economists, who, using a complex econometric model

Impacts and policy alternatives

9

and policy simulations, report findings that inflation targeting economies experience reductions in the volatility in inflation, without experiencing increased volatility in real variables such as real GDP (Batini et al., 2006). According to these estimates, IT central banks do enhance economic ‘stability’ relative to other monetary rules, such as pegged exchange rates and monetary rules. However, in the assessments of ‘stability’, these papers do not consider the issue of the stability of asset prices, including exchange rates, stock prices and other financial asset prices. As we discuss further below, asset price stability may need to be included in a full analysis of the impact of IT on overall economic stability. Asset price stability aside, while intriguing, these results are only as strong as the simulation model on which they are based and are only as relevant as the relevance of the questions they pose. Moreover, they are only as broad as the alternatives they explore. On all these scores, these results are problematic. First, they do not simulate the impact of IT relative to other possible policy regimes, such as targeting the real exchange rate as discussed below. Second, the model is based on estimates of potential output that are themselves affected by monetary policy (see, for example, Michl, 2007; Tobin, 1980). Hence, if monetary policy slows economic growth, it also lowers the rate of growth of potential output and, therefore, reduces the gap between the two, thereby appearing to stabilize the economy. Equally, if not more important, is the practical problematique of setting the targeted rate of inflation itself. Even if the advocated requisites of the IT regime are taken for granted, it is not yet clear what the practically targeted rate of inflation should be. Even though there appears to be a consensus among the advocates of the IT regime that the inflation target has to be ‘as low as possible’, there is no theoretical justification of this assertion; and as such, it sounds more of a recommendation than a careful calculation. Most disturbing is the common belief that what is good for the industrialized/developed market economies should simply be replicated by the developing countries as well. That this may not be the case is forcefully argued in Pollin and Zhu (2006). Based on their non-linear regression estimates of the relationship between inflation and economic growth for 80 countries over the period 1961–2000, Pollin and Zhu report that higher inflation is associated with moderate gains in GDP growth up to a roughly 15–18 percent inflation threshold. Furthermore they report that there is no justification for IT policies as they are currently being practiced throughout the middle- and low-income countries, that is, to maintain inflation with a 2–4 percent band. Moreover, we have other evidence on the negative consequences of monetary policy designed to produce extremely low levels of inflation in

10

Beyond inflation targeting

developing countries. Braunstein and Heintz show that contractionary monetary policy used to fight inflation often has a differentially negative impact on the employment rates of women relative to men (Braunstein and Heintz, Chapter 5, this volume). Given the possible negative costs of IT on output and employment, there should be some direct survey evidence indicating people’s preferences with respect to inflation and unemployment. While some studies have indicated that people have an absolute preference for low inflation, Arjun Jayadev (Chapter 4, this volume) reports on survey results asking people in different countries and income levels what is their bigger concern, high inflation or high unemployment. His main result is that, in his sample, poorer people are concerned more about high unemployment than high inflation, while richer respondents have the opposite preferences. Hence, concerns over employment and inflation probably have an important class dimension to them, something that economists and historians have suspected for many years. Finally, is the issue of the role of IT in the context of supply shocks, which periodically affect individual economies and, as we have seen recently, the world economy as a whole. Rigid IT rules can prove highly problematic in the context of supply shocks, where the main problem facing countries is too little supply, not too much demand. The solution is to help the economy absorb the loss of real income associated with the supply shock, without incurring collateral damage associated with greater income loss than absolutely necessary, while at the same time making the investments necessary to increase the supply of the key commodities or find substitutes for them over the medium term. Contractionary monetary policy is a decidedly highly inefficient tool for carrying out this complex task. This problem is exacerbated by rigid inflation targets and is only slightly ameliorated by the use of ‘core inflation’ indices which usually ostensibly exclude the first round costs of volatile energy and food prices. This is only a partial solution because in the medium term the increased costs of key commodities must filter their way through the input output structure and, even without real wage resistance on the part of workers which the central banks are presumably concerned with, will raise the core inflation rate temporarily. Hence, rigid adherence to IT in this context will lead to possibly large unnecessary costs in slower economic growth and income. An overall picture on the selected macroeconomic indicators of the inflation targeters can be obtained from Tables 1.2 and 1.3. In Table 1.2 we provide information on the observed behavior of selected macro aggregates as the annual average of five years before the adoption of the IT versus the annual average after the adoption date to the current period.

Impacts and policy alternatives

11

Table 1.3 keeps the same calendar frames and reports data on key macro prices, viz. the exchange rate and the interest rates. Evidence on the growth performance of the IT countries is mixed. Taking the numbers of Table 1.2 at face value, we see that seven of the 21 countries report a decline in the average annual rate of real growth, while three countries (Canada, Hungary and Thailand) have not experienced much of a shift in their rates of growth. Yet clearly it is quite hard to disentangle the effects of the IT regime from other direct and indirect effects on growth. One such factor is the recent financial glut in the global asset markets and the associated surge of the household deficit spending bubble, which is now bursting. The Institute of International Finance data reveal, for instance, that the net capital inflows to the developing economies as a whole has increased from $47 billion in 1998, to almost $400 billion in 2006, surpassing their peak before the Asian crisis of 1997. As the excessive capital accumulation in telecommunications and the dot.com high tech industries phased out in late 1990s, the global financial markets seem to have entered another phase of expansion, and external effects such as these make it hard to isolate the growth impacts of the IT regimes. Despite the inconclusive verdict on the growth front, the figures on unemployment indicate a significant increase in the post-IT era. Only three countries on our list (Chile, Mexico and Switzerland) report a modest decline in their rates of unemployment in comparison to the pre-IT averages. The deterioration of employment performance is especially pronounced (and puzzling) in countries such as the Philippines, Peru and Turkey where rapid growth rates were attained. The increased severity of unemployment at the global scale seems to have affected the IT countries equally strongly, and the theoretical expectation that ‘price stability would bring growth and employment in the long run’ seems quite far from materializing yet. The adjustment patterns on the balance of foreign trade have been equally diverse. Ten of the 21 countries in Table 1.2 achieved higher (improved) trade surpluses (balances). While there have been large deficit countries, such as Turkey, Mexico, the Philippines and Australia, there were also sizable surplus generators such as Brazil, Korea, Thailand, Canada and Sweden. Not surprisingly much of the behavior of the trade balance could be explained by the extent of over-valuation of the exchange rates. This information is tabulated in Table 1.3. Table 1.3, like Table 1.2, calculates the annual averages of the five-year period before the IT versus annual averages after IT to date. Focusing on the inflation-adjusted real exchange rate movements, we find a general tendency towards appreciated currencies in the aftermath of adoption of the IT regimes. Mexico, Indonesia, Korea and Turkey are the most significant currency appreciating countries, while Brazil, and to some

12

Beyond inflation targeting

Table 1.2

Selected macroeconomic aggregates in the inflation targeting countries

Before : Annual average of 5 years prior to adoption of IT; After : Annual average of adoption of IT to current Growth rate

Unemployment rate

Trade Balance (external balance on goods and services) /GDP (%)

Year IT Before After Before After Before After started New Zealand 1990 Canada 1991 United Kingdom 1992 Sweden 1993 Australia 1994 Israel 1997 Czech Rep.a 1998 Poland 1998 Brazilb 1999 Colombia 1999 Mexico 1999 South Africac 2000 Switzerland 2000 Thailand 2000 Korea 2001 Hungary 2001 Perud 2002 Philippines 2002 Indonesia 2005 Turkeye 2006 Turkeye 2001Q2

2.7 2.9

3.0 2.8

4.2 8.4

6.9 8.7

2.2 0.8 2.2 5.8 4.5 7.9 3.2 3.3 1.7 2.6 1.4 1.5 4.6 4.2 2.0 3.1 4.6 4.5 4.0

2.7 2.7 3.9 3.1 3.2 3.7 2.3 2.3 4.8 3.8 1.7 1.7 4.5 4.2 5.2 5.1 5.6 7.8 4.5

7.4 2.8 8.6 8.5 4.0 14.3 7.0 11.1 2.7 n.a 4.1 1.9 4.4 8.0 7.8 10.2 6.5 9.9 6.6

5.2 6.1 7.3 9.4 8.9 16.7 9.8 15.8 1.9 27.7 3.1 2.4 3.7 6.1 10.2 11.5 10.3 10.4 10.0

0.4 0.5

1.3 2.7

–2.5 –1.6 1.3 6.2 –0.6 –1.3 –14.6 –7.2 –3.4 –1.8 0.0 –4.1 –1.7 1.0 –6.0 –0.5 –0.5 –1.9 0.0 0.0 0.1 0.1 0.0 0.1 0.0 0.0 –1.3 –2.9 –3.2 –0.4 –3.6 –0.7 7.3 –4.6 –9.8 –11.0 –7.5 –9.8

Central bank foreign reserves (US$ Million)

Before

After

2 897.9 11 964.0

4 623.2 24 256.0

39 666.5 37 408.5 15 399.0 18 521.8 13 777.9 20 337.1 7 567.3 24 421.1 9 172.5 21 686.5 12 591.8 31 581.8 47 701.3 42 304.5 7 567.3 24 421.1 20 630.9 51 396.6 15 860.0 9 580.0 38 277.1 40 646.5 32 556.1 40 474.8 55 299.5 157 739.2 9 918.1 13 652.1 9 264.8 11 222.9 11 281.6 14 006.6 31 326.7 32 989.2 33 237.4 56 990.4 20 083.4 33 237.4

Notes: a. The period before the inflation targeting refers to the period of 1994–97 for ‘Growth’ and ‘CPI inflation’ for the Czech Republic. b. The period before the inflation targeting refers to the period of 1996–98 for reserves in Brazil. c. The period before the inflation targeting refers to the period of 1994–97 and after inflation targeting refers to the period of 1999–2004 for unemployment rate in South Africa. Note that due to change in methodology and data coverage, unemployment figures are not directly comparable before and after apartheid. d. The period after the inflation targeting refers to the period of 2003–04 for unemployment rate in Peru. e. Official adoption date for Turkey is 2006. However, Turkish central bank declared ‘disguised inflation targeting’ in the aftermath of the 2001 February crisis. Source:

IMF (2008).

Impacts and policy alternatives

Table 1.3

13

Macroeconomic prices in the inflation targeting countries

Before: Annual average of 5 years prior to adoption of IT; After: Annual average of adoption of IT to current Inflation rate (variations in CPI)

Exchange rate real depreciationg,h

Year IT Before started

After

Before

After

New Zealand 1990 11.6 Chilea 1991 19.7 Canada 1991 4.5 United Kingdom 1992 6.4 Sweden 1993 6.9 Australiab 1994 4.2 Israel 1997 11.3 Czech Republicc 1998 9.1 Polandd 1998 24.1 Brazil 1999 819.2 Colombia 1999 20.4 Mexico 1999 24.5 Thailand 2000 5.1 South Africae 2000 7.3 Switzerland 2000 0.8 Korea 2001 4.0 Hungary 2001 15.2 Peru 2002 5.0 Philippines 2002 6.3 Indonesia 2005 8.0 Turkeyf 2006 28.3 Turkeyf 2001Q2 74.1

2.2 7.2 2.1

–7.6 –6.0 –7.5

–0.6 –4.0 –1.7

2.6 1.5 2.5 3.1

–2.4 –8.5 –6.9 –4.2

–2.2 1.2 –1.1 0.9

3.1 4.7 7.9 7.5 7.2 2.2 5.1 1.0 3.3 5.9 1.9 5.0 10.5 10.5 28.3

CB real interest rateh,i Before

Public assets real interest ratei,j

After

Before

After

7.0 .. 6.0

5.5 0.0 2.6

2.1 –16.0 5.8

5.1 –4.6 2.5

5.4 2.8 7.1 2.0

3.0 1.7 3.2 5.0

5.0 5.0 6.3 1.5

2.8 2.9 4.0 5.0

–6.6 –6.2 1.9 –4.5 –4.6 1.6 –428.0 5.5 –782.6 –9.5 0.5 18.4 2.8 –4.6 7.5 4.5 –1.0 4.9 4.3 –2.5 8.6 1.6 –3.7 0.2 6.0 –5.0 –0.2 2.5 –12.4 2.0 –1.6 1.4 9.3 8.7 –3.0 5.1 –6.2 –1.9 4.2 –6.3 –1.2 11.7 –9.9 –6.3 –13.3

0.7 6.2 15.7 6.6 5.0 1.6 4.4 0.1 –1.0 3.4 2.0 0.9 2.3 7.5 12.7

0.0 1.8 –786.9 1.5 3.2 4.7 7.3 0.9 6.5 2.3 3.8 5.2 4.1 14.8 23.9

0.9 11.6 12.4 2.0 3.8 3.1 4.2 0.3 2.1 3.4 –0.5 1.2 –2.4 10.5 15.5

Notes: a. The period after the inflation targeting period refers to the period of 1993–2005; the period before the inflation targeting refers to the period of 1987–90. b. Treasury Bill: the period after the inflation targeting refers to the period of 1994–2000; CB Rate: the period after the inflation targeting refers to the period of 1994–95. c. The period before the inflation targeting refers to the period of 1994–97. d. Treasury Bill rates; the period after the inflation targeting refers to the period of 1998–2000. e. Treasury Bill: the period before the inflation targeting refers to the period of 1994–2000. f. Official adoption date for Turkey is 2006. However, Turkish central bank declared ‘disguised inflation targeting’ in the aftermath of the 2001 February crisis. g. A rise in value indicates depreciation. Annual average market rate is used for: UK, Canada, Turkey, Australia, New Zealand, Brazil, Peru, Israel, Indonesia, Korea and Philippines. Annual average Official Rate is used for: Colombia, Thailand, Hungary, Poland and Switzerland. Principle rate is used for: South Africa, Mexico and Czech Republic.

14

Beyond inflation targeting

Table 1.3 h. i. j.

(continued)

Nominal values are deflated by the corresponding inflation averages (CPI column). Sweden, New Zealand, Canada: Bank Rate; Mexico: Banker’s Acceptance. Colombia: Interbankaria TBS; Peru and Chile: Saving Rate; New Zealand Newly issued three months Treasury bill rates; Indonesia: three Months Deposit Rate; Korea: National Housing Bond Rate; Thailand: Government Bond Yield Rate.

Source:

IMF (2008).

extent Columbia, have pursued active export promotion strategies and maintained real depreciation. The Czech Republic, Switzerland and Hungary are observed to have experienced nominal currency appreciation, and Poland seems to have maintained an appreciating path for its real exchange rate. Clearly much of this generalized trend towards appreciation can be explained by reference to the increased expansion of foreign capital inflows due to the global financial glut mentioned above. With the IT central bankers announcing a ‘no-action’ stance against exchange rate movements led by the ‘markets’, a period of expansion in the global asset markets has generated strong tendencies for currency appreciation. What is puzzling, however, is the rapid and very significant expansion in the foreign exchange reserves reported by the IT central banks. As reported in the last two columns of Table 1.2, foreign exchange reserves held at the central banks rose significantly in the aftermath of the IT regimes. The rise of reserves was especially pronounced in Korea, the Philippines and Israel where almost a five-fold increase had been witnessed. Of all the countries surveyed in Table 1.3, the UK and Brazil are the only two countries that had experienced a fall in their aggregate reserves.3 This phenomenon is puzzling because the well-celebrated ‘flexibility’ of the exchange rate regimes were advocated precisely with the argument that, under the IT framework, the central banks would gain freedom in their monetary policies and would no longer need to hold reserves to defend a targeted rate of exchange. In the absence of any officially stated exchange rate target, the need for holding such sums of foreign reserves at the central banks should have been minimal. The proponents of the IT regimes argue that the central banks need to hold reserves to ‘maintain price stability against possible shocks’. Yet, the acclaimed ‘defense of price stability’ at the expense of such large and costly funds that are virtually kept idle at the IT central banks’ reserves is questionable at best in an era of prolonged unemployment and slow investment growth, and needs to be justified economically as well as socially. We now turn to the issue of exchange rate policy more formally.

Impacts and policy alternatives

1.3

15

THE ROLE OF THE EXCHANGE RATE UNDER INFLATION TARGETING

As stated above, part of the broader requirements surrounding the IT system is often argued to be the implementation of a ‘floating/flexible’ exchange rate system in the context of free mobility of capital. Thus, ‘exchange rate flexibility and floating exchange rate system’ became the new motto, and to many advocates, central bank ‘policy’ has typically been reduced to mean merely ‘setting the policy interest rate’. The exchange rate and macro prices are, in theory at least, thereby left to the unfettered workings of the global finance markets. The role of the exchange rate as an adjustment variable has clearly increased over the last decade since the adoption of the floating exchange rate systems. In the meantime, however, the role of the interest rates and reserve movements has declined substantially as counter-cyclical instruments available to be used against shocks4 (see Table 1.2). Against this background a number of practical and conceptual questions are inevitable: what is the role of the exchange rate in the overall macroeconomic policy when an explicit IT regime is adopted? Under what conditions should the central bank, or any other authority, react to shocks in the foreign exchange market? And perhaps more importantly, if an intervention in the foreign exchange market is regarded necessary against, say, the disruptive effects of an external shock, what are the proper instruments? To the proponents of IT, the answer to these questions is simple and straightforward: the central bank should not have any objective in mind with regards to the level of the exchange rate, yet it might interfere against the volatility of the exchange rate in so far as it affects the stability of prices. However nuances remain. To what may be grouped under ‘strict conformists’, the central bank should be concerned with the exchange rate only if it affects its ability to forecast and target price inflation. Any other response to the foreign exchange market represents a departure from the IT system. Advocated in the seminal works by Bernanke et al. (1999) and Fischer (2001), the approach argues that attending to IT and reacting to the exchange rate are mutually exclusive. Beyond this assertion, the conformist view also holds that intervention in the foreign exchange market could confuse the public regarding the ultimate objective of the central bank with respect to its priorities, distorting expectations. In a world of credibility game, such signals would be detrimental to the central bank’s authority. Yet, while maintaining the IT objective, one can also distill a more active role for the exchange rate in the literature. As outlined by Debelle

16

Beyond inflation targeting

(2001), this ‘flexible IT’ view proposes that the exchange rate can also be a legitimate policy objective alongside the inflation target. More formally, an operational framework for the ‘flexible IT’ view was envisaged within an expanded Taylor rule. Taylor (2000) argued, for instance, that an exchange rate policy rule can legitimately be embedded in a Taylor rule that is consistent with the broad objectives of targeted inflation rate and the output gap. In contrast to all this, the structuralist tradition asserts that irrespective of the conditionalities of foreign capital and boundaries of IT, it is very important for the developing economies to maintain a stable and competitive real exchange rate (SCRER) (see, for example, Cordero, Chapter 3, this volume; Frenkel and Taylor, Chapter 2, this volume; Galindo and Ros, Chapter 8, this volume; Frenkel and Ros, 2006; Frenkel and Rapetti, Chapter 9, this volume). They argue that the real exchange rate can affect employment, and the economy more generally, through a number of channels: (1) by affecting the level of aggregate demand (the macroeconomic channel); (2) by affecting the cost of labor relative to other goods and thereby affecting the amount of labor hired per unit of output (the labor intensity channel); and (3) by affecting employment through its impact on investment and economic growth (the development channel). While the size and even direction of these channel effects might differ from country to country, maintaining a competitive and stable real exchange rate is likely to have a positive employment impact through some combination of these effects. The gist of the structuralist case for SCRER rests on a recent (and unfortunately not well understood or appreciated) paper by Taylor (2004). Resting his arguments on the system of social accounting identities, Taylor argues that the exchange rate cannot be regarded as a simple ‘price’ determined by temporary macro equilibrium conditions. The mainstream case for exchange rate determination rests on the well-celebrated Mundell (1963) and Fleming (1962) model where the model rests on an assumed duality between reserves (fixed exchange rate system) versus flexible exchange rate adjustments. The orthodox mainstream model, according to Taylor, presupposes that a balance of payments exists with a potential disequilibrium that has to be cleared. This, however, is a false presumption. The exchange rate is not an ‘independent’ price and has no fundamentals such as a given real rate of return (or a trade deficit) that can make it self-stabilizing. In Taylor’s words, ‘the balance of payments is at most an accumulation rule for net foreign assets and has no independent status as an equilibrium condition. The Mundell-Fleming duality is irrelevant, and in temporary equilibrium, the exchange rate does not depend on how a country operates its monetary (especially international reserve) policy’ (Taylor, 2004, p. 212).

Impacts and policy alternatives

17

The preceding discussions clearly underscore that the real world behavior of exchange rates is quite complex and the focus of the inflation targeting regime for floating exchange rates (in expectation of dropping it from the policy agenda altogether) is a mirage. This view of exchange rates helps to explain why many believe that there are no viable alternatives to IT as a mode of central bank policy. However, as this book tries to demonstrate, and as we briefly discuss in the next section, this view of no viable alternatives to inflation targeting is not correct.

1.4

SOCIALLY RESPONSIBLE ALTERNATIVES TO INFLATION TARGETING CENTRAL BANK POLICIES

One reason that ‘inflation-focused monetary policy’ has gained so many adherents is the common perception that there is no viable alternative monetary policy that can improve growth and employment prospects. There are two main factors accounting for this perception. First, as we discussed in the previous section, in an internationally financially integrated economy with high levels of international capital flows, monetary policy can be extremely challenging. In particular it might be very difficult to gear monetary policy by targeting monetary aggregates, or by pegging an exchange rate along with trying to promote employment growth. This is often seen as the so-called ‘trilemma’ which commands that central banks can only have two out of three of the following: open capital markets, a fixed exchange rate system and an autonomous monetary policy geared toward domestic goals. While this so-called ‘trilemma’ is not strictly true as a theoretical matter, in practice it does raise serious issues of monetary management (see the above arguments cited from Taylor, 2004 and Frenkel and Taylor, 2006). From our perspective, the real crux of the problem turns out to be the very narrow interpretation of the constraints of the trilemma: central banks are often thought to be restricted to choose two ‘points’ out of three. Yet, the constraints of the trilemma could as well be regarded as the boundaries of a continuous set of policies, as would emerge out of a bounded, yet continuous depiction of a ‘policy triangle’. Thus even within the boundaries of the trilemma a menu of choices does exist, ranging from administered exchange rate regimes to capital management/control techniques. In fact, many successful developing countries have used a variety of capital management techniques to manage these flows in order, among other things, to help them escape the rigid constraints of the so-called ‘tri-lemma’ (Epstein et al., 2005; Ocampo, 2002). In this section we report on a series of country studies undertaken by a

18

Beyond inflation targeting

team of researchers working on a Political Economy Research Institute (PERI) (University of Massachusetts, Amherst)/Bilkent project on alternatives to inflation targeting, as well as a United Nations Development Programme (UNDP) sponsored study of employment targeting economic policy for South Africa. A range of alternatives were developed by the researchers, all the way from modest changes in the IT framework to allow for more focus on exchange rates and a change in the index of inflation used, to a much broader change in the overall mandate of the central bank to a focus on employment targeting, rather than IT. Some of the alternative policies focus exclusively on changes in central bank policy, while for other countries changes in the broad policy framework and in the interactions of monetary, financial and fiscal policy are proposed. Some incorporate explicit goals and targets, while others prefer more flexibility and somewhat less transparency. It has to be noted at the outset that ‘inflation control’ is revealed among the ultimate objectives in all country studies summarized below. Thus there is a clear consensus among the country authors that controlling inflation is important and desirable. However all agree that the current prescription insisting on ‘very low’ rates of inflation at the 2–4 percent band is not warranted, and that responsibilities of the central banks, particularly in developing countries, must be broader than that. Accordingly, the policy matrix of the central banks should include other crucial ‘real’ variables that have a direct impact on employment, poverty and economic growth, such as the real exchange rate and/or investment allocation. They also agree that in many cases, central banks must broaden their available policy tools to allow them to reach multiple goals, including, if necessary, the implementation of capital management techniques. Table 1.4 presents a summary of the alternatives proposed in the PERI/ Bilkent project and is discussed further in what follows. 1.4.1

Modest but Socially Responsible Adjustments to the Inflation Targeting Regime

Some of the country studies in the PERI/Bilkent project proposed only modest changes to the IT regime. In the case of Mexico, for example, the authors argue that the IT regime has allowed for more flexible monetary policy than had occurred under regimes with strict monetary targets or strict exchange rate targets (Galindo and Ros, Chapter 8, this volume). They suggest modifying the IT framework to make it somewhat more employment friendly. In the case of Mexico, Galindo and Ros find that monetary policy was asymmetric with respect to exchange rate movements – tightening when exchange rates depreciated, but not loosening when

19

Inflation control, export promotion, investment expansion

GDP growth, inflation control, export promotion

India

Inflation control, activity level and employment expansion, external sustainability

Argentina

Brazil

Ultimate targets

Country

Slightly undervalued (competitive) exchange rate

Discretion

Discretion

Discretion

SCRER, interest rate

SCRER

Strict target or discretion

Intermediate targets

Sterilization, reserve requirements (other prudential requirements), capital management techniques Interest rate, asymmetric managed float (moving floor on exchange rate), bank reserves, bank capital requirements, bank capital requirements Interest rate, capital management techniques, if necessary

Tools/instruments

Integrated

Integrated/ coordinated with the fiscal and antipoverty objectives

Coordinated

Central bank: independent, integrated or coordinated?

Table 1.4 PERI/Bilkent alternatives to inflation targeting project, summary of policy recommendations

20

Ultimate targets

Inflation, SCRER

Employment generation, inflation control

Inflation control; employment generation; solvency of public debt; consolidate and expand social infrastructure

Mexico

South Africa

Turkey

(continued)

Country

Table 1.4

Real interest rate, non-appreciated exchange rate

Implementation of ‘domestic’ inflation measure, SCRER, ‘sliding floor’ on exchange rate Real GDP growth

Intermediate targets

Capital management techniques if necessary, labor-tax reform, increased public investments in social capital

Integrated/ coordinated with the fiscal and employment objectives

Integrated

Interest rate; credit allocation techniques (e.g. asset-based reserve requirements, loan guarantees etc.), capital management techniques

Strict employment target (coordinated with other institutions), looser inflation constraint Discretion

Integrated

Capital management techniques

Discretion

Central bank: independent, integrated or coordinated?

Tools/instruments

Strict target or discretion

21

Incomes growth, export promotion, inflation control

Vietnam

SCRER

SCRER

Discretion

Discretion

Interest rate, capital management techniques, prudential supervision of banks, targeted credit; incomes policies Interest rate, capital management techniques, prudential supervision of banks, targeted credit, incomes policies Integrated

Integrated

Source:

IMF (2008).

Note: SCRER: Stable and Competitive Real Exchange Rate; Central Banks: integrated means integrated into governmental macroeconomic policy making framework; coordinated means independent but committed to close coordination with other macroeconomic policy making institutions.

Inflation control; solvency of public debt

Philippines

22

Beyond inflation targeting

exchange rates appreciated. This lent a bias in favor of an over-valued exchange rate in Mexico. So they propose a ‘neutral’ monetary policy so that the central bank of Mexico responds symmetrically to exchange rate movements and thereby avoids the bias toward over-valuation without fundamentally changing the IT framework. In their own words, ‘the central bank would promote a competitive exchange rate by establishing a sliding floor to the exchange rate in order to prevent excessive appreciation (an “asymmetric band”. . .). This would imply intervening in the foreign exchange market at times when the exchange rate hits the floor (i.e. an appreciated exchange rate) but allows the exchange rate to float freely otherwise’. They point out that such a floor would work against excessive capital inflows by speculators because they would know the central bank will intervene to stop excessive appreciation. If need be, Galindo and Ros also propose temporary capital controls, as do some of the other authors from the PERI/Bilkent project. In his study of Brazil Nelson Barbosa-Filho 2008 Chapter 7, (this volume) also proposed extending the IT framework, but in a more dramatic way. According to Barbosa-Filho: ‘because of Brazil’s past experience with high inflation, the best policy is to continue to target inflation while the economy moves to a more stable macroeconomic situation. However, the crucial question is not to eliminate inflation targeting, but actually make it compatible with fast income growth and a stable public and foreign finance’. Given Brazil’s large public debt, Barbosa-Filho proposes that the targeted reduction in the real interest rate would reduce the Brazilian debt service burdens and help increase productive investment. In terms of the familiar targets and instruments framework, he proposes that the Brazilian central bank choose exports, inflation and investment as ultimate targets, and focus on the inflation rate, a competitive and stable real exchange rate and the real interest rate as intermediate targets. Furthermore, in order to achieve these goals, the central bank can use direct manipulation of the policy interest rate, bank reserve requirements and bank capital requirements. Brazil is not the only highly indebted country in our project sample. Turkey is another case with that problem. Here too the authors raise concerns to the conformist straightjacket of IT, and develop an alternative macroeconomic framework. Using a financial-linked computable general equilibrium model (CGE) for the case of Turkey, Telli et al. (Chapter 10, this volume) illustrate the real and financial sectorial adjustments of the Turkish economy under the conditionalities of the twin targets: on primary surplus to GNP ratio and on the inflation rate. They utilize their model to study the impact of a shift in policy from a strict IT regime, to

Impacts and policy alternatives

23

one that calls for revisions of the primary fiscal surplus targets in favor of a more relaxed fiscal stance on public investments on social capital, together with a direct focus on the competitiveness of the real exchange rate. They further study the macroeconomics of a labor tax reform implemented through reduction of the payroll tax burden on the producers, and an active monetary policy stance via reduction of the central bank’s interest rates. They report significant employment gains due to a policy of lower employment taxes. They also find that the economy’s response to the reduction of the central bank’s interest rate is positive in general; yet very much dependent on the path of the real exchange rate, thus they also call for maintaining a stable real exchange rate path à la Frenkel, Ros and Taylor. Frenkel and Rapetti (Chapter 9, this volume), in the case of Argentina, show that targeting a stable and competitive real exchange rate has been very successful in helping to maintain more rapid economic growth and employment generation. In the case of India, Jha (Chapter 12, this volume) also argues against an IT regime, and in favor of one that ‘errs on the side of undervaluation of the exchange rate’ with possible help from temporary resort to capital controls. Jha argues that, to some extent, such a policy would be a simple continuation of policies undertaken in India in the past. In Vietnam Packard concludes: ‘a strict inflation targeting regime is not appropriate for Vietnam. Inflation targeting’s rigid rules constrain policy makers to operate in a framework that requires inflation to take priority over more pressing development objectives. Thus, a stable and competitive real exchange rate is a superior alternative, precisely because it sets as a target a key macroeconomic relative price that is realistic, sustainable and growth enhancing’ (Packard, Chapter 14, this volume). 1.4.2

More Comprehensive Alternatives to Inflation Targeting

Other country case studies propose more comprehensive policy alternatives to simple inflation-focused monetary policy, including IT. Joseph Lim (Chapter 13, this volume) proposes a comprehensive alternative to IT for the case of the Philippines. He argues that the Philippines’ government has been seeking to achieve a record of dramatically higher economic growth, but that its monetary policy has been inadequate to achieving that goal. He therefore proposes an ‘alternative’ that ‘clearly dictates much more than just a move from monetary targeting to inflation targeting’ with the following proposals: (1) Maintenance of a competitive real exchange rate, either by pegging the exchange rate or intensively managing it as in South Korea. (2) Implementation of capital management techniques, as in China and Malaysia, to help manage the exchange rates. This should

24

Beyond inflation targeting

include strong financial supervision to prevent excessive undertaking of short-term foreign debt, and tax based capital controls on short-term capital flows, as was used, for example in Chile. (3) An explicit statement of output and employment goals, as the central bank transits from a purely IT regime. (4) Incomes and anti-monopoly policies to limit inflation to moderate levels. (5) Targeted credit programs, especially for export oriented and small and medium sized enterprises that can contribute to productivity growth and employment. These policy proposals in broad outlines are similar to those proposed by Epstein (Chapter 11, this volume) for the case of South Africa, which, in turn, have been developed in a much broader framework and in more detail by Pollin et al. (2006). Pollin et al. developed an ‘employmenttargeted economic program’ designed to accomplish this goal, with a focus on monetary policy, credit policy, capital management techniques, fiscal policy and industrial policy. The purpose of the program is to reduce unemployment rate by half in line with the government’s pledge to reduce the official unemployment rate to 13 percent by 2014.5 Here, ‘employment targeting’ replaces IT as the proposed operating principle behind central bank policy, and moderate inflation becomes an additional formal constraint which the central bank must take into account when formulating its policies.

1.5

CONCLUDING COMMENTS

In this introductory chapter we have argued that the current day orthodoxy of central banking – namely, that the top priority goal for central banks is to keep inflation in the low single digits – is, in general, neither optimal nor desirable. This orthodoxy is based on several false premises: first, that inflation, in any magnitude, has high costs; second, that in a low inflation environment economies will naturally perform best, and in particular, will generate high levels of economic growth and employment; and third, that there are no viable alternatives to this ‘inflation-focused’ monetary policy. In fact, moderate rates of inflation episodes reveal to have very low or no costs; and whether countries where central banks have adopted formal or informal IT have not performed better in terms of economic growth or employment generation is a matter of dispute. Per contra, there are viable alternatives to IT, historically, currently, and looking forward. Historically, countries both in the currently developed and developing worlds had central banks with multiple goals and tools, and pursued broad developmental as well as stabilization goals. Currently, very

Impacts and policy alternatives

25

successful economies such as Argentina, China and India have central banks that are using a broad array of tools to manage their economies for developmental purposes. And looking forward, the PERI/Bilkent project on alternatives to IT and PERI’s UNDP work on South Africa have developed an array of ‘real targeting’ approaches to central banking which we believe are viable alternatives to IT and, in particular, do a better job than mere IT in balancing the developmental and stabilization functions of central banks.

NOTES 1. We are indebted to Hasan Comert, Luis Rosero and Lynda Pickbourn for their diligent research assistance, and to Roberto Frenkel, Jose Antonio Ocampo, K.S. Jomo, Geoffrey Woglom, Refet Gürkaynak, James Heintz, Leonce Ndikumana, Arjun Jayadev and Robert Pollin for their valuable comments and suggestions on previous versions of the chapter. Research for this chapter was completed when Yeldan was a visiting Fulbright scholar at the University of Massachusetts, Amherst for which he acknowledges the generous support of the J. William Fulbright Foreign Scholarship Board and the hospitality of the Political Economy Research Institute at the University of Massachusetts, Amherst. We are also grateful to the funders of the PERI/Bilkent ‘Alternatives to Inflation Targeting’ project, including UN-DESA, Ford Foundation, Rockefeller Brothers Fund and PERI for their support. Needless to mention, the views expressed in the chapter are solely those of the authors and do not implicate in any way the institutions mentioned above. 2. Note that with the use of the term ‘conditionality’ here we refer not to the IMF’s standby rules in the narrow sense of balance of payments stabilization, but to the broader set of reforms and structural adjustment agenda as advocated by the international finance community and the transnational corporations. Often dubbed as the (post-) Washington consensus, the warranted set of policies range from IT central banks and flexible foreign exchange markets to broader institutional reforms such as flexible labor markets, privatization and increased governance. See Williamson (1993) for the original deployment of the term, and Rodrik (2003) for further discussion. 3. Brazil’s case is actually explained in part by the recent decision (late 2005) of the Lula government to close its debt arrears with the IMF with early payments out of its reserves. 4. Though note the one sided ever increase in the aggregate reserves of the central banks. The social desirability and economic optimality of this phenomenon in the aftermath of the adoption of floating exchange rate systems is another issue that warrants further research. 5. As of March 2005, South Africa had an unemployment rate of anywhere from 26 percent to 40 percent, depending on exactly how it is counted.

REFERENCES Adelman, I. and E. Yeldan (2000), ‘The minimal conditions for a financial crisis: a multi-regional inter-temporal CGE model of the Asian crisis’, World Development, 28 (6), 1087–100.

26

Beyond inflation targeting

Akyuz, Y. (2003), The Developing Countries and World Trade. Performance and Prospects, London: ZED Books. Akyuz, Y. (2006), ‘From liberalization to investment and jobs: lost in translation’, paper presented at the Carnegie Endowment for International Peace Conference, 14-15 April 2005, Washington, DC, pp. 45–6. Akyuz, Y., H. Flassback and R. Kozul-Wright (2006), ‘Globalization, inequality and the labour market’, in A. Kose, F. Senses and E. Yeldan (eds), Neoliberal Globalization as New Imperialism: Case Studies on Reconstruction of the Periphery, New York: NOVA Science. Ball, L. and N. Sheridan (2003), ‘Does inflation targeting matter?’, International Monetary Fund working paper no. 03/129. Batini, N., P. Breuer, K. Kochhar and S. Roger (2006), ‘Inflation targeting and the IMF’, International Monetary Fund staff paper, Washington, DC, March. Bernanke, B.S., T. Laubach, A.S. Posen and F.S. Mishkin (1999), Inflation Targeting: Lessons from the International Experience, Princeton, NJ: Princeton University Press. Buiter, W. (2006), ‘Rethinking inflation targeting and central bank independence’, paper presented at the Turkish Economic Association Conference, Ankara, September. Corbo, V., O. Landerretche and K. Schmidt-Hebbel (2001), ‘Does inflation targeting make a difference? Central Bank of Chile working paper no. 106, accessed 15 January 2007 at www.bcentral.cl/Estudios/DTBC/doctrab.htm. Debelle, G. (2001), ‘The case for inflation targeting in East Asian countries’, in Future Directions for Monetary Policies in East Asia, Sydney: Reserve Bank of Australia, pp. 65–87. Edwards, S. (2005), ‘The relationship between exchange rates and inflation targeting revisited’, National Bureau of Economic Research, working paper no. W12163. Epstein, G., I. Grabel and K.S. Jomo (2005), ‘Capital management techniques in developing countries: an assessment of experiences from the 1990s and lessons for the future’, in Gerald Epstein (ed.), Capital Flight and Capital Controls in Developing Countries. Cheltenham, UK and Northampton, MA, USA: Edward Elgar, pp. 301–33. Feldstein, M. (1997), ‘Capital income taxes and the benefits of price stability’, National Bureau of Economic Research working paper no. 6200, September. Fischer, S. (2001), ‘Distinguished lecture on economics in government’, Journal of Economic Perspectives, 15 (2), 3–24. Fleming, J. Marcus (1962), ‘Domestic financial policies under fixed and under flexible exchange rates’, International Monetary Fund staff papers 9, p. 369–79. Freeman, R. (2004), ‘Doubling the global workforce: the challenges of integrating China, India, and the former Soviet Bloc into the world economy’, paper presented at the conference on ‘Doubling the Global Work Force’, Institute of International Economics, 8 November, Washington, DC. Freeman, Richard B. (2007), ‘The great doubling: is your job going to Bombay or Beijing?’, in America Works; Critical Thoughts an the Exceptional US Labor Market, New York: Russell Sage Foundation, pp. 128–40. Frenkel, R. and J. Ros (2006), ‘Unemployment and the real exchange rate in Latin America’, World Development, 34 (4), 631–46. Frenkel, R. and L. Taylor (2008), ‘Real exchange rate, monetary policy, and

Impacts and policy alternatives

27

employment’, United Nations Department of Economic and Social Affairs (DESA) working paper no 19, February, New York. Heintz, J. (2006), ‘Globalization, economic policy and employment: poverty and gender implications’, Employment Strategy Department employment strategy paper 2006/3, International Labour Organization, Geneva. International Labour Organization (ILO) (2004), Global Employment Trends 2004, Geneva: ILO. International Monetary Fund (IMF) (2008) International Financial Statistics, CD Rom, Washington, DC: IMF. Michl, Thomas (2007), ‘Tinbergen rules the Taylor rule’, Levy Economics Institute working paper no. 444. Mishkin, F.S. and K. Schmidt-Hebbel (2001), ‘One decade of inflation targeting in the world: what do we know and what do we need to know?’, National Bureau of Economic Research working paper no 8397, July accessed at www.nber.org/ papers/w8397. Mundell, R.A. (1963), ‘Capital mobility and stabilization policy under fixed and flexible exchange rates’, Canadian Journal of Economics and Political Science, 29, 475–85. Ocampo, J.A. (2002), ‘Capital-account and counter-cyclical prudential regulations in developing countries’, World Institute for Development Economics Research discussion paper, August. Pollin, R. and A. Zhu (2006), ‘Inflation and economic growth: a cross-country nonlinear analysis’, Journal of Post Keynesian Economics, 28 (4) (Summer), 593–614. Pollin, R., G. Epstein, J. Heintz and L. Ndikumana (2006), An Employment-Targeted Economic Program for South Africa, Cheltenham, UK and Northampton, MA, USA: Edward Elgar. Rodrick, D. (2003), ‘Growth strategies’, National Bureau of Economic Research working paper no. 10050. Roger, S. and M. Stone (2005), ‘On target? The international experience with achieving inflation targets’, International Monetary Fund working paper WP/05/163, Washington, DC. Setterfield, M. (2006), ‘Is inflation targeting compatible with post keynesian economics?’, Journal of Post Keynesian Economics, 28 (4), 653–71. Taylor, J.B. (2000), ‘Low inflation, pass-through and the pricing power of firms’, European Economic Review, 7 (44), 1389–408. Taylor, L. (2004), ‘Exchange rate indeterminacy in portfolio balance, MundellFleming and uncovered interest parity models’, Cambridge Journal of Economics, 28, 205–27. Tobin, J. (1980), ‘Stabilization policy ten years after’, Brookings Papers on Economic Activity, 1, 19–71. Williamson, J. (1993), ‘Democracy and the Washington Consensus’, World Development, 21 (8), 1329–36.

2.

Real exchange rate, monetary policy and employment: economic development in a garden of forking paths Roberto Frenkel and Lance Taylor

Dejo a los varios porvenires (no a todos) mi jardín de senderos que se bifurcan. [I am leaving my garden of diverse paths to some (but not all)] Jorge Luis Borges, ‘El jardín de los senderos que se bifurcan’

2.1

INTRODUCTION

The exchange rate affects any economy through many channels. It scales the national price system to the world’s, influences key macro price ratios such as those between tradable and non-tradable goods, capital goods and labor, and even exports and imports (via the costs of intermediate inputs and capital goods, for example). The exchange rate is an asset price, partially determines inflation rates through the cost side and as a monetary transmission vector, and can have significant effects (both short and long run) on effective demand. Correspondingly the exchange rate can be targeted toward many policy objectives. In developing and transition economies five have been of primary importance in recent decades: 1.

Resource allocation: through its effects on the price ratios just mentioned, the exchange rate can significantly influence resource allocation, especially if it stays stable in real terms for an extended period of time. Through effects on both resource allocation and aggregate demand, a relatively weak rate can help boost employment, a point of concern in light of stagnant job creation in many developing economies over the past 10–15 years. 2. Economic development: often in conjunction with commercial and industrial policies, the exchange rate can be deployed to 28

Real exchange rate, monetary policy and employment

3.

4.

5.

29

enhance overall competitiveness and thereby boost productivity and growth. Finance: the rate shapes and can be used to control expectations and behavior in financial markets. Exchange rate policy ‘mistakes’ can easily lead to highly destabilizing consequences. External balance: the trade and other components of the current account usually respond to the exchange rate, directly via ‘substitution’ responses and (at times more importantly) to shifts it can cause in effective demand. Inflation: the exchange rate can serve as a nominal anchor, holding down price increases via real appreciation and/or maintenance by the authorities of a consistently strong rate. As will be seen below, it can also serve as an important transmission mechanism for the effects of monetary policy.

All these objectives have figured in recent policy experience. Use of the exchange rate to try to improve external balance has been central to countless stabilization packages over the decades, especially in small poor economies. The inflation objective became crucial in middleincome countries in the last quarter of the twentieth century (and is notably less urgent as of 2005). Along with capital market liberalization, fixed rates were significant contributors to the wave of financial crises in the 1990s. But in many ways the resource allocation and developmental objectives can be the most important in the long run – the central point of this chapter. We trace the reasons why in the following section on channels of influence. We then take up the policy implications, contrasting the use of the exchange rate as a development tool in conjunction with its other uses (often in coordination with monetary policy) to maintain external balance, contain inflation and stabilize asset markets,

2.2

RESOURCE ALLOCATION, LABOR INTENSITY, MACROECONOMICS AND DEVELOPMENT

Following Frenkel (2004), in this section we trace out three ways in which the exchange rate can have medium- to long-term impacts on development. We begin with overall resource allocation, and go on to the labor market and macroeconomics.

30

2.2.1

Beyond inflation targeting

Resource Allocation

The traditional 2 3 2 trade theory model is a useful starting point. It does focus on the key role of relative prices. It does not take into consideration important non-price components of industrial and commercial polices. Both themes are woven into the following discussion. The Lerner Symmetry Theorem (1936) is a key early result. Its basic insight is that if only the import/export price ratio is relevant to resource allocation, then it can be manipulated by either an import or an export tax-cum-subsidy. There is ‘symmetry’ between the two instruments, so that ‘under appropriate conditions’ (at hand in the textbooks) only one need be employed. A now obvious extension is to bring three goods into the discussion: exportable, importable and non-tradable in a ‘Ricardo-Viner’ model. Two price ratios – say importable/non-tradable and exportable/non-tradable – in principle guide allocation. The real exchange rate (RER or r) naturally comes into play as the relative price between the non-tradable and a Hicksian aggregate of the two tradable goods.1 These observations lead to two important policy puzzles. The first has to do with ‘level playing fields’. As applied in East Asia and elsewhere, industrial policy often involved both protection of domestic industry against imports by the use of tariffs and quotas, and promotion of exports through subsidies or cheap credits. In the case of a tariff on imports, the domestic price Pm becomes Pm 5 e (1 1 t) P*m

(2.1)

with e as the nominal exchange rate (defined as units of local currency per unit of foreign), t the tariff, and P* m the world price. Similarly if the internal price Pe for exports is set from abroad we have Pe 5 eP*e / (1 2 s)

(2.2a)

with P*e as the world price and s as the subsidy rate. The level playing field rests on the trade theorists’ notion that internal and external relative prices of tradable goods should be equal, Pm /Pe 5 P* m /P* e . This situation can be arranged if t 5 s 5 0 or more generally (1 1 t) 5 1/ (1 2 s) . The mainstream argument asserts that if all that industrial policy does is give more or less equal protection to both imports and exports, then its costs, administrative complications, and risks of rent-seeking and corruption are unjustifiable. You might as well set t 5 s 5 0 and go to a free trade equilibrium.

Real exchange rate, monetary policy and employment

31

In a Ricardo-Viner set-up, with Pn as a price index for non-tradables the price ratios Pe /Pn and Pm /Pn become of interest. Positive values of t and s move domestic relative prices in favor of tradable goods. From a more or less mainstream perspective (Woo, 2005) this outcome can be interpreted as a justification for industrial policy. The world, however, is a bit more complicated. If the home country is exporting a differentiated product, for example, a more appropriate version of Equation (2.2a) is P*e 5 Pe (1 2 s) /e

(2.2b)

so that the foreign price of home exports is set by the subsidy and exchange rate. Presumably, a lower value of P*e stimulates sales abroad. Moreover, if the economic bureaucracy has the requisite motivation and organization, it can tie export subsidies to the attainment of export, productivity and other targets and so pursue a proactive industrial policy. In such a context, import protection and export promotion serve different purposes: the former allows domestic production to get started along traditional infant industry lines, while the latter enables national firms to break into international markets.2 Now focus on the exchange rate. An increase in the nominal rate e would also switch incentives toward production of tradables, without the need for extravagant values of s and t. This simple observation is in fact a strong argument in support of the use of a depreciated RER as a developmental tool. If we define r as r 5 [ mPm 1 (1 2 m) Pe ] /Pn

(2.3)

with m as the weight in a tradable goods price index, then a high value of e means that the RER r will also be weak or depreciated. Of course, a weak RER may not be a sufficient condition for long-term development. For example, it may usefully be supplemented by an export subsidy or tariff protection to infant industries with their additional potential benefits as mentioned above. Even without an effective bureaucracy, generalizing Lerner symmetry to a Ricardo-Viner world suggests that more than one policy instrument may be helpful because there are two relative price ratios that can be manipulated (Lerner, 1936). The rub is that a strong exchange rate implies that commercial/industry policy interventions also have to be strong, with correspondingly high intervention costs. A weak RER may be only a necessary condition for beneficial resource reallocation to occur, but a highly appreciated RER is likely to be a sufficient condition for ‘excessive intervention’ in a situation in which

32

Beyond inflation targeting

development cannot happen. It is hard to find examples of economies with strong exchange rates that kept up growth for extended periods of time. 2.2.2

Labor Intensity

Continuing with the allocational theme, it is clear that the exchange rate will affect relative prices of imported intermediates and capital goods on the one hand, and labor on the other. Moreover, the RER largely determines the economy’s unit labor costs in terms of foreign currency. To explore the implications, we can consider the effects of sustained real appreciation on different sectors. Producers of importables will face tougher foreign competition. To stay in business they will have to cut costs, often by shedding labor. If they fail and close down, more jobs will be destroyed. If home’s export prices P*e are determined by a relationship like Equation (2.2b), similar logic applies to that sector. In non-tradables, which will have to absorb labor displaced from the tradable sectors, jobs are less likely to open up insofar as cheaper foreign imports in the form of intermediates and capital goods substitute for domestic labor. On the whole, real appreciation is not likely to induce sustained job creation and could well provoke a big decrease in tradable sector employment. Reasoning in the other direction, RER depreciation may prove employment-friendly. In both cases it is important to recognize that a new set of relative prices must be expected to stay in place for a relatively long period if these effects are going to work through. Changes in employment/output ratios will not happen swiftly because they involve restructuring firms and sectoral labor market behavior. This must take place via changes in the pattern of output among firms and sectors, by shifts in the production basket of each firm and sector, and adjustments in the technology and organization of production. These effects arise from a restructuring process in which individual firms and the organization of economic activity adapt to a new set of relative prices. Gradual adjustment processes are necessarily involved. Finally, in the long run if per capita income is to increase there will have to be sustained labor productivity growth with employment creation supported by even more rapid growth in effective demand. Macroeconomics comes into play. 2.2.3

Macroeconomics

The question is how a weak exchange rate (possibly in combination with other policies aimed at influencing resource allocation among traded goods) fits into the macroeconomic system. Much depends on labor

Real exchange rate, monetary policy and employment

33

market behavior in the non-traded sector. Following Rada (2005) we work through one scenario here, to illustrate possible outcomes. Assume that output in the tradable sector is driven by effective demand, responding to investment, exports and import substitution as well as fiscal and monetary policy. The level of imports depends on economic activity and the exchange rate (along with commercial/industrial policies). A worker not utilized in tradable sectors must find employment in nontradables, become under- or unemployed, or leave the labor force. For concreteness, we assume that almost all labor not employed in tradables finds something to do in non-tradable production as a means of survival. Typical activities would be providing labor services in urban areas or engaging in labor-intensive agriculture. If Lt is tradable sector employment and L is the economically active population, then employment in non-tradables is Ln 5 L 2 Lt. With wn as the non-tradable wage, the value of labor services provided is Yn 5 wnLn. The tradable sector wage rate wt is determined institutionally, at a level substantially higher than wn. The non-tradable sector’s demand-supply balance thus takes the form Yn 2 wnLn 5 Yn 2 wn (L 2 Lt) 5 0

(2.4)

Demand for Yn is generated from the value of tradable sector output PtXt. At the same time, real output Xt determines Lt and thereby Ln. Suppose that Pt is set by mark-up pricing on variable costs including labor and imports. Then from both the demand and supply sides an increase in Xt leads to a tighter non-traded labor market which should result in an increase in wn. Equation (2.4) becomes the upward-sloping ‘Non-tradable equilibrium’ schedule in Figure 2.1. Non-tradable labor services become more valuable when economic activity rises. In national accounting terms this signals a productivity increase in the sector because each worker producers a higher value of output in terms of tradable goods, or a general price index. In other words, an endogenous productivity level is built into the specification. If workers in both sectors don’t save, then their behavior does not influence overall macroeconomic balance. Leaving aside a formal treatment of fiscal and monetary instruments for simplicity, the equation takes the form It 1 Et 2 spXt 2 eP*m (1 1 t) aXt /Pt 5 0

(2.5)

Demand injections come from investment It, exports Et and changes in the magnitude of the import coefficient a via import substitution. Saving leakages come from profits with p as the tradable sector profit share and

34

Beyond inflation targeting Non-tradable sector wage wn

Macroeconomic equilibrium Non-tradable equilbrium

Tradable sector output Xt

Trade deficit

Figure 2.1

Equilibrium between tradable and non-tradable sectors

s the saving rate as well as from ‘foreign saving’ in the form of imports. Equation (2.5) is the vertical ‘Macroeconomic equilibrium’ line in Figure 2.1. Together, the two schedules determine Xt and wn. In the lower quadrant the trade deficit is assumed to be an increasing function of tradable sector output in the short run. Now consider the outcomes of a devaluation. It will have impacts all over the economy, including a loss in national purchasing power if imports initially exceed exports, redistribution of purchasing power away from low-saving workers whose real wages decrease, a decline in the real value of the money stock and capital losses on the part of net debtors in international currency terms. Presumably exports will respond positively to an RER depreciation but that may take time if ‘J-curve’ and similar effects matter. Another positive impact on the demand for tradables will

Real exchange rate, monetary policy and employment Labor productivity growth rate ξ Lt

35

Employment growth contours Output growth

Kaldor-Verdoorn

Output growth rate Xˆ t

Figure 2.2

Output, labor productivity and employment growth in the tradable sector

come from import substitution, reducing the magnitude of the coefficient a. One implication is that for a given level of output, the trade deficit should fall with devaluation, or the corresponding schedule should shift toward the horizontal axis in the lower quadrant. If devaluation is contractionary, the macroeconomic equilibrium schedule will shift leftward in the upper quadrant, reducing Xt, wn, and the trade deficit further still. In this case real devaluation should presumably be implemented together with expansionary fiscal and monetary policies. As discussed in detail below, exchange rate strategies must be coordinated with other policy moves. If export demand and production of import substitutes are stimulated immediately or over time by a sustained weak RER, the macroeconomic equilibrium curve should drift to the right, driving up economic activity and employment in the medium to long run. So far, the analysis has taken labor productivity as a constant. Mediumand long-run considerations have to take into account the evolution of productivity. For the tradable sector, this question can be analysed in terms of Figure 2.2, sketched verbally but not actually drawn by Kaldor in his 1966 Inaugural Lecture (published in Kaldor, 1978). To the traditional diagram we follow Rada and Taylor (2004) by adding dashed

36

Beyond inflation targeting

‘Employment growth contours’ with slopes of 45 degrees. Each one # shows combinations of the output growth rate (X^ t 5 (dXt /dt) /Xt 5 Xt /Xt) and labor productivity growth rate (xLt) that hold the employment growth rate (L^ t 5 X^ t 2 xLt) constant. Employment growth is more rapid along contours further to the south east. Movements across contours show the effects on employment growth of shifts in the diagram’s two solid curves. The ‘Kaldor-Verdoorn’ schedule represents a ‘technical progress’ function of the form proposed by Verdoorn (1949) and Okun (1962), xLt 5 xLt 1 gX^ t

(2.5)

in which the productivity trend term xLt could be affected by human capital growth, industrial policy, international openness, population growth and other factors. The ‘Output growth’ curve reflects the assumption that more rapid productivity growth can make output expand faster, for example, by reducing the unit cost of exports. The diagram presupposes that this effect is rather strong because the slope of the ‘Output growth’ line is less than 45 degrees, implying that 0X^ t /0xLt . 1. If a depreciated RER stimulates net export growth, the ‘Output growth’ $ curve will shift to the right, causing xLt, Xt and L^ t all to increase. One might also imagine that the trend rate xLt of productivity growth could rise in the new regime. The ‘Kaldor-Verdoorn’ schedule would shift upward, and with a relatively flat output growth curve, all three growth rates would rise. However, if the slope of the ‘Output growth’ curve were to exceed 45 degrees, effective demand would not increase as rapidly as productivity so that L^ t would have to fall. What happens to wages and productivity in the non-tradable sector? Let l 5 Lt /L be the share of tradable sector employment in the total. Then lL^ t 1 (1 2 l) L^ n 5 L^ where L^ is overall employment growth. Nontradable employment expansion becomes L^ n 5

1 [ L^ 2 l (X^ t 2 xLt) ] . 12l

(2.6)

Let the elasticity of demand for non-tradables with respect to Xt be u. Differentiating Equation (2.4) then gives w^ n 5 uX^ t 1

1 [ l (X^ t 2 xLt) 2 L^ ] 12l

(2.7)

Even taking into account the favorable effects on employment of a weak exchange rate that were mentioned above, a low demand elasticity u and

Real exchange rate, monetary policy and employment

37

fast labor force growth L^ could mean that a strong export performance translates into weak or even negative wage and productivity growth in the non-traded sector. A case like this calls for fiscal and social policies intended to foster demand for non-tradables and compensate for the negative effects on income distribution and employment.

2.3

MACROECONOMIC POLICY REGIMES FOR A STABLE COMPETITIVE REAL EXCHANGE RATE

For the reasons just indicated, a competitive and stable RER can make a substantial contribution to economic growth and employment creation. Programming the RER, however, is no easy task. It is most directly impacted by the nominal exchange rate, itself influenced by many factors, but also depends on the overall inflation rate and shifting relative prices. Nor can the RER be the only macro policy objective. In any economy there are bound to be multiple and partially conflicting objectives. And all policies – exchange rate, fiscal, monetary and commercial/industrial – are interconnected and have to be coherently designed and implemented. The following discussion focuses on these exchange rate coordination issues in the context of middle income economies with at least sporadic access to private international capital markets. Although they are not addressed in detail here, somewhat similar questions can easily arise in low income countries receiving official capital inflows, especially if they jump to levels of 10–20 percent of GDP as suggested in the discussion of the Millenium Development Goals (MDG). So how can policy makers target the RER while at the same time controlling inflation, reducing financial fragility and risk, and aiming toward full employment of available resources? Our focus necessarily has to shift from the ‘real economy’ to encompass monetary and expectational considerations. The principal emphasis is on the degrees of freedom available to the monetary authorities, if only because they have been at center stage in recent policy debates. 2.3.1

What Determines the Nominal Exchange Rate?

To set the stage, a few observations about how the nominal exchange rate fits into the macroeconomic system make sense. Theories that can reliably predict the level of the rate and its changes over time when it is not strictly pegged do not exist. (The fact that pegs not infrequently break down means they do not have 100 percent predictive

38

Beyond inflation targeting

power either.) In present circumstances in developing and transition economies (especially those at middle income levels) it is not unreasonable to assume that a more or less floating rate is determined in spot and future asset markets; in effect the spot rate floats against its ‘expected’ future values. The quotation marks mean that we view expectations along Keynesian lines as emerging from diverse opinions on the part of market participants about how the rate may move. ‘Beauty contests’ which magnify small shifts in average market opinion and other sources of seemingly capricious market behavior can easily come into play (Eatwell and Taylor, 2000). With regard to the level of the rate, it is useful to think about a simple bond market equilibrium condition such as # i 5 f (e, eexp, M)

(2.8)

# with i as the local interest rate, e the spot exchange rate, eexp the expected (as an aggregate of by market perceptions) change in the rate over time and M an index of monetary relaxation. A high or depreciated value of e means that national liabilities are cheap as seen from abroad. It should be associated with high local bond prices or low interest rates. If expected # depreciation eexp rises, on the other hand, foreign wealth-holders will want to shift away from local liabilities and i will increase. Open market bond purchases will increase M and be associated with a reduction in i. Over the past couple of decades under conditions of external liberalization, most developing economies have been afflicted by high local interest rates and appreciated currencies. This unfavorable constellation of ‘macro prices’ is consistent with Equation (2.8). The dynamics of the exchange rate will be influenced by interest rates, because it is an asset price. One crucial question is whether lower domestic rates will tend to make the nominal rate depreciate or appreciate. If it tends to rise (or depreciate) over time, then exchange rate dynamics can be a powerful mechanism for transmitting the effects of expansionary monetary policy into inflation by driving up local production costs. Standard arbitrage arguments as built into interest rate parity theorems # imply that the expected change in the spot rate eexp should be an increasing function of the difference between domestic and foreign rates. If myopic perfect foresight applies, the expected change will be equal to the observed change (up to a ‘small’ error term). Hence a lower local interest rate should cause appreciation over time. On Wall Street, such an analysis of exchange rate movements is called an ‘operational’ view. A ‘speculative’ view is that the exchange rate will depreciate when the local interest rate decreases.3 This view makes intuitive sense insofar as low

Real exchange rate, monetary policy and employment

39

interest rates should make national liabilities less attractive. It was perhaps first advanced macroeconomically by Minsky (1983) and can be made consistent with the parity theorems if it is assumed that there is a relatively strong positive feedback of expected exchange rate increases into the domestic interest rate via the bond market equilibrium condition Equation (2.6). Recent macroeconomic history (Frenkel, 2004) suggests that the speculative view is the more accurate description of exchange rate behavior in middle income economies. 2.3.2

Avoiding Catastrophes

The most fundamental justification for avoiding a persistently strong exchange rate is that it is an invitation to disaster. Exchange appreciation is always welcome politically because it may be expansionary (at least in the short run), is anti-inflationary and reduces import costs (including foreign junkets for those who can afford them). However, for the reasons discussed above, it can have devastating effects on resource allocation and prospects for development. Moreover, fixed or quasi-fixed strong real rates can easily provoke destabilizing capital flow cycles as perhaps first described analytically by Frenkel (1983) and re-enacted many times since. The existence and severity of these cycles is in practice a powerful argument for a stable exchange rate regime built around some sort of managed float (details below). A floating rate does appear to moderate destabilizing capital movements in the short run, and is therefore a useful tool to deploy. At the same time, the central bank has to prevent the formation of expectations that there will be RER appreciation, which can easily become self-fulfilling along beauty contest lines. A commitment to a stable rate, backed up by forceful intervention if necessary, is one way the bank can orient expectations around a competitive RER. 2.3.3

Trilemmas

Possibilities for central bank intervention are often said to be constrained by a ‘trilemma’ among (1) full capital mobility, (2) a controlled exchange rate and (3) independent monetary policy. Supposedly, only two of these policy lines can be consistently maintained. If the authorities try to pursue all three, they will sooner or later be punished by destabilizing capital flows, as in the run-up to the Great Depression around 1930 and Britain and Italy’s difficulties during the European Exchange Rate Mechanism (ERM) crisis more than 60 years later. The trilemma as just stated is a textbook theorem which is, in fact, invalid. Even with free capital mobility, a central bank can undertake

40

Beyond inflation targeting

transactions in both foreign and domestic bonds (not to mention other monetary control maneuvers) to regulate the money supply, regardless of whatever forces determine the exchange rate (Taylor, 2004). Nevertheless, something like a trilemma can exist in the eye of a beholder. There are practical limits to the volume of interventions that a central bank can practice, along with complicated feedbacks. Possibilities for sterilizing capital inflows or outflows are bounded by available asset holdings. Volumes of flows depend on exchange rate expectations which in turn can be influenced by central bank behavior and signaling. So how does the market decide when a perceived trilemma is ripe to be pricked? The fact that no single form of transaction or arbitrage operation determines the exchange rate means that monetary authorities have some leeway in setting both the scaling factor between their country’s price system and the rest of the world’s and the rules by which it changes. However, their sailing room is not unlimited. A fixed rate is always in danger of violating what average market opinion regards as a fundamental. Even a floating rate amply supported by forward markets can be an invitation to extreme volatility. Volatility can lead to disaster if asset preferences shift markedly away from the home country’s liabilities in response to shifting perceptions about fundamentals or adverse ‘news’. Unregulated international capital markets are at the root of any perceived trilemma. It is a practical problem that must be evaluated in each case, taking into account the context and circumstances of policy implementation. 2.3.4

Monetary and Exchange Rate Policies and Capital Flows

The implication is that if it wishes to target the RER, the central bank has to maintain tolerable control over the macroeconomic impacts of cross-border financial flows in a world with relatively open foreign capital markets. For the sake of clarity, it makes sense to analyse situations of excess supply and excess demand for foreign capital separately. Large capital inflows can easily imperil macro stability. Indeed, central bank attempts to sterilize them by selling domestic liabilities from its portfolio may even bid up local interest rates and draw more hot money. Preservation of monetary independence in this case may well require capital market regulation. Measures are available for this task.4 They do not work perfectly, but can certainly moderate inflows during a boom. Booms never last forever; the point is that the authorities can use capital market interventions to slow one down to avoid an otherwise inevitable crash. If there are capital outflows too large to manage with normal exchange rate and monetary policies, the authorities certainly do not want to engage in recession-triggering monetary contraction. If the exchange rate has been

Real exchange rate, monetary policy and employment

41

maintained at a relatively weak level, the external deficit is not setting off financial alarm bells, and inflation is under control, then there are no ‘fundamental’ reasons for market participants to expect a maxi-devaluation. Under such circumstances the way for the authorities to maintain a policy regime consistent with a targeted RER is to impose exchange controls and restrictions of capital outflows. Contrary to IMF-style opinion that all runs against a currency must be triggered by poor fundamentals (even if they momentarily escape the notice of the authorities and IMF officials), it is perfectly clear that they can arise for reasons extraneous to economic policy – think of a political crisis, the fallout from mismanagement of an important bank or the impacts of financial contagion from a regional neighbor. In all such cases outflow controls can be used to maintain an existing policy package in place. They may not have to be utilized for very long.5 2.3.5

Monetary Policy

In a developmental policy regime, monetary policy must be designed in view of its likely effects on the RER, inflation control and the level of economic activity. There is nothing very surprising here – in practice central banks always have multiple objectives. In the USA, despite lip service to controlling price inflation, the Federal Reserve certainly responds to the level of economic activity and financial turmoil (witness the 1990s stock market bubble and the long-term capital management (LTCM near-crisis). In many developing countries central banks intervene more or less systematically in the exchange markets. The proposal here is that these interventions should help support a developmentally oriented RER for the reasons presented above. That is, the nominal rate should move to hold the RER in the vicinity of a stable competitive level for an extended period of time. Inflation targeting, on the other hand, is the current orthodox buzzword. The nominal exchange rate and other policies should be programmed to ensure a low, stable rate of inflation. A trilemma-like argument is involved. If exchange market interventions target the RER as opposed to the nominal exchange rate and the central bank cannot manage the money supply, there is no nominal anchor on inflationary expectations. The inflation rate cannot be controlled. As we have seen, in practical terms the trilemma can be circumvented, allowing the monetary authorities to bring developmental objectives into their remit. But they have to take at least five important considerations into account in monetary management. First, many developing countries now have low to moderate inflation rates, demoting inflation control in the hierarchy of policy objectives.

42

Beyond inflation targeting

Second, will low interest rates tend to set off inflationary nominal depreciation (under ‘speculative’ exchange rate dynamics as discussed above)? RER targeting can help the central bank steer away from this problem. Third, shifts in aggregate demand likely to result from changes in the exchange rate and monetary policy must be taken into account, and appropriate offsetting policies deployed. Fourth, also as mentioned above, some mix of temporary capital inflow or outflow controls may be needed to allow the central bank to regulate monetary aggregates and interest rates rather than be overwhelmed by attempts at sterilization. Finally, unstable money demand and other unpredictable factors mean that the monetary authorities have to be alert and flexible. Indeed, ‘inflation targeting’ is a codeword for orthodox recognition that quantitative monetary and even interest rate targets are impractical. It is a means for giving more discretion in trying to attain a single target. The point being made here is that discretion can and should serve other ends. A stable competitive RER in coordination with sensible industrial and commercial policies can substantially improve prospects for economic development. Surely that should be the overriding goal of the monetary and all other economic authorities in any developing or transition economy.

NOTES 1. Just to be clear, we will treat the RER as the ratio of tradable to non-tradable price indexes. Real devaluation or weakening the RER means that r increases. 2. Again, these arguments are old. Ocampo and Taylor (1998) provide a recent summary. # 3. To be more precise, the change over time in the spot rate e 5 de/dt will turn negative when i decreases if the operational view applies and positive when the speculative view is true. 4. For an ample menu, see papers by Deepak Nayyer, Eric Helleiner and Gabriel Palma in Eatwell and Taylor (2002) and Epstein et al. (2003). Salih Neftci and Randall Dodd assess the possibilities of using financial engineering to circumvent controls. 5. Argentina, for example, successfully managed exchange controls and capital outflow restrictions in mid 2002. The measures were transitory. They were gradually softened as buying pressure in the exchange market diminished.

REFERENCES Eatwell, J. and L. Taylor (2000), Global Finance at Risk: The Case for International Regulation, New York: The New Press Eatwell, J. and L. Taylor (eds) (2002), International Capital Markets: Systems in Transition, New York: Oxford University Press.

Real exchange rate, monetary policy and employment

43

Epstein, Gerald, Ilene Grabel and K.S. Jomo (2003), ‘Capital management techniques in developing countries’, in Ariel Buira (ed.), Challenges to the World Bank and IMF; Developing Country Perspectives, London: Anthem Press, reprinted in Gerald Epstein (ed.) (2005), Capital Flight and Capital Controls in Developing Countries, Cheltenham, UK and Northampton, MA, USA: Edward Elgar. Frenkel, R. (1983), ‘Mercado Financiero, expectativas cambiales, y movimientos de capital’, El Trimestre Economico, 50, 2041–76. Frenkel, R. (2004), ‘Real exchange rate and employment in Argentina, Brazil, Chile, and Mexico’, paper prepared for the Group of 24, Washington, DC: September. Kaldor, N. (1978), ‘Causes of the slow rate of growth of the United Kingdom’, in Further Essays on Economic Theory, London: Duckworth. Lerner, A.P. (1936), ‘The symmetry between import and export taxes’, Economica, 3, 306–13. Minsky, H.P. (1983), ‘Monetary policies and the international financial environment’, Washington University Department of Economics mimeo, St Louis, MO. Ocampo, J.A. and L. Taylor (1998), ‘Trade liberalization in developing economies: modest benefits but problems with productivity growth, macro prices, and income distribution’, Economic Journal, 108, 1523–46. Okun, A.M. (1962), ‘Potential GNP: its measurement and significance’, reprinted in Joseph Pechman (ed.) (1983), Economics for Policy-Making, Cambridge, MA: MIT Press. Rada, C. (2005), ‘A growth model for a two-sector open economy with endogenous employment in the subsistence sector’, Schwartz Center for Economic Policy Analysis, New School University, New York. Rada, C. and L. Taylor (2004), ‘Empty sources of growth accounting, and empirical replacements à la Kaldor with some beef’, Schwartz Center for Economic Policy Analysis, New School University, New York. Taylor, L. (2004), ‘Exchange rate indeterminacy in portfolio balance, MundellFleming, and uncovered interest rate parity models’, Cambridge Journal of Economics, 28, 205–27. Verdoorn, P.J. (1949), ‘Fattori che Regolano lo Sviluppo della Produttivita del Lavoro’, L’Industria, 1, 3–10. Woo, W.T. (2005), ‘Some fundamental inadequacies in the Washington Consensus: misunderstanding the poor by the brightest’, in Jan Joost Teunissen (ed.), Stability, Growth, and the Search for a New Development Agenda: Reconsidering the Washington Consensus, The Hague: Forum on Debt and Development, (FONDAD).

3.

Inflation targeting and the real exchange rate in a small economy: a structuralist approach Jose Antonio Cordero

3.1

INTRODUCTION

In the past few years several nations, especially encouraged by the International Monetary Fund (IMF), have decided to move into a monetary regime based on explicit inflation targets. This regime has been strongly recommended to economies that were struggling to bring their inflation rates down to the level of the more advanced countries. The application of this regime has been very effective in reducing inflation but, as argued in Epstein and Yeldan (2008), the final effects on employment and economic growth are not so clear. A rigorous analysis of the consequences of inflation targeting in open economies requires a formal model exploring the interaction among money, the foreign sector, output and employment. Within the structuralist tradition several models have been developed in order to examine growth and distribution in open economies. These works, however, do not always include the monetary sector and, when they do, they rarely establish the links between the latter and the balance of payments; they are not designed to examine the association between the monetary regime and the foreign exchange rate system. This chapter attempts to fill that void by means of a framework that allows comparing two alternative monetary regimes: one resulting from inflation targeting, and another resulting from real exchange rate targeting. The model allows formalizing the adjustment process in three periods: first, output and employment vary with given prices, second, inflation clears an explicit monetary sector and, finally, capital accumulation, income distribution and the real exchange rate reach a stable steady state. The growth of labor productivity, through learning effects, leads to endogenous growth. Inflation is explained here as a monetary phenomenon, but effective 44

Inflation targeting and the real exchange rate in a small economy

45

demand remains a critical determinant of output, employment and economic growth. The chapter thus also contributes to a better understanding of the interaction between the real and the monetary sector. In Section 3.2 the central bank adjusts the nominal exchange rate to hit a real exchange rate target, with inflation control playing an important but secondary role. In this case the growth and employment performance are favorable but the economy suffers the difficulties arising from the ‘trilemma’. It is shown that these complications may be mitigated by reducing the emphasis on open market operations, and utilizing the possibilities provided by the legal reserve requirement or capital controls. The model in Section 3.3 presents a monetary system in which the nominal interest rate is chosen to hit an inflation target, and the nominal exchange rate floats freely to equilibrate the balance of payments. Under this inflation targeting scheme, price stability becomes the only goal of monetary policy. We show that this arrangement is very effective in bringing inflation down, but overdetermination hurts both the employment and growth potential of the economy. For each of the models developed in Sections 3.2 and 3.3, the dynamic properties of the long-run equilibrium are also analysed. Section 3.4 presents some concluding remarks.

