r 8 ;41#Stit23196483., THE WORLD BANK - *,i C n n sDisi, atanon 8 deg agt b ility, nd sOCZ4i ----1E S sm ii ,sw M , ndoiaEu 241813c 24556 10 F ' ~June 2002 i UL B ECK E RMAN I9IRX 90 L I tOA 1,NO i I TORS 0 0 i2 de 1E --~-~ "i .- -g A ' - ___.4_ ? ,~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~L iSIIE , - ,/s . , u .-,t~~~~. - - DIRECTIONS IN DEVELOPMENT Crisis and Dollarization in Ecuador Stability, Growth, and Social Equity Paul Beckerman and Andres Solimano Editors THE WORLD BANK Washington, D.C. © 2002 The International Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, USA All rights reserved Manufactured in the United States of America First printing June 2002 123404030201 The findings, interpretations, and conclusions expressed here are those of the author(s) and do not necessarily reflect the views of the Board of Executive Directors of the World Bank or the governments they represent. The World Bank cannot guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply on the part of the World Bank any judgment of the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is copyrighted. No part of this work may be reproduced or transmitted in any formn or by any means, electronic or mechanical, including photo- copying, recording, or inclusion in any information storage and retrieval system, with- out the prior written permission of the World Bank. The World Bank encourages dis- semination of its work and will normally grant permission promptly. For permission to photocopy or reprint, please send a request with complete infor- mation to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750-4470, www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. Cover photograph by Ray Witlin, the World Bank Photo Library. ISBN 0-8213-4837-X Library of Congress Cataloging-in-Publication Data has been applied for. Contents Preface .................................................... ix Abbreviations ....................................................xi 1. Crisis and Dollarization: An Overview .................................................1 Andr6s Solimano Introduction ....................................................1 Historical and Structural Features of Ecuadoran Economy and Society ....................................................4 Dollarization: Lessons and Challenges Ahead ........................................6 Social Impact of Economic Crisis and Dollarization ............................ 12 Organization of This Book ................................................... 13 Notes ................................................... 13 References ................................................... 14 2. Longer-Term Origins of Ecuador's "Predollarization" Crisis ... 17 Paul Beckerman 1. Introduction ................................................... 17 2. Historical Background of Ecuador's Predollarization Crisis ......... 19 3. Ecuador's Economic Structure Going into the Predollarization Crisis ................................................... 34 4. Ecuador's Predollarization Crisis ................................................... 51 .5. Conclusion: Underlying Causes of Ecuador's Predollarization Crisis ................................................... 59 Notes .................................................... 76 Bibliography ................................................... 78 3. Ecuador under Dollarization: Opportunities and Risks ................... 81 Paul Beckerman and Herndn Cortes-Douglas 1. Introduction ................................................... 81 2. Evidence and Theory from the Literature on Dollarization ........... 83 3. Lessons from Panama's Dollarization ............................................... 86 iii iv CRISIS AND DOLLARIZATION IN ECUADOR 4. Lessons from Argentina's Currency-board Experience .................. 90 5. Ecuador's Dollarization System ................................................... 95 6. Transition Issues: Deposit Unfreezing and Price Adjustment ..... 100 7. Ecuador's Macroeconomic Performance under Dollarization in 2000 and 2001 ................................................... 108 8. The Longer Term under Dollarization ............................................. 117 9. Conclusions .................................................... 120 Notes ................................................... 121 Bibliography ................................................... 123 4. Ecuador: Crisis, Poverty, and Social Protection ................................ 127 Suhas Parandekar, Rob Vos, and Donald Winkler 1. Introduction ................................................... 127 2. Inequality and Poverty .................................................... 130 3. Vulnerable Groups ................................................... 134 4. Human Development ................................................... 138 5. Targeted Poverty Programs .................................................... 146 6. Future Prospects ................................................... 162 7. Strategic Options ................................................... 170 Notes ..................................................... 174 References ................................................... 175 5. Gender Dimensions of Vulnerability to Exogenous Shocks: The Case of Ecuador ................................................... 177 Maria Correia 1. Introduction .................................................... 177 2. Assessing Vulnerability to Exogenous Shocks: A Framework ..... 179 3. Gender and Household Vulnerability in Ecuador ......................... 181 4. Institutional Context .................................................... 203 5. Conclusions and Recommendations ................................................ 204 Notes ..................................................... 207 References ................................................... 211 Tables 2.1 Ecuador: Governments, 1979-2001 .................................................. 29 2.2 Ecuador: Dollarization Indicators ................................................... 47 2A.1 Ecuador: Selected Annual Macroeconomic Indicators, 1991-2000 ................................................... 62 2A.2a Ecuador: Selected Monthly Macroeconomic Indicators, 1998 ..... 64 2A.2b Ecuador: Selected Monthly Macroeconomic Indicators, 1999 ..... 66 2A.3 Ecuador: National-Income Accounts ............................................... 68 2A.4 Ecuador: Nonfinancial Public-Sector Accounts (Percentage of GDP) ................................................... 70 CONTENTS v 2A.5 Ecuador: Balance-of-Payments Accounts (US$ Million) ................. 72 2A.6 Ecuador: Year-End Extemal Debt Outstanding and Disbursed (US$ Million) .................................................. 73 2A.7 Ecuador: Year-End Monetary and Commercial-Bank Aggregates (Percentage of GDP), 1987-1999 ................................................... 74 3.1 Panama: Selected Macroeconomic Indicators, 1970-99 .................. 87 3.2 Argentina: Selected Macroeconomic Indicators, 1989-2000 ........... 92 3.3 Ecuador: The Central Bank's Four Balance Sheets, March 10, 2000-December 31, 2001 .................................................. 97 3.4 Ecuador: Exchange Rate (Sucres per U.S. Dollar), Consumer Prices, and Real-effective Exchange Rate, December 1998-December 2001 .................................................. 105 4.1 Extreme Poverty in Ecuador ........................... 132 4.2 Consumption-based Comparison of Poverty in Ecuador ............. 133 4.3 Ecuador: Household Characteristics by Consumption Quintile .......................... 135 4.4 Ecuador: Correlates of Poverty............: ............. 137 4.5 Ecuador: Composition of Social Spending, 1995-2001 ................. 149 4.6 Public Expenditure Incidence of Social Spending by Type of Program and Per-capita Expenditure Quintiles, 1999 .................. 150 4.7 Vulnerable Groups Among the Poor and Ecuador's Social- Protection Programs ................................................. 151 4.8 Ecuador: Targeting Errors of the Bono Solidario by Eligibility Criteria, 1999 ................................................. 153 4.9 Ecuador: Eligible Population and Actual Recipients of the Bono Solidario, 1999 ................................................. 153 4.10 Ecuador: Coverage and Targeting: Programs for Children under Six Years .............. . 157 4.11 Ecuador: Coverage and Targeting: Programs for School-Age Children ........................................ 159 4.12 Ecuador: The Annual Cost of Alternative Coverage, Level, and Targeting Changes in the Bono Solidario (in US$) ................... 165 4.13 Strategies for Social Protection ................................................ 171 5.1 Structure of Employment by Sex and Status (Quito, Guayaquil, and Cuenca Averages from March 1998 to March 2000) ................................................ 195 5.2 Average Level of Education of the Labor Force by Sex and Labor Market/Unemployment Segment (Quito, Guayaquil, and Cuenca Averages from March 1998 to March 2000) ..................................... 195 5.3 Unemployed, According to Education and Sex (Quito, Guayaquil, and Cuenca Averages from March 1998 to March 2000) .............. 196 5.4 Labor-market Segmentation by-Sex (Quito, Guayaquil, and Cuenca Averages from March 1998 to March 2000) ...................... 196 vi CRISIS AND DOLLARIZATION IN ECUADOR 5.5 Unemployed Population by Relationship with Household Head (Quito, Guayaquil, and Cuenca Averages from March 1998 to March 2000) .................................................... 198 5.6 Indicators of Land Ownership in Ecuador .................................... 202 Figures 2.1 Ecuador: Year-end Public and Publicly Guaranteed External Debt (US$ million), 1970-2000 .....................................................27 2.2 Ecuador: Per-capita Real GDP, Real Private Consumption, and Year-End Per-capita Public External Debt in- 1999 U.S. Dollars and Prices, 1972-2000 ................................................... 31 2.3 Ecuador: Gross Domestic Capital Formation (at 1975 Prices, 1975 GDP = 100), 1965-2000 ................................................... 32 2.4 Ecuador: Gross Fixed Capital Formation, Gross Domestic Saving, and Net Imports of Goods and Nonfactor Services (Percent of GDP), 1971-2000 ................................................ 33 2.5 Ecuador: Exchange Rate (Sucres per U.S. Dollar); Trade-Weighted. Real-Effective Exchange Rate (+ = Depreciation; 1990 = 100), June 1970-September 2001 .................................................... 33 2.6 Ecuador: Per-Capita Nonfinancial Public-Sector Revenue (in 1998 U.S. Dollars at 1998 Prices), 1990-2000 ............................... 37 2.7 Ecuador: Monthly Average Crude Oil-Export Price, June 1995-September 2001 .................................................... 38 2.8 Ecuador: Per-Capita Central Government Expenditure (in 1998 U.S. Dollars at 1998 Prices), 1990-2000 .............................. 39 2.9 Ecuador: Nonfinancial Public-Sector Overall and Primary Surplus (US$ Million at 1998 Prices and Exchange Rate), 1990-2000 ..................................................... 42 2.10 Ecuador: Onshore Commercial-Bank Deposits (US$ Million) ... 48 2.11 Ecuador: Onshore Commercial-Bank Loans Performing Normally and in Arrears .................................................... 49 2.12 Ecuador: Indicators of Macroeconomic Imbalance, 1988-2000 ... 52 2.13 Ecuador: Consumer Prices, 1995-2000 . ............................................. 55 3.1 Ecuador: Monthly Trade-weighted Exchange-rate Competitive- ness, December 1994-December 2001 ................................................ 103 3.2 Ecuador: Monthly Increases in Consumer Prices, January 1999-December 2001 .................................................... 103 3.3 Ecuador: Consumer Prices and Weighted Trading-partner Prices at the Current Exchange Rate, December 1997-December 2001 .................................................... 104 3.4 Ecuador: Quarterly Real GDP (1998 average =100), 1997.4-2001.4 .................................................... 109 CONTENTS vii 3.5 Ecuador: Quarterly Nonfinancial Public-sector Revenue (US$ million), 1998.2-2001.4 ................................................... 110 3.6 Ecuador: Quarterly Nonfinancial Public-sector Expenditure (US$ million), 1998.2-2001.4 ................................................... 111 3.7 Ecuador: Quarterly Performance of the Main Components of the Current Account of the Balance of Payments (US$ million), 1998.1-2001.4 ................................................... 113 3.8 Ecuador: Monthly Merchandise Trade and Real-effective Exchange Rate (December 1996-October 2001) ............................... 117 4.1 Ecuador: Real Wage and Urban Poverty Trends (index, 1990 = 100) ................................................... 134 4.2 Malnutrition Rates in Latin America ............................................... 139 4.3 Malnutrition in Ecuador: Stunting (%) by Consumption Quintiles, 1998 and 1999 ................................................... 139 4.4 Malnutrition in Ecuador: Stunting (%) by Area and Region, 1998-2000 ................................................... 140 4.5 Ecuador, Jamaica, Honduras: Years of Educational Attainment by Age Cohort ................................................... 141 4.6 Ecuador: Educational Attainment by Rural and Urban Areas (persons over 24 years old ) .................................................... 143 4.7 Ecuador: Gender Gap in Education ................................................. 144 4.8 Ecuador: Percentage of 18-Year-Olds Completing School ............ 145 4.9 Ecuador: Percent of Children Working, 1999 .................................. 146 4.10 Ecuador: Social Spending per Capita, 1995-2001 (in US$) ........... 148 5.1 Factors Affecting Vulnerability to Exogenous Shocks .181 Preface Understanding the nature of deep economic crises and social disarray and formulating adequate exchange rate and other policies for stabiliza- tion, growth, and social equity are topics of great importance in develop- ing countries and emerging economies in the turbulent world of the early 21st century. The experience of Ecuador in the late 1990s and the early 21st century showcases a country with structural problems of low growth, regional divides, and social and ethnic fragmentation made more acute by a severe currency and banking crisis in the late 1990s. Ecuador's response to the crisis centered on the adoption of foreign money-dollarization- as a last-resort measure to cope with total distrust in the national cur- rency and domestic institutions after repeated cycles of failed stabilization and crisis. This book assesses several aspects of the Ecuado- ran experience, including a historical analysis of the main features of the country's economic development and the main political economy fea- tures that set the background for the most recent cycle of crisis and stabi- lization. The book analyzes in detail the characteristics of the economic crisis of 1998-99 and the subsequent experiment with dollarization and its initial results. Then the book turn's to the impact of the crisis and sub- sequent stabilization through dollarizaton. on poverty, inequality, mar- ginalization, gender, and the Ecuadoran family. The book also assesses the ability of existing social-protection institutions to cope with a severe economic crisis and subsequent stabilization. Most of the material for this book was initiated when several of the authors belonged to what was then the World Bank's Country Depart- ment for Ecuador, Colombia, and Venezuela. The work benefited from first-hand involvement-at times at the highest political level-in Ecuador until mid-2000. We want to acknowledge several people and former colleagues who made this book possible. David de Ferranti, World Bank Vice President for Latin America and the Caribbean, provided generous financial sup- port to fund this publication and encouraged a free analysis of events and ix x CRISIS AND DOLLARIZATION IN ECUADOR policies in Ecuador. David Yuravlivker, Vicente Fretes-Cibils, and Eduardo Wallentin also provided useful insights based on their knowl- edge of Ecuadoran economy and society. Luis Jacome, former governor of the Central Bank of Ecuador, provided helpful comments on chapters 2 and 3. Diana Cortijo, Mario Aventino, and Hazel Vargas gave important logistic support, and Paola Scalabrin was instrumental in publishing this project. We also appreciate efficient editorial work by Thea Clarke. Book design, editing, production, and dissemination were coordinated by the World Bank Publications team. As ever, the authors alone are responsible for any errors of fact or judgment this book may contain. Abbreviations AGD Agencia de Garantia de Dep6sitos (Deposit Insurance Agency, Ecuador) CAF Andean Development Corporation CAN Andean Community of Nations CDC Centers for Disease Control, U.S. CEPLAES Centro de Planificaci6n y Estudios Sociales (World Bank Poverty Group) CONAMU el Consejo de las Mujeres del Ecuador (National Council of Ecuadoran Women) DDSR debt-and-debt-service reduction DINAMU Direcci6n Nacional de la Mujer (National Directory for Women, Ecuador) ECD early childhood development ECV Encuesta de Condiciones de Vida (Ecuador LSMS survey) EAP economically active population EU European Union FISE Fondo de Inversi6n Social de Emergencia (Emergency Social Investment Fund, Ecuador) GDP gross domestic product IDB Inter-American Development Bank IESS Instituto Ecuatoriano de Seguro Social (Ecuadoran Social Security Institute) IMF International Monetary Fund NEC Instituto Nacional de Estadistica y Censo (National Institute of Statistics and Census, Ecuador) INNFA Instituto Nacional del Nifo y de la Familia (National Institute of the Child and the Family, Ecuador) LIBOR London interbank offered rate LSMS Living Standards Measurement Study MERCOSUR Mercado Comuin del Sur (Southern Common Market) NCHS National Center for Health Statistics, U.S. NGO nongovernmental organization OECD Organisation for Economic Co-operation and Development xi xii CRISIS AND DOLLARIZATION IN ECUADOR OPEC Organization of Petroleum Exporting Countries ORI Operaci6n Rescate Infantil (Child Rescue Program, Ecuador) PACMI Programa de Alimentaci6n Complementaria Materno-Infantil (Maternal-Infant Nutrition Program, Ecuador) PAHO Pan American Health Organization PANN Programa Nacional de Alimentaci6n y Nutrici6n (Food and Nutrition Program that replaces PACMI in Ecuador) PDI Programa de Desarrollo Infantil (Child Development Program, Ecuador) PRONEPE Alternativo Programa Nacional de la Educaci6n Prescolar (National Alternative Preschool Education Program, Ecuador) SIISE Sistema Integrado de Indicadores Sociales del Ecuador (Integrated System of Social Indicators of Ecuador) SIMUJER Situation of Women and Gender Inequality Indicators database, Ecuador STFS Secretaria Tecnica del Frente Social (Technical Secretariat of the Social Front, Ecuador) VAT value added tax WHO World Health Organization Crisis and Dollarization: An Overview Andres Solimano Introduction On January 9, 2000, Ecuador decided to adopt the U.S. dollar as its national currency, its domestic medium of exchange, and its unit of account,1 thus becoming the first country to officially dollarize its econ- omy in the 21st century. The purpose of this book is to analyze the con- text within which dollarization took place in Ecuador and some of its economic consequences. It describes the initial conditions, accompanying policies, and response of the economy to the official adoption of a foreign currency as the legal tender and the issues the still-new Ecuadoran expe- rience with dollarization suggests for other countries considering the adoption of a new monetary regime. Another important theme of the book is the social impact of the crisis of the late 1990s and of subsequent dollarization. The end of the 20th century caught Ecuador in one of the more serious economic crises-compounded by a governance crisis-in its Republican history. The country was on the verge of hyperinflation in late 1999 with the price level increasing at a rate of near 30 percent per month. The national currency, the sucre, was in free fall. The government had inter- vened in the banking system, and a large part of the deposits of the pub- lic was frozen. Internationally, in late 1999 the country was in partial arrears with private creditors and bondholders and, for various reasons, the International Monetary Fund (IMF) had withheld for nearly a year a crucial loan to support the balance of payments. This, in turn, forced the World Bank and the Inter-American Development Bank (IDB) to post- pone their own policy-based lending to Ecuador in 1999, attendant to the The author was Country Director of the World Bank for Ecuador, Colombia, and Republica Bolivariana de Venezuela between 1997 and 2000. He is currently Regional Advisor for the United Nations Economic Commission for Latin America and the Caribbean. Comments on this chapter by Paul Beckerman are appreciated. 