0804752788 Crisis of Demo

THE CRISIS OF DEMOCRATIC REPRESENTATION IN THE ANDES

THE CRISIS OFDEMOCRATIC

REPRESENTATION IN THE ANDES

❖

Edited by

Scott Mainwaring, Ana María Bejarano,and Eduardo Pizarro Leongómez

❖

Stanford University PressStanford, California 2006

Stanford University PressStanford, California

© 2006 by the Board of Trustees of the Leland Stanford Junior University. All rights reserved.

No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying

and recording, or in any information storage or retrieval system without the prior written permission of Stanford University Press.

Printed in the United States of America onacid-free, archival-quality paper

Library of Congress Cataloging-in-Publication Data

The crisis of democratic representation in the Andes / edited by Scott Mainwaring, Ana María Bejarano, and

Eduardo Pizarro Leongómez.p. cm.

Includes bibliographical references and index.ISBN-13: 978-0-8047-5278-7 (cloth : alk. paper)

ISBN-10: 0-8047-5278-8 (cloth : alk. paper)1. Representative government and representation—Andes Region.

2. Democracy—Andes Region. 3. Political culture—Andes Region. 4. Andes Region—Politics and government. I. Mainwaring, Scott, 1954-

II. Bejarano, Ana María. III. Pizarro Leongómez, Eduardo.JL1866.C76 2006

320.98—dc22 2006005159

Typeset by G&S Typesetters, Inc. in 10/12 Bembo

List of Tables and Figures viiAcknowledgments xi

List of Contributors xiii

1. The Crisis of Democratic Representation in the Andes: An Overview 1

Scott Mainwaring, Ana María Bejarano, and Eduardo Pizarro Leongómez

PART I. PARTY SYSTEMS, POLITICAL OUTSIDERS, AND THE CRISIS OF DEMOCRATIC REPRESENTATION

2. From Crisis to Collapse of the Party Systems and Dilemmas of Democratic Representation: Peru and Venezuela 47

Martín Tanaka

3. Giants with Feet of Clay: Political Parties in Colombia 78Eduardo Pizarro Leongómez

4. Ecuador: The Provincialization of Representation 100Simón Pachano

5. Outsiders and Neopopulism: The Road to Plebiscitary Democracy 132

René Antonio Mayorga

PART II. DECENTRALIZATION, LEGISLATURES, AND DEMOCRATIC REPRESENTATION

6. Decentralized Politics and Political Outcomes in the Andes 171Kathleen O’Neill

7. The Nature of Representation in Andean Legislatures and Attempts at Institutional Reengineering 204

Brian F. Crisp

CONTENTS

PART III. POPULAR POLITICS AND THE CRISIS OF DEMOCRATIC REPRESENTATION

8. Urban Citizen Movements and Disempowerment in Peru and Venezuela 227

Daniel H. Levine and Catalina Romero

9. Indigenous Politics in the Andes: Changing Patterns of Recognition, Reform, and Representation 257

Deborah J. Yashar

PART IV. CONCLUSION

10. State Deficiencies, Party Competition, and Confidence in Democratic Representation in the Andes 295

Scott Mainwaring

Index 347

vi Contents

Tables

Table 1.1. Support for Democracy, Latin America, 2005 9Table 1.2. GDP per Capita in the Andes, 1960 –2002 10Table 1.3. Citizen Trust in Representative Institutions, Andean

Countries 17Table 1.4. Confidence in Parties and Parliament, Select Countries,

World Values Survey 18Table 1.5. Electoral Volatility and Share of Vote for New Parties in

Lower-Chamber Elections, Andean Countries 19Table 1.6. Average Share of Vote Won by Outsider Presidential

Candidates in Five Most Recent Presidential Elections, Andean Countries 22

Table 1.7. Programmatic Representation in Latin America and Spain 26Table 2.1. Peru: Vote Percentages for the Major Political Parties,

1978–2000 49Table 2.2. Venezuela: Vote Percentages for Presidential Elections,

1973–1998 50Table 2.3. Peru and Venezuela: Inflation, GNP, and Subversive Acts,

1980 –1999 51Table 2.4. Venezuela: Trends in Party Identification 52Table 2.5. Venezuela: Percentage of Seats in the Chamber of Deputies,

1973–2000 60Table 2.6. Venezuela: Number of Governors Elected, by Party,

1989–2000 62Table 3.1. Number of Lists that Registered for Senate and Lower-

Chamber Elections, 1958–2002 84

TABLES AND FIGURES

Table 3.2. Electoral Performance of Lists That Elected More Than OneSenator, 1991–2002 86

Table 3.3. Senate Seats Won by Political Parties and Movements,1991–2002 88

Table 4.1. Ecuador: Main Political-Electoral Reforms, 1983–2003 105Table 4.2. Share of Congressional Vote Won by Four Major Parties,

1979–2002 106Table 4.3. Share of National Electorate and Number of Deputies per Prov-

ince, 2002 107Table 4.4. Share of Presidential Vote in First Round, 1978–2002 117Table 4.5. Size of Legislative Delegations, 1979–2002 118Table 4.6. Electoral Strongholds of the Main Political Parties,

1979–2002 122Table 4.7. Territorial Distribution Index (TDI) of Main Parties,

1979–2002 124Table 4.8. Regional Distribution of Origin of Deputies, by Party,

1979–2002 125Table 5.1. Bolivia: Party Votes and Seats, Lower-Chamber Elections, June

30, 2002 156Table 6.1. Expenditure Decentralization in Latin America 174Table 6.2. Popular Election of Subnational Executives 175Table 7.1. Pre-Reform Intraparty Characteristics 208Table 7.2. Partisan Composition of Andean Legislatures Prior to Constitu-

tional Reform 210Table 7.3. Partisan Composition of Andean Legislatures after Constitu-

tional Reform 214Table 7.4. Bill Targets in the Colombian Senate before and after Electoral

Reform 217Table 7.5. Public Confidence in Congress, 1996 219Table 10.1. State Performance and Perceptions Thereof in the Andes,

1996 –2005 299Table 10.2. Pearson Correlation Coefficients between State Performance

and Confidence in Parties and Parliaments, Country-LevelIndicators 300

Table 10.3. Determinants of Confidence in Political Parties in the Andes 302

Table 10.4. The Impact of Assessment of the National Economic Situationand Interpersonal Trust on Confidence in Parties 303

Table 10.5. Determinants of Confidence in Congress in the Andes 304Table 10.6. Attitudes about Democracy and Representation in the

Andes 307

viii Tables and Figures

Table 10.7. Citizen Confidence in Institutions in the Andes 309Table 10.8. Secondary Education and Urbanization in the Andes 315

Figures

Figure 1.1. Presidential turnout as a percentage of the eligible electorate inthe Andes 24

Figure 1.2. Turnout as a percentage of the eligible electorate in lower-chamber elections in the Andes 25

Figure 6.1. Turnout in millions of voters, by level 181Figure 6.2. Percent of the vote won by traditional parties, by level 187Figure 7.1. Interparty and intraparty incentives of legislators in the

pre-reform Andean countries 211Figure 7.2. Public satisfaction with democracy 220Figure 10.1. Votes cast in presidential elections as a percentage of total popu-

lation, Andean countries, 1950s to 2004 314Figure 10.2. Effective number of parties, lower chambers in the Andean

countries 318Figure 10.3. Effective number of parties in the Senate: Peru, Venezuela,

Bolivia, and Colombia 319

Tables and Figures ix

Our first and foremost gratitude is to three institutions without which thisbook would not have come to fruition: the Kellogg Institute for InternationalStudies at the University of Notre Dame, the Ford Foundation, and the Coca-Cola Company. The distant background to the book is that in the fall of 1998Colombia’s foremost human rights lawyer, Gustavo Gallón of the ColombiaCommission of Jurists, came to spend what was originally intended to be a se-mester at the Kellogg Institute as a Visiting Fellow. The deteriorating humanrights situation in Colombia made it unwise for Gallón to return to Colombiaas planned and instead he stayed at Notre Dame for an additional semester. Dur-ing his year at Notre Dame, the Colombian Commission of Jurists, the KelloggInstitute, the Kroc Institute for International Peace Studies at Notre Dame, andthe Center for Civil and Human Rights, also at Notre Dame, hatched a collab-orative project on “Democracy, Human Rights, and Peace in Colombia.” TheFord Foundation’s Santiago office generously supported this project (Grant1000-0727). This project enabled Colombian scholars at risk to spend time inresidence at the Kellogg Institute, thus allowing them to work productively andsafely at a time when they could not live safely in Colombia.

Through this program, Ana María Bejarano, Eduardo Pizarro Leongómez, and four other distinguished Colombian scholars came to the Kellogg Insti-tute for one-year periods between 2000 and 2003. The three editors of this volume became friends. We spent many hours discussing the Colombian situa-tion and the deterioration of democracy in other Andean countries—issues ofdeep normative concern and intellectual interest to all three of us. Along withMiriam Kornblith, René Antonio Mayorga, Simón Pachano, and Martín Tanaka,we conceived a broader project on the Andean crisis. The Ford Foundationonce again provided support for this project (Grant 980-0350-3). Its support enabled Kellogg, CEBEM (Centro Boliviano de Estudios Multidisciplinarios),

ACKNOWLEDGMENTS

xii Acknowledgments

FLACSO-Ecuador, IESA (Instituto de Estudios Superiores de Administración,Caracas), the Instituto de Estudios Peruanos (IEP), and the Universidad Los An-des (Bogotá) to have some scholarly exchanges within the Andean region and tosponsor a scholarly workshop in Quito in July 2003.

A generous gift from the Coca-Cola Company to the Kellogg Institute hasfunded a series of conferences on some of the outstanding problems confrontingLatin America since the 1990s, and it enabled us to host an eponymous confer-ence at the Kellogg Institute in May 2002. We are grateful for this support. Thestaff of the Kellogg Institute spent long hours arranging the travel of participants,designing posters, and hosting the participants. At the conference, we benefitedfrom the stimulating comments of Michael Coppedge, Humberto de la Calle,Paul Drake, Myles Frechette, Luis Gallegos-Chiriboga, Frances Hagopian,Miriam Kornblith, Curtis Kamman, Ricardo Luna, Alejandro Reyes, and SamuelValenzuela.

Since its inception in 1982, the Kellogg Institute has promoted research ondemocracy and other important normative issues that confront humanity. Webenefited from the rich intellectual debate at the Kellogg Institute on this theme.During the years of gestation of this project, among the contributors to this vol-ume, Ana María Bejarano, René Mayorga, Simón Pachano, Eduardo PizarroLeongómez, Catalina Romero, and Martín Tanaka were Visiting Fellows at theKellogg Institute. Brian Crisp, Daniel Levine, and Deborah Yashar, who alsocontributed to this volume, were Visiting Fellows at the Kellogg Institute inearlier years.

A grant by the Fulbright Educational Partnerships of the U.S. Departmentof State enabled Kellogg to host several Visiting Fellows from Venezuela andPeru starting in academic year 2003–2004 (Grant SPE50003GR057). This granthelped the Kellogg Institute to continue developing its ties with the Andeancountries, especially with our institutional partners, the Instituto de Estudios Pe-ruanos (IEP) and the Instituto de Estudios Superiores de Administración (IESA),as we brought this book to fruition. Through this program, Martín Tanaka was aVisiting Fellow at Kellogg as we worked on this book; he wrote Chapter 2 ofthis volume.