3.2

A SMALL OPEN ECONOMY WITH A REAL EXCHANGE RATE TARGET

Here a real exchange rate target is set by the central bank at a level that is just the one required to maintain the competitiveness of exports. Maintaining exports will help employment so that, in a way, a real exchange rate target is analogous to an employment target. The economy operates in the following way. The foreign exchange rate system is better defined as one in which the nominal rate is adjusted (crawling peg) by the central bank to bring the actual real rate to the specified target. This framework analyses the difficulties the central bank faces when attempting to control both the money supply and the foreign exchange rate under an open capital account. This is the well known problem of the ‘trilemma’ of monetary policy: with a rigid nominal exchange rate, attempts to reduce the money supply (and therefore inflation) by raising the interest rate cause an increase in international monetary reserves which, in turn, offsets the reduction in money supply (and thus inflation) that authorities were originally looking for. This situation leads to inflation rates that are higher than internationally accepted levels. But it also allows the central bank to make the

46

Beyond inflation targeting

adjustments that are needed to maintain the competitiveness of exports and the rate of economic growth. 3.2.1

Assumptions of the Model

Some assumptions below are rather general, but others reveal the structuralist approach that is followed in building the model.1 1. 2. 3.

4.

5.

6.

7. 8.

9.

The country produces a consumption good ‘Q’, which could be exported or consumed domestically. The country imports a composite good that may be used for consumption or investment purposes. All capital used in the country is imported. The small economy determines domestically the price of local production PQ, but it cannot affect the international price of imports PM, assumed exogenous. Local prices PQ, are assumed fixed in the short run, but may change in the medium run. This assumption, in relation to the price of local production, allows analysing the effect that inflation may have on the competitiveness of exports. The price of imports, in terms of foreign currency is PM, assumed exogenous as seen above. In terms of local currency, the price of imports is ePM, where ‘e’ represents the nominal exchange rate (number of units of local currency that have to be paid for one unit of foreign currency). There are two social classes: producers and workers. Workers spend all their wage income on consumption of both the local good and the imported good, while producers save a fixed portion ‘s’ of their profit income and the rest they spend on consumption of both the domestic and imported goods. Good Q is produced by means of a fixed-coefficient production function: Q 5 Min [ La, uK ] where ‘L’ represents the labor input, ‘a’ is a technical coefficient, ‘K’ is capital and ‘u’ represents the output/capital ratio. Capital does not depreciate, and firms are assumed to hold excess capacity so that ‘u’ is a variable that moves up (down) when capacity utilization increases (decreases). The interest rate is a policy variable fixed by the central bank for stabilization purposes. The capital stock K and the labor force N (and of course their ratio, denoted here as k) are given in the short and medium run, but they may change in the long run. The nominal exchange rate ‘e’ is a policy variable that remains fixed in the short and medium run, but is adjusted in the long run by the central bank. The adjustment process follows a rule that allows hitting a real exchange target.

Inflation targeting and the real exchange rate in a small economy

47

10.

The real exchange rate, which determines the competitiveness of M exports, is defined as h 5 eP PQ . 11. The wage share is assumed given in the short and medium run, and is defined as A 5 Wa PQ . In the short run W and the technical coefficient a are also fixed, but they adjust in the long run. 12. The analysis is conducted in three stages: in the short run output adjusts while prices, the real exchange rate and the distribution of income remain given; in the medium run inflation will move to clear the monetary market; and in the long run the wage share, the real exchange rate and the capital/labor force ratio are allowed to vary. 3.2.2

General Overview

In this section we present the goods market, and define the conditions for macroeconomic equilibrium in the short run (equality between the investment and savings rates). With the previous assumptions we may find equilibrium values for the rate of growth of the capital stock, the rate of profit, the output/capital ratio and the employment rate; inflation only adjusts in the medium run, as seen below. Within the foreign sector, in the short run, the given levels of the nominal interest rate and real exchange rate help us determine the increase in international monetary reserves; this then enters into the determination of money supply growth. At this point we have already determined the short run variables, and the monetary sector may then clear in the medium run through adjustments in the inflation rate. Once the whole model is solved we examine the impact that open market operations have on inflation and economic growth. It is at this point that the ‘trilemma’ arises and the real exchange rate emerges as a leading character in the story. The relevance of this variable stems also from the fact that it is, along with the interest rate, the only one capable of stimulating the level of economic activity.2 3.2.3

Equations of the Model

Goods market gs 5 sr 1 B

(3.1)

gd 5 b0 1 b1r 2 b2i

(3.2)

u r 5 (1 2 A) h

(3.3)

48

Beyond inflation targeting

PQ 5 PQ

(3.4)

gd 5 gs

(3.5)

L LQK 5 5 l 5 auk N QKN

(3.6)

M^ d 5 h1p 1 h2u 2 h3i 2 h4pE

(3.7)

M^ S 5 R 1 t0 2 t1i

(3.8)

M^ S 5 M^ d

(3.9)

i 5 i0

(3.10)

B5F2R

(3.11)

B 5 B (h) ; Bh , 0

(3.12)

F 5 F (i) ; Fi . 0

(3.13)

e^ 5 W [ hT 2 h ]

(3.14)

Monetary market

Foreign sector

In following the structuralist tradition, several variables are measured as ratios over the value of the capital stock ePMK: desired investment ( gd ) , total saving (gs), the current account deficit (B), capital inflows (F), the increase in international monetary reserves (R) and output (u). The nominal interest rate (i), the actual inflation rate (p), the expected inflation rate (pE ), and the rates of growth of money supply and money demand (M^ s, M^ d ) are written in percentage terms. The real exchange rate, the wage share and the capital/labor force ratio are denoted by h, A and k respectively, and the symbols h1, h2, h3, h4, t0, t1, b0, b1, b2 all represent positive parameters. Equation (3.1) defines total saving as domestic saving ‘sr’, plus foreign saving ‘B’. In Equation (3.2) desired investment depends on the profit and interest rates. Then Equation (3.3) shows the rate of profit in terms of the flexible output/capital ratio, the real exchange rate and the wage share, and Equation (3.4) indicates that local prices remain fixed in the short run. Goods market equilibrium is defined in Equation (3.5), while Equation

Inflation targeting and the real exchange rate in a small economy

49

(3.6) shows the employment rate (l) as a function of the output/capital ratio (for given levels of K/N, denoted here as k). In the monetary market equilibrium is attained when the rates of growth of money supply and money demand (in nominal terms) are equal. In Equation (3.7) money demand is determined by inflation, the output/ capital ratio, the nominal interest rate and an exogenous expected inflation term. Agents face a dual decision regarding price increases: on the one hand, inflation raises their demand for money as they attempt to maintain the value of their real balances. But, on the other hand, agents will also try to get rid of (that is, spend) their holdings of money if they expect price increases, the public will attempt to win the race against inflation. The money supply in Equation (3.8) grows with the accumulation of foreign exchange reserves, but decreases with the exogenous interest rate. This latter component attempts to formalize central bank policy making: in order to reduce money growth, the bank conducts open market operations and drives interest rates up, and the opposite would happen if policy makers wished to increase money growth. This is more elaborate than the ‘Taylor rule’ that appears in the ‘new consensus macroeconomics’ (Arestis and Sawyer, 2003a; Romer, 2000): we argue here that changes in the interest rate will affect the money supply and this may or may not affect inflation; while the Taylor rule (Taylor, 1993) is set up as a mechanical relationship between inflation and the central bank’s reaction. Finally, the constant parameter t0 captures the effect of monetary policy that is not related to open market operations (changes in the legal reserve requirement or the discount rate, for example). Equation (3.9) presents money market equilibrium, and Equation (3.10) shows the exogenous nominal interest rate. Inflation is here a monetary phenomenon: excess money supply growth leads to excess spending, and hence to higher inflation. But also the lack of money supply results from high interest rates, which hurts employment and output growth. Thus, effective demand continues as a relevant determinant of economic activity. Here money is not neutral, so, unlike several versions of the ‘new consensus’ macroeconomics (Meyer, 2001), money does matter. Our approach also differs from the post-Keynesian view of inflation as resulting from cost-push and social conflict (as presented, for example, in Arestis and Sawyer (2006)). This model creates a bridge between monetarist notions of the money market, and Keynesian views on the importance of effective demand. Finally, we need to keep in mind that, ceteris paribus, the use of the interest rate to control inflation, will only lead to a stable adjustment process if money supply is more sensitive to changes in the interest rate than money demand (that is, t1 is bigger than h3 ). Only in this case will the

50

Beyond inflation targeting

higher interest rate lead to a reduction in the excess money supply, and to a lower inflation rate.3 The foreign sector is described in Equations (3.11) through (3.14). In Equation (3.11) the current account deficit is financed by capital inflows and depletion of international monetary reserves. The current account deficit depends negatively on the real exchange rate and net capital inflows depend positively on the interest rate. In Equation (3.14) we see how, in the long run, the monetary authority devalues the nominal exchange rate to bring the real rate to its target (hT); Ω is a positive parameter. 3.2.4

Formal Solution of the Model

The process starts when the central bank defines the nominal interest rate and the target for the real exchange rate. These two policy variables are utilized, respectively, to control the money supply, and to maintain the competitiveness of net exports. The current account deficit and net capital inflows can then be determined, and with this we can solve for R in Equation (3.11). In the goods sector Equations (3.1), (3.2), (3.12) and (3.5) determine the profit rate. In Equation (3.3), given A and h, we find the output/capital ratio, and Equation (3.6) then leads to the employment rate (l). Again in Keynesian fashion, employment is determined in the goods market (not in the labor market as monetarist and new classical approaches would suggest). The short-run solution for the goods sector is shown in Figure 3.1. We may now move into the medium run and use Equations (3.7), (3.8), (3.9) and (3.10) in the monetary sector to solve for inflation, as shown below: p0 5

R0 1 t0 1 (2 t1 1 h3) i0 2 h2u0 1 h4pE h1

(3.15)

where u0 represents the level of the output/capital ratio found in the goods market, and i0 denotes the interest rate fixed by the central bank. The relationship between the interest rate and inflation is shown in Figure 3.2. The money market equilibrium schedule (MM line) in Figure 3.2 has a negative slope when the conditions specified in Note 3 hold. This schedule provides the combinations of inflation and interest rate that are consistent with equilibrium in the monetary market. Here p0 is the inflation rate that clears the money market when the central bank fixes the interest rate at i0. In order to bring inflation down, the central bank raises the interest rate to reduce money supply growth; as the availability of money declines, aggregate spending decreases and inflation goes down (the economy moves to the left along the MM line).

Inflation targeting and the real exchange rate in a small economy

51

gs

g

gd g0

r u

u u=

l = auk

rh 1–A

u0

Figure 3.1

r

l

l

Goods market equilibrium in the short run

i

i0

MM ␲0 Figure 3.2

Money market equilibrium schedule

␲

52

Beyond inflation targeting

But, in the foreign sector, a higher ‘i’ will attract more capital flows, making international monetary reserves go up. The money supply will then increase (the MM line shifts out), compensating (at least partially) the initial attempt to reduce inflation pressure. Thus the monetary authorities cannot control both the exchange rate and interest rate (and the money supply) under an open capital account. Second, in the goods market, as ‘i’ goes up, investment demand goes down; the gd line in Figure 3.1 shifts down and the rate of growth, the rate of profit, the output/capital ratio and the employment rate decrease. Stabilization is not too successful and hurts economic activity. Fortunately, the central bank can decide to increase the target level for h, which will cause foreign savings (the current account deficit) to go down. The gs line in Figure 3.1 will shift to the right (down) leading to faster growth, higher profit rate, and higher output/capital ratio and employment rate. We have thus confirmed the double edged character of the real exchange rate: it is to blame for the difficulties the central bank faces in controlling inflation; had it had more flexibility in the short run, there would had been no accumulation of international reserves and we would not have lost ground in the fight against inflation. But, on the other hand, it is the ability of the central bank to move the real exchange rate to a targeted level, which provides the economy with the possibility to reach faster growth and higher employment rates. Arguments and evidence in favor of targeting the real exchange rate may be found in Frenkel (2004) and in Frenkel and Ros (2006). Now a final comment on monetary policy: there is no reason for the central bank to only use open market operations to control the money supply. Other policy instruments, like the legal reserve requirement, could play an important control role without a direct impact on interest rates.4 A higher reserve requirement ratio may be seen in our model as a reduction in the term t0 in Equation (3.15). In Figure 3.2 this will show as a downward shift of the MM schedule, indicating that there will now be a lower level of inflation at every level of the interest rate. The central bank may thus be able to curve down price growth, but without changes in international monetary reserves, breaking in this way the ‘trilemma’. The alternative presented here is also in line with the arguments in Frenkel (2004), who goes even further to point out that capital controls are feasible; a similar argument on the convenience of capital controls is provided by Ffrench Davis (2003). Many central banks, however, would not like to use the legal reserve requirement, or capital controls as instruments for monetary policy: they prefer open market operations because they are, well, more market

Inflation targeting and the real exchange rate in a small economy

53

oriented.5 Policy makers are thus opting for more flexible exchange rate arrangements, combined with inflation targeting regimes, as will be explained in Section 3.3. 3.2.5

Long-run Dynamics

We follow Cordero (1995) in analysing the interaction between the wage share (A) and the real exchange rate (h), which we assumed given in the short and medium run. But we have an additional state variable here: the capital/labor force ratio k, so this is really a 3 × 3 system. We assume that, in the long run, the variables which adjusted in the short run to clear the goods market (u, g, r, l ), and in the medium run to clear the monetary sector (p) remain at their equilibrium levels. The motion of A is given by A^ 5 W^ 1 a^ 2 p

(3.16)

where the hats denote rate of growth. The coefficient a was defined as Q so we can write that y 5 L 5 1a where ‘y’ represents average labor productivity. Following Arrow (1962) it is argued that labor productivity depends on learning, which is assumed to be positively related to production. We also follow Dutt (1994) in assuming that production is associated to the investment rate ‘g’: L Q

y^ 5 2a^ 5 Y ( g 2 g) , Yg . 0

(3.17)

with Yg the partial derivative of productivity growth with respect to the investment rate, and g is an arbitrary constant number. The specification of wage growth assumes that workers try to increase their nominal wage whenever the actual wage share falls below an exogenous level AW desired by workers as indicated in Equation (3.18): W^ 5 q [ AW 2 A ]

(3.18)

Finally, the inflation rate that clears the monetary market depends on the real exchange rate and the wage share: p0 5 p (h, A) ; ph , 0; pA , 0

(3.19)

where ph, pA denote partial derivative of p with respect to h and A, respectively. The dependency of p0 on ‘h’ and ‘A’ comes from the presence of the

54

Beyond inflation targeting

equilibrium output/capital ratio (u0) in Equation (3.15); and from the fact that this u0 was determined in the goods market (through the interaction between gd and gs for given h and A). We may now put Equations (3.17) through (3.19) in (3.16) to get a differential equation in h and A: A^ 5 q [ AW 2 A ] 2 Y [ g (h) 2 g ] 2 p [ h, A ]

(3.20)

The dynamic behavior of the real exchange rate is described by: h^ 5 e^ 2 p

(3.21)

According to Equation (3.14), reproduced below, the nominal exchange rate moves up when the actual level of the real exchange rate falls below the target chosen by the central bank: e^ 5 W [ hT 2 h ]

(3.14)

Next use Equations (3.19) and (3.14) in Equation (3.21) to get: h^ 5 W [ hT 2 h ] 2 p [ h, A ]

(3.22)

We should recall now that we had a third state variable (k), but the system with Equations (3.20) and (3.22) may be solved independently of k, so we may proceed as if we had a 2 × 2 system with A and h as the state variables.6 The stability of this system (Equations 3.20 and 3.22) is analysed by means of the Jacobian matrix: 0A^ 0A Det (J) 5 ≥ ^ 0h 0A

0A^ (2 q 2 pA) 0h ¥ 5 £ 0h^ 2pA 0h

(2Yggh 2 ph)

§

(2 W 2 pA)

The determinant of the Jacobian is: Det (J) 5 (q 1 pA) (W 1 ph) 2 pA (Yggh 2 ph) 5 q [ W 1 ph ] 1 pA [ W 2 Yggh ] which will be positive if W . ph and W . Yggh. In other words, Det (J) will be positive if W is large. The trace is given by Tr (J) 5 2q 2 pA 2 W 2 ph, which will be

Inflation targeting and the real exchange rate in a small economy

55

negative if q, and Ω are large. So the long-run equilibrium will be stable if wages are flexible and the central bank is very committed to maintaining the real exchange rate close to the chosen target. Finally, as the shortrun equilibrium level of growth depends on h, and h is determined in the system formed by Equations (3.20) and (3.22), we may say that economic growth is endogenous in this model.

3.3 3.3.1

THE SMALL ECONOMY UNDER AN INFLATION TARGETING REGIME The Inflation Targeting Regime

In the previous section we saw the problems that may arise if, with a fixed real exchange rate, the central bank attempts to control inflation by using open market operations. Although, as we saw, there are several ways in which the ‘trilemma’ could be faced successfully, there is a strong tendency to move to an inflation targeting regime and turn the exchange rate system into a flexible one. This bias, according to Epstein (2005), is part of a global change in the practice of central banking, promoted mostly by institutions like the IMF. In this view central banking practices must be based on independence, inflation as the goal of policy (including application of inflation targeting regimes), and the use of indirect methods of monetary policy (mostly open market operations). Arestis and Sawyer (2003b) relate this bias to the ‘new consensus’ macroeconomics in which inflation targets become the focus of monetary policy. In this section we want to emphasize that, once this approach is adopted, everything in the economy will be tied to the inflation target; the result is great success in bringing inflation down, but a rather disappointing performance in terms of growth and employment. According to Galindo and Ros (2008) this monetary regime leads to possible appreciation of the real exchange rate, and increased vulnerability to monetary shocks coming from the external sector. In the case of Mexico, these authors show that inflation targeting has led to an appreciation of the real exchange rate, which has had a negative impact on output growth. Epstein (2002) reports that the utilization of this monetary system in South Africa led to high interest rates, accompanied by low employment and investment rates. His recommendation is that South Africa move to a scheme in which employment is targeted, but subject to an inflation goal. He also recommends the use of capital controls in order to better control the effect of the world economy on South Africa. Setterfield (2006) also

56

Beyond inflation targeting

argues that governments should pursue output and employment targets, in addition to inflation goals. Other more general cross-country studies have found that there is no reason to keep inflation below the 3 to 5 percent range, especially because in middle income economies higher single digit rates of inflation could stimulate economic growth (Pollin and Zhu, 2005). Another study conducted by Ball and Sheridan (2003) concludes that inflation targeting has generated no major benefits in economic performance (other than a decline in inflation rates). In this section we develop a model to examine the effects of inflation targeting on growth and employment. 3.3.2

Revised Equations

Money market i 5 G (p) ; Gp , 0; p 5 p (i) , pi , 0

(3.23)

M^ s 5 t0 2 t1i

(3.24)

M^ d 5 h1p 1 h2u 2 h3i

(3.25)

p 5 pT

(3.26)

M^ s 5 M^ d

(3.27)

B5F

(3.28)

B 5 B (h) ; Bh , 0

(3.29)

F 5 F (i) ; Fi . 0

(3.30)

gd 5 gd (r, i) , gdr . 0, gdi , 0

(3.31)

G 5 gd (r, i) 2 B (h)

(3.32)

S 5 sr

(3.33)

G5S

(3.34)

Foreign sector

Goods sector

Inflation targeting and the real exchange rate in a small economy

57

PQ 5 PQ

(3.4)

u r 5 (1 2 A) h

(3.3)

l 5 auk

(3.6)

In Equation (3.23) the interest rate is the instrument the central bank uses to control inflation. Once a target is defined in Equation (3.26), open market operations allow bringing the interest rate to the level that is exactly needed to hit the chosen inflation target. As for money demand (Equation 3.25), we use a formulation that is similar to the one in Section 3.2. Finally, and in accordance to this monetary system, we assume that the central bank announces an inflation target in the understanding that this goal is prior to any other objective the bank might have. The priority assigned to hitting this level will be clear as the model is explained in more detail. The monetary sector operates in the following manner. First, the central bank recognizes that the interest rate and the inflation target are not chosen independently of one another: the bank knows that there is an inverse relationship between nominal interest rates and money supply. Thus, in order to fight inflation, the bank conducts open market operations (to reduce the money supply) and pushes interest rates up. In other words, higher interest rates, are associated to lower inflation rates, and this is exactly what Equation (3.23) tells us. Again this is clearly different from the ‘Taylor rule’ that appears in the new synthesis models that we referred to in previous sections of this chapter. Our Equation (3.23) is a behavioral relationship indicating how inflation reacts to changes in the nominal interest rate. The so-called Taylor rule is a more mechanical expression indicating how the central bank moves the interest rate when inflation goes up. In the new synthesis models this rule replaces the monetary sector. In this chapter, however, we argue that economic authorities in less developed countries still believe that their policy decisions should be based on their perception of the monetary sector; hence there is no Taylor rule in this model, but there is a fully structured monetary market. Next we turn to the foreign sector where we start by getting rid of the accumulation of international monetary reserves. As Equation (3.28) shows, the current account deficit has to be financed by net capital inflows. This is made possible by allowing the real exchange rate to fluctuate as required to satisfy Equation (3.28).7 In the goods sector the specification looks a bit different from Section 3.2, but it still holds the same meaning: Equation (3.32) defines demand

58

Beyond inflation targeting

injections, while (3.33) defines domestic savings. Equations (3.3), (3.4) and (3.6) are borrowed from the previous section to describe prices, the rate of profit and the employment rate. 3.3.3

Solution of the Model

Price rigidities are an important feature of this model: in the short run prices remain fixed by Equation (3.4), and will adjust only in the medium run (the growth of prices in the medium run is determined by the inflation target chosen by the central bank). In the short run the burden of adjustment falls on the output/capital ratio and the real exchange rate. Once the inflation target pT is announced, Equations (3.26) and (3.23) allow defining the nominal interest rate. In the foreign sector the real exchange rate adjusts to equilibrate the balance of payments and we have: h 5 h [ i ] , hi , 0

(3.35)

Next with i and h known we move to the monetary sector: Equations (3.24), (3.25) and (3.27) provide an expression for money market equilibrium: t0 2 t1i 2 h1p (i) 2 h2u 1 h3i 5 0

(3.36)

This equation describes a relationship between the interest rate and the output/capital ratio, and the slope is given by [ 2 t1 2 h1pi 1 h3 ] 0u 5 ,0 h2 0i As before, the sign of this slope results from the conditions in Note 3 (Figure 3.3 presents the corresponding graph) and we write that: u 5 u [ i ] , ui , 0

(3.37)

Once we know the inflation target we can plug the known interest rate i0 in Equation (3.36) to find the output/capital ratio (u) that clears the money market. This implies that, in order to derive u, it is not necessary that savings and investment be equal: the model is overdetermined. In the next step we go to the goods sector, and examine the behavior of the rate of profit. From Equation (3.3), and also using Equations (3.35) and (3.37) we may write the following expression: r5

[ u (i) ] (1 2 A) [ h (i) ]

(3.38)

Inflation targeting and the real exchange rate in a small economy

59

u

MM

i Figure 3.3

Money market equilibrium schedule under inflation targeting

and the response of the rate of profit to changes in the interest rate is given by: (1 2 A) 0r [ u h 2 uhi ] 5 ri 5 [ h (i) ] 2 i 0i

(3.39)

We do not have information to define the sign of the term [ uih 2 uhi ] , so we can just examine two possible cases: r 5 r [ i, (1 2 A) ] , ri , 0, rA , 0

(3.40a)

r 5 r [ i, (1 2 A) ] , ri . 0, rA , 0

(3.40b)

or

Signing the derivatives of r with respect to ‘i’ is very important as that will allow to also sign the derivatives of demand injections (G) and domestic saving (S) with respect to ‘i’. Case 1 ri , 0, rA , 0

60

Beyond inflation targeting

From Equation (3.32) and from Equation (3.40a) we get: G 5 gd { r [ i, (1 2 A) ] , i } 2 B [ h (i) ]

(3.41)

and the slope of this line is 0G 5 gdrri 1 gdi 2 Bhhi , 0 0i

(3.42)

Next from Equations (3.33) and (3.40a) obtain: S 5 sr [ i, (1 2 A) ]

(3.43)

and the slope of the savings function is 0S 5 sri , 0 0i

(3.44)

Just as in other Keynesian models, the slope of the S line has to be steeper than that of the G line in order to secure stability of equilibrium in the short run: P

0G 0S P,P P 0i 0i

(3.45)

The graphical solution appears in Figure 3.4. In the first panel choosing an inflation target fixes the interest rate by means of Equation (3.23). The foreign sector appears in the second panel, where the known interest rate leads to the real exchange rate that clears the balance of payments. The monetary sector in the third panel shows that, given the interest rate, the ‘MM’ schedule helps us determine the output/capital ratio. Then we examine the behavior of demand injections (G) and domestic saving (S), and overdetermination becomes clear. If the interest rate is below the one that brings G and S to equality, then the economy will be at a point on the G line. Any reduction in the inflation target (and thus increase in the interest rate) will lead to real appreciation, higher unemployment and a lower rate of economic growth. A reduction of the inflation target will reduce the level of activity even if the interest rate is to the right of the one that makes G and S equal. As the nominal exchange rate is no longer a policy variable, the central bank cannot stimulate economic activity and reduce the inflation rate at the same time. Case 2 ri . 0, rA , 0

Inflation targeting and the real exchange rate in a small economy

61

␲ ␲T ␲ = p (i) i

i0 h

B (h) = F (i) u

i

u

l = uak

MM i l G, S

G S i0

Figure 3.4

i

An open economy under inflation targeting and overdetermination in the goods sector

Here the G function keeps the negative slope, but the S function will definitely have a positive slope in the S versus i plane, as shown in Figure 3.5. With the interest rate at the given level i0 (which is, again, lower than the one that brings G and S to equality), the economy must be at a point on the

62

Beyond inflation targeting

G, S S

G

i

i0 Figure 3.5

Overdetermination on the goods sector when savings slopes up

S schedule (investors cannot satisfy their desires). A reduction in the inflation target will require a rightward movement of the interest rate, with the economy moving along the domestic saving line towards the right and to a higher rate of growth. But notice that investment desires will suffer a downward movement along the G schedule. So what we are observing here is that, as inflation is brought down, the investment drive also declines. Thus in this case we see that within the inflation targeting regime, overdetermination causes a gap between actual and desired investment. When the given interest rate lies to the right of the level that brings domestic savings and demand injections to equality, the economy is positioned at a point on the G schedule; in this case stabilization policies lead to a lower rate of growth. 3.3.4

Long-run Dynamics

In the long run we analyse two state variables: A and k, where k represents the capital (K) to labor force (N) ratio. The analysis here is based on Cordero (2002). The motion of A is described by Equation (3.16) reproduced below: A^ 5 W^ 1 a^ 2 p

(3.16)

Inflation targeting and the real exchange rate in a small economy

63

and for the technical coefficient ‘a’ we use Equation (3.17) also reproduced below: y^ 5 2a^ 5 Y (g 2 g) , Yg . 0

(3.17)

The growth of nominal wages is again depicted by Equation (3.18) W^ 5 q [ AW 2 A ]

(3.18)

but the desired wage share is now endogenized and made dependent on the employment rate: L L QK AW 5 a0 1 a1 a b 5 a0 1 a1 a b 5 a0 1 a1 (auk) N Q KN

(3.46)

Finally, we recall that the local inflation rate was set by a target defined by the central bank, according to Equation (3.26): p 5 pT

(3.26)

The use of Equations (3.17), (3.18), (3.46), (3.37) and (3.26) in Equation (3.16) leads to a differential equation for the evolution of A: A^ 5 qa0 1 qa1au [ i ] k 2 qA 2 Y [ g 2 g ] 2 pT

(3.47)

The motion of k is described by: k^ 5 K^ 2 N^ where K^ is the investment rate (g), and N^ is assumed fixed at a level denoted by ‘n’: k^ 5 g 2 n

(3.48)

In order to finish writing Equations (3.47) and (3.48) in terms of A and k, we have to replace g with its short-run equilibrium value.8 Before we proceed we must generate expressions for the desired investment rate and total savings. The first of these is provided by Equation (3.31) which, after using Equation (3.40a), becomes: gd 5 gd { r [ i, (1 2 A) ] , i } , gdr . 0, gdi , 0, ri , 0, rA , 0

(3.49)

64

Beyond inflation targeting

Total savings, on the other hand, is defined as domestic savings (sr) plus foreign savings [B(h)]. From Equations (3.33), (3.40a) and our definition of foreign savings, we get gs 5 s { r [ i, (1 2 A) ] } 1 B [ h ] , s . 0, ri , 0, rA , 0

(3.50)

For the case depicted in Figure 3.4 the g level is determined by the corresponding value of gd in Equation (3.49).9 The system formed by Equations (3.47) and (3.48) may be rewritten as: A^ 5 qa0 1 qa1au [ i ] k 2 qA 2 Y [ gd { r [ i, (1 2 A) ] , i } 2 g ] 2 pT (3.51) k^ 5 gd { r [ i, (1 2 A) ] , i } 2 n

(3.52)

The stability of the system is analysed by means of the Jacobian matrix: 0A^ 0A Det (J) 5 ≥ ^ 0k 0A

0A^ (2q 2 YggdrrA) 0k ¥ 5 £ 0k^ gdrrA 0k

qa1au [ i ]

§

0

We have: Det (J) 5 2 [ gdrrA ] [ qa 1au [ i ] ] . 0 and Tr (J) 5 (2q 2 YggdrrA) Since rA , 0, the long-run equilibrium will be stable when nominal wage flexibility is important (q large) compared to learning effects (Yg). This does not mean that the economy has to stay away from productivity gains; it means instead that wage flexibility is necessary to accommodate productivity growth.

3.4

CONCLUDING REMARKS

In this chapter we compare the effects of different monetary regimes on the growth and inflation performance of an open economy. With a real exchange rate target the decisions of the central bank are affected by the ‘trilemma’ of monetary policy, but the real exchange rate can still be

Inflation targeting and the real exchange rate in a small economy

65

utilized to push employment and economic growth. Thus, although stabilization based on open market operations has a tendency to lower economic activity, the central bank can always target a level of the real exchange rate which compensates, at least partially, the contractionary effects of stabilization. The use of monetary instruments like the legal reserve requirement may help bring inflation down without the problems associated with the ‘trilemma’. The long-run equilibrium is stable if nominal wages are flexible, and the central bank maintains a competitive real exchange rate. Under an inflation targeting regime, the problems of the ‘trilemma’ disappear and prices are more easily stabilized. In this setting all economic results depend on monetary policy: the goods sector is not necessary to find equilibrium levels for the relevant variables. This situation causes overdetermination as there is no level of the rate of profit which is capable (except by accident) of bringing saving and investment to equality. Stabilization policy is harmful to growth and employment and, even if savings increased, investment desires remain unsatisfied. The inflation targeting regime leads to a stable long-run equilibrium if nominal wages are flexible enough to accommodate the productivity gains resulting from learning. The most important result in this chapter is the recognition of the critical importance that the real exchange rate has on the determination of the rates of accumulation and employment of the small open economy. Thus countries which have already decided to embark on inflation targeting regimes should allow themselves to tolerate higher inflation goals to avoid excessive appreciation of the real exchange rate. For countries which still find themselves battling the difficulties of the ‘trilemma’ the recommendation would be to combine the use of open market operations with the legal reserve requirement or even with some control over the capital account of the balance of payments.

NOTES 1. The structuralist approach is described in Taylor (1991) and in Dutt (1992). 2. In other structuralist models (Dutt, 1984; Taylor, 1985, for example) income distribution plays a critical role in the determination of economic growth. Blecker (1989) then extends those models to examine how income distribution affects growth in an open economy. In this chapter, however, the investment function (our Equation (2)) looks more like the NeoKeynesian version used by Marglin (1984), and the ouput/capital ratio has been taken out of the desired accumulation function. This modification prevents the wage share from having an effect on the rate of growth. 3. The adjustment mechanism may be described by 0i/0t 5 f (Ms 2 Md) , f . 0; and this will be stable only if f (2t1 1 h3) , 0. 4. This point was analysed with a formal model in Cordero (2005).

66

Beyond inflation targeting

5. The mainstream view argues that higher levels of the legal reserve requirement cause higher administrative costs, and thus may lead to higher spreads and margins for financial intermediation. But this is not necessarily true: in the case of Costa Rica, for example, the reserve requirement fell from 15 percent in 1999 to 5 percent in 2002, but in the same period the spread between active and passive interest rates declined only half a percentage point. Then, in 2003, when the reserve requirement jumped from 5 percent to 10 percent, the spread went up by only 1.5 percent (Cordero, 2005). 6. The long-run motion of k is given by: k^ 5 g 2 N^ . The short-run equilibrium level of g depends on h: g0 5 g (h) , gh . 0. Assume that N^ 5 n (k) with nk . 0, and we get k^ 5 g (h) 2 n (k) . The system formed by Equations (3.20) and (3.22) provides a solution for h, which we use to find the level of k required for k^ 5 0. The equilibrium is stable if nk . 0. 7. With prices fixed in Equation (3.4), h adjusts as a result of variations in the nominal exchange rate e. 8. Of course we also analysed a second case which here would require that we plug Equation (3.40b) in Equation (3.31) for the gd. expression, and in Equation (3.33) for the gs. equation. The long-run dynamics will be analogous to that of the first case, but will not be presented in the chapter. 9. If the interest rate that is consistent with the inflation target falls to the right of the level making G equal to S, then the g level is determined by the corresponding value of gs in Equation (3.50). Again the dynamics of this case is analogous to the one resulting from Figure 3.4 and will not be analysed here in detail.

REFERENCES Arestis, P. and M. Sawyer (2003a), ‘New consensus, new Keynesianism, and the economics of the Third Way’, Levy Economics Institute of Bard College working paper no. 364. Arestis, P. and M. Sawyer (2003b), ‘Inflation targeting: a critical appraisal’, Levy Economics Institute of Bard College working paper no. 388. Arestis, P. and M. Sawyer (2006), ‘The nature and role of monetary policy when money is endogenous’, Cambridge Journal of Economics, 30 (6), 847–60. Arrow, K. (1962), ‘The economic implications of learning-by-doing’, Review of Economic Studies, 29 (3), 155–73. Ball, R. and N. Sheridan (2003), ‘Does inflation targeting matter?’, National Bureau of Economic Research working paper no. 9577, Cambridge, MA. Blecker, R. (1989), ‘International competition, income distribution and economic growth’, Cambridge Journal of Economics, 13 (3), 395–412. Cordero, J. (1995), ‘Essays on growth and distribution for open economies’, unpublished PhD dissertation, University of Notre Dame, Indiana. Cordero, J. (2002), ‘A model of growth and conflict inflation for a small open economy’, Metroeconomica, 53 (3), 261–89. Cordero, J. (2005), ‘Inflacion, politica monetaria y regimen cambiario en Costa Rica’, in J.R. Vargas and Y. Xirinachs (eds), La formación de economistas: Ensayos en honor de Pepita Echandi, San Jose, Costa Rica: Posgrado en Economia, Universidad de Costa Rica. Dutt, A. (1984), ‘Stagnation, income distribution and monopoly power’, Cambridge Journal of Economics, 8 (1), 25–40. Dutt, A. (1992), ‘Two issues on the state of development economics’, in A. Dutt and K. Jameson (eds), New Directions in Development Economics, Aldershot, UK and Brookfield, VT, USA: Edward Elgar, pp. 1–34.

Inflation targeting and the real exchange rate in a small economy

67

Dutt, A. (1994), ‘On the long run stability of capitalist economies’, in A. Dutt (ed.), New Directions in Analytical Political Economy, Aldershot, UK and Brookfield, VT, USA: Edward Elgar, pp. 93–120. Epstein, G. (2002), ‘Employment-oriented central bank policy in an integrated world economy: a reform proposal for South Africa’, Political Economy Research Institute working paper series no. 39, University of Massachusetts, Amherst. Epstein, G. (2005), ‘Central banks as agents of economic development’, Political Economy Research Institute working paper series no. 104, University of Massachusetts, Amherst. Epstein, G. and E. Yeldan (2008), ‘Inflation targeting, employment creation and economic development: assessing the impacts and policy alternatives’, International Review of Applied Economics, 22 (2) (March), 129–30. Ffrench Davis, R. (2003), ‘Macroeconomic balances in emerging economies: the conflict between purely-financial and real-economy macrobalances’, paper prepared for the Task Force on Macroeconomic Policy at the Initiative for Policy Dialogue, Columbia University, New York. Frenkel, R. (2004), ‘Real exchange rate and employment in Argentina, Brazil, Chile and Mexico’, paper prepared for the G24, Centro de Estudios de Estado y Desarrollo, Argentina. Frenkel, R. and J. Ros (2006), ‘Unemployment and the real exchange rate in Latin America’, World Development, 34 (4) (April), 631–46. Galindo, L. and J. Ros (2008), ‘Alternatives to inflation targeting in Mexico, International Review of Applied Economics, 22 (2), 201–14. Marglin, S. (1984), Growth, Distribution, and Prices, Cambridge, MA: Harvard University Press. Meyer, L.H. (2001), ‘Does money matter?’, Federal Reserve Bank of St Louis Review, 83 (5), 1–16. Pollin, R. and A. Zhu (2005), ‘Inflation and economic growth: a cross country non-linear analysis’, Political Economy Research Institute working paper no. 109, University of Massachusetts, Amherst. Romer, D. (2000), ‘Keynesian macroeconomics without the LM curve’, Journal of Economic Perspectives, 14 (2), 149–69. Setterfield, M. (2006), ‘Is inflation targeting compatible with Post Keynesian economics’, Journal of Post Keynesian Economics, 28 (4) (Summer), 653–72. Taylor, J. (1993), ‘Discretion versus policy rules in practice’, Carnegie-Rochester Conference Series on Public Policy, 39, 195–214. Taylor, L. (1985), ‘A stagnationist model of economic growth’, Cambridge Journal of Economics, 9 (4), 383–403. Taylor, L. (1991), Income Distribution, Inflation and Growth: Lectures on Structuralist Macroeconomic Theory, Cambridge, MA: MIT Press.

PART II

Thematic issues: class relations and gender impacts of inflation targeting

4.

Income, class and preferences towards anti-inflation and anti-unemployment policies Arjun Jayadev1

4.1

INTRODUCTION

From one important point of view, indeed, the avoidance of inflation and the maintenance of full employment can be most usefully regarded as conflicting class interests of the bourgeoisie and the proletariat respectively, the conflict being resolvable only by the test of relative political power in the society. Harry Johnson (Johnson, 1968, p. 986)

Among the casualties of the advent of the rational expectations revolution in the 1970s was a rich vein of political economy which analysed the macroeconomic dynamics of unemployment and inflation as deriving from distributive struggles between capitalists and workers (Bach and Stephenson, 1974; Boddy and Crotty, 1975; Hibbs, 1977; Rosenberg and Weisskopf, 1981; Rowthorn, 1977). This approach often made explicit the class conflict innate in Keynesian accounts of the Phillips curve trade-off and in considerations of the natural rate of unemployment.2 While the details of these studies varied, the argument, with slight modifications, remained the same: workers and capitalists had opposing and irreconcilable differences in the trade-off between inflation and unemployment. Such ‘conflict theories’ not only gave political explanations for the trajectory of inflation and unemployment (most notably in the work of Hibbs (1977)) but also thereby provided direct predictions for the preferences of individuals towards anti-unemployment and anti-inflation policies based on their position within the social stratification. Specifically, the working class, broadly classified, is more concerned about reducing unemployment than firm owners. Unemployment reduces the lifetime income of workers directly and also exerts a downward pressure on wages by reducing the bargaining power of workers.3 Inflation, on the other hand, worked to the advantage of those workers with low savings. With the rise of new classical macroeconomics in the 1970s and its 71

72

Beyond inflation targeting

subsequent hegemony any trade-off between inflation and unemployment came to be seen as essentially short term and certainly could not be utilized by policy makers to affect macroeconomic outcomes without incurring severe macroeconomic costs. As a consequence, much of the macroeconomic literature moved away from class-based, political economy models of inflation and unemployment towards what Iversen and Soskice (2006, p. 425) elegantly call macroeconomics which ‘[f]ocuses attention on what democratic governments can do wrong in the short term’. The prescription that followed was to replace government with independent central banks, and discretionary macroeconomic policy with rules-based approaches such as inflation targeting (Barro and Gordon, 1983; Cukierman, 1992). While post-Keynesian and heterodox approaches never abandoned the idea of demand management, the advent of New Keynesian economics restored the space in mainstream economics for economic policy to have beneficial macroeconomic outcomes. As opposed to rational expectations, there is a role for demand management and other policies when imperfections arise due to wage and price rigidities whereby involuntary unemployment can be reduced. The now long literature on the non-accelerating inflation rate of unemployment (NAIRU) continues to suggest that combating inflation and unemployment involves two independent and potentially opposing targets.4 Given that there has been a restitution of space for macroeconomic policy, this chapter seeks to make a contribution in reconstructing the distributional consequences of anti-inflationary and anti-unemployment policies, and attitudes towards them. Recent research on the impact of such policies on the poor and the rich has used social survey data to answer part of this question. Scheve (2004) and Jayadev (2006) both find strong results that the relatively poor are less likely than the relatively rich to prioritize tackling inflation as opposed to tackling unemployment, presumably because they take the latter to be a more important problem facing them. Other research (Jayadev, 2008) continues this line of research by assessing the class content of disinflationary and expansionary policy. This research finds support for the contention that those in contrary class positions respond very differently to policies engineered to combat inflation versus those that are designed to combat unemployment. There are substantial class-based differences in what may be termed ‘relative aversion’ to inflation and unemployment (a preference that policy is designed to keep inflation down rather than unemployment down). Capitalists and highly skilled workers are more likely to display relative inflation aversion than less skilled and unskilled workers. I briefly extend the analysis to look at the relationship between

Anti-inflation and anti-unemployment policies

73

relative inflation aversion and other attitudes towards macroeconomic policy. Individuals who support broadly pro-labor, solidaristic and redistributionary policies for the government are significantly less relatively inflation averse (or what is the same thing, more relatively unemployment averse). This chapter summarizes the findings of these two lines of research (that is, the income and class-based preferences for combating inflation rather than unemployment). As such, the chapter provides some indication for the degree and kind of political support that can be drawn for policies that aim to combat inflation versus those which combat unemployment. The results have implications for the debate on antiinflationary policies and inflation targeting in particular as they apply to different countries. I very briefly summarize relevant research on the income and class impacts of inflation and unemployment and attitudes towards these problems in Section 4.2. Section 4.3 describes the data that I use. Section 4.4 presents the results of various logistic regressions. Section 4.5 summarizes and concludes.

4.2

INCOME AND ATTITUDES TOWARDS INFLATION AND UNEMPLOYMENT

The theoretical impetus for the shift in focus mentioned in Section 1 was provided by the rational expectations revolution of the 1970s and subsequent neo and new classical analyses of the state which suggested that expansionary policies were short-term palliatives at best, and in the medium to long run, simply inflationary. That is to say, expansion, when driven ‘artificially’ by the state using fiscal or monetary authorities, may well have a positive effect on the poor in the short run by increasing the growth and employment rates in the economy; but in the long run will simply lead to more inflation and a fall in employment back to its natural rate. The impact on the poor, it follows, is possibly a net deterioration in their welfare, as unemployment remains at previous levels and goods become more costly. Thus the standard prescription of expansionary policy in the 1960s and 1970s was replaced by the orthodoxy of inflation targeting in the 1980s and 1990s, at least partly as a manner in which the poor were to be protected from a perhaps well meaning, but misguided government. The theoretical linkages between inflation and poverty do not, however, lead to the direct conclusion that inflation is necessarily bad for the poor. A brief account of the channels by which expansionary and potentially inflationary policy affects the poor as identified in Romer and Romer

74

Table 4.1

Beyond inflation targeting

Some hypothesized linkages between inflation and poverty

Channel

Effect on welfare of poor

Short run Growth

Real wage and transfers

Expansionary demand policy, in raising average income (without a change in the distribution of income) has a direct effect on reducing poverty. To the extent that wages and transfers are unindexed to inflation, and that certain commodities (for example, food) form a larger portion of the consumption basket of the poor, the real income of the poor falls.

Medium to long term Growth

Credit market

Uncertainty

Romer and Romer (1998) suggest a strong positive correlation between high inflation and low growth, as well as high inflation and instability of demand, thus suggesting that short-run gains for the poor from expansionary demand policy is eroded in the long run. It is assumed that unanticipated inflation is positive for debtors and negative for creditors. Given that the poor fall overwhelmingly in the former category, it should be the case that they benefit to some extent. However, the size of this effect depends on how much the poor owe and how much the real wage declines. Uncertainty caused by inflation causes financial market dislocations, increases capital flight and reduces the rate of growth through reducing physical and human capital investment; all of which have a negative impact on the poor.

(1998), Easterly and Fischer (2001) and Cardoso (1992) among others is summarized in Table 4.1. Given these ambiguous theoretical relationships, both at the macroeconomic and microeconomic levels, the link between inflation and the poor is strongly an empirical matter, and certainly highly dependent on conditioning structural factors. One of the more influential papers linking anti-inflationary policy and the poor is that by Easterly and Fischer ( 2001). In a novel analysis the authors turn to an unimpeachable authority on the subject of inflation

Anti-inflation and anti-unemployment policies

75

and the poor – the poor themselves. Easterly and Fischer use a large social survey conducted in 38 countries in 1996 to analyse the attitudes towards inflation of individuals at different income levels. Their study suggests a robust effect: the self-identified poor were more likely than the selfidentified rich to mention inflation5 as among the top three concerns for the economy. They find, by contrast, that there is no statistically significant difference between the rich and the poor in terms of their attitude towards recession and unemployment. From their study it is easy to draw the conclusion that anti-inflationary policy is preferred by the poor to policy designed to combat unemployment or recession. Expansionary policy which leads to inflation can be thought of as being akin to a tax:6 redistribution from the public at large to facilitate increased public expenditure. As such, asking about people’s preferences towards inflation without presenting the alternative is similar to asking them about their preferences towards lowering taxes. From the macroeconomic policy viewpoint, an important and relevant opportunity cost of lower inflation is higher unemployment and informalization, especially when it is achieved by reducing government spending. As such, a more useful question to ask would be to ask the poor and the rich their relative preferences between inflation and unemployment. Ideally, the question that should be posed is of the form: ‘if the government was to reduce inflation, even if it meant increasing unemployment a little, would you support it?’ While the Roper Starch Survey used by Easterly and Fischer does not unfortunately provide such an option, there is another dataset which poses the relevant question more closely and directly (more on which in Section 4.3). As a result, I use this survey to analyse the attitudes of the poor towards inflation and unemployment.

4.3

CLASS AND ATTITUDES TOWARDS INFLATION AND UNEMPLOYMENT

The radical political economy approaches to macroeconomics mentioned in the introduction saw unemployment as acting as ‘a regulator of class conflict’ (Rowthorn, 1977). Unemployment was the central fulcrum of the labor-capital confrontation. Specifically, increases in unemployment maintained a downward pressure on wages while tight labor markets increased factor income going to labor by exerting upward pressure on wages (Boddy and Crotty, 1975).7 Conflict theories of inflation suggested that the rise in inflation was seen as the result of perpetual claims by workers for wage increases ahead of productivity. The detrimental impact of inflation was primarily on those

76

Beyond inflation targeting

who had nominally denominated assets whose value was eroded with increasing prices, although there were also pressures on firms which were unable to pass on higher wages as higher prices. To that extent the negative impact of inflation was more pronounced on the capitalist and rentier class. As a result these theories would predict pronounced differences in attitudes towards inflation and unemployment depending on the respondent’s class position. Drawing from this political economy research, empirical studies in the 1970s and 1980s had begun to establish the class character of individuals’ preferences for inflation versus unemployment aversion. Given objective evidence that periods of relatively low unemployment and relatively high inflation coincided with an equalization of the personal distribution of income, a larger share of national income going to labor versus capital and a reduction in poverty as well as losses to those with savings in nominally denominated assets (typically the rich), attitudes towards inflation and unemployment had a class character. Hibbs (1977, p.1470) summarizes the central findings from US and UK surveys: Popular concern about unemployment and inflation is class-related. Low and middle income and occupational status groups are more averse to unemployment than inflation, whereas, upper income and occupational status groups are more concerned about inflation than unemployment . . . it does appear that the subjective preferences of class or status groups are at least roughly in accordance with their objective economic interests . . .

Recent studies of attitudes towards inflation and unemployment have largely ignored class. While there is substantial empirical evidence from opinion research that both inflation and unemployment are seen by respondents as disutilities (see, among others, Di Tella et al., 2001; Easterly and Fischer, 2001; Shiller, 1997), there is less consensus on the relative importance that individuals in different classes place on reducing each of these. Part of the issue is simply that there have been few surveys done which explicitly ask the respondent to rate their aversion to inflation versus unemployment if these were alternative outcomes. As such, the data has limited researchers’ agenda. Equally, there has been little interest in the characteristics of individuals who might support a policy designed to combat inflation versus one that tackled unemployment. To the extent that this has been undertaken, it has been to assess inflation aversion among the rich versus the poor (Easterly and Fischer, 2001; Jayadev, 2006).8 There has been little or no recent work which attempts to look at politics and in particular class politics in the determination of preferences towards anti-inflation and anti-unemployment policy. It is to this exercise that we now turn.

Anti-inflation and anti-unemployment policies

4.4 4.4.1

77

DATA Measuring Inflation Aversion

The International Social Survey Program (ISSP) conducted by the Interuniversity Consortium for Political and Social Research in 1996 focuses on the preferences of more than 30 000 individuals in 279 different countries regarding the role of government in society. The countries included include OECD economies, former eastern bloc economies and unfortunately do not include any low income countries apart from China.10 Among the questions asked in this survey is the following: If the government had to choose between keeping down inflation and keeping down unemployment to which do you think it should give highest priority?

This is the key variable of analysis for the rest of the chapter. I define a variable ‘relative inflation aversion’ as taking a value of one when the respondent prefers that the government prioritize reducing inflation rather than unemployment and zero when the opposite holds. Summary statistics for all variables used are provided in Table 4.1. While this provides a direct measure of the weights placed in an individual welfare function on inflation as opposed to unemployment, it is not without some limitations. Ideally, such a question might ask how much inflation the individual might accept for reducing the level of unemployment and vice versa so as to have a more direct calibration of the marginal rate of substitution in the social and individual welfare function. However, the measure is certainly superior to questions which ask about inflation without reference to unemployment or any other macroeconomic policy objective, thereby providing no implicit budget constraint. Figure 4.1 summarizes the average preference for keeping inflation rather than unemployment down by country. Some interesting observations suggest themselves. Nearly 42 percent of the overall sample report being relatively inflation averse. However, this masks large differences in the average relative inflation aversion between countries, from a low of less than 20 percent of respondents in France to a high of above 60 percent of respondents in the Czech Republic. In only five countries out of 20 is the percentage of relatively inflation averse respondents over half (and only in two countries – West Germany and the Czech Republic – is the percentage overwhelmingly above the midway mark). In order to assess the relative inflation aversion of the rich and the poor, I use the income question of the survey which asks the respondent their personal income in the last year. The income of the individual is grouped

Beyond inflation targeting

20

21.5

20.6

30

41.5

54.8

50.2

49.5

47.1

47.6

47.0

46.3

44.0

39.5

34.9

34.0

22.4

30.0

40

18.3

Percentage (%)

50

42.4

60

53.4

61.8

70

64.9

78

10

Is

ra

Fr

a el nce -A ra b Sp s ai n Ita ly La tv Ire ia la n C d hi C na an Au ada st ra R lia Is us r s U ael- ia ni te Jew d St s at N Po es ew l a Ze nd al a Sl nd ov en G er i m Ja a an pa y (E n G er H ast m un ) C any gar ze y ch (W R est ep ) ub O lic ve ra ll

0

Figure 4.1

Percentage of respondents who prefer that the government keeps inflation down rather than unemployment down

according to within country quintiles and quintile dummies take a value of 1 if the income of the respondent is in that quintile (and zero otherwise). I also define a ‘relative inflation aversion’ dummy that takes the value of 1 if the respondent prefers the government to keep down inflation rather than unemployment. 4.4.2

Measuring Class

Constructing readily comparable and objective measures of social class are fraught with difficulties (see Leiulfsrud et al. (2005) and Wright (1997) for an exposition on some of these). Contemporary measures of class differ based on the elements that researchers consider important to their theoretical approach – for example, along lines of ownership, management, career prospects, income, status, education or other such categories (Carchedi, 1977; Erikson and Goldthorpe, 1993; Esping-Andersen, 1992; Wright, 1985). However, not all of these are likely to bear directly upon the question of relative inflation aversion. Class has relevance in as much as it reflects the labor market position of the respondent and therefore their preferences to policies which enhances their position. It is desirable,

Anti-inflation and anti-unemployment policies

79

therefore, to utilize a definition which closely reflects the respondent’s occupation and position as employer or worker. In this chapter, I utilize a few different measures of class. The ISSP dataset provides an occupational variable based on the International Standard Classifications of Occupations (ISCO) 1988 classification which has previously been used by researchers to construct a wide array of measures of stratification. A first (crude) method is to utilize the traditional Marxist division between firm owner and employee. If the respondent is classified as self-employed with employees, they are classified as a firm owner, and an employee otherwise. I define a dummy variable called firm owner which takes the value of 1 if the respondent is self-employed and has more than one employee, and 0 otherwise. Sociologists have expended enormous effort in providing more useful and sophisticated categorizations of class. In order to perform a more satisfactory class analysis, I replicate Wright’s (1985) scheme which divides the labor force into owners and wage laborers and wage laborers in turn into three categories – experts, skilled and low skilled. Wright uses this to operationalize his idea of class locations and contradictions therein. Workers may be divided according to their relative privilege in the labor process. Based on a careful cataloging of occupations, Wright defines experts as those whose jobs require skills (and in particular accredited or credentialed skills) and who are in scarce supply relative to their demand by the market. Semi-skilled and unskilled class positions by contrast are held by those who have uncredentialed or no skills and who are thus in abundant supply. Using this approach has significant advantages. It makes theoretical sense for the question at hand to conceive of class measures which reflect the respondent’s relationship to the labor market and therefore to their bargaining power and probability of continued employment. An expert, for example, will typically enjoy a credential rent and be more likely to have both a higher level of bargaining power and a lower probability of being replaced than a low-skilled worker. They may therefore have opposing ideological and political interests based on other workers. At the same time both an unskilled worker and a skilled worker are more concerned about unemployment than a capitalist.11 Appendix 4.A1 details the creation of the class variables. Figure 4.2 shows the average relative inflation aversion in each country for each grouping of wage laborers. As is evident, in most countries ‘experts’ are more relatively inflation averse than semi-skilled and unskilled workers,12 as might be expected given the logic that more highly skilled workers enjoy greater bargaining power and a lower probability of unemployment than lower skilled workers, but are equally likely to see their wages eroded by inflation.

80

Beyond inflation targeting 90 Low-skilled workers Skilled workers Experts

80

Percentage (%)

70 60 50 40 30 20 10

st

ra l

C

C

Au

an

ia ad a ze ch Ch R ina Ea ep u st G blic er m an Fr y an H ce un ga I ry Is rela ra el nd Is Ara ra el bs Je w s N L ew at vi Ze a al an Po d la nd R us Sl sia ov ak ia U ni Sp te a d in W es Sta t G te er s m an y

0

Figure 4.2

Percentage of each category who are relatively inflation averse

Another approach is to look at subjective evaluations of class categories. The ISSP dataset asks respondents their own evaluation of their social class (the categories are lower middle class, upper working class, middle class, upper middle class and upper class). Unlike more objective measures, subjective perceptions of class probably depend on an amalgam of factors such as the respondent’s education, status, income and gender as well as factors which are of direct relevance to the question of relative inflation aversion – the respondent’s position and prospects in the labor market and the asset market. Nevertheless, it is useful to look at this as a check on the robustness of the earlier measure of class. I define three subjective class categories – the variable subjective lower class takes a value of 1 when the respondent identifies as being in the lower class or in the working class and 0 otherwise. Another variable subjective middle class takes a value of 1 if the respondent is from the lower middle class or middle class and 0 otherwise. Finally, the value of the variable subjective upper class takes a value of 1 when the respondent is from the upper middle or upper class.

4.5

RESULTS AND EXTENSIONS

In order to assess the income and class character of relative inflation aversion, I undertake a series of logistic regressions of relative inflation aversion on the respondent’s class position. Tables 4.2, 4.3, 4.4 and 4.5 provide the detailed results of these exercises.

Anti-inflation and anti-unemployment policies

Table 4.2

81

Logistic regression on the likelihood that a respondent will prefer the government to keep inflation down rather than unemployment down

Variable

Coefficient

z-statistic

Lowest income quintile Second income quintile Third income quintile Fourth income quintile Number of observations

0.56* 0.67* 0.76* 0.82* 14 340

(−10.5) (−6.6) (−4.8) (−3.4)

Note: * 5 significant at the 1 percent level. Omitted variable is richest quintile. Country intercept dummies are included but not shown.

Table 4.3

Logistic regression on the likelihood that a respondent will prefer the government to keep inflation down rather than unemployment down, with controls

Variable Lowest income quintile Second income quintile Third income quintile Fourth income quintile Some primary education Some secondary education Person in 20’s Person in 30’s Person in 40’s Person in 50’s Person in 60’s Person in 70’s Person above 70’s Unemployed Female Trade union member Voted for right wing party Number of observations

Coefficient

z-statistic

0.67* 0.75* 0.83* 0.87** 0.89** 0.86* 1.31** 1.52* 1.29* 1.25*** 1.43* 1.30** 1.26 0.82* 0.72* 0.81* 1.39* 14 340

(−6.5) (−4.5) (−3.3) (−2.3) (−2.1) (−3.1) (2.3) (3.6) (2.2) (1.9) (3.0) (2.0) (1.4) (−5.4) (−3.6) (−4.3) (7.6)

Note: * 5 significant at the 1 percent level, ** 5 significant at the 5 percent level, *** 5 significant at the 10 percent level. Omitted variable is ‘richest quintile’ for the income variables, ‘at least some university education’ for the education variables and ‘persons less than 20 years of age’ for the age variables. Country intercept dummies are included but not shown.

82

Beyond inflation targeting

Table 4.2 reports the results from a logistic regression of the relative inflation aversion dummy on the income quintile dummies. The fifth, or richest, quintile is the omitted variable, so that the coefficients on the income quintile dummies measure the difference between the coefficient on that income category and on the richest quintile. In order to control for country differences in the national averages of relative inflation aversion, the regression includes average individual country relative inflation as dummies, but these are not shown in the table. A strong and consistent result is immediately evident. Relative inflation aversion (or the likelihood of preferring the government to keep inflation down rather than unemployment down) is decreasing in the income of the respondent. The odds ratio falls monotonically with the fall in the income category, and the coefficients are all highly significant. The odds ratios suggest that a respondent in the lowest quintile is only about half as likely as someone in the richest quintile to display a preference for keeping inflation down rather than unemployment. Someone in the second and third quintile is about two-thirds and three-fourths as likely as someone in the top quintile to display relative inflation aversion, respectively. Table 4.3 repeats the analysis of Table 4.2 but controls for several other characteristics of the respondents. The first control is for the level of education. The prior expectation for this variable is unclear: a lower level of education makes an individual’s welfare potentially more vulnerable to both unemployment (lower human capital reduces employment opportunities) and inflation (less education means less knowledge and ability to protect one’s income from inflation). The coefficients from column 2 suggest, however, that less educated individuals are less relatively inflation averse than more educated individuals. While statistically significant, these effects are not large: both primary and secondary educated individuals are about 90 percent as likely as university educated people to prefer that the government keep down inflation rather than unemployment if it had to choose one of the two. A second control used is the age of the respondent. We may expect that the elderly are more concerned about inflation versus unemployment than the young, since the elderly are likely to be living off accumulated assets and pensions (which can easily be eroded through inflation) rather than wage income. The regression suggests that it is certainly the case that respondents in all age groups are more relatively inflation averse than those in the omitted group (those below age 20). However, those in their 60’s and 70’s do not appear to be more relatively inflation averse than those in their 20’s and 30’s. The last four rows report some interesting results. Not surprisingly, the unemployed are less likely than those who have employment to prefer

Anti-inflation and anti-unemployment policies

Table 4.4

83

Logistic regression of relative inflation aversion on class variables (without controls)

Variable

I Odds ratio

II Z-statistic

Firm owner 1.47*** (5.53) Unskilled Semi-skilled Subjective lower class Subjective middle class Number of observations 23 824

III

Odds ratio

Z-statistic

0.69*** 0.70***

(−6.66) (−6.44)

13 955

Odds ratio

Z-statistic

0.69***

(−6.37)

0.84***

(−3.15)

20 437

Note: *** 5 significant at the 1 percent level, ** 5 significant at the 5 percent level, * 5 significant at the 10 percent level. Omitted variable is experts (high skilled) in column II, and subjective upper class in column III. Country average inflation dummies are included but not shown.

that the government pursue a policy of keeping inflation down rather than unemployment. Women, too, are less relatively inflation averse than men and are about seven-tenths as likely as men to be so. Trade union members display statistically significantly less relative inflation aversion than nonunion members, perhaps because union members can more easily bargain their wages upwards than non-union members in the face of rising prices. Finally, political affiliation has an extremely strong predictive effect on the question at hand. Those who voted for the right wing party in the country in the last election are about 1.4 times as likely as those who did not to display relative inflation aversion. As the coefficients on the income quintile dummies show, the result that the poor are less likely than the rich to prefer that the government keep inflation down rather than unemployment remains true in the presence of controls. A respondent in the poorest quintile is about two-thirds as likely as one in the richest quintile to say so. This central result remains robust to the inclusion of several additional controls, such as occupation, household size and whether the individual works for the public or private sector. Moving now to considering the results on class or job attachment, columns I-III in Table 4.4 are the results from performing the logistic regression on the three definitions of class without any controls. As is evident from column I, firm owners are significantly more likely than workers to be relatively inflation averse (or less likely to be relatively

84

Table 4.5

Beyond inflation targeting

Logistic regression of relative inflation aversion on class variables (with controls)

Variable

I Odds ratio

Firm owner Unskilled Semi-skilled Subjective lower class Subjective middle class Lowest income quintile Second income quintile Third income quintile Fourth income quintile Unemployed Female Union member Age Number of observations

II

Z-statistic

1.20**

III

Odds ratio

Z-statistic

0.87** 0.82***

(−1.98) (−2.86)

Odds ratio

Z-statistic

0.75***

(−3.72)

0.86**

(−2.08)

(2.11)

0.63***

(−7.95)

0.66***

(−5.51)

0.68***

(−6.28)

0.73***

(−5.17)

0.75***

(−3.65)

0.76***

(−4.24)

0.80***

(−3.97)

0.81***

(−3.18)

0.85***

(−2.67)

0.86** 0.70*** 0.82*** 0.79*** 1.00

(−2.50) (−3.80) (−5.43) (−5.10) (0.97)

0.92 0.74** 0.80*** 0.81*** 1.00

(−1.17) (−1.86) (−4.75) (−3.90) (0.53)

0.89** 0.71*** 0.80*** 0.80*** 1.00

(−1.86) (−3.58) (−5.76) (−4.76) (1.24)

14 245

9 273

13 469

Note: *** 5 significant at the 1 percent level, ** 5 significant at the 5 percent level, * 5 significant at the 10 percent level. Omitted variable are experts (high skilled) and highest income quintile in column II, and subjective upper class and highest income quintile in column III. Country average inflation dummies are included but not shown.

unemployment averse). Column II uses the classification for wage laborers developed by Wright. The omitted dummy is expert workers and hence the results suggest that as compared to experts, both low-skilled and skilled workers display less relative inflation aversion suggests that in comparison with the omitted group. Similarly, in column III in comparison to those who consider themselves upper class, those who consider themselves middle and lower class are much less relatively inflation averse. Columns I-III in Table 4.5 show that these results persist in the presence of a variety of plausible controls, including dummies for income, gender, age, employment status and union membership. Column I shows

Anti-inflation and anti-unemployment policies

85

that a firm owner is about a fourth more likely to prefer anti-inflation to anti-unemployment policies as a worker. Column II shows that relative to ‘experts’, semi-skilled are about eight-tenths and unskilled workers are about nine-tenths as likely to report supporting anti-inflation to antiunemployment policies. The results in column III show that as compared to the subjective upper class, subjective lower classes display significantly less inflation aversion. A respondent who considers themself as being part of the lower or lower middle class is about three-fourths as likely as someone who is in the upper class to prefer that the government keep inflation down. A respondent in the middle or upper middle class is about eight-tenths as likely as someone in the upper middle or upper class to prefer that inflation be kept down rather than unemployment. What kinds of other policy preferences are associated with relative inflation aversion? If the class content narrative is correct and not simply an artifact of the occupational categorization used in this chapter to define class, we should find that there are other attitudes in common with respect to class politics. To test this, I use questions from the ISSP dataset which speak to the respondent’s stance towards pro-business and pro-labor policies and attitudes, on the one hand, and towards redistributionary policies on the other. The dataset provides a host of questions which could be used for the purpose and while those used here are not exhaustive, using other indicators provides very similar results. In order to gauge attitudes towards labor and towards business I construct several dummy variables which could be said to represent anti-labor and pro-business sentiment. I define a dummy variable pro wage control when the respondent replies being strongly in favor or in favor of wage control by law. Another dummy variable anti jobs for all refers to the situation when the respondent believes that it should not be the government’s responsibility to provide a job for everyone who wants one. If the respondent is against the idea of the government financing of new jobs then they are said to be anti government jobs. The variable pro deregulation refers to the situation the respondent is for government deregulation of business. Anti protect jobs takes a value of 1 if the respondent is against the protection of declining industries in order to protect jobs. Finally the dummy variables power of labor and power of business refer to situations where the respondent believes the political power of labor and the political power of business to be too strong respectively. Column I in Table 4.6 shows the results of a logistic regression of the inflation aversion variable on these dummies. In concordance with the narrative of class-based differences, all the odds ratios are statistically significant and in the expected direction. There is a strong correspondence between pro-business and anti-labor preferences, on the one hand, and relative inflation aversion, on the other.

86

Beyond inflation targeting

Table 4.6

Logistic regression of relative inflation aversion on other attitudes

Variable

I

II

Odds ratio

Z-statistic

1.10** 1.51*** 1.49*** 1.37*** 1.56*** 1.12*** 0.86***

(2.55) (10.03) (5.39) (2.70) (9.96) (2.97) (−3.22)

Pro wage control Anti jobs for all Anti government jobs Pro deregulation Anti protect jobs Power of labor Power of business Anti redistribution Anti equalization Lower social services Number of observations

14 533

Odds ratio

Z-statistic

1.32*** (6.28) 1.45*** (8.98) 1.40*** (10.26) 17 743

Note: *** 5 significant at the 1 percent level, ** 5 significant at the 5 percent level, * 5 significant at the 10 percent level. Country average inflation dummies are included but not shown.

Attitudes towards redistribution are approximated by three dummies. A respondent is anti redistribution if they respond that the government should not redistribute wealth, and is anti equalization if they respond that it is not the government’s responsibility to reduce income differences between the rich and the poor. If the respondent prefers reducing taxes even if this means lower social services then the dummy variable lower social services takes a value of 1. Column II in Table 4.6 shows the results of a logistic regression of the inflation aversion variable on these dummies. The odds ratios are all statistically significant and show that relative inflation aversion is stronger among respondents who are also against government action to redistribute income and wealth and to provide social services at the potential cost of higher taxation.

4.6

CONCLUSION

In the last three decades the hopeful message of Keynesian demand management has fallen out of favor with policy makers; as more faith has been placed in market-based solutions, in independent central banks and in microeconomic interventions to handle the problems of inflation and unemployment. The theoretical impetus for this shift was provided by the rational expectations revolution of the 1970s and subsequent new

Anti-inflation and anti-unemployment policies

87

classical analyses of the state. These in turn argued that democratic governments were often bad for macroeconomic efficiency as they would tend to increase deficits and be unable to credibly tackle inflationary pressures. Worse still, any attempt to artificially reduce the unemployment rate would lead inevitably to higher inflation with no effect beyond the very short term on the unemployment rate. Thus, the standard prescription of earlier times was replaced by the orthodoxy of central bank independence and increasingly a narrow focus on inflation targeting (Bernanke et al., 1999). For proponents of this view, delegating responsibility to an authority which can credibly commit to a single target is beneficial in increasing macroeconomic efficiency and protecting the public from a perhaps well meaning, but misguided government. A movement away from commitment to full employment and towards low targeted inflation has potentially profound distributional consequences. Despite the claims made by some that anti-inflationary policy is, for example, pro poor, it is an empirical question as to whether inflation or unemployment is seen as a bigger problem by different individuals if there is a trade-off between these. Both inflation and unemployment are more likely, for example, to affect the poor, but this study as well as recent research on inflation versus unemployment aversion (Jayadev, 2006; Scheve, 2004) found that the rich are more relatively inflation averse than the poor. Other results presented in this chapter further strengthen this idea by finding that relative inflation aversion is more pronounced among the privileged or elite broadly defined in class terms. Since class is a critical variable in determining an individual’s labor market opportunities as well as the source and variability of their income, the results make sense. The findings are in concordance with conflict-based models of unemployment and inflation which argue that macroeconomic policies may have systematically differential effects on the welfare of workers and owners (as well as on different segments of the working class) and that the preferences of individuals in separate class positions reflect these differences. These findings have important consequences for research on the implications of antiinflation versus anti-unemployment policies in general and on the more current debate around inflation targeting in particular.

NOTES 1. 2.

The author is grateful to Gerald Epstein, Christian Weller, Sam Bowles, Kade Finnoff, Ozgur Orhangazi, James Heintz, Suresh Naidu and Erinc Yeldan for useful comments. Indeed, as Pollin (1998) points out, class conflict is the implicit mechanism that drives the natural rate of unemployment even in orthodox neoclassical accounts. As he puts

88

3. 4. 5.

6. 7.

8. 9. 10. 11. 12.

Beyond inflation targeting it ‘Marx and Kalecki also share a common conclusion with natural rate proponents, in that they would all agree that positive unemployment rates are the outgrowth of class struggle over distribution of income and political power. . . . Of course, Friedman and the New Classicals reach this conclusion via analytic and political perspectives that are diametrically opposite to those of Marx and Kalecki. To put it in a nutshell, mass unemployment results in the Friedmanite/New Classical view when workers demand more than they deserve, while for Marx and Kalecki, capitalists use the weapon of unemployment to prevent workers from getting their just due’ (Pollin, 1998). The idea of unemployment as a labor disciplining device has, of course, a provenance from Marx. Marxist theories of the labor market maintain this as a key fact of the labor market (for example, Bowles, 1985). For a recent review see Ball and Mankiw (2002). At the outset, it is important to note a potentially serious problem with the data. The choice for inflation as worded in the survey is for ‘inflation and high prices’. As Easterly and Fischer themselves note, this raises the problem that in complaining about high prices, the poor are only naturally complaining about low incomes. They suggest that the low correlation between this variable and another option available to the respondent ‘money not enough to live by’ is proof that the poor are indeed protesting inflation and not low income. This is not, however, truly satisfactory. It is not entirely clear that a poor respondent would list both ‘inflation and high prices’ and ‘money not enough to live by’ as being within the top three problems facing the economy if they interpret the first option as being about their low income. Indeed it is often spoken about as being a cruel ‘tax’, whereby the public’s real earnings are eroded and transferred to the state. Recent empirical studies, especially about the US experience of the 1990s boom, support these ideas indirectly. Bernstein and Baker (2003) find that the low unemployment period in the US economy of the late 1990s aided in improving the welfare of workers according to several metrics. Real wages increased after a generation of decline, and the inflation adjusted income of low income families grew by twice the amount that they did in the 1980s expansion (when average unemployment was higher). Abraham and Haltiwanger (1995) who review mainly US evidence suggest that real wages are more likely to be pro-cyclical than counter-cyclical. Scheve (2003, 2004) remains an exception in providing more detailed evidence for the characteristics of individuals supporting each policy. In the survey respondents from Israel and Germany are divided in two separate categories each. The former is divided between Israeli Arabs and Israeli Jews and the latter is split between East and West Germans. Survey was conducted during 2006. This is an example of what Wright terms a contradictory class location. The major exception is Israel – perhaps because of the impact that the period of hyperinflation in the 1970s had upon even skilled workers.