2 CRISIS AND DOLLARIZATION IN ECUADOR stalemate with the loan by the IMF. At a time when hyperinflation had abated in Latin America, the Ecuadoran case of extreme monetary insta- bility was clearly a regional anomaly for the late 1990s. Most of the ingredients of high inflation and acute monetary instabil- ity were present: (a) a flight from national money and de-facto dollariza- tion2 as nationals and foreigners in Ecuador lost all confidence in the capacity of the sucre to serve its store-of-value function, (b) large fiscal deficits, (c) a sharp contraction in real economic activity, and (d) a severe banking crisis.3 The increasingly cornered government, led by President Jamil Mahuad, a highly educated and intellectually sophisticated social- democrat, could not gather congressional support for passing crucial tax legislation and other measures to stabilize the economy. This situation, combined with the near paralysis of the international financial institu- tions based in Washington, helped bring about an economic meltdown manifested in very high inflation, a banking-crisis, economic depression, and social disarray during most of 1999. It is important to recognize that the Ecuadoran crisis took place in a delicate situation of security within the Andean region. On the one hand, Ecuador and Peru were trying to consolidate an historic peace agreement signed by Ecuadoran President Jamil Mahuad and Peruvian President Alberto Fujimori in October of 1998. On the other hand, Ecuador was exposed to the potentially desta- bilizing effects of acute mtensification of the armed conflict in Colombia, a country that shares a long border with Ecuador. In this setting, and in one of the more dramatic experiments in recent monetary history, the Ecuadoran government decided, in January of 2000, to adopt, de facto, unilaterally, and apparently without much exter- nal consultation, the U.S. dollar as its national currency. This was a "pol- icy of last resort," an almost desperate move to restore some degree of monetary and price stability in a country that needed an urgent monetary anchor to stabilize expectations, avoid hyperinflation, stop uncontrolled currency depreciation, and enable resumption of normal economic and financial activity. Official dollarization had a political motivation as well. In late 1999, constitutionally elected President Mahuad was facing a sharp plunge in his popularity. His presidency was being challenged by a particularly adverse set of events: a severe economic crisis, an active and militant indigenous movement with radical political and economic demands, a badly divided and fragmented parliament, and a restive army. In these circumstances, a radical change in the monetary regime toward dollar- ization was seen by President Mahuad as a way to regain the initiative for his government, by changing the focus of the national debate away from purely political issues toward much needed economic stabiliza- tion. In spite of the announcement of official dollarization, President CRISIS AND DOLLARIZATION: AN OVERVIEW 3 Mahuad was deposed on January 21, 2000, following an indigenous uprising that seized the parliament building with support from units of the army. After late-night negotiations involving rebellious colonels, members of the political class, the U.S. Embassy in Quito, and the Orga- nization of American States, the rebels stood down and Vice President Gustavo Noboa was sworn in as the new President of Ecuador in the Ministry of Defense with the support of the army. The "constitutional order" was restored. The new government of Gustavo Noboa ratified the change of monetary regime initiated by President Mahuad, and official dollarization was adopted, in haste, and under very fragile conditions. At this stage, consul- tations were initiated with the U.S. government, whose currency was to be adopted. The reluctant IMF, which had distanced itself from the Mahuad administration, resumed lending in April 2000, and entered into full col- laboration to ensure the success of the change in the monetary regime.4 The mechanics and economic effects of dollarization are important subjects of this book. Supportive economic and financial legislation-the Law of Economic Transformation-was approved in March 2000. This legislation included a number of structural changes in several areas. In August 2000 Ecuador successfully carried out a bond exchange, which reduced its massive Brady debt by roughly a third, and its bilateral exter- nal debt was rescheduled in September 2000 by the Paris Club. The econ- omy benefited from an increase in the international price of oil, which helped to improve the fiscal accounts and the balance of payments. At the same time, important efforts to improve tax collection were undertaken. The income tax, which had been suspended in January 1999, was rein- stated. The fiscal accounts improved sharply, passing from a fiscal deficit of near 5 percent of gross domestic product (GDP) in 1999 to a small sur- plus in 2000. The balance of payments also improved, as a result of a combination of favorable oil prices, the repatriation of flight capital helped by the lib- eralization of dollar deposits in the banking system associated with offi- cial dollarization and, very importantly, by a surge in foreign remittances of Ecuadorans following massive emigration after the crisis that began in 1998-99.5 As a consequence of all these factors, the current account of the balance of payments registered a surplus of nearly 10 percent of GDP in 2000 compared with a deficit of roughly the same magnitude in 1999. The progress in solving the banking system crisis was slower than in other areas. In spite of intensive work to rationalize, dispose the assets of nonviable banks, privatize intervened banks, and other measures, as of 2001 a considerable segment of the Ecuadoran banking system still remained in the hands of the Deposit Insurance Agency (Agencia de 4 CRISIS AND DOLLARIZATION IN ECUADOR Garantia de Dep6sitos, AGD), which underwent several changes in its management structure in 2000 and 2001. Dollarization succeeded in stabilizing expectations, as reflected in declining interest rates and induced capital repatriation. Banks registered an increase in their deposits from the public. Dollarization did not stop inflation immediately, because adjustment to a new equilibrium level for the real exchange rate, undervalued when dollarization was launched, was reached through inflation. In addition, GDP started to recover fol- lowing official dollarization, helped by a gradual recovery of confidence and favorable external shocks. In turn, unemployment has slowly declined and real wages have become more stable, although real wage levels are rather depressed in dollar terms. Historical and Structural Features of Ecuadoran Economy and Society The deep economic crisis of the late 1990s that preceded dollarization in Ecuador was, as argued in this book (see chapters 2 and 3), the culmina- tion, in dramatic overtones, of an economic and governance crisis associ- ated with several structural characteristics of Ecuadoran economy and society. Historically, the emergence of Ecuador as an independent state from the the Confederation of Gran Colombia in 1830 created a country with two main competing regions: a coastal area (Costa) centered around the city of Guayaquil and the Sierra or highlands around the capital city of Quito. The two regions have different social, economic, cultural, and ethnic characteristics. Regional disputes have been an important source of social and political instability in Ecuador throughout the 19th and 20th centuries. Ecuador's main political parties are formed along regional lines, weakening central authority and forcing a style of policymaking that allocates resources, taxes, and quotas of political power in an effort to maintain regional balance.6 Economy-wide objectives such as eco- nomic growth and monetary stability are often displaced by the needs of regional balancing, redistribution, and rent-seeking. In turn, Ecuador, like most Latin American countries, is a highly socially stratified country. Wealthy people, elite landowners, and financial and industrial entrepre- neurs coexist with a population that is mostly poor (see chapter 4) and with a large (at times politically active) indigenous population. This social structure superimposed on the regional divide often hampers the capacity of governments to undertake national policies that gamer wide social consensus. During the 20th century the country endured repeated constitutional reforms, presidential crises, and cycles of military govern- ments followed by civilian rule (see Solimano 2002), both trying to ensure stable governance and economic development but with often disappoint- CRISIS AND DOLLARIZATION: AN OVERVIEW 5 ing results. The difficulties of building stable governing coalitions were exacer- bated in the late 1990s. In fact, since 1996 Ecuador has had four different Presidents: Abdala Bucaram, Fabian Alarc6n, Jamil Mahuad, and Gus- tavo Noboa.7 Over the same period there were about 10 finance ministers plus a frequent rotation of the technocracy working in government. Many of the most able and qualified people left the country. A common feature of the Ecuadoran economy in the 20th century has been the dependence of real economic activity, the fiscal accounts, and the balance of payments on exports of a few commodities, such as cacao, bananas, shrimp, and oil. This dependence has made the econ- omy prone to volatility associated with cycles in the international prices of commodities and climatic changes. This dependence on com- modity prices was accentuated in the 1970s with the oil price boom. Although the oil boom allowed a doubling of the yearly real growth rate of GDP of previous decades, from an annual rate of growth of 4.7 percent per year in 1950-60 to 9.4 percent in the 1970s, this dynamism was ultimately short-lived. In the 1980s and 1990s the economy reverted to average GDP growth rates on the order of 2 percent, lower than the historic average of the past 50 years in Ecuador and in Latin America. In the 1980s Ecuador, like other Latin American economies, suffered a foreign debt crisis after the windfall of oil revenues of the 1970s, and the cycle of foreign overborrowing of that decade. As a consequence, GDP growth declined to around 2 percent in the 1980s, down from more than 9 percent in the previous decade. In the 1990s, Ecuador started reforms that were never completed, suffered several large external shocks and natural disasters, and then culminated the decade with the disruptive economic and financial crisis we have already dis- cussed and which is analyzed in further detail in the next chapter of this book. An important cause of Ecuador's unsatisfactory economic perfor- mance is weak institutions. The fiscal structure has traditionally been very dependent on the revenues of oil and taxes on other commodities and, until recently, has suffered from the widespread practice of tax eva- sion. Public expenditure is far from efficient and well directed. In turn, the crisis of the banking system that started in 1998 also revealed serious shortcomings in the regulatory structure of the system, a pattern of loan concentration, and the vulnerability of the bank's portfolios to high real interest rates and overall economic decline. Still, there is room for (cautious) optimism. Ecuador is a country with significant economic potential. It has a strong natural resource base and talented people, its geographical proximity to major international mar- 6 CRISIS AND DOLLARIZATION IN ECUADOR kets in the "center" makes it a favorable location for international trade and foreign investment, and, in spite of its complex social and regional structure, it is generally a country of social peace. Dollarization: Lessons and Challenges Ahead The experience of Ecuador with dollarization is of interest for the rest of Latin America and other emerging economies wrestling with the adop- tion of the adequate exchange-rate regime in a world of increased finan- cial integration but also of volatility and instability. We can highlight six important areas in which the Ecuadoran experience is relevant for other nations. Dollarization under Fragile Initial Conditions The choice of a monetary regime by a country is a far-reaching decision that, under normal circumstances, must be preceded by a period of inter- nal discussion of the merits and possible disadvantages of possible alter- natives. Moreover, the introduction of a foreign currency to replace the national currency needs to be accompanied by adequate preparation and by legal reforms in several areas of the economy. A solid banking system, a sustainable fiscal position, and wage and price flexibility are all eco- nomic preconditions for successful dollarization. On the legal side, basic legislation must be introduced to legally sanction the new currency and allow contracts (wages, rents, and so on) to be made in foreign currency (now also the national currency). Also, the accounting systems of banks and corporations have to adopt new practices and conventions in line with the fact that a foreign currency is the legal tender after official dol- larization is adopted. The decision of when to dollarize (for example, its timing and sequenc- ing) is, however, a matter of debate. Some people adopt the position that dollarization need not wait on these other reforms to be in place and believe, on the contrary, that dollarization can accelerate the overall process of reform.8 As shown earlier, adequate fiscal, financial conditions, and accounting practices were not present when Ecuador announced dollarization in Jan- uary 2000. It is apparent that dollarization was not a decision made under controlled conditions to ensure its success. Rather, it was a bold move to reverse a situation of near hyperinflation and massive flight away from domestic currency, debased after a long period of monetary instability. Also, as already mentioned, the fiscal budget was in a sizable deficit dur- ing the year preceding dollarization and the state intervened in a large CRISIS AND DOLLARIZATION: AN OVERVIEW 7 part of the banking system, with several important banks having nega- tive net worth. Important pieces of legislation regarding the banking system, the new accounting systems, the conversion of contracts from sucres to dollars, labor laws, and other laws were passed after dollarization was launched. In fact, the legal approval of dollarization came in March 2000 and it was fully implemented in September 2000. The degree of public support for dollarization was mixed. Various groups, such as the indigenous people's movement and left-wing political parties, opposed dollarization, in part on nationalistic grounds. The middle class, industrialists, and bankers, however, supported dollarization both in Guayaquil and Quito. Very importantly, Congress ultimately supported dollarization. The United States was initially very cautious in supporting the measure taken by Ecuador. In the end dollarization was launched, implemented, and as of early 2002 consolidated. In a metaphoric sense, dollarization was a revo- lutionary regime change in the monetary system of the country and, like many revolutions, starting from dramatically deteriorated conditions, it still succeeded in holding. Of course, other countries considering dollar- izing would certainly benefit from more stable and balanced initial con- ditions. This was, indeed, the case of El Salvador, which decided to dollarize in January 2001 in far more comfortable fiscal and financial con- ditions than those of Ecuador just a year before. Indeed, El Salvador had maintained a fixed exchange-rate regime for almost a decade, and dollar- ization was seen as a "natural" consequence of a long period of a fixed exchange rate, low domestic inflation, and a largely dollarized banking system. A more distant case of dollarization is, Panama, which adopted the system in 1903 and has nearly a century of economic history with a foreign currency as the national currency.9 The Dynamics of Inflation, the Real Exchange Rate, and Output Dollarization was adopted in Ecuador mainly to stop very high infla- tion.10 In the last quarter of 1999 the consumer price index rose by 60 per- cent; the wholesale price index rose by 187 percent. However, the domestic price level continued to rise rapidly after dollarization was adopted, following a sharp depreciation of the currency from 18,000 to 25,000 sucres per dollar." There were two main reasons for the large depreciation of the currency preceding dollarization that was fueled by wild expectations of Ecuador's financial markets: (1) the intent to avoid a real appreciation after dollarization on account of "residual inflation" and (2) the need to increase the purchasing power of a limited level of international reserves (dollars) to buy (cheaply) the monetary base in 8 CRISIS AND DOLLARIZATION IN ECUADOR sucres at a more depreciated exchange rate. This latter factor was impor- tant since Ecuador had a very low level of international reserves at the time dollarization was implemented. Domestic prices nearly doubled over 2000. For 2001, however, inflation was only about 25 percent. Clearly, after official dollarization the speed of convergence of the domes- tic price level to a new international parity was gradual and spread over at least two years after the new currency was introduced. A similar speed of inflation convergence was observed in Estonia, a country that introduced a currency board in 1992. In Estonia inflation converged to moderately low levels, only two years later, in 1994.12 The real exchange rate in Ecuador depreciated mildly in 1998 (about 3.5 percent), but depreciated about 40 percent in 1999. After an additional real depreciation in January of 2000 following the "last" maxi-deprecia7 tion of the sucre, the real exchange rate (a somewhat peculiar concept in a dollarized economy) began steadily appreciating in February 2000 and afterwards as a consequence of the slow process of convergence of the domestic price level already noted above (see chapter 3 for a more detailed analysis of these trends). This pattern of rapid real depreciation of the national currency before the change in the monetary regime followed by a real appreciation of the currency was observed in three countries that adopted currency boards in the early 1990s: Argentina in 1991, Estonia in 1992, and Lithuania in 1994. As the acute crisis of Argentina in late 2001 and early 2002 is show- ing rather dramatically, the failure to correct the real appreciation of the currency through domestic deflation, cuts in nominal wages, and unem- ployment can be so costly as to generate an economic and political crisis of large proportions leading, among other things, to abandonment of the seemingly irreversible currency board regime. The growth cycle before and after official dollarization in Ecuador was the following: real GDP