We are grateful to our editors at Stanford University Press, Amanda Moran andJared Smith, who gave their steadfast support to this project. Anna EberhardFriedlander was a capable project manager, and Ruth Steinberg did the copy ed-iting. Elizabeth Rankin ably prepared the manuscript for publication and createdthe index. Julia Smith and Elizabeth Station translated several chapters fromSpanish to English. The comments of Paul Drake and an anonymous reviewerhelped all of us sharpen our arguments.

Ana María Bejarano is Assistant Professor of Political Science at the Univer-sity of Toronto. She holds a Ph.D. in Political Science from Columbia Univer-sity. She previously was Professor of Political Science at the Universidad de LosAndes in Bogotá, where she also served as Director of the Center for Social andLegal Research (CIJUS). She co-edited Elecciones y democracia en Colombia,1997–1998 (Universidad de los Andes, Fundación Social, Veeduría ElecciónPresidencial, 1998), and co-authored the chapter on Colombia in FrancesHagopian and Scott Mainwaring, eds., The Third Wave of Democratization in LatinAmerica: Advances and Setbacks (Cambridge University Press, 2005). Recent publications include articles in Constellations and the Canadian Journal of LatinAmerican and Caribbean Studies. She is finishing a book on the historical originsand divergent trajectories of democracy in Colombia and Venezuela. Her cur-rent research deals with regime change, institution-building, and constitution-making in the Andes.

Brian F. Crisp received his Ph.D. from the University of Michigan and is cur-rently an Associate Professor in the Department of Political Science at Wash-ington University in St. Louis. His work on the institutional mechanisms con-structed to formalize state– civil society relations and the impact of theserelations on policy choices has appeared in the American Political Science Review,American Journal of Political Science, and Journal of Politics. His book, Democratic In-stitutional Design: The Powers and Incentives of Venezuelan Politicians and InterestGroups, was published by Stanford University Press (2000).

Daniel H. Levine is the James Orin Murfin Professor of Political Science at theUniversity of Michigan. He has published extensively on issues of democracy, de-mocratization, social movements, Venezuelan politics, and religion and politics

CONTRIBUTORS

xiv Contributors

in Latin America. His books include Conflict and Political Change in Venezuela(Princeton University Press, 1973), Religion and Politics in Latin America: TheCatholic Church in Venezuela and Colombia (Princeton University Press, 1981), andPopular Voices in Latin American Catholicism (Princeton University Press, 1992).

Scott Mainwaring is Eugene Conley Professor of Political Science and Direc-tor of the Kellogg Institute for International Studies at the University of NotreDame. His books include The Third Wave of Democratization in Latin America: Ad-vances and Setbacks (Cambridge University Press, co-edited, 2005), Democratic Ac-countability in Latin America (Oxford University Press, co-edited, 2003), Chris-tian Democracy in Latin America (Stanford University Press, co-edited, 2003),Rethinking Party Systems in the Third Wave of Democratization: The Case of Brazil(Stanford University Press, 1999), Presidentialism and Democracy in Latin America(Cambridge University Press, co-edited, 1997); Building Democratic Institutions:Party Systems in Latin America (Stanford University Press, co-edited, 1995), andmany others. He received a John Simon Guggenheim Memorial Foundationfellowship in 2000 for work on a project on authoritarianism and democracy inLatin America, 1945–2000.

René Antonio Mayorga is a Senior Researcher at the Centro Boliviano de Estudios Multidisciplinarios (CEBEM). He is also Professor of Political Scienceat the Facultad Latinoamericana de Ciencias Sociales (FLACSO) in Ecuador,and in the Joint Master’s Program of CEBEM, FLACSO, and the UniversidadComplutense of Madrid. He has been a visiting professor at Brown Universityand at the Universities of Salamanca, Berlin, and Notre Dame, among others.His books include De la anomia política al orden democrático (CEBEM, 1991); Antipolítica y neopopulismo (CEBEM, 1995); and La cuestión militar en cuestión:Democracia y fuerzas armadas (CEBEM, 1994). He is editor of Democracia y go-bernabilidad en América Latina (ILDIS, CEBEM, 1992). He also contributedchapters to The Third Wave of Democratization in Latin America (Cambridge Uni-versity Press, 2005), Transitional Justice and the Rule of Law in New Democracies(Notre Dame University Press, 1997), and Mixed-Member Electoral Systems: TheBest of Both Worlds? (Oxford University Press, 2001).

Kathleen O’Neill is studying law at New York University. She received a Ph.D.in Political Economy and Government from Harvard in 1999, and she was anAssistant Professor of Government at Cornell University from 2000 to 2003.

Simón Pachano is a Researcher at FLACSO in Quito, Ecuador. He has alsotaught in Spain and Bolivia, and he has published extensively on Ecuadorian pol-itics. Among his many publications are Democracia sin sociedad (ILDS, 1996) andLa representación caótica: Análisis del sistema electoral ecuatoriano (FLACSO, 1998).

Eduardo Pizarro Leongómez is Professor at the Universidad Nacional deColombia in Bogotá and served as Director of its Instituto de Estudios Políticosy Relaciones Internacionales (IEPRI). He was a Visiting Fellow at the KelloggInstitute for International Studies in 2000 –2001. He has published severalscholarly analyses of the conflict in Colombia, including Una democracia asediada:Balance y perspectivas del conflicto armado en Colombia (Editorial Norma, 2004), andhe is a weekly contributor to the national newspaper, El Tiempo.

Catalina Romero is Professor of Sociology at the Department of Social Sci-ences and Dean of the Social Sciences at the Pontificia Universidad Católica delPeru. She obtained her Ph.D. at the New School for Social Research in 1989.She was Director of the Instituto Bartolomé de Las Casas in Lima from 1982 to1992. She has published several articles and books about the Catholic Church,social movements, and civil society in Peru.

Martín Tanaka holds a Ph.D. in Political Science from the Facultad Latino-americana de Ciencias Sociales (FLACSO) in Mexico City. He is Director of theInstituto de Estudios Peruanos in Lima. Tanaka has written about Peruvian pol-itics, Latin American politics, and social movements and participation in Peru.He is author of Los espejismos de la democracia: El colapso del sistema de partidos enel Perú (Instituto de Estudios Peruanos, 1998). He contributed chapters toFrances Hagopian and Scott Mainwaring, eds., The Third Wave of Democratiza-tion in Latin America: Advances and Setbacks (Cambridge University Press, 2005),and to Carol Wise, Riordan Roett, and Guadalupe Paz, eds., Post-StabilizationPolitics in Latin America (Brookings Institution Press, 2003).

Deborah J. Yashar is Associate Professor of Politics and International Affairs andDirector of the Program in Latin American Studies at Princeton University. Sheis the author of Contesting Citizenship in Latin America: The Rise of Indigenous Move-ments and the Postliberal Challenge (Cambridge University Press, 2005) and De-manding Democracy: Reform and Reaction in Costa Rica and Guatemala, 1870s–1950s(Stanford University Press, 1997). Her current research focuses on the intersec-tion of civil wars, reconstruction, and democratization, and on inequality.

Contributors xv

THE CRISIS OF DEMOCRATIC REPRESENTATION IN THE ANDES

This book analyzes and explains the crisis of democratic representation in five Andean countries: Bolivia, Colombia, Ecuador, Peru, and Venezuela. In this region, disaffection with democracy, political parties, and legislatures has spread to an alarming degree. In Bolivia (2003), Ecuador (1997, 2000, and 2005), Peru (2000), and Venezuela (1993), democratically elected presi-dents were not able to finish their terms of office because of popular and elitediscontent. In Peru and Venezuela, massive discontent with existing party op-tions gave rise to surprising collapses of the party system in the 1990s. Politicaloutsiders with anti-establishment discourses have flourished as traditional par-ties have faded. In Ecuador, a successful military coup removed a democratic-ally elected president in 2000; in Peru, a successful palace coup led to a demo-cratic breakdown in 1992; and in Venezuela, a coup in 2002 removed thedemocratically elected president for one day, although he made a rapid come-back to the presidency. Some parties that in the recent past were major electoralcontenders and won the presidency have seemingly suffered their terminal demise.

The crisis of democratic representation in the Andes is important both intel-lectually and politically. Understanding what has gone wrong with democracyin Latin America and many other “third wave” democracies has become one ofthe outstanding intellectual challenges of our day. The widespread dissatisfactionwith democratic representation is a core ingredient in the crisis of democracyin the Andes and throughout much of Latin America. In recent years, as thewave of transitions to democracy and semi-democracy in Latin America hasebbed, intellectual and political attention has turned to how to build more robust democracies that satisfy the aspirations of more citizens—and how tocomprehend the grave shortcomings of most existing democracies and semi-democracies in the region. Because the Andean countries provide clear examples

1

The Crisis of Democratic Representationin the Andes: An Overview

❖

Scott Mainwaring, Ana María Bejarano, and Eduardo Pizarro Leongómez

of the weakness of mechanisms of democratic representation, they are an excel-lent set of countries for examining this problem.

Politically, this subject is important because the Andean countries have thepotential to be negative role models in a region (Latin America) that has histor-ically had strong demonstration and diffusion effects in terms of regime changes(Mainwaring and Pérez-Liñán, forthcoming). Moreover, a deep discrediting ofmechanisms of democratic representation can have grave implications for de-mocracy. In Peru, disenchantment with traditional mechanisms of democraticrepresentation helped pave the way to a democratic breakdown in 1992. In Ven-ezuela, the growing disaffection with conventional vehicles of democratic rep-resentation led Hugo Chávez to the presidency in 1998. Under his leadership,democracy in Venezuela has eroded, and the country has polarized sharply be-tween his followers and foes.

We hope that the book makes five main contributions to political science and to understanding Latin American politics. First and foremost, we hope to contribute to the broadening of theoretical and empirical horizons about dem-ocratic representation by studying a region in crisis. Our work shifts the main-stream thinking about representation in three ways. Most of the work on dem-ocratic representation focuses on the advanced industrial democracies, andalmost all of it analyzes how representation works. Analyzing the Andes suggestsa more innovative (in relation to the existing literature) question that is moreimportant for our region and some other parts of the world: why representationsometimes fails to work.1 This issue is paramount because in the Andes as wellas some other parts of the developing world the perceived failures of democraticrepresentation are widespread and profound. In the extensive literature on po-litical representation, to the best of our knowledge this is only the second bookto focus on a crisis of democratic representation (see Novaro 1994). Many pre-vious works have dealt with a related subject, namely, a decline of political par-ties, but most of this literature has focused on the advanced industrial democ-racies, where (perhaps excluding Italy) there is nothing resembling the crisis ofparties and of democratic representation that has plagued the Andean region.

Much of the existing literature assumes that programmatic convergence be-tween voters and legislators is at the core of democratic representation and ex-clusively analyzes such convergence. In contrast, such programmatic or ideo-logical representation is very weak in the Andes. To understand representationin this region, it is essential to look beyond programmatic and ideological con-vergence between voters and their representatives.

Most of the literature on the advanced industrial democracies posits that pat-terns of political representation remain relatively stable over time (Bartolini andMair 1990; Converse 1969; Lipset and Rokkan 1967). When we turn to manypost-1978 democracies, however, it is important to think about a range in pat-terns of democratic representation, running from more to less legitimate and

2 Mainwaring, Bejarano, Pizarro Leongómez

stable. We pose new theoretical questions about why patterns of democraticrepresentation in many countries do not achieve the stability that most of thetheoretical literature posits.2

Second, we aspire to enrich empirical knowledge about democratic represen-tation in the Andes. New work on this type of representation in Latin Americais needed because of its importance in democratic theory and the paucity of em-pirical work on it for Latin America (see Chalmers, Martin, and Piester 1997;Hagopian 1998; Luna and Zechmeister 2005; Roberts, forthcoming). For LatinAmerica, there is extensive literature on legislators and legislatures and on partiesand party systems, and a growing literature on voters. There is little, however, ondemocratic representation, which involves the relationship between voters andparties or elected politicians.