REFERENCES Abraham K.G. and J.C. Haltiwanger (1995), ‘Real wages and the business cycle’, Journal of Economic Literature, 33, 1215–64. Bach, G. and J. Stephenson (1974), ‘Inflation and the redistribution of wealth’, Review of Economics and Statistics, 61, 1–13. Ball, L. and G.N. Mankiw (2002), ‘The NAIRU in theory and practice’, Johns Hopkins University working papers no. 475. Barro, R. and D. Gordon (1983), ‘A positive theory of monetary policy in a natural rate model’, Journal of Political Economy, 31, 589–610.

Anti-inflation and anti-unemployment policies

89

Bernstein, J. and D. Baker (2003), The Benefits of Full Employment: When Markets Work for People, Washington, DC: Economic Policy Institute. Bernanke, B., T. Laubach, F. Mishkin and A.S. Posen (1999), Inflation Targeting: Lessons from the International Experience, Princeton, NJ: Princeton University Press. Boddy, R. and J. Crotty (1975), ‘Class conflict and macro-policy: the political business cycle’, Review of Radical Political Economics, 7, 1–19. Bowles, S. (1985), ‘The production process in a competitive economy: Walrasian, non-Hobbesian, and Marxian models’, American Economic Review, 75 (1), 16–36. Carchedi, G. (1977), The Economic Identification of Social Class, London: Routledge and Kegan Paul. Cardoso, Eliana (1992), ‘Inflation and Poverty’, National Bureau for Economic Research working paper no. 6793, March, accessed at http://ssrn.com/ abstract5293237. Cukierman, A. (1992), Central Bank Strategy, Credibility, and Independence, Cambridge, MA: MIT Press. Di Tella, R., R. MacCulloch and A. Oswald (2001), ‘Preferences over inflation and unemployment: evidence from surveys of happiness’, American Economic Review, 91, 335–41. Easterly, W. and S. Fischer (2001), ‘Inflation and the poor’, Journal of Money, Credit, and Banking, 33 (2), 160–78. Erikson, R. and J.H. Goldthorpe (1993), The Constant Flux, Oxford: Oxford University Press. Esping Andersen, G. (ed.) (1992), Changing Classes: Stratification and Mobility in Post-Industrial Societies, London: Sage. Hibbs, D. (1977), ‘Political parties and macroeconomic policy’, American Political Science Review, 71, 1467–87. Iversen, T. and D. Soskice (2001), ‘An asset theory of social preferences’, American Political Science Review, 95 (4) (December), accessed at www.people.fas. harvard.edu/~iversen/data/ISCO_conversion_tables.htm. Iversen, T. and D. Soskice (2006), ‘New macroeconomics and political science’, Annual Review of Political Science, 9, 425–53. Jayadev, A. (2006), ‘Differing preferences between anti-inflation and antiunemployment policy among the rich and the poor’, Economics Letters, 91, 67–71. Jayadev, A. (2008), ‘The class content of preferences towards anti-inflation and anti-unemployment policies’, International Review of Applied Economics, 22 (2), 161–72. Johnson, H. (1968), ‘Problems of efficiency in monetary management’, Journal of Political Economy, 76, 986. Leiulfsrud, H., Bison, I. and H. Jensberg (2005), ‘Social class in Europe’, European Social Survey 2002/3, NTNU Social Research Ltd., unpublished, accessed http://ess.nsd.uib.no/files/2003/ESS1SocialClassReport.pdf. Pollin, R. (1998), ‘The natural rate of unemployment: it’s all about class conflict’, Dollars and Cents Magazine, (September/October). Romer, Christina D. and David H. Romer (1998), ‘Monetary policy and the wellbeing of the poor’, National Bureau for Economic Research working paper no. 6793. Rosenberg, S. and T. Weisskopf (1981), ‘A conflict theory approach to inflation in

90

Beyond inflation targeting

the postwar US economy’, American Economic Review Papers and Proceedings, 71 (2), 42–7. Rowthorn, R.E. (1977), ‘Conflict, inflation and money’, Cambridge Journal of Economics, 1 (3), 215–39. Scheve, K. (2003), ‘Public demand for low inflation’, Bank of England working paper no. 172, London. Scheve, K. (2004), ‘Public inflation aversion and the political economy of macroeconomic policymaking’, International Organization, 58 (1), 1–34. Shiller, R. (1997), ‘Why do people dislike inflation?’ in C. Romer and D. Romer (eds), Reducing Inflation: Motivation and Strategy, Chicago, IL: University of Chicago Press, pp. 13–71. Wright, E.O. (1985), Classes, London: New Left Books. Wright, E.O. (1997), Class Counts: Comparative Studies in Class Analyses, Cambridge: Cambridge University Press.

Anti-inflation and anti-unemployment policies

APPENDIX 4.A1

91

CLASS MEASURES

The ISSP dataset provides ISCO-88 classifications for all but four of the countries. For each of these countries we have ISCO-68 classifications. Iversen and Soskice (2001) provide a bridge between these coding mechanisms based on previous work by Ganzebloom. Using this code (available from Iversen’s webpage at the Harvard School of Government), I recode all respondents as per ISCO-88 codes. I drop the 300 or so observations for which there is no bridge available. Wright (1997) provides a mechanism by which to classify ISCO-88 classifications into three skill categories, experts, skilled and low skilled workers. Using the codes provided by Leilsfrud et al. (2005), I replicate these categorizations for the ISSP dataset. The code is available upon request. Table 4.X

Summary statistics

Variable

Source

Inflation down Expert Semi-skilled Unskilled Subjective lower class Subjective middle class Subjective upper class Firm owner

First income quintile Second income quintile Third income quintile Fourth income quintile

Obs

Mean

Std dev.

Min.

Max.

V63: ISSP Constructed from V202: ISSP Constructed from V202: ISSP Constructed from V202: ISSP

23 824 35 313

0.41 0.07

0.49 0.25

0 0

1 1

35 313

0.23

0.42

0

1

35 313

0.24

0.43

0

1

V221: ISSP

35 313

0.31

0.46

0

1

V221: ISSP

35 313

0.49

0.50

0

1

V221: ISSP Constructed from V213 and V214: ISSP dataset Constructed from V217: ISSP Constructed from V217: ISSP Constructed from V217: ISSP Constructed from V217: ISSP

35 313 35 313

0.07 0.04

0.26 0.19

0 0

1 1

35 313

0.15

0.36

0

1

35 313

0.11

0.31

0

1

35 313

0.13

0.34

0

1

35 313

0.11

0.31

0

1

92

Table 4.X

Beyond inflation targeting

(continued)

Variable

Source

Fifth income quintile Union member Female Unemployed Age Pro wage control

Constructed from V217: ISSP V222: ISSP V200: ISSP V206: ISSP V201: ISSP Constructed from V17: ISSP Constructed from V36: ISSP Constructed from V20: ISSP Constructed from V21: ISSP Constructed from V23: ISSP Constructed from V33: ISSP Constructed from V34: ISSP Constructed from V16: ISSP Constructed from V42: ISSP Constructed from V56: ISSP

Anti jobs for all Anti government jobs Pro deregulation Anti protect jobs Power of labor Power of business Anti redistribution Anti equalization Lower social services

Obs

Mean

Std dev.

Min.

Max.

35 313

0.10

0.30

0

1

35 313 0.18 35 228 0.52 35 313 0.06 35 109 44.82 33 798 0.33

0.38 0.50 0.23 16.65 0.47

0 0 0 15 0

1 1 1 97 1

33 820

0.27

0.44

0

1

34 188

0.06

0.24

0

1

28 350

0.92

0.28

0

1

34 035

0.21

0.40

0

1

30 614

0.65

0.48

0

1

29 484

0.84

0.37

0

1

33 542

0.24

0.43

0

1

33 096

0.30

0.46

0

1

26 652

0.52

0.50

0

1

5.

The gendered political economy of inflation targeting: assessing its impacts on employment1 Elissa Braunstein and James Heintz

5.1

INTRODUCTION

Central banks in developing countries are increasingly maintaining low inflation rates as the central goal of monetary policy, without much consideration of how these policies impact real outcomes like employment, investment and growth (Epstein, 2003). Although targeting very low inflation rates seems to have done little to raise economic growth, these policies remain a key feature of neoliberal approaches to monetary policy (Epstein, 2000). Gerald Epstein and Juliet Schor argue that anti-inflation policy and neoliberal approaches to central banking emerged from the ‘contested terrain’ of central banks – the class and intra-class conflicts over the distribution of income and power in the macroeconomy (Epstein, 2000; Epstein and Schor, 1990). Their work underscores the importance of understanding monetary policy from a political economy perspective, as the distribution of the gains and costs of economic policy proffers insight into both a policy’s genesis and its longer term consequences. In this chapter we build on their analysis by considering the employment costs of inflation reduction in developing countries from a gender perspective. We explore two empirical questions: (1) what is the impact, if any, of inflation reduction on employment, and is the impact different for women and men, and (2) how are monetary policy indicators connected to deflationary episodes and gender-specific employment effects? We find a common pattern among countries undergoing what we term contractionary inflation reduction, or periods of declining inflation that are accompanied by a loss of employment. After controlling for long-term employment trends, we find that the ratio of women’s to men’s employment tends to decline during these periods in the majority of countries examined. During periods of expansionary inflation reduction, however, there are no clear patterns to the relative changes in women’s and men’s 93

94

Beyond inflation targeting

employment. In terms of monetary policy, we find that countries that respond to inflation by raising real interest rates or tightening the real money supply (both relative to their long-run trends) are also more likely to experience employment contractions, with concomitantly higher costs for women’s employment. Conversely, those that maintain competitive real exchange rates are likely to reverse the negative impact of contractionary inflation reduction on women’s relative employment. That the costs of inflation reduction, at least in terms of employment, are inequitably distributed means that the contested terrain of monetary policy is also gender specific, with the result that the costs of implementing these sorts of policies are actually quite different – and potentially higher – than is generally presumed. After a discussion of the literature on the differences in women’s and men’s unemployment in developing countries, and presenting our empirical results, we will develop this last point in more detail in the closing section.

5.2

GENDER DIFFERENCES IN EMPLOYMENT AND UNEMPLOYMENT IN SEMIINDUSTRIALIZED COUNTRIES

In thinking about the differences between women’s and men’s unemployment, it is helpful to think in terms of supply-side factors and demand-side factors (Seguino, 2003). On the supply side, differences in human capital are probably the most commonly considered. However, gender-based differences in education, skill and experience are themselves rooted in workers’ productive roles outside the factory door and the institutional, social and material contexts in which they live. A useful way of considering these differences is through what Folbre (1994) terms ‘structures of constraint’ – the preferences, norms, assets and rules that shape individual choice. Women and men make decisions about whether or not to look for wage work. However, self-perception, what individuals value, and what choices they perceive as possible are constituted by the social world, and are different for women and men (Sen, 1990), and so individual preferences must be understood in this light. Norms are the traditional structures of gender and kinship that constitute the social expectations of women and men in the household. For example, women are primarily associated with the care of the family, and much of their work time is spent outside of the market, whereas men’s work is typically viewed as directly productive and more fully incorporated into the market sphere. Assets also engender labor supply. Systematic differences by gender in ownership of assets are

The gendered political economy of inflation targeting

95

common, and partly determine how much wage employment women seek. Similarly, laws that confer patriarchal property rights, where eldest men have the right to claim and apportion the fruits of household members’ labor time, can create incentives for high fertility and lower female labor force participation (Braunstein and Folbre, 2001). Conversely, not having a legal claim on a spouse’s income in the event of separation means that a paying job can be an insurance policy against loss of that support (Folbre, 1997). On the demand side, gender discrimination results in differential access to employment opportunities. The presumption that men should bear the primary financial responsibility for provisioning families has been linked with higher unemployment for women relative to men in OECD countries (Algan and Cahuc, 2004). Women are laid off first because employers presume that it is more important for men to be able to fulfill their traditional breadwinning responsibilities (Azmat et al., 2004). Differential hiring practices also contribute to gender segregation in employment. For example, labor-intensive exporters prefer to hire women both because women’s wages are typically lower than men’s, and because employers consider women to be more productive in these types of jobs due to employer perceptions that women possess ‘nimble fingers’, are less prone to worker unrest, are better suited to tedious work and are more reliable workers than men (Anker and Hein, 1985; Elson, 1996; Elson and Pearson, 1981; Fernández-Kelly, 1983).2 By extension, women may lose their comparative advantage when industries upgrade, leading to a defeminization of employment as has happened in Mexico, India, Ireland and Singapore (Elson, 1996; Fussell, 2000; Ghosh, 2001; Joekes, 1999). As a result of these gender differences in labor demand and supply, changes in macroeconomic structure and policy have differential effects on men’s and women’s work (Seguino, 2003). For example, feminists have long argued that the interaction between gender relations and structural adjustment programs (SAPs) have implications both for the distribution of costs and benefits between different groups of women and men, and for the achievement of the economic objectives of the SAPs themselves (Bakker, 1994; Benería and Feldman, 1992; Benería and Roldan, 1987; Cagatay et al., 1995; Elson, 1991, 1995). Macroeconomic policies and variables have been shown to influence women’s labor supply decisions. Cagatay and Ozler (1995) pool cross country data for 1985 and 1990 to show that SAPs led to increased feminization of the labor force via worsening income distribution and openness. Others have also argued that, in the case of Latin America, macroeconomic crises place pressures on household resources which increase labor force participation and hours worked, particularly for women (Arriagada,

96

Beyond inflation targeting

1994; Cerrutti, 2000). Similarly, research into the determinants of women’s labor supply in post-apartheid South Africa has shown that women’s labor force participation responds positively to growing unemployment, thereby further increasing the country’s average unemployment rate (Casale, 2003). Trade liberalization and structural adjustment policies can cause a reallocation of paid employment from traditional activities, which are adversely affected, to export-oriented production, which is encouraged (Cagatay and Ozler, 1995; Standing, 1999). In addition, female-intensive export-oriented industries are often more cyclically volatile than men’s industries, resulting in higher overall rates of unemployment (Howes and Singh, 1995). Emphasis on export-oriented industrialization has also been associated with increases in informalization of certain types of employment due to increases in competitive pressures (Carr et al., 2000; Standing, 1999). As female labor force participation rises in the context of crisis and structural adjustment, the increasing dominance of informal work has become a key new reality for women (Arriagada, 1994; Benería, 2001; Patnaik, 2003). Similar work has been done on the gendered employment effects of the Asian financial crisis in 1997–98 (Aslanbeigui and Summerfield, 2000; Lim, 2000; United Nations, 1999). Women were typically the first to be laid off both because they worked in more cyclically volatile firms, such as small export-oriented enterprises, and because of efforts to protect the jobs of ‘male breadwinners’. The lack of formal employment opportunities and pressures for greater labor force participation led to an increase in women’s informal employment in some cases. A slightly different pattern was found by Lim (2000) in the Philippines, where the post-crisis decline increased male unemployment more than female unemployment despite a rapid displacement of women from the manufacturing sector (especially in traded goods). The reason was the relative resilience of the service and trade sectors, which are more female intensive. Women did, however, increase their labor force participation to deal with male unemployment, and their total work hours relative to men increased as well. These gender dynamics have implications for macroeconomic outcomes. For example, increases in women’s employment that are associated with lower unit labor costs have important implications for inflation dynamics. For the non-tradable and import-competing sectors in which some degree of competition exists, lower unit labor costs associated with growth in women’s employment reduce inflationary pressures in the countries where the women work. Clearly, there are significant structural differences between women’s and men’s labor markets on both the supply and demand sides that are

The gendered political economy of inflation targeting

97

differently affected by macroeconomic structure and policy. The literature reviewed above on semi-industrialized countries suggests that economic contractions have a larger negative effect on women’s formal employment than men’s, though women tend to increase their labor force participation at the same time to protect household income. In the next section we turn to a more direct test of this employment hypothesis, looking at whether the process of inflation reduction in particular can be associated with gender differentials in employment.

5.3

INFLATION REDUCTION AND WOMEN’S EMPLOYMENT

The empirical exercise we present here explores the effects of inflation reduction on women’s and men’s formal employment. We compiled data for 51 ‘inflation reduction episodes’ in 17 low- and middle-income countries.3 To assess the employment effects of inflation reduction periods, we looked at actual employment trends during each inflation reduction episode, disaggregated by gender, and compared these to long-run employment trends, estimated by applying a Hodrick-Prescott filter to the employment series. We also examine indicators that suggest how monetary policy responded during inflation reduction episodes using a similar approach. We compare average short-term real interest rates, growth rates of the real money supply, and indicators of the real exchange rate to their long-run trends to see if these variables deviated from trend during inflation reduction episodes. We drew from three different data sources in conducting this exercise: employment data came from the International Labor Organization’s (ILO) LABORSTA online database, and the macroeconomic data was compiled from World Development Indicators 2005 (Washington, DC: World Bank) and International Financial Statistics (Washington, DC: IMF, October 2005).4 The methodology used is drawn from the literature on measuring ‘sacrifice ratios’ – that is, the loss of output or employment associated with a given reduction in inflation. Sacrifice ratios can be measured simply as the slope of the Phillips Curve showing an employment-inflation trade-off. However, this approach assumes that the sacrifice ratio remains constant over long periods of time. Other approaches examine changes in employment or output over specific deflationary periods. This technique allows the sacrifice ratio to change over time and makes it possible to analyse the behavior of economic variables during specific deflationary periods. This is the approach followed here. Ball (1993) suggests one method for identifying deflationary periods.

98

Beyond inflation targeting 100

14 Interest rate

12

80 10

70 60

8

50 6

40 30

Interest rate (percent)

Percent of sample countries undergoing disinflation

90

4

20 2 10

Figure 5.1

2001

1999

1997

1995

1993

1991

1989

1987

1985

1983

1981

1979

1977

1975

1973

0 1971

0

Inflation reduction episode frequencies and the real US interest rate, 1971–2003

Inflation figures are notoriously volatile, making the identification of turning points difficult. A moving average of inflation – in our case, a threeyear moving average, encompassing one previous year and one subsequent year – can be used to smooth the series. Peaks and troughs in the smoothed inflation series are identified. Peaks occur when the value in a particular year exceeds the values of immediately adjacent years. Troughs occur when the value in a given year falls below the values of the adjacent years. A deflationary period runs from a peak year to the next trough year. For the purposes of this exercise, we use the term ‘inflation reduction episode’ to refer to these deflationary periods. The reason for this is that, during some of the periods identified, employment actually expands more rapidly than its long-run trend. It seems confusing to refer to these periods as ‘deflationary’. Therefore, we use the terms ‘expansionary inflation reduction episode’ and ‘contractionary inflation reduction episode’. The inflation reduction episodes identified for the 17 countries examined are remarkably similar in terms of the timing of the episodes. That is, inflation reduction episodes occur in many of these countries simultaneously. Figure 5.1 illustrates this pattern. The figure shows, for each year

The gendered political economy of inflation targeting

99

1970–2003, the percent of all countries in our study in which an inflation reduction episode occurred. Inflation reduction episodes are concentrated over certain sub-periods: 1974–77, 1980–86, 1991–94 and 1997–2000. Also Figure 5.1 clearly demonstrates that inflation reduction episodes were much more common throughout the 1990s than throughout the prior two decades, the 1970s and 1980s. This pattern of inflation reduction suggests that one or more common factors determine inflation rates and monetary variables in the different countries examined. One obvious possibility is the world interest rate. Figure 5.1 shows the real yield on US Treasury bills over the entire period as a proxy for short-term, inflation-adjusted world interest rates. An increase in the real yield of US Treasury bills preceded three out of the four common inflation reduction episodes. Only in the most recent episode, from approximately 1997 to 2000, did interest rates fail to increase beforehand.5 Because the data for the different countries show similar trends in the variables analysed, we emphasize deviations from long-run trends, rather than absolute levels or changes, in order to investigate common patterns and divergent trends. We restricted the countries used as follows: ● ●

●

Only low- or middle-income countries were examined.6 Countries must have at least 20 years of gender-disaggregated employment data (it would be hard to estimate a meaningful ‘longrun trend’ with a shorter series). Time series with missing values were used. However, time series with two consecutive missing observations were rejected. Missing values were estimated for the purposes of computing a long-run trend by extrapolating between the previous and subsequent values in the series.

Changes in employment across inflation reduction episodes were calculated as the annualized value of the overall rate of change in employment across the entire peak-to-trough period. As mentioned earlier, values for the long-run employment trends were computed by applying a HodrickPrescott filter to the actual employment time series for each country (men, women and total). The employment time series used most likely underestimates the magnitude of informal employment in these countries. Therefore, the results of this analysis should be interpreted bearing this in mind. Table 5.1 summarizes the results for all the inflation reduction episodes studied. The table shows the country name, the dates of each inflation reduction episode, and the deviation from the long-run trend for women’s

100

Table 5.1

Beyond inflation targeting

Inflation reduction episodes and deviations from long-run employment trends, disaggregated by gender A. Contractionary inflation reduction episodes period

Barbados

Brazil Colombia Costa Rica India

Jamaica Kenya Malaysia Mauritius

Philippines

Singapore South Korea

Sri Lanka Taiwan

Thailand

Trinidad and Tobago

Deflation as percent of avg. inflation

Deviations from long-run employment trends (percent) W

M

Ratio

1980–86 1990–94 1996–99 1993–99 1980–85 1982–85 1973–77 1982–86 1991–94 1997–02 1974–76 1992–2000 1975–80 1981–87 1981–86 1980–86 1989–93 1994–96 1973−76 1980−82 1984–87 1974−76 1981–86 1980–85 1991–94 1997–2000 1981–86 1997–99 1974–76 1980–85 1991–02 1974–76 1980–85 1990–93 1997–00

−143 −113 −117 −305 −29 −123 −175 −37 −26 −93 −45 −190 −36 −66 −181 −182 −46 −21 −69 −55 −130 −171 −267 −196 −45 −82 −106 −56 −106 −243 −197 −88 −180 −31 −136

−1.8 −2.5 −1.0 −0.7 −3.2 −1.5 −0.2 −0.1 0.1 −1.0 −0.5 −0.5 −2.2 0.8 −0.4 −0.6 −1.3 −1.8 −1.6 0.2 −2.4 −6.7 −1.8 −1.4 −0.4 −1.2 −0.7 −0.7 −4.9 0.6 −0.4 −1.3 −2.6 −0.8 −0.8

−0.8 −2.6 0.3 −0.1 −2.5 −0.1 −0.4 −0.2 −1.2 0.9 −0.2 0.1 −0.1 −0.3 −0.8 −1.6 –0.3 −0.9 −0.4 −0.3 0.0 −0.7 −2.0 −0.9 0.0 −1.1 0.1 −2.6 0.3 −0.5 −0.2 −0.8 −0.7 0.1 −0.7

−1.0 0.1**** −1.4 −0.6 −0.7 −1.4 0.2**** 0.1**** 0.2**** −0.2 −0.3 −0.6 −2.1 1.1**** 0.4**** 0.9**** −0.9 –0.8 –1.2 0.5**** –2.3 −5.9 0.1**** −0.5 −0.4 −0.1 −0.8 1.9**** −5.1 1.1**** 0.2**** −0.5 −1.8 −0.9 −0.1

1980–87

−56

−1.0

−0.6

−0.4

The gendered political economy of inflation targeting

Table 5.1

101

(continued)

Brazil Chile Costa Rica Jamaica Kenya Malaysia Mauritius Philippines Singapore Sri Lanka Trinidad and Tobago

1989–92 1984–88 1991–93 1979–82 1985–88 1993−96 1992–96 1974–77 1990–94 1990–99 1974–76 1989–94 1974–77 1989–92 1993–96

−39 −39 −42 −101 −90 −134 −27 −59 −60 −139 −100 −50 −48 −37 −69

1.9 0.8 0.2 0.1 2.9 0.9 0.6 3.9 0.2 0.1 1.9 8.9 0.2 3.8 0.7

−0.8 2.3 1.6 0.5 0.5 −0.3 1.4 1.6 0.3 0.1 3.1 3.6 1.0 1.6 0.6

2.8**** −1.5 −1.4 −0.4 2.3**** 1.3**** −0.8 2.1**** −0.1 0.0 −1.1 4.9**** −0.8 2.1**** 0.2****

Note: **** Inflation reduction episodes in which the ratio of women’s to men’s employment increased more rapidly than the long-run trend.

Summary: Contractionary inflation reduction episodes with declining female:male employment ratios relative to the trend: 67 percent. Contractionary inflation reduction episodes with declining female:male employ ratios relative to the trend (excluding India): 72 percent. Expansionary inflation reduction episodes with increasing female:male employment ratios relative to the trend: 53 percent.

employment, men’s employment, and the female to male employment ratio. Negative values indicate that the series grew more slowly than the long-run trend (a negative value could also indicate a more rapid decrease in the actual value compared to the long-run trend). Table 5.1 is divided into contractionary and expansionary inflation reduction episodes. During contractionary inflation reduction episodes the rate of increase of total employment fell below its long-run trend. During expansionary inflation reduction episodes, the rate of increase of total employment was equal to or greater than its long-run trend. In 67 percent of contractionary inflation reduction episodes the rate of change of the female to male employment ratio fell below its longrun trend, indicating that women’s employment was disproportionately affected by the slowdown. If India were excluded from the sample, this proportion would rise to over 72 percent.7 However, in expansionary

102

Beyond inflation targeting

inflation reduction episodes there was no clear distinction. The female to male employment ratio increased faster than trend in 53 percent of cases and at or below trend in 47 percent of cases – nearly an even split. The contractionary inflation reduction episodes were associated with a larger decline in inflation from peak to trough relative to the average inflation rate for the entire episode. Averaging across all contractionary episodes, the decline in inflation was 115 percent of the average inflation rate during the episode in question. For expansionary episodes, the decline was 69 percent. The difference in employment experiences across countries during inflation reduction episodes – for example, expansion or contraction of employment – might be explained, in part, by policy choices. For example, if real interest rates rose above the long-run trend in reaction to an acceleration of inflation, this could trigger a contractionary inflation reduction episode. However, if real interest rates were not raised above the long-run trend (for example, they were kept in line with the long-run trends of global interest rates), a contraction of employment might be avoided. To examine this possibility, we looked at average real short-term interest rates across the inflation reduction episodes.8 In most cases short-term rates linked directly to monetary policy choices were used (for example, a discount rate or bank rate). If these rates were unavailable, yields on short-term (three-month) Treasury bills were calculated instead.9 If actual average real interest rates were negative, these inflation reduction episodes were tabulated separately. Table 5.2 shows patterns in short-term interest rates over expansionary inflation reduction episodes. Average actual real interest rates are compared to the average interest rates associated with the long-run trend, calculated by applying a Hodrick-Prescott filter to the real interest rate series. The difference in average real interest rates (actual rates minus the rates associated with the long-run trend) is expressed as a percentage of the average long-run trend in real interest rates over the inflation reduction episode. Only inflation reduction episodes with positive actual average real interest rates over the episode in question are included in Table 5.2. In almost all of the inflation reduction episodes, actual real interest rates were, on average, kept below the long-run trend. This is consistent with the argument that countries that do not raise interest rates in the face of growing inflationary pressures are less likely to experience employment losses, or a slow-down in the growth rate of employment, during inflation reduction episodes. There were only two exceptions: Sri Lanka 1974–76 and Trinidad and Tobago 1989–92. One problem with looking at interest rates over the entire inflation reduction episode is that policy makers might increase interest rates in

The gendered political economy of inflation targeting

Table 5.2

103

Patterns in average real short-term interest rates over inflation reduction episodes in which total employment expanded (only episodes with positive average real interest rates included). Period

Chile Costa Rica Kenya Malaysia Philippines Singapore Sri Lanka

Trinidad and Tobago

Difference in actual and long-run average real interest rates as a percent of the long-run average

1984–88 1991–93 1993–96 1992–96 1990–94 1990–99 1974–76

−31.3 −9.8 −2.1 −9.7 −42.8 −6.9 124.0

1989–94 1989–92

−5.6 16.3

1993–96

−3.0

Notes

1973–75: avg. neg. interest 1988–91: 15.4 per cent

anticipation of inflation or in response to domestic economic conditions prior to the year in which inflation peaks. Therefore, we also examined average interest rates over an alternative periodization in order to see if the timing of the interest rate changes influenced our interpretation of the nature of the inflation reduction event. We looked at the period one year before the beginning of an inflation reduction episode and extending half-way into the episode. For example, if the inflation reduction episode spanned the years 1980 to 1984, then the alternative period over which interest rates were compared would be 1979 to 1982. This alternative periodization for the real interest rate analysis shows that real interest rates were negative on average in Sri Lanka from 1973 to 1975. Therefore, the increase in real interest rates documented in Table 5.2 might represent an effort by policy makers to move from negative real interest rates to positive (yet still relatively low) rates. Raising negative interest rates (that is, making them more positive) could have a different impact on employment than raising positive real interest rates – a point to which we will return shortly. However, the alternative periodization does not shed much light on the case of Trinidad and Tobago from 1989 to 1992. Table 5.3 shows a similar set of calculations for another sub-set of

104

Table 5.3

Beyond inflation targeting

Patterns in average real short-term interest rates over inflation reduction episodes in which total employment contracted (only episodes with positive average real interest rates included) Period

Barbados

Difference in actual and long-run average real interest rates as a percent of the long-run average

1980–86 1990–94 1996–99

136.7 14.4 −3.3

Colombia

1980–85

−1.1

India

1982–86 1991–94

114.2 −9.3

1997–02

−18.4

1992–2000

−8.4

Jamaica

Malaysia Mauritius

Singapore Sri Lanka Taiwan

Thailand

1981–86 1989–93

151.9 −79.2

1994–96 1981–86 1981–86 1997–99 1980–85 1991–02

116.4 117.9 1398 12.9 124.3 −6.5

1980–85 1990–93 1997–2000

16.6 15.2 114.2

Notes

1995–98: 11.0 percent 1979–83: 134.6 percent 1990–93: −19.3 percent 1996–2000: −12.8 percent Interest rates are t-bill rates. 1991–96: avg. neg. interest 1988–91: −76.3 percent

1990–97: −11.9 percent

inflation reduction episodes. This time, contractionary inflation reduction episodes are featured. As with Table 5.2, only inflation reduction episodes with positive actual average real interest rates over the episode in question are included. In most cases the opposite pattern observed previously in Table 5.2 is

The gendered political economy of inflation targeting

105

evident. During contractionary inflation reduction episodes actual interest rates were kept above the long run trend on average. In two cases, Barbados 1996–99 and Colombia 1980–85, actual real interest rates fell below the long-run trend during the entire inflation reduction episode. However, if we apply the alternative periodization we can see that real interest rates increased before inflation peaked and during the earlier stages of the inflation reduction period. For example, in Colombia from 1979 to 1983 average actual real interest rates were nearly 35 percent higher than the long-run average over this same period. There were five contractionary inflation reduction episodes in which real interest rates behaved differently: India 1991–94, India 1997–2002, Jamaica 1992–2000, Mauritius 1989–93 and Taiwan 1991–2002. In one case – Jamaica – the alternative periodization shows that interest rates were negative on average going into the inflation reduction episode. This alters the impact on employment of keeping rates below their long-run trend. However, in the other four cases interest rates and employment exhibited a different pattern when compared to the other episodes in Table 5.3. This discussion suggests that negative real interest rates are important in analysing the employment trends across inflation reduction episodes. Table 5.4 summarizes trends in real interest rates for those inflation reduction episodes in which average actual real interest rates were negative. In the majority of the cases in which average real interest rates were negative, interest rates were kept below the long-run average and employment grew slower than the trend rate of growth. This suggests that keeping interest rates negative and below the long-run trend will not help to increase employment. In two cases employment expanded despite low, negative real interest rates: Jamaica 1979–82 and Trinidad and Tobago 1974–77. However, these appear to be the exception rather than the rule. One reason why negative real interest rates tend to be associated with a contraction in employment relative to its long-run trend is that many of these inflation reduction episodes are ‘stagflationary’. That is, supplyside shocks produce a situation in which inflation accelerates, economic growth slows down, resulting in negative real interest rates. Real interest rates are only one set of variables that potentially link monetary policy to inflation reduction and employment dynamics. Two other possibilities are the real exchange rate and the growth rate of the real money supply. Table 5.5 examines trends in these two variables across the various inflation reduction episodes. Specifically, it shows differences between the actual average annual growth rate and the growth rate associated with the long-run trend for the real exchange rate and the real money supply across various inflation reduction episodes. The

106

Table 5.4

Beyond inflation targeting

Patterns in average real short-term interest rates over inflation reduction episodes in which average real interest rates were negative Period

Costa Rica India Jamaica

Kenya Mauritius Philippines

Singapore South Korea

Taiwan Trinidad and Tobago

1982–85 1973–77 1974–76 1979–82 1985–88 1975–80 1980–86 1973–76 1980–82 1984–87 1974–76 1980–85 1991–94 1997–2000 1974–76 1974–77 1980–87

Actual interest rates above or below long-run trend

Employment: expansionary or contractionary

below below below below above below below below below below below below below above below below below

contract contract contract expand expand contract contract contract contract contract contract contract contract contract contract expand contract

inflation reduction episodes are separated into contractionary and expansionary periods. In terms of the real exchange rate, for the purposes of this study, we measured the rate as the nominal US dollar exchange rate adjusted for changes in the US GDP deflator relative to the specific country’s GDP deflator. A decrease in the value of our real exchange rate measurement, therefore, represents an appreciation in the real exchange rate. Likewise, an increase in value represents a depreciation. There do not appear to be any systematic patterns with respect to changes in the real exchange rate across inflation reduction episodes and whether the episode was contractionary or expansionary. In 34 percent of inflation reduction episodes the average annual percent change in the real exchange rate was below that of the long-run exchange rate (that is, the exchange rate appreciated relative to its long-run trend); in 60 percent of the episodes the difference in average growth rate was positive (that is, the actual real exchange rate depreciated relative to the long-run trend); and in 6 percent

The gendered political economy of inflation targeting

Table 5.5

Country

107

Differences in the average actual annual growth rate and the average annual growth rate associated with the long-run trend for the real exchange rate (RER) and the real money supply across inflation reduction episodes Period

RER (percent)

Money supply (percent)

Total employment contracts on average 1980–86 –0.3 1990–94 11.1 1996–99 −1.2 Brazil 1993–99 11.1 Colombia 1980–85 12.2 Costa Rica 1982–85 −6.7 India 1973–77 11.6 1982–86 10.8 1991–94 13.6 1997–2002 10.4 Jamaica 1974–76 −9.0 1992–2000 −0.6 Kenya 1975–80 −3.5 1981–87 11.7 Malaysia 1981–86 14.5 Mauritius 1980–86 0.0* 1989–93 11.8 1994–96 −4.0 Philippines 1973–76 −3.9 1980–82 10.8 1984–87 0.0 Singapore 1974–76 13.9 1981–86 12.8 South Korea 1980–85 17.0 1991–94 −0.4 1997–2000 16.7 Sri Lanka 1981–86 −0.7 1997–99 11.7 Thailand 1974–76 −0.2 1980–85 13.7 1990–93 −1.9 1997–2000 17.0 Trinidad and Tobago 1980–87 12.7

−1.1 −1.5 13.0 10.3 10.2 −9.4 −1.6 10.6 −0.4 11.7 –0.7 10.5 11.9 11.8 −1.4 −2.2 −1.9 −1.4 −1.6 −0.4 −10.2 −0.6 −2.4 −1.9 −2.7 16.2 −1.1 0.0 −0.3 −2.5 12.8 −0.4 −1.0

Total employment expands on average 1989–92 0.0 1984–88 12.2

14.9 12.7

Barbados

Brazil Chile

108

Beyond inflation targeting

Table 5.5

(continued)

Country

Period

Costa Rica Jamaica Kenya Malaysia Mauritius Philippines Singapore Sri Lanka Trinidad and Tobago

Note:

*

1991–93 1979–82 1985–88 1993–96 1992–96 1974–77 1990–94 1990–99 1974–76 1989–94 1974–77 1989–92 1993–96

RER (percent) −4.8 −2.2 −7.8 15.7 −4.6 n.a. −1.8 10.4 −1.0 −0.3 −1.0 −3.8 11.4

Money supply (percent) −1.1 11.9 14.9 111.4 15.1 15.4 −0.5 11.4 −4.8 −0.7 −1.0 −2.7 15.4

Based on 1981–86 average, due to data limitations.

Source: Authors’ calculations based on data from the IMF publication International Financial Statistics. Real US interest rate is T-bill rate less the inflation rate.

of the episodes there was no difference between the growth rate of the actual and long-run real exchange rates. These ratios were approximately the same for contractionary and expansionary inflation reduction episodes.10 However, real exchange rates appear to have an impact on the gender bias observed in contractionary inflation reduction episodes. Recall that, in the majority of cases, women’s formal employment was disproportionately affected by the slow-down in employment growth. However, about a third of the time the ratio of women’s to men’s employment actually improved when compared to its long-run trajectory. In each of these cases the real exchange rate either depreciated or showed no deviation relative to its long-run trend. In other words, maintaining a competitive exchange rate may offset some of the gender bias observed during contractionary inflation reduction. Why would this be the case? As previously noted, in many countries the growth of women’s employment – particularly formal employment and wage employment – has tended to be concentrated in tradable sectors, either export-oriented or import-competing (Benería, 2003; Elson, 1996; Elson and Pearson, 1981; Kabeer and Mahmud, 2004). A real depreciation of the exchange rate favors tradable sectors and could help protect women’s employment in certain cases. Turning to the real money supply,11 there is some indication that money

The gendered political economy of inflation targeting

109

supplies grew more slowly during contractionary episodes. In 67 percent of all contractionary episodes for which data is available actual average annual growth rates in the real money supply fell below the long-run trend. In 60 percent of all expansionary episodes actual annual growth rates in the real money supply were greater than the average for the longrun trend. Relative to the gendered effects of inflation reduction then, tightening the real money supply may have a tendency to be associated with greater sacrifices in women’s employment. This analysis suggests a number of preliminary findings. However, it is important to recognize that a myriad of factors affect the variables examined here – in particular, employment, inflation, interest rates, exchange rates and the money supply – and therefore any generalizations must be tentative. Although some commonalities do arise, the diversity of country experiences prevents us from drawing definitive conclusions without additional in-depth analysis and country-specific case studies. Nevertheless, we can make some general observations from the analysis presented. ●

●

●

●

Inflation reduction episodes occur simultaneously across a large number of countries. This suggests that there are common, external factors that influence inflation dynamics in low- and middle-income countries. If employment contracts during an inflation reduction episode, it is likely that women will experience a larger loss of employment, in percentage terms, than men. However, during inflation reduction episodes in which employment expands, the gender-specific impact is ambiguous. Countries that respond to inflationary pressures by raising real interest rates above the long-run trend are more likely to experience a slow-down in the growth of employment relative to those countries that keep interest rates in line with or below the long-run trend, with concomitantly higher losses for relative female employment. However, countries with negative real interest rates do not appear to be able to increase employment growth by lowering real interest rates still further. We did not find a link between changes in the real exchange rate and the impact of inflation reduction on employment in general. However, we did find that real exchange rates seem to impact the gender bias of contractionary inflation reduction episodes. In all cases where women experienced relative employment gains during employment contractions, exchange rates either depreciated or showed no deviation relative to long-run trends.

110 ●

Beyond inflation targeting

Tightening the real money supply also seems to be negatively associated with employment in general and women’s employment in particular.

These results suggest that contractionary monetary policy aimed at reducing inflation often has a disproportionately negative impact on women’s employment, an effect that may be eased by maintaining a competitive exchange rate. Conversely, non-contractionary inflation reduction is not necessarily favorable to women’s formal employment in all circumstances.

5.4

DISCUSSION AND CONCLUSIONS

The empirical analysis presented here concerns the short-run, genderspecific impacts of policy responses during inflation reduction episodes. The results say little about the long-run impact of different policy responses. Supporters of inflation targeting frequently acknowledge that short-run trade-offs might exist, but the long-run benefits of low inflation for growth and development are more significant. This argument is problematic when transitory policy shocks have long-run consequences for real economic variables (Fontana and Palacio-Vera, 2004). Similarly, short-term gender-specific shocks can have long-run effects for a country’s human and economic development. A number of empirical studies suggest that gender-based inequities in employment and unemployment have implications for long-term development. For example, this body of research shows that a positive relationship exists between gender equality (measured most commonly as educational equity) and economic growth in developing countries (Dollar and Gatti, 1999; Hill and King, 1995; Klasen, 1999). Some of the effects are quite large: Klasen (1999), in a panel data study between 1960 and 1992, finds that had South Asia and sub-Saharan Africa had more gender equity in education, growth would have been 0.9 percent per year faster. Investing in girls makes for a higher productivity workforce, but higher rates of unemployment and cyclical volatility in women’s jobs will discourage these types of investments at both the individual and community levels. In a related sense, lower incomes and higher income volatility for women could lead to lower investments in human capital overall, thereby lowering long-term growth. Theory and evidence have aptly demonstrated a higher co-incidence between a mother’s income and the family’s basic needs than a father’s income (Benería and Roldan, 1987; Blumberg, 1991; Chant, 1991), a finding underlying what has been termed the ‘good mother

The gendered political economy of inflation targeting

111

hypothesis’. Income that is controlled by women is more likely to be spent on children’s health and nutrition (Dwyer and Bruce, 1988; Hoddinott et al., 1998). In many countries a large proportion of fathers provide little or no economic support for their children (Folbre, 1994). But faced with cyclically higher rates of unemployment during disinflation, ‘good mothers’ will have fewer opportunities to invest in their children. In the nearer term systematic discrimination in hiring and firing results in a misallocation of workers as the most able are not matched with the most appropriate jobs. Such misallocations can result in efficiency losses with effects on output and growth. Tzannatos (1999) evaluates the effects of eliminating gender-based occupational segregation in a number of developing countries, and finds on average a one-time gain in output of 6 percent. His results are significant but incomplete. He only considers occupational segregation, not industrial segregation, which can be a much more extensive part of gender segregation in some countries, as in China. In addition, Tzannatos only considers the static picture, and not the dynamic efficiency losses from being on a particular growth path characterized by a high degree of gender segregation (Elson, 1999). Less instrumental and ultimately more central than the economic implications of gender-biased inflation reduction discussed above is the issue of equity, as we find that women as a group shoulder a disproportionate share of the costs of contractionary inflation reduction. From a social justice perspective, then, it is important to better understand and redress these gender-biased outcomes. One way of doing so would be to focus research and policy on better understanding the link between inflation targeting and overall social welfare. To get at these connections, central banks should incorporate gender-specific indicators in the creation of targets, such as gender disaggregated employment figures or gender aware inflation rates (that account for gendered consumption and employment patterns). These market-based targets should also be supplemented by longer term gender aware human development targets to ultimately get at the links among inflation targeting, equity and well-being. Treating gender equity as secondary to other, more ‘general’ economic concerns may also be instrumental in the political sustainability of inflation targeting in monetary policy. As our analysis implies, incorporating concerns over gender equity into monetary policy formulation would involve a move away from inflation targeting as it is currently practiced and could harm the interests of those invested in a low inflation, high interest rate environment. Moreover, if women’s labor force participation keeps unit labor costs and inflation lower than it would otherwise be, then a focus on gender equity within the context of sustainable levels of inflation could require other mechanisms for price control that are more

112

Beyond inflation targeting

consistent with long-run development. Such a move might be resisted by those that benefit (perhaps only in the short run) from women’s more precarious employment – for example, their employers and employed men. From this perspective, gender-biased central bank policy may help solve the political problems introduced by inflation targeting in that gender bias concentrates the costs of these policies on a less powerful segment of society – women. As such, inflation targeting policies should be considered in terms of their social content (for example, what are the social structures that underlie this policy) as well as its social impact (Elson and Cagatay, 2000). Such an approach underscores the gendered nature of the contested terrain in macroeconomic policy making.

NOTES 1.

2. 3. 4. 5. 6. 7. 8. 9. 10.

11.

This chapter is part of the PERI/Bilkent project on ‘Alternatives to Inflation Targeting’. We thank the Ford Foundation, the UN-DESA, the Rockefeller Brothers Fund and the Political Economy Research Institute at the University of Massachusetts for financial support. We would also like to thank Gerald Epstein, Erinc Yeldan and Diane Elson for their help and comments, as well as the participants (and our hosts) of the AIT meeting at CEDES in Buenos Aires in June 2005. All mistakes are our own, of course. Similar reasoning can be applied to the ‘pink collar’ aspects of the international production of services, a sector where women predominate in the administrative support aspect of the trade. Our choice of countries was limited to those for which reasonably reliable, genderdisaggregated formal employment data were available. In addition, the GDP deflator for the USA used to construct the real exchange rate measure was taken from the online database of the US Bureau of Economic Analysis. Other real US interest rates did increase around the time of the last inflation reduction episode, for example, the yield on three-year government bonds. The sample of countries includes Singapore which could arguably be classified as a high-income country today. However, for much of the period considered in this chapter, 1970–2003, Singapore can be considered a middle-income country. India is notable in terms of its frequent deviation from the behavior exhibited by other countries, both with respect to the gender dynamics of slower employment growth and the observed trends in real interest rates during inflation reduction episodes. Due to its extreme volatility, Brazil was excluded from this analysis. Only in the case of Jamaica were the treasury-bills used to determine actual interest rates and to estimate long-run trends. In contractionary episodes 36 percent of the episodes showed an appreciation, 58 percent a depreciation and 6 percent no difference in the average growth of the real exchange rate relative to its long-run trend. In expansionary episodes the percentages were 34 percent, 60 percent and 7 percent, respectively. The definition of the money supply used is M2, money plus quasi-money, which includes cash; checking accounts; and relatively liquid deposits like savings and money market deposit accounts.

The gendered political economy of inflation targeting

113

REFERENCES Algan, Y. and P. Cahuc (2004), ‘Job protection: the macho hypothesis’, Society for Economic Dynamics 2004 meeting papers no. 332, September. Anker, R. and C. Hein (1985), ‘Why Third World urban employers usually prefer men’, International Labour Review, 124 (1), 73–90. Arriagada, I. (1994), ‘Changes in the urban female labour market’, CEPAL Review, 53, 92–110. Aslanbeigui, N. and G. Summerfield (2000), ‘The Asian crisis, gender, and the international financial architecture’, Feminist Economics, 6 (3), 81–103. Azmat, Ghazala Maia Gvell and Alan Mangging (2004), ‘Gender gaps in unemployment rates in OECD countries’, London School of Economics Centre for Economic Policy discussion paper no. 604. Bakker, I. (ed.) (1994), The Strategic Silence: Gender and Economic Policy, London. Zed Books with the North-South Institute. Ball, L. (1993), ‘What determines the sacrifice ratio?’, National Bureau for Economic Research working paper no. 4306. Benería, L. (2001), ‘Shifting the risk: new employment patterns, informalization, and women’s work’, International Journal of Politics, Culture and Society, 15 (1), 27–53. Benería, L. (2003), Gender, Development, and Globalization: Economics as if All People Mattered, London: Routledge. Benería, L. and S. Feldman (eds) (1992), Unequal Burden: Economic Crises, Persistent Poverty, and Women’s Work, Boulder, CO: Westview Press. Benería, L. and M. Roldán (1987), The Crossroads of Class and Gender: Industrial Homework, Subcontracting, and Household Dynamics in Mexico City, Chicago, IL and London: University of Chicago Press. Blumberg, R. (1991), ‘Income under female versus male control’, in R. Blumberg (ed.), Gender, Family and Economy: The Triple Overlap, Newbury Park, CA: Sage Press, pp. 97–127. Braunstein, E. and N. Folbre (2001), ‘To honor and obey: efficiency, inequality and patriarchal property rights’, Feminist Economics, 7 (1), 25–44. Cagatay, N. and S. Ozler (1995), ‘Feminization of the labor force: the effects of long-term development and structural adjustment’, World Development, 23 (11), 1883–94. Cagatay, N., D. Elson and C. Grown (eds) (1995), ‘Gender, adjustment and macroeconomics’, World Development, 23 (11) (special issue), 145–390. Carr, M., M. Chen and J. Tate (2000), ‘Globalization and home-based workers’, Feminist Economics, 6 (3), 123–42. Casale, D. (2003), ‘The rise in female labour force participation in South Africa: an analysis of household survey data, 1995–2001’, PhD dissertation, Division of Economics, University of KwaZulu-Natal. Cerrutti, M. (2000), ‘Economic reform, structural adjustment and female labor force participation in Buenos Aires, Argentina’, World Development, 28 (5), 879–91. Chant, S. (1991), Women and Survival in Mexican Cities: Perspectives on Gender, Labour Markets and Low-Income Households, New York: Manchester University Press. Dollar, D. and R. Gatti (1999), ‘Gender inequality, income and growth: are good

114

Beyond inflation targeting

times good for women?’, World Bank policy research report on gender and development working paper series no. 1, Washington, DC. Dwyer, D. and J. Bruce (1988), A Home Divided: Women and Income in the Third World, Palo Alto, CA: Stanford University Press. Elson, D. (ed.) (1991), Male Bias in the Development Process, Manchester: Manchester University Press. Elson, D. (1995), ‘Gender awareness in modeling structural adjustment’, World Development, 23 (11), 1851–68. Elson, D. (1996), ‘Appraising recent developments in the world market for nimble fingers’, in A. Chhachhi and R. Pittin (eds), Confronting State, Capital and Patriarchy: Women’ Organizing in the Process of Industrialization, New York: St Martin’s Press, pp. 35–55. Elson, D. (1999), ‘Labor markets as gendered institutions: equality, efficiency and empowerment issues’, World Development, 27 (3), 611–27. Elson, D. and N. Cagatay (2000), ‘The social content of macroeconomic policies’, World Development, 28 (7), 1347–64. Elson, D. and R. Pearson (1981), ‘Nimble fingers make cheap workers: an analysis of women’s employment in Third World export manufacturing’, Feminist Review, 7, 87–107. Epstein, G. (2000), ‘Myth, mendacity and mischief in the theory and practice of central banking, mimeo, University of Massachusetts Amherst. Epstein, G. (2003), ‘Alternative to inflation targeting monetary policy for stable and egalitarian growth: a brief research summary’, Political Economy Research Institute working paper no. 62. Epstein, G. and J. Schor (1990), ‘Macropolicy in the rise and fall of the golden age’, in S. Marglin and J. Schor (eds), The Golden Age of Capitalism: Reinterpreting the Postwar Experience, Oxford: Clarendon Press, pp. 126–52. Fernández-Kelly, M. (1983), For We are Sold, I and My People: Women and Industrialization in Mexico’s Frontier, Albany, NY: SUNY Press. Folbre, N. (1994), Who Pays for the Kids? Gender and the Structures of Constraint, London and New York: Routledge. Folbre, N. (1997), ‘Gender coalitions: extrafamily influences on intrafamily inequality’, in L. Haddad, J. Hoddinott and H. Alderman (eds), Intrahousehold Resource Allocation in Developing Countries. Models, Methods, and Policy Baltimore, MD: Johns Hopkins University Press, pp. 263–74. Fontana, G. and A. Palacio-Vera (2007), ‘Are long-run price stability and shortrun output stabilization all that monetary policy can aim for?’, Metroeconomica, 58 (2), 269–98. Fussell, E. (2000), ‘Making labor flexible: the recomposition of Tijuana’s maquiladora female labor force’, Feminist Economics, 6 (3), 59–79. Ghosh, J. (2001), ‘Globalisation, export-oriented employment for women and social policy: a case study of India’, paper prepared for United Nations Research Institute for Social Development project on Globalization, ExportOriented Employment for Women and Social Policy, Geneva. Hill, M. and E. King (1995), ‘Women’s education and economic well-being’, Feminist Economics, 1 (2), 21–46. Hoddinott, J., H. Alderman and L. Haddad (eds) (1998), Intrahousehold Resource Allocation in Developing Countries: Methods, Models and Policy, Baltimore, MD: Johns Hopkins University Press. Joekes, S. (1999), ‘A gender-analytical perspective on trade and sustainable

The gendered political economy of inflation targeting

115

development’, in Trade, Sustainable Development and Gender, New York and Geneva: United Nations Conference on Trade and Development, pp. 33–59. Kabeer, N. and S. Mahmud (2004), ‘Globalization, gender, and poverty: Bangladeshi women workers in export and local markets’, Journal of International Development, 16, 93–109. Klasen, S. (1999), ‘Does gender inequality reduce growth and development? Evidence from cross-country regressions’, World Bank policy research report on gender and development working paper series no. 7, Washington, DC. Lim, Joseph (2000), ‘The effects of the East Asian crisis on the employment of women and men: the Philippine case’, World Development, 28 (7), 1285–306. Patnaik, U. (2003), ‘Global capitalism, deflation and agrarian crisis in developing countries’, United Nations Research Institute for Social Development, social policy and development programme paper no. 15. Seguino, S. (2003), ‘Why are women in the Caribbean so much more likely than men to be unemployed?’, Social and Economic Studies, 52 (4), 83–120. Sen, A. (1990), ‘Gender and cooperative conflicts’, in I. Tinker (ed.), Persistent Inequalities: Women and World Development New York and Oxford: Oxford University Press, pp 123–49. Standing, G. (1999), ‘Global feminization through flexible labor: a theme revisited’, World Development, 27 (3), 583–602. Tzannatos, Z. (1999), ‘Women and labor market changes in the global economy: growth helps, inequalities hurt and public policy matters’, World Development, 27 (3), 551–69. United Nations (1999), World Survey on the Role of Women in Development. Globalization, Gender and Work, New York: United Nations.

6.

Inflation and economic growth: a cross-country non-linear analysis Robert Pollin and Andong Zhu1

6.1

INTRODUCTION

This chapter presents new cross-country evidence between 1961 and 2000 on the relationship between inflation and economic growth. Despite the central importance of this inflation-growth relationship for macroeconomic theory and policy, there is nothing close to a professional consensus as to what the empirical evidence tells us about this relationship. The results we present here have direct relevance to the debate on inflation targeting as an appropriate framework for conducting monetary policy. Over the past decade governments throughout the world have embraced inflation targeting as a dominant policy framework. For the most part, this specifically means that they have set a low band of acceptable inflation rates as a primary target in the conduct of economic policy. This band is usually between a 3–5 percent annual inflation rate. They have then maintained sufficiently high short-term interest rates as the intermediate policy instrument for preventing inflation from exceeding that target band. Higher interest rates are aimed, in turn, at reducing economic growth. Slower economic growth should then dampen inflationary pressures. At least in the short run, the costs in terms of slower growth of containing inflation within this 3–5 percent band are evident. But proponents of inflation targeting hold that, over a longer term framework, maintaining low inflation will itself yield benefits for growth that exceed these short-term costs.2 Some limitations of inflation targeting have been widely recognized by mainstream economists and even US central bankers Ben Bernanke and Alan Blinder (see Bernanke et al., 1999; Blinder, 1998). The Bernanke/ Blinder view is that inflation targeting does not provide an approach to maintaining low inflation that is clearly superior to other approaches. This is true as such, but this concern about inflation targeting as an operating procedure alone begs a more important question. This crucial question is whether maintaining inflation within a band of 3–5 percent itself is, as a 116

Inflation and economic growth: a cross-country analysis

117

generalization, supportive of economic growth, regardless of the technique being used to maintain inflation within that low band. It is this broader question that we address in this chapter. That is, are countries making sacrifices in terms of their economic growth path by focusing macroeconomic policy on maintaining inflation at no more than 3–5 percent? In Section 6.2 we briefly review the overarching and longstanding analytic debates on the relationship between inflation and economic growth, then focus specifically on the recent econometric research that has explored that relationship. In Section 6.3 we present basic descriptive data from our data sample, then examine the main results from our various economic exercises. In the concluding Section 6.4 we consider the broader implications of our findings, especially as they relate to policy debates around inflation targeting and possible alternative approaches to inflation control.

6.2 6.2.1

LITERATURE OVERVIEW Analytic Perspectives

We begin by separating out the phenomenon of hyperinflation, which we broadly define as being annual inflation rates in excess of 40 percent per year. Hyperinflations occur through a variety of specific factors. But regardless of their specific origins, hyperinflations represent a breakdown of economic functionings. We will assume that hyperinflations correspond with, and are detrimental to, a positive economic growth path. We are therefore leaving aside here the possibility that there may be some positive correspondence between inflation above 40 percent and economic growth. Hyperinflations aside, the relationship between inflation and growth has been at the very center of macroeconomic theory debates since the monetarist counterrevolution against Keynesianism beginning in the 1960s.3 The main progeny of that counterrevolution – the ‘natural rate of unemployment’, the vertical Phillips Curve, and New Classical Economics more generally – have been focused largely around demonstrating that there can be no positive benefits for economic growth or employment of operating an economy at anything above a minimal inflation rate in the range of 2–3 percent. From this perspective, inflation impedes efficient resource allocation by obscuring the signaling role of relative price changes, which, in turn, is the most important guide to efficient economic decision making. This position contrasts sharply with the Keynesian perspective and the early Phillips Curve models, which held that inflation and economic

118

Beyond inflation targeting

growth can be positively associated when inflationary pressures emerge as a byproduct of rising aggregate demand. In this Keynesian framework it is not the case that inflation is itself a positive engine of growth, certainly not a primary growth-inducing force. The point is rather that if rising aggregate demand is leading to increased growth, then some inflationary pressures are likely to emerge in this scenario as a relatively benign byproduct. Within this Keynesian framework, there could also be reasons for inflation and growth to be negatively correlated. This would occur when inflation results from monopolistic pricing practices, exchange rate volatility or supply shocks. These problems can also be compounded when adequate policy interventions do not occur to dampen the inflationary impulses induced by monopolistic pricing, exchange rate volatility or supply shocks. 6.2.2

Recent Empirical Studies

Probably the most influential recent contribution to the econometric literature on inflation and growth is that of Bruno and Easterly (with results presented in both Bruno (1995) and Bruno and Easterly (1998)). Bruno and Easterly examined the relationship between inflation and economic growth for 127 countries between 1960 and 1992. Their examination of this data set is historical and descriptive. They do not present a formal econometric model. Their key conclusion was that there is no robust evidence from this data sample demonstrating a trade-off between output growth and inflation. More specifically, only on the basis of two conditions could one observe a negative growth-inflation relationship at all in their data sample. These were: (1) the inclusion in the data sample of very high inflation experiences, that is, rates of inflation of 40 percent and higher and (2) increasing the frequency of the data observations. As they write, ‘The results get stronger as one goes from the cross-section to ten year averages to five year averages to annual data’ (1998, p. 4). Once one controls for these two factors, Bruno and Easterly found that average growth rates fell only slightly as inflation rates moved up to 20–25 percent. For inflation rates below 20 percent, Bruno concluded that ‘there is no obvious empirical evidence for significant long-run growth costs’ (Bruno, 1995, p. 38). Moreover, of particular importance for our concerns with aggregate demand effects on inflation and growth, Bruno found that during 1960–72, economic growth on average increased as inflation rose, from negative or low rates to the 15–20 percent range. This is because, as Bruno explained, ‘in the 1950s and 1960s, low-to-moderate inflation went hand in hand with very rapid growth because of investment demand

Inflation and economic growth: a cross-country analysis

119

pressures in an expanding economy’ (ibid., p. 35). Thus, inflation that results directly from economic expansion does not, according to Bruno’s findings, create any significant barriers to expansion. Despite these findings, Bruno still makes clear in his single-authored paper that he does not advocate complacency with respect to inflation rates in the 20 percent region. According to Bruno, once inflation moves into the 20 percent region, it is difficult to contain at this level. This is because, within the 20 percent inflation region, the systems of indexing wages and financial assets, as well as exchange rate adjustments, become more frequent. This then creates a momentum toward accelerating inflation. Neither Bruno alone nor Bruno and Easterly provide systematic evidence on behalf of Bruno’s concerns about inflation within the 20 percent region. Nevertheless, Bruno is clear in his conclusion that ‘getting inflation down to single digits is important even for longer-term growth reasons’ (ibid., p. 38). But even within this less systematic discussion on the dangers of inflation in the 20 percent range, it is still notable that Bruno never suggests that inflation needs to be pushed below a single-digit threshold – and specifically down into the 3–5 percent range advocated by proponents of inflation targeting. Since the Bruno and Easterly study, various researchers have examined the output growth-inflation relationship through more formal techniques than those employed by Bruno and Easterly while still searching out, as with Bruno and Easterly, potential non-linearities. For example, in a 1998 paper, International Monetary Fund (IMF) economists Atish Ghosh and Steven Phillips combine panel regression techniques with non-linear treatment of the inflation-growth relationship. They also utilize a decision-tree technique that, in their view, is more robust to outliers and non-linearities than is standard regression analysis. Their model draws from a data sample of IMF member countries over 1960–96. According to this model, they find evidence of a negative inflation threshold at 2.5 percent. But they also acknowledge that thresholds of 5 or 10 percent generate statistical results very similar to the 2.5 percent threshold. A 2001 publication by two separate IMF economists, Moshin Khan and Abdelhak Senhadji, offers two innovations relative to Ghosh and Phillips (Khan and Senhadji, 2001). The first is their use of conditional least squares, a new non-linear estimation technique. The second, and more straightforward, innovation was to divide their data sample into industrial and developing countries. Based on this approach, they find that the threshold level above which inflation significantly slows growth is 1–3 percent for industrial countries and 11–12 percent for developing countries. More recently still, a 2004 paper by Burdekin et al. followed Khan and

120

Beyond inflation targeting

Senhadji in allowing for different threshold effects among the industrial and developing countries (Burdekin et al., 2004). They also allow for non-linearities in the growth-inflation relationship through utilizing spline estimation techniques. The results from this research diverge sharply from Khan and Senhadji. In terms of point estimates, they found that the turning point for industrial countries was 8 percent while that for developing countries was 3 percent. In short, all of these studies are in broad concurrence with Bruno and Easterly as to the presence of non-linearities in the growth-inflation relationship. They also broadly concur with Bruno’s conclusion that the negative effects of inflation will occur somewhere below a 20 percent threshold, most likely in the single-digit range. However, they diverge sharply as to where the turning point occurs within a range of roughly 12 percent inflation or less. Moreover, the two studies that adopted the simple innovation of dividing the sample between industrial and developing countries reached opposite conclusions as to which set of countries had a higher inflation threshold. Thus, despite the deployment of sophisticated techniques for capturing the impact of non-linearities in the growth-inflation relationship, major questions remain unresolved. In particular, there remains no robust evidence in support of a policy goal of maintaining an inflation target in the range of 3–5 percent.

6.3

DESCRIPTIVE DATA AND ECONOMETRIC EVIDENCE

Our own model is a straightforward panel data model, in which we aim to isolate the effects of inflation on economic growth through including a series of control variables as well as allowing for a non-linear component to the growth-inflation relationship. Our data sample runs from 1961 to 2000, including data from a total of 80 countries. We have excluded from the model countries whose population is less than 2 million people. We do this to focus our empirical exercises on countries whose economies are minimally large enough so that the countries’ patterns of economic activity can be understood as having features that are distinct to that country. Appendix 6.1A provides a full list of the countries in our data sample. 6.3.1

Descriptive Statistics

We first provide some basic descriptive statistics from our data sample in both Table 6.1 and Figure 6.1. Table 6.1 shows both means and standard deviations for inflation and growth, for the full sample, and broken out

Inflation and economic growth: a cross-country analysis

Table 6.1

GDP growth Mean Standard deviation Inflation Mean Standard deviation

121

Descriptive statistics on GDP growth and inflation 80 country sample, 1961–2000 All countries (80 countries) (%)

OECD countries (21 countries) (%)

Middle-income countries (32 countries) (%)

Low-income countries (25 countries) (%)

1.9

2.6

1.8

0.9

2.7

1.8

2.7

3.4

10.2

7.0

12.8

11.3

7.2

4.9

8.3

6.6

Note: Israel and Singapore are not included in the country groupings because they are nonOECD high-income countries. Source:

See Appendix 6.1A.

according to our three income-level groupings. For all countries in the sample, as we see, the average rate of GDP growth is 1.9 percent and the average inflation rate is 10.2 percent. However, from the standard deviations – 2.7 percent for GDP growth and 7.2 percent for inflation – we also see that there are wide disparities among the observations in the sample. The disparities do diminish as we break out the full sample of countries into income-level groupings. Not surprisingly, the OECD countries experience the highest average rate of economic growth (virtually by definition; see Note 4) and the lowest average inflation rates. Average growth is significantly faster in the middle-income countries relative to the low-income countries, but average inflation is somewhat lower in the low-income countries. The four scatter plots in Figure 6.1 show the range of values for our data sample more fully. No strong patterns at all emerge from these figures in terms of the inflation/GDP growth relationship. Of course, these data plots do not control for factors other than inflation that could be affecting economic growth. We label in the four diagrams in Figure 6.1 the data points that emerge as outliers through simple observation. This provides some useful perspective. For example, with the full set of countries, the most rapid growth spurt was experienced by Haiti from 1996 to 2000. Haiti grew on average by 15.2 percent in this period, even while inflation was rising at an average

122

Beyond inflation targeting All countries

16

OECD countries 12

Haiti 1996–2000

10

8 4 Ecuador 1991–95 Venezuela 1986–90 Zimbabwe 1996–2000 Costa Rica 1981–85

0 –4 Rwanda 1991–95

–8 –10

0

10

GDP growth rate (%)

GDP growth rate (%)

12

30

8 6

Puerto Rico 1976–81

4 2

Puerto Rico 1981–85

40

0

50

4

8

Ecuador 1971–75

Venezuela 1986–90

Costa Rica 1981–85

GDP growth rate (%)

Ecuador 1991–95

0

Haiti 1996–2000

8 4 Ghana 1986–90

0 Zimbabwe 1996–2000

Jordan 1986–90 Rwanda 1991–95

–8 0

10

20

30

40

50

0

10

Inflation rate (%)

Note: Source:

24

–4

–8 –10

20

12

4

Venezuela 1971–75

16

Low-income countries 16

–4

12

Inflation rate (%)

Middle-income countries

8

Greece 1986–90

Finland 1991–95

–2

12 China 1991–95

New Zealand 1976–80

Switzerland 1991–95

Inflation rate (%)

GDP growth rate (%)

S. Korea 1966–70

S. Korea 1986–90

0

Nicaragua 1976–81

20

Japan 1966–70 Ireland 1996–2000

Nicaragua 1976–80

20

30

40

Inflation rate (%)

Annual data are grouped into five-year averages. See Appendix 6.1A.

Figure 6.1

Inflation and economic growth, 1961–2000

of 14.9 percent. In terms of other outliers in the all-country diagram, we see that the very high inflation and/or very low growth outliers are all lowor middle-income countries, with Rwanda, Nicaragua and Zimbabwe all experiencing severe political conflicts during their low growth/high inflation years. With the OECD and middle-income country diagrams, we see that the countries able to experience the most rapid economic growth rates were Japan, Ireland, South Korea and China. In all cases the rapid GDP growth was tied to reaching new levels of export success. Inflation in these

Inflation and economic growth: a cross-country analysis

123

countries over the relevant years ranged widely, between 2.6 and 13.2 percent. Clearly, it is difficult to offer generalizations from these figures as to the interrelationship between inflation and economic growth. It is evident that we need to examine this relationship more systematically, the task to which we now turn. 6.3.2

Econometric Model

Our approach has been to build a formal model that is still consistent with the main strength of the Bruno-Easterly framework, which is its simplicity. To do this, we work with a panel model that incorporates non-linearities through two relatively simple procedures. The first feature of our non-linear model is to simply exclude from our data set all observations in which inflation exceeded 40 percent. As we mentioned above, we accept the finding of Bruno and Easterly that inflation in that high range will produce negative effects on growth. We are therefore effectively asking with our model whether an annual inflation rate below 40 percent exerts a negative effect on economic growth, and if so, at what point are such negative effects likely to emerge? The second way that we introduce non-linearity in our model is to include the squared term on inflation as an explanatory variable, which means we are estimating the regression equations as a second-degree polynomial. This is a straightforward, widely used technique for estimating non-linear relationships, through allowing for changes in slopes as a function of changes in the independent variable. In this case the slope of the estimating equation can vary with changes in the inflation rate. This enables us to observe turning points in the relationship between inflationgrowth and inflation-equality. We can observe such possible turning points through this calculation: Turning point 5 2 ((inflation coefficient) / (2 * (inflation 2 squared coefficient)) . Within this framework, we then also pursue robustness tests through three sets of straightforward procedures: 1. We utilize four different panel data techniques: pooled ordinary least squares (OLS), between effects fixed effects, and random effects. In principle, researchers are supposed to establish through diagnostic exercises which of the four techniques is appropriate with a given data sample. In practice, however, it is frequently difficult to know

124

2.

3.

Beyond inflation targeting

which technique is the most reliable. Each of the techniques has both strengths and weaknesses. A pooled OLS model implicitly assumes there are no problems of omitted variables in a model, which is not likely to be true, even through frequently the problems may not be serious enough to substantially distort one’s results. A between effects model averages the data for each country into one observation. It is therefore testing more narrowly for variation on a countryby-country basis, as opposed to considering variation between time periods as well as countries. With the fixed effects model, we are allowing for intercept shifts to occur for each country, based on the range of possible omitted variables in evaluating country-by-country determinants of economic growth. But the fixed effects model effectively creates dummy variables for each country in the sample, which reduces degrees of freedom. Finally, the random effects model also allows for a different intercept for each country in the sample. But the random effects model isolates these individual country effects in the error term, and therefore does not reduce degrees of freedom in the manner of the fixed effects estimator. But, at the same time, to be an unbiased estimator, the random effects model requires that the omitted variable effects will be uncorrelated with the explanatory variables. Given this range of concerns with the various techniques, we report results utilizing all four techniques. By examining results generated by all four techniques, we are able to assess the robustness of our findings across the range of panel data estimators. We run regressions both on the full set of countries as one sample, then through dividing the countries into three groupings, OECD countries, middle-income countries and low-income countries. We are therefore able to observe the extent to which differences in the results are due to broad differences in the various countries’ level of development, as distinct from the individual country differences that we control for through the fixed and random effects models.4 We decompose the full time period into four decade-long sub-periods. This enables us to examine how the relationships may have changed over time. We are especially interested in following up on Bruno’s observation that inflation and growth were positively correlated from the 1960s up until the 1973 oil shock. This was a period in which, as Bruno said, inflation emerged out of explicit efforts to stimulate aggregate demand. Beyond these distinct features of our model, we also incorporate a set of control variables in each specification of the model. These control variables are standard in cross-country estimates of the

Inflation and economic growth: a cross-country analysis

125

determinants of economic growth. They include (1) the initial level of GDP; (2) the share of investment spending in GDP; (3) the share of government spending in GDP; (4) the fiscal deficit; (5) educational levels; (6) the level of overall health, as measured by life expectancy; (7) the change in terms of trade; (8) the effects of natural disasters; and (9) the effects of wars. In addition, we include dummy variables for each year in the pooled OLS, fixed effects and random effects models to control for the time effects within each set of country observations. Full descriptions of each of the control variables is reported in the Appendix 6.1B. We do not report here the full set of results on the control variables, but these results are available on request.5 We report the key findings of our econometric models in Tables 6.2 and 6.3. Both tables report the coefficients and t-statistics for the inflation and inflation-squared variables only for each of the regressions. We also report the turning points estimated by each equation when inflation switches from becoming a positive to negative, or negative to positive, influence on growth. Results for the full time period The results for the full time period are presented in Table 6.2. Considering first the data for all countries in the sample, we see that the sign of the inflation coefficient is consistently positive across specifications, and is statistically significant in all but the fixed effects specification. Moveover, the coefficient values across specifications are similar, ranging between 0.09–0.15. The coefficients on the inflation-squared terms are also similar and statistically significant in all cases. With the coefficients on the inflation and inflation-squared terms, we can then calculate the turning points as being between a 15.2 and 18.6 inflation rate. Overall, this first set of tests with the full data sample suggest that the rate of economic growth rises by between about 0.1 and 0.15 percent for every percentage point increase in the inflation rate up to a 15–18 percent threshold. Inflation then becomes a damper on growth beyond this threshold. The clear findings we obtain with the full data set is, however, not maintained when we consider OECD, middle-income and low-income countries separately. With the data grouped by income levels, we expect that the significance levels will go down due to the smaller sample sizes. And we do indeed observe generally lower significance levels with the results grouped by income levels. More specifically, in the case of the OECD countries, none of the coefficients for inflation or inflation-squared are statistically significant in any of the specifications. Moreover, the signs on the inflation variable shift to negative in the pooled OLS, between effects and random effects models. In

126

356 0.11* (2.55) −0.003** (−3.42) 18.3

Fixed effects 127 0.028 (0.31) −0.001 (−0.84) 14.0

Pooled OLS

127 0.06 (1.12) −0.002* (−2.45) 15.0

127 0.057 (0.84) −0.002 (−1.67) 14.3

Random effects

Middle-income countries

356 0.091 (1.61) −0.003* (−2.41) 15.2

356 0.11* (2.49) −0.003** (−3.73) 18.3

Random effects

32 0.129 (1.19) −0.004 (−1.71) 16.1

Between effects

80 0.149* (1.99) −0.004* (−2.35) 18.6

Between effects

86 0.359 (1.38) −0.01 (1.55) 18.0

Pooled OLS

135 −0.055 (−0.66) −0.005 (−1.43) −5.5

Pooled OLS 135 −0.034 (−0.36) −0.005 (−1.37) −3.4

Random effects

86 0.559* (2.38) −0.012* (2.20) 23.3

Fixed effects

86 0.386* (2.29) −0.01* (2.36) 19.3

Random effects

Low-income countries

135 0.025 (0.23) −0.007 (−1.79) 1.8

Fixed effects

OECD countries

25 0.238 (0.73) −0.008 (0.72) 14.9

Between effects

21 −0.130 (−0.30) 0.001 (0.06) 65.0

Between effects

Notes: 1. Data grouped as five-year averages; dependent variable is GDP growth; t-statistics in parentheses; p , 0.05 5 *, p , 0.01 5 **. 2. All countries with less than 2 million people were excluded from the sample. All observations with inflation above 40 percent were excluded. Five-year period average data were used instead of yearly data in the regression. ‘All countries’ refers to all the countries in this table.