contracted sharply in 1999, falling by 7.3 percent that year, with unemployment rising from 11 to 15 percent. As chapters 2 and 3 of this book document, this situation was the combined effect of several fac- tors: external shocks (a decline in oil prices in 1998/99), natural disasters (the El Nifio phenomena in 1997/98), domestic instability, and a severe banking crisis. The latter clearly amplified the contractionary effects of the other shocks. GDP grew by 2.3 percent in 2000, as a consequence of an improve- ment in domestic confidence following dollarization (domestic interest rates fell) and by a recovery in international oil prices. Real growth reached 5.4 percent in 2001 as the gradual stabilization in inflation consolidated, confi- dence recovered, and construction of the second Transandean oil-pipeline generated employment and income. However, social conditions in Ecuador postdollarization still remain precarious (see chapters 4 and 5). CRISIS AND DOLLARIZATION: AN OVERVIEW 9 Dollarization in Ecuador and Exchange-Rate Regimes in the Andean Area Ecuador is a member of the Andean Community of Nations (CAN), a free-trade area. Although exchange-rate regime harmonization among its member countries is not practiced in the CAN, and monetary integration is still not on their agenda, the fact is that Ecuador's new monetary sys- tem adds to the already large variety of exchange-regimes in the Andean region. At present (mid-2002) we have floating exchange-rate regimes in Peru, Colombia, and Republica Bolivariana de Venezuela, a crawling peg system in Bolivia, and a foreign-currency regime in Ecuador. The fact that two trade partners (and neighboring countries) of Ecuador-Peru and Colombia-are in a floating exchange-rate regime while Ecuador is dol- larized creates the potential for Ecuador to lose regional competitiveness, should these countries depreciate their currencies, an option unavailable to Ecuador. In the context of MERCOSUR (Mercado Comi:n del Sur) countries, this is what exactly happened to Argentina when Brazil sharply devalued its national currency, the real, in early 1999, causing Argentina to suffer an important loss of competitiveness. Argentina, with its currency board system, could not adjust its exchange-rate parity to maintain competitiveness. A similar situation is starting to face Ecuador, so this can be considered a vulnerability of the new system. A more gen- eral lesson here is that decisions made by one country on its exchange- rate regime should consider the interdependences with the exchange-rate regimes of other member countries of the same integration bloc. The CAN and MERCOSUR are just starting to put in place mechanisms of consultation on monetary and fiscal policy among their member cowu- tries. Such consultations are still far behind experiences of macroeco- nomic coordination and harmonization such as that of the European Union (EU), in which the exchange-rate regimes were defined in a collec- tive way. Of course, the degree of integration in goods, capital, and labor markets in the EU is far higher than in the CAN (or MERCOSUR).13 Still, the development of practices of mutual consultation in monetary and exchange-rate matters among member countries is worth pursuing. Seigniorage and Lender of Last Resort A classic argument in the case for national money'4 is that, by giving up the use of national money and adopting a foreign currency, a country loses a source of revenue, given by the difference between the real command of resources that the creation of money entails and the low cost of producing (paper) money. This difference is called seigniorage. For ranges of low to moderate inflation and with "normal" demand for money, seigniorage can 10 CRISIS AND DOLLARIZATION IN ECUADOR represent several points of GDP. By adopting the U.S. dollar as its national currency, Ecuador loses this source of revenue and transfers seigniorage to the Federal Reserve Bank of the United States: However, the quantitative importance of the loss of seigniorage in Ecuador is bound to be modest, as the economy was already highly demonetized and de-facto dollarized before the U.S. dollar was officially adopted. In any case, it should not be ruled out that in the future some arrangement could be made for the seigniorage to be shared with Ecuador. Another feature of a dollarized system is the apparent absence of a lender of last resort. Because the Central Bank, still in existence in Ecuador, cannot create money any more, banks will be unable, unlike in the past, to resort to bailouts and credits from the Central Bank. In the absence of national money, the Central Bank ceases to play the role of lender of last resort. However, as the commercial banking system was in such a fragile condition in Ecuador at the time of dollarization, a special contingency fund for banks in distress was created following officiai dol- larization. From this perspective, this fund can be viewed something like a lender of last resort in the event of a banking crisis. Moreover, as has been the case in history, for example in the United States during episodes of banking crisis before 1913, the year the Federal Reserve System was created, the resolution of banking crisis or liquidity shocks was arranged by private financiers such as J. P. Morgan. In other cases, the resources for performing the functions of lender of last resort can come from the fiscal budget or from foreign borrowing. The Adjustment Mechanism of the Dollarized Economy An economy operating with a foreign currency as the legal tender works in several respects like the economies under the gold standard of the pre- 1913 world. The so-called price-specie flow mechanism of David Hume described such a system as one in which balance of payments disequilib- ria had a domestic monetary counterpart (money expands when there is a balance of payments surplus and contracts when there is a deficit). These changes in the money supply affect domestic prices relative to world prices, thereby automatically correcting the balance of payments disequilibria and, in this way, restoring macroeconomic balance. This sys- tem rests on a combination of policy rules and price and wage flexibility. A critical point of the mechanism is that it requires both downward as well as upward wage and price flexibility. In particular, when there is a loss of external competitiveness a deflation of prices and salaries is needed to correct external and internal imbalances. By adopting official dollarizationi Ecuador entered into the world of tight rules in economic policy. As mentioned before, in the new system, CRISIS AND DOLLARIZATION: AN OVERVIEW 11 the Central Bank can neither print money nor adjust the exchange parity between a national and foreign currency since the national currency was abolished. Fiscal deficits cannot be monetized, and commercial banks cannot receive credit from the Central Bank in national currency to resolve financial difficulties. The new system also puts strong require- ments on fiscal solvency and domestic financial stability. This, needless to say, implies a strong departure from previous practices in the conduct of monetary, fiscal, and exchange-rate policies in Ecuador. The other component of David Hume's price-specie flow mechanism is wage and price flexibility. Certainly Ecuador has had a lot of upward price flexibility in the recent past. The point, however, is to what extent there is also downward wage and price flexibility in Ecuador to correct relative prices in the wake of external shocks and natural disasters, to which the Ecuadoran economy has been quite prone in the recent past (see chapter 2). A subtle point is that although in Argentina there was some downward wage and price flexibility, but this was still not enough to reverse a real appreciation of the currency. In addition, cutting nominal wages, as antic- ipated by Keynes long ago, can be very unpopular and practically costly in a modem contractual economy. Thus it is not a recommendable course of action on which to rely to correct currency misalignments. Dollarization and Hard Pegs In the recent discussion of exchange-rate regimes a "bi-polar" view emerged. According to this view, for a financially integrated economy, two regimes are bound to be viable: "hard pegs" (currency boards, dol- larization, or currency union) or exchange-rate flexibility.15 "Intermedi- ate" exchange-rate regimes such as (soft) fixed-exchange rates, crawling pegs, and others are bound to be susceptible to crisis and failure in a con- text of high capital mobility. Only hard pegs and flexible rates would endure according to the bi-polar view. After the current Argentine crisis, this view is severely challenged. Several emerging economies have been in the hard peg group: Argentina (until December 2001), Bulgaria, and Hong Kong, China, all have had currency boards; Panama, Ecuador, and El Salvador are coun- tries that use the U.S. dollar as their national currency. In turn, for devel- oped economies, the countries of the EU have decided to adopt a common currency, the euro, another form of hard peg from the perspective of each member country. Argentina and Bulgaria are cases of countries that adopted currency boards after experiencing periods of very high inflation or hyperinflation. The other countries entered into a hard-peg currency arrangement in more gradual fashion and after a preparation period. Ecuador shares with Argentina and Bulgaria the fact that it adopted a 12 CRISIS AND DOLLARIZATION IN ECUADOR hard-peg regime because of the urgent need to restore credibility after experiencing extreme monetary instability.16 Although the recent experience with hard pegs suggests that they were mostly successful in stopping high inflation, often in a gradual way, and helped to restore stability, an important issue is the capacity of the system to last over long periods of time. This leads us to the com- plex problem of the "exiting option." As the events of early 2002 in Argentina show, the exiting from a currency board, if not well prepared and anticipated, can be extremely traumatic, possibly involving an implosion of the economy. In general, once a country has adopted a hard peg it is not expected to exit. The recent abandonment by Argentina of its currency board is starting to shatter this long-held view. The adoption of a hard peg is a kind of open-ended choice, almost irre- versible. In fact, hard pegs are conceived to side step the main weakness of "soft pegs" (fixed exchange rates, crawling pegs), namely, that fre- quent exits from the fixed system are often unanticipated and disrup- tive and often entail credibility loss for the monetary authorities. However, loss of the exit option should ultimately be considered a lim- itation of hard pegs if exiting is needed in extremis. Social Impact of Economic Crisis and Dollarization Economic crisis can have very adverse social consequences. In the late 1990s Ecuador suffered a sharp recession and a large increase in unem- ployment. Output contraction and job losses reduced economic welfare of the citizens, particularly that of the unemployed. In addition, as the eco- nomic crisis came with instability, continuous currency depreciation, and high and volatile inflation, there was a reduction in real wages, affecting workers and their families as well as other low-income groups and classes whose incomes grow (if at all) at a slower pace than the exchange rate and average prices. In the case of Ecuador, as documented in chapter 4, unem- ployment, poverty, and inequality all worsened in this period. From a longer-term perspective, the low (and volatile) rate of GDP growth of the 1980s and 1990s implied almost stagnant income per capita for a long period, with minimal poverty reduction, persistent inequality, and social marginalization of minorities. This social situation worsened further because of the economic crisis of the late 1990s. The social impact of dol- larization has to be evaluated against this background. Gender biases, in turn, seem to make crises affect women more adversely (see chapter 5). Dollarization, as we document in this book, has not been costless in Ecuador. The exchange rate chosen for conversion of the money supply in sucres to dollars (25 sucres per dollar) was a very undervalued rate. As a consequence, there was a sharp reduction in real wages in dollars. As CRISIS AND DOLLARIZATION: AN OVERVIEW 13 inflation continued at a significant (although declining) pace after dollar- ization, real wages suffered from continuous inflation and slack in the labor market. However, as of early 2002, about two years after dollariza- tion was adopted, the real exchange rate has started to appreciate and real wages to recover. In addition, following the pattern of other exchange-rate-based stabilization plans, a recovery of consumption and an increase in relative prices of nontradable goods and assets accompa- nied a recovery in economic activity, with declining unemployment and some improvement in deteriorated social conditions. Still, the medium- term effects on the external competitiveness of Ecuador of currency appreciation must not be neglected. In retrospect, the social impact of dollarization was affected by the pre- carious nature of social safety nets in Ecuador that were unable to shield vulnerable groups, the poor, women, and the unemployed from the social costs associated with both the economic crisis of 1998-99 and the stabi- lization efforts afterwards. Organization of This Book This volume comprises four other chapters. In chapter 2, Paul Beckerman presents a broad analysis of long-term characteristics of the Ecuadoran economy covering several dimensions: economic structure, geography, social structure, and regional divides; frequency of governance crisis; dependence on volatile commodity prices; fiscal and financial structure; and exposure to natural disasters. The chapter places the late 1990s eco- nomic crisis that preceded official dollarization in historical perspective. Chapter 3, by Paul Beckerman and Hernan Cortes-Douglas, provides an in- depth analysis and documentation of the Ecuadoran experience following dollarization. It analyzes in detail the workings of the new monetary sys- tem, the behavior of the fiscal and banking systems, the adjustment in prices and the real exchange rate, and real economic activity after official dollarization in 2000 and 2001. Chapter 4, by Suhas Parandekar, Rob Vos, and Donald Winkler, elaborates ancl carefully documents the effects of the crisis on unemployment, real wages, and income distribution as well as the effects and limitations of policies to counteract these adverse social effects. Finally, chapter 5 by Maria Correia discusses the gender and family dimen- sions of Ecuador's severe economic crisis of the late 1990s. Notes 1. The old national currency, the sucre, retained de jure legal status under the Constitution, essentially because the government believed it would simply have been too complicated to change the Constitution. 14 CRISIS AND DOLLARIZATION IN ECUADOR 2. By December of 1999, around 66 percent of total deposits in the Ecuadoran financial system were in U.S. dollars and nearly 90 percent of the credit was in dollars. 3. Historically, not all experiences with very high inflation and hyperinflation came along with a banking crisis; see Solimano (1990a, 1991). 4. See Fischer (2001b) for an account of the relationship between Ecuador and the IMF, from the perspective of the latter. 5. It is estimated that around 1 million Ecuadorans left the country between 1998 and 2001. 6. See Hurtado (1993) for a thorough analysis of Ecuador's economic, social, and political structure in both the colonial period and in the Republican era. 7. See Arteta and Hurtado (2002) for a recent political economy analysis of Ecuador. 8. See Eichengreen (2002) for analysis of alternative views on the timing and sequencing of dollarization. Berg and Borenztein (2000) and Calvo (1999) review several pros and cons of dollarization as an exchange-rate regime. 9. See Moreno-Villalaz (1999) for an analysis of the Panama experience as a dol- lar economy. 10. A more acute situation of near hyperinflation before adopting a hard peg regime was observed in Bulgaria in the 1990s. That country adopted a currency board in July of 1997. Preceding the currency board, inflation reached 500 percent in January 1997 and more than 2,000 percent in March of that year; see Gulde (1999). For an early analysis of inflation dynamics in post-Communist Bulgaria, see Solimano (1990b). 11. See Arteta (2001) for an analysis of that period. 12. See Balino and Enoch (1997). 13. See Scandizzo (2001). 14. See Fischer (1982 and 1993) for an assessment of the arguments for national money and their empirical significance. 15. See Fischer (2001a) for an interesting discussion of the bi-polar view of exchange-rate regimes. 16. See Calcagno, Manuelito, and Titelman (2001) for a comparison of Ecuador's dollarization with Argentina's currency board. References The word processed describes informally reproduced works that may not be commonly available through libraries. Arteta, Gustavo. 2001. "Dollarization in Ecuador: Experiences, Challenges, and Lessons.!" Americas' Insights, September. CRISIS AND DOLLARIZATlON: AN OVERVIEW 15 Arteta, Gustavo, and Osvaldo Hurtado. 2002. "The Political Economy of Ecuador." Project Political Economy of Andean Region. ECLAC. Processed. Balifno, Tomas, J.T., and Charles Enoch. 1997. "Currency Boards Arrangements. Issues and Experiences." Occasional Paper 151. International Monetary Fund, Washington, D.C. Berg, Andrew, and Eduardo Borenzstein. 2000. "The Pros and Cons of Full Dollarization." International Monetary Fund Working Paper, WP/00/50. Calcagno, Alfredo, Sandra Manuelito, and Daniel Titelman. 2001. "Dollarization in Ecuador: A Parallel with Argentine ConvertibiLity." Economic Commission for Latin America and the Caribbean. Processed. Calvo, Guillermo. 1999: "On Dollarization." University of Maryland. Processed. Eichengreen, Barry. 2002. "When to Dollarize." Journal of Money, Credit and Banking 34(1):1-24. Fischer, Stanley. 1982. "Seigniorage and the Case for National Money." Journal of Political Economy 90(April):295-313. . 1993. "Seigniorage and Official Dollarization." In Nissan Liviatan, ed., Proceedings of a Conference on Currency Substitution and Currency Boards, pp. 6-10. World Bank Discussion Paper 207. World Bank, Washington, D.C. 2001a. "Exchange Rate Regimes: Is the Bi-Polar View Correct?" Distinguished Lecture on Economics in Government, American Economic Association and Society of Government Economists. . 2001b. "Ecuador and the IMF." Hoover Institution Conference on Currency Unions, Palo Alto, California. Gulde, Anne-Marie. 1999. "The Role of Currency Board in Bulgaria's Stabilization." International Monetary Fund Policy Discussion Paper, PDP/99/3. Hurtado, Osvaldo. 1993. El Poder Polftico en Ecuador. Editorial Planeta, Quito. 16 CRISIS AND DOLLARIZATION IN ECUADOR Moreno-Villalaz, Juan Luis. 1999. "Lessons from the Monetary Experience of Panama: A Dollar Economy with Financial Integration." Cato Journal 18(3). Scandizzo, Stefania. 2001. "Options for Monetary Integration in the Andean Community." Andean Development Corporation, Caracas, Republica Bolivariana de Venezuela. Processed. Solimano, Andres. 1990a. "Inflation and the Costs of Stabilization: Historical and Recent Experiences and Policy Lessons." World Bank Research Observer 5(2):167-85. . 1990b. "Inflation and Growth in the Transition from Socialism: The Case of Bulgaria." World Bank Policy Research Working Paper Series # 659. . 1991. "The Economies of Central and Eastern Europe: An Historical and International Perspective." In Vittorio Corbo, Fabrizio Coricelli, and Jan Bossak, eds., Reforming Central and Eastern European Economies. Initial Results and Challenges. The World Bank. . 2002. "Crisis in the Andean Region: A Political Economy Analysis." Project on Political Economy of the Andean Region. United Nations Economic Commission for Latin America and the Caribbean, Santiago, Chile. Processed. 