Our third contribution revolves around explaining why a crisis of democraticrepresentation occurs. When voters have free choices from an ample array ofparty options, why do they remain deeply dissatisfied? Why can’t they find aparty option that satisfies them?

We do not definitively resolve why a crisis of democratic representationerupted in the Andes. This is a new research question that demands further ex-amination, and disagreement over it is intractable. Two authors, Brian Crisp andSimón Pachano, focus on institutional arrangements to explain deficiencies indemocratic representation. Crisp (Chapter 7) argues that in the Andes institu-tional incentives foster either too much or too little focus on national program-matic issues as opposed to district-level constituency demands. Pachano (Chap-ter 4) claims that many deficiencies of representation in Ecuador stem frominstitutional rules of the game. These rules of the game favor party system frag-mentation, impede the formation of stable ruling coalitions, and encourage afocus on provincial and local constituency service rather than programmaticnational issues. Both chapters are emblematic of institutionalist approaches tounderstanding the shortcomings of democratic representation. Both authorsimply that citizens’ deep dissatisfaction with democratic representation could beattenuated with well-conceived institutional reforms.

In their chapters, in contrast, René Antonio Mayorga (Chapter 5) and ScottMainwaring (Chapter 10) see the crisis of democratic representation as stem-ming from governance (Mayorga’s term) or state deficiencies (Mainwaring’s fo-cus). Mayorga asserts that deep dissatisfaction with democratic performance un-derlies the crisis of democratic representation and the rise of political outsiders.In the book’s Conclusion, Mainwaring argues that institutional rules of the gameare not generally at the core of the dissatisfaction with democratic representation.He asserts that the main cause of the crisis of democratic representation in theAndes has been state deficiencies in many arenas, ranging from citizen securityto corruption and economic performance. For both Mayorga and Mainwaring,the rise of political outsiders, declining confidence in parties, high electoral

The Crisis of Democratic Representation in the Andes 3

volatility, and the other manifestations of a crisis of democratic representationare products of bad performance by democratic regimes. Bad performance hasbred dissatisfaction with politicians and parties.

To this performance-based argument, Mainwaring adds one other explana-tion for the deep dissatisfaction with democratic representation. Mainwaring ar-gues that the zero-sum nature of party competition and the media focus on neg-ative images of parties and assemblies help account for the discrediting of theseagents of democratic representation. This explanation resonates with construc-tivist approaches to social science because it calls attention to the way in whichpolitical competition and media images help construct citizen conceptions ofpolitics, and specifically of parties and assemblies.

Although neither this volume nor any other can definitively resolve what hascaused the crisis of democratic representation in the Andes, our volume makesa contribution by explicitly putting this question on the intellectual agenda and by staking out three of the most important explanations: institutionalist,performance-based, and constructivist.

Our fourth contribution is to advance understanding of the consequences ofa crisis of democratic representation. This issue comes to the fore in the chap-ters by Martín Tanaka (Chapter 2), René Antonio Mayorga (Chapter 5), andDaniel Levine and Catalina Romero (Chapter 8). Tanaka and Mayorga examinethe consequences of the discrediting of democratic representation for demo-cratic regimes. In many post-1978 democracies and semi-democracies, citizenshave become disillusioned with the mechanisms of democratic representation.As both authors show, the discrediting of the conventional mechanisms of dem-ocratic representation in several Andean countries had ominous consequencesfor democratic regimes. In Peru and Venezuela, the perceived failures of tradi-tional mechanisms of democratic representation, including most dramaticallythe collapse of the old party systems, paved the way for an erosion (Venezuela)and breakdown (Peru) of democracy. In both cases, political outsiders took ad-vantage of the discrediting of the old parties, won the presidency, and began toattack and dismantle some key democratic institutions. In Bolivia, the decay ofthe major parties and the discrediting of conventional mechanisms of represen-tation led to the forced ouster of President Gonzalo Sánchez de Lozada in 2003,closing a chapter in Bolivian history during which the prospects for democracyin a poor, ethnically divided country temporarily improved.

One lesson of Mayorga’s and Tanaka’s chapters is that the deep discrediting ofagents of democratic representation is often dangerous for democracies. In thisrespect, Tanaka’s and Mayorga’s analyses are relevant for the large number ofcountries where there is a crisis of democratic representation. In this sense, theproblems that we address have implications for democratic and semi-democraticregimes in Africa, Asia, the post-Communist world, and elsewhere in LatinAmerica.


Levine and Romero ask a different question about the consequences of a crisis of democratic representation. While Mayorga and Tanaka examine theconsequences of a crisis of democratic representation for democratic regimes,Levine and Romero analyze the consequences for how poor urban citizens pur-sue their interests. In both Peru and Venezuela, politicians and parties failed todeliver what citizens wanted. Levine and Romero argue that the “discredit anddecay of established leaders and parties combined with institutional failure andsustained economic crisis opened the way . . . for a wide range of movementsto emerge and claim a voice as ‘civil society.’” Stated more generally, a crisis ofdemocratic representation has profound consequences for citizen politics.

Our fifth contribution is conceptual. We define and operationalize a crisis ofdemocratic representation in this introductory chapter. In our definition, dem-ocratic representation is the relationship by which voters authorize representa-tives to govern. We argue that citizen satisfaction with the agents of democraticrepresentation (politicians, parties, and assemblies) varies widely, and that thisvariance is expressed in both subjective and behavioral indicators. At the sub-jective level, citizens express more or less confidence in parties, politicians, andassemblies, and they view these agents as having more or less legitimacy. At thebehavioral level, citizens vote or withdraw from electoral participation. Theyremain loyal to the same party over time, or they might switch party preferencewith frequency in order to find a more satisfactory agent to represent them.They continue to vote for establishment parties or search for anti-system candi-dates because of their dissatisfaction with the existing party options. We use theterm “crisis of democratic representation” to refer to contexts in which, at thesubjective level, citizens do not trust or confer legitimacy to agents of demo-cratic representation. At the behavioral level, they are more likely to supportanti-system candidates and parties, to turn to new parties, to switch electoralpreferences with frequency, and to withdraw from electoral participation.

The book also addresses other important questions. Can innovations in rep-resentation at the subnational level offset deficiencies at the national level? Towhat extent can institutional reforms of the formal mechanisms of democraticrepresentation overcome perceived deficiencies in the system? Have the mech-anisms designed to enhance representation of indigenous groups been good orharmful to democracy? Deborah Yashar (Chapter 9) sees these new mechanismsas advancing democracy, whereas René Mayorga (Chapter 5) argues that theyhave made democracy more inclusionary but that some indigenous groups havea utopian, anti-liberal-democracy discourse and practice.

This book primarily addresses the literatures on political representation anddemocracy. It is one of the first books in English to analyze democratic repre-sentation in Latin America. By examining the crisis of democratic representa-tion, we hope to contribute to the literature that seeks to understand why manycompetitively elected regimes around the world have huge deficiencies.


In the vast literature on democracy, this book contributes to recent work onthe severe shortcomings of many competitively elected regimes (e.g., O’Don-nell 2003). In most of Latin America and in some post-Communist countries,competitively elected governments have failed to deliver the goods, generat-ing widespread dissatisfaction with democracy and concern about its future. We contribute to this literature by looking at the deficiencies of democratic representation.

One of the fundamental arguments of this book involves the relationship be-tween these two literatures, and in particular between a crisis of democratic rep-resentation and regime or state deficiencies. Tanaka, Mayorga, and Mainwaringargue that the crisis of democratic representation resulted largely from regime orstate deficiencies. The profound delegitimation and repudiation of parties andpoliticians has paved the way for democratic breakdowns and erosions. Tradi-tional agents of democratic representation—above all, political parties—may bedeeply flawed, but democracy without parties is at best severely deficient, and atworst, as Schattschneider (1942, 1) wrote long ago, simply unthinkable.

In the remainder of this introductory chapter, we undertake four tasks. First,we explain our reason for focusing on the Andes. Second, we examine why it hasbecome meaningful to think of the five Andean countries as facing some com-mon challenges. Third, we define the concept of democratic representation.Finally, we explain what a crisis of democratic representation is and examine em-pirical manifestations thereof.

Case Selection: Why Focus on the Andes?

If the problems that we are addressing are common throughout the world today,why focus on one specific region within Latin America rather than adopting across-regional research strategy such as that successfully pursued by Beissingerand Young (2002) in their book on state failure? We have two reasons. First, theAndean region is widely perceived as being in crisis, and its international im-portance has grown as a result of the crisis. The perceived deficiencies of dem-ocratic representation and the discrediting of parties are more acute in the An-des than in most of the rest of Latin America. Therefore, it is a particularly goodregion for examining the subject at hand. Yet the Andean region, as a region, hasnot been studied in much detail. In contrast to the situation with the SouthernCone and Central America, there are few works on the Andes as a region (Burtand Mauceri 2004; Conaghan and Malloy 1994; Crandall et al. 2005; Drake andHershberg, forthcoming; O’Neill 2005). An attempt to fill this gap is important.

This is not to claim, however, that the Andean region is unique in experi-encing a crisis of democratic representation or in experiencing severely deficientcompetitive regimes. On this score, the Andean region is illustrative of manystruggling democracies in and beyond Latin America.


Second, regions of the world, such as Latin America, and, within them, sub-regions, such as the Andes, are important in world politics (Gleditsch 2002;Mainwaring and Pérez-Liñán, forthcoming). Within the Andean region, thereare powerful cross-national influences and demonstration effects. The rise ofPresident Hugo Chávez in Venezuela (1998–present), for example, influencedthe electoral victory of President Lucio Gutiérrez in Ecuador (2003–present), aswell as the emergence of Evo Morales, Bolivia’s most famous leader of coca grow-ers, as a viable presidential candidate in 2002. Morales was subsequently electedpresident in 2005.

These five countries have created some regional organizations that have rein-forced common influences and the sense of a regional identity. On May 26,1969, the governments of these five countries signed the Cartagena Agreement,thus beginning an early process of regional integration. The current AndeanCommunity consists of a set of organizations known as the Andean IntegrationSystem, which includes the Andean Parliament, the Andean Tribunal of Justice,the Andean Presidential Council, the Andean Council of Foreign Ministers, andthe Andean Corporation of Promotion.3 Common institutions and some com-mon problems make a focus on the Andes a reasonable way to geographically de-limit our study.

Although our focus is the Andean region, an important part of our researchdesign, especially when we explain a crisis of democratic representation, in-volves comparing the Andes to a broader set of countries. Without such com-parison involving variance in the dependent variable (i.e., the extent to whichthe democratic representation is in crisis), it would be impossible to explain theoutcome.

The Andean Region in Crisis

The Andean region as understood in this book includes Bolivia, Peru, Ecuador,Colombia, and Venezuela. It does not include Chile and Argentina, althoughboth countries have borders along the Andean range. These two countries arepart of the Southern Cone. We include Venezuela as part of the Andean regioneven though most of its inhabitants see themselves as more geographically, eth-nically, and culturally aligned with the Caribbean than with the Andean region.The same is true of the Colombian population living along the Atlantic Coast.The reason for including them is that both countries were part of the set of re-publics whose independence was established by Simón Bolívar, and they havelong been part of the set of countries with common membership in Andean re-gional organizations.