Turning Point

(Inflation)

No. of observations Inflation

Model

Turning Point

(Inflation)

Fixed effects

Pooled OLS

All countries

Inflation and economic growth, 1961–2000

No. of observations: Inflation

Model

Table 6.2

Inflation and economic growth: a cross-country analysis

127

short, we do not obtain any reliable results on the inflation-growth relationship for the OECD countries. With the middle-income countries, the signs on the inflation coefficient are all positive. However, the coefficients are insignificant in all cases, the coefficient values correspondingly jump from 0.06 to 0.129. However, the estimated turning points in these equations are within a tight band of between 14–16 percent. Finally, with the low-income countries, we do again obtain consistently positive coefficient values from the inflation variable. These coefficients are also significant in the fixed effects and random effects models. The coefficient values in these regressions are substantially higher than with the other country groupings, ranging between 0.24 and 0.56. The inflationsquared terms are also strongly significant in the fixed and random effects models. The turning point estimates range between 15–23 percent. Results by decade In Table 6.3 we report results from regressions run separately for each of the four decades in our data sample. These regressions are run with annual data rather than five-year averages in order to generate a larger number of observations. In doing this, we recognize the point, emphasized by Bruno and Easterly, that we should expect more of a negative correspondence between inflation and growth as we move to higher frequency data samples. This is because of the likelihood that negative effects on growth will occur through short bursts of high inflation rates that approach our cut-off figure of 40 percent. Such short bursts of high inflation and slow or negative growth will be smoothed out when data are grouped at lower frequencies. One key point emerges from the results in Table 6.3: that the evidence for a positive association between growth and inflation is far stronger in the 1961–70 decade than in subsequent decades. For 1961–70, the coefficient values on inflation are all positive, though statistically significant only in the between effects model. The fixed effects model stands apart with a low inflation coefficient value of 0.065, but otherwise the coefficients for the other specifications are high, at 0.11 for the pooled OLS and random effects models and a very high 0.61 for the between effects model. With the 1971–80 sample, the inflation coefficients remain positive, but the coefficient values and levels of significance fall off, especially with the random effects and between effects models. For 1981–90, the inflation coefficients all turn negative, and there is a statistically significant negative value in the fixed effects model. Finally, for 1991–2000, we obtain negative inflation coefficients with two tests and close to zero coefficients for the other two.

128 718 −0.118* (1.98) 0.001 (0.61) 59.0

718 −0.016 (0.31) 0.00 (0.008) –

No. of observations Inflation

718 −0.016 (0.33) 0.00 (0.08) –

Random effects 85 −0.016 (0.17) 0.003 (0.91) 2.7

Between effects

59 0.61* (2.08) −0.024* (2.08) 12.7

Between effects

698 0.025 (0.31) −0.003 (1.30) 4.2

Pooled OLS

620 0.084 (0.99) −0.005* (2.00) 8.4

Pooled OLS 620 0.084 (0.92) −0.005* (2.03) 8.4

Random effects

698 −0.13 (1.69) −0.001 (0.43) −65.0

Fixed effects

698 0.01 (0.15) −0.003 (1.70) 1.7

Random effects

1991–2000

620 0.06 (0.61) −0.006* (2.21) 5.0

Fixed effects

1971–80

106 −0.189 (1.32) 0.004 (0.87) 23.6

Between effects

71 0.047 (0.20) 0.001 (0.13) −23.5

Between effects

Notes: 1. Annual data; dependent variable is GDP growth; t-statistics in parentheses; p , 0.05 5 *, p , 0.01 5 **). 2. All countries with less than 2 million people were excluded from the sample. All observations with inflation above 40 percent were excluded. Five-year period average data were used instead of yearly data in the regression. ‘All countries’ refers to all the countries in this table.

Turning Point

(Inflation)

Fixed effects

Pooled OLS

Model

Years

Turning Point

480 0.109 (1.24) −0.006 (1.74) 9.1

Random effects

1981–90

480 0.065 (.070) −0.007* (2.10) 4.6

480 0.112 (1.04) −.0005 (1.47) 11.2

No. of observations Inflation

(Inflation)

Fixed effects

Pooled OLS

1961–70

Inflation and economic growth all countries by decades

Model

Years

Table 6.3

Inflation and economic growth: a cross-country analysis

129

These results provide broad support for Bruno’s observation cited above, about inflation and growth moving positively together during the 1960s in correspondence with, as he put it, ‘very rapid growth because of investment demand pressures in an expanding economy’ (Bruno, 1995, p. 35). During the 1970s, demand management policies were still in favor to support growth. But the positive associations between growth and inflation as a byproduct of growth in this period were undermined by the two oil price shocks in 1973 and 1979. The overall inflation experience of this decade therefore is a combination of demand-pull effects from growth and supply side shocks. It is therefore not surprising that the inflation coefficients in the 1970s fall in value and lose significance. The 1980s marked the beginning of what Angus Maddison (2001), among others, has term the ‘neoliberal era’. Probably the single defining feature of this era is the virtual abandonment by governments throughout the world of Keynesian demand management policies as a tool for stimulating growth and employment. Thus, as a broad generalization, the inflation that is experienced in the 1980s and 1990s emerges almost entirely as a result of supply shocks and inertia, as opposed to demand-pull pressures. Within this context it is also not surprising that the inflation coefficients become consistently negative, albeit generally not to a statistically significant extent.

6.4

CONCLUSIONS

Considering first our full data set of 80 countries between 1961 to 2000, we have consistently found that higher inflation is associated with moderate gains in GDP growth up to a roughly 15–18 percent inflation threshold. However, the findings diverge when we divide our full data set according to income levels. With the OECD countries, no clear pattern emerges at all with either the inflation coefficient or our estimated turning point. Both the signs on the inflation coefficients as well as the turning points are highly sensitive to specifications. With the middle-income countries, by contrast, we return to a consistently positive pattern of inflation coefficients, though none are statistically significant. However, the turning points range within a narrow band in this sample, between 14.0 and 16.3 percent. With the low-income countries, we obtain positive and higher coefficient values on the inflation coefficient than with the middle-income countries. These coefficients are also statistically significant with the fixed and random effects models. Finally, with the groupings by decade, the results broadly indicate that inflation and growth will be more highly correlated to the degree that

130

Beyond inflation targeting

macroeconomic policy is focused on demand management as a stimulus to growth as opposed to macroeconomic austerity and inflation targeting. Overall, there is no evidence from this research supportive of a policy of maintaining inflation within a low band of about 3–5 percent, to the degree that government policy makers are interested in promoting economic growth and employment, rather than merely low inflation as an end in itself. At the same time, there is also no evidence that governments should allow inflation to rise above a 15–20 percent range in an effort to spur growth. This suggests that there is still a wide range of inflation rates that are very likely to be associated positively with economic growth. Certainly for the middle- and low-income countries, our results strongly suggest that allowing inflation to be maintained in the range of 10 percent or somewhat higher is very likely to be consistent with higher rates of economic growth. This is most especially the case when inflation is resulting from, as Bruno put it, ‘investment demand pressures in an expanding economy’. For the OECD countries, the primary conclusion that we can reach from our results is a negative one: that no generalization about the inflation-growth relationship is likely to find robust support from the available evidence. What appears likely for the OECD countries is that the wide range of relationships that emerge from the data reflect the differences in the sources of inflation – that is, whether inflation has resulted primarily from Keynesian type demand-pull forces as opposed to supply shocks and inertia. Some broad policy implications flow from these results. The first is that there is no justification for inflation targeting policies as they are currently being practiced throughout the world, that is, to maintain inflation with a 3–5 percent band and to adjust short-term interest rates as needed to dampen inflationary pressures beyond that targeted band. As a corollary, there is very likely to be positive growth benefits in middle- and low-income countries from allowing inflation to rise to a high single-digit range or even in some cases up to about 15 percent rather than dampening inflationary pressures through raising short-term interest rates. This is especially true to the extent that inflation within this range is resulting from demand-pull forces as opposed to supply shocks and inertia.6 A second implication is that researchers are likely to make productive contributions through giving increased attention on the inflation-growth relationship to some relatively underexplored aspects of the issue. The first is to be able to sort out with increased specificity the sources of inflationary pressures, given the likely wide disparities in the inflation-growth relationship depending on what is fueling inflation. A second is to focus more on policy measures for dampening inflation not at very low levels, but rather

Inflation and economic growth: a cross-country analysis

131

at levels approaching the upper limit of the positive growth-inflation association. This would be in the range of 10–15 percent for middle- and low-income countries. With the OECD countries, the acceptable range is likely to depend entirely on what are the primary sources of inflationary pressures. One well-known policy tool for maintaining inflationary pressures within a positive threshold range is some variation of incomes policies. Incomes policies have been widely used as an inflation control tool in a variety of contexts. One common situation has been in bringing down inflation after it has risen to a range above 40 percent. For example, in their paper ‘Moderate inflation’ Dornbusch and Fischer (1991) describe how Mexico in the 1980s drew upon experiences in Argentina, Brazil, Peru and Israel in developing a strategy to bring inflation down from the 100 percent range to something closer to 20 percent. Dornbusch and Fischer reported that the Mexicans learned two lessons from these experiences, (1) ‘that disinflation without fiscal discipline was unsustainable’; but that (2) ‘disinflation without incomes policy, relying solely on tight money and tight budgets, would be unnecessarily expensive’ (p. 31). In analysing the Israeli experience with disinflation over the 1980s, Bruno documents in detail the major contributions of incomes policies to the success of the effort (Bruno 1993, Chapter 5). A more directly relevant set of experiences with respect to inflations at more moderate levels have been the Nordic countries. This is because, in these countries, incomes policies have been used successfully as a tool for maintaining relatively low inflation over long periods of time rather than as primarily an instrument of disinflation after inflation exceeded 40 percent, as was true with Mexico and Israel. Sweden, for example, succeeded in maintaining unemployment at an average rate below 2 percent between 1951 and 2000 while still holding inflation at a 4.4 percent average rate. The application of incomes policies in Sweden, moreover, primarily took the form of centralized bargaining between unions and business, through which the aim of inflation control was recognized in the bargaining process. As such, the government did not have to rely on setting mandates for acceptable wage and price increases. The government did also utilize fiscal and monetary policies as tools for controlling inflation. But they did not have to apply these tools stringently, precisely because they were able to rely on their well-developed system of incomes policies as a complement to monetary and fiscal policies.7 The most basic critique of incomes policies is that, in order for the approach to have any chance of success, it is necessary that a country operate with a high level of organization among workers, and that there be some reasonable degree of common ground between workers and

132

Beyond inflation targeting

business. Otherwise, there will be no realistic prospect for economy-wide bargaining to yield results that will be honored widely. By its very nature, the relationship between unions and business in capitalist economies is likely to be highly contentious. But this could possibly diminish to the extent that both sides see the benefits accelerated economic growth and employment expansion as opposed to maintaining tight monetary policy for the purpose of holding inflation within a 3–5 percent band. This point brings us to a final issue for further research that includes both purely analytic as well as policy oriented implications. This is to examine the relationship between inflation and inequality in addition to the inflation-growth relationship. To the extent that inflation is associated with faster economic growth, it is likely to also be correlated with faster employment growth and thereby increased equality. At the same time, to the extent that wage agreements and social benefits do not include adequate cost-of-living adjustments, even a growth-generated inflation could yield greater inequality. In terms of policy implications, the issues that are central in the exploration of the inflation-inequality relationship will also be closely linked to the question of inflation control policies. For example, are incomes policies or inflation targeting a more effective means of promoting greater equality as well as economic growth? These are crucial questions that deserve substantial additional research in an effort to design more effective analytic foundations for the conduct of macroeconomic policy.

NOTES 1. This chapter was written as part of a broader project on pro-poor macroeconomic policies in sub-Saharan Africa, under the sponsorship of the United Nations Development Programme (UNDP). We are most grateful for the support of the UNDP. We are also grateful for comments on this specific aspect of the broader project from our co-workers Jerry Epstein, James Heintz and Leonce Ndikumana as well as from Michael Ash. 2. An excellent survey of inflation targeting and related issues in global monetary macroeconomics is Saad Filho (2005). 3. The literature on this issue is of course vast. Three references offering different perspectives are Cross (1995), Krueger and Solow (2001), and Saad Filho (2005). 4. Grouping the countries in the sample by average GDP levels does raise the potential for significant bias in the regression. This is because the dependent variable in the model is GDP growth. Strictly speaking, we are not dividing the sample based on the dependent variable, but there is obviously a close correspondence between the growth of GDP, our dependent variable and GDP levels, the variable on which we truncate the sample. To test for bias here, we have also divided the full sample based on pre-1960 GDP level groupings – that is, on the basis of data points that precede in time our sampling period. In this case the division is between current OECD countries and the non-OECD countries – that is, the demarcation between middle- and low-income countries was not so evident in the pre-1960 data, and only becomes evident over the 40 years that constitute

Inflation and economic growth: a cross-country analysis

133

our data sample. However, the results of this exercise do not vary substantially from those reported with the three GDP-level groupings reported here. This suggests that any potential bias from the GDP-level groupings is not a serious problem for our substantive understanding of the findings. 5. By exploring the inflation-growth relationship within this framework of a standard cross-country growth regression model, we are building in an assumption that causality in the relationship is running from inflation to growth. We do not explore the issue of simultaneity or reverse causality in this exercise, while we recognize it as an important issue for further research. 6. Even some of the most recent work by IMF economists has recognized that, at least for the low-income countries, inflation in the range of 5–10 percent is likely to be supportive of economic growth. See IMF (2005). 7. Different perspectives on the Nordic experiences are presented in Calmfors (1993), Pekkarinen et al. (1992), Flanigan (1999), Marshall (1994) and Iversen et al. (2000).

REFERENCES Bernanke, B., T. Laubach, A.S. Posen and F.S. Mishkin (1999), Inflation Targeting: Lessons from the International Experience, Princeton, NJ: Princeton University Press. Blinder, A.S. (1998), Central Banking in Theory and Practice, Cambridge, MA: MIT Press. Bruno, M. (1993), Crisis, Stabilization, and Economic Reform: Therapy by Consensus, Oxford: Clarendon Press. Bruno, M. (1995), ‘Does inflation really lower growth?’, Finance and Development, 32 (3) (September), 35–8. Bruno, M. and W. Easterly (1998), ‘Inflation crises and long-run growth’, Journal of Monetary Economics, 41, 3–26. Burdekin, R.C.K., A.T. Denzau, M.W. Keil, T. Sitthiyot and T.D. Willett (2004), ‘When does inflation hurt economic growth? Different nonlinearities for different economies’, Journal of Macroeconomics, 26, 519–32. Calmfors, L. (1993), ‘Centralization of wage bargaining and macroeconomic performance’, OECD Economic Studies, 21 (Winter), 161–91. Cross, R. (ed.) (1995), The Natural Rate of Unemployment: Reflections on 25 Years of the Hypothesis, Cambridge: Cambridge University Press. Dornbusch, R. and S. Fischer (1991), ‘Moderate inflation’, National Bureau of Economic Research working paper no. 3896. Flanigan, R.J. (1999), ‘Macroeconomic performance and collective bargaining: an international perspective’, Journal of Economic Literature, 37 (3) (September), 1150–75. Ghosh, A. and S. Phillips (1998), ‘Warning: inflation may be harmful to your growth’, IMF Staff Papers, 45 (4), 672–86. International Monetary Fund (IMF) (2005), ‘Monetary and fiscal policy design issues in low-income countries’, 8 August draft, Policy Development and Review Department and Fiscal Affairs Department, IMF, Washington, DC. Iversen, T., J. Pontusson and D. Soskice (eds) (2000), Unions, Employers, and Central Banks, Cambridge: Cambridge University Press. Khan, M.S. and A.S. Senhadji (2001), ‘Threshold effects in the relationship between inflation and growth’, IMF Staff Papers, 48 (1), 1–21.

134

Beyond inflation targeting

Krueger, A. and R.M. Solow (eds) (2001), The Roaring Nineties: Can Full Employment Be Sustained?, New York: The Russell Sage Foundation and the Century Foundation. Maddison, A. (2001), The World Economy: A Millennial Perspective, Paris: Development Centre of the Organisation for Economic Co-operation and Development. Marshall, M. (1994), ‘Lessons from the experience of the Swedish model’, in P. Arestis and M. Marshall (eds), The Political Economy of Full Employment, Aldershot, UK and Brookfield, US: Edward Elgar, Ch. 10. Pekkarinen, P., M. Pohjola and B. Rowthorn (1992), Social Corporatism: A Superior Economic System?, New York: Oxford University Press. Saad Filho, A. (2005), ‘Pro-poor monetary and anti-inflation policies: developing alternatives to the new monetary policy consensus’, manuscript, Department of Development Studies, School of Oriental and African Studies, University of London.

Inflation and economic growth: a cross-country analysis

APPENDIX 6.1A

135

COUNTRIES INCLUDED IN DATA POOL FOR ANALYSING THE INFLATION-ECONOMIC GROWTH RELATIONSHIP

Data Sample is 1960 to 2001 OECD countries

Middle-income countries

Non-OECD high-income countries

Low-income countries

Australia Austria Belgium Canada Denmark Finland France Greece Ireland

Algeria Argentina Bolivia Brazil Chile China Colombia Costa Rica Dominican Rep. Ecuador Egypt El Salvador Guatemala Honduras Hungary Iran, I.R. of Jamaica Jordan

Israel Singapore

Bangladesh Burundi Cameroon Central Afr. R. Congo Ghana Haiti India Indonesia

Italy Japan Korea Netherlands New Zealand Norway Portugal Spain Sweden Switzerland United Kingdom United States

Malaysia Mexico Panama

Paraguay Peru Philippines Poland South Africa Sri Lanka Syria Thailand Tunisia Uruguay Venezuela

Kenya Lesotho Malawi Mali Nepal Nicaragua Niger Pakistan Papua New Guinea Rwanda Senegal Sierra Leone Togo Uganda Zaire Zimbabwe

136

APPENDIX 6.2B

Beyond inflation targeting

SPECIFICATIONS OF VARIABLES IN THE FULL INFLATION/ ECONOMIC GROWTH MODEL

Economic growth. Real GDP per capita (Constant price: Laspeyres) growth rate. The nth year’s growth rate is calculated as the log value of the ratio of the nth year’s per capita GDP to the (n-1)th year’s per capita GDP (Penn World Table (PWT) 6.1, http://pwt.econ.upenn.edu/). Inflation. The increase of consumer price index (World Bank World Development Indicators (WDI), 2003). Initial output level. The log value of per capita GDP (Constant price: Laspeyres) at the beginning year of each period (PWT 6.1, http://pwt.econ. upenn.edu/). Investment. The share of gross investment in GDP (current prices) (PWT 6.1, http://pwt.econ.upenn.edu/). Fiscal policy. (1) The share of government consumption in GDP (current prices) (PWT6.1, http://pwt.econ.upenn.edu/). (2) Government budget deficit as percentage of GDP (WDI, 2003). Life expectancy. Life expectancy at birth (WDI CD-ROM, 2003, World Bank). Education level. Average years of secondary schooling of the total population aged 25 and over (R. Barro and J.W. Lee, 2000, http://www2.cid. harvard.edu/ciddata/barrolee/panel_data.xls). Terms of trade. The change of terms of trade weighted by foreign trade dependence ratio (the sum of exports and imports divided by GDP) (Easterly, William and Mirvat Sewadeh (2001), Global Development Network Growth database, www.worldbank.org/research/growth/ GDNdata.htm). Natural disaster. The share of population affected by the natural disasters happened in the year weighted by the share of agricultural output in GDP. Unreported natural disasters, if any, are treated as 0 (The natural disaster data come from the Centre for Research on the Epidemiology of Disasters (CRED), the OFDA/CRED International Disaster database, http://www.cred.be/emdat/intro.htm. The agricultural data are from WDI, 2003). War. A war is defined as an armed conflict with more than 25 deaths. Value 1 is given to those countries that experienced war within its border; –1 is given to those countries involved in war in other countries. Other situations are given value 0 (Gleditsch et al., 2002, Armed Conflict 1946– 2002 database, http://www.prio.no/cwp/ArmedConflict/).

PART III

Inflation targeting: critiques and country-specific alternatives

7.

Inflation targeting in Brazil: 1999–2006 Nelson H. Barbosa-Filho

Brazil adopted inflation targeting after a brief period of exchange rate targeting that ended up in a major currency crisis. More specifically, in 1994–98 the Brazilian government used high domestic interest rates and privatization to attract foreign capital and sustain an appreciated exchange rate peg. The main objective of the Brazilian economic policy at that time was to reduce inflation and the main side effect of exchange rate appreciation was a substantial increase in the country’s current account deficit and net public debt. Similar to what happened in Mexico and Argentina during the 1990s, the Brazilian macroeconomic stabilization strategy was heavily dependent on the continuous inflow of foreign capital and, as a result, the international financial position of Brazil became increasingly fragile after the contagion effects from the East Asian currency crises of 1997 and the Russian currency crisis of 1998. In fact, by the end of 1998 Brazil’s current account deficit reached 4.5 percent of GDP and the low stock of foreign reserves of the Brazilian Central Bank (BCB) did not allow a defense of the Brazilian currency, the real, in case another speculative attack hit the country. The inevitable currency crisis came in the beginning of 1999 and resulted in a ‘maxi-devaluation’ of the real. In numbers, the Brazilian real/US dollar exchange rate rose 57 percent in just two months, that is, from 1.21 in December 1998 to 1.90 in February 1999. After that the exchange rate dropped a little and then remained around 1.84 during the rest of 1999, that is, the exchange rate stabilized at a level 52 percent higher than before the crisis. The initial response of the BCB to such an abrupt devaluation of the real was a substantial increase in its base interest rate to stop the capital flight from Brazil and to reduce the pass-through of exchange rate depreciation to inflation.1 Then, after the exchange rate stabilized at a higher level in mid 1999, the government announced that it would start to target inflation. At that time the basic justification for such a move was that the government needed to substitute a price target for the former exchange rate target in order 139

140

Beyond inflation targeting

to coordinate market expectations and control inflation in a context of floating exchange rates. In terms of the classic policy ‘trilemma’ of open economies, the option for inflation targeting meant that Brazil chose to have an independent monetary policy, free capital flows and a floating exchange rate. In practice the situation was obviously different because the exchange rate was an important determinant of the Brazilian inflation rate during the period under analysis. On the one hand, the change in the domestic price of tradable goods in Brazil was (and it continues to be) basically determined by foreign inflation and exchange rate variations. On the other hand, the prices of some key non-tradable goods in Brazil were also tied to the exchange rate because, during the privatizations of the 1990s, the government allowed the price of some public utilities (basically in the telecommunication and energy sectors) to follow a price index that is heavily influenced by the exchange rate. The inevitable result was that a major part of the Brazilian inflation rate was linked to the exchange rate in 1999–2006.2 Given the centrality of the exchange rate for Brazilian inflation, it is no surprise that inflation targeting resulted in a disguised and loose exchange rate targeting by the BCB in 1999–2006. In theory the exchange rate was free to float, but in practice the country ended up with an asymmetric dirty floating regime, that is, a regime in which the BCB had to fight devaluations but to tolerate revaluations in order to meet the inflation targets set by the Brazilian government. As a result, the interaction between the international financial conditions, on the one side, and the Brazilian interest rates, on the other side, explains most of the behavior of the exchange rate in Brazil in 1999–2006, which in its turn explains most of the successes and failures of inflation targeting during that period. The objective of this chapter is to describe and analyse this experience.3

7.1

THE BRAZILIAN INFLATION TARGETING REGIME

The institutional basis of the Brazilian inflation targeting system can be described as follows: ●

The National Monetary Council (Conselho Monetário Nacional or CMN), formed by the Minister of Finance, the Minister of Planning and the President of the BCB, establishes the inflation targets based on the recommendations of the Finance Minister. All three members are appointed by the president and do not have fixed mandates.

Inflation targeting in Brazil: 1999–2006 ●

● ●

●

●

141

In June of every year the CMN establishes the inflation targets, and their corresponding intervals of tolerance, for the next two years. The target consists of the desired variation of a consumer price index (the IPCA index) estimated by the government’s statistical branch (the Instituto Brasileiro de Geografia e Estatística or IBGE). The BCB is responsible for achieving the target, but no specific instrument or strategy is specified.4 The Monetary Policy Committee (Comitê de Política Monetária or Copom), formed by the president and the directors of the BCB, decides the level of the BCB’s base interest rate, the so-known SELIC (Sistema Especial de Liquidação e Custodia) rate, needed to achieve the inflation target. Occasionally, additional actions such as changes in the banks’ reserve requirements can be taken.5 The target is considered met whenever the observed accumulated inflation during each calendar year falls within the interval of tolerance specified by the CMN. If the targets are not met, the president of the BCB has to issue an open letter to the Minister of Finance explaining the causes of the failure, the measures to be adopted to ensure that inflation returns to the target level, and the period of time needed for this to happen.

Since the BCB is not independent and its penalty for not meeting the inflation target is just to explain why that happened, we can conclude that the Brazilian inflation targeting regime is basically a loose way for the federal government to assure society, especially financial markets, that it will not let inflation run out of control.

7.2

MACROECONOMIC PERFORMANCE

The macroeconomic performance of Brazil since the adoption of inflation targeting has been mixed. First, considering inflation itself, the government’s targets were met when the international financial conditions allowed it, that is, inflation targeting was successful when the exchange rate dynamics helped the BCB efforts to control inflation. Second, when compared to the period of exchange rate targeting, inflation targeting brought a reduction in the base domestic real interest rate, but the Brazilian rate remained well above the international standards because the country needed to appreciate the exchange rate in order to meet its inflation targets. Third, the high real interest rate put an expansionary pressure on the net public debt and this had to be compensated by an increase in the government’s primary surplus, that is, the government budget surplus

142

Beyond inflation targeting

excluding net interest payments. Fourth, the growth performance of the economy under inflation targeting did not improve much when compared to the previous period of exchange rate targeting, and this happened even though the international environment was much more favorable for economic growth in 1999–2006 than in 1994–98.6 To facilitate the analysis, we look at each one of these issues separately below. Starting with inflation itself, Table 7.1 presents the annual inflation targets set by the CMN and the effective rates of inflation and exchange rate variation in 1999–2006. If we consider the whole period the inflation targets were met in five out of eight years, which on a first look seems to be a very good starting performance of inflation targeting in Brazil. However, when we look closer the real story cannot be considered that successful because the inflation targets were frequently changed according to the shocks that hit the Brazilian economy and, most importantly, disinflation was obtained mostly through exchange rate appreciation. In fact, if we exclude 1999 from the analysis, the inflation targets were met only when the exchange rate appreciated in nominal terms. To illustrate this point, Figure 7.1 shows that in 2000, 2004, 2005 and 2006 the target was met and the exchange rate appreciated, in 2001 and 2002 the target was not met and the exchange rate depreciated, and in 2003 the target was not met and the exchange rate appreciated. Now, before we move to the real interest rate, let us make a brief pause to present some of the history behind the numbers shown in Table 7.1. First, since the Brazilian inflation targeting regime started immediately after a currency crisis, there was a 7.2 percentage point (pp) increase in inflation in 1999 when compared to 1998.7 Despite such acceleration, the government inflation target for 1999 was met for two idiosyncratic reasons: first, the 1999 target was set in June of that year, when the government authorities already knew half of the annual inflation rate of 1999 and, second, the target for 1999 was not ambitiously low in order not to undermine the credibility of the new policy regime.8 Continuing with our inflation story, in 2000 the Brazilian exchange rate stabilized at a new and higher level, both in real and nominal terms, and this helped the BCB to reduce inflation and meet the inflation target set for that year. Then, in 2001, Brazil was hit by two major shocks that made the 6.0 percent inflation target unfeasible. First, because of an unexpected drought, there was a shortage in the supply of electricity from hydro-electric plants during 2001 and, since most of Brazilian electrical power comes from such a source, the adverse energy shock pushed inflation upwards. Second, and most important, the 2001 currency crisis in Argentina resulted in capital flight from and exchange rate depreciation in Brazil, pushing the domestic price of tradable goods up. Altogether

143

YES 36.7 56.4 0.3

n.a. 8.3 7.7 0.0 28.8 26.6

Nominal exchange rate variation End of the period Average of the period

GDP growth rate

Base interest rate Nominal base interest rate Real base interest rate 17.4 10.8

4.3

–2.2 0.9

YES

6.00 8.00 4.00 NO

6.0 7.0

2000 %

17.3 8.9

1.3

16.0 28.4

NO

4.00 6.00 2.00 NO

7.7 6.8

2001 %

19.2 6.0

2.7

35.1 24.3

NO

3.50 5.50 1.50 NO

12.5 8.4

2002 %

23.3 12.8

1.1

–15.3 5.4

4.00 6.50 1.50 YES 8.5 NO

9.3 14.7

2003 %

16.2 8.0

5.7

–4.8 –4.9

YES

5.50 8.00 3.00 NO

7.6 6.6

2004 %

19.1 12.7

2.9

–20.5 –16.8

4.50 7.00 2.00 YES 5.1 YES

5.7 6.9

2005 %

15.1 11.6

3.7

–1.3 –10.6

YES

4.50 6.50 2.50 NO

3.1 4.2

2006 %

Source:

Brazilian Central Bank (http://www.bcb.gov.br) and author’s calculation.

Note: The inflation targets are the targets set at least one year in advance. The exception is 1999, when the target was set in June of that year. The revised target is the target set in the corresponding year. The exchange rate variation refers to the Brazilian real/US dollar exchange rate, and the base interest rate is the SELIC interest rate, set by the BCB, and cumulated over the year. The real interest rate was obtained by deflating the nominal interest rate by the inflation rate cumulated over the year, according to the IPCA index.

25.6 15.3

8.00 10.00 6.00 NO

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

8.9 4.9

1999 %

Target inflation rate (end of the period) Initial target Ceiling Floor Was the target revised? Revised target Was the target met?

1.7 3.2

1998 %

Inflation, exchange rates, GDP growth and interest rates in Brazil

Effective inflation rate End of the period Average of the period

Table 7.1

144

Beyond inflation targeting 40 Inflation Exchange rate variation

35 30 25

Percent (%)

20 15 10 5 0 –5 –10 –15 –20 –25 1998

1999

2000

2001*

2002*

2003*

2004

2005

2006

Note: * Target not met. The inflation rate is the variation in the IPCA index, December to December, and the exchange rate variation the rate of change in the Brazilian real/US dollar exchange rate, also December to December. Source:

Brazilian Central Bank (http://www.bcb.gov.br).

Figure 7.1

Consumer inflation rate and exchange rate variation in Brazil

the final result of these two adverse shocks was a 1.7 pp increase in the Brazilian inflation rate in 2001, when compared to 2000.9 The macroeconomic turmoil of 2001 was followed by more instability in 2002, this time because of the Brazilian political cycle and the speculative international capital flows associated with it. More specifically, in 2002 Brazil had its presidential election and the looming victory of a leftist candidate, Lula, resulted in massive capital outflows and an unprecedented cut in Brazil’s access to foreign lines of credit. The exchange rate shot up as a result of such a move, reaching its highest level in real terms since the debt crisis of the early 1980s. The Brazilian inflation rate followed soon after and reached double-digit levels at the end of 2002. In fact, the exchange rate depreciation and inflation acceleration were so sharp and fast that they even made the Brazilian monthly base interest rate temporarily negative at the end of 2002. Altogether the inflation rate increased in another 4.8 pp in 2002, when compared to 2001. Lula was elected president in 2002 and his first term in office started with a sharp monetary and fiscal crunch in order to reduce inflation and restore the country’s access to foreign finance. Lula’s move and the very own excesses of the speculative attack to the Brazilian currency in 2002 quickly resulted in a return to normality, that is, a return to capital

Inflation targeting in Brazil: 1999–2006

145

4.0

3.5

3.0

2.5

2.0

1.5

7 l/0

Ju

6

07 n/

Ja

06

l/0

Ju

5

n/ Ja

05

l/0

Ju

4

n/

l/0

Ju

Ja

3

04

l/0

n/ Ja

Ju

2

03 n/

Ja

02

l/0

Ju

1

n/

l/0

Ju

Ja

0

01

l/0

n/ Ja

Ju

9

00 n/

Ja

99

l/9

Ju

n/

8

Source:

Ja

l/9

n/ Ja

Ju

98

1.0

IPEADATA (http://www.ipeadata.gov.br).

Figure 7.2

Nominal exchange rate in Brazil – Brazilian real per US dollar

inflows and exchange rate appreciation. Inflation decelerated to one-digit levels in 2003, with a reduction of 3.2 pp in relation to 2002. Despite this disinflation, the inflation target for 2003 was not met, even after the Lula administration revised it upwards because of the exchange rate depreciation of 2002.10 After the failure in 2001–03, the Brazilian inflation target regime started to perform well again in 2004–06. Most of this success can be credited to the exchange rate appreciation during these three years, which in its turn resulted from both the high domestic interest rates of Brazil and the favorable international trade and financial conditions in the rest of the world. More specifically, on the one hand, the BCB continued to practice high real interest rates in 2004–06 and this resulted in a gradual appreciation of the Brazilian exchange rate, in nominal and in real terms, as shown in Figures 7.2 and 7.3.11 On the other hand, the exchange rate appreciation of 2004–06 was also driven by the boom in Brazilian exports, pulled by the rise in the world demand for commodities, which in their turn increased the country’s trade and current account surpluses. It should be noted that the improvement in the Brazilian balance of payments happened despite the exchange rate appreciation for two reasons: first, the increase in international prices offset the reduction in the exchange rate for many sectors and, second, the depreciation in 2001 and 2002 was so

146

Beyond inflation targeting

180

160

140

120

100

80

Ja n/ 98 Ju l/9 8 Ja n/ 99 Ju l/9 9 Ja n/ 00 Ju l/0 0 Ja n/ 01 Ju l/0 1 Ja n/ 02 Ju l/0 2 Ja n/ 03 Ju l/0 3 Ja n/ 04 Ju l/0 4 Ja n/ 05 Ju l/0 5 Ja n/ 06 Ju l/0 6 Ja n/ 07 Ju l/0 7

60

Source:

Central Bank of Brazil (http://www.bcb.gov.br).

Figure 7.3

Index of the real effective exchange rate of Brazil (1992 5 100)

intense that only at the end of 2005 did the Brazilian real exchange rate return to the level verified in 1999. Altogether the inflation targeting performance of Brazil in 1999–2006 can be divided into three phases: (1) implementation, in 1999–2000, when the favorable international conditions and the modest inflation targets proved to be very successful; (2) crisis, in 2001–03, when a combination of adverse supply shocks and financial crises made the ambitiously low inflation targets unfeasible; and (3) consolidation, in 2004–06, when the very favorable international conditions and the high real domestic interest rates resulted in a quick reduction in inflation. Let us now turn to the evolution of the Brazilian base real interest rate throughout this process. Figure 7.4 shows the real base interest rate of Brazil since 1994 in order to allow a comparison between the exchange rate targeting and the inflation targeting regimes. The first thing that draws one’s attention is the very high real interest rate of Brazil during its exchange rate targeting regime and the substantial reduction brought by inflation targeting. In numbers, the average annual real base interest rate of Brazil was 21.9 percent in 1994–98, and 10.7 percent in 1999–2006.12 Despite such a substantial reduction in the real interest rate, it should also be noted that during the inflation targeting regime the Brazilian rate continued to be extremely high

Inflation targeting in Brazil: 1999–2006

147

30 INFLATION TARGETING

Percent (%)

25 20 15 10

EXCHANGE RATE TARGETING

5

07

-1

-1 20

06

-1 05

20

-1 20

04

-1 03

20

-1 20

-1

02 20

-1

01 20

-1

00 20

-1

99 19

-1

98 19

-1

97 19

-1

96 19

95 19

19

94

-1

0

Note: The real interest rate is the SELIC interest rate, set by the BCB, cumulated in the past 12 months, deflated by the consumer inflation rate also cumulated in the past 12 months, measured by the IPCA index. Source:

Central Bank of Brazil (http://www.bcb.gov.br) and author’s calculation.

Figure 7.4

Real annual base interest rate in Brazil

by international standards and, more important, the Brazilian real interest rate was extremely high when compared to the average GDP growth rate of the economy in 1999–2006, that is, 2.7 percent. Focusing our analysis on 1999–2006, Figure 7.4 shows that inflation targeting started with very high real interest rates in Brazil. As we mentioned earlier, the reason for this was the Brazilian 1999 currency crisis, which led the BCB to increase interest rates abruptly and substantially to stop the capital outflows from the country. Then, after inflation targeting was formally introduced, in mid 1999, the real base interest rate started to fall until it reached 9.5 percent at the end of 2000. After that the real interest rate fluctuated between 8 percent and 10 percent until the end of 2002, when the devaluation of the real associated with Lula’s election and the subsequent increase in inflation made BCB’s base interest rate temporarily negative. ‘Normality’ was quickly restored at the beginning of 2003, when the Lula administration raised the real interest rate of the economy to 14.3 percent in order to stop devaluation and decelerate inflation. When it became clear that such an effort was successful, in mid 2003 the real base interest rate started to fall gradually. The interest rate cuts persisted until the end of 2004, when the acceleration of GDP growth made the BCB fear that the market’s expectation for inflation would accelerate above the

148

Beyond inflation targeting

target set for 2005.13 The BCB’s response to such an expectational threat was another interest rate hike, which made the Brazilian real interest rate reach 13.4 percent in mid 2006. The increase in the interest rate in 2005 resulted in a sharp appreciation of the real and a deceleration of GDP growth. By the end of 2005 it was clear that the BCB had exaggerated in its response to the increase in the level of economic activity at the end of 2004, as well as that the huge discrepancy between Brazilian and international interest rates could compromise the country’s fiscal stability. The BCB’s response was another round of gradual interest rate cuts, which made the Brazilian real interest fall to ‘just’ 11.6 percent at the end of 2006, that is, a still abnormally high real interest rate by international standards and well above the GDP growth rate of the economy in that year, that is, 3.7 percent. Moving to fiscal policy, the main side effect of the Brazilian high real interest rates was the expansionary impact of net interest payments on public expenditures and debt. Table 7.2 presents the main fiscal numbers for the period, from which we can identify two distinct phases in the evolution of the net public debt of Brazil. First, in the first four years of inflation targeting most of the government net debt was directly or indirectly tied to the exchange rate because the Brazilian government was a net debtor in foreign currency and most of its domestic debt was formally indexed to either the exchange rate or the interest rate.14 So, in the wake of the 1999 currency crisis and the interest rate hike that followed it, the net interest payments by the Brazilian public sector shot up to 12.5 percent of GDP, and the ratio of its net public debt to GDP increased in 5.6 pp in just one year. Then, as the exchange rate stabilized at a new and higher level, the net interest payments by the public sector fell to less than 10 percent of GDP in 2000–01, but the net domestic debt of the government continued to rise gradually in relation to GDP because of the huge difference between the real interest rate, on the one side, and the growth rate of the economy, on the other side. The first phase of debt dynamics ended in 2002, when Brazil experienced another sharp and abrupt exchange rate depreciation, the net interest payments by the public sector shot up to 13 percent of GDP, and the net public debt of the country rose to 50.5 percent of its GDP. The second phase in the dynamics of public debt began in 2003, when the exchange rate appreciation started to reduce the government net interest payments. The ratio of net public debt to GDP followed the same path in 2004–06, when the Brazilian government was able to repay most of its foreign debt and accumulate a huge amount of foreign reserves. This movement was obviously facilitated by the favorable international trade and financial conditions of the period and, at the end of 2006, the Brazilian government became a net creditor in foreign currency. In contrast, on

149 15.8 15.1 −0.3 0.5

16.4 14.5 0.2 2.1

−3.0 12.5 0.9 11.5 9.5

35.2 4.2 31.0 9.4 44.5

1999

16.5 14.7 0.0 1.7

−3.3 7.4 0.8 6.6 4.1

36.5 3.9 32.7 9.0 45.5

2000

17.2 15.6 0.0 1.7

−3.4 8.2 1.1 7.1 4.8

38.9 3.9 34.9 9.6 48.4

2001

17.9 15.7 0.0 2.2

−3.7 13.0 1.2 11.8 9.3

37.5 4.2 33.3 13.0 50.5

2002

17.4 15.1 0.0 2.3

−3.9 7.2 1.1 6.1 3.3

41.7 4.2 37.5 10.7 52.4

2003

18.1 15.6 0.2 2.7

−4.2 6.5 0.9 5.6 2.3

40.2 4.4 35.8 6.8 47.0

2004

18.8 16.4 0.1 2.6

−4.4 7.1 0.7 6.5 2.8

44.1 4.7 39.4 2.3 46.5

2005

19.4 17.2 0.1 2.2

−3.9 6.8 0.3 6.5 2.9

47.6 5.1 42.5 –2.7 44.9

2006

Source:

Brazilian Central Bank (http://www.bcb.gov.br) and author’s calculation.

Note: The primary deficit is the government budget deficit excluding net interest payments. The nominal deficit is the total budget deficit, that is, the primary deficit plus net interest payments by the public sector. The numbers for the whole Brazilian public sector refer to the federal, state and municipal governments, and the state-owned enterprises.

Primary budget of the federal government in % of GDP Revenues Expenditures Net error and omissions Balance

0.0 7.4 0.3 7.1 7.4

Budget balance of the public sector in % of GDP Primary deficit Net interest payments On foreign debt On domestic debt Nominal deficit

1998 33.2 4.0 29.2 5.8 38.9

Fiscal policy in Brazil

Public debt in % of GDP Net domestic debt Monetary base Non-monetary debt Net foreign debt Net domestic and foreign debt

Table 7.2

150

Beyond inflation targeting

the domestic front, the high real interest rate continued to increase the Brazilian net public domestic debt, which reached the unprecedented high level of 47.6 percent of GDP at the end of the period. Altogether, we can say that in this second phase of debt dynamics inflation targeting was characterized by an increase in the ratio of net public domestic debt to GDP and a substitution of domestic debt for foreign debt. To complete our fiscal outlook, Table 7.2 also shows the evolution of the federal government primary surplus and nominal budget deficit in the inflation targeting period.15 Compared to the last year of exchange rate targeting, most of the inflation targeting period was characterized by a gradual increase in the federal primary surplus.16 Most of this increase happened in 1999, when the Brazilian federal government cut its spending and increased its revenues in proportion to GDP in order to avoid an explosive increase in its public debt. After that the federal government’s revenues and expenditures tended to grow together in relation to GDP, which in its turn had a small positive impact on economic growth through the balanced-budget multiplier. The second main fiscal crunch happened in 2003, when the federal government once more cut its spending in proportion to GDP in order to avoid an explosive increase in the country’s net public debt. However, differently from what happened in 1999, the federal government revenues also fell in proportion to GDP in 2003, which ended up stabilizing the primary surplus in terms of GDP.17 The federal government revenues and expenditures resumed growing faster than GDP in 2004–06, and fiscal policy became expansionary at the end of the period.18 Finally, considering GDP growth, the inflation targeting period had a slower rate of expansion than the exchange rate period, but a smaller volatility as well. To illustrate this point, Figure 7.5 shows the moving-average annual GDP growth rate of Brazil in 1994–2006. In numbers, on the one hand, the average GDP growth rate was 3.8 percent during the exchange rate targeting period, against 2.7 percent during the inflation targeting period. On the other hand, the maximum and minimum growth rates during exchange rate targeting were 8.5 percent and zero, respectively, whereas during inflation targeting the maximum and minimum rates were 5.7 percent and minus 0.8 percent, respectively. Finally, it should also be noted that the trend growth rate was stationary or declining during the exchange rate targeting period, but rising during the inflation targeting period. Figure 7.5 also reveals clearly the sequence of booms and busts experienced by the Brazilian economy since 1994. The downturns are usually associated with adverse shocks, and the upturns with expansionary macroeconomic policy. More specifically, the period starts with the boom

Inflation targeting in Brazil: 1999–2006 10

EXCHANGE RATE TARGETING

151

INFLATION TARGETING

Percent (%)

8 6 4 2 0 –2 95

19

96

19

-1

-1

-1

-1

-1

94

19

97

19

98

19

-1

-1

-1

99

19

00

20

01

20

03

20

04

20

05

20

-1

-1

-1

-1

-1

-1

02

20

06

20

07

20

Note: The growth rate is the rate of change of the average GDP during period t through t-3, in relation to the average GDP during period t-4 through t-7. Source:

IBGE (http://www.ibge.gov.br) and author’s calculation.

Figure 7.5

Moving-average of the annual GDP growth rate of Brazil

brought by inflation reduction in 1994. This boom ended in 1995, when the Brazilian economy suffered the contagion effect from the Mexican crisis and the Brazilian government adopted restrictive macroeconomic measures to avoid exchange rate depreciation and inflation acceleration. The second boom started after the Mexican crisis was absorbed by Brazil and lasted until 1997. Then, in 1997–99, the Brazilian economy experienced another growth deceleration because of the contagion effects from the East Asian crises of 1997 and the Russian crisis of 1998, and because of the recessive impact of their very own Brazilian crisis of 1999. The economy resumed growth in 2000, but the expansion was quickly curtailed by the Argentine crisis and the Brazilian energy rationing of 2001. The Brazilian economy started another recovery in 2002, but the expansion was short-lived because of the speculative attack to the Brazilian currency during the presidential elections of that year. The exchange rate depreciation and the restrictive macroeconomic measures adopted by the Brazilian government pushed economic growth down in 2003, and only in the beginning of 2004 did the economy start to recover. However, differently from all of the previous episodes, the 2004 expansion was cut for domestic reasons, namely the fear of the BCB that the economy was overheating at the end of that year. As we saw earlier, interest rates were raised and economic growth decelerated in approximately two

152

Beyond inflation targeting

percentage points in 2005, but it still remained well above the previous troughs. Then, at the end of the period under analysis, in the second half of 2006, the growth rate of the Brazilian economy started once again to accelerate.19

7.3

INFLATION TARGETING, REAL EXCHANGE RATES AND POTENTIAL OUTPUT

Whether or not Brazil started a new growth cycle in recent years is an open question that depends, among other things, on the management of monetary policy. From the recent history of inflation targeting in Brazil, there are two main macroeconomic challenges to the Brazilian authorities that will eventually have to be addressed. First, so far the success of inflation targeting depended heavily on a favorable behavior of exchange rate and this cannot go on indefinitely. By definition real exchange rate appreciation cannot go on forever, otherwise the country’s currency will become infinitely expensive and, more important, an excessively appreciated real exchange rate can reduce the growth prospects of the economy. Second, so far inflation targeting has been characterized by a slow GDP growth, both in comparison to Brazil’s previous history and to the growth rate of the rest of the world. Since also by definition any growth acceleration results in a temporary increase in the level of economic activity, it tends to create inflationary pressures and requires compensatory measures by monetary policy. However, if monetary policy is too conservative in setting a low inflation target and a fast speed of convergence to it, the economy may end up locked in a slow-growth equilibrium where the BCB ‘kills’ any growth acceleration for fears of rising inflation. To complete our analysis of inflation targeting in Brazil, let us look at these two challenges in more detail. First, regarding exchange rate, stable inflation requires a stable real exchange rate, but the level of the exchange rate is not determined a priori. There are multiple equilibria and, therefore, the real exchange rate can be stabilized at a competitive or an uncompetitive level. Thus, when the inflation target is ambitiously low and the speed of convergence to it too fast, there will be a tendency to real exchange rate appreciation, which in its turn may end up increasing the financial fragility of the economy in the medium run, especially if appreciation is sustained by speculative capital inflows. The usual logical sequence here is that high real interest rate leads to capital inflows and exchange rate appreciation, which in their turn leads to a reduction in the trade and current account balance of the economy, which in their turn make the economy more vulnerable to shifts in the

Inflation targeting in Brazil: 1999–2006

153

international liquidity. So, when inflation targeting is too ambitious and financed by speculative capital inflows, the result tends to be a sequence of booms and busts, during which GDP growth fluctuates around a low rate. Another possibility is that inflation targeting stabilizes the real exchange rate at a new low level without compromising the balance of payments of the economy.20 However, even in this case ambitious inflation targeting can still result in a slow growth rate because an appreciated real exchange rate does not stimulate the development of the domestic tradable sector. In other words, productivity growth is usually faster in the tradable sector and, therefore, an appreciated real exchange rate may end up reducing the overall growth rate of the economy.21 The natural solution to the exchange rate challenge outlined above is to combine inflation targeting with an asymmetric dirty floating that maintains the country’s real exchange rate stable and competitive in the long run. The dirty floating will have to be asymmetric because the instruments to combat appreciation are different than the ones use to combat depreciation in order to keep the real exchange rate competitive. In other words, in face of depreciation the government should use its traditional restrictive macroeconomic tools to stop the process and avoid an increase in inflation, without selling much of its foreign reserves. However, in the face of appreciation the government should buy the foreign currency to slow down the process, but with no open compromise with a specific exchange rate. The auxiliary measures to combat appreciation are obviously to increase imports and reduce domestic interest rates so that, at the end of the day, the asymmetric dirty floating ends up creating a sliding floor for the exchange rate.22 Second, regarding growth, inflation targeting is usually aimed at stabilizing economic growth at its potential level, so that it eliminates excess demand pressures that tend to increase inflation, and insufficient demand pressures that do the opposite. The logical form of the orthodox standpoint is that the long-run growth rate of the economy is given from outside of macroeconomic issues, by preferences and institutions, and the only thing a responsible central bank can do is to make the economy grow at its potential rate in the long run. So, according to the orthodox view of economic policy, attempts to grow faster than such supply-determined limits end up increasing inflation, with either no or negative permanent real effects on the economy. The main problem with such a view is that it fails to recognize that the potential output of an economy is an endogenous variable and, therefore, it can and usually is affected by macroeconomic policy itself, including monetary policy. More specifically, potential output is an unobservable variable that is estimated from the past and expected behavior of the economy.23 Since

154

Beyond inflation targeting

expectations are usually heavily influenced by the recent past, the estimates of potential output tend to extrapolate current trends to the near future and, in this way, they may create a self-fulfilling monetary policy. An example helps to illustrate the point. Suppose that a conservative central bank estimates a slow potential growth rate for the economy and, based on such a pessimistic estimate, it combats a growth acceleration for fearing that it will increase inflation above the pre-specified target. In doing so the very own actions of the central bank reduce the effective growth rate of the economy, especially of investment, and when potential output is re-estimated again ex post, the data will confirm that the central bank was right not necessarily because its initial estimates were right, but because the growth deceleration produced by the central bank is automatically translated in a lower estimate of the growth potential of the economy.24 In short, since the estimates of potential output are highly uncertain at the end of the sample and continuously revised every time a new observation is incorporated into the analysis, the actions of the central bank may seem to be correct ex post because its very own actions produced the scenario that justified it ex ante. The endogeneity of potential output at the end of the sample is especially dangerous for an economy, as Brazil, that aims to accelerate its growth rate. The danger here is that because the growth rate in the recent past was low, any strong growth acceleration tends to generate an apparent excessive output gap in the short run. So, if the central bank stops the process too soon, there won’t be enough time for the estimates of potential output to pick the structural change in the growth prospects of the economy. The final result is either that the economy never takes off, or that it takes off very slowly. On the other hand, the endogeneity of potential output does not obviously mean that authorities live in a sort of ‘field of dreams’, in which, if you believe, growth will come.25 There are objective limits to the growth rate of the economy that cannot be avoided by optimistic expectations like, for instance, the maximum supply of energy, the stock of foreign reserves and a zero unemployment rate. The endogeneity of potential output only means that there may be more than one equilibrium position for the GDP growth rate and, therefore, the speed of convergence to the inflation target is also a relevant variable for economic growth. As usually happens in economics, central banks face a trade-off in this area: on the one hand, quick disinflation may lock the economy in a slow-growth equilibrium, but, on the other hand, slow disinflation may lock the economy in a high-inflation equilibrium. Even though this trade-off has yet to be emphasized by modern macroeconomic theory, it is an unavoidable matter for policy makers who deal with the real world.

Inflation targeting in Brazil: 1999–2006

7.4

155

CONCLUSION

Inflation targeting represents an improvement in relation to the previous monetary policy regime adopted by Brazil, but it can and should be improved in order to increase the growth rate of the economy. In general terms the main conclusions from the previous sections can be summarized in ten points. (1) Inflation targeting managed to reduce inflation in Brazil after its 1999 and 2002 currency crises, with a substantial help of exchange rate appreciation. (2) Economic growth was slower under inflation targeting than under exchange rate targeting, but with a smaller volatility and with an apparent upward trend. (3) Inflation targeting reduced the real interest rate of the economy, which nevertheless remained well above international standards and more than three times higher than the GDP growth rate of Brazil in 1999–2006. (4) The high real interest rate required a substantial increase in fiscal austerity by the Brazilian government, but so far this has not been sufficient to stop the increase in the ratio of net public domestic debt to GDP. (5) The high domestic real interest rates and the favorable international trade and financial conditions in the rest of the world allowed the Brazilian government to accumulate foreign reserves, repay most of its foreign debt and reduce its dependence of foreign capital in 2003–06. (6) Inflation targeting can be combined with an asymmetric dirty floating regime, in which the central bank combats exchange rate depreciation with restrictive monetary policy, and accumulates foreign reserves to slow down appreciation. (7) The asymmetric dirty floating should aim at a stable competitive real exchange rate, in order to promote the fast growth of the domestic tradable sector, and in this way push the country’s exports and imports up without increasing the foreign financial fragility of the economy. (8) Aggregate estimates of potential output are not good guides for the demand pressures on inflation in moments of structural growth acceleration or deceleration. (9) Because of the endogeneity of potential output at the end of the sample, inflation targeting should be done with moderation in order to avoid a self-fulfilling monetary policy that locks the economy in a slow-growth equilibrium. (10) A fast convergence to a low inflation target may produce a permanently low growth rate, but a slow convergence to a moderate inflation target can produce a permanently high inflation.

NOTES 1.

In this chapter we define the exchange rate as the domestic price of foreign currency, so that a depreciation of the exchange rate means an increase in the exchange rate, that is,

156

2.

3. 4. 5.

6.

7. 8.

9. 10.

11. 12. 13. 14. 15. 16. 17. 18. 19. 20.

Beyond inflation targeting a devaluation of the domestic currency. By analogy, appreciation means a reduction in the exchange rate, that is, a revaluation of the domestic currency. For instance, according to the estimates for that time (Belaisch, 2003), 23 percent of exchange rate variations tended to pass through to consumer prices (measured by the IPCA index) in the long run, whereas the pass-through to the general price level (measured by the IGPDI index) was 71 percent. For a more technical analysis of inflation targeting in Brazil, see Bogdanski et al. (2000), Minella et al. (2003), Tombini and Alves (2006) and Bevilaqua et al. (2007). According to the Brazilian Presidential Decree 3088, of 21 June 1999, it is up to the BCB ‘to execute the necessary policies to meet the specified targets’. In 1999–2005 the Copom met every month to determine interest rates. Starting in 2006 the Copom has been meeting eight times a year, that is, one meeting approximately every six weeks. If necessary, an extraordinary meeting can be called by the president of the BCB. According to the IMF estimates, the average world GDP growth rate was 3.7 percent in 1994–98 and 4.2 percent in 1999–06. In its turn, the inflation targeting period can be divided in two phases regarding the world GDP growth rate, that is, slow-growth, in 1999–06, when the average world annual GDP growth rate was 3.5 percent, and fast growth, in 2003–06, when the average annual growth rate was 4.9 percent. Unless stated otherwise, all numbers are annual figures. The inflation rate of 1999 was high when compared to 1998, but modest when compared to the magnitude of the exchange rate depreciation verified in 1999. As we will see later in this section, most of the small pass-through of the exchange rate to domestic prices in 1999 can be credited to the high real interest rate and the abrupt fiscal crunch practiced by the Brazilian government during that year. As we will also see later in this section, the high real interest rate practiced by the BCB and the very own recessive impact of the two supply shocks on the level of economic activity helped to reduce the magnitude of inflation acceleration. One of the first measures of the Lula administration was to increase interest rates, cut public spending and increase the inflation target for 2003, from the 4.0 percent set by the previous administration to 8.5 percent because of the adverse ‘electoral’ shock of 2002. A complete analysis of Lula’s economic policy is beyond the scope of this chapter. The interested reader can find more information in Barbosa-Filho (2007) and Arestis et al. (2007). Note that most of the real exchange rate appreciation was concentrated in the first semester of 2003 and in 2005, which coincided with the periods of increases in the Brazilian base interest rate, as we will see later. These are geometric averages and the real interest rate was calculated ex post. At the time there was an intense debate in the media on whether or not the economy was really overheating, since the increase in inflation at the end of 2004 also reflected an increase in an important indirect tax rate, the so-known PIS-COFINS rate. Indexation to the interest rate increased the impact of exchange rate depreciation on domestic debt because the BCB usually increased its base rate abruptly after a devaluation of the domestic currency. The nominal deficit is the difference between the total government expenditures and revenues, that is, it is the net interest payments minus the primary surplus. Only in 2006 the primary surplus fell in relation to GDP, but it was still higher than the level verified in 2002. Nevertheless, the combined reduction in public revenues and spending had a negative impact on economic growth in 2003 through the balanced-budget multiplier. In 2003–06 all of the increase in government expenditures in percentage of GDP was directed to an increase in income transfers through social security, social assistance and unemployment benefits (Barbosa-Filho, 2007). The expansion, started in 2006, gained force in 2007, when the Brazilian GDP grew 5.4 percent. For a formal model of this case, see Barbosa-Filho (2006).

Inflation targeting in Brazil: 1999–2006 21. 22. 23. 24.

25.

157

For an analysis of the link between real exchange rates and development, see Frenkel and Taylor (2006). This idea has been first proposed by Ros (1995) for Mexico. For a more recent analysis of inflation targeting in Mexico, see Galindo and Ros (2006). For an analysis of the methods of estimation of potential output applied to Brazil, see Barbosa-Filho (2005). So far most of the mainstream studies on the uncertainty associated with estimates of potential output have been focused on the reliability of forecasts based on realtime data, rather than on the implications of a self-fulfilling monetary policy. For an example of the mainstream approach, see Orphanides and van Norden (2005). Field of Dreams is a 1989 movie in which a farmer becomes convinced by a mysterious voice that, if he constructs a baseball diamond in his corn field, the ghosts of deceased star baseball players will come to play. The voice repeatedly says to the farmer: ‘If you build, he will come’, and the Hollywood story obviously ends up confirming the voice’s prediction.

REFERENCES Arestis, P., L.F. de Paula and F. Ferrari (2007), ‘Assessing the economic policies of President Lula da Silva in Brazil: has fear defeated hope?’, Oxford Centre for Brazilian Studies working paper CBS-81-07. Barbosa-Filho, N.H. (2005), ‘Estimating potential output: a survey of the alternative methods and their applications to Brazil’, IPEA working paper no. 1092. Barbosa-Filho, N.H. (2006), ‘Exchange rate, growth and inflation’, paper presented at the Annual Conference on Development and Change, Campos do Jordão, Brazil. Barbosa-Filho, N.H. (2007), ‘An unusual economic arrangement: the Brazilian economy during the first Lula administration, 2003–2006’, International Journal of Politics, Culture and Society, 19, 193–215. Belaisch, A. (2003), ‘Exchange rate pass-through in Brazil’, International Monetary Fund working paper no. 03-141. Bevilaqua, A.S., M. Mesquita and A. Minella (2007), ‘Brazil: taming inflation expectations’, Central Bank of Brazil working paper no. 129. Bogdanski, J., A.A. Tombini and S.R.C. Werlang (2000), ‘Implementing inflation targeting in Brazil’, Central Bank of Brazil working paper no. 1. Frenkel, R. and L. Taylor (2006), ‘Real exchange rate, monetary policy and employment’, United Nations, Department of Economic and Social Studies working paper no. 19. Galindo, L.M. and J. Ros (2006), ‘Alternatives to inflation targeting in Mexico’, University of Massachusetts Political Economy Research Institute, alternatives to inflation targeting: central bank policy for employment creation, poverty reduction and sustainable growth working paper no. 7. Minella, A., P.S. de Freitas, I. Goldfajn and M.K. Muinhos (2003), ‘Inflation targeting in Brazil: constructing credibility under exchange rate volatility’, Central Bank of Brazil working paper no. 77. Orphanides, A. and S. van Norden (2005), ‘The reliability of inflation forecasts based on output gap estimates in real time’, Journal of Money, Credit and Banking, 37 (3), 583–601. Ros, J. (1995), ‘La crisis mexicana: causas, perspectivas, lecciones’, Nexos, (May), 46. Tombini, A.A. and S.A.L. Alves (2006), ‘The recent Brazilian disinflation process and costs’, Central Bank of Brazil working paper no. 109.

8.

Alternatives to inflation targeting in Mexico Luis Miguel Galindo and Jaime Ros1

8.1

INTRODUCTION

Inflation targeting has become increasingly popular over the past decade. As a nominal anchor for monetary policy with a public and explicit commitment to maintain economic discipline, inflation targeting is being promoted as a general framework in order to reduce and control the inflation rate, improve inflation predictability, accountability and transparency, reduce the expected inflation variability (Sheridan, 2001), improve the output-inflation trade-off (Clifton et al., 2001), help solve the dynamic consistency problem and even reduce output variability (Svensson, 1997, 1998). Several Latin American countries such as Chile (1990), Peru (1994), Mexico (1999), Brazil (1999) and Colombia (1999) have moved their monetary regimes to an inflation targeting framework (Corbo et al., 2002; Schmidt-Hebbel and Werner, 2002). There are also a number of concerns regarding the adoption of an inflation targeting regime. There is an important concern about the ability of the central bank to control the inflation rate, in particular under the presence of fiscal or external shocks. There are also fundamental doubts about the economic consequences of an inflation targeting regime. For example, Ball and Sheridan (2003) argue that the reduction of inflation and the reduction in output variability in the OECD countries is a general trend that is not necessarily related to the instrumentation of inflation targeting and that this kind of regime does not affect output variability. Furthermore, Newman and von Hagen (2002) claim that the evaluation of inflation targeting has been misleading because it has not properly considered the problem of regression to the mean given that some inflation targeting countries were worse off before their implementation than the countries without an inflation targeting framework. In the case of emerging market economies there is a special concern about the relationship of inflation targeting with exchange rate movements, imperfect and poorly regulated financial markets and weak 158

Alternatives to inflation targeting in Mexico

159

monetary institutions. In effect, there are important concerns about the effectiveness of an inflation targeting regime under weak fiscal conditions, poorly regulated financial systems, large potential external shocks, low institutional credibility and currency substitution phenomena (Fraga et. al., 2003; Calvo and Mishkin, 2003). There is also a major concern that important shifts in exchange rates will affect the general competitiveness of the economy, the current account deficit and generate external shocks on the inflation rate. For example, abrupt changes in the exchange rate might lead to inflation paths that are inconsistent with the original inflation target and therefore the central bank might try to use the nominal exchange rate as a nominal anchor (Svensson, 1998). In this case exchange rate shocks to the inflation rate are followed by a general appreciation of the real exchange rate affecting the general performance of the economy. Goldfajn and Gupta (2003) argue also that an appreciation of the real exchange rate is related with higher nominal interest rates or a tight monetary policy that makes economic recovery more difficult. This chapter addresses Mexico’s experience with inflation targeting. Section 8.2 presents some background on the operation of monetary policy in Mexico since the 1994–95 peso crisis. Section 8.3 assesses empirically a number of important issues in the evaluation of inflation targeting: the effects of the real exchange rate on output, the question of the passthrough of exchange rate movements into inflation and the asymmetric response of monetary policy in the face of exchange rate shocks. Section 8.4 considers alternatives to inflation targeting as currently implemented. Section 8.5 concludes.

8.2

INFLATION TARGETING IN MEXICO: SOME BACKGROUND

In the 1990s Mexico experienced a variety of monetary and exchange rate regimes. More precisely, the monetary regime went through three stages: nominal exchange rate targeting with a crawling band regime before the 1994 crisis, monetary targeting for a short period after the crisis, followed by a transition to inflation targeting that by now has been largely completed. With the 1994 crisis the exchange rate regime shifted from a crawling band to floating. The experience with the crawling band regime and its crisis has been analysed widely (see, for example, Lustig and Ros, 1998; Ros, 2001). It was during this period, in 1993, that the central bank was given independence and a mandate to preserve price stability. After the 1994–95 peso crisis, monetary policy focused on the growth of monetary aggregates and

160

Beyond inflation targeting

limits to credit growth as a means to control the inflationary effects of the sharp peso devaluations and rebuild the damaged credibility of the Banco de México. More precisely, the central bank established as its nominal anchor an intermediate target for the growth of the monetary base. As the country moved to a flexible exchange rate regime, an assumption of no international reserves accumulation was made. The ceiling on the growth of the monetary base was thus in essence a growth ceiling on net domestic credit. This monetary policy framework was soon abandoned as the policy failed to stabilize inflationary expectations, the exchange rate and inflation itself. The main reasons for this failure were the instability of the relationship between the monetary base and inflation and the fact that the central bank has no control of the monetary base in the short run. The demand for bills and coins in circulation largely determines the monetary base and this demand is very interest inelastic in the short term (Carstens and Werner, 1999). The peso crisis had strong inflationary effects, taking the inflation rate from single digits to over 50 percent. These inflationary effects were largely brought under control by means of a tight fiscal policy and an incomes policy negotiated with unions and business confederations. After this, the main objective of monetary policy became the reduction of inflation in a gradual and sustainable way. At the same time, the monetary policy regime moved towards influencing the level of interest rates, establishing borrowed reserves as its key instrument with the monetary base becoming less relevant and the inflation target more important in the conduct of policy. Under this framework the Banco de México establishes that banks should seek a null balance in their accounts at the central bank. A penalty rate (double the government treasury bill rate, the CETES rate) is applied to overdrafts and positive balances are not remunerated. Thus, the banks seek this balance in particular because if the daily average balance were negative they would have to pay the penalty rate. By providing more or less liquidity through daily monetary auctions, the overall net daily average balance of all current accounts held by banks at the Banco de México may close the day at a predetermined amount. If negative, the central bank puts the banking system in ‘short’ (‘corto’) in which case at least one credit institution has to pay the penalty interest rate. This is the way in which the central bank exerts pressure on interest rates. Although the corto represents only a small amount of the total liquidity (normally between 0.09 and 0.068 of the monetarty base), increasing it induces banks to bid up interest rates to avoid paying overdraft charges and puts upward pressure on market interest rates. The reverse mechanism applies when the central bank targets a positive balance (‘largo’) (Banco de México, 1996, Appendix 4; Carstens and Werner, 1999, pp. 15–16; OECD, 2002, 2004).

Alternatives to inflation targeting in Mexico

161

The transition to inflation targeting was accelerated in January 1999 when the Banco de México announced a medium-term inflation objective, and since 2000 the central bank publishes quarterly inflation reports to monitor the inflationary process, analyse inflation prospects and discuss the conduct of monetary policy. By now Mexico is considered to have in place the main components of an inflation targeting framework: an independent monetary authority (since 1993) that has inflation as its only policy objective, a flexible exchange rate regime, the absence of other nominal anchors and a ‘transparent’ framework for the implementation of monetary policy (Schmidt-Hebbel and Werner, 2002). The implementation involves the choice of an inflation index to be adopted as target, the target range and the time horizon. The national consumer price index (CPI) is used to determine the inflation objective. Since the beginning of 2000 the bank also tracks a core CPI which excludes volatile items – such as agricultural and livestock products, and also education (tuition fees) – and prices controlled by or agreed with the public sector. In the transition from high to moderate inflation rates, the inflation objective was specified in terms of a value that should not be exceeded. Today policy specifies a target range of plus or minus a percentage point. From January 1999 onwards the target of monetary policy has been framed in terms of a medium-term inflation objective of bringing down inflation to the rate prevailing in Mexico’s main trading partners by the end of 2003, interpreted as an annual growth of the CPI of 3 percent. Table 8.1 shows the inflation targets since the 1994–95 crisis, actual inflation performance, the growth projections and performance as well as the evolution of the real exchange rate. After a rough start, when in 1995 the inflation objective was missed by over 30 percentage points, the central bank’s record has been improving over time with its inflation target being met in five years since 1999 and inflation tending to converge towards its medium-term target range of 3 percent plus or minus one percentage point. Along with the success in bringing down inflation, the growth performance has been disappointing. More precisely, in the second half of the 1990s, under the stimulus of a very competitive exchange rate, the economy rapidly recovered from the 1995 recession and grew at rates that often surpassed the growth rate projected by the central bank and the government. However, from 2001 to 2003, growth sharply decelerated to rates that for three years in a row were below or barely above the rate of population growth. That is, the economy recorded a decline in per capita incomes from 2000 to 2003 before recovering in 2004. Overall performance since 1994 to 2004 has been unsatisfactory with GDP growth below 3

162 127.8

141.4

114.1

15 15.7 .4 6.8

1997

116.1

12 18.6 5 4.9

1998

105.8

13 12.3 3 3.7

1999

100.0

10 9.0 4.5 6.6

2000

92.5

6.5 4.4 n.a. –0.1

2001

102.3

4.5 5.7 1.5 0.7

2002

2004

2005

106.2

103.2

98.5

3±1 3±1 3±1 4.0 5.2 3.3 3.0 3.0–3.5 3.8 1.3 4.2 3.0

2003

98.7

3±1 4.1 3.6 4.5

2006

Source: Banco de México, www.banxico.org.mx/tipo/estadisticas/index.html. The real exchange rate is defined as the nominal exchange rate to the US dollar, multiplied by the ratio of foreign (CPI of USA) to domestic price levels (CPI of Mexico).

10 27.7 .3 5.1

1996

19 52.0 n.a. –6.2

1995

Inflation, growth and the real exchange rate

Inflation target (%) Actual inflation (%) GDP growth projection Actual GDP growth Real exchange rate (end of the year) (index)

Table 8.1

Alternatives to inflation targeting in Mexico

163

percent per year, well below the historical rates of the period 1940 to 1980 (6 to 6.5 percent).

8.3

THE EMPIRICAL EVIDENCE

In this section we address three relevant questions for the evaluation of the inflation targeting regime. First, what are the short-term and long-run effects of the real exchange rate on output? Have these effects changed with trade liberalization and integration with the US economy? Second, how important is the pass-through of the exchange rate to prices? Has inflation targeting modified the pass-through? Third, does inflation targeting have a bias towards exchange rate appreciation? If so, and the real exchange rate has long-run effects on output, the regime is not neutral with respect to economic growth. However, inflation targeting monetary policies are relatively recent making it difficult to use several econometric techniques. In this sense, these results must be considered only preliminary. The database consists of quarterly data for the period 1980 to 2003 and 1999 to 2003. The selection period was due to data limitations. A definition of the variables is included in the Appendix 8.1A. 8.3.1

The Real Exchange Rate and Output

The net effect of the real exchange rate on output is not clear-cut because there are alternative transmission channels with opposite effects in the short and long runs.2 In order to assess the long-run impact of the real exchange rate on output we proceed to estimate a vector autoregressive model (VAR) model including output (Yt), investment (INVt), US output (YUSt) and the real exchange rate (SRt).3 The unit roots tests of these variables are summarized in Table 8A.1 in Appendix 8.1A. These tests indicate that these variables are all I (1) and therefore it is necessary to consider the option of possible cointegration4 among them. The statistics of the Johansen (1988) procedure5 (Table 8.2) indicates the presence of at least one cointegrating vector (lower case letters refer to the natural logarithm of the series) already correcting for certain instability in the cointegrating space (Johansen et al., 2000). Normalizing this vector as an output equation (Equation 8.1) indicates the presence of a positive relationship between output, investment, US output and the real exchange rate. Therefore, a devaluation of the real exchange rate has a positive impact on Mexican economic growth in the long run.