2 Longer-Term Origins of Ecuador's "Predollarization" Crisis Paul Beckerman 1. Introduction On January 9, 2000, Ecuador's government fixed its exchange rate, which had been floating for nearly 11 months, and announced that it would sub- mit legislation to the Congress to fully dollarize the economy. At that moment, Ecuador's sucre was in apparent free fall, having lost two-thirds of its U.S.-dollar value during 1999 and a quarter during the first week of the new year alone. Real GDP had fallen 7.3 percent in 1999, and the recession was apparently still deepening. Commercial banks were in deep crisis: several large ones had failed, and credit operations were vir- tually suspended. A liquidity crisis loomed as banks prepared for March 2000, when release of time deposits, frozen for a year in March 1999, was to commence. On January 21, President Jamil Mahuad, who had been elected to a five-year term in mid-1998, was forced from office, mainly because of dis- satisfaction with the economy and opposition by some people to dollar- ization. After an unsuccessful coup attempt by some military officers and leaders of Ecuador's indigenous people, the vice president assumed the presidency, just barely maintaining constitutional normality. Seeing no alternative, the new government pressed forward with dollarization. In early February it submitted the necessary legislation to the Congress, which approved it after rapid debate. Some left-of-center parties expressed their opposition by not participating. Once the president The writer is an independent consultant. The writer thanks Heman Cortes Douglas, Luis Jacome, and Andres Solimano for valuable comments on earlier drafts. The writer alone is responsible for any errors of fact and judgment. Views expressed here do not necessarily reflect views of the World Bank or any other institutions with which the writer has been associated. Please do not cite this work without the writer's authorization. 17 18 CRISIS AND DOLLARIZATION IN ECUADOR approved the law in early March, the Central Bank began purchasing sucres from circulation, and the country adopted the U.S. dollar as its cur- rency. The conversion was complete well before the year's end. The crisis that precipitated the move to dollarization was triggered in late 1997 and 1998 by a combination of exogenous external and climatic shocks. These shocks included (a) plunging oil-export prices, (b) heavy damage from El Nifno rains, and (c) various effects of the East Asian, Russian, and Brazilian financial crises. The shocks widened the 1998 cur- rent-account deficit and made the pre-announced crawling-band exchange-rate policy unsustainable, forcing the authorities to float the sucre in early 1999. By reducing revenue, increasing the domestic-cur- rency equivalent of the public-debt service, and forcing increased expen- diture to cope with the El Nifno disaster, the shocks widened the 1998 fiscal deficit. In addition, the shocks damaged commercial banks' loan portfolios. The exchange-rate depreciation was especially hard on the banks, because their balance sheets were partially dollarized. Although they had made a point of keeping dollar assets matched with dollar lia- bilities, many of the banks' dollar borrowers were themselves unmatched, with sucre earnings backing dollar liabilities. Exchange-rate depreciation therefore increased banks' nonperforming assets and reduced cash repayments. Depositors, fearing for banks' safety, began withdrawing, intensifying banks' illiquidity, even as the depreciation swelled the dollar deposit stocks in sucre terms. The authorities tried to deal with the banking crisis in late 1998 by fully guaranteeing all bank deposits, and then, in March 1999, by freez- ing deposits. The deposit freeze threw the economy into disarray, how- ever, and the authorities found they had no choice but to unfreeze checking and savings accounts gradually. Withdrawals intensified, how- ever, and the Central Bank found it had little choice but to provide the banks credit to prevent a payments-system collapse. This domestic-credit creation more than doubled the monetary base during 1999, inducing a sharp exchange-rate slide toward the year's end. If the authorities had not dollarized when they did, hyperinflation was pretty much inevitable. This chapter examines the underlying causes of this "predollarization" crisis -that is, the reasons why the 1998 shocks produced a crisis of such magnitude. Similar shocks had serious consequences in neighboring economies at the same time, but in no case so devastating as in Ecuador. The main thesis is that a combination of specific characteristics of Ecuador's economic and political systems accounted for the severity of the crisis. These characteristics included (a) the heavy dependence of public revenue on volatile oil earnings, (b) the banking system's exposure to Ecuador's volatile and risky activities, (c) inadequate banking super- vision, (d) political fragmentation, (e) weak public administration, (f) the LONGER-TERM ORIGINS OF ECUADOR'S "PREDOLLARIZATION" CRISIS 19 political system's tendency to maintain energy subsidization, and-par- ticularly important-(g) the financial systern's partial dollarization. These characteristics were in turn the consequences of deeper geographic and historical realities, including (a) rivalry between Ecuador's coastal and highland regions, (b) the volatility of Ecuador's commodity export markets, and (c) the country's exposure to natural disasters, including earthquakes, volcanic eruptions, episodes of excessive rainfall, and drought. In addition, since the 1970s, the economy has suffered from the interrelated consequences of (d) excessive public-debt accumulation, (e) lagging and uneven structural reform in the public and financial sectors, and (fl the exchange-rate instability deriving from the need to cope with the external debt. This chapter will first set out the longer-term background to the predollarization crisis and then show how the characteristics of the econ- omy and society affected the evolution of the crisis in 1998 and 1999. Part 2 describes the deep historical origins of the crisis. Its sections discuss (A) the historical background of Ecuador's regionalism, political fragmenta- tion, and administrative weakness; (B) the economy's unusual vulnera- bility to "contingencies"; and (C) the ways in which the structural changes carried out by the military government during the 1970s oil boom led to heavy external-debt accumulation and to many of the struc- tural problems that still awaited reform when the crisis began.1 Part 3 describes the state of structural reform going into the crisis. Its sections focus on (A) the public sector; (B) the financial system and the crucial problem of its partial dollarization; and (C) certain additional areas in which structural-reform agendas remain-formal labor markets and trade policy. With this background, Part 4 recounts the evolution of the 1998-99 crisis. Part 5 then presenis summary conclusions regarding the underlying causes of the crisis. 2. Historical Background of Ecuador's Predollarization Crisis The root causes of the crisis that preceded Ecuador's move to dollariza- tion were the country's deeper historical and geographical circumstances. The political and administrative weakness of Ecuador's government helps explain its inability to act rapidly and forcefully to deal with the combination of contingencies that thrust the economy into crisis. The essential conditions of the crisis, however, were (a) the structural depen- dence on oil exports beginning in the 1970s, (b) the massive accumulation of public debt that resulted, and (c) the need to generate an export sur- plus to service the debt, which, together, led to (d) the partial dollariza- tion of the financial system. 20 CRISIS AND DOLLARIZATION IN ECUADOR A. Historical Roots of Ecuador's Governance Problems Complex governance problems, including regional rivalry, political frag- mentation, weak public administration, and pervasive corruption, figure centrally among the reasons for Ecuador's poor growth performance, par- ticularly during the past two decades. These problems derive in large mea- sure from the peculiar nature of Ecuador's historical formative process.2 Rivalry between the "Sierra" (mountain highlands), centered on the capital, Quito, and the "Costa" (coastal lowlands), centered on the port city of Guayaquil, has been a standing theme of Ecuador's history. The two regions have always been culturally and economically distinct. The balancing of regional interests has imparted a marked style to political and administrative structures and decisionmaking. One consequence of the regional rivalry is that the central government's political and admin- istrative powers and capacities have been limited. Even at moments of national crisis, government policies and actions have often had to subor- dinate broader national interests to reconciliation of regional interests. The 1998 banking crisis (see Part 4) is a clear example of this: The worst- affected banks were Guayaquil-based, but regional sensitivities and polit- ical interference prevented the banking supervisors from taking timely, effective action. Before the Republic's creation in 1830, what are now its two main regions had been quite separate from one another. The present Sierra region was an important part of the Inca Empire; in contrast, the Incas subjugated the pre- sent coastal areas only a few decades before the Spanish arrived in 1532, and had not yet integrated them into their empire.3 During the three centuries of Spanish rule, Quito and surrounding areas developed in isolation from the coastal lowlands, which consisted largely of self-governing indigenous communities. Guayaquil developed as a center for intracolonial sea trade (mostly illegal under Spanish mercantilist rule). Although many crucial events in the continental independence campaign of the early 1800s took place in what is now Ecuador, events in the highlands and on the coast were basically unrelated.4 Quito's and Guayaquil's initial uprisings, for example, were uncoordinated. In their July 1822 meeting in Guayaquil, Bolivar apparently persuaded San Martin to permit Guayaquil to join Quito in a "District of the South" within the Gran Colombia confederation (compris- ing modern Colombia, Panama, the Repiublica Boliviariana de Venezuela, and Ecuador), rather than Peru. Guayaquil's inhabitants were fearful of Lima's dominance. In 1830, when Gran Colombia disintegrated, this District of the South constituted itself as the "Republic of the Equator." The new republic underwent decades of internal struggle to determine how it would be governed. During the republic's first century, ideologi- cal and regional interests coincided: proclerical, centralizing, landowning LONGER-TERM ORIGINS OF ECUADOR'S "PREDOLLARIZATION" CRISIS 21 "Conservatives" were based in Quito, while anticlerical, decentralizing, commercial "Liberals" were based in Guayaquil. From 1860 until 1895 the Conservatives were generally dominant. Between 1860 and 1875 Ecuador achieved some material progress under Gabriel Garcia Moreno's procler- ical, centralizing dictatorship, but this government engendered regional resentment. From 1895 until 1925 the Liberals generally dominated the country, largely on the basis of the rise of the coastal cacao economy. The Liberals' own factional conflicts subjected the country to instability, how- ever. Over the course of its first century, Ecuador repeatedly changed its constitution and governmental structure, introducing and removing an upper legislative chamber, changing the role of the Church, extending the suffrage, and so on.5 The regional rivalry has led, first, to a set of standing compromises limiting central political and administrative powers, and, second, to the political prioritization of regional balance. Fearful that an administration largely from one region might impose its will on the other, the writers of Ecuador's constitutions tended to limit central political and administra- tive power, in particular, executive power. Twentieth-century constitu- tions made it relatively easy for the Congress to impeach and remove cabinet ministers, for example, even on purely political grounds. Legal intimidation-prosecuting or suing public officials, for example-has often been used to limit political and administrative power. Administra- tive institutions have been kept weak, both in mandates and capabilities. The determination of the regions to protect their positions and interests manifests itself in many other ways. For example, the country's long- standing practice of tax earmarking ensures, among other things, that particular localities and interest groups receive "fair shares" of national resources, at the cost of complicating expenditure programming, and the traditional cross-party regional caucuses of the Congress work together to advance regional interests. Since July 1925, when the Conservatives and Liberals gave way to new political groupings in the wake of the Revoluci6n Juliana (see section B below), Ecuador's political parties have been fragmented and unstable. As of mid-2001, representatives of 10 parties sit in the (unicameral) national Congress. The parties are difficult to classify ideologically. Pop- ulism figures heavily in their styles and substance. Of the four largest, two are relatively, if inconsistently, center-right and center-left parties based mainly in the Sierra and two are relatively center-right and center- left parties based mainly in the Costa.6 Another party (Pachakutik) claims exclusively to represent indigenous ethnic minorities. During 1998 and 1999, party fragmentation made it difficult to pass emergency legislation that was essential precisely because of the limitations of the central gov- ernment's executive and administrative powers (see Part 4). 22 CRISIS AND DOLLARIZATION IN ECUADOR A paradoxical consequence of Ecuador's regionalism has been a long- staniding failure to develop effective subnational, governments. While regional interests have sought to limit central political and administrative power since Independence, they have also generally sought -to limit regional autonomy, for fear that it could break the country apart. As a consequence, provincial administrations have simply not had the powers and the resources they needed for full governmental effectiveness. Some municipal administrations have proved more effective, but in general they have suffered from inadequate resources. Although Ecuador is small by comparison with its neighbors, it has a large territory with diverse populations. Proper application of the subsidiarity principle would undoubtedly improve administrative efficiency and enhance political participation and accountability. In recent years, Ecuadorans have been formulating and debating proposals for regional autonomy and decen- tralization. It remains to be seen, however, whether these proposals can be implemented in ways that would be politically, administratively, and financially feasible. B. Ecuador's Vulnerability to Economic Shocks and Natural Contingences Just as its deeper political history helps explain its political fragmentation and administrative weakness, geography and topography are funda- mental to explain the "contingency" to which Ecuador is subject. Ecuador has always depended heavily on volatile commodity export earnings and has always been especially vulnerable to seismic and climatic disasters. This section reviews the historical processes leading up to the oil boom of the 1970s, discusses Ecuador's exposure to natural contingencies and shocks, and briefly discusses the more recent instability of external finan- cial flows, which has become an additional dimension of vulnerability. DEPENDENCE ON COMMODrry ExPoRTs. Throughout its history, reliance-on primary commodity exports has subjected Ecuador, like many other Latin American economies, to debilitating boom-and-bust cycles. During the 20th century, three commodity exports-cacao, bananas, and oil- played crucial roles in the country's economic and political development. Cacao, produced in coastal agricultural areas, gradually became Ecuador's first large-scale commodity export during the latter part of the 19th century. Cacao exporters' growing wealth powered the Liberal Party's rise. During the first two decades of the 20th century, the Liberal Party fell under the sway of a group of Guayaquil cacao producers and the trading firms and banks associated with them.7 The party's control of the government enabled some business figures to divert revenue flows to LONGER-TERM ORIGINS OF ECUADOR'S "PREDOLLARIZATION" CRISIS 23 -themselves or their supporters, while the banks extended profitable loans to cover a growing public deficit. The banks created money (there was neither a central bank nor a banking supervisor), causing an inflation problem. This "system" collapsed in the 1920s when a series of external shocks struck the cacao economy. Fungal disease sharply reduced output, while growing cacao exports by British colonies (and later, the onset of the world Depression) drove downi world prices. As a result, real wages and incomes fell sharply, and deflation set in. In the early 1920s Guayaquil workers carried out a general strike, while in the Sierra peas- ants organized protest movements. The government repressed these vio- lently. In July 1925 a "League of Young Army Officers" seized power (the Rev- oluci6n Juliana). Its stated objective was to end the Conservatives' and Lib- erals' dominance and to begin carrying out modernizing reforms. It organized a provisional government, naming Isidro Ayora, a wealthy opponent of the Liberals from Guayaquil, to serve as president. His gov- ernment's accomplishments included drafting a new constitution (Ecuador's 13th), which enhanced the power of the legislature and diminished the power of the executive; wide-ranging fiscal and monetary reforms (recommended by an advisory mission headed by Edwin Kem- merer of Princeton University), including establishment of the Central Bank; and progressive social reforms, including establishment of the state pension system. The collapse of the cacao economy prevented economic recovery, however: Cacao exports fell from US$15 million in 1928 to US$7 million in 1931 and US$5 million in 1932. The government could not deal effectively with the economy, and in 1931 another military coup removed Ayora from office. From the standpoint of export-commodity dependence, the half-cen- tury between the Revoluci6n Juliana and the rise of the "oil economy" in the 1970s can be divided roughly into three periods. In the first period, from approximately 1925 until the late 1940s, Ecuador's economy contin- ued to stagnate in the aftermath of the cacao collapse. A populist move- ment emerged in the early 1930s, under the personalist, charismatic leadership of Jos6 Maria Velasco, who served as president five times between 1934 and 1972. The 1930s and 1940s were a period of political instability, during which the government alternated among (a) represen- tatives of the Quito and Guayaquil elites; (b) Velasco and his supporters, who sought generally to implement expansionary public-expenditure policies; and (c) the military, who intervened several times to change gov- ernments but did not themselves retain power over extended periods. The second period, from 1948 to 1958, was characterized by political sta- bility, mainly because it coincided with the rise of Ecuador's banana economy. During the 1940s, after disease devastated Central American 24 CRISIS AND DOLLARIZATION IN ECUADOR banana plantations, Ecuador began producing and exporting heavily. Banana exports grew from US$2 million to US$20 million between 1948 and 1952 as volumes and prices rose. Three presidents, including Velasco, served full four-year constitutional terms between 1948 and 1960. Until the late 1950s the favorable effects of the banana boom on the economy and hence on government revenue enabled