Until the late 1980s or the early 1990s, a book on democratic representationin the Andes would have made little intellectual sense. There would have beenno compelling grounds for grouping these five countries together in terms of


their systems of representation. The five countries faced very different politicalchallenges from 1958 until the late 1980s.

During these decades, Venezuela and to a lesser degree Colombia were amongthe most successful democracies in Latin America. In 1976 –77, they were ex-ceptional cases in the region; along with Costa Rica, they were two of three is-lands of democracy in a sea of authoritarianism. In contrast, the other three coun-tries analyzed in this volume had only short-lived experiments with democracybefore 1978. Bolivia had a semi-democratic regime from 1956 until 1964, fol-lowed by a string of mostly harsh military dictators from 1964 until 1982, inter-rupted only by two very short-lived efforts to install democracy in 1979 and 1980.Ecuador had semi-democratic regimes from 1948 until 1962 and from 1968 to1970, but they were punctuated by military coups. Until 1980, Peru’s only expe-rience with democracy was short-lived, lasting only from 1963 until 1968. Perualso had semi-democratic regimes from 1945 to 1948 and from 1956 to 1962.

Economically, too, there was a sharp contrast among these five countries un-til the late 1980s. In most of the post-1945 period, Venezuela had the highestper capita income in Latin America. Colombia was well behind Venezuela, buthad a per capita income and a standard of living far higher than that found inBolivia and Ecuador. In contrast, Bolivia has been one of the poorest countriesin Latin America since the early twentieth century. Ecuador was also muchpoorer than Venezuela. These economic differences underpinned contrasts indemocratic representation during those interludes in which democracy existedin Bolivia, Ecuador, and Peru. Venezuela was a wealthier, more educated, andmore urban society than the other four countries, with corresponding differ-ences for democratic representation.

That today these five countries face some important common challenges indemocratic representation is remarkable in view of their very divergent histories.The emergence of an intellectually interesting common puzzle about a crisis ofdemocratic representation in the 1990s reflects the confluence of striking changesin these five countries. One change is positive. With the exception of the Peru-vian breakdown of democracy in 1992, all five countries have had democratic orsemi-democratic regimes since the Bolivian transition to democracy in 1982.While Colombia and Venezuela underwent early transitions to competitive po-litical regimes in 1958 and 1959, respectively,4 Ecuador, Peru, and Bolivia re-stored such regimes during what Huntington (1991) called the “third wave of de-mocratization” at the end of the 1970s and the beginning of the 1980s.

Other changes have been inauspicious, and a shared sense of crisis that hasroiled the five countries has contributed to the relevance of analyzing them as asubregion within Latin America. Venezuela’s once solid democracy began to faceserious challenges in 1989 with the outbreak of massive popular protests againstPresident Carlos Andrés Pérez. In 1992, a military coup led by future presidentHugo Chávez failed, but it nonetheless signaled the growing disenchantment


with the existing political system. Deepening repudiation of the establishmentparties made possible Chávez’s election in 1998. Colombia’s democracy alsoeroded in the 1990s, victim of an armed conflict between drug lords, paramili-tary forces, left-wing guerrillas, and of a weakened state in the rural areas (Be-jarano and Pizarro Leongómez 2005). Once Venezuela and Colombia stood outas more democratic than their Andean counterparts; by the 1990s, the challengesthey faced were more similar to those of their Andean neighbors than had beenthe case in the previous four decades.

Thus, a region that was once characterized by profound contrasts in terms ofdemocratic representation, running the gamut from relatively stable and legiti-mate patterns of democratic representation in Venezuela and Colombia to dicta-torships in the other three countries during much of the 1960s and 1970s, startedto acquire some similarities. In all five countries, political outsiders burst onto thescene and challenged for the presidency—successfully in Bolivia (2005), Ecuador(2002), Peru (1990), and Venezuela (1998).5 In all five countries, electoral volatil-ity escalated, reflecting citizen discontent with existing party options. In all butVenezuela,6 public opinion reflected poor evaluations of parties and Congress,two pillars of democratic representation. Support for democracy is fairly low inall the countries except Venezuela, as Table 1.1 shows. The table gives the per-centage of survey respondents who agreed that “Democracy is always the bestform of government.” Respondents were given two other options: (1) “Forpeople like me, the form of government does not matter”; and (2) “Under someconditions, an authoritarian regime is better.”

Economically, too, there has been some convergence among these five coun-tries, mainly as result of Venezuela’s protracted economic demise. A countrythat once prospered relative to its Andean neighbors no longer stands out so dis-tinctively. In 1960, Venezuela’s per capita income was 3.4 times greater thanColombia’s; by 2002, its per capita income was only 31 percent higher. Duringthese four decades, Venezuela’s per capita income slid from $3,720 to $2,979,


Table 1.1

Support for Democracy, Latin America, 2005

% of respondents who unconditionally

Country or region favor democracy

Bolivia 49Colombia 46Ecuador 43Peru 40Venezuela 76

Average—Andean Region 51Total—Latin America 53

SOURCE: Latinobarómetro survey, 2005.

while the per capita income of Colombia and Ecuador increased substantially,and that of Bolivia and Peru increased modestly (11 percent and 26 percent, respectively) (Table 1.2).

From 1980 to 2002, per capita GDP fell in four of the five countries, all butColombia, with a particularly protracted and steep decline in Venezuela. Co-lombia, which enjoyed modest economic growth during the 1980s and first halfof the 1990s, has experienced economic stagnation since the mid-1990s.

The negative per capita economic growth for the region, coupled with poorjob generation, has led to increasing poverty. According to data of the EconomicCommission for Latin America and the Caribbean, in 2001, 61 percent of Bo-livians, 60 percent of Ecuadorians, 55 percent of Colombians, 49 percent of Pe-ruvians, and 49 percent of Venezuelans lived in poverty. Poor economic growthand increased poverty have bred dissatisfaction with democracy, resulting inpeople’s deteriorated image of two of the main pillars of representative democ-racy: parties and parliament.

A third factor that has fostered convergence across these five countries interms of representation has been the social dislocation caused by a market-oriented model of economic development. The industrial crisis due to the de-mise of import substitution industrialization and the turn toward market-oriented policies in the 1980s and 1990s was a turning point in Latin America’spolitical development. Government withdrawal, fiscal crises, and policies favor-ing economic austerity limited the flow of resources needed to sustain partiesfounded on clientelistic (Colombia) and corporatist (Venezuela) networks (Hago-pian 1998; Roberts, forthcoming). Some parties and party systems in Latin Amer-ica (Chile, Uruguay, and Costa Rica) have adapted to the new challenges ush-ered in by the era of market-oriented economic policies, while others have not.

Market-oriented models of economic growth and the decline in living standards for large sectors of society deepened the social chasm in most LatinAmerican countries, especially in the Andean region, between groups either incorporated or unincorporated into the formal economy, social security, stable


Table 1.2

GDP per Capita in the Andes, 1960 –2002(constant 1995 U.S. dollars)

% change, % change,1960 1980 2002 1960 –2002 1980 –2002

Bolivia $ 848 $1,014 $ 940 11 �7Colombia 1,104 1,868 2,282 107 22Ecuador 1,090 1,816 1,796 65 �2Peru 1,875 2,569 2,380 26 �7Venezuela 3,720 3,991 2,979 �20 �25

SOURCE: World Bank, World Development Indicators Database.

employment, unions, public services, and legalized neighborhoods. The un-incorporated sectors form Hugo Chávez’s electoral base in Venezuela. They arealso the source of widespread social and political movements responsible forpopular protests in Bolivia, Ecuador, and Peru.

Since the mid-1990s, the Andean region has been the most volatile in LatinAmerica. It has also been the region within Latin America of greatest concern toU.S. policy makers. Insufficient economic growth, rising poverty, increased eco-nomic inequality, disillusionment with the results of the democratic process, drugtrafficking, and the risk that the armed conflict in Colombia will overflow thecountry’s borders highlight the gravity of the situation throughout the region.7

The Concept of Democratic Representation

Because democratic representation is the central subject of this book, it is cru-cial to be clear about what we mean by this term. We use the term “represen-tation” to denote a principal–agent relationship whereby A (the principal) au-thorizes B (the agent) to act on her behalf. The clearest relationships ofrepresentation are those in which a clearly defined principal (an individual, agroup, an association, the electorate, etc.) explicitly delegates a clearly definedagent to undertake a task.8 Examples of explicit acts of delegation include vot-ing for someone to represent one’s interests or formally designating them to doso, creating a union to represent workers’ interests, and hiring a lawyer to rep-resent someone legally.

Our definition of representation is narrower and more clearly delineated thansome. Some prominent definitions of representation are impossible to opera-tionalize. For example, Manin, Przeworski, and Stokes (1999a, 2) define repre-sentation as “acting in the interest of the represented,” or as “acting in the bestinterest of the public.” We downplay whether the agent is acting in the interestof the public or the represented and instead focus on whether a principal authorizes an agent to act on her behalf. It is extremely difficult to establishwhether elected representatives are acting in the best interest of the public or ofthe represented.9 Hence, by Manin et al.’s definition, it is very difficult to es-tablish when a relationship of representation exists.10 In a similar vein, Pitkin(1967, 209) argues that representing “means acting in the interests of the repre-sented in a manner responsive to them.”

The definitions in Manin et al. and Pitkin do not stipulate that representationrequires a principal–agent relationship. By their definitions, a vast and ill-definedrange of actions purportedly undertaken on behalf of some people or of the pub-lic good could be understood as “representation.” In contrast, our definition iseasy to operationalize. We make no claim that elected representatives actually acton behalf of their constituents or the public good because of the difficulties injudging such claims.


Our understanding of “representation” also diverges from a usage found inmany empirical studies in the advanced industrial democracies. Many such stud-ies (Barnes 1977; Converse and Pierce 1986; Hill and Hurley 1999; McCroneand Kuklinski 1979; Pierce 1999; Thomassen 1999; Weissberg 1978; Wlezien2004) measure representation by the degree of programmatic convergencebetween voters and representatives.11 They implicitly assume that representationmust be programmatic. In contrast, and in agreement with Kitschelt (2000), ourdefinition allows for the possibility of clientelistic forms of representation, a pos-sibility that is widely practiced in the Andean region (see the chapters by Pachanoand Pizarro Leongómez). The degree of programmatic convergence betweenvoters and representatives might be important for the quality of representation,but it is an empirical issue rather than a defining characteristic of a relationshipof representation. Although some definitions and uses of “representation” di-verge from the one we employ, others are close to it (e.g., Manin 1997, 6 –7).

Democratic representation refers to the specifically democratic form of repre-sentation that is established when a voter (the principal) chooses an agent (a politi-cian or a party) to represent her interests in a democratic regime. The core ofdemocratic representation lies in the relationship between citizens, on the onehand, and elected politicians, parties, and assemblies, on the other. In this rela-tionship, voters are the principals, and elected politicians, parties, and assembliesare the agents.

Elections are the mechanism through which the relationship of democraticrepresentation is produced and reproduced (Manin 1997). They not only providethe means to elect the representatives. They are also a mechanism through whichcitizens make their preferences known, sending a message—albeit a blunt one—to their representatives about their policy preferences. In principle, electionsshould guarantee that electoral accountability will be a periodic event.12 For thisreason, the institutions that regulate the way elections are structured and carriedout, that is, electoral systems, are key in structuring representative institutions andrelationships ( Jackisch 1997).