164

Table 8.2

Beyond inflation targeting

Statistics of the Johansen procedure including output, investment, US output and the real exchange rate, period 1981:01–2003:04 yt 5 b1*invt 1 b2*yust 1 b3*srt**

H0 r50 r≤1 r≤2 r≤3

Constant

Trend

Trace

95%

m0 m0 m0 m0

0 0 0 0

63.09* 20.57 6.10 0.01

47.21 29.68 15.41 3.76

Notes: * Significant at the 5 percent level; Trace 5 Trace test; r 5 number of cointegrating vectors. Number of lags in the VAR 5 4; VAR includes constant unrestricted. ** y: output; inv: investment; yus: US output; sr: real exchange rate.

yt 5 0.478*invt 1 0.320*yust 1 0.019*srt

(8.1)

The series included in the estimated VAR are all considered initially as endogenous variables. Under these circumstances, the causal relations among these variables are not clear while the normalization of the cointegrating vector already implies a particular classification between endogenous and exogenous variables. Thus, it is important to take the results with some caution due to the potential endogeneity problem (Lütkepohl and Reimers, 1992). Therefore, we perform an impulse response exercise in the VAR including, in the following order: yust, yt, invt and srt. The impulse response function incorporates information on contemporaneous effects since it ignores the contemporaneous correlations of residuals. In order to avoid this pitfall it is common to orthogonalize innovations. The orthogonalized innovation impulse response function should be interpreted with caution since orthogonalization imposes a pre-order which implies a semi-structural interpretation of the model. The variable which enters first in the system acts as the most exogenous. The movements in these variables at one point in time precede the movements of the rest of the variables which come afterwards in the system. The impulse response analysis, using the Cholesky decomposition and the new order of the variables in the system, confirms the positive effect of investment and US output on Mexico’s economic growth and indicates that the real exchange rate has an initial negative effect that tends to disappear in the long run (Figure 8.1). The finding that an appreciation of the real exchange rate has longrun contractionary effects on economic growth in Mexico contradicts

Alternatives to inflation targeting in Mexico Response of output (y) to US output (yus)

0.03

0.02

0.01

0.01

0.00

0.00

–0.01

–0.01

–0.02

Response of output (y) to output (y)

0.03

0.02

165

–0.02 1

2

3

4

5

6

7

8

9 10

Response of output (y) to investment (inv)

0.03

1

0.02

0.01

0.01

0.00

0.00

–0.01

–0.01

3

4

5

6

7

8

9 10

Response of output (y) to real exchange rate (sr)

0.03

0.02

2

–0.02

–0.02 1

2

3

4

5

6

7

8

9 10

1

2

3

4

5

6

7

8

9 10

Note: The horizontal axis refers to the ten periods or quarters while the vertical axis shows the response of the inflation rate to a one standard deviation shock in each variable in the system.

Figure 8.1

Impulse – response analysis for yust, yt, invt and srt

previous results of impulse response analysis that the long-run relationship between output and the real exchange rate is negative (Kamin and Rogers, 1997). It is also worth noting that the effect of the real exchange rate on output has significantly increased after the beginning of the North American Free Trade Agreement (Table 8.3). Therefore, the Mexican economy seems to be more sensitive to exchange rate changes today than in the past. 8.3.2

The Nominal Exchange Rate and Inflation (Pass-through)

The positive impact of the nominal exchange rate on the inflation rate (pass-through) is one of the main concerns of the monetary authorities. Under these conditions, an inflation targeting regime is prone to suffer from external dominance. That is, the high sensitivity of inflation to any

166

Table 8.3

Beyond inflation targeting

Cointegration vectors of the Johansen procedure including output, investment, US output and the real exchange rate yt 5 b1*invt 1 b2*yust 1 b3*srt

Period 1982(1)–1993(4) 1994(1)–2003(4)

invt

yust

srt

0.852 0.761

0.328 0.420

0.475 0.855

exchange rate depreciation could cause external inflation shocks that will make it difficult for the central bank to achieve its inflation target. It has also been argued that an inflation targeting regime reduces the pass-through problem due to its ability to increase the credibility of the monetary authorities and therefore the importance of forward-looking variables (Fraga, et al., 2003; Schmidt-Hebbel and Werner, 2002). We evaluate the pass-through effect in Mexico. First, we estimate a VAR including the inflation rate, the output gap and changes in the nominal exchange rate. The sample period covers quarterly data from 1986:01 to 2003:04. This specification is relatively similar to a traditional Phillips curve and includes some elements of the VAR specification used in Schmidt-Hebbel and Werner (2002) or Fraga et al. (2003). That is, the output gap should have a positive impact on the inflation rate due to cost pressures through the labor market and input costs while a devaluation increases import costs and tradable goods prices and has a positive impact on the inflation rate. Additionally, it is possible to argue that this VAR includes variables that are all I (0) (Table 8A.1 in Appendix 8.1A). Therefore, this VAR, including the inflation rate, the output gap and changes in the nominal exchange rate, was used to generate the impulse response (Figure 8.2) and indicates that the output gap and the change in the exchange rate have both a positive impact on the inflation rate also reflecting the relevance of the passthrough effect. Second, in order to evaluate the relevance of the pass-through in the Mexican economy we can consider the ‘rolling’ correlation coefficient between the inflation rate and exchange rate depreciation. This correlation coefficient is estimated by adding one observation sequentially since 1989 (1). Figure 8.3 indicates the presence of a strong relationship between these two variables. The initial reduction of the pass-through, arguably related with the instrumentation of the inflation targeting regime, is not continuous and there is a persistence of a positive relationship between inflation and exchange rate movements.

Alternatives to inflation targeting in Mexico Response inflation rate to inflation rate

167

Response inflation rate to output gap

0.04

0.04

0.03

0.03

0.02

0.02

0.01

0.01

0.00

0.00

–0.01

–0.01 –0.02

–0.02 1

2

3

4

5

6

7

8

9

10

9

10

1

2

3

4

5

6

7

8

9

10

Response inflation rate to nominal exchange rate 0.04 0.03 0.02 0.01 0.00 –0.01 –0.02 1

2

3

4

5

6

7

8

Note: Period 1986:01–2003:04. The horizontal axis refers to the ten periods or quarters while the vertical axis shows the response of the inflation rate to a one standard deviation shock in each variable in the system.

Figure 8.2

8.3.3

Impulse – response analysis for the inflation rate, the output gap and changes in the nominal exchange rate

The Response of Monetary Policy to Exchange Rate Shocks

In this section we look for asymmetric effects of the exchange rate on monetary policy. The hypothesis is that monetary policy shows an asymmetric response to movements in the exchange rate. That is, the central bank raises the interest rate in response to a depreciating exchange rate but does not modify the interest rate in response to an appreciating exchange rate. The final result of this monetary policy is an appreciation of the real exchange rate. Table 8.4 shows the evolution of the ‘corto’. This table shows the dates of the changes and the quantities (with negative sign) of the monetary base to which the penalty rate is applied. The dates without changes in the ‘corto’ are not reported in Table 8.4. Over the period between January 1996 and December 2004, the Banco de México has increased the ‘corto’

168

Beyond inflation targeting

0.72

Correlation coefficient

0.70 0.68 0.66 0.64 0.62 0.60 0.58 1990

Figure 8.3

1992

1994

1996

1998

2000

2002

Rolling correlation coefficient between the inflation rate and exchange rate depreciation, quarterly data 1989: 1–2003:4

on 34 occasions and reduced it six times. Never over the period has the banking system been put in a ‘largo’. Increases in the ‘corto’ were preceded by or were simultaneous to depreciations of the peso on 18 occasions (53 percent of the number of increases) and the increase in the corto was followed by appreciations of the peso on 14 occasions. Only on four occasions (August and November 1996 and May and July 2001) were appreciations followed by a reduction of the ‘corto’ and only on one occasion did the central bank allow a sustained depreciation of the peso to take place without increasing the ‘corto’ (from June through December of 2003). We also explore the possibility of an asymmetrical response of monetary policy to exchange rate movements with an econometric exercise. In order to assess the relevance of this asymmetric effect we use a two-step procedure similar to Cover (1992), Karras (1996) and Kim, Ni and Ratti (1998). This procedure contains initially an equation in order to obtain an equilibrium exchange rate trough the use of the purchasing power parity (PPP) hypothesis (Hallwood and MacDonald, 2000). We consider that a value above the equilibrium position is an undervaluation while a value under the equilibrium position is an overvaluation. These values are, afterwards, used to test for a possible asymmetric response of monetary policy to movements in the exchange rate.

Alternatives to inflation targeting in Mexico

Table 8.4

169

Changes in the corto, 1996–2004

Date of change 1996 23 January 25 January 7 June 21 June 5 August 19 August 14 October 26 November 1998 11 March 25 June 10 August 17 August 10 September 30 November 1999 13 January 2000 January May June July October November

Corto (millions of pesos) −5 −20 −30 −40 −30 0 −20 0 −20 −30 −50 −70 −100 −130 −160 −180 −200 −230 −280 −310 −350

Date of change 2001 12 January 18 May 31 July 2002 8 February 12 April End of September 6 December 2003 10 January 7 February 28 March 2004 20 February 12 March 27 April 23 July 27 August 24 September 22 October 26 November 10 December

Corto (millions of pesos) −400 −350 −300 −360 −300 −400 −475 −550 −625 −700 −29 (daily) −33 −37 −41 −45 −51 −57 −63 −69

Source: Banco de México, www.banxico.org.mx. Informe Anual and Informe sobre Politica Monetaria, various issues.

Hence, Equation (8.2) describes the exchange rate using the purchasing power parity condition (ibid.): St 5 b0 1 b1Pt /P*t 1 ut

(8.2)

where St denotes the nominal exchange rate, Pt denotes the national price index and P*t denotes the foreign price index. The database includes quarterly information from 1995(1) to 2004(4) on the nominal exchange rate and the price indexes of Mexico and the USA. The estimation of Equation (8.2) using ordinary least squares (OLS)

170

Beyond inflation targeting

indicates that the nominal exchange rate has a long-term relationship with the price index differentials. The estimated long-run equation is as follows: St 5 3.36 1 1.202 * Pt /P*t

(8.3)

ADF(1) 5 23.07 Then using Equation (8.3) we define two shock functions based on the differences between the fitted values and the actual values of the exchange rate (U). Pos U1 t 5 max (shock, zero) Neg U2 t 5 min (shock, zero) The first function is associated with an undervaluation and the second one refers to an overvaluation. The second step of the procedure consists in estimating an interest rate equation including the positive and negative shocks given by the previous functions. 2 Rt 5 b0 1 b1U1 t 1 b2Ut 1 b3Rt21 1 et

(8.4)

Equation (8.4) describes the impact of exchange rate deviations on the interest rate. This equation is estimated to find evidence of asymmetric effects of the exchange rate on the interest rate. The interest rate is the three months nominal interest rate on government bonds (CETES) (Mexico’s central bank database). Making use of such a procedure we get the following estimates: 1 Rt 5 20.087*U2 t21 1 0.145*Ut21 1 0.76*Rt21 2 23.6*D98 (3) 220.7*D95 (4)

(t) ( p)

(21.01) (26.72) (0.26) (0.00)

(3.16)

(25.70)

(27.95)

(0.00)

(0.00)

(0.00)

R2 5 0.82 Normality test:6 Jarque-Bera: X2(2) 5 2.11(0.35) Autocorrelation: LM(4) 5 0.169(0.952) Heteroskedasticity: ARCH(4) 5 1.211(0.326) Period: 1995:1–2004:4

Alternatives to inflation targeting in Mexico

171

The long-run solution is: 1 Rt 5 20.371*U2 t21 1 0.615*Ut21

(8.5)

The main results indicate that only in the case of an undervalued exchange rate do we find a statistically significant coefficient. According to such estimates a tight monetary policy is carried out by the central bank whenever there is an undervalued exchange rate; however an overvalued exchange rate is not followed by an easy money monetary policy. The previous estimates thus confirm that the exchange rate has an asymmetric impact on monetary policy.

8.4

PROBLEMS WITH INFLATION TARGETING AND ALTERNATIVES

As discussed in Section 8.1, the recent past has been characterized by success on the inflation front and a poor growth performance. Two of our findings in Section 8.2, the positive effect of the real exchange rate on output and the significant pass-through of the exchange rate movements on prices, suggest that those two aspects, success on the inflation front and poor growth performance, are linked through the evolution of the real exchange rate (see Table 8.1). After the sharp depreciation of 1995, the real exchange rate has been appreciating over time, a trend that was only interrupted in 2002–03. From the end of 1995 to 2002, the real exchange rate fell by 35 percent (and by 27 percent from 1995 to 2004). With a relatively stable nominal exchange rate, in an increasingly open economy, government policy has provided a strong disinflationary pressure. At the same time, real appreciation, through the loss of competitiveness of the economy, has contributed to a poor growth performance. Our third finding in Section 8.2, on the asymmetric response of monetary policy to exchange rate shocks, indicates that real appreciation may be the result of a built-in bias in monetary policy towards real exchange rate appreciation. This bias has to do with the high pass-through. Over the past few years the central bank has been trying to break the link between exchange rate and prices by increasing the ‘short’ in response to ‘sharp’ depreciations. In doing so, it has reversed the depreciation itself. Because the economy has been in a process of disinflation in which the central bank has barely met its inflation targets, the process is not symmetrical, that is, there is no similar incentive to reverse the appreciations that may take place as a result of shocks to the exchange rate.

172

8.4.1

Beyond inflation targeting

Alternatives

Inflation targeting is certainly a more flexible framework for monetary policy than the previous monetary frameworks such as the use of some monetary aggregate to control the inflation rate. That is, inflation targeting considers additional factors in order to control inflation such as the exchange rate. Additionally, the hypotheses that prices and the monetary aggregate have a stable relationship and that money can be considered as the exogenous variable do not appear to be valid. For example, Carstens and Werner (1999) find that base money is essentially accommodating to exogenous shocks in the Mexican case. Therefore, inflation targeting represents a new framework with could include additional possible options. A first option is to move monetary policy towards a more neutral stance (that is, a more symmetric response to exchange rate shocks). In this sense, monetary authorities should try, at least, to use a reduction in the amount of ‘corto’ in the case of an appreciation of the real exchange rate. This is more likely and feasible than in the past as inflation tends to converge towards the long-run target and, as a result of either an enhanced credibility of the central bank or a less inflationary environment, the pass-through of the exchange rate on inflation tends to fall. In this case, monetary policy has more degrees of freedom because the authorities have established a reputation against inflation. In this context, a neutral monetary policy does not imply that the central bank will lose credibility in its commitment to control inflation. A neutral monetary policy will have at least two main impacts. First, it will allow a monetary policy with less bias against economic growth. That is, if the side effect of a contractionary monetary policy is a reduction, at least in the short run, of economic growth, a neutral monetary policy will contribute, in the margin, to a more dynamic economic growth. Second, a neutral monetary policy will not contribute to an overvaluation of the real exchange rate and, therefore, will not have an additional bias against economic growth through the real exchange rate channel. A second option is to shift from a CPI target to a domestic inflation target (that is, a measure of inflation purged from the direct effects of the exchange rate on imported goods in the CPI). Such a move would further reduce the pass-through effect of the exchange rate on the targeted price index and contribute to eliminate the asymmetric response of monetary policy to exchange rate shocks. In this case, a transitory shock on the nominal exchange rate will not generate a direct response in the interest rate. Under these circumstances monetary policy will be more focused on the long-term path of the inflation rate and therefore it will not overreact to transitory shocks.

Alternatives to inflation targeting in Mexico

173

These two options preserve the inflation targeting framework. A third, more radical departure from the current framework, would be to combine inflation targeting with real exchange rate targeting. More precisely, the central bank would promote a competitive exchange rate by establishing a sliding floor to the exchange rate in order to prevent excessive appreciation (an ‘asymmetric band’ with a floor and no ceiling as in Ros 1995). This would imply intervening in the foreign exchange market at times when the exchange rate hits the floor but allow the exchange rate to float freely otherwise. Such a proposal is free from some of the orthodox objections that have been made to real exchange rate targeting. In particular, it does not require knowledge of the adequate or equilibrium real exchange rate but only of the danger zone in which overvaluation severely hurts the growth process. This is because under our proposal the central bank does not target a particular real exchange rate but only establishes a floor on its value, leaving the real exchange rate to move freely above this threshold. Moreover, there is no problem with the amount of reserves required to defend the exchange rate since the central bank only defends the floor (which requires to accumulate rather than deplete reserves as would be the case if the central bank defended a ceiling). Of course, there is an additional orthodox objection when it comes to defending a floor to the exchange rate and this is that the central bank may lose control of the money supply and this could imply giving up the achievement of the inflation target (for a fuller discussion, see Frenkel and Rapetti, 2004). The problem arises at times of excess supply of foreign currency as a result, in particular, of massive capital inflows. It is worth noting, however, that speculative capital inflows will tend to be deterred to the extent that the central bank clearly signals that it will prevent the appreciation of the domestic currency, thus stabilizing exchange rate expectations. If necessary, however, the central bank can impose capital account regulations on short-term capital flows in order to recover control over the money supply.

8.5

CONCLUSIONS

The main conclusions that emerge from our analysis can be summarized as follows. Inflation has declined, in part during the inflation targeting regime, from levels over 50 percent in 1995 to around 5 percent in 2004. There has also been a reduction of the pass-through of exchange rate movements into inflation possibly as a result of an enhanced central bank credibility (with its stabilizing influence on inflation expectations) or of lower inflation itself. These achievements have, however, come with a cost,

174

Beyond inflation targeting

an almost continuous appreciation of the real exchange rate that has had contractionary effects on output in the long run. Our empirical analysis clearly supports these assertions (the reduction of the pass-through and the negative effect of real appreciation on output). It also suggests that real appreciation has been fed by an asymmetrical response of monetary policy to exchange rate movements: depreciations are followed by a tightening of monetary policy while appreciations are not reversed by a relaxation of monetary conditions. The alternative to the current framework is to give a more prominent place to the achievement of a competitive and stable real exchange rate in the design of monetary and exchange rate policy.

NOTES 1. We are indebted for comments to two anonymous referees, Jerry Epstein, Roberto Frenkel, Lance Taylor, Erinc Yeldan and other participants at the Amherst/CEDES Conference on Inflation Targeting, Buenos Aires, 13-14 May 2005. The usual caveat applies. 2. The literature on the contractionary effects of devaluation is very large and includes, among others, Diaz-Alejandro (1963), Cooper (1971), Krugman and Taylor (1978), Kamin (1988), Edwards (1989), Edwards and Montiel (1989), Morley (1992) and Razmi (2006). On the empirical evidence for Mexico, see Kamin and Rogers (1997), López and Guerrero (1998) and Kamin and Klau (1998). 3. The real exchange rate is defined as: SRt 5 St(Pus/P)t, where S is the nominal exchange rate; Pus is the consumer price index of the USA; and P is the consumer price index of Mexico. 4. The time span of the data is certainly not enough to consider the results of the cointegrating vector as a long-run solution. Therefore, it is not possible to make long-term inferences on the basis of these results but it represents a valid approximation for the period. 5. The cointegration tests without any dummies are reported in Appendix 8.1A in Table 8A.2. 6. The non-normality can be attributed to an outlying value in 1995:2 when debt and financial crisis took place in Mexico.

REFERENCES Ball, L. and N. Sheridan (2003), ‘Does inflation targeting matter?’, National Bureau for Economic Research working paper no. 9577, March. Banco de México (1996), Informe Anual, México: Banco de México. Calvo, G.A. and F.S. Mishkin (2003), ‘The mirage of exchange rate regimes for emerging market countries’, Journal of Economic Perspectives, 17 (4), 98–118. Carstens, A. and A. Werner (1999), ‘Mexico’s monetary policy framework under a floating exchange rate regime’, Banco de México, Documento de Investigación No. 90-05, May. Clifton, E.V., H. Leon and C.H. Wong (2001), ‘Inflation targeting and the unemployment-inflation trade off’, International Monetary Fund IMF working paper WP/01/166.

Alternatives to inflation targeting in Mexico

175

Cooper, Richard N. (1971), ‘Currency devaluation in developing countries,’ in G. Ranis (ed.), Government and Economic Development, New Haven, CT: Yale University Press, pp. 472–513. Corbo, V., O. Landerretche and K. Schmidt-Hebbel (2002), ‘Does inflation targeting make a difference?’, in N. Loayza and R. Soto (eds), Inflation Targeting: Design, Performance, Challenges, Santiago: Banco Central de Chile, document 106, pp. 1–65. Cover, J.P. (1992), ‘Asymmetric effects of positive and negative money-supply shocks’, Quarterly Journal of Economics, 107 (4): 1261–82. Diaz-Alejandro, Carlos F. (1963), ‘A note on the impact of devaluation and the redistributative effects’, Journal of Political Economy, 71, 577–80. Dickey, D. and W.A. Fuller (1981), ‘Likelihood ratio statistics for autoregressive time series with unit root’, Econometrica, 49 (4), 1057–977. Edwards, Sebastian (1989), Real Exchange Rates, Devaluation, and Adjustment, Cambridge, MA: MIT Press. Edwards, Sebastian and Peter J. Montiel (1989), ‘Devaluation crises and macroeconomic consequences of postponed adjustment in developing countries’, IMF Staff Papers, 36, 875–903. Fraga, A., I. Goldfajn and A. Minella (2003), ‘Inflation targeting in emerging market economics’, National Bureau for Economic Research working paper no. 100019, pp. 1–50. Frenkel, R. and M. Rapetti (2004), ‘Políticas macroeconómicas para el crecimiento y el empleo’, Paper prepared for the International Labor Organization. Goldfajn I. and P. Gupta (2003), ‘Does monetary policy stabilize the exchange rate following a currency crisis?’, IMF Staff Papers, 50 (1), 90–114. Hallwood C. and R. MacDonald (2000), International Money and Finance, Malden, MA: Blackwell Publishers. Johansen, S. (1988), ‘Statistical analysis of cointegrating vectors’, Journal of Economic Dynamics and Control, 12 (2–3), 231–54. Johansen, S., R. Mosconi and B. Nielsen (2000), ‘Cointegration analysis in the presence of structural breaks in deterministic trends’, Econometrics Journal, 3, 216–49. Kamin, S.B. (1988), ‘Devaluation, external balance, and macroeconomic performance in developing countries: a look at the numbers’, Princeton Essays in International Finance no. 62. Kamin, S.B. and J.H. Rogers (1997), ‘Output and the real exchange rate in developing countries: an application to Mexico’, Board of Governors of the Federal Reserve Bank international finance discussion paper no. 580, May. Kamin, S.B. and M. Klau (1998), ‘Some multicountry evidence on the effects of real exchange rates on output’, Board of Governors of the Federal Reserve Bank international finance discussion paper no. 611, May. Karras, G. (1996), ‘Are the output effects of monetary policy asymmetric? Evidence from a sample of European countries’, Oxford Bulletin of Economics and Statistics, 58 (2), 267–78. Kim, J., S. Ni and R. Ratti (1998), ‘Monetary policy and asymmetric response in default risk’, Economics Letters, 60 (1), 83–90. Krugman, P. and L. Taylor (1978), ‘Contractionary effects of devaluation’, Journal of International Economics, 8 (3), 445–56. Kwiatkowski, D., P.C.B. Phillips, P. Schmidt and Y. Shin (1992), ‘Testing the null hypothesis of stationary against the alternative of a unit root’, Journal of Econometrics, 54 (1–3), 159–78.

176

Beyond inflation targeting

López, J. and C. Guerrero (1998), ‘Crisis externa y competitividad de la economía mexicana’, El Trimestre Económico, October–December, LXV (260), 582–98. Lütkepohl, H. and H.E. Reimers (1992), ‘Impulse response analysis of co-integrated systems’, Journal of Economic Dynamics and Control, 16, 53–78. Lustig N. and J. Ros (1998), ‘Reforma estructural, estabilización económica y el síndrome mexicano’, Desarrollo Económico, 37 (148), 503–31. Maddala, G.S. and I. Kim (1998), Unit Roots, Cointegration and Structural Change, Cambridge: Cambridge University Press. Morley, Samuel A. (1992), ‘On the effect of devaluation during stabilization programs in LDCs’, Review of Economics and Statistics, LXXIV, 21–7. Neumann, M.J.M. and J. von Hagen (2002), ‘Does inflation targeting matter?’, Federal Reserve Bank of St Louis Review, (July–August). OECD (2002), Economic Surveys, Mexico City: OECD. OECD (2004), Economic Surveys, Mexico City: OECD. Phillips, P.C.P. and P. Perron (1988), ‘Testing for unit root in the time series regression’, Biometrica, 75 (2), 335–46. Razmi Arslan (2006), ‘The contractionary short-run effects of nominal devaluation in developing countries: some neglected nuances’, University of Massachusetts Amherst, Department of Economics, working paper 2005-09. Ros, J. (1995), ‘Después de la crisis: la política económica’, Nexos, accessed October 1996 at www.nexos.com.mx/articulosEspeciales.php. Ros, J. (2001), ‘Del auge de capitales a la crisis financiera y más allá: México en los noventa’, in Ricardo Ffrench-Davis (ed.), Crisis financieras en países ‘exitosos’, Santiago: CEPAL-McGraw Hill, pp. 119–57. Schmidt-Hebbel, K. and A. Werner (2002), ‘Inflation targeting in Brazil, Chile and Mexico: performance, credibility, and the exchange rate’, Central Bank of Chile working paper no. 171, July. Sheridan, N. (2001), ‘Inflation dynamics’, PhD dissertation, Johns Hopkins University. Svensson, L.E. (1997), ‘Inflation forecast targeting: implementing and monitoring inflation targets’, European Economic Review, 41 (6), 1111–46. Svensson, L.E. (1998), ‘Open economy inflation targeting’, National Bureau for Economic Research working paper no. 6545, May.

Alternatives to inflation targeting in Mexico

177

APPENDIX 8.1A Table 8A.1

Unit root tests

Variable

yt Δyt yust Δyust invt Δinvt srt Δsrt pt Δpt ΔΔpt

p*t Dp*t rt Δrt

r*t Dr*t st Δst gapt Δgapt

ADF

PP(4)

KPSS(10)

A

B

C

A

B

C

hm

ht

−3.19(8) −3.91(8) −2.68(2) −3.45(8) −3.69(4) −4.11(8) −4.05(4) −10.8(0) −1.86 (3) −4.01 (4) −9.09 (1) −1.39 (3) −7.64 (0) −2.64 (3) −5.28 (2) −2.62 (3) −2.67 (8) −2.10 (3) −4.18 (2) −3.99 (8) −4.75 (8)

0.54(8) −3.64(8) −0.64(2) −4.04(2) −0.97(5) −3.88(8) −3.26(4) −10.90(0) −3.50 (3) −2.01 (2) −9.08 (1) −1.84 (3) −3.66 (2) −0.96 (3) −5.14 (2) −1.24 (3) −2.49 (8) −3.35 (3) −3.17 (2) −4.00 (8) −4.79 (8)

1.99(8) −2.54(8) 4.27(2) −2.67(1) 0.34(2) −3.83(8) −0.02(0) −10.96(0) −0.55 (3) −1.30 (2) −9.14 (1) 2.55 (3) −2.63 (2) −0.82 (3) −5.12 (2) −1.82 (3) −2.17 (8) −1.71 (4) −2.42 (2) −4.05 (8) −4.81 (8)

−4.85 −21.52 −2.69 −7.57 −2.44 −8.82 −2.77 −10.81 −0.37 −3.59 −9.80 −3.48 −7.75 −2.70 −9.82 −1.66 −10.63 −1.02 −7.73 −8.67 −23.38

−0.33 −21.54 0.12 −7.63 −1.12 −8.81 −2.49 −10.82 −3.91 −2.60 −9.85 −4.60 −6.50 −1.10 −9.69 −0.32 −10.44 −3.21 −6.89 −8.71 −23.42

2.32 −18.91 6.98 −4.51 0.51 −8.82 −0.03 −10.87 0.99 −1.79 −9.91 7.87 −3.42 −0.72 −9.72 −1.71 −10.31 −1.62 −5.57 −8.75 −23.54

0.946 0.096 0.967 0.072 0.663 0.156 0.262 0.108 0.903 0.613 0.068 0.981 0.627 0.615 0.236 0.769 0.140 0.868 0.512 0.043 0.052

0.171 0.047 0.065 0.065 0.148 0.066 0.101 0.071 0.238 0.074 0.063 0.242 0.087 0.097 0.077 0.076 0.101 0.229 0.068 0.043 0.048

Notes: Test statistics in bold indicate a rejection of the null hypothesis. Critical values at 5 percent significance level for the Augmented Dickey-Fuller and Phillips-Perron tests for a size T5100 are −3.45 including constant and trend (model A), −2.89 including constant (model B) and –1.95 without constant and trend (model C) (Maddala and Kim, 1998, p. 64). hμ and ht are the KPSS tests for the null hypothesis of stationarity around a level and deterministic linear trend, respectively. Both tests are calculated with a lag window size equal to ten. The 5 percent critical values for the two tests are 0.463 and 0.146, respectively (Kwiatkowski et al. 1992, p. 166). Small letters represent the values in logarithms. Sources:

Dickey and Fuller (1981) and Phillips and Perron (1988).

Description of Variables y 5 Real Gross Domestic Product (GDP) in millions of Mexican pesos of 1993, Instituto Nacional de Geografía, Estadística e Informática (INEGI), http://www.inegi.gob.mx. yus 5 Real Gross Domestic Product (GDP) in billions of US dollars at

178

Table 8A.2

Beyond inflation targeting

Mis-specifications tests of the Vector Autoregressive Model: st, (pt 2 p*t ) and (rt 2 r*t )

Variable

Autocorrelation: LM(4)

Heteroskedasticity ARCH(4)

NormalityJ-B

st pt 2 p*t rt 2 r*t

F(3,14) 5 0.40 [0.75] F(3,14) 5 1.59 [0.24] F(3,14) 5 2.05 [0.15]

F(3,11) 5 0.095 [0.96] F(3,11) 5 0.31 [0.82] F(3,11) 5 0.09 [0.97]

c2(2) 5 8.39[0.02]* c2(2) 5 3.57[0.17] c2(2) 5 9.38[0.01]**

Note: *, ** indicate a rejection of the null hypothesis to 5 percent and 1 percent significance level. Period 1995(1)–2004(4).

prices of 2000, (the series is already seasonally adjusted), US Department of Commerce: Bureau of Economic Analysis. p 5 Mexican consumer price index (base 2002 5 100), Bank of Mexico, http://www.banxico.org.mx. p* 5 US consumer price index (base 1982–84 5 100), US Department of Commerce, Bureau of Economic Analysis. GAPY 5 Deviation of the real Gross Domestic Product (GDP) from potential output obtained using the Hodrick-Prescott filter. S 5 Exchange rate (pesos per US dollar), 48-hour interbank exchange rate. The data is taken from the last day of each quarter, Bank of Mexico, http://www.banxico.org.mx. Sr 5 Real exchange rate defined as S (P*/P) where S is the nominal exchange rate, P refers to domestic prices and P* to foreign prices. r 5 Nominal interest rate on three-month treasury bills (CETES 91 days). Average of the last month of the quarter, Bank of Mexico, http://www. banxico.org.mx. r* 5 Nominal interest rate three-month treasury bill, Board of Governors of the Federal Reserve System

9.

Five years of competitive and stable real exchange rate in Argentina, 2002–07 Roberto Frenkel and Martín Rapetti1

9.1

INTRODUCTION

In 1991 Argentine authorities established the convertibility regime, which implied the pegging of the peso (AR$) to the US dollar ($) by law and the validation of contracts in foreign currencies. The new monetary arrangement also stipulated that the central bank must fully back the monetary base with foreign reserves,2 what in practice turned the central bank into a currency board. The convertibility regime was the pillar of a broader stabilization program, intended to take the economy away from the high inflation regime settled since the mid 1970s, which had led to two brief hyperinflationary episodes in 1989 and 1990. The program also included an almost complete liberalization of trade flows and the full deregulation of the capital account of the balance of payments. It was jointly applied with an impressive process of market-friendly reforms, targeting the privatization of a large proportion of state-owned firms. The program successfully managed to stop inflation and initially spurred rapid growth. However, as happened with many other stabilization programs in the region, it led to the appreciation of the real exchange rate, which made economic growth highly dependent on external debt accumulation.3 Since the Asian and Russian crises, and especially after the Brazilian devaluation in 1999, the deceleration of capital inflows put the economy into a deflationary trend that ended up in a financial and external crisis in 2001–02. Between the last days of 2001 and the beginning of 2002, Argentina declared the default of its international debt and devalued the peso. The collapse of the convertibility regime implied a 21 percent contraction in GDP with respect to the peak of mid-1998 and a rise in the unemployment rate up to 21.5 percent, taking half of the population below the poverty line. However, only one quarter after the devaluation and default economic 179

180

Beyond inflation targeting

activity gradually started to recover. By the end of 2002, once the government managed to stabilize domestic financial markets, the recovery gained momentum and since then the economy has shown an impressive performance. In five years – from the first quarter of 2002 to the same period in 2007 – GDP has been growing at 8.1 percent annual rate, reaching a peak 18.1 percent higher than the one in mid-1998. The investment rate rose to 22 percent of GDP (on seasonally adjusted basis), which is the maximum range of the official time series beginning in 1993 and continues to grow at a higher pace than GDP. During these five years exports have expanded at a slightly higher rate than GDP, but its rate of growth has substantially increased from mid 2004 onwards. Current economic evolution contrasts with Argentina’s economic performance of the last 60 years. Since the Second World War economic growth has been low and very volatile, especially in the second financial globalization period beginning in the mid 1970s. For the first time in 30 years Argentina has grown five years in a row. More importantly, current expansion is based on solid macroeconomic fundamentals. The volatility of Argentine growth has been typically associated with current account and fiscal deficits. Between the mid 1940s and the mid 1970s, macroeconomic evolution was characterized by stop-and-go cycles related to external imbalances. During the second financial globalization period, the availability of external funds momentarily relaxed the external constraint to growth, but it led to two episodes of explosive fiscal and external debt accumulation,4 one in the late 1970s and the beginning of the 1980s and the other during the convertibility regime period. In contrast to those traditional fiscal and external imbalances, the current macroeconomic configuration stands out with the existence of external and fiscal surpluses. Certainly, the debt restructuring in 2005 – implying a $67 billion reduction in the nominal stock – softened both external and fiscal requirements, releasing resources for private sector spending. Similarly, favorable external conditions – especially the high prices of some commodities – have also played a role. However, in our view the main factor behind the current success is the official policy aiming at preserving a stable and competitive real exchange rate (SCRER). The SCRER has been a key factor explaining the current account adjustment, which passed from a $14.5 billion deficit in 1998 to $7.6 billion surplus in 2006. From this $22 billion adjustment, $20 billion came from the improvement in the trade balance, which is mainly attributable to the effects of the exchange rate depreciation. The influence of the SCRER on the fiscal accounts performance has also been important. After devaluation, the government introduced taxes on traditional exports, mainly agricultural products and oil. In practice,

Five years of competitive and stable real exchange rate in Argentina

181

this measure implied the introduction of multiple exchange rates that contributed to reduce the pass-through of devaluation to wage-goods prices, but also to capture part of the rent that these sectors obtained from the competitive real exchange rate. In 2006 the federal administration recorded a primary surplus of 3.5 percent of GDP and a total surplus of 1.8 percent of GDP, from which taxes on exports accounted for 63 percent of the former and 122 percent of the latter. In our view, the positive effects of the SCRER policy are the principal factors explaining the rapid growth experienced so far. This policy promotes economic growth not only by preserving external and fiscal accounts sustainability, but also by providing incentives to the tradable sector and thus encouraging the expansion of its production, employment and investment. Although the success of the SCRER strategy throughout these five years has undoubtedly had a persuasive impact among analysts, skepticism remains.5 The SCRER policy collides with conventional wisdom, particularly with the trilemma paradigm. In this chapter, we argue that a macroeconomic regime based on a SCRER is both desirable and manageable for a developing open economy. The next section describes the evolution of monetary and exchange rate policies in Argentina in the post-convertibility period. Section 9.3 discusses the usual criticisms against the SCRER policy and shows the conditions in which this policy is sustainable. Section 9.4 presents some concluding remarks regarding the management of a macroeconomic regime with a SCRER as an intermediate target.

9.2

MONETARY AND EXCHANGE RATE POLICIES IN THE POST-CONVERTIBILITY PERIOD

The deceleration of capital inflows that led to the 2001–02 crisis began in mid 1998. This process took place simultaneously with a persistent rise of the sovereign risk premium. However, the divergent trends in the domestic financial market that triggered the collapse of the convertibility regime only started in October 2000, associated with the political turmoil caused by the Vice-President’s resignation. The process followed a simple dynamics. Devaluation expectations and the perception of a higher risk of default led the private sector to withdraw deposits and run against the central bank’s international reserves. There were no bankruptcy reports of failing banks because the central bank supported the liquidity of the banking system. Despite several signals issued by the government aiming at changing the expectations, the intensification of this process could not be stopped. In December 2001 restrictions on

182

Beyond inflation targeting

capital outflows and on the withdrawal of deposits (the so-called ‘corralito’) were established. After the abandonment of the convertibility regime, the government aimed to restrain the capital outflow and stabilize the foreign exchange (FX) market by introducing a dual exchange rate regime. The idea was to use this scheme only temporarily, in order to stabilize the nominal exchange rate while the domestic prices absorbed the impact of the devaluation, and then pass to a floating rate regime. The authorities also decided to convert to pesos most of the domestic debts contracted in dollars (bank credits, rents and so on) at a AR$/$ 1 rate (plus indexation to consumer price index (CPI) inflation), thus neutralizing most of the effects of relative price change on the debtors’ balance sheets. In contrast, banks’ deposits originally denominated in dollars were ‘pesoificated’ at a AR$/$ 1.40 rate (plus indexation to the evolution of CPI inflation).6 Together with the ‘pesoification’, the authorities unilaterally decided to extend the maturity and duration of all deposits, including those originally contracted in pesos. In exchange, private depositors received certificates for the reprogrammed deposits. In February 2002 the FX market was unified and the peso started to float freely. Given the political and economic uncertainty, the exchange rate skyrocketed fed by self-fulfilling expectations. Interestingly, this process developed in an illiquid environment because of the restrictions on the withdrawal of cash from banks. The erratic monetary policy that followed in the first quarter of 2002 also failed to stabilize the exchange rate. The authorities delayed the launching of a domestic asset that could perform as a potential substitute for foreign currency. Given the distrust in banks and in the Treasury, the economic depression and the growing inflation, the international currency appeared as the only asset available to allocate financial savings. Only two and half months after the devaluation the central bank started to issue notes (the Lebac) in order to supply a financial instrument that could compete with the dollar. All these factors contributed to deepen the perverse dynamics of the financial variables during the first semester of 2002. The capital flight from domestic assets between March 2001 and mid 2002 is illustrated in Figure 9.1, which shows the large fall in private bank deposits7 and international reserves, while the nominal demand for cash remains stagnant. These developments provide evidence for the substitution of local assets (cash and deposits) in exchange for external assets (international reserves). The result of the asset substitution affected the FX market. The nominal exchange rate (NER) and real exchange rate (RER)8 rose continuously in the first semester of 2002 (around 260 percent and 180 percent, respectively). Their paths are shown in Figure 9.2. Real exchange rate

Five years of competitive and stable real exchange rate in Argentina 50 000

183 100 000

Exchange rate stabilization

Pesoification

90 000

40 000 80 000 30 000

70 000 60 000

20 000

50 000 10 000

Demand for cash Central bank reserves Private bank deposits Lebac

0

30 000 2001

Source:

40 000

2002

2003

2004

2005

Central Bank of Argentina.

Figure 9.1

Demand for cash, central bank international reserves, Lebac and private bank deposits (right axis) (in millions of pesos and dollars)

overshooting was so pronounced that in June 2002 its value was almost 193 percent higher than the 1980–2001 period average value, and 309 percent higher than the convertibility decade average. These disruptive trends began to revert in July 2002. The turning point was the exchange rate stabilization. Several factors contributed to this outcome. Controls on FX transactions9 had been introduced in November 2001 – before the convertibility collapse – and they were further tightened in March 2002. Since June 2002 controls and interventions in the FX market were strengthened in order to conduct a systematic policy to stabilize the exchange rate. The decision that export revenues surpassing $1 million had to be sold directly to the central bank was especially important in this regard. This became the main source of international reserves accumulation for the monetary authority, which in turn agreed to increase the volume of its interventions in the FX market. Financial market behavior itself also contributed to stop the bubble in the exchange rate. On the one hand, local interest rates skyrocketed (Figure 9.3). In July 2002 the average time deposits annual interest rate reached a

184

Beyond inflation targeting

4.0 3.5 3.0 2.5 2.0 1.5 MRER NER RER

1.0 0.5 2001

2002

2003

2004

2005

2006

Notes: * Calculated using US and Argentina consumer price indexes. ** Elaborated using monthly data of the exchange rates of the 18 principal trading partners of Argentina, adjusted by the respective consumer price indexes. Source:

Central Bank of Argentina.

Figure 9.2

Bilateral nominal (NER) and real exchange rate (RER) with the US and multilateral real exchange rate (MRER) (in pesos and indexes 1 5 December 2001)

76 percent peak, and the annual interest rate of the 14-day Lebacs reached almost 115 percent. Thus local financial assets began to appear more attractive as substitutes for the dollar. On the other hand, as mentioned above, the real price of the dollar reached very high and ‘abnormal’ levels in historical terms (that is, the prices in dollars of domestic assets, nontradable goods and salaries were perceived as abnormally low). In this context, once the authorities managed to stop the exchange rate bubble in July, the public rapidly changed expectations and the market started to show an appreciation trend. Thus, in the second half of 2002 a phase of monetary and financial variables normalization started. After reaching a peak of almost AR$/$ 4 during the last days of June, the exchange rate began to experience a smooth nominal appreciation trend. Although the inflation rate was already low and decelerating, the rise in domestic prices contributed to the real appreciation. In that context, local assets became increasingly

Five years of competitive and stable real exchange rate in Argentina Exchange rate stabilization

120

14d Lebac 91d Lebac Time deposits 30–59d Prime 30d

100

Monthly avg

185

80 60 40 20 0 2001

Source:

2002

2003

2004

Central Bank of Argentina.

Figure 9.3

Interest rates in pesos: Lebac (14 and 91 days), time deposits (30 to 59 days) and prime (30 days) (monthly average, in %)

attractive. Bank deposits began to grow, as did the demand for Lebac, local shares and the demand for cash (Figure 9.1). This portfolio substitution in favor of local assets resulted in a persistent drop in the interest rates (Figure 9.3). The normalization in financial activity dissipated, disrupting expectations and thus favored the recovery of private expenditure. Interestingly, this recovery took place without significant contribution from bank credits. Even though private deposits gradually improved allowing the recuperation of banks’ liquidity, credit to the private sector continued shrinking until late 2003. The financial crisis appeared to have persistent effects on the behavior of bank credit, which at the beginning of 2007 was still below the peak reached in 1998 (Figure 9.4). Domestic expenditure was mainly financed by private sector cash holdings. Figure 9.5 shows the increase in cash holdings since the fourth quarter of 2001. Both the monetary base/GDP ratio and the monetary base/total bank deposit ratio showed very high rates of growth and also relatively high levels in comparison to the convertibility period. Although the low interest rates on banks’ deposits (and the tax on financial transactions) have contributed to that performance, this behavior seems to be another persistent consequence of the financial crisis.

186

Beyond inflation targeting

Bank credit/GDP in percentage

28

24

20

16

12

8 96

98

99

00

01

02

03

04

05

Central Bank of Argentina; Ministry of Economy of Argentina.

Figure 9.4

Bank credit to private sector in relation to GDP seasonally adjusted

50

12 MB/GDP MB/Deposits

45 Monetary base/total deposits

06

11 10

40

9 35 8 30 7 25

6

20

5

15

4 1994

Source:

Monetary base/GDP

Source:

97

1996

1998

2000

2002

2004

Central Bank of Argentina; Ministry of Economy of Argentina.

Figure 9.5

Monetary base in relation to total bank deposits and with GDP seasonally adjusted

Five years of competitive and stable real exchange rate in Argentina

187

The nominal and real appreciation process stopped around mid 2003, when the government decided to manage the flotation of the exchange rate in order to preserve the SCRER. The impact of the SCRER on economic activity, employment and external and fiscal accounts was proving to be highly favorable. Thus, the government gradually started to recognize and make explicit reference to the importance of preserving the SCRER in the official economic strategy. On the contrary, central bank authorities never made any explicit statement regarding the existence of any exchange rate target. According to their official statements and documents, the permanent intervention in the FX market has been oriented to accumulate international reserves for a precautionary purpose, namely to protect the economy from international capital markets volatility. Statements aside, the joint intervention of the central bank and the Treasury in the FX market actually controlled the price of the dollar in a narrow range between AR$2.8 and AR$3.1. The resulting fluctuation of the exchange rate in this range made the multilateral real exchange rate remain stable around a level 129 percent higher than the one at the end of the convertibility regime. The bilateral real exchange with the US dollar also remained stable for some years, but since early 2005 it has shown a soft appreciation trend (see Figure 9.2). In 2002, when the Congress passed a law revoking the currency board, the government decided to keep the central bank’s independence with the mandate of pursuing low inflation rates as its primary mission. Given that the economy was still absorbing the effects of the crisis and the devaluation and that the domestic financial markets had shrunk significantly, the central bank disregarded the option of following an inflation targeting regime. The transmission mechanisms through the interest rate on aggregate demand were thought to be uncertain and weak.10 Instead, the authorities opted to follow a more pragmatic policy based on broad quantitative monetary targets. From 2003 on, targets have been announced at the beginning of every year throughout the central bank monetary programs, in which the authorities commit themselves to maintain monetary aggregates within a certain range. Given the uncertainty surrounding the effects of monetary policy, the central bank has tended to set these ranges sufficiently broadly. However, their upper bounds ended up being systematically lower than the monetary expansion arising from the intervention in the FX market to preserve the SCRER. Thus, since 2003 the central bank has dealt with two ‘conflicting’ objectives: the preservation of a competitive exchange rate by intervening in the FX market and at the same time the attainment of the targets of monetary expansion announced in the monetary program. The tension between these two policy objectives can be observed in Table 9.1, which shows the sources of variation of the monetary base. In the first

188

–1 450 1 281 1 374 1 931 2 352 3 584

Central bank FX intervention

406 1 674 809 483 186 2 113

Monetary base variation –1 856 –393 565 1 447 2 166 1 471

‘Excess’ of monetary expansion –216 –270 –420 –323 –836 –812

Central bank sterilization 1 426 86 –125 –601 –939 –204

Assistance to banks

124 250 –52 –543 –353 –446

Assistance to the Treasury

522 327 32 19 –39 –8

Others

n.a. n.a. 28 112 343 40

Treasury FX interventiona

Sources of variation of the monetary base (monthly average variation, in millions of pesos or dollars)

Source: Central Bank of Argentina.

Notes: a. In millions of dollors. b. Calculated for the period February–June 2002. c. A cancellation of a Banco Nación’s rediscount by the Treasury with assistance of the central bank in September 2002 for about AR$3500 million is omitted.

2002:01b 2002:02c 2003 2004 2005 2006

Table 9.1

Five years of competitive and stable real exchange rate in Argentina

189

semester of 2002 the central bank intervened in the FX market providing dollars to contain the depreciation pressures. Thus, the FX intervention operated as a source of monetary base contraction. Once the exchange rate was stabilized, the accumulation of international reserves resulted, on the contrary, in a source of monetary expansion. During the second semester of 2002 this source of monetary expansion was easily absorbed by the rapid growth in the demand for cash caused by the re-monetization of the economy. However, since 2003 the gradual deceleration of money creation established in the monetary programs in order to maintain inflation expectation under control started to conflict with the increasing expansion of monetary base generated by central bank’s intervention in the FX market aiming to preserve the SCRER. Since the amount of monetary base created to intervene in the FX market (first column in Table 9.1) exceeded the actual expansion of the monetary base to accomplish the monetary targets (second column), an ‘excess’ of monetary expansion (third column) had to be absorbed. This ‘excess’ of monetary expansion has been absorbed through several mechanisms. Throughout 2003 the sterilization operations implemented by the issuing of central bank notes were especially relevant. The need for sterilization increased during 2004 and 2005. However, the central bank could limit the issuing of Lebac because other compensatory mechanisms began to operate. In the first place, as liquidity grew the banks started to service the debt incurred with the central bank during the financial crisis. Hence, banks’ capital payments and especially the payment of interests operated as a source of monetary base contraction. In 2005 the central bank launched a program allowing the acceleration of banks’ debts amortizations, reinforcing this contractionary mechanism. By early 2006 most banks had cancelled their debts with the monetary authority. The Treasury also helped to absorb the ‘excess’ of monetary expansion. While in 2002 a net flow of financing from the central bank to the Treasury was observed, in 2003, and especially since 2004, the transactions between the Treasury and the central bank operated as a source of contraction of the monetary base. The Treasury’s purchases of international reserves with the proceeds of the primary surplus gave place to a monthly average contraction of the monetary base of AR$543 million in 2004. The main purpose of these operations was to continue servicing of the debt with the multilateral financial institutions. The Treasury and other official agencies also accumulated part of the fiscal surplus in foreign currency and thus intervened directly in the FX market to alleviate central bank’s management of the ‘conflicting’ objectives. These operations started in late 2002 and gradually expanded afterwards, thus becoming an important policy instrument (see last column of Table 9.1).

190

Beyond inflation targeting

In 2005 the sterilization needs through the issuing of Lebac increased again. Thus, in order to soften the appreciation pressures in the FX market, controls on the capital account were introduced in June. Basically, the new measures established that all capital inflows – excluding the issuing of new private and public debt, international trade financing and foreign direct investment – would be subject to a 30 percent unremunerated reserve requirement for at least 365 days. This strategy is inspired by that applied in Chile in the early 1990s and attempts to reduce short-term capital inflows. However, controls left open ways to avoid the reserve requirements. For instance, capital inflows can easily circumvent the reserve requirement by operating through the stock exchange market (by buying domestic assets abroad and selling them in the local market). There has been no evidence of a reduction in the supply of dollars in the FX market after the measures were implemented. Local analysts believe that controls are ineffective and even the authorities do not reject the idea that they were introduced more as a signal of the official willingness to maintaining the SCRER strategy rather than as an effective control mechanism. As from 2006 monetary policy stopped targeting the monetary base and started to focus on M2. The authorities argued that the change in the target was due to the increasing monetization of the economy and the gradual recovery of bank credit. In these conditions, it was argued, the use of a larger monetary aggregate represented a step forward toward the fine-tuning of monetary policy. In practice, the switch of the monetary aggregate target helped to relax the conflicting management of exchange rate and monetary policies. The central bank was facing increasing difficulties to accomplish the monetary base targets. As Table 9.1 shows, the ‘excess’ of monetary expansion had risen substantially between 2003 and 2005. The use of M2 as a target gave the authorities greater flexibility to conduct the two-target policy, allowing for greater intervention in the FX market and expansion of the monetary base. In summary, during the post-convertibility period the central bank has been able to conduct the two-target policy successfully. Moreover, while doing so it has obtained quasi-fiscal surpluses every year. Some analysts have argued that the management of monetary policy focusing on two targets has had an inflationary bias. Certainly, inflation accelerated during 2004 and 2005 and has remained stable around an annual rate of 10 percent since 2006. In our view, the acceleration of inflation is due not to inconsistencies in the management of monetary and exchange rate policies, but to the lack of coordination between these and the fiscal policy. The expansion of public spending well above the increase of tax revenues since 2006 has implied an expansionary fiscal impulse to an already fastgrowing aggregate demand. Given the fact that monetary and exchange

Five years of competitive and stable real exchange rate in Argentina

191

rate policies focus on preserving a SCRER – which is intended to put the economy in a high growth path – fiscal policy is the only macroeconomic instrument available to moderate aggregate demand when inflationary pressures arise.

9.3

THE ECONOMICS OF THE SCRER

The notion that a SCRER favors economic development has a long tradition in economic theory. Advocates of the outward orientation approach to development during the 1960s and 1970s pointed to the SCRER as a key element for that strategy (Balassa, 1971; Díaz Alejandro, 1979). According to this view, a competitive real exchange rate boosts economic growth because it softens the balance of payment constraint and favors the development of tradable activities, which tend to be more dynamic. The stability of the exchange rate is also important because low volatility reduces the risk and uncertainty of investment in tradable sectors. These arguments have been recently revitalized by modern scholars.11 Besides the traditional effects, modern advocates also emphasize that a development strategy based on a SCRER is market friendly (avoiding rent-seeking practices) and compatible with free trade agreements. In recent years many studies have documented a statistically and economically positive relationship between growth and real exchange rates.12 The preservation of a SCRER has also been invoked for other reasons. Maintaining a competitive real exchange rate typically involves intervention in the FX market and the accumulation of international reserves. It is a well documented fact that international financial integration may lead to macroeconomic instability and increases the likelihood of external crises. Some scholars argue that international reserves accumulation serves as a shield against volatile capital flows, especially for developing countries (Feldstein, 1999). Empirical studies show a positive relationship between reserve accumulation and growth (Polterovich and Popov, 2002). Another less studied motive is that competitive real exchange rates promote job creation. Besides the above-mentioned growth effects, a SCRER may impact on employment through a more intense use of labor (Frenkel, 2004). A competitive parity favors labor-intensive activities and sectors and also the substitution of expensive inputs (such as imports) in favor of labor across sectors.13 Probably because of all these reasons, the preservation of a SCRER does not attract much criticism by itself. Few scholars deny the beneficial aspects of stable and predictable relative prices and the positive effects on growth. In some cases welfare arguments against public intervention in

192

Beyond inflation targeting

the FX market are raised. But the optimality of the free market determination of the exchange rate and the argument that the public sector has no informational advantage over the private sector are not very appealing ideas in the specialized discussion about exchange rate regimes and policies. The apparent volatility of capital flows and the instability and unpredictability of free-floating exchange rates greatly lessen the relevance of those ideas (Frankel and Rose, 1995). Moreover, in some scholars’ view the free-floating exchange rate indeterminacy and unpredictability is precisely the deeper foundation of the need for managing the exchange rate (Blecker, 2005). This is particularly emphasized in countries in which the real exchange rate plays a crucial role in the economic performance. Skepticism towards the SCRER policy points to the ability of governments to conduct it. The main objection is that the real exchange rate – as any real variable – is not under the government’s control, at least in the long run. However, given the weak empirical support of real exchange rate determination models,14 the objections relevant for economic policy formulations are based on the trilemma or impossible trinity argument. The trilemma says that it is impossible for a country to simultaneously maintain free capital mobility, active monetary policy and the ability to manage the exchange rate. One of these features must necessarily be given up. In other words, the trilemma says that in an economy open to capital flows it is impossible for the authorities to simultaneously control the exchange rate and the interest rate (or the monetary base). There are at least two ways to express the objection to the SCRER policy based on the trilemma. One of them argues that targeting the exchange rate implies a central bank intervention in the foreign exchange market. In doing so, the central bank loses its ability to control money supply. Targeting the exchange rate and controlling the money supply can be simultaneously pursued only if capital flows are regulated. However, the effectiveness of capital regulation tends to decrease, because the private sector innovative capacity is greater than the public sector regulatory ability. The conclusion is that central banks have to choose between two poles (Fischer, 2001): active monetary and floating exchange rate or hard peg cum passive monetary policy. The second way to express the objection focuses on the argument of controlling inflation. If the interventions in the exchange market target the real exchange rate (instead of the nominal exchange rate), no nominal anchor remains for the public to configure inflationary expectations. Since the central bank cannot control the money supply, the inflation rate is completely out of control. The trilemma is essentially a policy argument, logically derived from interest rate parity theorems for open economies. When forwards exchange

Five years of competitive and stable real exchange rate in Argentina

193

markets are not fully developed the relevant theorem is the uncovered interest parity (UIP) condition. The UIP states that the returns of two perfect substitutes assets nominated in different currencies should be equal. This implies that the domestic interest rate (i) should equalize the sum of the foreign interest# rate (i*) and the expected variation of the nominal exchange rate (E (S) 5 SEt11 2 St /St). With the additional assumptions of small country (i*t 5 i*) and perfect foresight (SEt11 5 S), the UIP condition formally implies: it 5 i* 1

S 21 St

(9.1)

Equation (9.1) is a simple model with two unknowns: the domestic interest rate and the exchange rate (St). Under a credibly fixed exchange rate regime (St 5 S), the model is solved by determining the interest rate endogenously equal to the international rate. In other words, the government is able to set the exchange rate but loses control of the monetary policy. When the exchange rates floats freely, Equation (9.1) is solved by setting the domestic interest rate exogenously. This is the case in which governments have an active monetary policy at the cost of letting the exchange rate float. If both the interest rate and the exchange rate are exogenous, Equation (9.1) is overdetermined. The only way to avoid this situation is to consider the imposition of capital controls, which prevent arbitrage forces to make the parity hold. In any model conclusions critically depend on the assumptions. In the case of the trilemma one crucial assumption is that assets are perfect substitutes. If this assumption is relaxed the validity of the trilemma as a general theorem characterizing the performance of economies open to capital flows no longer holds.15 Moreover, it has been recognized for a long time in open economy macroeconomics that in the context of free capital mobility central banks have room to conduct active monetary policy and control the nominal exchange rate when assets are imperfect substitutes.16 The degrees of freedom of monetary policy vary inversely with the degree of assets substitutability. The degrees of freedom of monetary policy also depend on the institutional characteristics of the central bank, and the situation of the FX market. In a case of excess supply of foreign exchange at the targeted exchange rate, if the central bank is allowed to issue bonds to sterilize, it can control both the prevailing exchange and interest rates by purchasing all the excess supply of international currency in the FX market and sterilize the monetary effect of that intervention through the issuing of bonds in the monetary market. The central bank has two available instruments to perform its two targets: the intervention in the exchange market to control

194

Beyond inflation targeting

the exchange rate and the intervention in the money market to control the interest rate. Tinbergen’s maxim is fulfilled. The excess supply of international currency, at the exchange rate targeted by the central bank, implies an excess demand for domestic assets at the prevailing domestic interest rate. The fully sterilized intervention in the exchange market can be imagined as a policy implemented in two steps. In the first one, before sterilization, the central bank intervention generates a monetary base expansion. The resulting situation would show a higher amount of monetary base, the same amount of domestic bonds and an interest rate lower than the initial one. In the second step, the complete sterilization fully compensates for the change in the private portfolio that took place in the first step. The central bank absorbs the increment in the monetary base and issues an amount of domestic assets equal to the initial excess demand for domestic assets (the excess supply of international currency) turning the domestic interest rate to its previous level (Bofinger and Wollmerhäuser, 2003). Therefore, if assets are imperfect substitutes and sterilization is allowed, the central bank’s ability to simultaneously manage the exchange rate and the interest rate critically depends on the existence of an excess supply of international currency at the targeted exchange rate. In this setting the trilemma is invalid. It seems that this conclusion is not generally acknowledged because the literature discussing monetary autonomy and exchange regimes rarely considers situations of excess supply of international currency. It is mostly focused on balance of payments deficit situations.17 Certainly, in excess demand contexts the predictions of the trilemma are generally valid. Even when assets are imperfect substitutes, in these situations even powerful central banks have a limited capacity to intervene in the FX market. The limit is determined by the stock of international reserves. Consequently, it may be argued that even powerful central banks cannot simultaneously control the exchange rate and the interest rate in contexts of excess demand for international currency. But there is no symmetry between excess demand and excess supply situations. In the first case the trilemma is valid while not necessarily in the second one. The asymmetry lies in the fact that in the first case sterilization is constrained by a fixed stock (that is, the international reserves), while in the second sterilization may be done indefinitely because of a variable stock (that is, central bank’s bonds). The central bank’s ability to issue bonds but not international reserves is the key difference. This ability raises the question whether it is possible to carry the fully sterilized intervention policy under excess supply of foreign currency situations permanently. In order to do so, the central bank has to fulfill a sustainability condition: its net worth should not follow an explosive trend. Sustainability, therefore, depends on the magnitudes of the international

Five years of competitive and stable real exchange rate in Argentina

195

and the domestic interest rates and on the rate of variation of the nominal exchange rate. Taken as given the international interest rate and the trend of the nominal exchange rate, the sustainability condition depends on the domestic interest rate. The central bank enjoys autonomy to determine the domestic interest rate, but in order to be sustainable the policy must determine domestic interest rates lower than a certain upper limit. This limit can be formally determined as follows.18 Assume a central bank that holds international reserves (R) as its unique asset and issues monetary base (H) and remunerated liabilities (L) yielding the domestic interest rate set by the monetary authority (it). Therefore, the central bank’s net worth (N) at any point in time would be: Nt 5 StRt 2 (Ht 1 Lt)

(9.2)

In each period the central bank earns the yielding of international reserves – which for simplicity we assume are invested at the international interest rate – and serves the interest payments of its remunerated liabilities. There is also a valuation effect # on the international reserves due to the variation of the exchange rate (S). Since the changes in the stocks cancel out, central bank’s quasi-fiscal result is equal to the variation of its net worth. # dN 5 SR (i* 1 S) 2 iL

(9.3)

A simple (although restrictive) condition for the central bank’s net worth not to follow an explosive trend is to assume that the quasi-fiscal result has to be non-negative (dN $ 0). Under this sustainability condition, we obtain the maximum domestic interest rate that makes the fully sterilized intervention policy sustainable: # i* 1 S max it 5 (9.4) Lt /StRt It follows that there is a range of interest rates from zero to i max that makes the fully sterilized intervention policy sustainable. Given that central banks typically benefit from seigniorage and inflation tax revenues, the case in which Lt , StRt does not seem unlikely. In these cases, the upper limit of this range would be greater than the sum of the international interest rate and the rate of variation of the nominal exchange rate. It is important to notice that since i max depends on the behavior of Rt and Lt, the range of sustainable interest rates also evolves over time. Given a set of variables and parameters of the economy (such as the inflation rate, the elasticity of money demand and the rate of variation of the exchange rate), i max would tend to decrease as the interest rate set by the

196

Beyond inflation targeting

central bank increases. Thus, in order to keep the policy in a sustainable trend, the cumulative sterilization cost should be bounded and manageable. A key point for sustainability is, therefore, that the domestic interest rates set by the central bank should be ‘moderate’ in the mentioned sense.

9.4

CONCLUDING REMARKS

In this chapter we show that monetary and exchange rate policies targeting a SCRER are viable for developing open economies. We illustrated our argument with recent Argentine experience, which is just one of many other economies like China or India following this strategy. It is important to notice, however, that when a country is trying to preserve a SCRER, monetary, exchange rate and fiscal policies should be coordinated. Otherwise, potential conflicts between domestic goals – such as the exchange rate, inflation rate and employment – might arise. Recent inflationary pressures in Argentina could be an example of conflicts arising from the lack of coordination between economic policies. An outline of a macroeconomic regime targeting a SCRER in which monetary, exchange rate and fiscal policies are coordinated is briefly described as follows.19 First, it is important to mention that such a macroeconomic regime does not imply segmentation between objectives and instruments. The preservation of a SCRER, the level of employment and the control of inflation set the priorities and the restrictions that the economic policy must fulfill. Monetary, exchange rate and fiscal policies should be coordinated in order to guarantee the consistency between the multiple objectives. The exchange rate policy should focus on signaling the stability of the real exchange rate in the medium and long term, in order to set in motion the positive feedbacks mentioned in Section 9.3. In particular, the emergence of appreciation trends should be avoided to prevent self-fulfilling bubbles that increase the monetary ‘costs’ of buying FX interventions and also because real exchange rate appreciation may harm the profitability of tradable activities, making many of them non-viable and forcing firms to close. The preservation of a SCRER does not mean short-run indexation of the nominal exchange rate to domestic prices. The flexibility and advantages of floating the nominal exchange rate in the short run should be preserved. Central bank interventions in the FX market have to achieve two conflicting goals: they have to prevent expectations of real exchange rate appreciation and allow the nominal exchange rate to float in order to discourage short-term speculative capital flows. The interval of interventions

Five years of competitive and stable real exchange rate in Argentina

197

has to be narrow enough to perform the first function and wide enough to perform the second. The FX market behaves like an asset market. Buying and selling decisions are mostly based on expectations. If central bank interventions and signals stabilize expectations around the SCRER – a necessary condition for that is the consistency of monetary, exchange rate and fiscal policies and the robustness of the external sector accounts – the market forces by themselves will tend to stabilize the exchange rate. The monetary ‘costs’ of central bank interventions will be lower and fewer interventions will be required. For this reason, interventions should be firm, in order to clearly show to the market the willingness and strength of the monetary authority. It is implicit in the above presentation of the exchange rate policy that the buying and selling flows of international currency are manageable. This means that the central bank manage to keep the policy in a sustainable path. If capital inflows are massive the cost of sterilization may turn the monetary policy unsustainable. It is important to notice, however, that such a situation might arise as an endogenous consequence of the exchange rate policy itself. Massive capital inflows may result from an excessively stable nominal exchange rate in the short run, which turns speculative investments in one-way bets. Short-run volatility in the exchange rate increases the uncertainty of speculative investments and thus may reduce capital inflows, diminishing the amount and cost of interventions. This is the main reason why the exchange rate policy under a SCRER regime should preserve short-run volatility. However, massive capital inflows that make the policy unsustainable may occur even when the central bank induces short-run volatility in the exchange rate. This could happen in the context of high liquidity in international capital markets. In this kind of situations, it would make little sense to risk macroeconomic stability in order to preserve the capital account full openness principle. The preservation of the macroeconomic regime requires in this case capital account regulations, intended to restrict capital inflows and facilitate the management of exchange and monetary policies. There is a menu of measures able to accomplish this function that, even when they do not work perfectly well, evidence suggests that they contribute to soften capital inflows during booms.20 The need for controls is not permanent; they have to do their job only in a booming phase, and we now know well that booming phases do not last forever. When there is an excess demand for international currency that turns exchange and monetary policies unmanageable, FX interventions would cause an excessive monetary contraction and the consequent rise in the interest rate would trigger a recession. The defense of some nominal

198

Beyond inflation targeting

exchange rate may risk a speculative attack on the central bank reserves. The situation has similarities with a fixed exchange rate regime crisis. But there is an important difference. If there are no fundamental reasons for depreciation – generated, for instance, by expectation of balance of payments deficit – fiscal and monetary policies are consistent with the targeted real exchange rate, and inflation is under control, then the macroeconomic regime should be preserved. This would only be possible if exchange controls and restrictions on capital outflows were imposed. If there are no fundamental reasons inducing the excess demand for international currency, there is no need for the controls and regulations to last for long. As described in Section 9.2, Argentina successfully imposed exchange controls and capital outflow regulations in mid 2002, when the run into foreign currency was mainly caused by a self-fulfilling bubble in the exchange rate. The measures were gradually softened when the buying pressure in the FX market diminished. In a SCRER macroeconomic regime, monetary policy should not be exclusively focused on inflation. It is important to emphasize, however, that this regime performs a preventative role with respect to inflation acceleration. In most developing economies the exchange rate is the main transmission mechanism of monetary impulses to the inflation rate. The SCRER precisely encourages the central bank to implement monetary policies that avoid excessive fluctuations in both the nominal and real exchange rates. In contrast, for the same reason, an exclusive inflation focus of monetary policy generates incentives towards real exchange rate appreciation. In coordination with the other policies, monetary policy should be managed in order to attain multiple objectives. To manage monetary aggregates or set the interest rate to accomplish this goal, the central bank may have to compensate for the interventions in the FX market. Out of the extreme situations discussed above, this can be done through different instruments. The most common is the sterilization operations. In Section 9.3 we showed that fully sterilized interventions can be carried out in the context of excess supply of foreign exchange and we derived the range of interest rates that make the policy sustainable in time. Apart from sterilization, there are other instruments at hand to conduct the monetary policy under the SCRER regime. For instance, a central bank in possession of a significant amount of bank debt can manage it as an instrument for monetary control (Lavoie, 2001). Public sector deposits in the central bank can be used in an analogous way. Some prudential regulations can be oriented to the same target, particularly when the problem is to constraint money expansion. The central bank, for instance, can raise the cash requirements of the banking system and thus lower the

Five years of competitive and stable real exchange rate in Argentina

199

banking multiplier. Other prudential regulations can be directly focused on smoothing the selling pressure in the exchange market. For instance, if local banks are not allowed to back credits in domestic currency with liabilities in international currency and credits in international currency are limited, there are fewer incentives to the banks procuring international funding. The existence of public banks with a significant share of the financial market can facilitate the monetary management. Public banks can be coordinated in order to help the central bank in both the management of the liquidity and the FX interventions. Through the management of these instruments the central bank should be able to keep money expansion under control. Finally, fiscal policy should complement monetary policy in attaining inflation and employment targets in the short run. It should focus on the management of nominal aggregate demand: moderating it in cases of inflationary pressures and expanding it in the opposite situations. However, it is important to notice that since a SCRER regime is meant to promote development and growth, inflationary pressures may be more likely than deflationary ones. Since in this regime monetary policy typically has an expansionary bias through the buying interventions in the FX market, the role of moderating the aggregate demand to avoid inflation and real exchange rate appreciation would tend to rely on the fiscal policy. Conservative fiscal policy could also be necessary in the SCRER regime to ease the central bank needs to intervene in the FX market. Public sector surpluses may be used to buy part of the excess supply in the FX market. An anti-cyclical fiscal fund intended to perform this role could be a good institution to develop in a SCRER regime.

NOTES 1.

2. 3. 4. 5.

The authors would like to thank Nelson Barbosa-Filho, Erinc Yeldan, Jan Kregel and the participants in the workshop on ‘Alternatives to Inflation Targeting Monetary Policy for Stable and Egalitarian Growth in Developing Countries’ held at CEDES 13-14 May 2005 for their comments to previous versions of this chapter. We also thank Martín Fiszbein for his collaboration. The usual caveats apply. In 1992 the new central bank law slightly relaxed this constraint by setting narrow margins to the possibilities of purchasing public bonds and lending to the commercial banks. For an analysis of the macroeconomic performance during the convertibility period, see Damill and Frenkel (2007). An analysis of the evolution of Argentine debt, default and restructuring can be found in Damill et al. (2005). By mid 2007 energy supply shortages raised concerns about the sustainability of high rates of economic growth. These concerns, however, are not related to the SCRER policy.

200

Beyond inflation targeting

6.

Later on, the government issued new debt to compensate the banks for the balance sheet effect of the asymmetric ‘pesoification’. 7. Figure 9.1 shows a ‘jump’ in the private bank deposit series in January 2002. It reflects the accounting effect of the ‘pesoification’ at 1.40 pesos per dollar of deposits issued in foreign currencies, previously valued at a AR$/$ 1 rate. If we put this mere accounting effect aside, it is easy to see the drop in deposits. 8. Exchange rates are defined so that a rise in this variable implies a nominal or real depreciation. 9. They included the obligation to surrender the proceeds from exports in the local FX market. 10. For instance, the effects of the interest rate through the credit channel are very weak in an economy where the bank credit to private sector remains below 13 percent of GDP as in Argentina. 11. The literature emphasizing the positive effects of the SCRER on development increases day by day. See, for instance, Williamson (2003), Rodrik (2005), Dooley et al. (2004), Frenkel (2004), Frenkel and Taylor (2005) and Subramanian (2007). 12. See Hausman et al. (2005) and Prasad et al. (2007). 13. See Frenkel and Ros (2006) for an analytical and empirical study of the positive relationship between real exchange rate and employment in Latin America. 14. Evidence regarding short-run indeterminacy of the real exchange rate seems to be conclusive. Its behavior is almost completely determined by the nominal exchange rate. Although most scholars agree about the existence of an equilibrium real exchange rate in the long run, there is no consensus regarding the factors affecting its determination. The purchasing power parity (PPP) is the most accepted hypothesis (Taylor and Taylor, 2004). However, evidence regarding the PPP shows time series reverting to their means in very long periods (that is, half life of 3–5 years) and results are highly sensitive to data sets and estimation techniques. On the other hand, mean-reverting time series is no sufficient condition for the validity of the PPP hypothesis. 15. Another key assumption is that exchange rate expectations are formed with perfect foresight. If departures from the perfect foresight-rational expectation paradigm are considered, the predictions of the trilemma may no longer hold. For a critique of the trilemma in this vein, see Frenkel and Rapetti (2007). 16. See, for instance, Chapter 10 of Dornbusch (1980). 17. See, as an example, Canales-Kriljenko (2003). 18. The complete model is in Frenkel (2007). 19. For a detailed description of a macroeconomic regime proposal with a SCRER as an intermediate target, see Frenkel (2006). 20. See Epstein et al. (2003).

REFERENCES Balassa, B. (1971), ‘Trade policies in developing countries’, American Economic Review, 61 (2), 178–87. Blecker, R. (2005), ‘Financial globalization, exchange rates and international trade’, in G. Epstein (ed.), Financialization in the World Economy, Cheltenham, UK and Northampton, MA, USA: Edward Elgar, pp.183–209. Bofinger, P. and T. Wollmershäuser (2003), ‘Managed floating as a monetary policy strategy’, Economics of Planning, 36 (2), 81–109. Canales-Kriljenko, J.I. (2003), ‘Foreign exchange intervention in developing and transition economies: results of a survey’, International Monetary Fund working paper no. 03/95.