all three presidents to govem without having to confront serious economic crises.8 Renewed economic instability characterized the third period, begin- ning in the late 1950s. Falling banana prices brought about recession, unemployment, and intensified social protest. In 1960 Velasco was elected to his fourth term as president, promising to confront the eco- nomic downturn. Declining govemment revenue made it impossible for him to make good on his electoral promises, however, and he was forced to resign just over halfway through his term. Soon afterwards the military took power themselves, announcing that this time they intended to retain power long enough to carry out modernizing reforms. In 1964 this gov- ernment enacted a land reform that significantly changed land tenancy in the Sierra, although it preserved commercial holdings in the Costa. Per- sisting low commodity-export prices, however, made it no less difficult for the military government to manage the economy and the fiscal accounts effectively. Unable to agree on a policy program to confront the economic malaise, and increasingly unpopular because of political repression, the military decided to step down in 1966. Following an interim govemment, which produced a new constitution, new elections were held in 1968. Velasco was then elected President for the fifth time, winning a plural- ity of votes cast among five candidates. For three years, his government struggled in the face of inadequate legislative support and low banana export receipts to maintain a populist spending program. In June 1970 he assumed dictatorial powers, dissolving the Congress and dismissing the Supreme Court. Two months later he devalued the sucre from 18 to 25 per dollar (the rate of 18 had stood for nearly a decade), instituted capital controls, and decreed tax and tariff increases. In February 1972, however, largely to head off the election of a populist candidate for president they disliked, the military removed Velasco and assumed power. Earlier, in 1964, the government had granted prospecting and development conces- sions for the Amazon basin to several foreign oil companies, several of which made significant discoveries within several years. By the early 1970s, following construction of an oil pipeline from the producing fields over the Andes Mountains to the coast, sizable revenue flows seemed likely. Upon taking power in 1972, the military declared that they would retain power long enough to ensure that the oil earnings were applied to national development and social reform.9 Unfortunately, oil turned out to LONGER-TERM ORIGINS OF ECUADOR'S "PREDOLLARIZATION" CRISIS 25 be yet another volatile export commodity (section C below takes up the themes of oil-based growth and external debt since 1970). ExPosuRE TO FORCES OF NATURE. Along with world commodity markets, forces of nature have been another standing source of contingency for Ecuador's economy. Ecuador is. prone to earthquakes, landslides, vol- canic eruptions, and extended periods of. both drought and excessive rain. The record during the past 20 years indicates the nature of the prob- lem. In 1975 and 1983, Ecuador experienced damaging El Nifno episodes, with severe rain damage to coastal agricultural output and transport infrastructure and a substantial decline in fishing production. A new round of the phenomenon in 1998 was one of the shocks contributing to the predollarization crisis. Infrastructure accumulation and population growth imply that, as time passes, the consequences of any particular nat- ural disaster increase. Drought has been a recurring problem, affecting agricultural production and electricity generation in several recent years. A drought in 1995, for example, affected export and domestic food crops as well as electrical power supplies. Earthquakes are a standing hazard. In 1987, an earthquake tore apart 40 kilometers of the Transandean Pipeline, stopping oil production for five months (see section C follow- ing). Volcanoes are another hazard. During 1999, in the midst of the eco- nomic crisis, two eruptions-one (Guagua Pichincha) on the outskirts of Quito, the other (Tungurahua) near a rich agricultural and resort area- further disrupted economic activity and created uncertainty. Relatively few lives were lost, but property damage was considerable and tourism was affected. Many countries face standing risks from natural phenom- ena. Still, were one to list the world's economies according to the fre- quency and. variety of their natural disasters, Ecuador would surely rank relatively high. The likelihood of the occurrence of natural disasters dis- courages many kinds of investment, and the disasters themselves tend to have significant consequences for economic growth and stability. EXPOSURE TO SHIFTS IN CROSS-BORDER FINANCIAL FLOWS. More recently, inte- gration with world financial markets has exposed Ecuador to another dimension of volatility. Wealth-holders have been able to move resources with increasing ease between on- and offshore placements, as perceptions and realities of relative rates of return and bank safety evolve. As in other economies, this activity has imparted an additional dimension of vulner- ability to financial activities. In 1994 and 1995, Ecuador experienced a round of inflows followed by outflows (see section C below), which intensified the business cycle. Meanwhile, Ecuadoran financial institu- tions have continued to do business with foreign banks, coming to rely heavily on external funds for trade and working-capital credit. During 26 CRISIS AND DOLLARIZATION IN ECUADOR 1998 many' of these. lines were withdrawn-again, at a moment when their withdrawal was especially'inconvenient.both for banks and for the balance of payments (see Part 4). The benefits and drawbacks of cross- border financial-capital flows have been a subject of worldwide contro- versy. While presumably. beneficial for developing econornies, since they augment the resource base for capital forrmation, in Ecuador as elsewhere they are an additional source' of economic vulnerability. C. Oil, External Debt, and Ecuador's Exchange-rate Instability Ecuador's present external-debt problem can be traced back to the 1970s, when the start of large-scale oil exports generated a growth spurt and the private and public sectors began borrowing heavily. No less important, a large part of the present unfinished structural-adjustment agenda involves reversing changes the military government made in the 1970s. This section reviews Ecuador's macroeconomic evolution from the mid- 1970s up to the start of the predollarization crisis. The essential argument is that the need to service the external debt reduced ecbnomic growth and sustained the exchange-rate instability. The start of large-scale oil exports came at just the moment that the Organization of Petroleum Exporting Countries (OPEC) succeeded in rais- ing world oil prices. In 1972, soon after taking power, the military renego- tiated some of the concession contracts to increase the nation's share of the proceeds.10 In 1973 they took Ecuador into OPEC. On the basis of the surge in oil revenue, the military government increased public-sector employ- ment and capital formation rapidly, raising overall government expendi- ture by about two-thirds between 1972 and 1975. It reduced domestic taxation: non-oil public-sector revenue fell from 18.7. percent of GDP in 1972 to 13.8 percent in 1975, while oil revenue rbse from 2 to 8.4 percent. The government applied part of the oil earnings to subsidize domestic elec- tricity and oil derivatives. From 1970 to 1977 'annual real GDP growth exceeded 9 percent"l (compared with just below 6 percent in the 1960s). As the economy grew, Ecuador's private sector-mainly commercial banks-began borrowing from foreign banks engaged in "recycling" OPEC surpluses. During 1974 and 1975, however,' rising aggregate demand induced inflationary pressure. Oil revenue slipped as world recession drove down world oil prices. The 1975 El Nifio episode affected coastal agriculture and fishing, reducing government revenue. To avoid raising non-oil taxes or reducing expenditure and subsidies, the govern- ment began borrowing extemally to finance its deficit. At the end of 1979 the overall public external-debt stock reached US$4.5 billion (about 28 percent of GDP), compared with US$324 million (20 percent of GDP) at the end of 1970 (see figure 2.1). LONGER-TERM ORIGINS OF ECUADOR'S "PREDOLLARIZATION" CRISIS 27 Figure 2.1 Ecuador: Year-end Public and Publicly Guaranteed External Debt (US$ million), 1970-2000 Us$ million 16,000 14,000 12,000 _ F _ 10,000 8,000 =[|2~-^~~~1~~ 1970-2000 * Multilateral E Bilateral aD Bonds e Otherprivate sources Source: World Bank. In 1976 the military government, its support eroded by inflation and internal disputes, decided to restore constitutional government. In 1979, after a lengthy process including the elaboration of a new constitution, a popular Guayaquil political figure, Jaime Rold6s, was elected president. His government faced deepening economic problems. Some were of its own .making: Soon after taking office, the government raised the mini- mum wage and other wage benefits, which had significant fiscal conse- quences. The larger problem, however, was an external shock. Following the U.S. Federal Reserve's monetary tightening and the onset of world recession, rising interest rates and diminished oil and other commodity prices thrust Ecuador, like most other South American economies, into debt crisis. Higher interest rates on floating-rate debt led to a sharp dete- rioration in both private- and public-sector financial positions. President Rold6s' death in a May 1981 air disaster further complicated the govern- ment's problems. The vice president, Osvaldo Hurtado, immediately assumed the presidency and began steering the economy into adjustment to the new macroeconomic realities. 28 CRISIS AND DOLLARIZATION IN ECUADOR In 1982, under the pressure of surging interest rates on its external debt and recession induced by declining terms of trade, Ecuador's overall public-sector deficit reached 7.5 percent of GDP and its current-account deficit widened to nearly 8.5 percent of GDP. In May of that year, as part of an IMF-supported program, the government devalued the exchange rate, which had been fixed at 25 sucres per dollar since 1970, by 25 per- cent against the dollar. It also raised the (controlled) banking-system interest rates and raised the prices of a broad range of public-sector goods and sefvices. In March 1983 the authorities devalued again and then com- menced mini-devaluations to keep the exchange rate from slipping behind the price level. Consumer prices rose 63.4 percent over 1983 prices while real GDP declined 2.8 percent, largely because of that year's severe El Nino episode. On the basis of the IMF program (which succeeded in the sense that Ecuador met its conditionality and it was disbursed fully), foreign commercial banks reached agreement with Ecuador on debt rescheduling, and in July 1983 the Paris Club agreed to reschedule amor- tization due between June 1983 and May 1984. Repeated exchange-rate depreciation was a fundamental change for Ecuador, with at least two lasting consequences. First, it sharply increased the private sector's-in particular, commercial banks'-exter- nal debt-service obligations. Beginning in 1983, the Central Bank assumed the bulk of the private external debt (about US$1.5 billion, 11 percent of 1982 GDP) in exchange for sucre debt, under its sucretizaci6n policy12 (see Bayas and Somensatto 1994). Sucretizaci6n not only added significantly to the public debt stock, it also angered many citizens, who believed it was inappropriate to use public resources to bail out private borrowers and foreign creditors. The second consequence of the sliding exchange rate was that it set an incentive to move private wealth into dol- lars. At first this meant offshore placements, since until the 1990s bank operations in dollars were closely restricted. "Spontaneous" dollarization of informal contracts, real-estate valuations, professional services, and so on, also became increasingly widespread. Throughout the 1980s and 1990s, policymakers struggled to set an exchange rate that was both stable and set the incentives for the net export flow required to service the external debt. They never succeeded in doing so permanently, however. Real growth, inflation, exchange-rate depreciation, and the public deficit remained highly unstable. Under Le6n Febres Cordero's liberalizing government, which took office in August 1984, real GDP rebounded, growing 4.2 and 4.3 percent in 1984 and 1985, respectively, while inflation moderated to around 20 percent. In 1986, however, at the same time it began liberalizing commercial-bank interest rates, the government began a floating exchange rate for private- sector imports. Oil-export prices fell by more than half that year, how- LONGER-TERM ORIGINS OF ECUADOR'S "PREDOLLARIZATION" CRISIS 29 Table 2.1 Ecuador: Governments, 1979-2001 Acceded to Departure Period President office through from office Aug 79-May 81 Jaime Rold6s Election Accidental death May 81-Aug 84 Osvaldo Hurtado Vice Presiden't, Term concluded assumed office Aug 84-Aug 88 Le6n Febres Cordero Term concluded Aug 88-Aug 92 Rodrigo Borja Election Term concluded Aug 92-Aug 96 Sixto Duran Ballen Election Term concluded Aug 96-Feb 97 Abdala Bucaram Election Removed by Congress Feb 97-Aug 98 Fabian Alarc6n Designation by Term concluded Congress Aug 98-Jan 00 Jamil Mahuad Election Resigned Jan 00- Gustavo Noboa Vice President, assumed office ever, and in January 1987 the government suspended debt service to com- mercial banks. The March 1987 earthquake interrupted oil exports, and international reserves declined precipitously. The Central Bank tightened monetary policy to head off exchange-rate depreciation, but then finally stopped the float when the exchange rate came under speculative attack. Real GDP fell 6 percent in 1987 while consumer prices rose nearly 100 percent. Oil exports resumed in August 1987 after the pipeline was repaired, but with continuing high expenditure and subsidies the public deficit surged to more than 12 percent of GDP. The debt-service suspen- sion continued for seven years. Resumption of oil exports at somewhat higher prices led to a 10.5-per- cent GDP-growth rebound in 1988, but the public deficit remained mas- sive, on the order of 10 percent of GDP. The external-debt stock, now accumulating mainly through interest arrears, grew more slowly. In mid- 1988, soon after taking office following elections, a center-left government under President Rodrigo Borja announced a new policy package, including yet another large devaluation and introduction of a new exchange-auction system, with differentiated rates for private and public exporters and pri- vate importers. It then carried out mini-devaluations consistent with antic- ipated inflation of 30 percent. Consumer prices rose 54 and 50 percent in 1989 and 1990, respectively, however, and real growth slid to 0.3 percent in 1989. Over the next three years, real growth recovered to between 3 and 5 percent. Inflation persisted in the range of 50 to 60 percent, and the non- financial public-sector deficit remained around 6 to 7 percent of GDP, despite higher oil prices at the time of the Persian Gulf War. Building on the Cordero Government's reforms, the Borja Government undertook several 30 CRISIS AND DOLLARIZATION IN ECUADOR significant structural-adjustment initiatives, including a partial tax reform, trade liberalization, and progress toward financial-sector liberalization. In the run-up to the mid-1992 elections, however, government expenditure rose sharply, and, although the authorities maintained a relatively appreci- ated exchange rate, inflation persisted at high rates. In mid-1992, Sixto Duran Ballen was elected president on a platform of stabilization, liberalization, and structural reform. Soon after taking office, his government announced another large policy package, encom- passing a devaluation of 20 percent against the dollar and various fiscal measures, including increases in motor-fuel prices and electricity rates, a company-assets tax, expenditure cuts, and a public-employment freeze. These actions cut the 1993 public deficit nearly to zero. In August 1993 the authorities unified the foreign-exchange market and began a new policy of floating within a pre-announced crawling band. The idea was to set a nominal anchor that-would help gradually to reduce the inflation rate. This exchange-rate policy continued until 1998 (see Part 4). Meanwhile, the Duran Ballen Government began developing and implementing a substantial structural-reform program (see Part 3, section A). In late 1993 and 1994, following the introduction of the float-within-a- crawling-band exchange-rate policy, Ecuador experienced a short-term, financial-capital inflow (see Jaramillo 1994). Unlike other economies, where capital flows tended primarily to go to the stock markets, these inflows went mostly to short-term, fixed-income applications, since high, short-term interest rates were now available on sucre deposits: The place- ments were made mostly by Ecuadoran nationals, repatriating holdings taken abroad in the 1980s. The inflows themselves increased foreign- exchange reserves and so seemed to reduce exchange-rate risk, encour- aging further inflows. The high interest rates on short-term sucre deposits were a standing source of instability, however: Capitalization of rates on the order of 30 percent into deposit balances meant that this became a basic growth rate for these monetary stocks, helping to sustain this rate as the economy's "inertial" inflation rate during the 1990s. The period of arrears accumulation that commenced in January 1987 concluded in 1994, when, with IMF, World Bank, and IDB support, the authorities secured a debt-and-debt-service reduction (DDSR) deal with commercial-bank creditors, leaving Ecuador owing Brady bonds totaling just under US$6 billion (see figure 2.1). (The multilateral agencies financed the collateral for the Brady bonds.) Even so, by the end of 1998 total public and publicly guaranteed external debt outstanding stood at just over US$13 billion, about two-thirds of 1998 GDP-the heaviest bur- den by far among Latin Arherica's 10 largest economies.13 (In September 1999 Ecuador would become the first nation to suspend servicing of Brady debt. See Part 4.) LONGER-TERM ORIGINS OF ECUADOR'S "PREDOLLARIZATION" CRISIS 31 The national-accounts data suggest that in the mid-1980s, under the stress of its adjustment to the debt crisis, Ecuador's macroeconomy shifted to a lower-growth mode. Figure 2.2 shows that p'er-capita real GDP and real consumption grew relatively rapidly until the early 1980s, along with the external debt, but then leveled off, remaining nearly unchanged over the past two decades. (The years 1987 and 1988 were exceptional because of the earthquake damage to the oil pipeline and the subsequent recovery.) Figure 2.3 shows that the gross capital-formation rate rose through the late 1960s to relatively high levels. It remained high through the 1970s, but then declined abruptly in 1983, and has remained since then at lower levels. In 1999 and 2000 it fell precipitously on account of the economic crisis. (Capital formation surged in 1998,'the first year of the'crisis, largely because of reconstruction of infrastructure damaged by El Ninio rains.) Figure 2.4 shows that when per-capita real GDP leveled off in the early 1980s, the resource balance (net exports of goods and nonfactor services) Figure 2.2 Ecuadorj Per-capita Real GDP, Real Private Consumption, and Year-End Per-capita Public External Debt in 1999 U.S. Dollars and Prices, 1972-2000 1,400 ' 10 rPer-capita real GDP c=Per-capita real private consumption -Per-capita public external term debt (1999 U.S. dollars) Note: Data are reflated using the GDP deflator. Sources: International Monetary Fund, World Bank. 32 CRISIS AND DOLLARIZATION IN ECUADOR Figure 2.3 Ecuador: Gross Domestic Capital Formation (at 1975 Prices, 1975 GDP = 100), 1965-2000 1975 GDP = 100 30.0 25.0 20.0 15.0 10.0 5.00. 