To qualify as democratic, a relationship of representation requires that freeand fair elections take place. In addition, in the contemporary period, demo-cratic representation implies a political system that affords nearly universal adultsuffrage, respect for human rights and traditional civil liberties, and the subor-dination of the military to elected officials.

Of course, political parties and politicians are not the only vehicles that expresscitizens’ interests in democracies. Citizens also pursue their interests throughsocial organizations that aggregate, articulate, and express interests, as well asthrough intermediaries and movements that translate those interests into the po-litical arena and formulate public policy (see the chapters by Levine and Romeroand by Yashar). In a strict sense, however, democratic representation occurs onlybetween voters and their elected representatives in a democracy.


Neighborhood associations, social movements, and non-governmental orga-nizations do not fall within our definition of democratic representation in the nar-row sense—that is, that form of representation specific to democracies—becausethey can function under a democratic or a non-democratic regime. Vibrant so-cial movements can exist under authoritarian regimes. Their form of representa-tion is not specifically democratic. Indeed, in their classic work on transitions todemocracy, O’Donnell and Schmitter (1986) argued that social movements arefrequently less dynamic under democracy than in the waning phases of authori-tarianism. In addition, direct forms of collective action do not involve represen-tation. With direct collective action, individuals mobilize to work for some out-come rather than authorizing an agent to do so.

The web of representation under democratic regimes involves a wide rangeof different kinds of mechanisms. Democratic representation, however, impliesa more specific relationship, namely, that between voters and their elected rep-resentatives in a democratic regime. This is the form of representation specificto modern mass democracies.

Shirking, Accountability, and Democratic Representation

Democratic representation has been a perennial political problem. In all prin-cipal–agent relationships, agents inevitably acquire some autonomy with respectto the principal. The problem of democratic representation is the difficulty of en-suring that this autonomy is somewhat limited so as to promote some respon-siveness of politicians and parties to citizen interests or to the public good—thatis, to ensure that a formal relationship of representation works for the repre-sented, as democratic theorists and citizens hope will be the case. Curbing agents’autonomy is more problematic in the relationship between voters and electedpoliticians than in many principal–agent relationships because of a huge infor-mation asymmetry between voters and politicians, the blunt character of the pref-erences transmitted from voters to their representatives, the relative infrequencyof elections, and the difficulties of sanctioning agents who do not perform welluntil the next elections.

A rational citizen could willingly cede some autonomy in decision making to her representative on the grounds of the representative’s superior expertise on many issues (Dahl 1970, 28– 40; Pitkin 1967, 145; Rogowski 1981). Never-theless, a rational citizen would desire some responsiveness to her interests bythe representative. Through elections, the represented delegate to their rep-resentatives the power to make binding decisions, presumably in exchange forsome measure of responsiveness and electoral accountability (Fearon 1999; Ferejohn 1999; Manin 1997; Maravall 1999; Pitkin 1967, 209; Powell 2000). We understand responsiveness as the policy congruence between citizens andtheir representatives (Stimson, MacKuen, and Erikson 1995).13 By electoral


accountability, we mean that voters periodically have a chance to choose differ-ent representatives.

Elections do not guarantee that elected representatives will represent theirconstituents well (Downs 1957; Dunn 1999, 338–39; Fearon 1999; Ferejohn1999; Przeworski et al. 1999; Manin 1997; Manin, Przeworski, and Stokes1999b; Maravall 1999; Schumpeter 1946; Stokes 1999). Elected politicians havegreat opportunities to shirk. Elections occur intermittently, and nothing assuresthat elected representatives will behave according to voters’ preferences betweenelections. Tremendous information asymmetries between elected officials andthe average voter give the former ample opportunities to behave with autonomy.As Ferejohn (1999, 137) succinctly summarizes, “Electoral punishment . . . is afairly blunt instrument, and incumbent officials will be, at best, only moderatelyresponsive to public wishes.”14

If all relationships of democratic representation afford opportunities for shirk-ing, this problem is particularly acute in countries with more pronounced in-formation gaps between voters and politicians. In the Andes, the main agents ofdemocratic representation (political parties, elected politicians, and assemblies)until recently had too much autonomy with respect to most principals—in par-ticular, the large contingent of poor voters, who are formally represented byparties, politicians, and assemblies but whose capacity to influence political out-comes was seemingly marginal.15 By “autonomy,” we do not mean merely the in-dependence that representatives need in order to make good decisions for thepublic good on issues where their technical expertise exceeds that of the com-mon citizen. In the popular perception, representatives in the Andean countriesenjoy another, far more pernicious autonomy—the ability to turn their backson the electorate and function as a freewheeling, self-serving political class. InLatin America and in the Andean region in particular, until recently there was a chronic lack of political responsiveness to the masses. The informationasymmetries between voters and representatives are much greater in the Andes,where most voters have limited education and little information about politics,than in the advanced industrial democracies.

What Is a Crisis of Democratic Representation?

In this section, we explain what we mean by “a crisis of representation,” provideempirical measures thereof, and indicate the theoretical underpinnings of theseempirical measures. The idea of a crisis of representation has gained currency inLatin America (Cheresky 2003; Grompone 1996; Novaro 1994, 1995; Peruzzotti2004), and it finds faint echoes in the extensive literature on the decline of trustin institutions and the weakening of parties in the advanced industrial democra-cies. Yet the social science literature has not clarified what the concept “crisis ofdemocratic representation” means or how it can be empirically assessed. Without


explicit comparative benchmarks, the notion of a crisis of democratic represen-tation is underspecified.

The legitimacy of democratic representation is a continuous variable. We usethe term “crisis of democratic representation” to refer to one end of this con-tinuum, at which citizens do not believe they are well represented.

A crisis of democratic representation has an attitudinal/subjective and a be-havioral component. The subjective component involves citizen perceptions:large numbers of citizens are dissatisfied with the way in which they are repre-sented, or they may not feel represented at all. The represented (or those whoshould in principle be entitled to be represented but are not) believe that the pu-tative terms of the principal–agent relationship of delegation are being broken.They do not believe that the representatives are effectively acting on behalf ofthe represented or of some common good. The existence of a relationship ofdemocratic representation does not depend on whether the representatives areacting on behalf of the public good or of their own constituents, but citizens’perceptions of being adequately represented hinge on whether they believe therepresentatives are acting on behalf of some vision of the public good or of thecitizens’ interests. If citizens do not believe that representatives are acting on be-half of their constituents or of some vision of a public good, they have no rea-son to feel adequately represented. When such a perception of not being ade-quately represented is pervasive and more than transitory, it constitutes a crisisof democratic representation.

A citizen may feel adequately represented either because she believes heragents of representation are attempting to further some public good or becauseshe believes they are acting appropriately on behalf of her interests. A relation-ship of representation exists when citizens elect representatives, but citizen sat-isfaction with their representatives depends on their perception of how well thelatter perform their duties. Representatives in two countries could carry out theirjobs in the same way and obtain the same results in terms of government output,but citizen evaluations of their representatives could differ markedly in the twocountries.16

The continuum from greater to lesser satisfaction with and legitimacy of agentsof democratic representation also has a behavioral aspect. Even if there werewidespread citizen disaffection with existing agents of democratic representa-tion, we would be reluctant to affirm that a crisis existed unless there were be-havioral indications of repudiation of those agents. Citizens must reject existingmechanisms of democratic representation. They can do so by withdrawing fromelectoral participation, voting for new parties (especially anti-establishmentones), voting for political outsiders, turning to anti-system popular mobilizations,or joining revolutionary struggles, among other possibilities.

“Crisis” is a useful heuristic concept, although not one that has a precisesocial scientific demarcation. The degree of disaffection with and rejection of


democratic representation is best thought of as a continuum rather than a di-chotomy, with no precise cut point that enables one to categorize case A as a cri-sis and case B as a non-crisis. If there is no such precise cut point, how can weclaim that there is a crisis of democratic representation in the Andes? Some em-pirical measures indicate where different country cases fit on the continuum, andthese empirical indicators all locate the Andean cases toward the crisis end of thecontinuum. The concept “crisis of democratic representation” is not useful forintermediate cases, but it is useful for the unambiguous cases at this end of thecontinuum found in the Andes.

Confidence in Representative Institutions

This section begins our analysis of different measures of disaffection with and re-jection of democratic representation, that is, of our dependent variable. Our firstmeasure is based on survey data on confidence in the agents of democratic repre-sentation—specifically, parties and national legislatures.17 Trust in parties andparliament is a proxy for, not identical to, a judgment about whether a citizenfeels properly represented. Nevertheless, in the absence of direct survey questionsabout whether citizens feel properly represented, information about trust in theagents of democratic representation is a good proxy. A crisis of democratic rep-resentation involves a situation in which citizens have very low trust orconfidence in these agents of democratic representation.

In surveys, in almost all democracies, citizens express low trust in parties andthe National Congress. Hence, it is important to look at the data in compara-tive terms rather than absolute terms. How do citizen evaluations in country Xcompare with those in country Y, or how do citizen evaluations of institutionA compare with those of institution B, or how do evaluations of institution Zchange over time?

Data from the Latinobarómetro surveys (Table 1.3) show a profound lack oftrust in representative institutions in the Andes, especially in Ecuador, Colom-bia, and Bolivia. Even in the context of a region (Latin America) where demo-cratic institutions were performing poorly, the Andean subregion stood out interms of low citizen trust in parties and the national assembly.

Table 1.3 is based on responses to the question, “Please tell me how muchtrust [confianza] you have in each of the following groups, institutions, or per-sons mentioned on the list: a lot, some, a little, or no confidence?” We summedthe percentage of positive answers (“a lot” and “some”). In 1996, the five An-dean countries scored much worse than the Latin American average. Among theseventeen Latin American countries and Spain, Venezuelans and Colombiansexpressed the least trust in parties, and Bolivians were the fifth most negative. Asimilar story obtained in citizen evaluations of the National Congress. In 1996,citizen trust in Congress was worst among the eighteen countries in Colombia,third worst in Venezuela, and sixth worst in Bolivia.


Trust in parties and Congress increased significantly in Venezuela after HugoChávez’s election in 1998, but in the other four countries it has been chroni-cally low. The upsurge in trust in parties and Congress in Venezuela presumablystems primarily from pro-Chávez voters who were disaffected under the pre-1998 regime. In Bolivia, Ecuador, and Peru, between 1996 and 2003 trust inparties and the National Congress decreased.

Many authors have asserted that confidence in public institutions in the ad-vanced industrial democracies has eroded in recent decades (Dalton 1999; Do-gan 1997; Hetherington 1998; Lipset and Schneider 1983; Pharr and Putnam2000). How do levels of confidence in parties and parliament compare in West-ern Europe, the Andes, and the rest of Latin America? Table 1.4 looks at thisquestion based on data in the World Values Surveys, which included Colombia,Peru, and Venezuela, but unfortunately not Bolivia and Ecuador. The cross-regional comparison puts the Andean countries in a broader comparative per-spective.18 Confidence in parties and parliament was much higher in Western Eu-rope than in Colombia, Peru, and Venezuela. The average percentage of thoseexpressing confidence in parliament for the seventeen Western European coun-tries was 47 percent, and the average for the three Andean countries was only 23percent. The average percentage who expressed trust in political parties was 25percent for five Western European countries, 22 percent for the seven LatinAmerican countries outside the Andes, and only 15 percent for the three Andeancountries. Citizens in the Andes do not trust the institutions that are designed torepresent them.