Five years of competitive and stable real exchange rate in Argentina

201

Damill, M. and R. Frenkel (2007), ‘Argentina: macroeconomic performance and crisis’, in R. Ffrench-Davis, D. Nayyar and J. Stiglitz. (eds), Stabilization Policies for Growth and Development, Initiative for Policy Dialogue, Macroeconomics Policy Task Force, New York, forthcoming. Damill, M., R. Frenkel and M. Rapetti (2005), ‘The Argentinean debt: history, default and restructuring’, Columbia University Initiative for Policy Dialogue working paper series, New York. Díaz Alejandro, C. (1979), ‘Algunas vicisitudes históricas de las economías abiertas en América Latina’, Desarrollo Económico, 19 (74), 147–59. Dooley, M., D. Folkerts-Landau and P. Garber (2003), ‘An essay on the revived Bretton Woods system’, International Journal of Finance and Economics, 9, 307–13. Dooley, M., D. Folkerts-Landau and P. Garber (2004), ‘The US current account deficit and economic development: collateral for a total return swap’, National Bureau Economic Research working paper no. 10727. Dornbusch, D. (1980), Open Economy Macroeconomics, New York: Basic Books. Epstein, G., I. Grabel and K.S. Jomo (2003), ‘Capital management techniques in developing countries: an assessment of experiences from the 1990s and lessons for the future’, Political Economy Research Institute, working paper no. 56, University of Massachusetts Amherst. Feldstein, M. (1999), ‘Self-protection for emerging markets’, National Bureau for Economic Research, working paper no. 6907. Fischer, S. (2001), ‘Exchange rates regimes: is the bipolar view correct?’, Journal of Economic Perspectives, 15 (2), 3–24. Frankel, J. and A.K. Rose (1995), ‘Empirical research on nominal exchange rates’, in: G.M. Grossman and K. Rogoff (eds), Handbook of International Economics, vol. III, Amsterdam: Elsevier, pp. 1689–709. Frenkel, R. (2004), ‘Real exchange rate and employment in Argentina, Brazil, Chile and Mexico’, paper prepared for the G-24, Washington, DC, 24 August. Frenkel, R. (2006), ‘An alternative to inflation targeting in Latin America: macroeconomic policies focused on employment’, Journal of Post Keynesian Economics, 28 (4), 573–91. Frenkel, R. (2007), ‘La sostenibilidad de la política de esterilización’, CEPAL Review, 93 (December), 29–36. Frenkel, R. and J. Ros (2006), ‘Unemployment and the real exchange rate in Latin America’, World Development, 34 (4), 631–46. Frenkel, R. and L. Taylor (2005), ‘Real exchange rate, monetary policy, and employment’, paper prepared for the High-Level United Nations Development Conference, 14-15 March, New York. Frenkel, R. and M. Rapetti (2007), ‘Política cambiaria y monetaria después del colapso de la Convertibilidad’, Ensayos Económicos, 46, 137–66. Hausman, R., L. Pritchett and D. Rodrik (2005), ‘Growth accelerations’, Journal of Economic Growth, 10 (4), 303–29. Lavoie, M. (2001), ‘The reflux mechanism in the open economy’, in L.P. Rochon and M. Vernengo (eds), Credit, Interest Rates and the Open Economy: Essays on Horizontalism, Cheltenham, UK and Northampton, MA, USA: Edward Elgar, pp. 215–42. Polterovich, V. and V. Popov (2002), ‘Accumulation of foreign exchange reserves and long term growth’, New Economic School, working paper, Moscow.

202

Beyond inflation targeting

Prasad, E., R. Rajan and A. Subramanian (2007), ‘Foreign capital and economic growth’, Brooking Papers on Economic Activity, (1), 153–20. Rodrik, D. (forthcoming), ‘Growth strategies’, Brookings Papers on Economic Activity. Subramanian, A. (2007), ‘Capital account convertibility: a neglected consideration’, Economic and Political Weekly, 42 (25), 2413–18. Taylor, A. and M. Taylor (2004), ‘The purchasing power parity debate’, Journal of Economics Perspectives, 18 (4), 135–58. Williamson, J. (2003), ‘Exchange rate policy and development’, paper prepared for the project Capital Market Liberalization Task Force Initiative for Policy Dialogue, Barcelona, 2 June, 2003.

10.

A general equilibrium assessment of twin-targeting in Turkey Cagatay Telli, Ebru Voyvoda and A. Erinç Yeldan1

10.1

INTRODUCTION: MACROECONOMICS OF TWIN-TARGETING IN TURKEY

After a decade of failed reforms and deteriorated macroeconomic performance, Turkey entered the millennium under a Staff Monitoring Program signed with the International Monetary Fund (IMF) in 1998, and put into effect in December 1999. The program currently sets the macroeconomic policy agenda in Turkey and relies mainly on two pillars: (1) fiscal austerity that targets a 6.5 percent surplus for the public sector in its primary budget2 as a ratio to the gross domestic product (GDP); and (2) a contractionary monetary policy (through an independent central bank) that exclusively aims at price stability (via inflation targeting). Thus, in a nutshell the Turkish government is charged to maintain dual targets: a primary surplus target in fiscal balances (at 6.5 percent to the GDP); and an inflation targeting central bank whose sole mandate is to maintain price stability and is divorced from all other concerns of macroeconomic aggregates – hence the terms in the title: macroeconomics under twintargeting. According to the logic of the program, successful achievement of the fiscal and monetary targets would enhance ‘credibility’ of the Turkish government ensuring reduction in the country risk perception. This would enable reductions in the rate of interest that would then stimulate private consumption and fixed investments, paving the way to sustained growth. Thus, it is alleged that what is being implemented is actually an expansionary program of fiscal contraction. On the monetary policy front, the Central Bank of Turkey (CBRT) was granted its independence from political authority in October 2001. In what follows, the central bank announced that its sole mandate is to restore and maintain price stability in the domestic markets and that it will follow a disguised inflation targeting until conditions are ready for full targeting. 203

204

Beyond inflation targeting

Thus, over 2002 and 2003 the CBRT targeted net domestic asset position of the central bank as a prelude to full inflation targeting. Finally on 1 January 2006 the CBRT announced that it will adopt full-fledged inflation targeting. The purpose of this chapter is to provide an assessment of the key macroeconomic developments in Turkey over the post-2001 crisis period and to provide a general equilibrium analysis of the macroeconomic policy alternatives of the twin-targeters. We focus on three sets of issues: first, we study the macroeconomics of the expanded foreign capital inflows in resolving (temporarily) the macroeconomic impasse between the disinflation motives of the CBRT and imperatives of debt sustainability and fiscal credibility of the Ministry of Finance. Second, we study the reduction of the central bank’s interest rates. Third, we implement a labor market reform and study the implications of reducing/eliminating payroll taxes (paid by the employers). To these ends we construct a macroeconomic general equilibrium model with a full-fledged financial sector in tandem with a real sector. Across all policy simulations we exclusively focus on both the fiscal and financial adjustments and study the possible dilemmas of gains in efficiency in the labor markets versus the loss of fiscal revenues to the state. Our finding is that the current monetary strategy followed by the CBRT that involves a heavy reliance on foreign capital inflows along with a relatively high real rate of interest is effective in bringing inflation down; yet it suffers from increased cost of interest burden to the public sector and strains fiscal credibility. In contrast, our simulation results suggest that, given the ex ante constraints of the domestic economy in the short run, an alternative heterodox policy of reduction of the central bank interest rate and lowering of the payroll tax burden in labor markets may have strong employment and growth effects. The policy also achieves significant gains in fiscal credibility in the short run. Yet it suffers from increased inflationary pressures in the commodity and the financial markets. Even though observed to prevail at a modest scale in our simulation experiments, the ex ante constraints of maintaining inflationary expectations may lead to intolerance of the CBRT and render the policy ineffective. Thus, maintaining an integrated and coherent policy framework between the monetary and fiscal authorities is seen of prime importance for the success of the policy formulation at the macro scale. Our premise in this chapter is that a proper modeling of the general equilibrium linkages between the production-income generation and aggregate demand components across individual sectors as well as responses of the real macro aggregates to financial decisions are essential steps to understand the impact of the current austerity program on the evolution of output, fiscal, financial and external balances, and on employment.

A general equilibrium assessment of twin-targeting in Turkey

205

Accordingly, we develop a computable general equilibrium (CGE) model with a relatively aggregated productive sector, a segmented labor market and a full-blown public sector with a detailed treatment of fiscal balances and financial flows. The current model shares many of the analytical structure of the Agénor et al. (2006) design in the dynamics of financial transactions, especially with respect to formation of expectations and fragility. It is explicitly designed to capture the relevant linkages between the fiscal policy decisions, financialization constraints and external balances that we believe are essential to analyse the impact of disinflation and fiscal reforms on labor market adjustment and public debt sustainability. We pay particular attention to fiscal issues such as a high degree of debt overhang and fiscal dominance; the link between real and financial sector interactions, and interactions between external (current account) deficits private savinginvestment deficits and the public (primary balance) surpluses. We organize the chapter under four sections. First, we provide a broad overview of the recent macroeconomic developments in Turkey in Section 10.2. Here we study, exclusively, the evolution of the key macroeconomic prices such as the exchange rate, the interest rate and price inflation. Here we also comment on the external balances, the dynamics of external debt, fiscal policy issues and the labor market. In Section 10.3, we introduce and implement our CGE modeling analysis of the alternative policy scenarios to depict the short-run macroeconomic adjustments of the Turkish economy under the conditionalities of the IMF program targets on primary surplus to GDP ratio and on inflation rate. Finally, we provide a brief summary with concluding comments in Section 10.4.

10.2

MACROECONOMIC DEVELOPMENTS UNDER IMF’S STAFF MONITORING

The growth path of the Turkish economy over the post-1998 period had been erratic and volatile, mostly subject to the flows of hot money. Following the contagion effects of the Asian, Russian and Brazilian turmoil, the economy first decelerated in 1998 with a growth rate of 3.1 percent, and then contracted in 1999 at the rate of –5.0 percent. The boom of 2000 was followed by the 2001 crisis. The recovery was sharp as the economy has grown at an average rate of 7.1 percent over the 2002–06 period. Price movements were also brought under control through the year and the 12-month average inflation rate in consumer prices has receded from 45 percent in 2002 to 7.7 percent in 2005, and from 50.1 percent to 5.9 percent in producer prices. The post-2003 period has also meant a period of acceleration of exports,

206

Beyond inflation targeting

where export revenues reached $91.7 billion in 2006. Nevertheless, with the rapid rise of the import bill over the same period, the deficit in the current account reached $31.7 billion (or about 7.9 percent of the gross national product (GNP) in 2006). The current account deficit continued to widen in 2007 and reached $34 billion over 12 months cumulative period in the first quarter. On the public sector front one witnesses a very strong fiscal discipline effort. The ratio of central government budget deficit to the GNP was reduced from its peak of 16.2 percent in 2001 to 0.8 percent by 2006. Consequently, the public sector borrowing requirement (PSBR) as a ratio to the GNP fell from 16.1 percent to negative 3 percent, indicating a surplus, in 2006. Table 10.1 documents the main macro indicators of the post-1998 Turkish economy under close IMF supervision. 10.2.1

Macroeconomic Prices and the Monetary Policy

The CBRT initiated an open inflation targeting framework starting 1 January 2006. The CBRT’s current mandate is to set a ‘point’ target of 5 percent inflation of the consumer prices. Given internal and external shocks, the CBRT has recognized an internal (of 1 percent) and an external (of 2 percent) ‘uncertainty’ band around the point target. Thus, the CBRT will try to keep the inflation rate at its point target; however, recognizing a band of maximum 2 percentage points below or above the 5 percent target rate. The CBRT has announced that it will continue to use the overnight interest rates as its main policy tool to reach its target. It is stated explicitly that the ‘sole objective of the CBRT is to provide price stability’, and that all other possible objectives are out of its policy realm.3 Despite the positive achievements on the disinflation front, rates of interest remained slow to adjust. The real rate of interest remained above 10 percent for much of the post-2001 crisis era, and generated heavy pressures against the fiscal authority in meeting its debt obligations (Table 10.1). The persistence of the real interest rates, on the other hand, had also been conducive in attracting heavy flows of short-term speculative finance capital over 2003 and 2006. This pattern continued into 2007 at an even stronger rate. Inertia of the real rate of interest is enigmatic from the successful macro economic performance achieved thus far on the fiscal front. The credit interest rate, in particular, had been constrained by a lower bound of 16 percent despite the deceleration of price inflation. Consequent to the fall in the rate of inflation, the inertia of credit interest rates translates into increasing real costs of credit. High rates of interest were conducive in generating a high inflow of hot money finance to the Turkish financial markets. The most direct effect of the surge in foreign finance capital over this period was felt in the foreign

207

Contagion of emerging market financial crises 1999

Staff monitoring program initiated

1998

−9.2 −8.6 −34.9 −22.0 7.4 −24.8 17.5 19.0 −16.2 16.4 2.4 92.7

16.0 19.6 19.2 25.4 18.2 22.8 −10.9 11.8 −4.8 63.4

−7.4

2001

6.2 7.1

7.4

2000

12.7 −0.8 77.5

19.2 17.3 −14.3

−7.2 14.5 11.0 15.7

2.0 5.4

7.6

2002

9.3 −3.4 57.1

19.3 16.1 −11.2

20.3 −11.5 16.0 27.1

6.6 −2.4

5.8

2003

4.7 −5.2 50.4

20.2 18.4 −7.1

45.5 −4.7 12.5 24.7

10.1 0.5

8.9

2004

−0.4 −6.2 46.9

17.1 20.3 −2.0

23.6 25.9 8.5 11.5

8.8 2.4

7.4

2005

Financial IMF-Directed Post-Crisis Adjustments IMFcrisis directed Under the pragmatic and Under the disinwestern-friendly Islamism three-party flation of the AKP coalition program government

Basic characteristics of the Turkish economy under the IMF surveillance, 1998–2006

Real Rate of Growth GDP 3.1 −5.0 Consumption Expenditures Private 0.6 −2.6 Public 7.8 6.5 Investment Expenditures Private −8.3 −17.8 Public 13.9 −8.7 Exports 12.0 −7.1 Imports 2.3 −3.7 Macroeconomic Balances (as ratio to the GNP, %) Aggregate Domestic Savings 22.7 21.2 Aggregate Fixed Investments 24.3 22.1 Budget Balance −7.0 −11.6 Public Sector Borrowing Requirement 9.3 15.5 Current Account Balance 1.0 −0.7 Stock of Foreign Debt 55.4 71.0

Table 10.1

−3.0 −7.9 50.4

16.6 23.1 −0.8

17.4 −0.2 8.5 7.1

5.2 9.6

6.1

2006

208

(continued)

60.6 53.1 64.8 36.8 8.6 18.3

−0.9 5.5

1999

−2.6 15.6

28.6 51.4 54.9 4.5

2000

−14.4 −11.5

114.2 61.6 54.4 31.8

2001

−5.0 0.5

23.0 50.1 44.9 9.1

2002

Source:

4.8 4.7

−4.9 14.6 10.6 13.1

2004

1.6 7.9

−5.7 5.9 7.7 10.4

2005

SPO Main Economic Indicators; Undersecreteriat of Treasury, Main Economic Indicators; CBRT data dissemination system.

0.5 −5.3

−0.6 25.6 25.3 15.4

2003

Financial IMF-Directed Post-Crisis Adjustments IMFContacrisis gion of Directed Under the pragmatic and Under the disinemerging western-friendly Islamism three-party flation market of the AKP coalition financial program governcrises ment

71.7 71.8 84.6 29.5

1998

Staff monitoring program initiated

Notes: a. Deflated by the Producer Price Index. b. Based on real wage indexes (1997 5 100) in manufacturing per hour employed, Turkstat data.

Macroeconomic Prices Rate of Change of the Nominal Exchange Rate (TL/S) Inflation (PPI) Inflation (CPI) Real Interest Rate on GDIsa Real Wage Growth Ratesb Private Sector Public Sector

Table 10.1

1.9 −3.0

6.9 9.4 9.6 7.9

2006

A general equilibrium assessment of twin-targeting in Turkey

209

exchange market. The overabundance of foreign exchange supplied by the foreign financial arbiters seeking positive yields led significant pressures for the Turkish lira to appreciate. As the CBRT has restricted its monetary policies only to the control of price inflation, and left the value of the domestic currency to be determined by the speculative decisions of the market forces, the lira appreciated by as much as 40 percent in real terms against the US dollar and by 25 percent against the euro (in producer price parity conditions, over 2002–06). The overvaluation of the lira was the most important contributor in reducing the burden of an ever-expanding foreign indebtedness. While the aggregate foreign debt stock has increased from US$113.6 billion in 2001 to US$206.5 billion by the end of 2006, as a ratio to the GNP it has created an illusionary tendency to fall when measured in the overvalued lira units. 10.2.2

Fiscal Policy and Debt Management

The current fiscal policy stance in Turkey relies primarily on expenditure restraint. On the revenue side one witnesses a significant effort in raising tax revenues, both in real terms and also as a ratio to the GNP. Much of this effort can be explained by the rise in the share of indirect/excise taxes on goods and services (to 21 percent as a ratio to the GNP, or about 70 percent of total tax revenues), while the contribution of direct income taxes to the budgetary revenues are observed to fall especially after 2000. Data reveal a secular fall in the budget deficit through the post-2001 crisis adjustments and is now reduced to less than 1 percent to the GNP. As discussed above, much of the aggregate budget expenditures can be explained by the high costs of debt servicing, and the main logic of the current austerity program rested on maintaining the debt turnover via only primary surpluses. As a result, the boundaries of the public space are severely restricted, and all fiscal policies are directed to securing debt servicing at the cost of extraordinary cuts in public consumption and investments. Within total expenditures, public investments’ share has fallen from 12.9 percent in 1990 to 5.1 percent in 2003. As a ratio to the GNP, public investments stand at less than 2 percent currently. All of these painful adjustments on the fiscal front can be contrasted against the ‘gains’ over the existing debt burden of the public sector. Data from the Ministry of Finance4 reveal that, as a ratio to the GNP, gross public debt of the aggregate public sector has fallen from 68.1 percent in 2000 to 63.1 percent by the end of 2006, a decline of only 5 percentage points. This could have been achieved despite the very rapid rise in the rate of growth of GNP (7.2 percent per annum over the whole period), and the very strict fiscal austerity measures of primary surplus targets (of

210

Beyond inflation targeting

6.5 percent to the GNP for 2002 and beyond). Furthermore much of this decline has come only after 2005, and all of it is due to the decline in the ratio of foreign debt to the GNP. As a ratio to the GNP, public external debt declined from 25.2 percent in 2000 to 16.9 percent in 2006; while the domestic debt burden increased from 43.1 percent to 46.2 percent over the same period. It is a clear fact that the illusion of falling foreign indebtedness is a direct outcome of the real appreciation of the Turkish lira. As the increased external indebtedness of the public sector from $47.6 billion in 2000 to $69.6 billion in 2006, its ratio to the GNP had the effect of a fall when denominated in appreciated liras. In fact, appreciation of the lira disguises much of the fragility associated with both the level and the external debt induced financing of the current account deficits. A simple purchasing power parity (PPP) ‘correction’ of the real exchange rate, for instance, would increase the burden of external debt to 76.8 percent as a ratio to the GNP in 2005.5 This would bring the debt burden ratio to the 2001 pre-crisis level. Under conditions of the floating foreign exchange regime, this observation reveals a persistent fragility for the Turkish external markets, as a possible depreciation of the lira may severely worsen the current account financing possibilities. 10.2.3

Persistent Unemployment and Jobless Growth

Yet the most striking observation on the Turkish labor markets over the post-2001 crisis era is the sluggishly slow performance of employment generation capacity of the economy. Despite the very rapid growth performance across industry and services, employment growth has been meager. This observation, which actually is attributed to many developing economies as well,6 is characterized by the phrase ‘jobless growth’ in the literature. The rate of open unemployment was 6.5 percent in 2000, increased to 10.3 percent in 2002, and remained at that plateau despite the rapid surges in GNP and exports. Open unemployment is a severe problem, in particular, among the young urban labor force reaching 24.5 percent in 2005. On the other hand, the participation rate fluctuates around 48 percent to 50 percent, due mostly to the seasonal effects. It is known, in general, that the participation rate is less than the EU averages. This low rate is principally due to women choosing to remain outside the labor force, a common feature of Islamic societies, but its recent debacle depends as much on the size of the discouraged workers who had lost their hopes for finding jobs. According to Turkish Statistical Institute (Turkstat) data, the excess labor supply (unemployed plus underemployed) is observed to reach 13.6 percent of the labor force by the end of 2006 (Figure 10.1). Thus, to conclude, two important characteristics of the post-crisis

A general equilibrium assessment of twin-targeting in Turkey 20

54 Total unemployment rate (%) Labor participation rate

53

16

52

14

51

12

50

10 49

8

48

6 4

47

2

46

0

Participation rate (%)

18 Unemployment (%)

211

45 1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

Source: Author’s calculation using Turkish Statistical Institute (Turkstat), Household Labor Force Surveys database.

Figure 10.1

Labor participation rate and total unemployment

adjustment path stand out. First, the post-2001 expansion is observed to be concomitant with a deteriorating external disequilibrium, which in turn is the end result of excessive inflows of speculative finance capital. Second, the output growth contrasts with persistent unemployment, warranting the term ‘jobless growth’. The foregoing facts bring the following tasks to our agenda: (1) What are the viable policy choices in combating unemployment in the short run, and under the conditionalities of the ‘twin targets’? (2) Given our assessments of fragility conditions currently prevailing in Turkey, what are the short-run effects of a reduction in the interest cost of the central bank credit in terms of output, employment, foreign indebtedness and other macro aggregates? We now turn to the analytics of general equilibrium with the aid of our CGE model to study these questions.

10.3

COMPUTABLE GENERAL EQUILIBRIUM MODELING ANALYSIS

Given the overview of the recent macroeconomic developments, we now develop a real-financial CGE model for Turkey. In what follows, we provide a bird’s-eye overview of the model, and invite the interested reader to the Political Economy Research Institute website for a full algebraic description.7

212

10.3.1

Beyond inflation targeting

The Algebraic Structure of the Model and Adjustment Mechanisms

Product markets The model is fairly aggregate over its microeconomic structure but accommodates a relatively detailed treatment of the public accounts, and of realfinancial sector linkages. There are four production sectors as agriculture, industry, private services and public services. There is a financial sector with a full-fledged banking segment, a central bank, enterprises, government and household portfolio instruments. Sectoral production is modeled via multilevel functions. At the top level total output is given as a Leontieff specification of value-added and intermediate inputs. The value-added in each sector is generated by combining labor, as well as public and private physical capital. At the last stage of this multilevel production a sector-specific public capital combines with the composite input under a Cobb-Douglas specification. The composite primary input, in turn, is defined to be a combination of private capital and labor aggregate Li through a constant elasticity of substitution (CES) type of production function. Public capital is assumed to be fixed and sector-specific. Private capital is mobile across sectors and the movement is directed by the difference in the differentiated private profit rates. Labor’s wage rate is fixed in the short run and the labor market clears through quantity adjustments on employment. Households save a fraction 0 , sP , 1 of their disposable income. The saving rate is considered to be a positive function of the expected real interest rate in domestic currency denominated deposits: sP 5 sP0 a

H

1 1 intD s SAV b 1 1 E [ Inf ]

(10.1)

with E[Inf ], the expected inflation rate and sp0 is a scaling parameter. The portion of income that is not saved is allocated to consumption and that total flow of savings of the household is channeled to the accumulation of household financial wealth. Private capital investment is assumed to depend on a number of factors. The first is the growth rate of real GDP, which captures the regular accelerator effect. This effect is positive. The next one is the negative effect of the expected real cost of borrowing from the domestic banks. Specifically, private investment demand is represented by: DRealGDP21 sACC 1 1 intLD 2sINTL PK # PINV 5 a1 1 (10.2) b a b NomGDP DRealGDP21 1 1 E [ INF ]

A general equilibrium assessment of twin-targeting in Turkey

213

where NomGDP and RealGDP are the nominal and real values of the gross domestic product, respectively, valued at market prices. Financial markets, asset allocation and risk premia Household’s financial wealth is typically allocated to five different categories of assets: domestic money, HD, domestic currency denominated bank deposits held at home, DDH, foreign currency denominated deposits held domestically,8 FDDomH, holdings of government bonds, GDIH and portfolio investments abroad, PFIH.9 The household demand function for currency is positively related to consumption and negatively related to expected inflation and interest on domestic currency denominated deposits, intD. It also depends negatively on the interest on foreign currency denominated deposits, intDF, adjusted for the expected rate of depreciation (1 1 Δeexp): H

HD 5

H

H

H 0PRIVCON q CON (1 1 E [ Inf ]) 2q Inf (1 1 intD) 2q DD H [ (1 1 Deexp) (1 1 intDF ) ] qDF

(10.3)

Household allocation on domestic versus foreign currency deposits is a function of the interest rate on domestic currency denominated deposits as a ratio to the rate of return on foreign currency denominated deposits held at home. Total portfolio investments of households abroad is taken to be a fixed fraction of total household financial wealth, and the demand for government bonds by households is regarded as a function of the expected bond interest rate, E[intB]. A crucial decision of the enterprise sector is how to allocate their profits between funds to private investment and funds to government bonds. This decision depends on the average profit rate expected from production activities and, on the other hand, expected returns on government debt instruments: (1 1 E [ intB ]) 2sGDI PK # PINV E 5 m c d GDI (1 1 avgRPR21) DGDIE E

(10.4)

Banks set both deposit and lending interest rates. The deposit rate on domestic currency denominated deposits, intD, is set equal to the borrowing rate from the central bank, intR. The deposit rate on foreign currency deposits at home, on the other hand, is set on the basis of the (premium inclusive) marginal cost of borrowing on world capital markets: (1 1 intDF ) 5 (1 1 intFW )

(10.5)

Following Agénor et al. (2006), the risk-premium inclusive foreign

214

Beyond inflation targeting

interest rate is formulated as a function of the (risk-free) world interest rate, intFWRF, and an external risk premium: (1 1 intFW ) 5 (1 1 intFWRF ) (1 1 riskpr)

(10.6)

in which the risk premium is assumed to be a function of total foreign debt to exports ratio: riskpr 5 contag 1

k a ForDebt 2 a b 2 a Ei

(10.7)

In Equation (10.7) contag is an exogenous coefficient used to capture the characteristic changes in the ‘sentiments’ in world capital markets. The bank lending rate, intLD, in the last analysis is defined as a weighted average of the cost of borrowing from the central bank and the cost of borrowing from foreign capital markets. It also takes into account the (implicit) cost of holding required reserves. Public sector, credibility and expectations Since the government debt instruments constitute a relatively significant share of the assets in the domestic financial markets in Turkey, modeling the interactions between the public sector and the central bank (the socalled ‘fiscal dominance’) is one of the crucial concerns of this study. With a mandated target of the ‘primary surplus-GNP ratio’, the government’s fiscal policy is basically centered around the primary balance. A fiscal deficit is still realized if interest costs on the outstanding public debt exceeds the primary surplus. The public sector borrowing requirement, PSBR, is financed by either an increase in foreign loans or by issuing bonds. Of crucial importance is the realization of the interest rate on government bonds. The expected rate of return on this instrument is defined as: E [ intB ] 5 (1 2 PRdefault) intB

(10.8)

where PRdefault denotes the ‘subjective’ probability of default on the current stock of public debt as perceived by the ‘markets’. This variable is set to depend on, among various alternative measures, the current debt stock to tax revenues ratio with a one-period lag: PRdefault 5 1 2 e2g0

(DomDebtG 1 ForDebtG ) GTaxRev

(10.9)

The probability of default, PRdefault, has also a further effect on inflation expectations in such a way that the less the probability of default that

A general equilibrium assessment of twin-targeting in Turkey

215

is perceived, the higher the chances for the ‘declared’ inflation target to materialize. Following Agénor et al. (2006) the expected inflation rate is formulated as a function of the government’s ‘credibility inficator’, that is, the inverse of the probability of default, PRdefault, and the targeted rate of inflation in the previous period: E [ Inf ] 5 (1 2 PRdefault) Inf trgt 1 PRdefaultInf21

(10.10)

Note that, under such a setting, the demand for government bonds is affected by the probability of default. Private investors assign a non-zero probability of default in the current period. The expected rate of return will reflect the probability and will demand compensation in the form of higher nominal interest rates on government bonds. On the other hand, the larger the stock of debt, the higher the probability of default, and the higher the interest rate. For a given probability of default, a continued increase in the supply of bonds will require an increase in interest rates to evoke investors’ demand. Next an increase in the stock of debt will lead to a rise in the probability of default, which will also rise the prevailing interest rate on government bonds. Such a mechanism in the model tries to capture the structure of government trying to provide a signal of confidence to the markets under the current measures of the program. 10.3.2

General Equilibrium Analysis of Alternative Policy Environments

Now we utilize our CGE apparatus to provide a general equilibrium analysis of the macroeconomic policy alternatives under twin-targeting. In what follows we will focus on three sets of issues to depict three alternative policy environments: first, we highlight the important role of the expanded foreign capital inflows in resolving (temporarily) the macroeconomic impasse between the disinflation motives of the CBRT and imperatives of debt sustainability and fiscal credibility of the Ministry of Finance. Second, we implement a ‘fiscally benign’ monetary policy of reducing the interest rate charged by the CBRT. Third, we complement the interest rate reduction policy with a labor market reform and study the implications of reducing/eliminating payroll taxes (paid by the employers). In all of the policy simulations we exclusively focus on both the fiscal and financial adjustments, and study the possible dilemmas of gains in efficiency in the labor markets versus the loss of fiscal revenues to the state. Our simulation experiments are implemented as one-shot, comparative-static exercises. The results are tabulated in Table 10.2. As valid for all types of modeling exercises of the current genre, the simulation results should not be taken as

216

Table 10.2

Beyond inflation targeting

Experiment results

Macroeconomic Aggregates Real GDP (Bill 2003 TL) Real Private Consumption (Bill 2003 TL) Real Private Investment (Bill 2003 TL) Merchandise Imports (Bill US$) Merchandise Exports (Bill US$) Current Account Balance (Bill US$) Unemployment Rate (%) Average Profit Rate (%) As Ratios to the GDP Private Consumption Private Investment Imports Exports Current Account Balance Financial Rates and Prices Inflation Rate (CPI) Expected Inflation Rate Expected Depreciation Rate Realized Depreciation Rate Central Bank Interest Rate (intR) Interest Rate on Domestic Deposits (intD)

Base year (2003) data

EXP-1: Effects of increased foreign capital inflows

EXP-2A: Reduce central bank interest rate

EXP3: Reduce central bank interest rate and reduce payroll taxes

369.700

369.765

370.283

375.631

255.022

263.259

250.255

256.549

66.212

75.138

72.931

74.535

69.378

76.307

70.240

70.910

47.215

43.475

47.866

48.354

−9.201 10.55 16.15

−22.491 10.35 16.20

−9.234 10.69 16.20

−9.511 7.63 16.80

68.98 17.91 28.06 19.09

71.10 20.30 29.70 16.92

67.60 19.70 28.39 19.34

68.80 20.00 28.35 19.33

−3.72

−8.75

−3.73

−3.80

25.30 17.65

18.72 18.83

27.72 16.58

28.23 16.79

41.66

41.78

41.56

41.58

−1.01

−9.52

0.98

0.97

40.27

40.27

20.00

20.00

40.27

40.27

20.00

20.00

A general equilibrium assessment of twin-targeting in Turkey

Table 10.2

217

(continued)

Interest Rate on Private Domestic Debt (intLD) Interest Rate on Government Bonds (intB) Expected Interest Rate on Gov. Bonds (E[intB]) Risk Premium Incl. Foreign Int. Rate (intFW) Fragility Indicators Ratio of Gov. Dom. Debt to Tax Revenues Government’s Fiscal Credibility Index Perceived Probability of Default on Gov. Debt Ratio of Foreign Debt to Central Bank Foreign Reserves Ratio of Foreign Debt to GDP Risk Premium on Private Foreign Borrowing Currency Substitution (FX deposits/Tot. Deposits) Monetary Aggregates (as ratio to the GDP) Money Demand by HH Domestic Deposits of HH FX Deposits of HH Central Bank Foreign Reserves

46.50

47.16

37.65

37.65

36.56

60.37

5.49

5.92

18.28

25.63

3.13

3.29

3.41

4.09

3.24

3.23

156.98

196.26

122.34

127.76

0.50

0.42

0.57

0.56

0.50

0.58

0.43

0.44

235.87

264.88

223.08

223.77

48.93

52.89

46.25

45.92

1.39

2.10

1.20

1.20

90.86

90.99

93.85

93.85

2.64

2.75

2.68

2.68

21.27 19.32

20.76 18.89

20.75 19.47

20.60 19.33

0.21

0.20

0.21

0.21

218

Table 10.2

Beyond inflation targeting

(continued)

Fiscal Results (as ratios to the GDP) Government Aggregate Revenues Government Tax Revenues Government Consumption Exp. Government Investment Exp. Government Interest Exp. PSBR Primary Balance

Base year (2003) data

EXP-1: Effects of increased foreign capital inflows

EXP-2A: Reduce central bank interest rate

EXP3: Reduce central bank interest rate and reduce payroll taxes

39.13

39.44

38.76

37.03

33.35

33.79

33.15

31.48

11.95

12.05

11.84

11.31

4.36

4.53

4.16

3.20

16.60 14.28 6.50

42.40 28.01 6.50

3.99 1.00 6.50

4.13 1.09 6.50

a ‘forecast’ of the future, but rather ought to be regarded as quantitative insights on the relevant macroeconomic outcomes of alternative policy environments. EXP-1: Macroeconomics of foreign capital inflows The post-2001 Turkish economy has benefited from the recent surge of financial flows quite extensively. The increased buoyancy in the global financial markets led both to a fall in the rates of interest in the global markets and also served for provision of expanded liquidity, propelling consumption and investment expenditures. Mostly driven by the private portfolio flows, the net annual inflow of finance capital into the ‘new emerging market economies’ totaled $456 billion in 2005, before receding to $406 billion in 2006.10 These magnitudes exceeded the previous peaks hit in the global financial markets before the eruption of the 1997 Asian crisis. As outlined in Section 10.2, Turkey too had been one of the major beneficiaries of this financial glut. Balance of payments data indicate that the finance account has depicted a net surplus of $103.3 billion over the ‘AKP (Justice and Development Party) period’, 2003 through 2006.

A general equilibrium assessment of twin-targeting in Turkey

219

About half of this sum ($151.2 billion) was due to credit financing of the banking sector and the non-bank enterprises, while a third ($32.8 billion) originated from non-residents’ portfolio investments in Turkey. It is also observed that 64 percent of the total inflows (net financial flows plus errors and omissions) was used for financing the current account deficit which had totaled $71.8 billion over the same period; while 36 percent had been used for reserve accumulation of the CBRT. In this first policy experiment we first study the macroeconomic adjustment mechanisms against this continued inflow of finance capital into the Turkish economy. To this end, we exogenously increase the total inflow of portfolio investments from abroad, PFIROW, by a factor of $30 billion (roughly the realized net cumulative flow over 2003–06). No change in the CBRT’s current monetary policy stance is envisaged with respect to the level of interest rates and/or exchange rate administration.11 The exchange rate was left to full float to be determined by the free play of foreign exchange market transactors. Our results are tabulated under the column EXP-1 in Table 10.2. The immediate effects of the increased inflow of foreign capital are felt in the currency markets. The exchange rate appreciates by 9.5 percent and cost savings on the import side leads to a fall in the inflation rate to 18.7 percent, from 25.3 percent. Appreciation of the exchange rate leads to a rise in imports and the current account deficit widens to increase by about four-fold to reach $22.5 billion in 2003 prices. As a ratio to the GDP, it increases to 8.7 percent from its base value of 3.7 percent. The domestic counterpart of the widening current account deficits is the expansion of private investment (by 2.4 percentage points as a ratio to GDP) and of private consumption (by 2.1 percentage points as a ratio to GDP). The monetary base expands by 20 percent and serves for the liquidity requirements of this expansion. The aforementioned expansion of the economy is limited, however, only to the private sector. Given the fiscal constraint on the primary surplus target, the government’s room for maneuver is limited on the expenditure side. This constraint becomes even more binding as the domestic economy continues to operate with a significantly high real interest burden. It has to be remembered that a critical feature of the simulated policy environment is that the central bank continues to maintain its interest rate at the already high level. As the economy disinflates, however, the real cost of credit increases even further. The interest cost on the government’s debt instruments, in particular, expands to 60 percent from 36.3 percent. The government’s interest expenditures as a ratio to the GDP rises to 25 percent and that of the public sector borrowing requirement increases to 28 percent. Increased interest expenditures lead to a widening of the

220

Beyond inflation targeting

fiscal deficit. Consequently there is a worsening of the fiscal credibility of the government. The credibility index falls to 0.43 from its value of 0.50. The loss in fiscal credibility leads to a rise in the subjective probability of default as perceived by the private markets. Thus, the main result of the scenario unveils an important dilemma for the post-2001 Turkish economy: a policy of maintaining high real rates of interest along with heavy reliance on foreign finance proves to be disinflationary, and it also has expansionary effects on the private sector. The net result is that the CBRT achieves relative success in controlling inflationary pressures. In the meantime, however, the increased debt burden strains the already fragile fiscal balances and results in further loss of fiscal credibility. The predicament of controlling price inflation via high real interest rates and enhanced foreign capital inflows, on the one hand, and the imperatives of debt turnover and fiscal credibility, on the other, remains unresolved. This impasse is further accentuated with the rise of foreign indebtedness and consequent external fragility. As the results of EXP-1 suggest, stock of external debt increases both as a ratio to the GDP (from 48.9 percent to 52.9 percent) and to the foreign reserves of the central bank (from 235 percent to 264 percent). As a result of these adverse developments on external fragility, the risk premium on the Turkish liabilities in the world markets increase by 6 percentage points. Clearly, the realized quandary is not to be resolved by reliance on foreign capital and tight macro management alone, and postponing the necessary adjustments on the domestic front simply lead to culminated pressures on the fiscal side as well as on the external balances. EXP-2: Reduce the central bank interest rate Given the rather high costs of disinflation in terms of high fiscal and external fragility in the previous experiment, the natural policy question is to study the effects of a reduction in the interest rate charged by the central bank. In general, the burden of the interest rates has a significant contractionary effect on the Turkish financial sector. The cost of central bank liquidity is held responsible by many scholars for the external debt cycle and intensified inflows of speculative short-term finance into the Turkish economy. There is a general call for a reduction of the central bank’s rate of interest to escape the trap of speculative inflows of finance leading to appreciation and more inflows, with the consequent widening of the current account deficit and the rise of external indebtedness. Thus, in this experiment we reduce the central bank interest rate by half. Again, as above, no further change is envisaged, when experimenting with this simulation, in the policy instruments or in the parameterization of the exogenously set variables. Note that given the algebraic characterization of the loanable funds market, the central bank interest rate has a direct effect on the determination

A general equilibrium assessment of twin-targeting in Turkey

221

of the deposit interest rate of the domestic banks. Thus, intD is reduced by the same magnitude (50 percent) immediately. With declining rates of interest on deposits the banks find it possible to lower their credit interest rate charged to the enterprises (intLD falls by 8 percentage points). Private investment expenditures increase by 1.9 percentage points as a ratio to the GDP. The distinguishing adjustment mechanism at work is the expenditure switching of private consumption with investment. As private consumption expenditures are reduced by 2 percentage points as a ratio to the GDP, domestic savings could be generated to sustain the expansion in investments; thus, the net effect on the external balances remains modest. In other words, with a given level of domestic disposable income, an expanded level of investment demand could have been sustained. Thus, the current account balance is affected only marginally and the (nominal) adjustments in the exchange rate are revealed to be modest as well, with a realized depreciation of less than 1 percent. What brings forth this adjustment in the private expenditure patterns is the inflation tax. Price inflation accelerates by 2.7 percentage points, and causes a downward shift of aggregate private consumption. The acceleration of inflation further strengthens the decline of the real interest cost for all agents of the economy, private and public. In fact, the relative buoyancy of the economy, along with the decline of the public interest expenditures, leads to an improvement in the fiscal balances. Of particular interest is the decline in the public sector borrowing requirement to less than 1 percent of the GDP, and the increase of the credibility index by 7 percentage points. It is not clear, however, whether the central bank would be willing to tolerate the resultant increase of the inflation rate, which turns out to be the crucial adjusting variable to bring forth the warranted adjustments in the real economy. Yet a further issue is that even though the fiscal results of the policy are observed to be benign, the employment gains remain quite meager. Unemployment rate persists at above the 10.5 percent level, and it is this problem we aim to tackle in the next experiment. EXP-3: Complement central bank interest rate reduction with labor tax reform In this experiment we continue on the policy environment of the previous experiment and complement the central bank’s interest reduction strategy with a labor tax reform. Keeping the central bank’s interest rate at its reduced level (at half of the base run value), we now implement a further reduction on taxes paid by employers of labor. Turkey has one of the highest tax burdens on the labor markets. Employer-paid social security contributions averaged about 36 percent

222

Beyond inflation targeting

of total labor costs during 1996–2000; it has been argued that these high social security taxes create strong disincentives to job creation. More generally, many observers have called for a thorough overhaul of Turkey’s social insurance system. Ercan and Tansel (2006) too state that both the red tape and non-wage labor costs are higher in Turkey relative to, for instance, OECD averages. The authors consider the high tax burden on employment and high social security contributions among the institutional factors that contribute to the high level of unemployment and high level of undeclared work. Tunali (2003) indicates that employee contribution to the social security system can be as high as 15 percent while an employer in a typical risk occupation contributes as much as 22.5 percent. Thus in this experiment we study the implications of lowering the payroll tax paid by the employers on employment, production and fiscal balances. Maintaining the central bank rate of interest at half of its base run value, we reduce the payroll tax by half, from its base rate of 19 percent. The lower tax revenues are not compensated by any other taxes. The results of the experiment are depicted under column EXP-3 in Table 10.2. Clearly, the most important variable of this experiment is its effects on unemployment rate and the fiscal balances. Unemployment rate falls by around 3 percentage points, and the real GDP expands by 1.7 percent upon impact. We find, however, that the main adjustment falls on public investments and then on the price inflation. The first outcome is the direct result of the fiscal administration under the current austerity program. The logic of the fiscal balances is that, given the tax revenues and interest costs, the public sector is to maintain a primary surplus (of 6.5 percent) as a ratio to the GDP. Once this constraint is met the rest of the public expenditures are calculated. Thus, within the context of our experiment, as tax revenues are curtailed, the government finds it necessary to adjust public investments downwards. As a percentage of GDP, public investments are observed to fall to 3.2 percent from its base value of 4.4 percent (a significantly low rate itself). The effect on fiscal accounts is also emphasized in the ratio of government tax revenues against public debt stock. The aggregate tax revenues fall by almost 2 percentage points; yet, given the cost savings on the interest expenditures, the overall solvency of the public sector remains improved. Thus, the lower interest rate policy is enacted here as an important component of the labor tax reform policy. With a reduced interest burden over the public sector, the CBRT facilitates the fiscal authority to alleviate pressures on the fiscal balances that would have emerged as a result of reduced labor tax revenues. With accelerated growth in GDP and lower interest costs, the fiscal balances improve, with consequent gains in fiscal credibility.12

A general equilibrium assessment of twin-targeting in Turkey

223

At the outset, the trade-offs as suggested by the simulation exercise under EXP-3 seem modest and not severely binding: at a loss of 2.9 percentage point increase of the inflation rate (from 25.3 percent to 28.2 percent), the gains in fiscal credibility and employment are found to be relatively robust. Given that under the macroeconomic adjustments of the experiment the external balances were not strained any further, equilibrium in the foreign exchange market seems to be maintained, as well.

10.4

CONCLUDING COMMENTS AND POLICY DISCUSSION

In this chapter, we reported on the current state of the macroeconomic policy environment in the Turkish economy throughout the 2000s and studied the general equilibrium effects of two widely discussed policy changes in the current context: reduce payroll taxes and reduce the central bank interest rate. The current IMF-led austerity program operates with a dual targeting regime: a primary surplus target in fiscal balances (at 6.5 percent to the GDP); and an inflation targeting central bank whose sole mandate is to maintain price stability. Accordingly both policy questions are analysed within the constraints of the aforementioned dual targets set as outer conditionalities of Turkish macroeconomic decision making. Our policy experiments reveal that the current monetary strategy followed by the CBRT that involves heavy reliance on foreign capital inflows along with a relatively high real rate of interest, is effective in bringing inflation down, yet it suffers from increased cost of interest burden to the public sector and strains fiscal credibility. It also leads to excessive foreign indebtedness with increased external fragility. In the medium to long run the increased fiscal and external fragilities along with a persistent and high unemployment manifest a severe impasse, whose resolution will likely lead to onerous adjustments in the labor markets and the real sector. Against this background, we utilized the CGE model to search for applicable alternative policy regimes starting from the immediate short run. Our results indicate that a heterodox policy of (1) reducing the central bank’s interest rates along with (2) lowering the (payroll) tax burden in the labor markets offers a viable environment in the short run, with accelerated growth and improved employment outcomes. The first arm of the policy, viz. reduction of the CBRT interest rate, is important to facilitate the improvement in fiscal balance (and fiscal credibility) at a time when tax monies from labor taxes are expected to be reduced. The second critical element is lowering of the tax burden on employers. With lower payroll taxes levied on employment in production, the employers are led to increase employment

224

Beyond inflation targeting

demand (and also most probably be more willing to employ ‘formal’ labor, and reduce the unrecorded activities along with informalization of the labor markets; issues that our model is not well-equipped to address). With increased credibility of the public sector and lower rates of unemployment, the returns to the heterodox policy reform agenda are quite benign. However, all these come at visible opportunity costs; in particular on the inflation side. As lower interest rates boost domestic investment expenditures and the domestic economic activity is revived due to expanded employment, inflationary pressures accumulate in the commodity and financial markets. It is not so clear at the outset how tolerant would the CBRT be to the acceleration in inflation. Even though our results are quite modest on the pace of both realized and expected rates of inflation, it is clearly an important constraint that merits close observation in the Turkish macroeconomic environment. It is, in fact, mainly for this reason that we maintain some of the key features of the current austerity program with respect to expectations management in the short run. Of particular importance among these is the signaling effect of the primary surplus target. The policy environment of EXP-3 sets the fiscal balances with the programmed target of 6.5 percent primary surplus ratio to the GNP, rather than proposing a drastic break away from it. In fact, with a proper emphasis on dynamics, a direct case can clearly be proposed to stimulate domestic investment expenditures with a policy of lower interest rates, and advocating a fiscal policy of high public investments towards enhancing human capital formation and social infrastructure. Rather than cutting public investments on health, education and social infrastructure, a case can be forwarded to disregard the rising public sector borrowing requirement to GDP ratio in the short run, and implement a fiscal policy to maintain a level of household income capable of addressing the tasks of accumulating human capital. Yet, given the short-run framework of our current modeling framework, we choose to abstain from making ad hoc statements regarding the dynamic consequences of such a policy environment, an issue that had been dealt with elsewhere more effectively (see, for example, Gibson, 2005; ISSA, 2006; Voyvoda, 2003; Voyvoda and Yeldan, 2005). Above all, our simulation experiments clearly underscore the importance of maintaining an integrated and coherent policy framework between the monetary and fiscal authorities. Given the acuteness of the perceived dilemmas on disinflation and fiscal credibility, the resolution of the current impasse will surely necessitate a more tolerant view over the programmed targets (on both inflation and the primary surplus ratio) as well as a coherent and a mutually supportive macro policy design. Furthermore, there is a clear case for a acute need to design viable policies to diminish the

A general equilibrium assessment of twin-targeting in Turkey

225

exposure of the domestic economy (in particular of the financial markets) to short-term, speculative foreign capital. This, in turn, may necessitate implementation of capital management techniques to gear inflows towards longer maturities.

NOTES 1.

2. 3.

4. 5. 6. 7. 8. 9.

10. 11.

12.

Author names are in alphabetical order and do not necessarily reflect authorship seniority. We are indebted to Korkut Boratav, Yilmaz Akyüz, Jerry Epstein, Bill Gibson and to the members of the Independent Social Scientists’ Alliance for their valuable comments and suggestions on previous versions of the chapter. Previous versions of the chapter were presented at the Istanbul Conference of the EcoMod (June 2005); the 9th Congress of the Turkish Social Sciences Association (December 2005, Ankara); the Ankara congress of the Turkish Economics Association (September 2006); and seminars at Bilkent, METU, Bogazici, Utah, Massachusetts-Amherst, Connecticut and the central bank of Turkey. Research for this chapter was completed when Yeldan was a visiting Fulbright scholar at the University of Massachusetts-Amherst for which he acknowledges the generous support of the J. William Fulbright Foreign Scholarship Board and the hospitality of the Political Economy Research Institute at the University of Massachusetts-Amherst. Needless to mention, the views expressed in the chapter are solely those of the authors and do not implicate in any way the institutions mentioned above. That is, balance on non-interest expenditures and aggregate public revenues. The primary surplus target of the central administration budget was set 5 percent to the gross national product. Further institutional details of the central bank’s inflation targeting framework can be found in the December 2005 document, ‘General framework of inflation targeting regime and monetary and exchange rate policy for 2006’, accessed December 2006 at: www.tcmb.gov.tr/yeni/announce/2005/ANO2005-45.pdf. http://www.maliye.gov.tr. Measured in 2002 producer prices. If the PPP-correction is calculated in 2000 prices, the revised debt to GNP ratio reaches to 82.3 percent. See, for example, UNCTAD, Trade and Development Report (2002 and 2003), New York and Geneva. http://www.peri.umass.edu/Alternatives-to.382.0.html. By allowing households to hold foreign currency denominated deposits in the domestic banking system, we try to represent the high level of dollarized liabilities in the Turkish financial system (see Table 10.1). Both residents’ portfolio investments abroad, PFIH, and non-residents’ portfolio investments at home, PFIROW, are incorporated in the model in order to capture any real-economy effects of these ‘speculative’ means, which we believe are important in understanding the growth pattern of the Turkish economy in the last decade. See, for example, Institute for International Economics, http://www.iie.com. It has to be noted, as a reminder, that the current rate of interest set by the CBRT is already significantly high in real terms. Maintaining high real rates of interest was but one of the discretionary measures of the CBRT in an attempt to reduce inflation by curtailing domestic demand expansion, as well as to sustain the inflow of foreign capital to cover the widening current account deficit. Note, however, that this comparison is valid against the base run. One witnesses a slight loss in fiscal credibility relative to EXP-2.

226

Beyond inflation targeting

REFERENCES Agénor, P.R., H.T. Jensen, M. Verghis and E. Yeldan (2006), ‘Disinflation, fiscal sustainability, and labor market adjustment in Turkey’, in Richard Agénor, A. Izquierdo and H.T. Jensen (eds), Adjustment Policies, Poverty and Unemployment: The IMMPA Framework, Oxford: Blackwell Publishing, Chapter 7, pp. 383–456. Ercan, H. and A. Tansel (2006), ‘How to approach the challenge of reconciling labor flexibility with job security and social cohesion in Turkey’, in Reconciling Labour Flexibility with Social Cohesion: Facing the Challenge, Strasbourg: Council of Europe Publishing. Gibson, B. (2005), ‘The transition to a globalized economy: poverty, human capital and the informal sector in a structuralist CGE model’, Journal of Development Economics, 78, 60–94. ISSA (Independent Social Scientists Alliance) (2006), Turkey and the IMF: Macroeconomic Policy, Patterns of Growth and Persistent Fragilities, Penang, Malaysia: Third World Development Network. Tunali, İ. (2003), ‘Background study on the labour market and employment in Turkey’, paper prepared for the European Training Foundation, Turin, Italy, June. Voyvoda E. (2003), ‘Alternatives in debt management: investigation of Turkish debt in an overlapping generations general equilibrium framework’, unpublished PhD thesis, Bilkent University. Voyvoda E. and E. Yeldan (2005), ‘IMF programs, fiscal policy and growth: investigation of macroeconomic alternatives in an OLG model of growth for Turkey’, Comparative Economic Studies, 47, 41–79. World Bank (2000), ‘Turkey – country economic memorandum – structural reforms for sustainable growth’, vols. I and II, report no. 20657TU, September, Washington, DC.

11.

Employment targeting central bank policy: a policy proposal for South Africa Gerald Epstein1

11.1

INTRODUCTION

The story of South Africa’s struggle with apartheid is a dramatic and heroic one, a story that has culminated in a hugely successful political transition from a minority dominated, authoritarian and repressive government organized along racial lines before 1994, to an inclusive, democratic government, in which the majority rules, subject to important protections for minority rights. The story of South Africa’s economic transition, on the other hand, is not such a rosy one. Income and wealth are still highly unequally distributed, largely along ethnic lines with the white population still commanding the bulk of the national income, and still controlling the bulk of the national wealth. Economic growth has been moderate during most of the period since the creation of the new South Africa in 1994, running at an average of 3.4 percent and unemployment rates are hovering somewhere between 25 percent and 36 percent depending on exactly how one counts. Among the black population, in 2006 unemployment was over 30 percent, while among the white population it was less than 5 percent. Such problematic economic trends are in no small part due to the highly unequal economy inherited from the past. But some of the difficulties with the economic transition have been due to the ideologies and the policies, including macroeconomic policies, chosen by the new government in the last decade or so. The government has adopted many pages from the neoliberal play book, including financial liberalization, capital account liberalization, austere fiscal policy and, most relevant to this chapter, formal inflation targeting. These policies have achieved certain gains, but have done little to reduce unemployment and generate more economic equality. Many in South Africa are looking for some alternatives to these policies. 227

228

Beyond inflation targeting

While the government is still strongly committed to inflation targeting, recognition of its limitations are becoming more widespread. In this chapter I present employment targeting (ET) as an alternative to inflation targeting monetary policy framework for South Africa. It is a framework for monetary policy which attempts to incorporate some of the advantages normally claimed for a targeting framework – namely enhancing transparency and accountability – while focusing the goals of monetary policy more directly on critical macroeconomic problems facing the South African economy, namely, employment. Of course, no government, including that of South Africa, can ignore the inflationary impacts of policy, and any ET approach will have to include a goal of stabilizing inflation at sustainable levels.2 As will be seen as the argument develops, reorienting monetary policy toward generating employment will require that monetary policy redevelop and utilize a multiplicity of monetary policy and credit tools, including tools for credit allocation and capital management. While these policies had been in the toolkit of many central banks for decades, they have recently fallen out of favor, or have been made impotent by financial liberalization or legal changes (Epstein, 2007). So one will find a bit of ‘going back to the future’ in some of the proposals found in this chapter. But, as there is no simple return to the past, the ET monetary policy framework for South Africa takes into account important changes at both the national and international levels, especially in the operations of financial markets, which must be accommodated in any policy framework.

11.2

INFLATION TARGETING IN SOUTH AFRICA IN MACROECONOMIC CONTEXT

At the time of the 1994 victory in the struggle against apartheid a major debate ensued over the future direction of economic policy, which culminated in the adoption of the Department of Finance’s macroeconomic strategy, the ‘Growth, Employment and Redistribution’ (GEAR) policy. GEAR was similar to the neoliberal macroeconomic frameworks developed by the International Monetary Fund (IMF), and it is based on similar premises: the key to achieving employment growth and a rapid increase in living standards is to win the confidence of both domestic and foreign investors, by engaging in fiscal austerity and financial liberalization, among other policies. With respect to monetary policy, more specifically, soon after being elected in 1994, the South African government decided that price stability should be the central concern of monetary policy. In 1998 the Reserve

A policy proposal for South Africa

Table 11.1

Comparison of GEAR projections and actual outcomes (1996–2000 averages)

GDP Growth (%) Inflation (CPI) (%) Current Account Deficit (% GDP) Employment Growth (%) Non-Gold Exports Fiscal Deficit (% GDP) Source:

229

GEAR projections

Actual outcomes

4.2 8.2 2.4 2.9 8.4 3.7

2.8 6.7 1.1 0.7 7.1 3.2

Du Plessis and Smit (2005).

Bank adopted, for the first time, an informal inflation target range of 1–5 percent. Then in February 2002 the Reserve Bank adopted a formal inflation targeting approach. Under inflation targeting the Ministry of Finance establishes the target range and the reserve bank chooses the instruments to achieve the target. It chose a target of 3–6 percent rate of increase of CPI, which is a consumer price index excluding interest costs on mortgage rates and the Reserve Bank chose the repo rate as the main monetary policy instrument. Looking back over the initial years, however, one can see that the GEAR policies were implemented with mixed results (Table 11.1). The GEAR policy ‘delivered on its promises’ in the early years with respect to fiscal deficits, inflation and the current account, but failed in terms of economic growth and employment generation. Real GDP growth has been higher since that time but still not large enough to substantially reduce the rate of unemployment. Figure 11.1 shows the rate of real GDP growth before and after the transition to democratic rule. While GDP growth has gone up since the conflict-ridden and sanctions-laden period between 1984 and 1994, the 3.45 percent per year average growth rate is still relatively modest, though in the most recent years, largely due to the global commodity boom, growth rates have been above 4.5 percent. Given this relatively low per-capita growth in real GDP, it should not be surprising that employment performance has been fairly weak. Table 11.2 shows the official unemployment rate broken down by population group in 2001 and 2007. The contrast among the population groups is striking with African unemployment, by the official measure showing 26.8 percent unemployment in 2007, in contrast to the overall white rate of 5.8 percent. Both rates have come down somewhat since 2001 due again to the higher than average rate of economic growth during the latter years

230

Beyond inflation targeting 6 After democratic transition 5 4

Percent

3 2 1 0 –1 –2

Average growth: 3.45

Average growth: 1.03

–3 84 Source:

86

88

90

94

96

98

00

02

04

06

Reserve Bank of South Africa.

Figure 11.1 Table 11.2

Real GDP growth Official unemployment, September 2001 and September 2007 (percentage of labor force)

African Coloured Indian White Average Source:

92

September 2001

September 2007

35.7 21.2 18.8 5.8 29.4

26.8 20.6 8.2 3.8 22.7

Statistics South Africa.

of that period. This shows that in all likelihood higher economic growth does, in fact, lead to lower unemployment rates among all population groups. Table 11.3 shows the official unemployment rate in September 2001 and September 2007, broken down by gender and population group. These

A policy proposal for South Africa

Table 11.3

Male Black African Coloured Indian/Asian White Average Female Black African Coloured Indian/Asian White Average

231

Official unemployment by gender and ethnic group, September 2001 and September 2007 (percentage of labor force) September 2001

September 2007

31.5 19.5 15.7 4.7 25.8

23.3 20.0 7.4 3.5 19.8

40.7 23.1 23.5 7.4 33.8

30.9 21.3 10.2 4.2 26.1

Source: Labour Force Survey, March 2006, March 2007 (Pretoria: Statistics South Africa, September 2006 and 2007).

Table 11.4

Male Female Total

Official unemployment plus discouraged workers, September 2001 and 2007 (percentage of labor force) 2001

2007

33.8 47.0 40.0

31.8 34.9 34.0

Source: Author’s calculations from Labor Force Survey, March 2006 and March 2007 (Pretoria: Statistics South Africa).

show that unemployment rates among women are higher than among men for all population groups, but they too have gone down somewhat in response to higher economic growth rates. These data reflect the official unemployment statistics. Table 11.4 presents data that include ‘discouraged workers’ and therefore may be a more accurate measure of unemployment. They indicate that the ‘unofficial’ unemployment rate might be at 34 percent overall and over 35 percent for women in 2007. As mentioned earlier, since 2002, if not before, the Reserve Bank of South Africa has followed an inflation targeting regime for monetary policy; the policy has been directed, most recently toward keep CPI

232

Beyond inflation targeting

12 CPI inflation (Monthly, on an annual basis) 10

Percent

8

6

4

Inflation target range

2 2001 Source:

2002

2003

2004

2005

2006

Reserve Bank of South Africa.

Figure 11.2

CPI inflation

inflation within a target range of 3–6 percent. Indeed, inflation has been in that range since mid 2003 or so (Figure 11.2). Whether it has caused the decline is more difficult to answer. As Ball and Sheridan (2003) point out, however, inflation fell in many countries over the last decade, independently of whether the countries were adopting inflation targeting or not. What is clear is that this inflation targeting regime has led for most of the period to very high rates of real interest rates which, in turn, have contributed to relatively modest economic growth and, until the recent period of the global commodity boom, to anemic growth in employment.

11.3

AN EMPLOYMENT TARGETING FRAMEWORK FOR SOUTH AFRICA

In light of the problems with the inflation targeting regime in South Africa, and especially given South Africa’s major problems with unemployment and underemployment, an alternative framework for monetary policy is

A policy proposal for South Africa

233

badly needed. Here I propose a ‘real targeting’ framework for monetary policy, and then describe a specific type of real targeting framework that applies to South Africa, namely an ET framework for monetary policy. 11.3.1

A Real Targeting Framework for Monetary Policy

A real targeting framework for monetary policy should adhere to the following principles: ●

●

●

●

●

Context-appropriate monetary policy. Central bank policy goals and operating procedures must be based on the structure and needs of South Africa: no generic, one-size fits all approach is likely to be appropriate to every situation. Real economy-oriented monetary policy. Policy makers should recognize that very high rates of inflation can have significant costs, but that short of that, policy must also be oriented toward promoting investment, raising employment growth and reducing unemployment. Transparency and accountability. Taking a leaf from the targeting approach, central banks should be made more accountable to the public by making their objectives and approaches more transparent. They should tell the public what their targets for monetary policy are, describe the economic assumptions underlying their plans to reach those targets and, if they do not reach them, explain why while describing their plans for achieving them in the next period. And, most importantly, the goals of the central bank should be determined by a democratic process. Policy flexibility. A fundamental fact is that there is a great deal of uncertainty concerning the underlying structure of the economy and about the nature of national and international shocks at any particular time. Hence, adherence to any target has to be somewhat flexible. Sufficient tools to reach the targets. Monetary policy has been stripped by financial liberalization and neoliberal conceptions of policy tools needed to achieve a multiplicity of targets: credit allocation techniques, which used to be very important components of central bank policy in many countries, have been systematically eliminated, capital controls (capital management techniques) likewise, which in the past have been part of the arsenal of monetary policy, have been dramatically liberalized. These techniques need to be revitalized and modernized so that, where and when appropriate, they can be used as instruments to help central banks reach their goals.

234 ●

Beyond inflation targeting

Supporting institutions. Central bank policy is no panacea. Other important supporting institutions are also required, including strong tax institutions to enable the government to raise the revenue it needs to fund important public investments, and public financial institutions to channel credit in support of productive investment.

In addition to the direct advantages of a real targeting framework, there are several extremely important indirect advantages of the ET framework which themselves will contribute in a crucial way to the framework’s success. These include the accumulation of new knowledge about the connections between monetary policy and employment, and the implementation of new policies to generate more employment and economic growth. Both of these positive outcomes will be facilitated by focusing the attention of the central bank, with its enormous human and financial resources, on the key issue of employment growth. Making employment growth a target of monetary policy might also induce a profound shift in the attitude and the activities of the central bank. It might begin to assign its economists to study the relationships between monetary policy and employment growth; to study what monetary policy instruments are best used to achieve employment growth; and to organize conferences on employment growth and monetary policy. It might even give promotions and more resources to members of its staff that make breakthroughs in the understanding of these connections. In the case of South Africa, it will lead the Reserve Bank to link up with others outside the bank who have knowledge and experience with respect to employment generation and its relationship to financial variables. It could also lead the bank to design new programs to help it reach its targets.

11.4

EMPLOYMENT TARGETING MONETARY POLICY IN SOUTH AFRICA

In this section I describe the outlines of an ET policy for the Reserve Bank of South Africa. This policy has been developed in the context of an overall ET program for the South African economy developed by economists at the Political Economy Research Institute.3 This larger ET program includes fiscal stimulus, public credit allocation and development banking, capital management techniques, mechanisms of inflation control, tax reforms, including mechanisms such as an enhanced securities transactions tax to raise more revenue to finance employment policies, and

A policy proposal for South Africa

235

other sectoral policies, for example, anti-trust and competition policy to correct infrastructure bottlenecks preventing more growth and employment generation. In this chapter I focus on the role of the Reserve Bank and monetary policy within this ET framework. 11.4.1

The Role of Monetary Policy and the Reserve Bank in this ET Program

The basic structure of the proposed central policy is as follows: the Reserve Bank would set its interest rates to achieve an overall real growth rate consistent with the plan which has an employment target at its core. As part of its mandate, the Reserve Bank would try to reach an inflation constraint that is mutually decided upon as part of the overall program. In addition, the Reserve Bank would manage some of the credit allocation programs that are part of the targeted components of the overall ET macroeconomic policy. Finally, the Reserve Bank would manage the capital account as needed to maintain the exchange rate and exchange rate stability needed to implement the program. If the ET policy is to be fully successful, in the long run it will help considerably if the Reserve Bank makes certain institutional commitments. These might include some of the following. The Reserve Bank will launch a set of research programs both within the bank and outside to improve understanding of the relationship between monetary policy tools and employment growth, both in the formal sector and informal sector. The Reserve Bank will work with financial institutions, including development banks and major commercial banks, as well as smaller financial institutions, to develop instruments and programs to facilitate the allocation of credit for effective employment generating activities. If the rates of economic growth achieved through more expansionary monetary policy do not generate as much employment as projected, the Reserve Bank is committed to finding new policies and mechanisms, along with the rest of the government, to achieve these targets. 11.4.2

Targets and Instruments and All That

In the terminology of targets and instruments used in the monetary policy literature, in this framework the ultimate target of monetary policy is employment growth, while the intermediate targets are real GDP growth, inflation and exchange rates. The instruments are short-term interest rates (the repo rate in the case of current practice in South Africa), capital management techniques and credit allocation policies.

236

11.5

Beyond inflation targeting

SIMULATING THE IMPACT OF MONETARY POLICY: A VAR-BASED SIMULATION MODEL

In this section I present a simple exercise to show the impact of an expansionary monetary policy as part of an ET framework. Following this section, I consider the other instruments and intermediate targets to be utilized by the Reserve Bank in its ET policy. 11.5.1

Monetary Policy Experiments

In this section I present some simulations based on a series of simple structural vector auto-regression (VAR) models I built of the South African economy. These simulations are designed to estimate the impacts of employment targeted monetary policy on real GDP growth – a key determinant of employment generation – and on inflation and exchange rate stability which are crucial other variables of likely concern to the central bank. VAR’s-based simulation models are widely used in monetary policy analysis (see, for example, Aron and Muellbauer, 2002; Bernanke et al., 1997; Gallardo and Ros, 2008). This approach uses a minimum number of assumptions about the structure of the economy to estimate a small, simplified model of the macroeconomy. Based on that estimation, we simulate the model with changes in various policy tools and estimate the impact of those changes in policy on the economy. I do this by estimating the difference between the ‘baseline’ path of economic variables, such as inflation and economic growth, and the path these variables take under the new (hypothetical) settings of the monetary policy tools. The difference between these is taken to be the ‘impact’ of the policy change on the economy. It is important to point out that, despite their widespread use, these VAR models have many drawbacks. They are highly simplified models of very complex economies; and in situations where there are either short data series, or a large amount of structural change, these models may not be able to form good estimates of economic relationships or forecast with a great deal of accuracy. Still, for our purposes, we hope that these results can give us some ball-park idea of the impact of the approach I am proposing. This analysis addresses the impact of changes in the South African Reserve Bank monetary policy’s rule on exchange rates, inflation and economic growth. The goal is to estimate the quantitative impact of different interest rate settings on economic growth, inflation and exchange rate variability, and indirectly, then, on employment. It incorporates some important exogenous variables as well.

A policy proposal for South Africa

237

Before describing the estimates and the simulations, I first discuss our data variables and sources. 11.5.2

Data and Preliminary Statistical Tests

All data are either originally quarterly data, or have been transformed into quarterly data from monthly data. ●

●

●

● ● ●

●

Prime rate: the ‘prime rate’ is the ‘prime lending rate’. It tracks closely the changes in the repurchase (or ‘repo’) rate, which is the main tool of monetary policy. The reason we use the prime rate rather than the repo rate is that it has a longer data series.4 Exchange rate: the nominal rate relative to the US dollar. Our measure is the four quarter rate of change of the rand, measured so that an increase means an increase in the rate of depreciation of the rand.5 Inflation rate: we use the CPI variable, as calculated by Aron and Mullbauer (2002). This is the consumer price level, excluding mortgage interest. We use the four quarter rate of inflation.6 GDP growth: the four quarter rate of growth of real GDP. Source: South African Reserve Bank.7 US Tbill: the three-month US Treasury bill rate.8 Private credit: the four quarter rate of growth of private credit created by all monetary institutions; monthly data transformed to quarterly data. Source: South African Reserve Bank.9 Terms of trade: four quarter change in the terms of trade. Source: South African Reserve Bank.10

11.5.3

Stationarity Tests on Variables

All variables were tested for stationarity using the Augmented DickeyFuller tests. The results of these tests are presented in Table 11.5. For the estimates that follow, we report on results using the growth variable (of real GDP). We have done the estimates with de-trended GDP growth as well and the results are roughly the same. We report on the nonde-trended variable because of greater ease of interpretation. 11.5.4

VAR Estimates of Monetary Policy

I estimate a VAR model with four endogenous variables (prime rate, exchange rate change, inflation and growth) and one exogenous variable, the US Treasury bill rate (US Tbill), over the period 1989, first quarter

238

Beyond inflation targeting

Table 11.5

Stationarity tests

Variable

Stationary (significance level)

Intercept

Intercept and Trend

Exchange Rate Change Credit Change Growth Inflation Prime US Tbill

Yes (5%) Yes (5%) Yes (1%) Yes I (1) (1%) Yes I (1) (1%) Yes I (1) (1%)

Yes Yes No Yes Yes Yes

No No Yes No No

(1989q1) to 2004, fourth quarter (2004q4), with four quarter lags. I choose 1989q1 as a starting date because of concerns of significant structural breaks at that time. I then estimated the impulse response function using the Choleski de-composition, with the following ordering (prime, change in exchange rate, inflation, growth). I tried the results with different orderings and the results did not seem very sensitive to these changes. A positive shock in the prime reduces economic growth, has a modest, cyclical impact on inflation and is associated with increased variability in the exchange rate. Figure 11.3 presents the impulse response functions from this model. Note that in this VAR estimate, an increase in the prime rate has a stable impact on inflation, but it generates an increase in the inflation rate. This probably reflects the so-called ‘cost channel’ or ‘Patman Effect’, named after US Congressman Wright Patman, who warned that interest rate increases were inflationary. 5.5

Simulation Results

The next step is to transform this VAR model into a simulation model to be used for monetary policy experiments. The goal of the simulation exercise is to estimate the impacts of a medium- to long-term decline in the prime lending rate on exchange rate variability, inflation and economic growth. The basic idea is to estimate the likely impacts of a looser monetary policy by the Reserve Bank which attempts to expand real GDP growth in order to generate more employment, as part of an ET policy. In order to estimate these impacts, I undertook the following steps. I first solved the estimated VAR model using a dynamic simulation, over the period 1994q1–2004q4 and call the results of this dynamic simulation the ‘baseline’ results. These baseline estimates are then used in combination with data generated by ‘policy experiments’ to estimate

A policy proposal for South Africa Accumulated Response of Prime to Prime

12

Accumulated Response of Exchange Rate to Prime

30

10

20

8

10

6

0

4

–10

2

–20

0

239

–30 1

2

3

4

5

6

7

8

9

10

Accumulated Response of Inflation to Prime

5

1

2

4

5

6

7

8

9

10

9

10

Accumulated Response of Growth to Prime

1

4

3

0

3

–1

2 1

–2

0

–3

–1 –2

–4

–3

–5

–4 1

2

3

Figure 11.3

4

5

6

7

8

9

10

1

2

3

4

5

6

7

8

Accumulated response to Cholesky one SD innovations ± 2 SE

the impacts of changing the prime lending rate in the steps described below. As seen in Figure 11.4, the baseline estimates, when compared to the actual data, capture basic trends in the data, for the most part, but they do not do a very good job of tracking all the turning points in the data. Unfortunately, this is not an uncommon problem in such models which have a limited number of degrees of freedom. The next step is to create alternative scenarios in which the prime lending rate is adjusted to reflect an employment-oriented monetary policy. Here I report on one scenario: lowering the prime rate so that it remains 4 percentage points below the baseline and keeping it there for five years. Table 11.6 presents the numerical summary of the main effects of this policy experiment. Figure 11.5 describes the impacts by showing the difference from the baseline. The results show that GDP growth goes up on average by about

240

Beyond inflation targeting Exchange rate change

40

Growth 6

30

5

20 4

10

3

0 –10 –20 –30

Actual Growth (baseline)

2 Exchange rate change Exchange rate (baseline)

1 0

–40 94 95 96 97 98 99 00 01 02 03 04

94 95 96 97 98 99 00 01 02 03 04

Inflation 11

Prime rate 28

10 24

9 8

20

7 16

6 5 4

Actual Inflation (baseline)

12

Prime rate Prime rate (baseline)

8

3 94 95 96 97 98 99 00 01 02 03 04

94 95 96 97 98 99 00 01 02 03 04

US Tbill 7 6 5 4 3 2 1

Actual Baseline

0 94 95 96 97 98 99 00 01 02 03 04

Figure 11.4

Baseline estimate versus actuals 1994Q1–2004q4

6 percent for the period 2001–04 (about 0.5 on average for the whole fiveyear period), inflation increases by about 1 percentage point on average, while the exchange rate changes become more variable. Table 11.6 summarizes these results. In this simulation on average economic growth is raised by 0.6 of 1 percent, with only modest increases in inflation and exchange rate instability. The exchange rate instability

A policy proposal for South Africa

Table 11.6

241

Impact of 4 percent point decline in prime rate

Exchange Rate Change Growth Inflation Prime Rate US Tbill

2000

2001

2002

2003

2004

6.16 0.22 0.12 −4.00 0.00

5.74 0.79 1.29 −4.00 0.00

−1.91 0.44 0.96 −4.00 0.00

0.35 0.61 0.76 −4.00 0.00

1.29 0.55 0.85 −4.00 0.00

created is relatively small given the swings in exchange rates that South Africa has experienced in the last decade. And the increase in inflation stays well within the single digits. This increase in the rate of economic growth is itself modest, but accumulated over time, it could have a significant impact on employment growth. It should be obvious that, by itself, lowering interest rates will not ‘solve’ South Africa’s unemployment problem. The Reserve Bank, along with other economic institutions, will need to participate in more aggressive and creative initiatives to enhance employment opportunities.