111I 5140 Il`m nl _1~ _`1 _~1 _1 _~1 111111 0 10 10 I VP 1965-2000 * Nongoverrnent * Govermnent Source: National accounts of Ecuador. shifted from deficit to surplus: Gross domestic saving (the difference between gross investment and the resource deficit) generally exceeded gross investment after the early 1980s-that is, Ecuador was carrying out external dissaving to limit growth of external liabilities. Chronic exchange-rate instability (see figure 2.5) helped set the price incentive for this dissaving: Beginning in 1982, the various policy pack- ages and mini-devaluations left the real-effective exchange rate at a con- sistently more depreciated value than it had maintained over the 1970s and 1980s. This part has summarized a complex history, but several patterns seem clear. The oil euphoria of the 1970s not only failed to lead to sustained economic growth, it also left Ecuador with an overwhelning external- debt "overhang." Moreover, by using oil proceeds to increase public expenditure and subsidization and reduce non-oil taxes, the government of the 1970s set in place revenue and expenditure structures that deep- ened the economy's vulnerability to shocks. During the 1980s and 1990s, the need to generate a net-export surplus reduced the resources available LONGER-TERM ORIGINS OF ECUADOR'S "PREDOLLARIZATION" CRISIS 33 Figure 2.4 Ecuador: Gross Fixed Capital Formation, Gross Domestic Saving, and Net Imports of Goods and Nonfactor Services (Percent of GDP), 1971-2000. Percent of GDP 35.0 30.0 Gross fixed capital formation 25.0 A\ zt 20.0 - X,... -- . 15.0 Gross domestic savin -'. ; --- '- 10.0 \ Ilmports less exports of goods and nonfactor services S.C 0.~~~~~~~~~~~~~~~~~~~~~~~' -5~~0' 0' - E - +\z/' - ;N,¢/, M -5.c -10.0 -15.0 1971-2000 Source: National accounts of Ecuador. Figure 2.5 Ecuador: Exchange Rate (Sucres per U.S. Dollar); Trade-Weighted Real-Effective Exchange Rate (+ = Depreciation; 1990 = 100), June 1970-September 2001 Real-effective exchange rate Nominal exchange rate (log) (+ = depreciation; 1990 100) 100,000 ____ _______ _______ __ 200 175 10,000 _- 150 1 ,000I 1 25 10 / , v <,/ ~~~~~~~~~~~~~~~~~~~~~75 _-_ s ~~~~~~~~~~~~~' \_ ~~~~~~~50 I = 25 Monthly from June 1970 Source: International Monetary Fund. 34 CRISIS AND DOLLARIZATION IN ECUADOR for capital formation. Moreover, to set the incentives to bring about the export surplus, policymakers had to sustain a real-effective exchange-rate depreciation. Doing so subjected the exchange rate to instability and uncertainty, encouraging the spontaneous dollarization that would help make the 1998 crisis so devastating. 3. Ecuador's Economic Structure Going into -the Predollarization Crisis As it went into the predollarization crisis in 1998, Ecuador still had a large pending structural-reform agenda. Oil dependence, the large size of the public sector, and heavy external debt made the public finances particu- larly vulnerable. But the most immediately dangerous structural problem turned out to be that, because the authorities had relied so heavily on exchange-rate depreciation to maintain the export surplus and external- debt surplus, the economy's spontaneous dollarization was advancing inexorably. Partial dollarization left the financial system singularly vul- nerable to the exchange-rate depreciation. This part considers the public sector, financial system, and other sectors in turn. A. Public-sector Structure and Vulnerability As policymakers struggled to cope with the debt-problem during the 1980s, it became increasingly clear that lthe public-sector structure they had inherited from the 1970s was an obstacle to growth. The elected gov- ernments that followed the military regime found it difficult to limit pub- lic employment, target subsidies, and taxation sufficiently and efficiently. Public-sector management-in particular, tax administration, budget planning, and day-to-day expenditure administration-remained out- moded. The key oil, electricity, and telecommunications sectors were inefficient public monopolies, and the government owned and managed smaller enterprises in many other sectors. The reliance of public revenue on volatile oil earnings rather than on more stable revenue sources, together with the expenditure commitments deriving from the public- debt burden, the essentially tenured public-sector labor force, and man- dated transfers (required in many instances by revenue earmarking), made the fiscal accounts inherently vulnerable to exogenous shocks. The elected 'governments of the 1980s and 1990s differed ideologically, but all were persuaded of the practical need for public-sector reform. The Cordero and Borja Governments made significant advances in financial liberalization and trade liberalization, but found substantive public-sector reform difficult to achieve. The Duran Ball6n Government made some- what more progress. Its 1992 Public Budgets Law set a legal basis for mod- LONGER-TERM ORIGINS OF ECUADOR'S "PREDOLLARIZATION" CRISIS 35 ernization of the systems of formulating and implementing public bud- gets. The Modernization of the State Law (1993) established a ministerial- level "Modernization of the State Council" (CONAM) to plan and help bring about modernization and privatization. The Durin Ballen Govern- ment succeeded in divesting most of the smaller enterprises that had come under public control, and submitted legislation to the Congress to permit reorganization and eventual privatization of the telecommunications and electricity monopolies. During its first two years in office the Duran Bal- len Government also made substantial progress in controlling public finances. It reduced noninterest public expenditure, in part through a pro- gram that reduced public-sector staffing by nearly 10 percent. This action, combined with forceful Central Bank policy management and implemen- tation of the pre-announced crawling-peg exchange rate, helped reduce annual inflation to around 20 to 30 percent from the 40 to 50 percent pre- vailing in the early 1990s. These were hard-won reforms, secured in the face of broad political opposition. Even so, they were short of the compre- hensive public-sector overhaul that Ecuador needed. The revenue base was still heavily dependent on oil; non-oil taxes were inefficient and widely evaded; and the public-sector payroll remained larger than Ecuador could afford (averaging 7.3 percent of GDP over the 1990s). In 1995, the Duran Ballen Government's structural-reform effort fal- tered on account of unanticipated events, including (a) a border conflict with Peru in January; (b) extended drought in the middle part of the year; and (c) the resignation of the vice president, who had been managing eco- nomic policy, under corruption allegations. Under pressure from interest groups, the Congress held up progress on reorganization of the electricity and telecommunications sectors. The failure of two relatively large banks in 1995 and 1996 deepened the government's difficulties. The Central Bank provided liquidity credit to keep the banks, open and took direct ownership of one.14 Real growth slowed in 1995. No further disburse- ments were provided by the IMF under its 1994 program after the first, and disbursements of the World Bamk and IDB structural-adjustment loans were postponed on account of the failure to meet the conditionality. Ecuador relapsed into political instability after the Durin Ballen Gov- ernment left office in mid-1996. The government elected that year, under President Abdala Bucaram, was forced from office by the Congress after only six months because growing alarm over corruption, the President's unusual personal style, and, finally, in January 1997, sharp increases in gasoline prices following relatively high exchange-rate depreciation led to widespread protests. The Congress then installed an 18-month interim government under President Fabian Alarc6n, an anti-Bucaram leader in the Congress. This government had inadequate political support for any- thing more than caretaking. It made vigorous efforts to persuade the Con- 36 CRISIS AND DOLLARIZATION IN ECUADOR gress to enact tax reform and to advance privatization, but these efforts were largely fruitless. To increase revenue it enacted a tariff surcharge, reversing the long process of trade liberalization. Moreover, in early 1998, with the conclusion of its term close, this government had to deal with the El Nifio rains and declining oil-export prices that turned out to be the onset of the predollarization crisis. The core of Ecuador's public-finance problem was (and indeed remains) that revenue depended too heavily on volatile oil earnings, while inadequate non-oil revenue and overwhelming debt-service and payroll commitments narrowed the scope for developmental expendi- ture. The incomplete public-sector reform agenda encompassed (a) reform of tax policy and administration, both to increase revenue and mitigate the effects of taxation for allocative efficiency; (b) implementa- tion of oil-revenue stabilization mechanisms; (c) limitation and targeting of public subsidies; (d) reduction and improved management of public- sector staff; steps to ensure the efficiency and quality of (e) education, health, and social-welfare expenditure and of (f) public capital formation and maintenance; (g) modernization of legal and technical systems of budget planning and execution; (h) completion of privatization processes and regulatory development in the telecommunications, electric power, and hydrocarbons sectors; (i) modernization of the social-security sys- tem; and (j) implementation of politically, administratively, and fiscally viable public-sector decentralization. Figure 2.6 shows how heavily Ecuador's public revenue has depended on crude-oil exports and domestic sales of oil derivatives. Revenue from these sources has been remarkably variable. In the four years 1996 to 1999, overall public-sector oil revenue, including domestic sales, was 8.2, 6.4, 4.6, and 7.5 percent of GDP, respectively. Revenue from oil exports alone amounted to 4.9,3.2, 1.3, and 5.3 percent of GDP, respectively, in the same years.15 Export volume and production grew slowly because of delayed and inadequate investment in the sector, but unstable world oil prices accounted for most of the variability (see figure 2.716). (Implemen- tation of a "stabilization fund," which would even out the revenue flow by accumulating funds when oil proceeds were relatively high and releasing resources when oil proceeds were relatively low under rigorous rules, would help reduce oil-export revenue volatility.17) Although the domestic non-oil taxation system generates inadequate revenue and is characterized by inefficiency and inequity, the difficulty of reaching agreement among the various parties in the Congress has impeded modernization. The reforms the Congress has approved since the crisis began have tended to be piecemeal, stopgap, and temporary measures. Beginning in 1999, the Congress and the Executive agreed to set an unusual 1 percent tax on financial transactions while suspending LONGER-TERM ORIGINS OF ECUADOR'S "PREDOLLARIZATION" CRISIS 37 Figure 2.6 Ecuador: Per-Capita Nonfinancial Public-Sector Revenue (in 1998 U.S. Dollars at 1998 Prices), 1990-2000 US dollars per capita at 1998 prices and exchange rate 500.0 -- - - -- 450.0 - _ 435°°°°0 - .:!.. l250.0 m ;|^ l- 1500,o - ::: . . g 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 * Public enterprise operating surplus * Oil-export revenue 3 Oil revenue from domestic sales B3 Non-oil external-source revenue *Non-oil domestic-source revenue Source: Central Bank of Ecuador. the personal and corporate income taxes, which had low yields and were difficult to administer. In April 1999, however, a realignment of political coalitions in the Congress led to restoration of the income tax. In Novem- ber 1999, the Congress reduced the transactions-tax rate to 0.8 percent, allowing it to be credited against income tax, and lifted the value added tax (VAT) rate from 10 to 12 percent (but rejected an increase to 15 per- cent). The 12 percent VAT rate was still below the rates in comparable Latin American economies (Peru and Chile, for example, have 17 percent rates). The Congress essentially refused to consider integral reform pro- posals. A proper integral reform would rebuild the tax system as a whole, setting mutually appropriate rate structures for (a) VAT, (b) personal and company income tax, (c) excises, and (d) import levies, and would also modernize the national law covering provincial and municipal revenues, helping relieve financial constraints at those levels. (Kopits and others, 1999, discuss the taxation issues comprehensively.) Ecuador maintains intricate systems of revenue sharing between the central administration, provincial and municipal governments, and sev- 38 CRISIS AND DOLLARIZATION IN ECUADOR Figure 2.7 Ecuador: Monthly Average Crude Oil-Export Price, June 1995-September 2001 30.00 . ~~~~~Dollarization announcement 25.00 20.00 15.00 A 10.00 5.00 Monthly from June 1995 Source: Central Bank of Ecuador. eral specific autonomous public entities. In February 1997, just after the Bucaram Government was forced from office, the Congress and the interim government rapidly approved a measure under which 15 percent of central-government revenue would have to be transferred to subna- tional governments (9 and 11 percent in 1997 and 1998, respectively). In the event, this legislation proved difficult to implement, partly because of the continuing resource scarcity as the economy slid into crisis, but also because subnational governments themselves had inadequate absorption capacity and limited abilities to take on new spending responsibilities. Education, health, and most public works remain almost entirely central- government responsibilities, and a large political and administrative effort would be required to decentralize them. Meanwhile, Ecuador maintains its long-standing practice of revenue earmarking, which has limited budget planners' scope to make discretionary expenditure choices. Tax evasion is widespread, partly because tax administration remains inadequately developed, but also because taxpayers have been "demoralized," perceiving that their taxes go to such "questionable" pur- poses as debt service, bureaucracy, and corruption. On the expenditure side, the public payroll has been difficult for poli- cymakers to control: Most public workers are unionized and effectively tenured. Pay levels are, as in many economies, politicized and con- LONGER-TERM ORIGINS OF ECUADOR'S "PREDOLLARIZATION" CRISIS 39 tentious. About 90 percent of the roughly 240,000 workers on the central government's payroll work in the national education system, health (the central government runs public hospitals and various health programs), the national police, and the armed services. Employment in education, health, and the police will have to increase as Ecuador grows and devel- ops (although this should be accompanied by a decline in the ratio of administrative to line staff). In addition to workers on the central gov- ernment's direct payroll, mandated central-government transfers largely go to other public-sector workers (such as at universities). One conse- quence of the central government's chronically tight finances has been that its capital budget has been tight]y constrained: Public expenditure on capital formation and maintenance has lagged, affecting real growth. Figure 2.8 shows the evolution of the central government's expendi- ture over the 1990s. Expenditure is divided here into five functional cate- gories: (1) education, health, and social services; (2) transport and communications; (3) agricultural development; (4) all other noninterest expenditure, and (5) interest on the public debt. While overall social expenditure declined in the crisis years of 1998 and 1999, the total inter- Figure 2.8 Ecuador: Per-Capita Central Government Expenditure (in 1998 U.S. Dollars at 1998 Prices), 1990-2000 U.S. dollars at 1998 prices and exchange rate 400.0 3350.00 r L ; 00 1 1H 250.0 150.U0 0 . 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 * Education and health 13Transport and communications [ Agricultural development *Other-noninterest expenditure S Interest due Source: Central Bank of Ecuador. 40 CRISIS AND DOLLARIZATION IN ECUADOR est bill increased sharply. In 1998 the nonfinancial public sector's accrued interest bill was 5.1 percent of GDP (of which external interest accounted for 3.8 percentage points). In 1999, accrued interest reached 10.1 percent of GDP, of which external interest accounted for 6.4 percentage points (again, largely because in 1999 recession and exchange-rate depreciation sharply reduced the GDP figure in this ratio's denominator). It is important to note in this context that Ecuador's social safety net is especially limited (see chapter 4). Apart from the national pension system (discussed below), its most important component is the Bono Solidario set up in September 1998. This program provides small monthly stipends to mothers of poor families and to poor retirees who are registered by the Catholic Church. Survey work implies that beneficiaries include many people who should be ineligible'8 while excluding many who would be eligible. The Emergency Social Investment Fund (Fondo de Inversion Social Emergente, FISE), created in 1993 with funding from various external sources, is used for local infrastructure projects in areas affected by emer- gencies, the idea in part being to employ people temporarily in order to provide them income. These programs apart, while there are various pro- grams in education, health, and nutrition intended to benefit poorer peo- ple, none can be said to fulfill the objectives of a social safety net. The rising public interest bill was driven by three developments after the mid-1990s. First, after running at relatively low levels after 1992, the public deficit widened. Second, net domestic financing came to figure more heavily than net external financing in overall public financing. And, third, since public borrowing was dollar-denominated, the real-effective exchange-rate depreciation increased the value of the interest bill. Domestic Treasury debt in bonds and bills rose from 1.2 percent of GDP at the end of 1993 to 7.1 percent at the end of 1998 (88 percent of which was dollar-denominated). This figure rose to 15.4 percent at the end of 1999 (largely on account of the sharp decline in the dollar value of GDP, resulting from recession and sharp real-effective exchange-rate deprecia- tion). In addition, after December 1998, the Treasury added an additional US$1.6 billion in dollar-denominated bonds to recapitalize commercial banks and pay deposit guarantees (see Part 4). (In November 1999, as the predollarization crisis turned acute, the authorities unilaterally termed out the dollar-denominated domestic debt falling due through December 2000 for seven years, with two years' grace, at London interbank offered rate (LIBOR) plus 2 percent interest.) Although the volatile performance of public finances is basically the consequence of the volatility of oil revenue and the large interest bill and payroll, the problem has additional dimensions. The Public Budgets Law and various reform attempts by the Council for the Modernization of the State notwithstanding, administration of the various aspects of LONGER-TERM ORIGINS OF ECUADOR'S "PREDOLLARIZATION" CRISIS 41 public finance has remained weaker than the country requires. Tax administration was lax until the mid-1990s, when the revenue service began to focus on "large taxpayers," an approach that has had some suc- cess in other countries. In 1998, however, the Alarc6n Government car- ried out a thorough reform of tax administration, establishing a new agency and replacing a large proportion of the staff. This new agency has proved more successful than the one it replaced, largely because the transactions tax that went into effect in January 1999 provided vital information relevant for collection of other taxes. Customs administra- tion gave rise to so many complaints of corruption that in the mid-1990s the authorities turned it over to a foreign company. The Bucaram Gov- ernment restored it to civilian control, but complaints (and evidence) of corruption revived, and in 1997, the Alarc6n government turned it tem- porarily over to military control. Figure 2.9 shows a striking characteristic of Ecuador's nonfinancial public-sector accounts. Despite the economy's unsatisfactory growth per- formance in the 1990s, the primary balance has been strongly in surplus since 1990, excepting only the initial predollarization crisis year of 1998. Because the interest bill has been so large, however, the overall balance has tended to be in deficit. Ecuador's large primary balance raises several issues. One is the question of whether, when analyzing the effect of the primary deficit on the domestic economy, it is more appropriate to focus on the primary balance excluding oil-export revenue. A primary surplus (deficit) presumably indicates the net resource flow the fisc draws from (pumps into) an economy to pay down financial obligations or accumu- late assets. In Ecuador's case, however, oil-export revenue flows do not pass through the domestic private economy. Another important issue is that the primary surplus is calculated for the consolidated nonfinancial public sector. This misses the point that operating surpluses in, say, state enterprises cannot be transferred directly to other public entities. (More- over, there is a fundamental conceptual difference between a government deficit, which is a change in the public net liability position, and a state- enterprise's operating loss.) Analysis of Ecuador's fiscal accounts requires separate consideration of the various public-sector compo- nents-the central government, public enterprises, subnational govern- ments, and social security. In recent years, the central government has been the main source of the overall public-sector deficit, and has been the main source of its variation. Regional and municipal governments ran a consolidated overall deficit of about -0.3 percent of GDP from 1996 through 1999 (although they managed slight surpluses in 2000 and 2001). Public enterprises ran a combined operating deficit of around 1 to 1.3 per- cent of GDP in 1996-99 (they managed surpluses of about 0.3 percent in 2000 and 2001). 