Even if trust in parties and parliament has declined in the advanced industrialdemocracies, it remains far higher than in the Andes.19 The low level of trust in


Table 1.3

Citizen Trust in Representative Institutions, Andean Countries

% of respondents who express some or a lot of trust

1996 1997 2002 2003

political national political national political national political parties assembly parties assembly parties assembly parties

Bolivia 16.3 21.5 20.4 31.3 9 16.0 6.0Colombia 11.3 14.8 21.1 28.9 10 14.0 9.0Ecuador 18.3 26.9 15.5 19.5 7 9.0 5.0Peru 18.5 32.9 20.6 26.1 13 23.0 8.0Venezuela 11.3 18.8 20.8 29.6 19 37.0 14.0Average, five

Andean countries 15.1 23.0 19.7 27.1 11.6 19.8 8.4Average, twelve other

Latin Americancountriesa 23.5 29.3 31.4 38.1 NA NA 11.8

SOURCE: 1996, 1997, 2002, and 2003 Latinobarómetro.aArgentina, Brazil, Chile, Costa Rica, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, Panama,

Paraguay, and Uruguay.

the Andean parties and parliaments is troubling. We agree with Newton andNorris (2000, 52) that “an erosion of confidence in the major institutions of . . .representative democracy is a far more serious threat to democracy than a lossof trust in other citizens or politicians.”

Electoral Volatility

A crisis of democratic representation should also manifest itself in concrete,measurable behavioral results. In the next four sections we examine several suchbehavioral indicators using aggregate data about elections and patterns in partysystems.

Our first aggregate indicator is electoral volatility, the net share of votes thatshifts from one party to any other party from one election to the next (Pedersen1983; Przeworski 1975; Roberts and Wibbels 1999). High electoral volatilityshows large numbers of floating voters, that is, voters who do not support the sameparty in most elections. Persistently high volatility—high volatility in at least twoconsecutive electoral periods—is a possible sign of a crisis of democratic repre-sentation. It shows that large numbers of voters are repeatedly seeking alternativerepresentative vehicles, and hence suggests dissatisfaction with the quality of rep-resentation. Widespread dissatisfaction with representation could occur with


Table 1.4

Confidence in Parties and Parliament, Select Countries, World Values Survey

Confidence in Confidence in parties (%) parliament (%)

Average for Western European countries 24.6 46.9

Average for seven other Latin American countries 22.0 26.9

Venezuela 20.1 34.4Colombia 17.3 24.8Peru 7.9 9.6

SOURCE: 1995–97 and 1999–2001 waves of the World Values Surveys. Figures are for1999–2001 when a country was included in both waves. Peru and Venezuela are from the1999–2001 wave; Colombia is from 1997.

NOTE: Cell figures are the % of respondents who had a great deal or quite a lot ofconfidence in institutions. For confidence in parliament, we used all the Western European andLatin American countries included in the World Values Surveys of 1995–97 and 1999–2001.The averages for Western Europe and Latin America are unweighted. The seventeen WesternEuropean countries are Austria, Belgium, Denmark, Finland, France, Germany, Great Britain,Iceland, Ireland, Italy, Luxembourg, the Netherlands, Norway, Portugal, Spain, Sweden, andSwitzerland. The seven other Latin American countries are Argentina, Brazil, Chile, the Do-minican Republic, El Salvador, Mexico, and Uruguay. For confidence in parties, we used allseven Latin American countries; for Western Europe, we used Norway, Spain, Sweden,Switzerland, and West Germany.

low or moderate electoral volatility in the context of an oligopolistic electoralmarket,20 but usually when there is massive dissatisfaction with existing party op-tions, it opens the door to high volatility.

Table 1.5 provides data on volatility in lower-chamber elections in the fiveAndean countries. The data reinforce the idea that there is a crisis of democraticrepresentation in the Andes, especially in Peru, which has had one of the high-est levels of electoral volatility in the world. Since 1978, mean volatility in thelower chamber has been 22.1 in Colombia, 31.3 in Venezuela, 36.4 in Ecuador,39.8 in Bolivia, and 51.9 in Peru. Among these five countries, only Colombiahas had moderate volatility, and in Colombia volatility increased sharply in theelections of 1998 and 2002.

Mainwaring and Torcal (2006) calculated electoral volatility for thirty-ninecountries, including the five Andean countries, some advanced industrial de-mocracies, some other Latin American countries, and some post-Communistcases. In terms of rank order, Colombia was almost exactly in the middle, withthe 19th lowest volatility; Venezuela was 27th; Ecuador was 28th; Bolivia was31st; and Peru was 36th. Among the nine Latin American cases included in theanalysis, Peru, Bolivia, Ecuador, and Venezuela had the highest volatility. Bar-tolini and Mair (1990) analyzed electoral volatility in 303 electoral periods inthirteen Western European countries, and the mean volatility was 8.6, a smallfraction of what it has been in the Andean countries. Thus, the data on electoralvolatility support the argument that representative institutions in the Andes areundergoing intense citizen questioning.

Table 1.5 also provides data on the share of the lower-chamber vote won bynew parties. We operationalize new parties as those that competed for the firsttime within the last ten years. A high share of the vote allocated to new partiesreflects dissatisfaction with all of the traditional parties (Zoco, forthcoming). Inthe advanced industrial democracies, with rare exceptions, such as the Italian


Table 1.5

Electoral Volatility and Share of Vote for New Parties in Lower-Chamber Elections, Andean Countries

Mean electoral Elections Elections included volatility, lower included for for vote for Share of vote

chamber volatility new parties for new parties

Colombia 22.1 1978–2002 1991–2002 27.3Venezuela 31.3 1978–2001 1993–2001 39.0Ecuador 36.4 1979–1998 1996 –1998 17.0Bolivia 39.8 1980 –2002 1993–2002 32.4Peru 51.9 1980 –2001 1990 –2001 60.0

SOURCE: Electronic dataset available from Scott Mainwaring.NOTE: New parties are operationalized as those that first competed in lower-chamber elections within the

last ten years.

elections of 1993, even if citizens shift their votes away from one party to the next,most continue to vote for a party within the existing system. This measure istherefore a useful complement to the widely used data on electoral volatility. In-deed, for the purpose of assessing disgruntlement with the existing parties, it is amore useful measure.

As Table 1.5 shows, new parties have been able to burst on the scene and be-come successful electoral contenders in the Andes. The data are especially dra-matic for Peru, where on average 60 percent of the lower-chamber vote wentto new parties (with a high of 93 percent in 1995), and Venezuela, where 39percent did. Consistent with Simón Pachano’s argument (Chapter 4) on the rel-ative stability of the main party contenders in Ecuador, it is the Andean coun-try where new parties have on average registered the lowest share of the vote.

The flip side of the dramatic rise of new parties is the withering or disappear-ance of some of the traditionally major parties in these systems. In Venezuela, Ac-ción Democrática (AD, or Democratic Action) is a shadow of the party that wonthe presidency five of seven times between 1958 and 1988. COPEI, which wonthe presidency the other two times between 1958 and 1988, no longer exists. InPeru, three of the four main parties of the 1980s—IU (Izquierda Unida, or theUnited Left), AP (Acción Popular, or Popular Action), and the PPC (PartidoPopular Cristiano, or the Popular Christian Party)—have disappeared. In Bo-livia, the Acción Democrática Nacionalista (ADN, or Nationalist DemocraticAction), one of the three main contenders from 1982 until 2002, has been re-duced to irrelevance. The two other main parties, the MNR (Movimiento Na-cionalista Revolucionario, or Nationalist Revolutionary Movement) and the MIR(Movimiento de Izquierda Revolucionaria, or Movement of the RevolutionaryLeft) also suffered huge setbacks in 2005.

The traditional parties that have survived bear a faint resemblance to what theyonce were. Bolivia’s MNR led the 1952 revolution and inspired deep passion.The MNR, AD in Venezuela, and Peru’s APRA (Alianza Popular RevolucionariaAmericana, or American Popular Revolutionary Alliance) integrated the massespolitically and forged strong loyalties and identities. Little if any of that fervor re-mains. In Colombia, traditional parties have experienced an electoral erosion,and independents and minor parties have occupied growing political space (seeChapter 3, by Pizarro Leongómez).

Collapses of Party Systems

The collapse of a party system is a dramatic and unusual expression of a crisis ofdemocratic representation. It evinces repudiation not only of individual parties,but also of most of the existing parties. Citizens prefer to risk the unknownrather than sticking with the existing options. A party system collapse means aprofound rejection of existing agents of democratic representation.


Zoco (forthcoming) operationalizes a collapse of a party system as a situationin which new parties gain more than 45 percent of the votes over the course oftwo consecutive lower-chamber elections. For operational purposes, “new par-ties” are those that won less than 5 percent of the lower-chamber vote in the pre-vious election and did not have candidates for national political office (Congressor the presidency) in any election prior to that. We exclude the first two elec-tions after the inauguration of a democratic or semi-democratic regime becauseit does not make sense to think of a system collapsing before it forms. Accordingto this definition, there have been only three party-system collapses in recentdecades: those of Italy (1993), Peru (1995), and Venezuela (1998–2000). ThePeruvian party system remains in disarray more than a decade after its collapse.

In keeping with our argument that the magnitude of the crisis of representa-tion is distinctive in the Andes, two of the three recent party-system collapses inthe world’s set of democracies have taken place in this region. Consistent withour earlier argument that there should be some congruence between individual-level lack of trust in parties and aggregate behavioral indicators, we hypothesizethat countries in which citizens have very low trust in parties should be morevulnerable to party-system collapse.

This is in fact the case. Of the eighteen countries in the 1996 Latinobarómetrosurvey, Venezuela was the country with the lowest trust in parties, and it wassecond lowest to Colombia in trust in the national assembly. In the 1995 WorldValues Survey (WVS), Venezuela was among the countries with the lowestconfidence in parties and parliament. In the WVS, Peru registered the lowest ofany Latin American or Western European country in confidence in parties andparliament. In the 1980s and early 1990s, Italy was consistently the country inWestern Europe in which citizens expressed the least trust in parties and parlia-ment (Dogan 1997, 26; Listhaug 1995, 304). Thus, all three countries that expe-rienced a party-system collapse in the 1990s had been characterized by extremelylow trust in the institutions of representative democracy. This repudiation of par-ties and the national assembly was a key factor in the collapse of these three partysystems. Even with a small number of party-system collapses, the evidence sup-ports the hypothesis that collapse is more likely where trust in parties is low.

Outsider Presidential Candidates

We use a behavioral indicator to assess dissatisfaction with democratic represen-tation as expressed in presidential elections. Widespread dissatisfaction with rep-resentative institutions might affect presidential elections through the emergenceof electorally competitive outsider candidates. Substantial support for outsidercandidates expresses citizen dissatisfaction with conventional party options. Wedefine an outsider candidate as someone who runs as an independent or on anew party label. Electorally competitive independent presidential candidates and


candidates from new parties reflect disaffection with existing party options. Asnoted above, we operationalize a new party as one that won less than 5 percentof the lower-chamber vote in the previous election and did not have candidatesfor national political office in any election prior to the previous one. A candidatewho runs as an independent or on a new party is more of an outsider than onewho runs on an established party label. We exclude the first election after the in-auguration of a democratic or semi-democratic regime.