11.6

CREDIT ALLOCATION AND CAPITAL MANAGEMENT POLICIES TO SUPPORT EMPLOYMENT GENERATION

11.6.1

Credit Allocation Policies

As the empirical exercise in the previous section suggests, manipulation of interest rates will certainly have some positive impact on employment in South Africa, but ultimately, other tools will be needed to encourage employment at sufficient levels. Credit allocation tools can be an important set of policies to contribute to employment growth in South Africa if they are properly implemented and monitored. While some of this operation might occur outside of the purview of the South African Reserve Bank, the bank can and should play a significant role in this area. (See Pollin et al. (2006) for much more detail on credit allocation tools in South Africa.) In doing so, the Reserve Bank has many examples to draw on both from its own history and from the central banks of many highly successful developed and developing countries, including South Korea, Japan, France and the Taiwan Province of China and the People’s Republic of China (Amsden, 2001; Epstein, 2007). Still, experiences are not simply transferable in a ‘turnkey’ way, and even the earlier experience of directed

242

Beyond inflation targeting Exchange rate change

Growth

12

0.9

10

0.8

8

0.7 0.6

6

0.5

4

0.4

2

0.3

0

0.2

–2

0.1

–4

0.0 2000

2001

2002

2003

2004

Inflation

1.6

2000

–3.8

0.8

–3.9

0.4

–4.0

0.0

–4.1

–0.4

2002

2003

2004

2003

2004

Prime rate

–3.7

1.2

2001

–4.2 2000

2001

2002

2003

2004

2003

2004

2000

2001

2002

US Tbill

1.0 0.5 0.0 –0.5 –1.0 2000

Figure 11.5

2001

2002

Impact of 4 percent point drop in prime relative to baseline, 2000Q1–2004Q4

credited under the apartheid system is not something the current regime would want to replicate. So, the Reserve Bank, while learning from past experience, will need to develop credit allocation mechanisms to fit the current situation. These could include concessionary loans and increased capitalization for development banks, loan guarantees and asset-based reserve requirements.

A policy proposal for South Africa

243

To avoid corruption and misuse of these tools, careful monitoring and sharp incentives need to be put in place. 11.6.2

Capital Management Techniques

Another vulnerability of these programs is that, in a financially integrated economy, financial arbitrage and capital flight might undermine them. As just discussed, an expansionary monetary policy might induce unstable exchange rates and overshooting as well. Capital controls – or capital management techniques – have been widely used to manage exchange rates and create running room for monetary policy (see, for example, Epstein et al., 2005). South Africa has a long history of capital management techniques, including exchange and capital controls, dating back to at least 1939 (see Bruce-Brand (2002) for a brief history and overview). The current governing legislation was set out in 1961. Their application has ebbed and flowed over the years. With the reintegration of South Africa into the world following the abolishment of the apartheid government in 1994, after considerable discussion, the government decided to progressively liberalize the capital and exchange controls. Capital management techniques can help countries achieve more autonomy in the making of monetary policy, including reducing the variability and misalignment of the exchange rates, though there is still a great deal of debate on this point. A large literature has attempted to assess the ‘effectiveness’ of capital controls (see, for example, Lee and Jayadev (2005), Henry (2006) and Epstein et al. (2005), among many others, for surveys). Studies which have analysed the ‘effectiveness’ of capital controls have looked primarily at two things: first, at their ability to maintain a ‘wedge’ between domestic interest rates and foreign rates; or, second, at their ability to avoid ‘currency crises’, that is, either largescale changes in exchange rates or the forcing of countries off of a peg through reserve loss; more recent studies have looked at the ability of controls (especially on inflows) to prevent overvalued currencies and currency instability and the ability of controls to provide autonomy for central bank expansion. Many studies find that capital controls do seem to be effective in insulating domestic interest rates and exchange rates from international factors, but usually only to a moderate degree, especially if the time period is a long one; and only if the overall set of macroeconomic policies are internally consistent. More recent evidence suggests, however, that capital controls can help to avoid financial instability is stronger (Epstein et al., 2005). Countries that had controls on outflows and/or inflows were more likely

244

Beyond inflation targeting

to be able to avoid the contagion from the ‘Asian financial crisis’. In the long literature on controls on inflows for Chile, there is evidence that Chile was able to avoid an overvalued exchange rate and tilt toward a longer maturity structure of borrowing. In the case of Malaysia (Kaplan and Rodrik, 2002) controls on outflows temporarily allowed more expansionary policy. Hence, the evidence can be read that, properly implemented, controls can reduce instability and somewhat enhance macroeconomic autonomy, at least to a limited extent. South Africa has a lot of experience in implementing capital controls and should build on that experience to tailor these capital management techniques to modern circumstances.

11.7

CONCLUSION

This chapter has tried to develop an ET monetary policy framework for South Africa. I have emphasized that while monetary policy by itself cannot solve the unemployment problem in South Africa, it must make a contribution to doing so, and that simply stabilizing inflation will not do the trick. An ET framework will work best if it is part of an overall ET macroeconomic strategy and if the central bank, as an institution, is committed to cooperating with the government to implement such a policy. I have also argued that the central bank itself must bring to bear a coordinated set of tools, including interest rate policy, credit allocation policy, capital management techniques, and institutional development, to make an ET approach effective.

NOTES 1.

2. 3. 4. 5. 6. 7. 8. 9. 10.

Thanks to K.S. Jomo Terry McKinley, Butch Montes and Jose Antonio Ocampo for their encouragement, to my colleagues at PERI, Bob Pollin, James Heintz and Leonce Ndikumana for their indispensable contributions on our larger project on employment targeting in South Africa and to members of the ‘Alternatives to Inflation Targeting’ group for their suggestions. A final thanks to my project co-organizer, Erinc Yeldan for his contributions. All errors, of course, are mine. See Robert Pollin et al. (2006) for more details on a general macroeconomic ET framework for South Africa of which this central bank policy can be seen as a part. Ibid. International Financial Statistics (IFS), lending rate. IFS. http://www.csae.ox.ac.uk/resprogs/smmsae/datasets.html. http://www.reservebank.co.za/. IFS. http://www.reservebank.co.za/. Ibid.

A policy proposal for South Africa

245

REFERENCES Amsden, A. (2001), The Rise of the Rest: Challenges to the West from LateIndustrializing Economies, Oxford: Oxford University Press. Aron, J. and J. Muellbauer (2002), ‘Interest rate effects on output: evidence from a GDP forecasting model for South Africa’, IMF Staff Papers, 49 (November), 185–213, accessed at (www.csae.ox.ac.uk/conferences/2002-UPaGiSSA/papers/ Aron-csae2002.pdf); non-technical summary at http://www.csae.ox.ac.uk/resprogs/smmsae/nontechs/nontech07.html. Ball, L. and N. Sheridan (2003), ‘Does inflation targeting matter?’, International Monetary Fund working paper no. 03/129. Bernanke, B. et al. (1997), ‘Systematic monetary policy and the effects of oil price shocks’, Brookings Papers on Economic Activity, 1, 91–157. Bernanke, B.S., T. Laubach, A.S. Posen and F.S. Mishkin (1999), Inflation Targeting: Lessons from the International Experience, Princeton, NJ: Princeton University Press. Bruce-Brand, A.M. (2002), ‘Overview of exchange controls in South Africa’, statement to the Commision of Inquiry into the Rapid Depreciation of the Exchange Rate of the Rand, Reserve Bank of South Africa. Dooley, M.P. (1995), ‘A survey of academic literature on controls over international capital transactions’, National Bureau for Economic Research working paper no. 5352. Du Plessis, S. and B. Smit (2005), ‘Economic growth in South Africa since 1994’, conference paper for the Economic Policy under Democracy: A 10 year Review Conference, Stellonbosch University, 28-29 October. Epstein, G. (2007), ‘Central banks as agents of development’, in Ha-Joon Chang (ed.), Institutional Change and Economic Development, Helsinki: United Nations University Press, accessed at www.wider.unu.edu/publications/working-papers/ research-papers/2006/on-GB/rp2006-54/_files/78091779622373109/default/rp200654.pdf. Epstein, G., I. Grabel, K.S. Jomo (2005), ‘Capital management techniques in developing countries: an assessment of experiences from the 1990s and lessons for the future’, in Gerald Epstein (ed.), Capital Flight and Capital Controls in Developing Countries, Cheltenham, UK and Northampton, MA, USA: Edward Elgar, pp. 301–33. Farrell, G.N. (2001), ‘Capital controls and the volatility of South African exchange rates’, South African Reserve Bank occasional paper no. 15, July. Galindo, L.M. and J. Ros (2008), ‘Alternatives to inflation targeting in Mexico’, International Review of Applied Economics, forthcoming. IMF (2003), South Africa: Selected Issues, International Monetary Fund country report no. 03/18, January. International Monetary Fund (IMF) (2004a), International Financial Statistics, CD ROM. IMF (2004b), South Africa: Selected Issues, International Monetary Fund country report no. 04/379, December. Jayadev, A. and K.K. Lee (2004), ‘The effects of capital account liberalization on growth and the labor share of income: reviewing and extending the crosscountry evidence’, in Gerald Epstein (ed), Capital Flight and Capital Controls in Developing Countries, Cheltenham, UK and Northampton, MA, USA: Edward Elgar, pp. 15–57.

246

Beyond inflation targeting

Kaplan, E. and D. Rodrik (2002), ‘Did the Malaysian capital controls work?’, in Sebastian Edwards and Jeffrey A. Frankel (eds), Preventing Currency Crises in Emerging Markets, Chicago, IL: University of Chicago Press, pp. 393–441. Lee, K. and A. Jayadev (2005), ‘Capital account liberalization, growth and the labour share of income: reviewing and extending the cross-country evidence’, in G. Epstein (ed.), Capital Flight and Capital Controls in Developing Countries, Cheltenham, UK and Northampton, MA, USA: Edward Elgar, pp. 15–57. Pollin, R.N., G. Epstein, J. Heintz and L. Ndikumana (2006), ‘An employmenttargeted economic program for South Africa’, United Nations Development Program Poverty Center, Brazil.

A policy proposal for South Africa

APPENDIX 11.A1

247

VARIABLES, DEFINITIONS AND SOURCES FOR VAR MODEL

Variable

Definition

Units

Frequency

Source

Properties

RGDP Growth

Growth rate of real GDP, annualized

Percent

Quarterly

Stationary

Prime

Prime lending rate

Percent

Quarterly

Inflation

The inflation rate based on CPI metropolitan, total Nominal rand/ US dollar exchange rate (an increase is depreciation)

Percent

Quarterly

South African Reserve Bank South African Reserve Bank South African Reserve Bank

Rand per US dollar

Quarterly

Exchange Rate

IMF, International Financial Statistics (IFS)

Stationary

First difference is stationary

Log first difference is stationary

12.

Inflation targeting and the design of monetary policy in India Raghbendra Jha1

12.1

INTRODUCTION

With around 20 central banks adopting it as their basic monetary policy framework, inflation targeting (IT) has emerged as an important monetary policy framework. Over time IT has become more flexible in its interpretation of target and permitted other goals to be included in the basic framework. Central banks have enhanced their communication with their respective publics about their targets and modus operandi. Some authors have argued that for transition economies undergoing sustained financial liberalization and integration in world financial markets, IT is an attractive monetary policy framework. Consequently there is pressure for such economies to adopt IT. This chapter evaluates the case for IT in India. It begins in Section 12.2 by stating the objectives of monetary policy, especially that inflation control cannot be an exclusive concern of monetary policy in a country with mass poverty. An evaluation of the rationale for IT and nuances of implementation are spelt out in Section 12.3. Section 12.4 provides some evidence on the effects of IT in developed and transition economies. Section 12.5 discusses India’s experience with using nominal targets whereas Section 12.6 discusses some recent developments in Indian monetary policy. Section 12.7 reviews some reasons why India is not ready for IT. Section 12.8 shows that even if the Reserve Bank of India (RBI) – India’s central bank – wanted to, it could not pursue IT since the shortterm interest rate (the principal policy tool used to affect inflation in countries using IT) does not have significant effects on inflation. Section 12.9 concludes and sketches the contours of an alternative to an IT policy. A central feature of this policy is the use of capital controls. Appendix 12.A1 revisits the role that such capital controls played in ensuring that India did not get severely affected by the East Asian financial crisis of the late 1990s, even though India’s macroeconomic fundamentals at that time were comparable to those of countries that did experience such a crisis. 248

Inflation targeting and the design of monetary policy in India

Table 12.1 Year

1978–85 1985–90 1990–97 1997–02 1978–02

249

Growth and poverty alleviation in China Annual poverty reduction announced by the government (10 thousand)

Average annual growth rate of GDP per capita (%)

Average annual growth rate of farmers’ consumption level (%)

Average annual growth rate of farmers’ net income per capita (%)

1786 800 500 436 924

8.3 6.2 9.9 7.7 8.1

10.0 2.5 8.0 3.4 5.6

15.1 3.0 5.0 3.8 7.2

Source:

Chinese Statistical Abstract, various issues.

12.2

THE OBJECTIVES OF MONETARY POLICY IN INDIA

An overriding short-term concern of monetary policy is stabilization of the price level. However, India has had a long-standing problem of poverty and its alleviation has to be the cornerstone of the success of any policy, including monetary policy. Higher economic growth, along with some supporting redistributive measures has a crucial role in reducing poverty, even more than governance. Dollar and Kraay (2001) show for a broad cross-section of countries including India, that the incomes of the poorest 20 percent of the population rise in proportion to average income.2 China is an important example of the poverty reducing effects of economic growth (Table 12.1). For almost three decades Chinese per capita GDP has grown at more than 8 percent per annum. Poverty has declined at an average of 9 240 000 persons per year. In contrast India’s growth and poverty reduction record has been less spectacular (Table 12.2). Because of lower growth the reduction in poverty in India has been far lower than in China notwithstanding the fact that inequality in China has grown more sharply than in India (Jha, 2004). India’s national poverty headcount ratio fell only by about 12 percentage points between 1951–52 to 19973 and the rate of poverty reduction was higher in the 1980s than in the reform period, post-1991, indicating that the quality of growth in the

250

Beyond inflation targeting

Table 12.2

GDP and per capita GDP growth in India

Period

GDP growth (%)

1972–82 1982–92 1992–2002 Source:

Aggregate (average annual)

Per capita (average annual)

3.5 5.2 6.0

1.2 3.0 3.9

Kelkar (2004).

1980s was different from that in the 1990s, so that even with lower growth greater reduction in poverty could take place in the 1980s. This trend has recently been reversed. Jha et al. (2006) report that the rural headcount ratio, which is usually higher than the urban headcount ratio, was 20.6 percent in 2004 compared to higher than 26 percent in 1999. Hence there exists scope for poverty reduction through higher growth in India. Furthermore, unemployment has been rising in the post-reform period. In the organized sector employment barely changed between 1991 and 2001; from 1997 it actually fell. Total employment (organized and unorganized) is growing at about 1 percent per annum – half the projected growth rate of the labor force. Consequently, the unemployment rate for males in the rural sector rose from 5.6 percent in 1993–94 to 9 percent in 2004. In addition India also has substantial underemployment. If India is to reduce poverty rapidly, its GDP must grow at 8 percent or more on a sustained basis. Kelkar (2004) argues that growth could accelerate essentially because of a broad series of financial sector reforms, increased globalization and widening and deepening of product and financial markets, and beneficial structural changes – particularly on the supply side. In addition to its ‘surplus labor’ India is set to reap an important demographic dividend as the proportion of population in the working age group (15–64 age bracket) is expected to climb from 60.9 percent in 2000 to over 66 percent by 2030. The labor force is less nutritionally deprived and increasingly literate. Economic theory and international experience indicate that this could lead to sharp rises in labor productivity and an upward shift in the trend rate of growth. However, this labor force must be productively employed for these productivity gains to be realized (Jha, 2005). Low interest rates (to enhance investment) and a slightly undervalued exchange rate with low volatility (to boost exports) are critical to sustaining high growth rates. An appropriate monetary policy must address these requirements.

Inflation targeting and the design of monetary policy in India

12.3

251

EVALUATING THE RATIONALE FOR INFLATION TARGETING

The time inconsistency literature argues that a discretionary policy setting leads to higher inflation without any gains in output (Barro and Gordon, 1983; Kydland and Prescott, 1977). The high welfare cost of inflation raises the appeal of rules-based policy. Under a rules-based regime central banks set explicit values for intermediate targets, which they can control, and which are strongly related to the ultimate goals of monetary policy (viz. stabilization of output and inflation), which cannot be directly controlled. In recent times emerging market economies, including India, have experimented with three nominal targets: exchange rate, money supply growth and inflation.4 The relative advantages/disadvantages of exchange rate and money growth targeting are portrayed in Table 12.3. Bernanke and Mishkin (1997) argue that IT, in contrast to exchange rate targeting but like monetary targeting, enables monetary policy to focus on domestic considerations and to respond to shocks to the domestic economy.5 IT, like exchange targeting and unlike monetary targeting, has the advantage that people easily understand it. Since the central bank has a numerical inflation target, the chances of slipping into a time inconsistency trap are reduced. Svensson (1999), Bernanke and Mishkin (1997) and White (2004) argue that IT is ‘decision making under discretion’ with central banks following a targeting rule which sets interest rates to reduce the deviation between conditional inflation forecast (the intermediate target of monetary policy) and the inflation target to zero over the horizon. In an emerging market economy such as India the problems of monetary management, in general and inflation control, in particular, get compounded by low policy credibility (Calvo and Mishkin, 2003). This can lead to sudden stops of capital inflows, making such countries prone to financial crises. This increases the appeal of rules-based monetary regimes (like IT). Taylor (2002) argues that rules-based policies enhance the anticipation effects of monetary policy. Given less developed financial markets such anticipatory effects are likely to be lower. Yet monetary policy could still have considerable effects through movements of wages and property prices. With an IT regime shocks from the monetary regime may be lower. However, this rationale for IT is incomplete. For instance, Kirsanova et al. (2006) show, in the context of the UK, that an IT regime in itself cannot be considered to be an optimal monetary policy framework and optimal policy response must include terms of trade or exchange rate terms – an

252

Table 12.3

Beyond inflation targeting

Advantages and disadvantages of the nominal anchors of exchange rate targeting and monetary targeting Anchor: Exchange rate targeting

Advantages

1. This fixes the inflation rate for internationally traded goods and thus directly contributes to keeping inflation under control. It is especially useful for sharply reducing inflation in emerging market economies 2. If the exchange rate peg is credible, it anchors inflation expectations to the inflation rate in the anchor country to whose currency it is pegged 3. An exchange rate provides an automatic rule for the conduct of monetary policy that avoids the time-inconsistency problem 4. An exchange rate is simple and direct and, therefore, is well understood by the public

Disadvantages

1. An exchange rate target leads to loss of independent monetary policy (Obstfeld and Rogoff, 1996). Hence the ability of the monetary authorities to respond to shocks is compromised 2. The exchange rate peg may persuade large-scale foreign borrowing. In the case of emerging market economies such loans are invariably denominated in foreign currency. Large accumulation of such loans may lead to a crisis. In most developed countries a devaluation may have little direct effect on the balance sheets (since debts are denominated in home currency) but not so in emerging market economies since debts are denominated in foreign currency 3. Bernanke and Mishkin (1997) argue that exchange rate pegs can lead to financial fragility 4. Although exchange rate targeting may be initially successful in bringing inflation down a successful speculative attack can lead to a resurgence of inflation Anchor: Monetary targeting

Advantages

1. An advantage over exchange rate targeting is that monetary targeting enables a central bank to adjust its monetary policy to cope with domestic considerations 2. A monetary target is easily understood by the public – but not as well as an exchange rate target 3. Monetary targets have the advantage of being able to promote almost immediate accountability for monetary policy

Disadvantages

1. Typically the link between money growth and inflation is subject to long and uncertain lags 2. The demand for money may not be stable, there may be instability of velocity and the money supply may not be controllable (Jha and Rath, 2003). This is especially true of broad monetary targets such as M2 or M3 and less so of narrow money

Inflation targeting and the design of monetary policy in India

253

argument likely to hold with greater certainty for developing and transition countries such as India. Even if IT leads to price stability it may not guarantee financial stability. The RBI (2004) notes that the 1990s – a decade of relative price stability – witnessed a number of episodes of financial instability indicating that price stability is not sufficient for financial stability. Large movements in capital flows and exchange rates affect the conduct of monetary policy continually. Bernanke and Gertler (2001), Bernanke (2003), Bean (2003) and Filrado (2004) argue that even if price stability does not imply financial stability in the short run, price stability does not endanger financial stability and price stability, and financial stability would reinforce each other in the long run. In an economy with relative price stability the interest rate should not remain passive (as it would in an IT regime) in the presence of a sudden capital outflow. Thus the RBI (2004) notes ‘[while] there is very little disagreement over the fact that price stability should remain a key objective of monetary policy, reservations persist about adopting it as the sole objective of monetary policy’ (p. 56).6 Empirical evidence suggests that in emerging market economies central bank interest rates react more strongly to changes in the exchange rate rather than changes in the inflation rate or output gap (Mohanty and Klau, 2004). Hence, at this time, the standard tool to target inflation – short-term interest rate – is unlikely to be particularly useful.7 12.3.1

The Mechanics of Inflation Targeting

The modus operandi of a typical IT regime is as follows. The central bank revises its inflation and output forecast at a frequency determined by monetary policy committee meetings using updated information. If the conditional inflation forecast is higher than the target, the central bank raises the interest rate to minimize such deviation by the end of the targeting horizon, and vice versa. Households and firms then decide upon their consumption and investment plans. Blinder (1998) and Taylor (1993, 2002) argue that this is close to what many policy makers do in practice. It has become common to compare ex post the actual setting of policy rates by central banks with what would have been predicted by the Taylor rule which suggests that (short-term) interest rates (the federal funds rate in the USA or the bank rate in India) should be changed in response to deviation of inflation from a target and an output gap – the so-called reaction function of central banks (Clarida et al., 1998; Mohanty and Klau, 2004; Svensson, 1999). In applying his rule to the USA for the 1987–92

254

Beyond inflation targeting

period Taylor shows that it described the actual performance of policy well. IT is not applied mechanically and focuses on containing inflation as a medium-term goal (Bernanke and Mishkin, 1997).8 The choice of the price index to be used by an IT regime is critical. Typically the consumer price index (CPI) or, preferably, a measure of core inflation that ignores excessively volatile prices, for example, food and energy is used.9 Whether IT should respond to asset prices is debatable. Bean (2003) analyses an objective function minimizing output gaps and deviation from inflation targets and argues that a middle solution between completely ignoring asset prices and including asset prices in the price index number, if these affect inflationary expectations, be used for IT. Central banks operate in an environment of considerable uncertainty about the functioning of the economy and global capital flows. Hence the conduct of monetary policy must be informed by examining a number of indicators and cannot rely on just one intermediate target. Most central banks, including those of developed countries, practice liquidity management involving estimating market liquidity, autonomous of policy action, and initiate liquidity operations to steer monetary conditions and are thus able to switch between quantitative targets and interest rate targets in response to macroeconomic circumstances. Most central banks try to build in automatic stabilizers in the liquidity management framework, for example, by setting reserve requirements on an average basis to allow the financial system the leverage to adjust to temporary/seasonal liquidity shocks on its own account without central bank action and exercising an explicit preference for encasing short-term interest rates in a corridor around some optimal rate rather than at a point target.

12.4

THE PERFORMANCE OF INFLATION TARGETING

There is considerable debate about whether IT improves performance in regard to inflation and output. Ball and Sheridan (2003) argue that the adoption of IT does not lead to a systematic improvement in the growth-inflation trade-off; Hu (2004) argues otherwise. Fraga et al. (2003) focus on emerging market economies, including India, and show that economies working with IT have higher volatilities of output, inflation, interest rates and exchange rates than developed countries using IT (Table 12.4). Preparing for a switch to an IT regime requires considerable background work. Financial markets should be sufficiently developed

Inflation targeting and the design of monetary policy in India

Table 12.4

Countries

255

Volatility and average of selected variables for 1997:1–2002:2 (quarterly data) Volatility of basic variables Inflation Exchange rate*

Developed Economies Australia 2.05 Canada 0.83 Iceland 2.45 New Zealand 1.21 Norway 0.77 Sweden 1.11 Switzerland 0.54 United Kingdom 0.92 Average 1.24 Median 1.02 Emerging Market Economies Brazil 2.09 Chile 1.30 Colombia 5.43 Czech Republic 3.46 Hungary 4.09 Israel 3.18 Mexico 5.98 Peru 3.04 Poland 4.13 South Africa 2.13 South Korea 2.36 Thailand 3.25 Average 3.37 Median 3.22

Average

GDP growth**

Interest rate

GDP Inflation growth

0.13 0.04 0.15 0.16 0.10 0.12 0.08 0.06 0.11 0.11

1.96 1.30 3.13 3.61 2.25 2.41 1.14 0.79 2.07 2.11

0.58 1.14 3.02 1.47 1.46 0.44 0.92 1.13 1.27 1.13

4.78 3.57 4.17 3.09 2.66 2.58 1.79 2.61 3.16 2.88

5.89 1.96 4.05 1.65 2.44 1.24 0.85 2.46 2.57 2.20

0.31 0.17 0.25 0.09 0.16 0.10 0.07 0.11 0.11 0.26 0.14 0.14 0.15 0.14

2.06 3.25 3.38 2.73 – 3.36 3.17 3.45 2.40 1.11 6.38 6.13 3.40 3.25

7.06 – 10.02 5.81 1.13 3.34 7.26 5.50 4.14 3.65 5.52 6.72 5.47 5.52

1.81 3.11 0.81 1.18 – 2.98 4.05 2.11 3.85 2.26 4.31 0.08 2.41 2.26

5.89 3.88 12.51 5.31 11.21 4.35 11.72 3.89 8.40 6.51 3.73 2.88 6.69 5.60

Note: * refers to the coefficient of variation (standard deviation/mean); ** growth rate measured comparing the current quarter to the same quarter of the previous year. Source:

International Financial Statistics, IMF (quarterly data) (2003).

and global capital markets should have adequate confidence in these markets, thus enabling the adoption of a sufficiently flexible exchange rate regime and the central bank should have a high degree of independence and be able to use short-term interest rates as the main operating instruments. Such conditions may not be satisfied in many transition countries.

256

12.5

Beyond inflation targeting

RECENT INDIAN EXPERIENCE WITH NOMINAL TARGETING

The RBI has never pursued a pure nominal targeting regime, opting for a combination of rules-based and discretionary measures with the former changing over time. In the 1980s and early 1990s the RBI opted for a nominal exchange rate peg externally, and monetary control internally. However, both policy mechanisms have faltered. An inflexibly pegged exchange rate has proved to be unsustainable in the presence of strong capital flows10 whereas the instability of the money demand function as well as its supply (Jha and Rath, 2003) indicates that monetary targeting, by itself, is no longer a feasible option. Further, the RBI faces a persistent fiscal overhang and has to support high fiscal deficits. If fiscal policy is imprudent and the central bank does not help finance the deficit, the end result would still be inflationary as the public debt/GDP ratio would turn unsustainable in the medium term and the price level could at least partially be determined by the fiscal theory of the price level.11 Fiscal deficits are inflationary and put pressure on real interest rates and crowd out private investment (Engen and Hubbard, 2004). There is a vicious cycle between inflation and budget deficits – high deficits cause higher inflation, which raise interest rates, thus increasing the deficit by raising debt service payments. Further, higher inflation reduces the real value of tax collections. The literature has emphasized frameworks based on the clear mandates of central bank independence and fiscal responsibility legislation. Fiscal rules restrict government spending which checks excessive build-up of deficits and public debt, imparting stability to the economy. Concurrently fiscal rules may restrict the government’s ability to take countercyclical policy measures and hence contribute to increased business cycle volatility. Fiscal policy rules are likely to be effective if accompanied by strong commitments and increased transparency (Bayoumi and Eichengreen, 1995) – hence the widespread consensus for central bank independence backed by fiscal discipline to contain inflation and stabilize inflationary expectations. Although price stability, output growth, reduction of exchange rate volatility and financial stability are monetary policy goals for the RBI, none of these are under its direct control. The RBI sets intermediate targets which it can control and which have a stable relationship with the ultimate goals of monetary policy. A narrow target such as base money may be fully within RBI control but may have only weak links with the ultimate objectives of monetary policy. A broad target such as nominal income

Inflation targeting and the design of monetary policy in India

257

may be closely related to the ultimate objectives of monetary policy but not be amenable to RBI control. Both money supply and money demand have become unstable since the initiation of financial sector reforms. Other nominal targets have similar problems. Hence a purely rules-based monetary policy regime seems unhelpful.

12.6

RECENT DEVELOPMENTS IN MONETARY POLICY DESIGN IN INDIA

With the progressive widening of fiscal deficits from the 1960s, the burden of financing was borne by the RBI and the banking system. The support of the banking system to the government’s borrowing program involved progressive increases in the statutory liquidity ratio to 38.5 percent by the early 1990s. Although interest rates on government securities were steadily raised to enhance their attractiveness, it became increasingly difficult to get voluntary subscriptions even at high interest rates. The cash reserve ratio was increased from 3 percent in the early 1970s to almost 25 percent (if incremental reserve requirements are taken account of) by the early 1990s. Nevertheless, liquidity growth remained excessively high during the 1970s and 1980s and spilled over onto inflation. With expansionary fiscal policy there are limits to the effectiveness of monetary policy in containing inflation. The combined deficit of central and state governments has been close to 10 percent of GDP for more than 15 years and the share of net bank credit to the government in financing the fiscal deficit has hovered around 10 percent of GDP for much of the past decade. The 1985 Chakravarty Committee on Monetary Policy recommended that price stability emerge as the ‘dominant’ objective of monetary policy along with commitment to fiscal discipline (RBI, 2002, p. 67). Price stability was seen to be critical to sustain the process of reforms begun in 1991 (RBI, 1993). In the latter half of the 1990s, as the economy slowed down, monetary policy pursued an accommodative stance with an explicit preference for a softer interest rate regime with a constant vigil on inflation. The RBI formally adopted a multiple indicator approach in April 1998. These are to (1) maintain a stable inflation environment; (2) maintain appropriate liquidity conditions to support higher economic growth; (3) ensure orderly conditions in the exchange market to avoid excessive volatility in the exchange rate; and (4) maintain stable interest rates (RBI, 2002). Besides broad money, which remains an information variable, other macroeconomic indicators including interest rates, rates of return on money, capital and government securities markets along with data on

258

Beyond inflation targeting

currency, credit extended by banks and financial institutions, fiscal position, trade, capital flows, inflation rate, exchange rate, refinancing and transactions in foreign exchange available on a high frequency basis are juxtaposed with output data for drawing policy perspectives when formulating monetary policy. This ‘check list’ approach has been criticized as watering down the concept of a nominal anchor. However, it is difficult to find a variable that would encapsulate the large number of factors that need to go into monetary policy making at this stage of transition from a relatively autarkic administered economy to a relatively open market-oriented one. The RBI uses a mix of policy instruments including changes in reserve requirements, and standing facilities and open market operations which affect the quantum of marginal liquidity and changes in policy rates, for example, the bank rate and repo/reverse repo rates, which impact the price of liquidity with short-term interest rates signaling the stance of monetary policy. Shifts in monetary policy transmission channel necessitated policy impulses which would travel through both quantity and rate channels and episodes of volatility in foreign exchange markets emphasized the need for swift policy reactions balancing the domestic and external sources of monetization in order to maintain orderly conditions in financial markets. Even within the set of indirect instruments the preference is for marketbased instruments. Another serious challenge comes from the capital account, especially the high volatility of capital flows vis-à-vis trade flows.12 Since external borrowings are denominated in foreign currency, large devaluations are inflationary and cause serious currency mismatches with adverse effects on the balance sheets of borrowers. The need for reserves as self-insurance emanates from the volatile nature of capital flows (including sharp reversals) and reflects weakness in the existing international financial architecture. India’s ratio of net foreign assets to reserve money grew from 11.9 percent in 1990 to 44.5 percent in 1996, 65.8 percent in 2000 and 117.3 percent in 2003.

12.7

INDIA’S UNPREPAREDNESS FOR INFLATION TARGETING

That transition economies such as India may not be ready for IT is the considered view not just of the RBI but also International Monetary Fund (IMF) economists. Masson et al. (1997) argue that economic structures in developing countries (including India) are incapable of supporting an IT

Inflation targeting and the design of monetary policy in India

259

regime in the short to medium runs, because such countries do not satisfy a number of prerequisites for the successful implementation of IT. The authors consider these to be: 1.

2.

3.

Independence of the central bank: this refers both to operational conditions and the policy space within which the RBI can operate. There are limits to the effectiveness of monetary policy in containing inflation when faced with expansionary fiscal policy. Domestic and financial markets should have enough depth to absorb the placement of public and private debt instruments; and the accumulation of public debt should be sustainable. In the Indian case, while there is some evidence to suggest that the latter condition is satisfied (Jha and Sharma, 2004), the first is definitely not (Sharma, 2004). If both conditions are not satisfied the independence of monetary from fiscal policy is compromised and the interest rate transmission channel of policy is weak and incompletely evolved. In addition, the central government can, even with financial liberalization, apply subtle pressure on the RBI to alter monetary policy. I give two instances of these. In the latter half of 2004, when inflation topped 8 percent and real interest rates had become negative, the RBI wanted to raise the bank rate to lower inflation but could not, under government pressure. Similarly in early 2005 the Governor of the RBI publicly voiced concern over volatile FII inflows and suggested a fiscal approach to capping them. However, the Finance Minister almost immediately rebuffed him. Refraining from using other nominal anchors: successful adoption of IT requires that other nominal variables such as the exchange rate should not be targeted. However, India needs to maintain a stable and competitive exchange rate to encourage exports. Even in developed economies, which have explicitly opted for it, IT is associated with high exchange rate volatility. In view of their vulnerability to exchange rate crises, developing countries such as India are wary of excessive volatility. Predominance of demand as opposed to supply shocks: IT implicitly assumes that monetary policy has to respond primarily to demand shocks. Balakrishna (1991) has underscored the role of supply shocks in determining inflation in India.13 These make inflation dependent on monetary as well as non-monetary factors. If inflation rises because of a demand shock, the pursuit of IT will stabilize both inflation and output. However, if inflation rises because of an adverse supply shock, the pursuit of IT will exacerbate the recessionary effect on output by reducing demand (McKibbin and Singh, 2003).

260

4.

Beyond inflation targeting

Practical difficulties in the implementation of IT: the high frequency data requirements including those of a fully dependable inflation rate for targeting purposes are not yet available (RBI, 2004).

12.8

CHECKING FOR VIABILITY OF INFLATION TARGETING IN INDIA

A prerequisite for the RBI to pursue IT is that there should be a stable and significant relationship between the measure of inflation to be controlled and short-term interest rates. I test for this using monthly data over the period April 1992 to March 1998 from the RBI’s Handbook of Statistics on the Indian Economy (2000). The variables used are as follows: IIP: Index of industrial production (1980–815100); REER: Index of real effective exchange rate (36-country), 19855100; Narmon: Narrow money; Cmrate: Call money rate; Xrate: Exchange rate of Indian rupee vis-a-vis US dollar (monthly averages); CPI: Consumer price index for industrial workers (1982 5100); WPITR20: Trimmed WPI (Mohanty et al., 2000); WPI: Wholesale price index (1993–94 5100); WPIADM: Wholesale administered price index (ibid.). Monthly dummies were added and logs were taken of all variables except the Cmrate. Augmented Dickey Fuller tests (not reported here to conserve space) indicated that all series are I(1).14 The bivariate relationships between the three candidate inflation measures and the monthly economic indicators, the P values from bivariate Granger causality tests are reported in Table 12.5. Each entry gives the P values for the null hypothesis that the indicator does not cause the inflation measure, that is, the probability of obtaining a sample, which is even less likely to conform to the null hypothesis of no Granger-causality than the sample at hand. These Granger causality results are reported up to eight lags. The WPITR20 measure of inflation assumes that the WPI is the headline measure of inflation and defines the trimmed mean inflation index as: WPITRa 5

1

n21

(12.1) a wipi a i5k1l bb 100 where WPITRa is the trimmed WPI computed by ordering the component price change data pi and associated weights wi and removing the components on each tail of the distribution by a percent. The numbers of components trimmed from the left and right tails of the distribution are k and l respectively. When a 5 0 the trimmed mean would equal the weighted a1 2 2a

Inflation targeting and the design of monetary policy in India

Table 12.5 CPI Lags 1 2 3 4 5 6 7 8 WPITR20 Lags 1 2 3 4 5 6 7 8 WPIADM Lags 1 2 3 4 5 6 7 8

261

P Values from Bivariate Granger Causality Tests IIP

Exrate

Narmon

REER

Cmrate

0.22 0.4 0.69 0.1 0.01* 0.00* 0.00* 0.00*

0.67 0.72 0.87 0.5 0.25 0.13 0.12 0.16

0.00* 0.00* 0.00* 0.00* 0.00* 0.00* 0.00* 0.00*

0.64 0.99 0.61 0.46 0.36 0.26 0.12 0.03*

0.35 0.43 0.93 0.8 0.55 0.58 0.69 0.82

0.06 0.01* 0.00* 0.00* 0.00* 0.00* 0.00* 0.00*

0.00* 0.00* 0.00* 0.00* 0.00* 0.00* 0.00* 0.00*

0.07 0.01* 0.00* 0.00* 0.00* 0.00* 0.00* 0.00*

0.01* 0.00* 0.00* 0.00* 0.00* 0.00* 0.00* 0.00*

0.14 0.09 0.00* 0.04* 0.19 0.2 0.14 0.26

0.00* 0.00* 0.00* 0.00* 0.00* 0.00* 0.00* 0.00*

0.33 0.09 0.03* 0.00* 0.00* 0.00* 0.00* 0.00*

0.08 0.01* 0.00* 0.00* 0.00* 0.00* 0.00* 0.00*

0.75 0.56 0.44 0.2 0.01* 0.00* 0.00* 0.00*

0.45 0.23 0.18 0.97 0.44 0.41 0.4 0.12

Note: * 5 Significant at 5 percent level; CPI: Consumer price index; WPITR: Trimmed whoesale price index; WPIADM: Price index for the administered goods; IIP: Index of industrial production; Narmon: Narrow money; Exrate: Exchange rate Rs/$; REER: Real effective exchange rate; Cmrate: Call money rate.

mean whereas in the case of a 5 50 it would equal the weighted median. The root mean square error (RMSE) for any level of trimming is defined by: n

RMSEa 5

(pa 2 pt) 2 /n Åa t i51

(12.2)

262

Beyond inflation targeting

model 1: Cmrate –> LCPI 0.01

0

–0.01 95% CI for OIRF OIRF –0.02 0

10

20

30

step Note: The band indicates the 95 percent confidence interval for the orthogonalized impulse response function.

Figure 12.1

Orthogonalized impulse response function (OIRF) of call money rate on log CPI

where pat is the trimmed WPI with a trimming ratio of a percent from each of the tails of the price distribution at time t, pt is the 36-month centered moving average change in WPI at time t, and n is the number of samples. Mohanty et al. (2000) conclude that this RMSE is minimized for a 5 20. Data on this variable are available in Mohanty et al. (2000). The results of the Granger causality test indicate a weak relation between Cmrate and the measures of inflation. In fact only WPITR20 seems to have a causal relation with Cmrate. On the other hand, the links between the measures of inflation and IIP, Narmon, Xrate and REER are much stronger. Hence the causality tests do not provide support for using interest rates as instruments in a policy of IT. A drawback of the crude Granger causality testing is that it provides no information about whether the sign of the (dynamic) bivariate relationship is theoretically correct. Further, there may be omitted variables. A VAR on the variables lCPI, lIIP, lNarmon, lREER and Cmrate indicates that the conclusions of Table 12.5 are broadly correct.15 Figure 12.1 shows that the 95 percent confidence band for the

Inflation targeting and the design of monetary policy in India

263

orthogonalized impulse response function of Cmrate on lcpi is very wide, hence adding to our agnosticism about the efficacy of IT in India. The error correction model for lCPI is not significantly responsive to any of the error correction terms. Hence it appears that IT may be difficult to pursue in the Indian context.

12.9

CONCLUSIONS: OPTIONS FOR INFLATION TARGETING IN INDIA

This chapter has argued that the primary objective of Indian monetary policy, at least in the medium term, has to be the attainment of higher economic growth. Moreover, since India has high inflation aversion, this objective does not conflict with short-term stabilization. The design of monetary policy in India is circumscribed by the fact that the liberalization of financial markets is far from complete so that the interest rate transmission channel is incomplete. Further, the banking system has strong monopoly elements. Moreover, as the financial sector liberalizes some major government-owned mutual fund operations have had to be bailed out. The exacerbation of such contingent liabilities along with already high fiscal deficit aggravates monetary policy difficulties in the Indian context. Against this background this chapter has argued that the multi-objective formulation pursued by the RBI has merit and that such monetary policy should be pursued to maintain stable interest and inflation rates and a slightly undervalued currency in order to engineer higher export-led growth. This policy has, however, led to substantial capital inflows with attendant build-up of reserves and necessitated considerable sterilization operations. This has now emerged as a significant problem with its continuance at the current pace seemingly unsustainable if, for no other reason, than the fact that such reserves attract low yields. Currently two policy packages to address this issue have been discussed. The first is geared towards fiscal correction and monetary expansion. A second policy measure is weighted towards real exchange rate appreciation (more in line with IT) and would involve relatively larger current account deficits. Real appreciation could be secured by nominal appreciation or higher inflation. Both policies would lead to low inflation rates and reduced inflows of foreign capital and, therefore, lower accumulation of reserves at given rates of sterilization. Policy packages that use import liberalization would, like real appreciation, permit higher absorption via higher current account deficits but without penalizing exports. The optimal package for India is

264

Beyond inflation targeting

a judicious combination of these two broad sets of policies with greater emphasis on fiscal consolidation and import liberalization, rather than real exchange rate appreciation. These are essential elements of an appropriate monetary policy regime for India. Since rapid export growth is important, it makes sense to err on the side of undervaluation of the exchange rate, enabling India to capture a larger share of world markets. Growing exports, in turn, raise the incentive to invest. The propensity to save also rises in response to the increased profitability of export-oriented investment. Moreover, an undervalued exchange rate is likely to boost saving by raising the share of profits in national income. This argument does not imply that unlimited real depreciation is feasible or desirable, just that there should be a bias towards mild undervaluation because it can play a supportive role to complementary outward-oriented trade policies in generating a virtuous circle of higher saving, investment and growth. Along this path import demand would grow concomitantly and getting a current account surplus is not inevitable. Clearly India has been conducting some form of real exchange rate targeting leading to a sharp rise in foreign exchange reserves. Lal et al. (2003) indicate that this has come at high economic costs – a claim that has been successfully disputed.16 IT would require India to pursue a clean float reducing the need for large reserves. But the price to be paid is the possibility of a highly unstable or inappropriate exchange rate. India’s policy makers were wise to reject this regime and opt for managed floating plus selective controls on capital flows. Reserves are now at a very comfortable level and continue to rise rapidly. The question of whether and how to absorb foreign inflows is far more pertinent now than it was in the 1990s. Clearly sterilization has outlived its usefulness. Some sterilized reserve accumulation can continue to maintain the present ratio of reserves to GDP. Further increases in the ratio should be avoided except as a purely short-term response to manifestly short-term inflows. The policies espoused here have the advantage that, in addition to promoting balance of payments adjustment, they are desirable independently of the balance of payments and of the ‘temporary’ or ‘permanent’ character of the inflows. Naturally, due to political constraints, these policies can only be pursued at a moderate pace. If there is continued acceleration of inflows, despite the adoption of the suggested strategy, the government should consider tightening capital inflow controls17 so that the strategy is not derailed. A corollary would be that capital account convertibility is eschewed. It is not being suggested that India should resist an exchange rate appreciation indefinitely. Once Indian GDP has grown in excess of 8 percent for more than two decades, so that real incomes have gone up substantially

Inflation targeting and the design of monetary policy in India

265

and unemployment and poverty have dropped sharply, India could contemplate changes in the monetary policy regime, essentially in the direction of relaxing the undervaluation of the exchange rate. However, even at that point, given the importance of variables other than inflation in an optimal monetary policy framework, there would be no argument for moving to IT.

NOTES 1. 2. 3.

4. 5.

6. 7. 8.

9. 10.

11. 12. 13. 14. 15. 16.

I am grateful to Gerald Epstein and three anonymous referees for helpful comments and Anurag Sharma for research assistance. The usual disclaimer applies. At the very least there is no evidence that economic growth hurts poverty alleviation (Winters et al., 2002). Results from the 1999–2000 National Sample Survey (Central Statistical Organisation, 2000) indicate a larger drop in poverty; however, this survey’s methodology is not comparable with those of the earlier surveys, correcting for which reveals a modest drop in poverty. Nominal income, another intermediate target, is both hard to target and poorly related to the ultimate objectives of monetary policy. Another alleged advantage of an IT regime is that deviations from inflation targets are routinely allowed in response to supply shocks by excluding some combination of food and energy prices, indirect tax changes, terms of trade shocks and the direct effects of interest rate changes on the index. See also Epstein (2000). For an application to India, see Section 12.8. Seyfried and Bremmer (2003) discover that the Reserve Bank of Australia pays particular attention to inflationary pressures, as measured by the GDP gap. They find a relatively high degree of persistence and low speed of adjustment in the interest rate indicating that the central bank is interested in interest smoothing in addition to IT. See also Lomax (2005). India with one wholesale price index and four CPIs still does not have a single price index with widespread acceptability for IT. Measures of core inflation are not computed officially (Mohanty et al., 2000). Joshi and Sanyal (2004) indicate the RBI has been targeting the index of real effective exchange rate (REER) of the Indian rupee with regard to the currencies of five countries, USA, Japan, UK, Germany and France, at the 1993–94 level. Patel and Srivastava (1997) note that such targeting has had more than a transitory effect. However this benign relationship may break as reforms lead to greater capital mobility. In Latin America Jacome and Vazquez (2005) find no causal relationship between central bank independence and inflation, although the association between the two is strong. The Indian rupee is convertible on the current account but capital account convertibility is not permitted. Callen and Chang (1999) review the literature on inflation in India. The threshold of inflation over which price stability should take precedence over the growth objective has been estimated to be between 4 and 6.5 percent (Samantaraya and Prasad, 2001). Since the Mohanty et al. (2000) data set stops at 1998 we do not extend our analysis beyond that date. Detailed results are reported in the full version of the paper available on the PERI website, accessed 6 May 2007 at http://www.peri.umass.edu. Lal et al. (2003) argue that India’s growth rate in the 1990s could have been up to

266

17.

Beyond inflation targeting 2.7 percent per annum higher if foreign exchange inflows during the decade had been fully absorbed. However, Joshi and Sanyal (2004) argue that if net foreign inflows had been absorbed domestic spending (and not foreign exchange reserves) would have risen. Reserves as a proportion of GDP rose over the 1990s by an average of about 1.2 percent per annum. If the entire increase in reserves had been absorbed into investment each year, the ratio of investment to GDP averaged over the decade would have been 1.2 percent higher than it actually was. Given fixed incremental capital output ratio (ICOR) (of 2.8 in the 1990s) the increase in India’s growth rate of GDP would have been only 1.2/2.8 5 0.4 percent per annum (approx.) higher over the decade. This sacrifice would be even lower if (1) the ICOR would have risen (in line with the assumption of diminishing returns to capital); (2) some of the reserve accumulation spilled over onto higher consumption, thus reducing the growth rate; and (3) furthermore, the level of foreign exchange reserves in India was inadequate in 1991 and accumulating foreign reserves was necessary. Through a Chilean-type tax, for example.

REFERENCES Balakrishna, P. (1991), Pricing and Inflation in India, Delhi: Oxford University Press. Ball, L. and N. Sheridan (2003), ‘Does inflation targeting matter?’, National Bureau for Economic Research, working paper no. 9577. Barro, R. and D. Gordon (1983), ‘A positive theory of monetary policy in a natural rate model’, Journal of Political Economy, 91 (4), 589–610. Bayoumi, T. and B. Eichengreen (1995), ‘Restraining yourself: the implications of fiscal rules for economic stabilization’, IMF Staff Papers, 42, 32–48. Bean, C. (2003), ‘Asset prices, financial imbalances and monetary policy: are inflation targets enough?’, Bank for International Settlement working paper no. 140. Bernanke, B. (2003), ‘Constrained discretion and monetary policy’, BIS Review, 5, 1–8. Bernanke, B. and M. Gertler (2001), ‘Should central banks respond to movements in asset prices?’, American Economic Review, 91 (2), 253–7. Bernanke, B. and F. Mishkin (1997), ‘Inflation targeting: a new framework for monetary policy?’, Journal of Economic Perspectives, 11 (2), 97–116. Blinder, A. (1998), Central Bank in Theory and Practice, Cambridge, MA: MIT Press. Callen, T. and D. Chang (1999), ‘Modeling and forecasting inflation in India’, International Monetary Fund working paper WP/99/119, Washington, DC. Central Statistical Organisation (2000), National Sample Survey, New Delhi: Government of India. Clarida, R., J. Gali and M. Gertler (1998), ‘Monetary policy rules in practice: some international evidence’, European Economic Review, 42 (4), 1033–67. Dollar, D. and A. Kraay (2001), ‘Growth is good for the poor’, World Bank policy research paper no. 2587, pp. 1–50. Engen, E. and R. Hubbard (2004), ‘Federal governments and interest rates’, National Bureau for Economic Research working paper no. 1068. Epstein, G. (2000), ‘Myth, mendacity and mischief in the theory and practice of central banking’, accessed 6 May 2007 at www.umass.edu/peri.

Inflation targeting and the design of monetary policy in India

267

Filrado, A. (2004), ‘Monetary price and asset price bubbles: calibrating the monetary policy tradeoffs’, Bank for International Settlement working paper no. 155. Fraga, A., I. Goldfajin and A. Minella (2003), ‘Inflation targeting in emerging market economies’, National Bureau for Economic Research, working paper no. 10019. Hu, Y. (2004), ‘Empirical investigations of inflation targeting’, Mimeo, Institute of International Economics, Washington, DC, February. Jacome, L. and F. Vazquez (2005), ‘Any link between central bank independence and inflation? Evidence from Latin America and the Caribbean’, International Monetary Fund working paper no. WP/05/75. Jha, R. (2004), ‘Reducing poverty and inequality in India: Has liberalization helped?’, in A. Cornia (ed.), Inequality, Growth and Poverty in an Era of Liberalization and Globalization, Oxford: Oxford University Press, pp. 287–326. Jha, R. (2005), ‘The political economy of recent economic growth in India’, in R. Jha (ed.), Economic Growth, Economic Performance and Welfare in South Asia, Houndmills: Palgrave Macmillan, pp. 28–51. Jha, R. and D. Rath (2003), ‘On the endogeneity of the money multiplier in India’, in R. Jha (ed.), Indian Economic Reforms, Basingstoke: Palgrave Macmillan, pp. 51–72. Jha, R. and A. Sharma (2004), ‘Structural breaks, unit roots, and cointegration: a further test of the sustainability of the Indian fiscal deficit’, Public Finance Review, 32 (2), 196–219. Jha, R., R. Gaiha and A. Sharma (2006), ‘Mean consumption, poverty and inequality in India in the sixtieth round of the national sample survey’, ASARC working paper no. 2006/11, Australian National University. Joshi, V. and S. Sanyal (2004), ‘Foreign inflows and macroeconomic policy in India’, Mimeo, Oxford University, pp. 1–54. Kelkar, V. (2004), ‘India on the growth turnpike’, Narayanan Oration, presented at Australia South Asia Research Centre, Australian National University. Kydland, F. and E. Prescott (1977), ‘Rules rather than discretion: the inconsistency of optimal plans’, Journal of Political Economy, 87 (2), 473–92. Lal, D., S. Bery and D. Pant (2003), ‘The real exchange rate, fiscal deficits and capital flows – India: 1981–2000’, Economic and Political Weekly, 38 (47), 4965–76. Lomax, R. (2005), ‘Inflation targeting in practice – models, forecasts and hunches’, speech by the Deputy Governor of the Bank of England to the 59th Atlantic Economic Conference, 12 March. Masson, P., M. Savastano and S. Sharma (1997), ‘The scope for inflation targeting in developing countries’, International Monetary Fund working paper no. 97/130. McKibbin, W. and K. Singh (2003), ‘Issues in the choice of a monetary regime for India’, in R. Jha (ed.), Indian Economic Reforms, Basingstoke: Palgrave Macmillan, pp. 11–50. Mohanty, M. and M. Klau (2004), ‘Monetary policy rules in emerging economies: issues and evidence’, Bank for International Settlement working paper no. 149. Mohanty, D., D. Rath and M. Ramaiah (2000), ‘Measures of core inflation for India’, Economic and Political Weekly¸ 35 (5), 273–82.

268

Beyond inflation targeting

Obstfeld, M. and K. Rogoff (1996), Foundations of International Macroeconomics, Cambridge, MA: MIT Press. Patel, U. and P. Srivastava (1997), ‘Some implication of real exchange rate targeting in India’, Mimeo, Indian Council for Research on International Economic Relations, New Delhi, March. RBI (Reserve Bank of India) (1993), Annual Report 1991–92, Mumbai: RBI. RBI (2000) Handbook of Statistics on the Indian Economy, Mumbai: Reserve Bank of India. RBI (2002), Reserve Bank of India Bulletin, Mumbai: RBI. RBI (2004), Report on Currency and Finance, Mumbai: RBI. Samantaraya, A. and A. Prasad (2001), ‘Growth and inflation in India: detecting the threshold level’, Asian Economic Review, 43 (3), 414–20. Seyfried, W. and D. Bremmer (2003), ‘Inflation targeting as a framework for monetary policy: a cross-country analysis’, Australian Economic Review, 36(3), 291–9. Sharma, A. (2004), ‘Fiscal deficits, financial crises and adjustment in a semi-open economy’, unpublished PhD dissertation, Australian National University. Svensson, L. (1999), ‘Inflation targeting as a monetary policy rule’, Journal of Monetary Economics, 43 (3), 607–54. Taylor, J. (1993), ‘Discretion versus policy rules in practice’, Carnegie–Rochester Conference Series on Public Policy, 39, 195–214. Taylor, J. (2002), ‘The monetary transmission mechanism and the evaluation of monetary policy rules’, in N. Loayza and K. Schmidt-Hebbel (eds), Monetary Policy: Rules and Transmission Mechanisms, Santiago: Central Bank of Chile. White, W. (2004), ‘Making macro prudential concerns operational’, speech delivered at Symposium on Financial Stability at De Nederlandsche Bank, Amsterdam, March. Winters, A., N. McCulloch and A. McKay (2002), ‘Trade liberalisation and poverty: the empirical evidence’, Department of Economics, University of Sussex discussion paper no. 88, Brighton.

Inflation targeting and the design of monetary policy in India

APPENDIX 12.A1

269

CAPITAL CONTROLS: OR WHY DID INDIA ESCAPE THE EAST ASIAN CRISIS?

It is instructive to compare India and the East Asian countries in 1996 (that is, just before the East Asian crisis of 1997). The first six columns of Table 12.A1 indicate that, in most respects, India’s ‘fundamentals’ (fiscal balance, inflation, current account balance, non-performing assets, debt-exports ratio and debt-service ratio) were worse or no better than the crisis-countries. All these countries were on loose dollar peg and India was only marginally different from the rest in this regard except for the fact that India did not allow its real exchange rate to appreciate and was able to maintain its real exchange targeting posture. The critical difference between India and the crisis-countries can be seen in the last two columns of Table 12.A1. India managed to keep shortterm debt under control, both in relation to total debt and in relation to foreign exchange reserves and thus avoided an unstable debt structure, an outcome that was the direct result of controls on debt-creating short-term inflows. Table 12.A1

India Indonesia Korea Malaysia Philippines Thailand

Various countries: indicators of crisis-vulnerability, 1996 FB/ GDP (%)

ΔP/P (% p.a.)

CAB/ XGS (%)

NPA (%)

NCEDT/ XGS (%)

TDS/ XGS (%)

SDT/ SDT/ EDT RES (%) (%)

–9.0 –1.0 0.0 0.7 0.3 0.7

9.0 8.0 4.9 3.5 8.4 5.8

–11.7 –13.0 –14.6 –6.4 –9.9 –19.5

17.3 8.8 4.1 3.9 n.a. 7.7

103.6 180.5 82.0 40.4 80.1 110.9

21.2 36.6 9.4 9.0 13.4 12.6

5.3 25.0 49.4 27.9 19.9 41.5

27.1 166.7 192.7 39.7 67.9 97.4

Note: FB/GDP: Fiscal balance as a proportion of GDP; ΔP/P: Rate of consumer price inflation; CAB/XGS: Current account balance as a proportion of exports of goods and services; NPA: Non-performing assets of commercial banks as a proportion of total advances; NCEDT/XGS: Non-concessional external debt as a proportion of exports of goods and services; TDS/XGS: Debt service as a proportion of exports of goods and services; SDT/ EDT: Short-term external debt as a proportion of total external debt; SDT/RES: Short-term external debt as a proportion of foreign exchange reserves. Sources: FB/GDP, NPA: Bank of International Settlements Annual Reports (1997–98 and 1999–2000) and Government of India, Economic Survey (1999–2000). CAB/XGS, NCEDT/XGS, TDS/XGS, SDT/EDT, SDT/RES: World Bank, Global Development Finance (1999). ΔP/P: IMF International Financial Statistics (2000).

270

Beyond inflation targeting

India was able to resist the pressure to adopt capital account convertibility essentially because of three reasons: first, the ideology of laissezfaire is still not dominant in India, and second, foreign banks, which are normally a strong pressure group in favor of capital account convertibility, had a very small presence in the country. Finally, India was ‘too big to be bullied’ into adopting capital account convertibility by Wall Street, the IMF and the US Treasury (Joshi and Sanyal, 2004).

13.

Towards an alternative monetary policy in the Philippines Joseph Anthony Lim1

13.1

INTRODUCTION

The modern macroeconomic history of the Philippines had been marked by periodic balance of payment crises with massive devaluations that resulted in high inflation. These crises were followed by drastic monetary and fiscal recessionary programs implemented by the International Monetary Fund (IMF), as a precondition for the release of emergency funds. Monetary targeting was implemented in the Philippines from the 1980s to the 1990s under IMF sponsorship. It is generally accepted as contributing to the depth of recessions in the Philippines in the last 25 years. Since it was replaced by inflation targeting only in 2002, it is difficult not to include in this chapter a discussion of the experience with monetary targeting. The next section describes the monetary targeting experience of the Philippines. Section 13.3 discusses the shift from the monetary targeting regime to the inflation targeting regime starting 2002. Section 13.4 places monetary policy in the context of the complex macro situation and development needs of the Philippines. The last section gives detailed recommendations for an alternative monetary policy.

13.2

MONETARIST TARGETING AND POLICY IN THE PHILIPPINES: CRITICIZING THE DEMAND-SIDE CURE FOR INFLATION AND CURRENT ACCOUNT DEFICITS

The economic crises which occurred from the late 1940s to the present had been connected with balance of payment and foreign exchange crises. These have led to some very sharp recessions – especially the economic collapse in 1984–85 – that destroyed any chance of the Philippines becoming an East Asian success story. Carrying a strong belief that current account deficits and inflation 271

272

Beyond inflation targeting

are due to excessive aggregate demand caused by monetary expansion, monetarism’s response to balance of payment crises involved damaging pro-cyclical monetary and fiscal austerity that deepened the crisis and recession. Starting in the early 1960s the IMF had become the standard funder of last resort during crises. In the early 1980s quarterly monetary targeting became the norm. These monetary targets became very tight every time balance of payments deteriorated and inflation increased. Monetary targets were based on targets on the monetary base and achieved through: (1) high required reserve ratio; (2) high policy rates of the central bank; and (3) open market sale of central bank bills and government securities in order to reduce the monetary base. It was the use of the third instrument that was most damaging as it directly reduced liquidity and credit in the financial sector. 13.2.1

Supply-Side Causes of Inflation

Figure 13.1 gives a picture of inflation rate based on the consumer price index (CPI) and GDP growth rate. The graph shows that high inflation usually happens simultaneously with lower growth or recessions, rather than during periods of high growth and high aggregate demand. This is particularly true for the periods 1984–85 (economic collapse), 1990–91 (another recession) and 1998 (Asian crisis period). This is because these periods were periods of significant currency devaluations resulting from balance of payment crises. This brings about stagflation that explains the high inflation and recession. In fact, the role of currency devaluations and oil price shocks in explaining periods of high inflation in the Philippines is very clearly illustrated in Figure 13.2. It can be seen that the devaluation in 1970 brought high inflation in 1970–71. The first oil price shock in 1973–74 brought about high inflation in 1973 and, especially in 1974. Again the second oil price shock and worldwide inflation in 1979–81 brought about the high inflation during the same years. The massive devaluations in 1983 and 1984 led to high inflation in the economic collapse period of 1984–85. The moderate devaluation in 1990–91, plus the oil price shock due to the first Gulf War crisis brought about the inflation in 1990 and 1991. The significant devaluation during the Asian crisis in 1998 brought a slight uptick in inflation, but nowhere near the high inflation that occurred in previous devaluations. Thus, by just using two types of supply-side shocks – currency depreciation and oil price shock – one can explain practically all the above-10 percent inflation in modern Philippine history. High inflation periods are not triggered by high domestic demand but by supply-side shocks.

Towards an alternative monetary policy in the Philippines

273

50 CPI inflation rate GDP per capita growth 40

Percent (%)

30

20

10

0

Source:

71 19 74 19 77 19 80 19 83 19 86 19 89 19 92 19 95 19 98 20 01 20 04

68

19

65

19

62

19

59

19

56

19

53

19

19

19

50

–10

National Statistics Coordination Board; International Financial Statistics (IMF).

Figure 13.1

GDP per capita growth versus CPI inflation rate

If one uses monthly or quarterly data on inflation, one will also see that agricultural price shocks make their impact on price inflation due to weather disturbances leading to food shortages. 13.2.2

Monetary Contraction and Impact on GDP and Unemployment

Using monetary contraction to fight inflation aggravates recessions and unemployment, since the original supply shocks (devaluation, oil price shocks or food shortages) already have recessionary impact. It is important to point out that the serious stagflationary periods 1984–85, 1991 and 1998 were all periods when the Philippines undertook severe monetary and credit contraction. Figure 13.3 shows the relationship of lending rate and GDP growth rate. Lending rates rise during crises because of the higher inflation, but credit and monetary contraction goes beyond this since the authorities purposely contract the monetary base and increase required reserves and the central bank policy rate. As expected, one sees a strong inverse relationship between the lending rate and GDP growth rate.

274

Beyond inflation targeting 90 Growth rate of exchange rate CPI inflation rate

80 70 60 Percent (%)

50 40 30 20 10 0

Source:

73 19 76 19 79 19 82 19 85 19 88 19 91 19 94 19 97 20 00 20 03

70

19

67

19

64

19

61

19

58

19

55

19

52

19

19

19

49

–10

National Statistics Coordination Board; International Financial Statistics (IMF).

Figure 13.2

CPI inflation versus growth of exchange rate

The lending rates in Figure 13.3 actually underestimate the costs of borrowing since during periods of monetary contraction, many firms are credit-rationed. Figure 13.4 shows the relationship between growth of real money and GDP growth rate. Declines in real money are associated with declines in GDP growth in 1960, 1964, 1974, 1984–85, 1991 and 1998. Figure 13.5 shows that the recessions of 1984–85, 1991 and 1998 created significant upticks in the unemployment rate. 13.2.3

Monetary Contraction and External Deficits

Much of the demand-suppressing monetary austerity programs were implemented not only to reduce inflation but to reduce aggregate demand to ensure that current account deficits were also reduced during balance of payment crises. And it is often overdone to produce current account surpluses. Figure 13.6 shows a graph of the current account deficit (as percentage of GDP) and GDP per capita growth rate over the years. Concentrating

Towards an alternative monetary policy in the Philippines

275

35 GDP per capita growth Lending rate

30 25

Percent (%)

20 15 10 5 0 –5

19

76 19 78 19 80 19 82 19 84 19 86 19 88 19 90 19 92 19 94 19 96 19 98 20 00 20 02 20 04 20 06

–10

Source:

National Statistics Coordination Board; International Financial Statistics (IMF).

Figure 13.3

GDP per capita growth versus lending rate

on three recession periods – 1984–85, 1991 and 1998 – it is clear that every crisis was preceded by a year or two of very significant current account deficits. This is because import demand grows during boom years as developing countries are dependent on imported raw materials and capital goods. It is also clear that after every recession, the current account deficit improved, and in the case of the crisis in the mid 1980s and the Asian crisis the current account turned positive after the recession. It should be emphasized that the improvement in the current account balance was caused partly by the massive devaluation and partly by the recession which had been aggravated by the monetary contraction dictated by monetarist policy. Monetarist theory and the IMF interpret large trade deficits as ‘overspending’ and require monetary and fiscal tightness. Thus, although monetary targeting is supposed to target only one variable – inflation – it is also affected by foreign exchange depletion and capital outflows during crises. Finally the last possible reason for monetary tightening during balance of payment crises is to stem the flight of capital and attract them back into the country, and at the same time stave off the damaging exchange rate

276

Beyond inflation targeting 50 GDP per capita growth Real money supply growth rate

40

30

Percent (%)

20

10

0

–10

–20

Source:

88 19 91 19 94 19 97 20 00 20 03 20 06

85

19

82

19

79

19

76

19

73

19

70

19

67

19

64

19

61

19

58

19

55

19

19

19

52

–30

National Statistics Coordination Board; International Financial Statistics (IMF).

Figure 13.4

GDP growth rate versus real money supply growth

collapse. This was mainly the excuse in the 1984–85 debt crisis and the 1998 Asian crisis. However, in both cases it was not the high interest rates that stopped the currency depreciation and balance of payments outflow but mainly the devaluation-cum-recession which improved the current account by improving the trade accounts. Figure 13.7 tracks the current account and balance of payment balances. It can be seen in the graph that the balance of payments improvement from 1984 to 1987 following the 1984–85 collapse was mainly due to very significant improvements in the current account. In the 1998 Asian crisis even the improvement of the balance of payments in 1998 was less than the current account improvements, indicating net capital outflows in a period of high interest rates. But in subsequent years the positive balances in the current account exceeded that of the balance of payments, which meant that there was continuing net capital outflow in the balance of payment account from 1999 to 2002 (most likely due to the lackluster performance of the Philippine economy

Towards an alternative monetary policy in the Philippines

277

14 12 10 8

Percent (%)

6 4 2 0 –2 –4 –6

Unemployment rate GDP Growth rate

19 8 19 1 8 19 2 8 19 3 8 19 4 8 19 5 8 19 6 8 19 7 8 19 8 8 19 9 9 19 0 9 19 1 9 19 2 9 19 3 9 19 4 9 19 5 9 19 6 97 19 9 19 8 9 20 9 0 20 0 0 20 1 0 20 2 0 20 3 04

–8

Source: National Statistics Coordination Board; Labor Force Survey, National Statistics Office; International Financial Statistics (IMF).

Figure 13.5

Unemployment rate versus GDP growth rate

and the political troubles of the Estrada and Arroyo governments). In the crises of 1984–85 and 1998 the current account went into significant positive territory because of the joint effects of devaluation and output and demand contraction. This proves that monetarist aggregate demand suppression is successful in stopping foreign exchange outflows and current account deficits – not by stemming capital outflows and encouraging return of foreign exchange due to high interest rates but, as previous graphs prove, by creating sharp recessions and severe unemployment, which reduce current account deficits. Inflation rates are also lowered because recessions kill consumer and investment demand. An important lesson from the discussion above is that the roles of the current account, the exchange rate regime and issues concerning how to stem capital flight should be incorporated in any alternative to the current monetary policies. From the 1980s to 2000 the sole purpose of monetary policy was to fight ‘overspending’ that had ‘caused’ current account deficits and high inflation. It ensured that monetary policy could not be used for

278

Beyond inflation targeting 6

4

2

Percent (%)

0

–2

–4

–6

–8

Current account balance, % of GDP GDP per capita growth

Source:

05

03

20

01

20

99

20

95

97

19

19

93

19

91

19

89

19

87

19

85

19

83

19

81

19

79

19

19

19

77

–10

National Statistics Coordination Board; International Financial Statistics (IMF).

Figure 13.6

Current account balance as percentage of GDP versus GDP per capita growth

countercyclical and growth purposes. The adoption of financial liberalization policies starting in the 1980s also ensured that monetary policy could not be used to finance targeted or prioritized sectors of the economy. This complemented the prescription of the multilateral agencies for the country to abandon industrial policy and undertake trade liberalization. At the same time the policy is not matched by a confirmation of the theory behind the demand theory of inflation and ‘overspending’. Figure 13.8 graphs the inflation rate on the y-axis and unemployment rate on the x-axis for the years from 1959 to 2004 to find out any trace of a Phillips curve. One can see there is nothing that looks like it in the graph in any sub-period. This is because inflation is caused by supply shocks and not demand-led. (The graph looks like it is producing ‘natural rates of inflation’ rather than ‘natural rates of unemployment’.)

Towards an alternative monetary policy in the Philippines

279

8 6 4 2

Percent (%)

0 –2 –4 –6 –8 –10

BOP, % of GDP Current account balance, % of GDP

Source:

05

03

20

01

20

99

20

97

19

95

19

93

19

91

19

89

19

87

19

85

19

83

19

81

19

79

19

19

19

77

–12

National Statistics Coordination Board; International Financial Statistics (IMF).

Figure 13.7

Current account and balance of payments, percentage of GDP

13.3

FROM MONETARY TARGETING TO INFLATION TARGETING

13.3.1

A Switch in Regime: From Monetary Targeting to Inflation Targeting

In the Philippines research work in the central bank itself pointed to the ‘overkill’ created by monetarist policy in trying to tame inflation. This is mainly because the quantity theory of money equation assumes a unitary coefficient of money to prices with output holding steady, despite massive monetary contraction. The use of a single estimated inflation equation assumes all explanatory variables (exchange rate movements, output growth, and so on) to be exogenous and that only the monetary variable will decrease inflation. A study by a staff member of the central bank

280

Beyond inflation targeting 59

Inflation rate

49 39 29 19 9 –1 8.00

8.50

9.00

9.50

10.00

10.50

11.00

11.50

12.00

Unemployment rate Source: Labor Force Survey, National Statistics Office; International Financial Statistics (IMF).