42 CRISIS AND DOLLARIZATION IN ECUADOR Figure 2.9 Ecuador: Nonfinancial Public-Sector Overall and Primary Surplus (US$ Million at 1998 Prices and Exchange Rate), 1990-2000. US$ million at 1998 prices and exchange rate 150.0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 100.0 - | h °° I-r IF F -100.0 -150.0 1990-2000 * Primary surplus U Interest due E3 Overall surplus Source: Central Bank of Ecuador. Public finances have remained unstable partly because Ecuador's bud- get-management processes remain outdated. The processes by which annual budgets are formulated, considered by the Congress, codified into a payments calendar, adjusted over the course of budget exercise for unforeseen events, and finally executed have various shortcomings. These arise in part from the practical difficulty of planning properly i' an unsta- ble context; institutional complexities in the planning and implementation phases; and long-standing problems in the processing of information. Once dollarization brings about price stability, the need to alter the. budget in mid-year for unanticipated events should diminish. Since the mid-1990s, with Woild Bank support, the government has been developing and imple- menting a modern, computerized management -information system. The system was officially inaugurated for a core group of public entities in May 2000, and implementation has been proceeding since then. When this sys- tem is complete, policymakers will be in a far better position to plan and oversee public resource allocation. These changes will resolve only part of the problem of public-sector financial management, however: Apart from the "structural" problems of overreliance on oil revenue, the use of human LONGER-TERM ORIGINS OF ECULADOR'S "rREDOLLARIZATION" CRISIS 43 resources, and the external debt burden, Ecuador still needs to modernize its ways of determining public expenditure. Part 4 argues that the structure of the public finances magnified the consequences of the 1997 and 1998 shocks that set off the predollarization crisis. Reduced oil revenue and the need to increase public expenditure on account of El Nino, together with the inflexibility of public expenditure, combined to increase the public deficit. The authorities simply could not reprogram expenditure sufficiently in response to' their changed priorities. Three additional aspects of Ecuador's public-sector structural-reform agenda go beyond the government budget narrowly defined. These are (a) the need for pension reform, (b) the issue of political and administra- tive decentralization, and (c) the lagging privatization of publicly owned assets. Like many of South America's older national pension systems, Ecuador's pay-as-you-go social-security system has become financially unviable, and a fundamental reform, like those of Chile, Bolivia, Argentina, and Peru, is clearly necessary. Since the mid-1980s, in addition to contributions for its own staff, the central government has been pro- viding a subsidy to the IESS (the Instituto Ecuatoriano de Seguro Social, that is, the Ecuadoran Social Security Institute) covering 40 percent of pension payments due as well as certain specific pension deficits, including those of the national police and armed forces. The subsidy has been provided in cash, however, only to the extent the IESS actually needed it to meet its obligations. The balance has been capitalized into an interest-bearing loan from the IESS to the government, which amounted to about US$600 million by the end of 1999. At present, basic pension activity is restricted to the IESS, with no role for private. institutions. Various reform propos- als have been made, under which, in general, the central government would amortize the loan in cash over time, if and as the social-security system were placed on a viable financial basis and a role were created for the private sector. Many groups remain opposed to reform, however. Peasant groups in particular fear that reform could affect the Seguro Campesino, a program of pension and health benefits peasants receive but for which they make no direct contributions. Modernization of the social- security system is essential to ensure that no new pension-system bailout becomes necessary, that workers receive the pensions for which they con- tributed, and that appropriately regulated private-sector financial insti- tutions can participate in the system. In recent years, debate has revived in Ecuador on the potential for political and administrative decentralization. Participants in these debates refer often to positive and negative points of recent decentraliza- tion experiences in other countries. Proponents argue that, by applying the "subsidiarity" principle and devolving political and administrative 44 CRISIS AND DOLLARIZATION IN ECUADOR decisionmaking to more appropriate governmental levels, Ecuador could ease its regional rivalry (see Part 2) and make expenditure more respon- sive to local circumstances. Participants in these debates are well aware, however, of decentralization's pitfalls. Moreover, there is a wide range of views on precisely how it ought to be done. The fiscal aspects of decen- tralization are especially thorny. Ecuador's subnational governments are financially pressed, largely because existing legislation and outdated property values severely constrain their revenue-generation capacity and they have very limited financing capacities. As they elaborate their decentralization project, Ecuadorans will need to solve several problems simultaneously. Not only must they find a generally accepted political formula; they must ensure that expenditure responsibilities and financial resources of the various levels of governments are more or less balanced. (Brazil and Colombia offer striking examples of the dangers of decentral- izing without such matching.) The Duran Ballen Government successfully privatized many small enterprises (including several that had come under public control through insolvency), but privatization of the national telecommunica- tions, electricity, and hydrocarbon monopolies has been more difficult. Political and labor-union opposition was only part of the problem. The technical problem was that the enterprises themselves were not orga- nized on commercial bases, and so could not readily be transferred to pri- vate ownership. In the case of the electricity enterprise, establishment of new generating and transmission enterprises was delayed by the diffi- culty of allocating assets. Governments following the Duran Ballen Gov- ernment gradually succeeded in reorganizing the telecommunications and electricity sectors, formulating sectoral policies, and creating new regulatory agencies. Thus far, however, the enterprises remain unsold (an attempt in November 1997 to auction management contracts and minor- ity holdings in the telecommunications monopoly's two successor com- panies failed, and they are still in government hands). Moreover, rate setting remains subject to political pressure. The dollarization legislation approved in March 2000 included reforms increasing the shares of the telecommunications and electricity enterprises that private owners could hold. Thus far, the process of reorganizing and privatizing the state oil monopoly, PetroEcuador, has not yet been possible. To do so would require complex reorganization and legal reforms, and political and labor-union opposition remains powerful. B. Ecuador's Financial System and Partial Dollarization Ecuador's financial system proved especially vulnerable during 1998, and this was the main reason the crisis proved so devastating (see part 4 LONGER-TERM ORIGINS OF ECUADOR'S "PREDOLLARIZATION" CRISIS 45 below). Beginning in the mid-1980s, Ecuador's financial system had undergone liberalization, which changed the financial system from a "repressed" structure with directed-credit programs, high reserve require- ments, restricted foreign-exchange operations, and regulated interest rates to one in which private entities were essentially free to manage their affairs. The decade of liberalization culminated with the 1994 General Law of Financial Institutions. Directed-credit programs ended, reserve require- ments were reduced and rationalized, interest rates were freed, and banks were permitted to accept dollar deposits and provide dollar loans. Com- mercial banks were allowed to have offshore operations, on the reasoning that if Ecuadoran funds could not be prevented from going off shore, it should at least be possible to bring them home for credit operations. As it liberalized, Ecuador modernized its Central Bank. It phased out the Central Bank's export-subsidization programs and introduced mod- em monetary-management techniques, including more transparent pro- cedures for lending to commercial banks and liquidity management through repurchase operations and open-market operations in Central Bank liabilities. In 1992 a new Law of the Monetary Regime and State Bank substantially modernized the institution. This law prohibited Cen- tral Bank direct lending to the public sector and transferred various debt stocks (including the external debt remaining from sucretizaci6n) from the Central Bank to the Treasury. The institution's technical capacities and staff skills were significantly upgraded. Until 1998, however, the mone- tary authority retained an antiquated structure, under which the Central Bank executed policy set out by a "Monetary Board" whose members were the finance minister, the banking superintendent, representatives of commercial banks and the nonbank private sector, and a fifth member elected by the other four. That year, a constitutional reform abolished this institution, created a conventional presidency and board of directors, and made them completely independent once they were appointed by the president and confirmed by the Congress. Financial liberalization was not accompanied, however, by adequate development of banking supervision, for many of the same reasons why public administration generally had developed inadequately. Even after the onset of the 1998 banking crisis, many Ecuadorans perceived the exer- cise of banking supervision as political, not administrative. (There was a widespread view that if supervisors acted against Guayaquil banks, fair- ness required that they also act against Quito banks.) Banking supervi- sion was inadequate at various levels. Laws and norms covering such matters as connected lending, portfolio concentration, risk management, capital adequacy, accounting standards, documentation, income recogni- tion, and asset classification were uneven-in some instance outmoded, in some instances up-to-date. Enforcement was generally deficient, how- 46 CRISIS AND DOLLARIZATION IN ECUADOR ever, partly because supervisors' technical capacities were uneven, but also because bankers were often able to intimidate supervisory personnel administratively and legally. In any case, the main supervision problem was that the authorities had no formal means of intervening in banks short of liquidation. Banks maintained many long-standing risky practices. Connected lending and portfolio concentration were commonplace-indeed, many banks belonged to economic groups that used them to serve their own financing needs. These practices aggravated the risks of bank lending in a contingency-prone economy. Liberalization provided banks scope to engage in additional risky activities, including aggressive interest-rate competition, offshore banking, and U.S.-dollar operations. While interest- rate competition was a presumable objective of financial liberalization, absence of effective supervision meant banks could undertake riskier operations than they could safely manage. Managers of more conserva- tive banks found themselves having to engage in risky activities in response to competition. Offshore banking turned out to be a source of instability. Banks ran their offshore funding operations pretty much as if they were onshore, taking deposits and doing other business in branches within Ecuador. Although the Banking Superintendency nominally regulated these off- shore operations, it was unable in fact to work effectively outside Ecuador. After the crisis began in 1998, the authorities' inadequate knowledge of the offshore banks' situation complicated their ability to deal with it. This is why in December 1998, for example, the authorities had little choice but to extend the same guarantee to offshore deposits that they provided to onshore deposits. The March 1999 deposit freeze applied to the off-shore banks, but banking authorities in some places- in particular, the United States-did not recognize it.19 Since the onset of the crisis, the authorities have concluded that, however logical the argu- ment for allowing offshore operations may once have seemed, their inability to supervise such operations left the authorities little choice but to conclude them (the March 2000 dollarization legislation provided for a gradual phase-out). Important as the financial system's operational and supervisory inad- equacies were, the evolution of the predollarization crisis as described in Part 4 shows that the banking system's partial dollarization (see table 2.2) was probably the main reason the predollarization crisis evolved as it did. Strictly speaking, partial dollarization did not cause the crisis, but it did intensify the destabilizing effects of exchange-rate depreciation, mak- ing the crisis far harder to manage than it would otherwise have been. Partial dollarization meant that the economy was operating internally with two different units of account, subject to an unstable-that is, LONGER-TERM ORIGINS OF ECUADOR'S "PREDOLLARIZATION" CRISIS 47 volatile and uncertain-exchange rate. Exchange-rate depreciation not only increased the' sucre equivalent of the dollar component of the money supply, it also drove private firms and individuals with open, exposed positions into insolvency. Although commercial banks tried hard to main- tain matched positions on their own balance sheets,20 they apparently took less care to ensure that their borrowers had matching positions. The reason "spontaneous" dollarization became so pervasive was that long experience of inflation and exchange-rate depreciation made the value of the national monetary uncertain, encouraging people to choose a hard currency unit to denominate their assets. In December 1996, 24 percent of all onshore bank demand, savings, and time deposits were in dollars rather than sucres; in December 1998, this percentage had risen to 41, and in March 2000 it stood at 63 (see figure 2.10). Over this period, the commercial banks' overall deposit base declined about 30 percent in dol- lar terms, with sucre deposits falling by more than two-thirds while total dollar deposits grew. Offshore deposits were entirely in dollars. In addi- tion, a growing quantity of dollar currency circulated within Ecuador and came into increasing use for transactions. Banks' loan portfolios underwent a corresponding evolution. In December 1994, 33 percent of all bank loans were in dollars; by Decem- ber 1998, this percentage had risen to 60, and in March 2000 it stood at 91 (see figure 2.11). Moreover, after December 1998, the proportion of dollar- denominated loans classified as nonperforming rose sharply, in contrast to 'sucre loans. As the crisis progressed, banks collected sucre loans and gave far fewer; at the same time, their dollar lending stocks stabilized and turned increasingly nonperforming on account of real-effective deprecia- Table 2.2 Ecuador: Dollarization Indicators Year-end percentage in U.S. dollars of. Year Quasi'money Deposits Loan portfolio 1989 9.7 14.7 1.9 1990 7.4 13.3 1.5 1991 7.5 14.5 3.0 1992 10.8 20.0 6.8 1993 12.6 16.9 13.4 1994 15.7 15.6 20.3 1995 24.3 19.2 28.3 1996 28.0 22.3 32.6 1997 36.9 23.6 45.1 1998 43.9 36.9 60.4 1999 47.4 53.7 66.5 Source: Central Bank of Ecuador. 48 CRISIS AND DOLLARIZATION IN ECUADOR Figure 2.10 Ecuador: Onshore Commercial-Bank Deposits (US$ Million) US$ million 6,000 5,00 2,0 100 N N N N N. N1.111111.11'1111.1.1111.1.1 December 1996-February 2000 E U S. - dollar deposits * Sucre deposits Source: Central Bank of Ecuador. tion, as borrowers without dollar income found it harder to meet their obligations. As always, exchange-rate depreciation tended generally to increase the price level, if only by raising tradable-goods prices. With the money sup- ply dollarized, depreciation directly increased its local-currency value, and so increased the pressure on the price level more than it otherwise would. Moreover, since the inflationary aftermath of any exchange-rate depreciation eroded at least some of its (real-effective) effect, deeper depreciation was required in the partially dollarized system to achieve any given external-accounts objective. On this reasoning, all other things being equal, the farther dollarization proceeded in the bank-deposit base, the larger the inflationary consequences of any exchange-rate deprecia- tion were likely to be. Moreover, once the exchange rate and price level began rising, they affected the state of expectations and uncertainty regarding their future levels. This is true in any inflationary economy, but in an economy with two currency units the consequences were bound to be more complex and unstable. As the account in Part 4 indicates, once the exchange rate began depreciating sharply, commercial banks became acutely illiquid, first because debt service from dollar-denominated loans LONGER-TERM ORIGINS OF ECUADOR'S PREDOLLARIZATION" CRISIS 49 Figure 2.11 Ecuador: Onshore Commercial-Bank Loans Performing Normally and in Arrears US$ million 6,000 5,000 4,000 3,000 2,000 1,000 4000 . . . . . ls. 