Table 1.6 presents data on the share of the vote won by outsider presiden-tial candidates in the five Andean countries and, for a baseline comparison, theUnited States.21 On average, outsiders have won between three (Ecuador) and fiveand one-half (Peru) times the share of the vote that they won in the UnitedStates—and this includes a U.S. election, 1992, with the most successful outsidercandidate in recent U.S. history (Ross Perot). Outsiders won the election in Peruin 1990, Venezuela in 1993 and 1998, and Colombia and Ecuador in 2002.22 Thisis an extraordinary political occurrence that has happened in few other LatinAmerican countries. It manifests a repudiation of the existing system of demo-cratic representation. Another outsider (Evo Morales) made it to the runoff roundin the presidential election in Bolivia in 2002, and subsequently won in 2005 (buthe does not meet our operational definition of an outsider in 2005 because of hisparty’s success in 2002). The results of Bolivia’s 2002 and 2005 elections signaleda profound erosion of the parties that had dominated Bolivian politics from 1982until 2002 (Mayorga 2005).

In 1990, in Peru, Alberto Fujimori created a new party and easily defeatedrenowned author Mario Vargas Llosa in the presidential runoff. In 1993, in Ven-ezuela, Rafael Caldera was the first presidential winner from outside the two par-ties (Acción Democrática and COPEI) that had dominated presidential elections


Table 1.6

Average Share of Vote Won by Outsider Presidential Candidates in Five Recent Presidential Elections, Andean Countries

Share of vote Average share won by outsider won by outsidercandidates, most candidates, last

Country Elections included recent election five elections

United States 1984 –2000 0.3% 6.0%Ecuador 1988–2002 58.9 17.5Bolivia 1985–2002 51.3 22.1Venezuela 1983–2000 40.2 26.5Colombia 1986 –2002 66.5 28.5Peru 1985–2001 27.9 32.7

SOURCE: Electronic dataset available from Scott Mainwaring.

from 1958 until 1988. The founder of COPEI and an ex-president, Calderabroke with his own party, formed an independent political movement, and wonthe election. As an ex-president, he was not a political outsider, but he was a partyoutsider because he ran outside the established parties. Moreover, he ran andrailed against them (Crisp, Levine, and Molina 2003). His victory marked the be-ginning of the end of the two-and-a-half party system (COPEI and AD) thatdominated Venezuela from 1973 until 1988. Then, in 1998, Hugo Chávez, whohad led a failed 1992 military coup, won the presidential election in Venezuela asa political outsider.

Álvaro Uribe Vélez, the winner of the Colombian election of 2002, was thefirst winning presidential candidate from outside the Liberal or ConservativeParties since the nineteenth century. Like Caldera in Venezuela, Uribe defectedfrom his party (the Liberals) when he failed to win the presidential nomination.Although Uribe Vélez came from the ranks of the Liberal Party, he ran as an in-dependent backed by the Conservative Party, dissident liberals, and independentsectors. For the first time in the lengthy history of the Liberal Party, a dissidentcandidate defeated the official candidate (Horacio Serpa Uribe). It was also thefirst time since 1942 (except during the Frente Nacional of 1958–74) that theConservative Party did not present a presidential candidate. Finally, in 2002, an-other ex-golpista military leader, Lucio Gutiérrez, won the presidential runoff inEcuador. Gutiérrez led the 2000 coup that deposed President Jamil Mahuad.When Gutiérrez was himself overthrown in April 2005, his vice president, Alfredo Palacio, also a political outsider, assumed the presidency.

It is not only at the presidential level that political outsiders have displaced po-litical parties. Peru is the most extreme example among these five countries interms of the ability of outsiders to displace parties (Conaghan 2000). In Peru, in2004, independent regional movements controlled 13 of the 25 regional gov-ernments and 1,634 of the 2,281 jurisdictions.23

Among the Andean countries, only Venezuela and Colombia had even mod-erately institutionalized party systems before the 1980s. Nevertheless, the decayof parties and party systems across the Andes in the 1990s and in the first half-decade of the twenty-first century is notable.

Electoral Participation

Widespread dissatisfaction with democratic representation might lead to de-pressed electoral participation and/or increased numbers of spoiled ballots. If cit-izens lose their confidence that voting makes a difference in how well they arerepresented, they are presumably less likely to vote (Dalton 1988). On the otherhand, the relationship between a crisis of representation and diminished electoralturnout and/or more spoiled ballots might be less clear than is the case withthe other indicators we have used in this chapter, especially for cross-national


comparisons. Conclusions about cross-national differences in turnout are notstraightforward because of differences in incentives to vote. Voting is obligatoryin Bolivia, Ecuador, and Peru, and was obligatory in Venezuela from 1961 until1999, but it has not been obligatory in Colombia or in Venezuela since 1999.One would expect substantially higher turnout with obligatory voting, otherthings being equal. Longitudinal within-country comparisons should still beuseful, except perhaps for comparing Venezuela before and after 1999, becausethe switch from obligatory to optional voting could explain a decline in the firstfew years after 1999.

Figures 1.1 (presidential elections) and 1.2 (lower-chamber elections) providedata on electoral participation. The trends are very similar in presidential andlower-chamber elections, and they show one more dimension of the crisis ofdemocratic representation in the Andes: declining voting. In Bolivia, Ecuador,


Figure 1.1Presidential turnout as a percentage of the eligible electorate in the Andes

Perc

enta

ge

Presidential Elections

1960 1970 1980 1990 2000

Ecuador�Bolivia

ColombiaPeru�Venezuela

100�

80�

60�

40�

20�

0

Sources: Nohlen (1993); International Institute for Democracy and Electoral Assistance (IDEA),http://www.idea.int; Oficina Nacional de Procesos Electorales (ONPE), http://www.onpe.gob.pe;International Foundation for Electoral Systems, http://www.ifes.org; Latinamerica Press, http://www.latinamericapress.org, based on UNICEF 2001 and INEI 2002 reports; Political Database ofthe Americas, http://www.georgetown.edu/pdba; Consejo Nacional Electoral (CNE); World Fact-book 2002, http://www.odci.gov/cia/publications/factbook/; and Elections Around the World,http://www.electionworld.org.


Figure 1.2Turnout as a percentage of the eligible electorate in lower-chamber

elections in the Andes

100�

80�

60�

40�

20�

0

Perc

enta

ge

Parliamentary Elections

1960 1970 1980 1990 2000

Ecuador�Bolivia


Sources: Nohlen (1993); International Institute for Democracy and Electoral Assistance (IDEA),http://www.idea.int; Oficina Nacional de Procesos Electorales (ONPE), http://www.onpe.gob.pe;International Foundation for Electoral Systems, http://www.ifes.org; Latinamerica Press, http://www.latinamericapress.org, based on UNICEF 2001 and INEI 2002 reports; Political Database ofthe Americas, http://www.georgetown.edu/pdba; Consejo Nacional Electoral (CNE); World Fact-book 2002, http://www.odci.gov/cia/publications/factbook/; and Elections Around the World,http://www. electionworld.org.

and Peru there has been a modest decline in turnout over the extended periodof time shown in Figures 1.1 and 1.2. The data also show signs of disenchant-ment with democratic representation in Venezuela and Colombia, though of adifferent nature in the two countries. Venezuela, for decades characterized byvery high electoral participation, has experienced a sharp decline in turnout.Venezuela’s turnout dropped somewhat in 1978 and 1988, and then plunged in1993. Colombia has had chronically low turnout rates but without a clear neg-ative trend since 1962. Thus, in Colombia there are indications of a chronic citizen lack of enthusiasm about democratic representation. None of the fivecountries evinces a clear upward trend in the percentage of spoiled ballots (datanot shown).

Ta

ble

1.7

Pro

gram

mat

ic R

epre

sent

atio

n in

Lat

in A

mer

ica a

nd S

pain

Mea

n le

ft–

righ

t St

anda

rdiz

edre

pres

enta

tion

gap

coef

ficie

nt, P

arty

1 �

betw

een

part

ies

Stat

isti

cal s

igni

fican

ce,

Stat

isti

cal s

igni

fican

ce,

stan

dard

ized

and

thei

r m

edia

nC

ount

ryN

agel

kerk

e R

2Pa

rty

1 vs

. Ref

eren

ce P

arty

Part

y 2

vs. R

efer

ence

Par

tyco

effic

ient

, Par

ty 2

vote

rs (

1–10

sca

le)

Spai

n.5

6.0

00.0

004.

66N

AU

rugu

ay.5

5N

ot s

igni

fican

t at

.10

.000

2.86

1.93

Chi

le.1

8.0

00.0

051.

841.

39Pa

ragu

ay.2

5.0

00.0

241.

722.

15C

olom

bia

.04

.031

.021

1.68

2.55

Arg

enti

na.1

4.0

87.0

001.

641.

83E

l Sal

vado

r.2

3.0

00.0

001.

572.

11M

exic

o.2

1.0

00.0

001.

542.

64C

osta

Ric

a.0

9.0

00.0

051.

482.

90B

razi

l.1

9N

ot s

igni

fican

t at

.10

.000

1.32

2.83

Ven

ezue

la.0

9.0

73.0

031.

272.

63E

cuad

or.0

2N

ot s

igni

fican

t at

.10

.004

1.12

2.56

Hon

dura

s.0

2.0

24N

ot s

igni

fican

t at

.10

1.09

2.32

Peru

.04

Not

sig

nific

ant

at .1

0N

ot s

igni

fican

t at

.10

1.00

2.11

Gua

tem

ala

.02

Not

sig

nific

ant

at .1

0N

ot s

igni

fican

t at

.10

1.00

3.08

Nic

arag

ua.0

1N

ot s

igni

fican

t at

.10

Not

sig

nific

ant

at .1

01.

002.

45Pa

nam

a.0

1N

ot s

igni

fican

t at

.10

Not

sig

nific

ant

at .1

01.

001.

75B

oliv

ia.0

1N

ot s

igni

fican

t at

.10

Not

sig

nific

ant

at .1

01.

002.

52

SO

UR

CE

S:

1996

Lat

inob

aróm

etro

, Que

stio

ns 3

8 an

d 40

, for

the

mul

tino

mia

l log

isti

c re

gres

sion.

Sou

rces

for

mea

n ga

p be

twee

npa

rtie

s an

d th

eir

med

ian

vote

r: fo

r vo

ters

’ lef

t–ri

ght

posit

ions

, Wor

ld V

alue

s Su

rvey

199

5–

97 fo

r A

rgen

tina

, Bra

zil,

Chi

le, M

ex-

ico,

Uru

guay

, and

Ven

ezue

la; W

orld

Val

ues

Surv

ey 1

999

–20

01 fo

r Pe

ru; L

atin

obar

ómet

ro19

98 fo

r B

oliv

ia, C

olom

bia,

Cos

ta R

ica,

Hon

dura

s, E

cuad

or, G

uate

mal

a, N

icar

agua

, and

El S

alva

dor.

Sou

rces

for

mea

n sc

ore

of p

arti

es a

s pe

rcei

ved

by d

eput

ies

of a

llot

her

part

ies:

Man

uel A

lcán

tara

, Dir

ecto

r, P

roye

cto

de E

lites

Lat

inoa

mer

ican

as, U

nive

rsid

ad d

e Sa

lam

anca

(19

94–

2005

).N

OT

ES:

Col

umns

2 t

o 5

are

base

d on

res

ults

of m

ulti

nom

ial l

ogis

tic

regr

essio

ns. D

epen

dent

var

iabl

e: in

divi

dual

s’ pa

rty

choi

ce. I

ndep

ende

nt v

aria

ble:

indi

vidu

als’

left

–ri

ght

loca

tion

. And

ean

coun

trie

s in

ital

ics.