Figure 13.8

Inflation versus unemployment rate

(Dakila, 2001) shows that with a simultaneous equation system where exchange rate movements and output movements are endogenized, the degree of monetary tightening required to achieve a stipulated reduction of inflation is lessened. This is because monetary tightening cum recessionary policies tends to eventually cause some appreciation of the currency and a fall in output growth (both of which tend to be deflationary). When single-digit inflation rates were achieved in the mid 1990s under the Ramos administration, the central bank relaxed monetary targeting to what was called the ‘modified monetary targeting’ policy. This allowed the monetary targets to be exceeded as long as inflation targets were being met. This was the start of a hybrid system of inflation and monetary targeting. The official reason given was that, because of financial liberalization, the link between quantitative monetary targets and inflation had weakened due to ‘structural breaks’ in the income velocity of money and volatilities and instabilities in the money multiplier. Thus the high liquidity and large monetary expansion in the mid 1990s failed to have any impact on inflation as it remained below double-digit rates and continued to decline until the Asian crisis (see Guinigundo, 2005). The laxer monetary policy in the mid 1990s of course was shattered by the ‘contagion’ effects of massive depreciation pressures during the Asian crisis. The major ways used to stave off the depreciation (and indirectly

Towards an alternative monetary policy in the Philippines

281

avoid inflationary pressures) were constant and periodic raising of the reverse repurchase rate and the raising of liquidity reserve ratios.2 However, the massive currency depreciation (around 40 percent) during the Asian crisis had a very low ‘pass-through’ to inflation (see Figure 13.2). It did not even lead to (and only approached) double-digit inflation. This easily allowed inflation rates to fall after the crisis. Inflation rates reached a very low 3 percent in 2002 and 2003, replicating the inflation rates in the developed world. By January 2000, monetary authorities were already studying the alleged successes of developed and developing countries that had adopted inflation targeting (from a prior regime of exchange rate pegging or monetary targeting). The decision was to make the big switch to inflation targeting starting January 2002. The switch entailed the following elements: (1) continuation of the announcement of inflation targets based on a band over a two-year time period; (2) adoption of a more passive monetary quantitative policy, and employing more the repurchase and reverse repurchase rates (policy rates) as the monetary instrument; (3) increased sophistication in inflation rate estimation and in using single and multi-equation models to forecast inflation and setting of inflation targets; (4) the use of forward-looking models with monetary instruments reacting to and aiming to influence inflationary expectations rather than actual inflation; (5) creation of an advisory committee to recommend monetary policies based on the new inflation targeting regime; and (6) the issuance of a quarterly inflation report explaining the central bank’s policies and achievement or non-achievement of the inflation targets. But the most important elements are: (7) the stipulation of escape clauses that exempts the central bank from achieving the inflation targets. This means that if the inflation target is not achieved, the central bank can opt to not do anything if the reasons are: (a) sudden changes in prices of agricultural commodities and products; (b) natural calamities or catastrophic events; (c) volatility in oil prices; and (d) sudden changes in government policies, such as tax structures. Another important element is: (8) the setting up of a ‘core’ inflation rate, as opposed to the overall CPI or ‘headline’ inflation rate. The core takes out oil and agricultural products whose prices are easily affected by external shocks and weather disturbances. This is an important element since even if overall ‘headline’ inflation rate is not achieved, as long as the ‘core’ inflation rate is within the inflation target, the central bank can also opt not to do anything. A central bank report (Guingundo, 2005) mentions two important things: (1) the current inflation targeting method allows for ‘ample room for judgment and discretion of policy makers’; and (2) in explaining why

282

Beyond inflation targeting

overshooting of the inflation target in the fourth quarter of 2004 did not lead to increases in policy rates by the central bank, the publication gives the following explanation: (a) the inflationary pressures caused by supplyside shocks (oil price increase) were not susceptible to monetary action, since the latter would work on the demand side; (b) the central bank forecasts indicated that the pressures would subside in 2006; and (c) there were downside risks to the overall strength of economic activity. The last point is extremely important as it proves that current central bank policy does put weight on the strength or weakness of ‘economic activity’. 13.3.2

A More ‘Benign’ Policy?

So far ‘inflation targeting’ in the Philippines from 2002 to 2006 has not led to very drastic monetary tightening unlike ‘monetary targeting’. The improvement in inflation targeting over monetary targeting can be due to a number of factors. First, the nature of the policy tools makes inflation targeting more benign than monetary targeting. To illustrate, maintaining a monetary base target will entail credit tightening since it does not allow money supply and domestic credit to grow with the economy and with inflation. On the other hand, maintaining the policy rate of the central bank (that is, keeping the reverse repurchase rate constant within reasonable limits) does not entail credit tightening and allows money supply and domestic credit to grow moderately. Thus to make monetary targeting and inflation targeting equally restrictive will entail increasing policy rates drastically, which is much more obvious and prone to public criticism compared to reducing the monetary base target (which had not been transparent and kept in IMF memorandums). The Philippines, in its inflation targeting history (2002 to present), has not increased the policy rate drastically and so inflation targeting policy in the Philippines has not mimicked monetary targeting by drastically cutting money supply and credit. More importantly, the adoption of inflation targeting was not done as a precondition to IMF conditionality but was a conscious switch decided by the central bank. Thus, using policy rates as the key instrument allows the central bank to implicitly inject growth objectives in deciding what rates to maintain, even if on paper the main goal of inflation targeting is inflation reduction. Second, the escape clauses and the use of ‘core inflation’ allow the Philippines to use similar instruments as the US Fed and to decide on policy rate changes based on the strength of the economy and not only on inflation targets. The policy has been quite lax since world and domestic inflation has not been very high between 2002 and 2006. Figure 13.9 shows that since 2003, the core inflation has always been

Quarterly average in percent (2000 = 100)

Towards an alternative monetary policy in the Philippines

283

12 10 8 6 4

Headline inflation Core inflation

2 0 1995

Source:

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

Bangko Sentral ng Pilipinas (BSP).

Figure 13.9

Headline and core inflation rates: first quarter 1995 to first quarter 2006

below the headline inflation. Inflation targets of the central bank were exceeded in periods when oil prices were rising in the world market (2004– 05) – when inflation rates went up beyond 6 percent and 7 percent. But policy rates were raised only infrequently (the last one in October 2005) since inflation targeting was implemented in 2002. The justifications were: (1) core inflation was below the headline inflation; and (2) overshooting of the inflation target was due to the oil price shock – a supply rather than a demand factor. Between 2002 and 2006, the central bank raised policy rates by only 75 basis points (due to the higher inflation led by rising oil prices in 2005 and 2006) compared to the more than 400 basis points done by the US Fed in the same period. Of course, the Philippine policy rate started at a higher level of 9 percent compared to the US Fed rate of 1 percent. 13.3.3

Alternate Tightening and Loosening in 2007

2007 saw an initial monetary tightening but ultimate loosening of monetary policy. In the first half of the year the government mopped up liquidity by allowing banks and their customers to park their money in special deposit accounts with the central bank, receiving interest rates equivalent to the repurchase rate. The official reason is too much liquidity due to massive inflows of remittances of overseas workers. What is left unsaid, however, is that the real danger of the high liquidity is that, coupled with little credit going to private businesses, the high liquidity could possibly lead to the creation of bubbles in the fast rising equities and real property markets. But when inflation was seen to be hovering around 3 percent and below

284

Beyond inflation targeting

(way below the inflation target of 4 percent to 5 percent for 2007), the central bank reduced their policy rates by as much as 200 points in the first 11 months of 2007. An additional strong reason to do this is the alarming appreciation of the peso due to the massive inflow of remittances and portfolio inflows. In 2008 and 2009 when the global financial crisis erupted, the peso started to depreciate due to strong capital outflows. In 2008, when food and fuel inflation brought CPI inflation rate in the Philippines to the teens (see Figure 13.9), the Bangko Sentral ng Pilipinas (BSP) increased policy rates by 100 basis points in the period from June 2008 to August 2008. When food and fuel inflation waned in the fourth quarter of 2008 to the present period (March 2009), the BSP reduced policy rates by 125 basis points between December 2008 to March 2009. The current overnight deposit rate and overnight lending rates of 4.75 percent and 6.75 percent respectively, are the lowest since inflation targeting started in January 2002.

13.4

THE NEED FOR A HOLISTIC VIEW AND AN ALTERNATIVE MONETARY POLICY

Even with the laxer monetary policy, this has not led to adequate credit expansion for investment needs, better employment prospects and a more stable external financial sector. It is clear that the financial, monetary, fiscal, external and real sectors are integrally linked and affect one another critically, and one cannot expect monetary policy alone to solve the macro problems. 13.4.1

The Recent Fiscal Crisis

From the mainstream point of view, the biggest challenge to the current Philippine macroeconomy after the Asian crisis is the large fiscal deficits and public debt burden in the period 2002 to 2006. This problem is resurfacing in the global recession years of 2009 and beyond as tax collection is expected to fall with the economic slowdown and government spending will have to increase to pump prime the economy. Fiscal surplus had been achieved by the Ramos administration before the Asian crisis. The Asian crisis, however, brought back fiscal deficits. Table 13.1 shows fiscal deficits worsening after the crisis, with the national government deficit reaching more than 5 percent of GDP in 2002. Public sector borrowing requirements reached more than 6 percent in the same year as the deficit of government corporations, especially that of the National Power Corporation, became larger.

285

462.5 17.4 15.6 19.3 −1.9 −4.2 35.6 21.6 14.0

National Govt Revenues (P Billion)

As % of GDP: National Government (NG) Revenues of which: Tax National Government Expenditures National Government Surplus/Deficit (−)

Public Sector Borrowing Requirements

As % of NG Revenues: NG Debt Service Payments Interest Principal

101.5 32.8 68.7

59.6 32.9 26.8

42.9 22.2 20.7

−4.6

16.1 14.5 19.9 −3.8

478.5

1999

108.0 32.1 75.9

64.6 31.8 32.7

44.3 27.4 16.9

−5.2

15.3 13.7 19.3 −4.0

514.8

2000

106.0 32.7 73.3

65.7 34.4 31.3

48.4 30.8 17.6

−5.2

15.6 13.6 19.7 −4.0

567.5

2001

110.2 34.4 75.9

71.0 37.1 33.9

61.9 32.1 29.8

−6.8

14.6 12.8 19.9 −5.3

578.4

2002

118.2 35.7 82.5

78.2 39.7 38.5

73.5 35.4 38.1

−6.4

14.8 12.8 19.5 −4.6

639.7

2003

Bangko Sentral ng Pilipinas (BSP); Bureau of Treasury, National Economic and Development Authority.

94.6 35.2 59.5

As % of GDP: Total Public Sector Debt Domestic Foreign

Source:

56.1 31.9 24.2

Total National Government Debt Domestic Foreign

As % of GDP:

1998

Fiscal and public finance figures for the Philippines

Item

Table 13.1

109.8 35.4 74.4

79.0 41.5 37.5

85.1 36.9 48.2

−5.8

14.5 12.5 18.4 −3.8

706.7

2004

87.2 31.7 55.6

n.a.

16.3 14.3 17.4 −1.1

979.6

2006

93.4 33.0 60.4

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

As of July 2006 72.3 69.6 40.2 38.2 32.0 31.4

83.2 36.7 46.5

−3.4

15.1 13.0 17.8 −2.7

816.2

2005

286

Beyond inflation targeting

7

6

5

Economic services

Percent

Social services 4

Defense

3

General public services Net lending Interest payments

2

1

Source:

19 97 19 98 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06

19

96

0

Department of Budget and Management (DBM).

Figure 13.10

National government expenditures, percentage of GDP

The gravity of the fiscal problem is due to the fact that the post-Asian crisis economic recovery from 1999 to 2005 had failed to improve tax effort. The tax effort peaked at more than 17 percent of GDP in 1996 and 1997 and consistently fell after that and bottomed at 12.5 percent in 2004 (Table 13.1). The falling tax effort despite significant GDP growth forced the government to undertake substantial tax reforms to respond to downgrades by the rating agencies of the sovereign debt. The tax effort improved to around 14.3 percent in 2006 mainly due to the expanded coverage and higher value-added taxation (raised from 10 percent to 12 percent). The falling tax effort and rising public debt burden after the Asian crisis brought about a serious situation wherein public debt service – principal and interest debt payments – made up 85 percent of government revenues in 2005 and 87 percent in 2006 (Table 13.1).3 The improving fiscal deficits from 2003 to 2006 were brought about only because total non-debt expenditures of the government (as percentages of GDP) – especially social and economic services – were cut drastically as interest payments increased. Table 13.1 shows total national government expenditures falling to 17.4 percent of GDP in 2006 from more than 19 percent in 1998. Figure 13.10 shows the gravity of the situation as economic and social services as percentages of GDP continuously fell in recent years while the share of interest payments went up.

Towards an alternative monetary policy in the Philippines

287

Despite improvements in the tax effort in 2006, the high public debt burden is expected to continue for some time as Table 13.1 indicates. Public investment (and other economic and social spending) has yet to recover from cutbacks, while private investments remain low despite high economic growth. In the first quarter of 2007, despite a whopping 6.9 percent GDP growth, the tax effort fell to 12.2 percent (from 13 percent in the same period of 2006). The inability of revenue generation to keep pace with economic growth does not bode well for public investment and the infrastructure needs of the country, which have been neglected in recent years due to fiscal tightness. This is most relevant in 2009 when pumppriming is vital to prevent the economy from being adversely affected by the global economic recession, which already has reduced exports by 40 percent to 50 percent in the first two months of 2009. During the period of high issuance of government securities to finance the deficits in 2002 to 2006, there was no significant rise in the treasury bill rates since the financial institutions – teaming with liquidity – preferred government securities to private loan assets (after having been burnt in the Asian crisis). This enabled the Treasury to issue treasury bonds at even lower rates. 13.4.2

More Open Capital Accounts and More Volatile Exchange Rate Movements

Figure 13.2 shows that since the 1980s the exchange rate has been very volatile. Obviously the exposure to financial and capital account liberalization has created a Pandora’s box of dangerous short-term capital flows and volatile movements. Even the years following the Asian crisis – from 1999 to 2005 – have been periods of volatile short-term portfolio flows with international capital shunning the country due to political instability and fiscal problems. Starting 2005, short-term capital came into the country due to the increases in VAT and expectations of the narrowing of the fiscal deficits. But they intermittently flowed out also due to political instability, increases in the US interest rates in 2005 and early 2006, and ultimately the sub-prime fiasco in the US that brought stock markets tumbling down for high-yielding emerging markets like the Philippines. In late 2006 and throughout 2007 the strong appreciation of the peso due to remittances of overseas workers, short-term capital inflows and a generally weak dollar internationally is causing major concerns among exporters, overseas workers and domestic manufacturing competing with imports. On the other hand, other domestically oriented sectors are happy as inflationary pressures are stemmed and economic growth is stimulated.

288

Beyond inflation targeting 90.000 M2, % of GDP Domestic credit, % of GDP

80.000 70.000

Percent

60.000 50.000 40.000 30.000 20.000 10.000

Source:

54 19 57 19 60 19 63 19 66 19 69 19 72 19 75 19 78 19 81 19 84 19 87 19 90 19 93 19 96 19 99 20 02 20 05

51

19

19

19

48

0.000

International Financial Statistics (IMF).

Figure 13.11

M2 and domestic credit, percentage of GDP

Trade deficits may be expected to deteriorate (although this has not happened very significantly yet). But the latest picture of the Philippines has changed from one of high current account deficits during periods of high growth (see Figure 13.6) to consistently positive current account balances (despite trade deficits) because of the massive influx of foreign exchange earnings of the overseas workers. 13.4.3

Low Financial Confidence and Credit to the Private Sector

The post-Asian crisis period has been marked by low financial confidence due to the trauma of non-performing assets suddenly increasing and due to stringent financial supervision to achieve higher capital adequacy ratios and loan loss provisions. Figure 13.11 shows M2 (money plus quasimoney) and domestic credit as percentages of GDP. It is clear that the Philippines has not achieved substantial financial deepening as M2 and domestic credit fell drastically from its 1997 peak. The lack of financial deepening is partly caused by the various recession and monetary

Towards an alternative monetary policy in the Philippines

289

tightening periods. Another big reason is the decline in financial confidence due to the financial crises. This is very clear in the decline of domestic credit (as percent of GDP) from the mid 1980s until 1992 due to the financial and economic collapse of 1984–85. It is happening again in the post-Asian crisis period as both domestic credit and M2 as percentages of GDP declined from 1998 until 2004. But the picture of the Philippines in 2000 to 2008 changed from one of high current account deficits during periods of high growth (see Figure 13.13) to positive current account balances (despite trade deficits) because of the massive influx of foreign exchange earnings of the overseas workers. The positive current account balance is being threatened in 2009 and beyond as the global economic recession has significantly reduced exports and is threatening the remittances of the overseas Filipinos workers who are at risk of losing their jobs. The Asian crisis increased financial regulations on banks. The Basel international standard for minimum capital adequacy (net worth to risk asset) ratios of banks is currently at 8 percent. The Philippines has been more stringent and imposed a minimum capital adequacy ratio of 10 percent starting in 2001. The actual current average capital adequacy ratio for Philippine banks ran to a high 15 percent to 16 percent in 2005. This situation, together with the perennial political crisis, has so far discouraged banks from aggressively lending to the private sector. The banking system is awash with liquidity with a strong appetite for government securities – whether peso or dollar denominated – rather than private lending. This will constrain investments and employment generation in an economy with still underdeveloped long-term capital markets. 13.4.4

Persistently High Unemployment and Low Investment Rates

Figure 13.4 shows persistently high unemployment in the latest economic recovery period of 2000–04 despite positive economic growth. Figure 13.12 gives more detail on the employment picture. It shows that a major trend in the employment picture is the downward employment absorption capacity of agriculture and the low and stagnant employment absorption capacity of industry. The only sector adequately absorbing the growing labor force is the service sector. It must be pointed out that the industrial and agricultural sectors are the main tradable sectors. With increased trade liberalization, globalization and competition among countries, these sectors are now exhibiting increasing output-employment ratios as output increases are not matched by equivalent employment increases. This means labor shedding and labor-cost cutting in areas whose products are facing stiff competition from imports. Thus, services, which is largely a non-tradable sector, becomes

290

Beyond inflation targeting

45 40 35

Percent

30 Agriculture Industry Services Unemployed

25 20 15 10 5 0

1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

Source:

Labor Force Survey, National Statistics Office.

Figure 13.12

Employment by sector and unemployed, as percentage of labor force

the biggest absorber of employment. But this is not enough to absorb the expelled labor from the tradable sector, and the new labor force entrants (see Lim and Bautista, 2006). Of course the result is that unemployment remains persistently high, hovering at 10 percent and 11 percent. This situation leads us to explore whether a more employment-sensitive monetary policy can help alleviate unemployment and underemployment through an integrated scheme of credit allocation to labor-intensive and employment-generating activities. Related to this is the fact that investment rates have fallen since the Asian crisis (as is true for most of the East Asian economies). Investments as a share of GDP fell from above 24 percent before the Asian crisis in 1997 to 21 percent in 2000 and to 14.8 percent in 2006. The steady and continuous fall of the investment rate, partly due to the collapse in public and infrastructure investments due to the fiscal bind, does not bode well for future supply capacity and productivity improvements in the economy, even as the GDP growth rate has improved considerably in 2004 to 2007. This is another area where monetary policy may play a more active role. A lax and accommodating monetary policy is even more urgent in the global recession period of 2009 and beyond when exports are plunging and

Towards an alternative monetary policy in the Philippines

291

consumption and investments are being threatened by possible declines in remittances from overseas Filipinos and losses in domestic business confidence because of losses in the stock market, depreciating currency and adverse global conditions. 13.4.5

External Shocks and High Oil Prices

The volatile world oil price movements starting in the second half of 2004 has increased the average inflation rate in the Philippines from a low of 3 percent in 2002 and 2003 to more than double in 2005 and 2006. The continuing threat of higher oil prices may lead to tighter monetary policies at home and abroad. Thus, a policy to respond to increasing world oil prices and return to higher world inflation and interest rate regimes is now an urgent task. If this supply-led inflation leads to a policy of fighting inflation through demand-suppressing methods of increasing interest rates, recessionary pressures may again become a major monetary policy.

13.5

COMPONENTS OF AN ALTERNATIVE MONETARY AND FINANCIAL POLICIES

13.5.1

An Undervalued Currency and Tax-Based Capital Controls on Inflows

The first component of the recommended alternative scheme is a reasonably pegged exchange rate regime, targeted at an undervalued level during good and healthy periods of global trade, and at a slightly overvalued level during bad times of global trade contraction. (This recommendation is consistent with the policy prescriptions of Frenkel and Taylor, Chapter 2, this volume.) The peso has depreciated significantly in 2008 and 2009 due to the massive capital outflows from the stock market and the significant fall in exports during the current global economic recession. A weak currency is not healthy during collapsing global trade and shrinking export markets for this brings about stagflationary tendencies on the domestic economy without any positive effects from the export market. This puts the central bank in a dilemma as a lax monetary policy may aggravate the peso depreciation. This is true because of a flexible exchange rate regime coupled with complete capital account liberalization. A healthy economy requires an exchange rate that does not appreciate too much during good times when the economy and world economy is

292

Beyond inflation targeting

strong. It also requires an exchange rate that does not go on a freefall collapse during the bad times. This requires capital controls and a reversal on the capital account liberalization that most emerging markets had undertaken in the 1990s. It would therefore be practical to propose a tax-based (or market-based) capital controls on inflows like Chile and Malaysia, or quantitative restrictions on entry and exit of ‘hot money’ to and from the short-term equity and bond markets of the country, like China and India. Because of the difficulty in undertaking this (note the recent unsuccessful Thai attempt to tax ‘hot money’), it is suggested that a regional ASEAN13 effort to undertake a common policy on capital controls be explored. 13.5.2

Incorporate Output and Employment Targets to the Current Inflation Targeting Regime

If one envisions a transition away from the inflation targeting regime, it would be wise to recommend that output and employment goals be explicitly stated as part of the objectives of monetary policy. After all, not adversely affecting economic activity has explicitly been cited by the central bank as one of the reasons why policy rates were not increased drastically despite inflationary pressures from world oil prices in 2005 and 2006. 13.5.3

A More Active Role in Stimulating the Economy

The above monetary policy can have some beneficial impact, especially on the current problems of high fiscal deficits, lack of financial confidence and unemployment. The more accommodating monetary policy may be complementary to the moves to improve and increase financial loans in the post-Asian crisis period, and to offset the natural conservative tendencies in credit expansion due to higher capital adequacy ratios and loan-loss provisions. It is crucial in the current economic slowdown brought about by the global economic recession. Furthermore, this will create a better atmosphere for involving credit allocation in employment generating activities to be discussed in the next section. Finally, since oil prices have returned to normal levels and world interest rates are low in the current period of global economic recession, the more accommodating policy may allow some room for monetizing the fiscal deficits inasmuch as the pass-through of monetary increases to inflation has been accepted as weak and unstable. Fiscal expansionary policies – especially in social, economic and infrastructure spending – are vital in returning the system to quality and employment-generating growth. And due to the high debt-

Towards an alternative monetary policy in the Philippines

293

to-GDP and debt service ratios of the Philippines, it is difficult to support fiscal expansion with higher public debts. Monetizing the fiscal deficit will be difficult to implement practically since the current authorities and the IMF have succeeded in institutionalizing the non-monetization of fiscal deficits. Strong political will from the national government and monetary authorities will again be required. 13.5.4

Credit Allocation to Stimulate Investments in High Value-Added and Employment Generating Sectors

Inasmuch as unemployment is a major problem in the post-Asian crisis period and investment rates are low particularly in sectors that may have positive externalities for the economy and inasmuch as the financial sector is reluctant to use market mechanisms to lend to the private sector, it is worth considering whether targeted credit programs can help in alleviating the unemployment and low investment problem. However, the Philippines since the 1980s has moved away from subsidized and targeted credit schemes as the financial liberalization school has predicted distortionary and adverse effects from such policies, and because the bad experiences with the Marcos administration has convinced many economists and technocrats that subsidized and targeted credit to potential cronies is detrimental. Relying more on the financial markets and the private sector, they figure, is more beneficial than government interventions. Thus, by September 2002, the central bank rediscount window has been liberalized to allow a generalized and uniform access to the facility by all sectors of the economy at market rates. The use of the facility has been reoriented for money supply management (complementing open market operations) instead of selective credit allocation (such as to exports and small-scale industries) and development financing (Guinigundo, 2005). There are some targeted credit schemes outside the scope of the central bank administered by the Land Bank of the Philippines (LBP) and Development Bank of the Philippines (DBP) targeted at agricultural cooperatives, farmers’ groups and small-scale industries with funds from the Department of Agriculture (DA), Department of Agrarian Reform (DAR), other agencies and multilateral organizations. The Department of Trade and Industry (DTI) also has some credit lines targeted to small- and medium sized enterprises (SMEs). The current system has become rather schizophrenic as the formal system and the big government agencies such as the National Economic Development Authority (NEDA) and the central bank promote financial liberalization and reduction of targeted credit in the formal sector. But the Arroyo government currently promotes

294

Beyond inflation targeting

microfinance lending by government-supported agencies as one of its key anti-poverty strategies. Although there are some striking successes by some rural banks, cooperative banks and other microfinance units in providing credit to develop local economies and some economic sectors, this is not being mainstreamed. The targeting of SMEs and micro-enterprises, largely in the informal sector, for microfinance has positive aspects inasmuch as these entities have high employment generating potentials. However, the provision of microfinance has oftentimes been politicized and lacks a holistic approach of providing the SMEs and micro-enterprises access to markets and linkages in the formal sector, access to technology and good management practices, skills development and technical assistance for product development. This, unfortunately, is consistent with the overall trade and industrial policy that the government has been following for several decades, which prohibits ‘picking winners’ and frowns at promoting priority economic sectors (closely subscribing to the policies of the ‘Washington Consensus’ view and World Trade Organization (WTO) rules). Thus credit allocation in this setting becomes directionless and limited to providing small ‘livelihood’ programs for some targeted poor areas rather than to permanent, productive and growing industries and employment for a vibrant economy to benefit a wider pool of poor and low-income families. It is therefore recommended that a targeted credit allocation program be set up by the central bank and related institutions (such as key state banks) to give prioritized credit to key sectors that satisfy the following: 1.

2.

3. 4.

Exhibit strong potential for successful take-off, viability and high repayment but may require lumpy investments and/or suffer the ‘first mover’4 problem. Have strong employment generating effects, and/or strong interlinkages with the other sectors of the economy, which will lead to multiplier effects in the economy. Have technology or knowledge spillover effects, as well as economies of scale. Are part of an integrated set of industries that suffer coordination failure problems.

Market failures, endogenous growth and strategic trade theories are now mainstream economic theories that justify the above interventions. It is only the strong resistance of institutions that support the Washington Consensus that prevent developing countries from applying the policy implications of the new economic theories.

Towards an alternative monetary policy in the Philippines

13.5.5

295

Heterodox Policies of Income Policies, Price Controls and Price Stabilization

Since most of the inflationary pressures are from supply-side shocks, other means may be more productive than an immediate monetary demandreduction response. In many of the weather and natural calamity shocks the government has, for practical reasons and with some success, gone into temporary price controls and constant monitoring of basic foodstuffs and imposing heavy penalties on hoarding. Importation of agricultural products is also undertaken during periods of agricultural shortages. These policies should be continued and enhanced. Improvements are especially needed in the area of equitable, transparent and efficient distribution of imported foodstuffs during periods of agricultural shortages. The inefficient distribution system and lack of transparency in many of the sales of the National Food Authority (NFA) during periods of food and agricultural shortages call for an overhaul in the system and more participation of the private sector in the distribution of temporarily imported foodstuffs during periods of shortages. The oil price shocks had led many in the Philippines to question the deregulated structure of the oil industry. There is a perception that there is an asymmetry of quick price increases during times when world prices are rising, but slow and lagging price decreases when world prices are falling. The question of a domestic oil cartel has arisen (led by Shell, Caltex and a previously government-owned oil company, Petron), as many new players are finding it hard to compete in a setting of volatile world prices. Regulatory boards still regulate electricity charges and transportation fares, while the Department of Energy has clout to stop unreasonable price increases in oil and gasoline products. Thus, the importance of regulation in public utilities and oil/gasoline products become crucial in fostering competition and stopping cartel-like pricing. The Philippines also lacks anti-trust legislation that will remove monopoly and predatory pricing in key economic sectors. If world oil prices increase significantly again, it is recommended that an oil price stabilization fund (used moderately successfully in the 1970s by the Marcos government) be explored to tackle the situation. Equally important is the need to develop, in the medium and long run, alternative fuel sources and to re-enact laws giving special tax incentives for firms providing these alternative fuel sources. The Philippines has an incomes policy that deliberately keep wages low and lagging behind price increases even as their monitoring and implementation are grossly inadequate. With labor productivity increasing, this has led to lower inflation but has also led to a long-run decline in real wages, which may have contributed to worsening income distribution and

296

Beyond inflation targeting

poverty. A more balanced approach that this chapter recommends involves the following: (1) depending more on tripartite agreements derived by government, labor and employers to impose agreed-upon ceilings on price and wage increases; (2) enact laws and rules that punish monopoly and cartellike pricing (especially for the oil and telecommunications industries); and (3) do more monitoring and temporary regulation of key products, such as food, oil, telecommunications and rent-control housing, especially during periods of price instabilities. The above set of alternative monetary policies points to a need for a change in mindset from a dichotomy between the financial and real sectors to a viewpoint wherein the financial sector is integrally linked and supportive of the real sector. The strategy is to coordinate and link the current targeted credit programs in the anti-poverty strategy campaign with the initiatives of the economic departments and agencies in developing and promoting key priority economic sectors that have high value-added, high technology spillover, multiplier effects and employment generating potentials. Successes in this arena will hopefully spill over to the formal and big business sector of the economy. 13.5.6

Testing the New Alternative

Inasmuch as the central bank already has a long-term and a short-term macroeconometric model, it is suggested that the new alternative be tested employing similar types of macro model. The new econometric model poses some challenges: (1) How to model a more discretionary monetary policy based on multiple objectives of both output/employment generation and price stability (various simulation scenarios and quadratic loss functions might be used). (2) How to define and model ‘overheating’ and excessive aggregate demand in the model to identify when some monetary contraction or aggregate demand reduction may be beneficial to the economy and to identify when monetary contraction will be unreasonably recessionary. (3) How to model the incomes policies and temporary price controls. (4) How to model increased credit allocation and incorporating its impact on the growth of key economic sectors with high value-added and employment generation. No doubt different scenario simulations are needed (with low, medium and high scenario assumptions) to make the models more useful to policy makers. 13.5.7

A Difficult Endeavor

The above alternative set of policies runs counter to the current macro policies of the Philippine government and go beyond policies traditionally reserved

Towards an alternative monetary policy in the Philippines

297

for the central bank. Thus the adoption of the alternative monetary policy is a formidable endeavor that requires a change, not only in the inflation targeting regime, but also in the current macro framework of the country. To achieve this, one does not need to drastically change the mandate and autonomy of the central bank. What is needed is perhaps a gradual but sure and progressive change in its attitude. The first hurdle of shifting away from overemphasizing inflation over growth has been achieved. The next step is just to make this more explicit. The hardest part is convincing not only the central bank but the entire government, business and financial sectors of the country that high and sustained growth requires an active industrial policy supported by monetary and credit policies. The last hurdle is the most difficult one as it entails a rejection of the Washington Consensus and the adoption of a new and fresh perspective. This is what heterodox economists, including those involved in this volume, are working and striving for. The current global financial crisis – triggered by a deregulated financial system existing side by side inflation targeting regimes that jacked up global interest rates, which ultimately led to the subprime fiasco – may give many economists and policy makers cause to question the old monetary paradigm and pave the way for a framework that provides stronger and healthier links between the financial/monetary sector and the real and productive sector of the economy.

NOTES 1. The author is grateful to Francis Dakila, Digna Paraso and Zeno Abenoja of the Bangko Sentral ng Pilipinas for their support and help. He would also like to thank Gerald Epstein and Erinç Yeldan for their positive comments and suggestions in improving this chapter. All mistakes are the author’s. 2. These are cash and short-term government securities which a bank is allowed to keep as required reserves. 3. The increase in 2006 was, however, due more to conscious prepayment of foreign debt (as the peso became strong) and domestic debt (as interest rates started to fall). 4. Haussman and Rodrik (2002) refer to the market failure problem of ‘information spillover’ or ‘first mover’ problem as the problem wherein the first mover bears all the risks. If they succeed, others will imitate them and reduce their market share. If they fail, they will bear all the losses.

REFERENCES Dakila, F. (2001), ‘Evaluation of alternative monetary policy rules: a simulationbased study for the Philippines’, PhD thesis, University of the Philippines, Diliman.

298

Beyond inflation targeting

Frenkel, R. and L. Taylor (2006), ‘Real exchange rate, monetary policy and employment’, DESA working paper no. 19. Guinigundo, D. (2005), ‘Inflation targeting: the Philippine experience’, Bangko Sentral ng Pilipinas, Metro Manila. Haussman, R. and O. Rodrik (2002), ‘Economic development as self-discovery’, National Bureau for Economic Research working paper no. 8592. Lim, J. and C. Bautista (2006), ‘External liberalization, growth and distribution in the Philippines’, in L. Taylor (ed.), External Liberalization in Asia, Post-Socialist Europe, and Brazil, New York: Oxford University Press, pp. 267–310. Lim, J. and M. Montes (2002), ‘Structural adjustment after structural adjustment, but why still no development in the Philippines?’, The Asian Economic Papers, 1 (3) (Summer), 90–119.

14.

Monetary policy in Vietnam: alternatives to inflation targeting Le Anh Tu Packard1

14.1

INTRODUCTION

One of the most important challenges facing policy makers is to determine how monetary policy should be conducted in order to meet their country’s national development goals. In recent years a growing number of central banks have convinced each other that the siren song of inflation targeting is worth pursuing,2 even though a strong theoretical case that this monetary rule possesses superior welfare properties has yet to be established. Inflation targeting calls for the ‘explicit acknowledgment that low and stable inflation is the overriding goal of monetary policy’, which implies that a low inflation target should have supremacy over other development objectives.3 For Vietnam, the quest for a pro-development monetary policy has become more urgent because the country is entering a new developmental phase that will be shaped by the terms of its accession to the World Trade Organization (WTO) and commercial treaties with its trading partners. Mindful of both the opportunities and risks that come with this phase, the Vietnamese government has been looking into macroeconomic and monetary policy guidelines to manage this period of unprecedented exposure to the world economy. In keeping with recent fashion among central banks, the State Bank of Vietnam (SBV) expressed interest in exploring the feasibility of inflation targeting. However, the general consensus is that at present Vietnam does not meet the necessary conditions to implement an inflation targeting regime because the central bank lacks adequate tools to carry out an effective inflation targeting monetary policy. Additionally, other offices of economic management in the Ministries of Finance, Planning and Investment, Industry and Trade do not view inflation targeting as a matter of urgency, nor are they willing to cede to the SBV that degree of power. There is also the larger question of whether an inflation targeting regime is compatible with Vietnam’s development priorities, and whether it would help improve economic performance over the long run. 299

300

Beyond inflation targeting

The context for discussing monetary policy in Vietnam is as follows: upon launching sweeping reforms during the 1990s, the country has generally followed the East Asian ‘developmental state’ model. In the view of its political leaders, monetary policy should serve as a tool to meet the country’s socioeconomic development goals, which are rapid and sustainable growth, modernization, industrialization and poverty reduction. According to the 1998 Law on the State Bank of Vietnam, the task of the central bank is to stabilize the value of the currency, secure the safety of the banking system and facilitate socioeconomic development in keeping with the nation’s socialist orientation (Kovsted et al., 2002, Thao 2004). SBV senior officials consider this a mandate to control inflation and promote economic growth (Nghia, 2005). The implied assumption shared by policy makers and the public is that the nominal exchange rate and the domestic price level are closely linked. This link became highly visible during the Doi Moi (Renovation) reforms when Vietnam was grappling with hyperinflation and large depreciation of the parallel market exchange rate. Unlike many other developing countries, Vietnam has a strong domestic constituency for low inflation because of its earlier traumatic experience with hyperinflation, which heightened public sensitivity to price movements. That said, a strong preference for low inflation and willingness to carry out policies to support a low inflation environment does not mean that the central bank has a clear mandate to adopt formal inflation targeting. This chapter aims to contribute to the search for the right mix of macroeconomic and monetary policies that can best serve Vietnam in the coming period of greater openness and intensified competition in both domestic and global markets. It examines the factors that should guide monetary policy, taking into account the current state of Vietnam’s transition to a more market-oriented economy and the challenges posed by dollarization, financial repression, informal and underdeveloped financial markets, and rapid international economic integration. Not surprisingly, it finds that critical gaps in knowledge, institutional arrangements, tools and rules are impeding the effectiveness of monetary policy. Sharing the view that the main task of the central bank should be to maintain macroeconomic prices that are conducive to rapid and sustainable economic growth, an alternative to inflation targeting is proposed. To support Vietnam’s transition to a more market-oriented economy, the central bank should instead target a real exchange rate (RER) that is stable and competitive. It is argued that this key relative price has a more powerful influence on the allocation of labor and capital, and on the composition of domestic output, than administrative levers typically employed by centrally planned economies. Maintained over an extended period, a

Monetary policy in Vietnam: alternatives to inflation targeting

301

stable and competitive RER promotes an efficient allocation of resources and employment-creating growth, reinforces macroeconomic and financial stability, and encourages financial market development. The chapter argues that a stable and competitive RER is a superior intermediate target for several reasons. First, this target clearly is consistent with the Law on the State Bank of Vietnam,4 which states that the SBV’s task is to stabilize the value of the currency. Since Vietnam now has multiple important trading partners, the value of the domestic currency should be stabilized against a trade-weighted basket of currencies that take into account differences in their respective rates of inflation. Second, it improves the transparency of monetary policy and strengthens confidence in the central bank’s ability to conduct monetary policy effectively. In other words, the central bank is assigned a task that is realistic and therefore doable. Third, a stable and competitive RER can contribute substantially to economic growth and employment creation if it is supported by complementary fiscal, monetary and industrial policies. Fourth, it can have positive medium- to long-term impacts on structural change and development through a variety of channels: resource allocation, changes in production techniques and growth of capital stock including stock of human capital (Frenkel and Taylor, 2005).5 Fifth, compared to a strict focus on inflation targeting which tends to slow economic growth and lower employment growth (Epstein, 2003), a RER target is more likely to be a more effective stabilizing force and can do a better job in dampening output volatility during periods of global turbulence. A stable and competitive RER’s long-term positive impact on resource allocation and the composition of output takes place through its influence, both direct and indirect, on key macroeconomic prices such as the domestic interest rate, the relative price of traded to non-traded goods, the relative cost of capital and labor, and the import-export price ratio. The RER, used in conjunction with appropriate commercial and industrial policies, can serve as a development tool in coordination with other monetary policy instruments to strengthen the economy’s overall competitiveness, increase aggregate productivity, maintain external balance, contain inflation and stabilize asset markets (Frenkel and Taylor, 2005). International evidence from cross-country empirical research provides support for this view: instability (variability) of the RER is found to be negatively related to growth (Corbo and Rojas, 1995; Montiel, 2003), and overvaluation of the RER (in other words, an uncompetitive RER) has been linked with slower growth (Montiel, 2003; Razin and Collins, 1997). The chapter is organized as follows: Section 14.2 provides a brief history

302

Beyond inflation targeting

of Vietnam’s banking system. Section 14.3 analyses the macro economy from the perspective of identifying transmission mechanisms. Section 14.4 examines the issues surrounding the framework for monetary policy. Finally, Section 14.5 describes the merits of a stable and competitive RER as a superior alternative to inflation targeting, and offers some concluding comments.

14.2

VIETNAM’S BANKING SYSTEM: BRIEF HISTORY

From 1976 to 1989, like other centrally planned economies, Vietnam’s single-tier banking system was owned and controlled by the state. The SBV provided nearly all domestic banking services through a vast branch network. Bank lending was state directed, and credit rationing was imposed because financial resources were scarce. During this period, SBV offices served as the interface between state planning, the national budget and state entities including some 12 000 state-owned enterprises (SOEs).6 Under central planning, the SBV was not required to carry out many traditional functions of commercial banking, such as credit analysis or risk management, and its main task was to ensure that financial resources were allocated to economic units in accordance with the plan. The quantity of currency outside banks was very high (the ratio of currency outside banks to nominal GDP reached 9.2 percent in 1986) as the government attempted to monetize sharply rising fiscal deficits as revenue growth failed to keep pace with rising expenditures (World Bank, 1991).7 SOEs in Vietnam lacked fiscal discipline as they operated under the soft budget constraint that was common among socialist countries. To circumvent credit rationing, they engaged in unauthorized credit creation through various means such as abuse of the check payment system and use of supplier credits8 as a substitute for borrowing in credit markets. These practices had inflationary consequences, created financial problems for the SBV and contributed to deterioration in the consolidated balance sheets of SOEs. Before money and capital markets were established during the 1990s, household liquid and semi-liquid assets mainly consisted of the domestic currency, gold, hard currency notes, and easily tradable commodities, such as rice. Remittances from overseas Vietnamese9 contributed to the dollarization of the economy. Continued efforts by households and other economic agents to protect themselves from inflation by reducing their domestic currency holdings (causing the ratio of currency outside the

Monetary policy in Vietnam: alternatives to inflation targeting

303

14 Inflation rate: monthly Nominal household 3-month deposit rate: monthly Nominal household demand deposit rate: monthly Real household demand deposit rate: monthly

12 10 8 6 4

Sources:

May-90

Apr-90

Mar-90

Feb-90

Jan-90

Dec-89

Nov-89

Oct-89

Sep-89

Aug-89

Jul-89

Jun-89

–4

Apr-89

0 –2

May-89

2

State Bank of Vietnam, World Bank (1991, p. 86, 1992).

Figure 14.1

Inflation and interest rates, Vietnam 1989–90

banking system to GDP to decline from 9.2 percent in 1986 to 6.6 percent in 1988) only worsened the inflationary spiral. The 1987–89 macroeconomic crisis and hyperinflation provided the impetus for the comprehensive and coordinated Doi Moi reforms. In 1988 the Prime Minister signed Decree No. 53/ND which ended the monobank system and created a two-tier system consisting of the SBV as the central bank and four state-owned commercial banks (SOCBs). In addition, the government ended the state monopolies on financing foreign trade and on providing long-term finance. The intent was to increase management autonomy and responsibility, and to introduce the pressure of competition in order to improve bank performance (ibid.). In 1990 the government promulgated two banking ordinances for a two-tier banking system. These ordinances transformed the SBV into a central bank with oversight over the domestic banking system and provided the legal framework for commercial banks and other financial institutions. The government liberalized entry into the banking system and lifted rules on sectoral specialization of the SOCBs. Commercial banks were given responsibility for the operation and control of their finances and implementation of universal banking activities. The decision to raise the interest rate for household deposits in the formal banking system increased confidence in the domestic currency and encouraged households to deposit their dong assets in bank accounts (Figure 14.1). In 1989 RERs on households rose sharply, and encouraged a steady rise in the value of household deposits from VND 207 billion in March 1989 to VND 1348 billion by January 1990 (Figure 14.2). As regards the current conduct of monetary policy, the broad division of labor currently is as follows: the government’s task is to prepare a plan

304

Beyond inflation targeting Real household 3-month deposit rate: monthly Household monthly bank deposits

14 12

1 600 1 400 1 200

10

1 000

8

800 6

600

4

400

2

200

0

0 Apr-89

Sources:

Jun-89

Aug-89

Oct-89

Dec-89

SBV, World Bank (1992).

Figure 14.2

Household deposits, Vietnam

for monetary policy that includes an inflation forecast, while the National Assembly’s tasks are to set an annual target for the inflation rate that is consistent with the state budget and economic growth objectives, and to supervise the implementation of monetary policy. As it currently stands, Vietnam’s monetary policy strategy is but a component of the broader five-year socioeconomic development strategy that is formulated by the government and the ruling Communist Party. Within this framework, the SBV’s role is to come up with a concrete action plan for the banking sector, which includes setting targets for the amount of liquidity needed by the economy. Specifically, the SBV announces annual targets for total liquidity (M2 growth) and credit growth. However, the goal of keeping within credit growth targets does not appear to receive high priority, as actual credit growth in both 2004 and 2005 were well above the targeted growth rate of 25 percent. The SBV is also tasked with stabilizing the exchange rate,10 an appropriately vague mandate that ought to direct the authorities to maintain a trade-weighted inflation-adjusted exchange rate that promotes sustainable (non-inflationary) growth over the medium to long term. Like China and Singapore, the Vietnamese currency is officially pegged to a basket of currencies, and like China, the authorities do not disclose the currency composition in the basket nor the basket weights. From January 2000 to December 2006, Vietnam’s real effective exchange rate has depreciated measurably, especially in comparison to China and its South East Asian neighbors (Figure 14.3). This depreciation corresponds to a moderate acceleration of Vietnam’s GDP growth rate from 6.8 percent in 2000 to 8.4 percent in 2005, and a somewhat more rapid acceleration of the inflation rate.

Monetary policy in Vietnam: alternatives to inflation targeting

305

130.00 Vietnam Singapore

Jan 2000 = 100

120.00

China Indonesia

Thailand Malaysia

110.00 100.00 90.00 80.00

Sep-2006

Jan-2006

May-2006

Sep-2005

Jan-2005

May-2005

Sep-2004

May-2004

Jan-2004

Sep-2003

Jan-2003

May-2003

Sep-2002

Jan-2002

May-2002

Sep-2001

May-2001

Jan-2001

Sep-2000

Jan-2000

Figure 14.3

May-2000

70.00

Real effective exchange rate

14.3

DESCRIPTION OF THE MACRO ECONOMY

14.3.1

GDP and Macro Aggregates: Mechanisms of Adjustment

To carry out monetary policy effectively, policy makers in Vietnam need to have a much better grasp of the actual mechanisms of transmission and adjustment than they do at present. For example, the transmission of monetary policy via the interest rate channel is unclear because credit market segmentation, financial repression and credit rationing add additional layers of murkiness to the process. Through its short-term policy rate and commercial bank reserve requirement, the SBV is able to influence the commercial bank lending rate and activity levels of enterprises that borrow from the formal financial sector. However, its influence over credit growth in the informal financial sector and informal lending rates is not at all clear.11 Furthermore, the picture is obscured by the country’s ongoing structural transformation that has led to a gradual flattening of the investment spending curve, as investment spending becomes more sensitive to interest rate changes. This is illustrated in Figure 14.4, where the expected link between gross capital formation by enterprises and real lending rates has not emerged until after 1994. Even so, investment spending in Vietnam continues to be less sensitive to interest rate movements compared to other countries with more developed financial sectors. This is because retained earnings continue to be the main source of financing for business capital spending. To elaborate further on the problem of incomplete information, both

306

Beyond inflation targeting 30.0

Inflation rate (CPI) Lending rate, working capital (short-term), real

25.0

Gross capital formation by enterprises, % of GDP

Percent (%)

20.0 15.0 10.0 5.0

Figure 14.4

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

1993

–5.0

1992

0.0

Inflation rate, lending rate and gross capital formation by enterprises

aggregate money supply and important elements of the money demand function are unknown (Hauskrecht and Nguyen, 2004) because the economy is partially dollarized and there is no reliable data about the quantity of US dollars and stock of gold outside the banking system that are used as a medium of exchange and a store of value.12 In particular, in Vietnam the aggregate ‘true’ stock of money is hard to estimate because it includes M2 (recorded by the SBV), foreign deposits held in banks, as well as two significant unobserved variables, private sector foreign currency holdings and gold in circulation. It is also likely that the domestic and foreign currency will have different velocities (ibid.) with different trajectories, thus complicating the formulation based on the simple formulations that rest on the ‘quantity theory of money’. Figures 14.5A and 14.5B present the velocity time path for currency outside banks (V1) and for total liquidity M2 (V2), which includes currency outside banks, domestic currency deposits and foreign currency deposits. Two mutually offsetting influences on velocity deserve mention. Ongoing structural reform of the financial sector and improvements in the payments system increases velocity. At the same time, in a multi-currency economy a large-scale portfolio switch to the domestic currency can lower velocity. Figures 14.5A and 14.5B illustrates this: from 1991 to 1994 households and firms switched to the domestic currency and reduced their non-bank foreign currency and gold holdings as they trusted more the government’s ability to control inflation. This brought about a decline in velocity and also led to greater monetary deepening.

307

Velocity 1: nominal GDP/currency outside banks Velocity 2: nominal GDP/M2 Inflation rate

16

9 8 7 6 5 4 3 2 1 0 –1

14

10 8 6 4 2 0 1986

1988

Figure 14.5A

1990

1992

1994

1996

1998

2000

2002

Velocity time paths, 1986–2002

16.00

900.0

Velocity 1: nominal GDP/currency outside banks Velocity 2: nominal GDP/M2 Inflation rate

14.00

800.0 700.0

Velocity rate

12.00

600.0

10.00

500.0

8.00

400.0

6.00

300.0 200.0

4.00

100.0

Figure 14.5B

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

1993

1992

1991

1990

1989

1988

–100.0 1987

0.0

0.00 1986

2.00

Percent per annum (%)

Velocity rate

12

Percent per annum (%)

Monetary policy in Vietnam: alternatives to inflation targeting

Velocity time paths, 1986–2005

Accurate tracking of domestic credit growth is also critical to the effective conduct of monetary policy. This is yet another problem for the central bank, because key variables that affect financial sector development and domestic credit growth in Vietnam are difficult to estimate. These include the magnitude of inter-firm credit as percent of aggregate credit creation and the quality of their accounts receivable (which may pose a significant risk to the banking system). Finally complicating the task is the murky link between bank credit growth, the inflation rate and actual borrowing by business enterprises (see Figure 14.4) due to the coexistence of formal and informal financial markets, and the role of inter-firm credit. As regards exchange rates, the impact of the central bank’s exchange

308

Beyond inflation targeting

rate policy on the real economy is difficult to determine because it is not possible to predict how Vietnam’s informally pegged exchange rate regime affects the growth of monetary aggregates. The SBV does not provide information on its interventions in the foreign exchange market, and there is no explicit sterilization policy. Yet, thus far, SBV actions to manage the informal peg do not appear to have negative consequences. The M2 growth rate has not been overly volatile and the inflation rate has been low.

14.4

MACROECONOMICS AND INSTITUTIONAL FRAMEWORKS OF CENTRAL BANKING

14.4.1

Issues Surrounding Scope for Inflation Targeting

Although there is interest in inflation targeting on the part of the SBV, and a steady stream of international expert advice is provided on inflation targeting for the SBV senior management, the consensus view is that at present the conditions to support a formal inflation targeting monetary framework are not met. The reasons are evident (see Sections 14.1 and 14.2) when we consider the four main conditions outlined by the International Monetary Fund (IMF) that are deemed necessary to support such a framework (Carare et al., 2002): ●

● ● ●

The central bank has a clear mandate to make inflation targeting the primary objective of monetary policy and is publicly accountable for meeting this objective. The inflation target will not be subordinated to other objectives and monetary policy will not be dominated by fiscal priorities. The financial system is developed and stable enough to implement the inflation targeting framework. The central bank has adequate policy instruments to be able to influence inflation.

At present, the SBV has limited scope to implement monetary policy using market-based indirect instruments to influence inflation,13 although this has long been its declared objective, because financial markets are thin and not well developed (the government securities market is segmented and illiquid). In addition, as explained above, the government is only at the early stage of building the necessary foundations (including timely access to a high frequency databank of key economic and financial variables needed for policy analysis) for developing a ‘reasonable understanding of the links between the stance of policy and inflation’.

Monetary policy in Vietnam: alternatives to inflation targeting

309

Nevertheless, there are many additional reasons why a rigid inflation targeting framework is not appropriate for Vietnam, even if the conditions for inflation targeting are met. First, it gives primacy to the wrong target (inflation), forcing policy makers to operate in a framework that implicitly accords higher priority to inflation than to other more pressing development objectives. For example, it obliges the central bank to automatically adopt a tightening stance whenever the inflation indicator rises above its target range, or risk being branded as incompetent for failing to stick to the inflation target. A rigid inflation targeting framework also sets false standards for judging the quality of monetary policy, distracting policy makers from more serious and arduous efforts to understand the actual workings of their economy. Second, it is not so easy to determine what should be the right rate of inflation to target, and the SBV may find it much too tempting to simply follow the lead of other central banks even when that may not suit Vietnam’s particular circumstances. Third, it sends the wrong message about what is needed to ensure good policy making. The implicit underlying justification for inflation targeting is that policy makers cannot be trusted to make sound policy decisions. The assumption is that they tend to give in to short-sighted political demands that can harm national social welfare over the long run. Therefore, to protect against this, policy makers must be bound to tight rules and explicit objectives, and they must be held publicly accountable to meeting these objectives. Although there are valid points in these arguments, tying their hands with rigid rules and wrong targets could very well push the economy onto a different path that departs sharply from the country’s development goals. 14.4.2

Quality of Monetary policy

Surprisingly, the government managed to achieve favorable macroeconomic results in spite of having to operate somewhat in the dark (given the critical gaps in information described in some detail above) with the crudest of monetary tools to influence aggregate demand. One explanation for this success is that these constraints did not prevent the government from pursuing the ‘right’ fiscal and monetary policies (see Section 14.2). The economic administration seems to have maintained macroeconomic stability and succeeded in keeping inflation under control over a prolonged period, from 1990 to 2005. The gains from achieving this credibility can be seen in the progress made in monetary deepening, as the ratio of M2 to nominal GDP more than doubled from its nadir in 1993. The re-intermediation of foreign currency previously

310

Beyond inflation targeting

held outside the banking system also indicates greater confidence in the banking system. Thus far, policy making credibility has remained strong – despite poor information and weak monetary instruments – because key fiscal and monetary policies have been well managed. Both the stock and flow of government debt (including debt denominated in foreign currency) have stayed at prudent levels, the outlook for fiscal balance remains healthy and the expected trajectory for the current account deficit does not give cause for concern. However, the monetary authorities do not have adequate tools at present to protect the economy from exogenous shocks, which means that Vietnam’s vulnerability to external shocks will increase as its economy becomes more integrated with the global economy. Moreover, the risk of policy error is likely to increase if the government fails to address the issue of critical gaps in information. As noted in Section 14.2, the coexistence of formal and informal financial markets and the role of inter-firm credit in liquidity creation have made the relationship between bank credit growth and actual borrowing by business enterprises blurry and hard to predict. Without better information, the central bank runs the risk of misinterpreting data and may respond inappropriately, with dire consequences. For example, the SBV may attribute a ‘too high’ rate of credit growth to excessive monetary or fiscal easing, when in fact these high numbers may actually be the result of credit reallocation due to a secular rise in the formalization of credit and decline in informal sector lending.

14.5

INVESTIGATING ALTERNATIVES TO INFLATION TARGETING

14.5.1

The Real Exchange Rate is a Better Target

An alternative policy target should be consistent with and support wideranging development priorities. In other words, an important criteria for the monetary target is that it should play a positive development role and actively support the economy’s structural transformation. To this end, we need to specify the critical components of this economic transformation in order to sharpen monetary policy’s role, and to ensure that it becomes a coherent part of the nation’s development strategy. Taking into account these considerations, a stable and competitive real exchange rate (SCRER) is regarded to serve as a better intermediate target because it helps to advance Vietnam’s national priorities, which

Monetary policy in Vietnam: alternatives to inflation targeting

311

are rapid and sustainable growth, modernization, industrialization and poverty reduction. In contrast to inflation targeting regimes that have been found to push many economies into a lower employment growth trajectory, SCRER targeting would contribute to growth and employment creation through its impact on resource allocation, rewarding firms that adopt forward-looking production technologies and encouraging them to develop promising new businesses. To lay the foundations for sustainable growth, strategies that give priority to developing the medium to large enterprise (MLE) sector may be more effective than strategies that advocate concentrating resources on developing the SME sector. This is because mid-size enterprises have better capacity for learning and for technological innovation, and can create more jobs faster. Development of this sector will accelerate formalization of the economy (enabling policy makers to better monitor economic activity), promote human capital development, technological development and development of management skills, and strengthen the competitiveness of domestic firms. Given that the long-term survival and growth of enterprises depends on their being able to maintain a healthy profit rate, the stability of employment growth in the formal sector is closely linked to an environment that is conducive to MLE growth. An important aspect of this environment is that monetary policies send consistent signals to affirm the basic stability of key macroeconomic relative prices including the RER. This is needed so that enterprises will be confident enough to proceed with their investment plans in order to develop in areas that are most likely to be profitable. A SCRER targeting framework for monetary policy is key to promoting rapid expansion of the MLE sector. This is necessary to ensure that employment in the formal sector (which is dependent on MLE growth) will increase at a rate that can absorb Vietnam’s rapidly growing labor force (about 1.2 million new entrants to the workforce every year). The economic well-being of Vietnamese workers depends on this, because wage rates in the formal sector are significantly higher than informal sector wage rates. A SCRER target also helps the government to reduce its reliance on administrative levers to bring about desired changes in the economy. Officials have less justification to yield to pressure from firms in import-substitution sectors for special protections. If firms in sectors such as paper, steel and cement are unable to survive and prosper in a favorable price environment created by a SCRER, the government should conclude that they are unlikely to achieve long-term commercial viability. Consequently, the economy would achieve better resource allocation if these firms were to close down their operations.

312

Beyond inflation targeting

NOTES 1.

2. 3. 4. 5.

6.

7. 8. 9.

10. 11.

12.

13.

Earlier versions of this chapter were presented to the May 2005 CEDES/Amherst Research Conference in Buenos Aires and the July 2005 Da Nang Symposium on Continuing Renovation of the Economy and Society. Financial support for this project has been provided by the Ford Foundation, UNDESA and the Rockefeller Brothers Foundation. My gratitude and thanks go to two anonymous referees, Gerald Epstein, Erinç Yeldan, Jaime Ros, Lance Taylor, Per Berglund, Dang Nhu Van and Phillipe Scholtes for their insightful comments and valuable ideas. I am responsible for all remaining errors and omissions. Bernanke and Mishkin (1997). However, Mishkin (2000), an advocate of inflation targeting, acknowledges that price stability is ‘a means to an end, a healthy economy, and should not be treated as an end in itself’ and that ‘central bankers should not be obsessed with inflation control’. Law on the State Bank of Vietnam is dated 12 December 1997. Frenkel and Taylor (2005) emphasize that the RER must be kept at a stable and competitive level for a relatively long period if the positive effects are to take place. The reason is that responses to the new (competitive) set of relative prices take time because they involve restructuring firms and sectoral labor market behavior. This takes place over time via changes in the pattern of output among firms and sectors, and adjustments in technology and organization of production. In 1989 the SOE sector was made up of about 12 000 enterprises, of which 3100 were in industry; while the remaining were in trade, construction, agriculture and services. Most SOEs were provincial or district enterprises that were managed by the Industrial Bureaus of the provincial or district People’s Committees (World Bank, 1991). The reform of state enterprises, a key component of the Doi Moi reforms, subjected the SOEs to a hard budget constraint. By 1992 the number of SOEs fell by nearly half to 6545 enterprises, and their labor force was cut from 2.7 million to 1.7 million (IMF, 1998). The expenditures included the costs of maintaining a large military force, direct subsidies to SOEs and indirect subsidies associated with price controls. This is done by delaying or failing to repay credit extended by their suppliers which generally were other SOEs. The Vietnamese term employed by SOE managers to describe this practice is chiem von nhau (conquering each other’s working capital). This usually took place through informal channels due to unfavorable regulations governing formal money transfers. Recipients were forced to take the money in Vietnamese currency at an exchange rate which effectively gave them half or sometimes only a third of the amount they could get in the open market (Beresford and Dang Phong, 2000). SBV intervention in the foreign exchange market to support the targeted rate of exchange takes the form of buying and selling foreign currency or engaging in foreign exchange swaps. As in other countries where credit market segmentation play an important role, firms in Vietnam that have access to the formal banking system become key actors in the process of credit creation. They act as financial intermediaries to credit-constrained firms by providing the latter with trade credit. In other words, institutional factors help to turn inter-firm credit into an imperfect substitute for bank credit. Both function as ‘a quasi second legal tender’ or ‘parallel currency’ in the economy (Hauskrecht and Nguyen, 2004). The government can track the quantity of currency outside banks and the quantity of dong and dollar deposits. However, the quantity of gold and hard currency held by households and other economic agents that are used as a medium of exchange and store of wealth is not known. As regards indirect monetary policy instruments, during the mid 1990s the SBV introduced and has been using required reserves, refinancing and discount lending facilities, open market operations and foreign exchange interventions. The refinancing rate and the discount rate together define the upper and lower band for the open market operations rate although on occasion this rate does move outside the band (Camen, 2006).

Monetary policy in Vietnam: alternatives to inflation targeting

313

REFERENCES Beresford, Melanie and Dang Phong (2000), Economic Transition in Vietnam: Trade and Aid in the Demise of a Centrally Planned Economy, Cheltenham, UK and Northampton, MA, USA: Edward Elgar. Bernanke, B. and F. Mishkin (1997), ‘Inflation targeting: a new framework for monetary policy?’, Journal of Economic Perspectives, 11 (2). Carare, A., A. Schaechter, M. Stone and M. Zelmer (2002), ‘Establishing initial conditions in support of inflation targeting’, International Monetary Fund working paper WP/02/102. Camen, U. (2006), ‘Monetary policy in Vietnam: the case of a transition country’, Bank for International Settlement papers no. 31, accessed at www.bis.org/publ/ bppf/bispap31t.pdf. Epstein, Gerald (2003), ‘Alternatives to inflation targeting monetary policy for stable and egalitarian growth: a brief research summary’, working paper no. 62. Frenkel, R. and L. Taylor (2006), ‘Real exchange rate, monetary policy, and employment’, DESA working paper no. 19, United Nations, February; extended version accessed in Alternatives to Inflation Targeting, 2, from the Political Economy Research Institute, accessed at www.un.org/esa/desa/papers/2006/ wp/9_2006.pdf. Hauskrecht, A. and T.H. Nguyen (2004), ‘Dollarization in Viet Nam’, Indiana University, Kelley School of Business, Department of Business Economics and Public Policy working paper, accessed at http://ideas.repec.org/p/iuk/ wpaper/2004-25.html. IMF (International Monetary Fund) (1998), ‘Vietnam: selected issues and statistical annex’, IMF staff country report no. 98/30. IMF (1999), ‘Vietnam: selected issues’, IMF staff country report no. 99/55. IMF (2003), statement by Sean Nolan, IMF Division Chief, Asia and Pacific Department, at the Consultative Group Meeting for Vietnam, accessed at www. imf.org/external/np/dm/2003/120203.htm. IMF (2004), staff report for the 2004 Article IV consultation. Kovsted, J., J. Rand, F. Tarp, N.D. Tai, N.V. Huong and T.M. Thao (2002), ‘Financial sector reforms in Vietnam: selected issues and problems’, CIEM/ NIAS discussion paper no. 0301, Hanoi. Mishkin, Frederic S. (2000), ‘Inflation targeting in emerging market countries’, American Economic Review, 90 (2) (May), 105–9. Montiel, Peter (2003), Macroeconomics in Emerging Markets, Cambridge: Cambridge University Press. Monetary Authority of Singapore (2001), ‘Singapore’s exchange rate policy’, accessed 8 August 2006 at www.mas.gov.sg. Nghia, L.X. (2005), ‘Orientations of the banking sector development strategy towards international integration’, paper presented at the 18 January seminar on Preparation for the 2006–2010 Five Year Social Economic Development Plan in the Banking Sector, Hanoi. Razin, Ofair and Susan M. Collins (1997), ‘Real exchange rate misalignments and growth’, National Bureau for Economic Research working paper no. w6174. Thao, N.D. (2004), ‘Strategy for the development of banking services in Vietnam in the context of international economic integration’, Project VIE/02/009 Banking Service Sector, Hanoi, August.

314

Beyond inflation targeting

Thanh, V.T., D.H. Minh, D.X. Trong, H.V. Thanh and P.C. Quang (2001), Exchange Rate in Vietnam: Arrangement, Information Content and Policy Options, Hanoi: Central Institute for Economic Management, February. Tho, T.V., N.N. Duc, N.V. Chinh and N. Quan (2000), Kinh Te Vietnam. 1955– 2000. Tinh Toan Moi, Phan Tich Moi [Vietnam Economy. 1955–2000. New Estimates, New Analysis], Hanoi: Statistical Publishing House, December. UNIDO (2004), ‘Vietnam Agenda 21. National Conference and Forum on Sustainable Development in Vietnam’, Document prepared by UNIDO Country Office in Vietnam for the Forum on Sustainable Industry and Business, Hanoi, December. World Bank (1991), ‘Transforming a state owned financial system: a financial sector study of Vietnam’, report no. 9223-VN, Washington, DC, April. World Bank (1992), ‘Vietnam: restructuring public finance and public enterprises’, report no. 10134-VN, Washington, DC, April.

Index conditional least squares 119 Conselho Monetário Nacional 140 contagion effects 139, 151 convertibility 181–3, 185, 187, 190 corto 160, 167–9, 172 crawling band 159 crawling peg 45 credit allocation techniques 20–21 credit targeting 24 currency crisis 139, 142, 147–8 current account 139, 145, 152, 180, 271, 274–9, 288–9 Czech Republic 77, 78, 80

accountability 3, 4 aggregate demand 16, 28, 42, 118, 124, 187, 190–91, 199 see also effective demand anti-inflationary policy 6, 75, 87 anti-unemployment policy 73–5 Argentina 23, 131, 179–202 Augmented Dickey–Fuller Test 260 balance of payments 58, 272, 276, 279 ‘black’ market see parallel markets Brazil 22, 131, 139–55, 158 Brazilian Central Bank 139–43, 145, 147–8, 151–2 call money rate 260–61 capital adequacy ratio 289 capital controls 22–4, 41, 44, 55, 190, 193, 197–8, 233, 243–4, 248, 264, 269–70, 291–2 capital flight 139, 142, 243 capital flows, international 40, 204, 215–18, 220, 223 capital management techniques 18–24, 225, 233–5, 243–4 central banks 3–24 credibility of 159–60, 166, 172, 173 and development 5 net worth of 194–5 central bank independence 4, 8, 87, 187 Chakravarty Committee 257 Chile 24, 158, 190, 244 China 24, 112, 249 class 71–87 class conflict 71, 75, 93 cointegration 163, 166 Colombia 158 Comitê de Política Monetária 141 computable general equilibrium model see general equilibrium model

deflation 93, 97–101 demand management 72, 86, 129–30 devaluation 179–82, 187, 272–3, 275–7 see also exchange rate, real exchange rate development channel 16 dirty float 140, 153, 155 disinflation 142, 145, 154, 171 Doi Moi reforms 300, 303 East Asia 248, 269–70 economic growth 132, 179–81, 191, 249, 257, 263 see also inflation and economic growth, SCRER effective demand 29, 32–3, 36 see also aggregate demand 28 emerging market economies 158 employment 24–5 targeting 18, 24, 227–47 equilibrium, macroeconomic 34–7, 47 expectations 181–2, 184–5, 192, 196–7 exchange rate 28 appreciation 139, 142, 145, 148, 152, 153, 155 depreciation 139, 142, 144, 145, 148, 151, 153, 155

315

316

Beyond inflation targeting

devaluation 179–82, 187 and economic development 28, 30–31 and economic policy 28, 37 and external balance 28 and finance 29 and inflation 29 and monetary policy 159, 167–71 over-valuation 301 pass-through 8, 139, 171–4 pegged 291 and resource allocation 28, 29 targeting 139–42, 146–7, 150, 155, 179–99, 252, 310–11 volatility 180, 187, 191–2, 197 see also real exchange rate, nominal exchange rate expansionary policy 72–3, 75 expansionary fiscal contraction 203 export sector 108 see also tradable goods external balance 29 external fragility 220, 223 Federal Reserve 41 financial crisis 146 financial instability 5 fiscal balance 23, 180–81, 187–9, 190, 284–7, 292–3 fiscal dominance 3, 205, 214 fiscal overhang 256 fiscal policy 160, 190–91, 196–9 flexible exchange rate 160–61 see also exchange rate France 77–8, 80 GEAR (Growth, Employment and Redistribution) 228–9 gender 93–112 gender bias 108–9 and central bank policy 111–12 see also inflation targeting gender discrimination 95 gender equality 110 gender equity 110–12 general equilibrium model 24, 211–23 Granger Causality 260–62 Greenspan, Alan 3 Haiti 121–2 heterodox economics 72

Hodrick-Prescott filter 97, 99, 102 hyperinflation 117 IMF (International Monetary Fund) 5, 8, 9, 119 and conditionality 5 see also inflation targeting impulse response function 164, 238, 262–3 income distribution 4, 8 incomes policies 21, 131–2, 160 India 248–70 inflation aversion 78–87 conflict theories of 75 core 254, 281–3 and economic growth 9, 56, 116–36 expected 48 and inequality 132 and poverty 73–4, 76 and unemployment 93–112 inflation preferences 71–87 inflation rate targets 9, 18 inflation targeting 4, 41–2, 45, 55 alternatives to 172–4, 291–7 and asset price stability 9 and budget deficit 149–50 and central bank accountability 8 and economic growth 10–11, 56, 116–36 and escape clauses 281–2 and exchange rate stability 16 and foreign exchange reserves 15 and gender bias 108–9, 111 and IMF 5, 44 and industrial policy 30–31 and inflation rates 8 in open economies 44 and price stability 3 and primary fiscal surplus 150 and real exchange rates 44–67 and sacrifice ratios 9 in small economies 55 and trade balance 11–17 and unemployment 11–12, 15, 24, 56, 71–87, 93–112 inflation targeting countries, list 6 inflation targeting, socially responsible alternatives 17–24

Index inflation variability 158 ILO (International Labour Organization) 4 India 23, 95–107 interest rate 38, 45, 57, 60–61, 116–30, 159–60, 167, 170, 172, 183–5, 187, 192–8 see also real interest rate interest rate parity 192–3 Ireland 95, 122, 135 Israel 95, 131, 135 Instituto Brasileiro de Geografia e Estatística 141 International Social Survey Program 77–92 J-curve 34 Jacobian matrix 54, 64 jobless growth 210–11 Kaldor 35 Keynesian models 49, 50, 60 labor intensity channel 16 labor market 29, 32–3 Lebac 182–90 Lerner Symmetry Theorem 30 liberalization 4 logistic regression 81, 83–6 Lula 144–5, 147 macroeconomic channel 16 macroeconomic instability 5 Malaysia 24, 244 Mexico 18, 55, 95, 131, 135, 158–78 Minsky, Hyman 39 monetarism 117 monetary base 179, 185–90, 192, 194, 195 monetary policy 41–2, 45, 93–4, 97, 102, 105, 110–11, 132, 182, 187, 190, 192–3, 197–9 pro-development 299 money supply growth 94, 97, 105–6 and women’s employment 108–10 monetary targeting 159, 190, 251–2, 256, 271–2, 275, 279–82 modified 280

317

macroeconomics new classical 71, 73 new consensus 49, 55 New Keynesian 72 see also post-Keynesian economics and structuralist macroeconomics NAIRU (non-accelerating inflation rate of unemployment) 72, 117 narrow money 252, 260–61 Nicaragua 122 nominal anchor 258 nominal exchange rates 37–39, 41, 42, 45, 46, 182, 192, 193, 195–7 and middle income countries 39 and targeting 159 see also exchange rate, real exchange rate nonlinearity 119 non-tradables see non-traded goods non-traded goods 28, 30–36 North American Free Trade Agreement (NAFTA) 165 open market operations 45, 57 outlier 119, 120–21 output gap 166–7 output-inflation trade-off 158 output variability 158 parallel market 300 pass-through 163, 165–6, 171–4 see also exchange rate Peru 131, 158 Philippines 23–4, 271–98 Phillips Curve 71, 117, 166, 278 political economy 71–2, 75–6 and gender 93–112 policy space 7 post-Keynesian economics 49, 72 potential output 152–5 poverty 73–4 alleviation of 249 price rigidities 58 price stability 9, 11, 14 primary surplus 141,150 primary surplus target 203, 210, 219, 223–4

318

Beyond inflation targeting

productivity 29, 31–3, 35–7 prudential regulations 198–9 rational expectations 71–2, 86 real exchange rate 28–42, 37–68, 94, 97, 105–9, 179–202, 263–4, 269–70, 300 appreciation of 32, 39, 55, 171 and capital flows 40–41 and central bank intervention 39, 41 depreciation of 31, 166, 168, 171, 180, 189, 198 and employment 36, 65 fixed 55 and output 163–5 overvaluation 301 and targeting 44, 45, 173, 251–2, 264 and wage share 53 see also exchange rate, nominal exchange rate real interest rate 94, 97, 102–6, 109, 141–3, 145–8, 150, 152, 155, 206, 208, 212, 220, 232 see also interest rate relative inflation aversion see inflation aversion remittances 283–4, 287, 289, 291, 302 Reserve Bank of India 248 reserve requirements 45, 52 Ricardo-Viner Model 30 root mean square error 261 Rwanda 122 sacrifice ratio 8, 97 SCRER (stable and competitive real exchange rate) 16, 17, 180–81, 187, 189–92, 196–9, 310–11 seigniorage 195

SELIC (Sistema Especial de Liquidação e Custodia) interest rate 141, 143, 147 Singapore 95, 100–108 stagflation 105 state-owned enterprises 302 supply shock 10, 146 South Africa 24, 55 South Korea 24, 122 State Bank of Vietnam 299–301 sterilization 188–90, 194, 196–8 structuralist macroeconomics 16–17, 44–5 surplus labor 250 Sweden 131 Taylor Rule 49, 57 Tinbergen Rule 194 tradables see traded goods traded goods 28, 30–36, 96, 108–9 transparency 3, 8, 18 trilemma 8, 17–18, 39, 41, 45, 47, 52, 140, 181, 192–4 Turkey 22–3, 203–26 twin targeting 203–26 uncovered interest parity condition 193 under-employment 232 unemployment, male 96 unemployment, female 96 unemployment preferences 71–87 VAR (vector auto-regression) models 236–38, 262 Vietnam 23, 299–314 volatility 8, 9, 15 Zimbabwe 122