'C N N N N 'C 'c X XO o~ o~ o~ Ns Ns 5N oh a, oE o0 o0 0 o0 o a, December 1996 -February 2000 * U.S.- dollar credit performing normally El U.S.-dollar credit in arrears * Sucre credit performing normally * Sucre credit in arrears Source: Central Bank of Ecuador. diminished, and second because depositors, knowing or fearing that the banks' condition had deteriorated, began withdrawing. In summary, the structural weaknesses of Ecuador's banking system meant that the system would play a central role in the evolution of the crisis. As a consequence of uneven structural reform, going into the crisis Ecuador had a liberalized, but inadequately supervised, banking system working in two units of account. IPartial dollarization was, in hindsight, the crucial problem. Better supervision might have enabled the authori- ties to manage it better and faster, but, as Part 4 argues, a partially dol- larized financial system would inevitably prove too vulnerable to sharp exchange-rate depreciation. C. Labor-market and Foreign-trade Reform Going into the crisis in 1998, the most important aspects of Ecuador's incomplete structural-reform agenda were those having to do with the public and financial sectors. Nevertheless, important structural-adjust- ment agendas remained in several other aspects of Ecuador's economy. 50 CRISIS AND DOLLARIZATION IN ECUADOR Two subjects worth mentioning in this connection are (a) formal labor markets and (b) the trade regime. FORMAL LABOR MARKETS. Ecuador's formal labor markets were (and are still) highly regulated. Job-tenure and severance-payment rules were especially so, contributing to the large size of- the informal sector. In addition, until full dollarization commenced in 2000, Ecuador main- tained an unusual formal-sector wage regime. Wages came to be set every six months by special sectoral commissions, with labor-union, employer, and government representatives. A central commission sets the minimum wage, which guides many of the sectoral wage commis- sions. At the outset of the crisis, formal-sector wages comprised a large number of components, including thirteenth, fourteenth, fifteenth, and sixteenth wage payments made in different months of the year, cost-of- living adjustments, and a remarkably extensive array of other allowances and benefits. The dollarization legislation (see chapter 3) included provisions introducing gradual change. One would gradually "unify" all these wage components in both the private and public sec- tors; the other introduces temporary hourly employment contracts. Full unification, further "flexibilization," and gradual extension of "formal- ity" to the labor market as a whole are essential to making labor markets work efficiently. THE TRADE REGIME. During the 1950s, 1960s and 1970s, llke many devel- oping economies, Ecuador constructed a restrictive trade regime. In the late 1980s and early 1990s, however, it liberalized and modernized it to a considerable degree. Formally, Ecuador's external tariff is the Common External Tariff of the Andean Group, as liberalized in the late 1980s, although its basic tariff rate is somewhat lower than the Common Exter- nal Tariff. During 1998 this afforded Ecuador some scope to set tariff sur- charges ranging between 2 and 10 percent-mainly as a revenue measure, but partly (as their differentiation reveals) as a protection mea- sure. These surcharges remained in place until sometime after dollariza- tion commenced. The present tariff system has an anachronistic structure favoring raw materials over final products. Raw and intermediate goods pay tariffs ranging between 5 and 15 percent, capital goods pay 15 per- cent, and consumer goods pay tariffs of 20 percent (new automobiles remain subject to a 35 percent rate). Although Ecuador eliminated most nontariff and restrictive barriers, several remain. Some agricultural imports are subject to levies based on reference prices, and certain imports (including used automobiles) are banned. Although the trade regime is not at present the most pressing structural-reform priority, com- pletion of the reform process-in particular, movement toward a uniform LONGER-TERM ORIGINS OF ECUADOR'S "'REDOLLARIZATION" CRISIS 51 external tariff--would encourage development of industrialized exports, and to this extent reduce the economy's reLiance on commodity exports. . Ecuador's Predollarization Crisis A. The Onset of the Crisis, 1998 Ecuador's predollarization crisis commenced at the end of 1997, about halfway through the interim government (February 1997 to August 1998) of Fabian Alarc6n, appointed by the Congress after it deposed Abdala Bucaram's government. The shocks that produced the crisis included (a) declining crude-oil export prices (see figure 2.7 above); (b) severe damage from the 1997-98 El Nifio rains, affecting coastal populations, agriculture, and infrastructure;21 and then, during 1998, (c) the effects of the financial crises involving East Asia, the Russian Federation, and Brazil. These last included recession in export. markets (Ecuador temporarily lost a new Russian flower-export market, for example), intensified competition from economies with depreciated exchange rates, and the retraction of finan- cial flows to developing economies generally. The pubLic dLeficit surged in 1998 as a consequence of the shocks. Pub- lic-sector oil-export revenue fell to 1.3 percent of GDP in 1998 from 4.9 and 3.2 percent in 1996 and 1997, respectively. The El Nifno damage reduced tax revenue and forced heavy emergency and reconstruction expenditure. The crisis stiffened political resistance to tax increases. Even before the severity of the crisis became clear, it was generally recognized that the tax system needed to be modernized to increase yields and improve efficiency. In early 1998, however, Congress considered but rejected several integral tax-reform proposals. Meanwhile, as explained in Part 3, section A above, the interim government's policymakers had lit- tle scope to limit public expenditure. An additional problem was that the government removed from office in February 1997 had stopped the auto- matic adjustment system for motor-fuel prices in effect since 1994. Elec- tricity rates ancl the price of cooking gas were tending to lag behind their production costs as well, effectively creating large subsidy flows. As a consequence, the government found itself in a tight cash squeeze. Presi- dential elections held in mid-1998 made it politically harder to tighten fis- cal poLicy. The 1998 nonfinancial public deficit reached 5.6 percent of GDP,22 compared with 2.9 and 2.5 percent in 1996 and 1997, respectively (see figure 2.12). As oil-export earnings decLined and the trade accounts deteriorated, importers concluded that the authorities would be forced to end the pre- announced crawLing-peg exchange-rate policy in effect since 1993, and began advancing orders for inventory In late March and mid-September 52 CRISIS AND DOLLARIZATION IN ECUADOR Figure 2.12 Ecuador: Indicators of Macroeconomic Imbalance, 1988-2000 Percent of GDP (deficits) Annual percentage rate (inflation) 12.0 100.0 10.0 90.0 8.0 80.0 6.0 70.0 4.0 60.0 2.0 50.0 0.01 140iL i 1990 '1991 1992 '1993 19 95W619 9 9 4. -2.0 30.0 -4.0 .20.0 -6.0 10.0 -8.0 0.0 _Current-account deficit/GDP ~ Nonfinancial public deficit/GDP - Growth of consumer prices (December/December) Source: International Monetary Fund, World Bank. 1998 the authorities carried out devaluations beyond the pre-announced crawling-peg band. Despite these devaluations, the 1998 current-account deficit surged to 11 percent of GDP, compared with 3.6 percent in 1997 (see figure 2.12). A large proportion of the financing came through private capital transfers, including withdrawals from offshore deposit holdings, as well as some international-reserve loss. The shocks affected commercial banks particularly seriously. They had lent heavily in affected sectors, including coastal agriculture, oil-sector services, and exports generally, and their nonperforming loans began ris- ing. In addition, the emerging-markets crises and the country's worsen- ing prospects persuaded foreign banks to retract credit lines to Ecuadoran banks, intensifying liquidity pressure and forcing banks to reduce credit to companies that depended on these lines for working cap- ital. Given the developments on the external and fiscal accounts, how- ever, the Central Bank took the view that it had to tighten credit and raise interest rates to limit exchange-rate depreciation and inflation. During the first half of 1998 several commercial banks underwent episodes of illiquidity and heavy deposit withdrawals. LONGER-TERM ORIGINS OF ECUADOR'S "PREDOLLARIZATION" CRISIS 53 In August 1998, following a two-round election in May and July, a new government heacled by President Jamil Mahuad began what was to have been a five-year term, under a constitution revised by a constitutional assembly earlier in the year. The revisions were an ambitious attempt to overcome Ecuador's long-standing governance problems. They included measures to strengthen the president's authority, limit the Congress' powers to increase taxes and public expenditure, make it harder to impeach cabinet members and judges on purely political grounds, set minimum electoral support requirements for political parties to be repre- sented in the Congress, make the Central Bank fully independent, elimi- nate mid-term congressional elections (which tended to weaken the president), decentralize fiscal responsibilities, and make the judiciary more independent. Unfortunately, however, the predollarization crisis commenced even before the new constitution went into effect in August 1998. As this part shows, the crisis subjected the new constitution to severe stress: Much of what happened over the next two years violated the spirit and letter of the constitution-not least, of course, the Mahuad government's forced departure and the move to dollarization (see Arteta and Hurtado 2002). Nevertheless, the 1998 constitution remains in effect, and, now that d[ollarization has brought about more normal circum- stances, it is fair to hope that it will not be subjected to such severe stress. A month after taking office in August 1998, the, Mahuad government carried out an essential reform of the tong-standing cooking-gas and elec- tricity subsidies, raising the prices and introducing a new direct cash- transfer mechanism (called the Bono Solidario) targeted to poorer households. These subsidies had been expensive, amounting at some times-depending on the current exchange rate and prices-to several percentage points of GDP. The Mahuad Government also began consid- ering ambitious proposals to advance longer-term structural reform. Its attention was focused, however, on negotiations with Peru to settle a long-standing border dispute (which led to an historic peace accord in November 1998). Over the latter part of 1998, however, the banking crisis deepened. An important bank failed in August 1998, and virtually all banks experienced intensified portfolio and liquidity problems. A Sep- tember 1998 World Bank mission, noting that the authorities had no way to intervene in problem banks short of the traumatic procedure of liqui- dation, recommended that the authorities institute a universal deposit guarantee and establish an agency capable of taking over and restructur- ing banks in crisis. In December 1998, the president approved emergency legislation extending a virtually unlimited Treasury guarantee to all deposits, even to the trade-credit lines owed to foreign banks. This legis- lation establishecd Ecuador's first Deposit Insurance Agency and estab- lished modalities through which the authorities could intervene in 54 CRISIS AND DOLLARIZATION IN ECUADOR troubled banks. The Congress had taken a month to debate 'the.'emer- gency legislation, partly because legislators from coastal.areas feared that their region's banks would be unfairly affected, and partly because it was combined with the unusual tax reform that substituted a-financial-trans- actions tax while suspending the income tax (see below) The guarantee raised the stakes for the authorities and- inevifably. created doubts. It was credible as such only for deposits in smaller institutions: Itwas' clear-that if large banks failed, the authorities would either have to keep them open or honor the guarantee in some sense less than.fully. While the guarantee may have limited deposit withdrawals for a time, in retrospect its inher- ent lack of credibility undermined its effectiveness. On December 1, 1998, the day it began operating, the Deposit :Insur- ance Agency took over Ecuador's largest bank (Guayaquil-based Filan- banco), which had been foundering for several months. In.a pattern to be followed during 1999 for-larger failing banks, the authorities kept it open; recapitalizing it with special 10-year dollar-denonAinated Treasury bonds paying 12 percent annual interest. The bank used.these bonds as collat- eral for Central Bank liquidity loans to help meet withdrawal demand. In the first weeks of 1999 several'smaller banks failed and eventually liqui. dated. The Treasury paid their- deposits (after. several months' delay) through the Deposit Insurance Agency. The same legislation also replaced the poorly performing personal and company income taxes with a 1 percent tax on all financial transactions, including checks. The government needed to raise revenue, but the trans- actions levy was the only tax measure for which'a legislative majority could be formed. The president's Democracia Popular Party, although the largest in the Congress, had far less than a majority of the seats in the Congress and relied for support on other parties. The next largest party, the Partido Social'Cristiano, opposed tax reform generally but was will- ing to substitute the transactions tax for the income tax. Although it proved an effective revenue source during 1999 and 2000, the transac- tions tax encouraged financial disintermediation and set an added incen- tive for deposit withdrawals at a moment when the banks were already in severe crisis. B. The Deepening Crisis, 1999 In February.1999 the Central Bank floated the exchange rate to limit inter- national-reserve loss.23 Because the 1999 budget incorporated an exchange-rate assumption that the float would have made implausible, the Central Bank delayed the float, at a substantial cost in reserve loss, until the moment the Congress approved the budget. Over the following four weeks the exchange rate lost 30 percent of its U.S.-dollar value. Bank LONGER-TERM ORIGINS OF ECUADOR'S "PREDOLLARIZATION" CRISIS 55 loan portfolios deteriorated accordingly. Consumer prices were 13.5 per- cent higher in March than in February (see figure 2.13), and there were intensifying fears of hyperinflation. Several large banks were now in acute danger, notably the large Guayaquil-based bank (Banco del Pro- greso) that had operated with a high-interest-rate policy and a heavily dollarized loan portfolio.24 Ecuador's regional politics were deeply involved here: The bank's principal shareholder claimed that authorities were seeking to close the bank in order to damage coastal interests, and for a time secured broad regional political support on this basis. In mid-March in an attempt to limit inflation pressure and prevent further bank failures, President Mahuad first announced a bank holiday and then, after several days, a deposit freeze: All checking and savings deposits were frozen for one year and time deposits were frozen for one year from the original maturity date.25 He also announced that interna- tional firms would be contracted to audit the banks to determine their true capital adequacy. In response to the freeze, the Central Bank's presi- dent and several board members tendered their resignations. The freeze temporarily reversed the exchange-rate depreciation and slowed the inflation-by mid-April the exchange rate had appreciated to nearly its prefloat value. Inevitably, however, it severely damaged depositor confi- dence. Banking-system credit operations, already shrinking, nearly Figure 2.13 Ecuador: Consumer Prices, 1995-2000 Percent per month 16 Dollarization announcement l 14 12 10I 8 A 2 1l V^4 1\2 % O~~~~ v Dc95 Jun96 Dec96 Jun97 Dec97 Jun98 Dec98 Jun99 Cec99 un 00 Dec 00Jun01 Dec 01 -2 December 199SaDerember 2001 Source: Internationa.l Monetary Fund. 56 CRISIS AND DOLLARIZATION IN ECUADOR ceased, and this largely explains why the economy slid into such a deep recession in 1999. This recession aggravated Ecuador's deepening poverty (see chapter 4) and worsened most social indicators. Ecuador had gone into the crisis with some of Latin America's most unfavorable indicators of poverty incidence and income inequality. In 1998, at the outset of the crisis, 46 per- cent of the population was poor, compared with 34 percent in 1994. Dur- ing the same years, extreme poverty (insufficient income for a minimum food basket) had worsened from 15 to 17 percent. In 1998 69 percent of the rural population was impoverished, compared with 56 percent in 1994. Worsening inequality accompanied this trend of deepening poverty: The overall income Gini ratio worsened from 0.54 in 1994 to 0.58 in 1998. During 1999 poverty worsened even further. Health and nutri- tion standards plunged. Another indication of the social devastation the crisis caused is the measured urban unemployment rate, which roughly doubled between June 1998 and June 1999 and remained at a high level into 2000. Family structures throughout the society came under intensi- fied pressure, and by the middle of 1999 this in itself had become a crisis of overwhelming proportions (see chapter 5). Ecuador's public sector lacked the institutional means, let alone the resources, to cope directly with its population's immiserization. Apart from the reviving energy subsidies (resulting from lagging adjustment of the relevant prices), the Bono Solidario introduced in September 1998 constituted the nation's social safety net, and its real value was deteriorating rapidly as prices surged. Ecuador had essentially no emergency employment program, no emergency nutrition programs for vulnerable children, and no means to encourage students from impoverished families to remain in school. On July 30, 1999, on the basis of the bank audit results, the Banking Superintendent affirmed that 19 of the 32 banks examined (including 3 closed earlier) were sound, but closed 6 banks (including the Banco del Progreso) and took 4 relatively large banks into enhanced monitoring, recapitalization, and restructuring programs under the Deposit Insurance Agency. Three of these four banks failed within two months. The author- ities nevertheless kept these banks in operation, merging them with other banks previously taken over. In all, by the end of September 1999, com- mercial banks accounting for roughly 60 to 70 percent of total banking assets were under public stewardship. Within weeks of the freeze, the authorities began accelerating the unfreezing schedules for checking and savings deposits, hoping in this way to restore normal banking operations more rapidly. They also set up a scheme under which time deposits could be converted into marketable "Reprogrammable Certificates of Deposit," which could be negotiated, held, or used to service bank loans at par. Unfreezing led to deposit with- LONGER-TERM ORIGINS OF ECUADOR'S "PREDOLLARIZATION" CRISIS 57 drawals, however. Between April and December, some US$465 million (about 3.1 percent of 1999 GDP and about 16 percent of the end-June broad money supply) in checking and savings deposits (in both on- and offshore offices of the banks) was unfrozen. About one-third was with- drawn from the banking system. These withdrawals contributed to capi- tal flight and renewed pressure on the exchange rate. In November 1999, the ConstitutionThis document did not complete OCR process. <==**"