Programmatic Representation in the Andes

An extensive literature has analyzed programmatic linkages between voters andtheir representatives (Barnes 1977; Converse and Pierce 1986; Hill and Hurley1999; Luna and Zechmeister 2005; Kitschelt et al. 1999, 309– 44; McCrone andKuklinski 1979; Miller and Stokes 1963; Weissberg 1978; Wlezien 1996; Woodand Andersson 1998).24 Programmatic representation is not the only form ofdemocratic representation. Nevertheless, programmatic convergence betweenvoters and their agents is an important ingredient of democratic representa-tion. For democracy to function well, many elected politicians must be con-cerned about the success of public policy and hence about programmatic issues.If elected politicians focus only on supplying selective (clientelistic) goods fortheir constituents, democracy cannot function well (Guevara Mann 2001).

We presume that in countries in which citizens are deeply dissatisfied withdemocratic representation, programmatic representation tends to be weak. Ifprogrammatic representation is strong, that is, if representatives and voters con-verge in their preferred ideological or programmatic positions, voters are prob-ably more likely to be satisfied with parties and politicians. Conversely, voters areprobably more likely to be dissatisfied if programmatic representation is weak.

Programmatic representation occurs along broad ideological lines more thanaccording to specific issues (Converse and Pierce 1986; Hinich and Munger 1994;Kitschelt et al. 1999, 336 –39; Thomassen 1999). Therefore, if programmaticrepresentation is functioning well, and if the left–right dimension effectively cap-tures most of the salient issues in party competition (Sani and Sartori 1983), thereshould be a correspondence between voters’ ideological position and their partychoice. For this reason, one way of assessing programmatic representation is tolook at the extent to which voters’ ideological positions predict their preferredparty. If voters’ ideological position is a poor indicator of their party choice, pro-grammatic representation is probably weak.

Columns 2 through 5 of Table 1.7 assess the variance in ideological voting forseventeen Latin American countries and Spain, based on a multinomial regres-sion analysis using the 1996 Latinobarómetro. The dependent variable was partychoice among the three largest parties (according to the survey responses) in eachcountry. For each country, the reference category was the party that, based onbinary logistic regressions for the three possible pairs of the three largest parties(results not shown), had the lowest standardized coefficient. The only indepen-dent variable in the equations was voters’ left–right position along a 0 –10 scale.25

We rank-ordered the countries by multiplying the standardized coefficient forParty 1 by the standardized coefficient for Party 2.26 (The third party in eachcountry is the reference category, so there is no coefficient for it.) The resultingproduct provides one rough indicator of how important ideological voting is indetermining party choice.


Ideological voting, and by extension programmatic representation, wereweak throughout Latin America except Uruguay and Chile.27 Ideological vot-ing was extraordinarily weak in Ecuador, Bolivia, and Peru. Colombia’s rank inthe penultimate column is deceptive because a small leftist party, with only fiveparty supporters in the survey, drove it up. In four of the five Andean countries,all but Venezuela, the Nagelkerke Pseudo R2 was extremely low, under .05, andVenezuela was not much higher at .09. In Bolivia and Peru, in the competitionamong the three largest parties, voters’ ideological positions provided no lever-age in predicting their party choice; the coefficients for voters’ ideological po-sitions are not significant at p � .10.

The final column of Table 1.7 presents a second empirical indicator of pro-grammatic representation in Latin America. For each party, we measured themean distance between each of its voter’s positions on a left–right scale from1 to 10 and its mean position on this scale as evaluated by deputies of other par-ties.28 We took each voter’s distance from the elite mean because calculating therepresentation gap by taking a mean score for all voters of the party could be mis-leading. In principle, scores for each party range from 0 (all voters of a given partyposition themselves at exactly the same point on the 1 to 10 scale as the mean forthe deputies) to 9 (all voters locate themselves at one extreme end of the scale; allelites locate the party at the other extreme end). We then weighted each party byits share of voters in the survey to generate a country-level score.

With the partial exception of Peru, the Andean countries had high represen-tation gaps. In Venezuela, which had the largest representation gap in the Andeanregion, the mean distance between where a voter located herself on the 1–10left–right scale and the mean elite position for her party was 2.63 points. A pro-grammatic representation gap of 2.63 for an individual party indicates that if theparty were located at 5.00 at the elite level, its median voter in terms of the meandistance from the 5.00 score at the elite level would locate herself at 7.63 or at2.37—a veritable chasm. The scores for Ecuador (2.56), Colombia (2.55), andBolivia (2.52) also reflect very large programmatic representation gaps betweenparties at the elite level and their voters.

Programmatic representation has increased in Venezuela since 1998 and prob-ably in Bolivia since 2002, when Evo Morales first ran for president. The emer-gence of strong leftist presidential candidates produced political polarization.The sharp polarization has clarified political options and raised the stakes of pol-itics, and presumably therefore intensified programmatic competition. Never-theless, programmatic competition was historically fairly weak in Venezuela.

The weakness of programmatic representation in the Andes underscores thatit is important to expand the scope of thinking about representation beyond whathas been done for the advanced industrial democracies. An extensive literature onrepresentation in the advanced industrial democracies has focused on program-matic representation. Yet an exclusive focus on programmatic representation


does not take us very far in understanding the Andean region. The most impor-tant conclusion about programmatic representation in the Andes is its weakness.More important in the Andes is analyzing the widespread discrediting and turn-ing away from conventional agents of democratic representation. As scholars ex-pand the geographic scope of studies of representation beyond the advanced in-dustrial democracies, it will be important to not only examine programmaticrepresentation but also to look at alternative forms of representation, such asclientelism, and to consider not only how representation works but why it oftenfails to work in the perception of citizens.

Building on Kitschelt (2000), with minor modifications, we distinguish between programmatic, clientelistic, personalistic (mainly populist), and institutional-affective linkages between voters and parties. These linkages referto the primary basis upon which a given voter supports a party or politician.First, with programmatic linkages, a voter chooses a party or candidate becauseof the congruence between her programmatic/ideological positions and theparty’s or candidate’s.

Second, a voter may choose a party or candidate primarily on the basis of selective incentives that will personally benefit the voter or some non-programmatically defined group (e.g., a neighborhood) of which the voterforms a part. In this case, a voter might cast a ballot for a politician or party eventhough a competitor is ideologically closer to her preferred position. By secur-ing clientelistic goods, voters can advance their material interests in a way thatwould not be possible through public goods. When this occurs, clientelisticlinkages are dominant. Third, a voter may choose a candidate on the basis of the candidate’s personality, without a strong link to ideological preferences or tosociological location. Finally, by institutional-affective linkages we mean that avoter supports a party based on a sense of loyalty to it—a cultural/symbolicidentification with the party—above and beyond what can be explained on thebasis of the voter’s programmatic and clientelistic interests. This kind of linkagehas received no attention in the literatures on representation and on parties, yetit deserves some consideration. Examples where institutional-affective linkagesprobably help explain why voters remained attached to parties include the sup-port of most poor voters for the Justicialist Party in Argentina even after itturned to market-oriented policies in the 1990s (Levitsky 2003); the Con-servative and Liberal Parties in Colombia in the 1960s and 1970s, when the pro-grammatic differences between them were narrow yet many citizens retainedpowerful traditional party loyalties that are probably not fully explained byclientelistic benefits (Archer 1995); and the Blancos and Colorados in Uruguay,where strong party loyalties persisted for decades despite relatively small pro-grammatic differences between them (González 1991).

Programmatic linkages are weak in the Andes, with the partial exception ofVenezuela since 1998. Traditional affective linkages between citizens and parties


have profoundly eroded. Clientelistic linkages are alive and well, but in an era ofstate shrinking (especially in Bolivia and Peru among the Andean countries), thesupply of public sector resources available to politicians for building clientelis-tic linkages has diminished. Finally, as several chapters in this volume under-score (see especially Tanaka and Mayorga), personalistic linkages have flourishedgiven the deep discrediting of conventional institutional channels of democraticrepresentation.

The Crisis of Institutionalized Democratic Representation

Our empirical indicators in this chapter have focused on the crisis of parties andassemblies at the national level. What is in crisis in the Andean region and in manystruggling democracies throughout the world is these institutionalized channelsof democratic representation. We have not focused on individual politicians asagents of democratic representation, an issue that is central to the chapters byTanaka and Mayorga.

The delegitimation and decay of party systems and the discrediting of assem-blies has, as Mayorga’s and Tanaka’s chapters show, paved the way for plebiscitar-ian forms of representation in which populist presidents displace parties as the pri-mary vehicles of expressing the popular will (O’Donnell 1994; Weyland 1999).Thus, it might be argued that personalistic, plebiscitarian representation is simplydisplacing more institutionalized democratic representation. If this is true, is itaccurate to speak of a crisis of democratic representation?

We believe that the answer to this question is affirmative. The institutionalagents of democratic representation—assemblies and parties—have long held aprivileged position in democratic theory, and for good reason. In a democracy,representatives should programmatically advance the interests of voters. The ex-pansion of personalistic representation subverts that central democratic principle.

As Tanaka and Mayorga argue in their chapters, plebiscitarian representationeasily erodes into less-than-democratic forms of governing, and indeed may pavethe way to authoritarian or semi-democratic regimes, as occurred with PresidentAlberto Fujimori in Peru in the 1990s and with Venezuelan President HugoChávez after 1998. Leaders who are elected on the basis of direct populist appeal,sometimes with demagogic claims and often with the express intention of weak-ening institutional forms of democratic representation, often undermine ratherthan strengthen democratic institutions (O’Donnell 1994). What starts out asplebiscitary representation easily erodes into non-democratic or even anti-democratic representation.

Parties are key agents of democratic representation for three reasons. First, theyprovide indispensable information shortcuts to voters (Downs 1957; Hinich andMunger 1994). If elections were organized exclusively around individual candi-dates rather than partly through parties, voters would face daunting difficulties in


obtaining the significant information needed to make good electoral judgments.Second, for this reason, parties are important mechanisms of electoral accounta-bility. If citizens want to vote the bums out, they need to be able to punish notonly individual officeholders, but also in most cases the political parties respon-sible for governing. Third, parties connect citizens to the state in a different waythan other vehicles of interest-articulation because through elections they offer ameans to state power. For good reason, Schattschneider (1942, 1) wrote that“modern democracy is unthinkable save in terms of the parties.” Thus, a pro-found erosion of parties is closely linked to a crisis of democratic representation.The rise of plebiscitarian representation does not compensate for and mayeven—as has happened frequently in the Andes—exacerbate the crisis of dem-ocratic representation.

Our focus in most of the book is democratic representation at the nationallevel. As O’Neill makes clear (Chapter 6), democratic representation can occurat the subnational level. Citizens might be satisfied with these subnational levelseven if they are dissatisfied with their agents of representation at the national level.Many proponents of decentralization believe it builds closer linkages betweenvoters and representatives, enhances electoral accountability and responsiveness,and hence can improve democratic representation. Subnational politics has be-come more important in addressing citizen needs as the process of decentraliza-tion has taken hold; hence the centrality of O’Neill’s chapter in understandingdemocratic representation today in the Andes.

Even if citizens were satisfied with their subnational agents, the overall systemof democratic representation would be compromised if voters were disgruntledwith their national agents. The national-level agents of democratic representa-tion have the first and foremost responsibility to resolve many pressing needsthat citizens face. The national macroeconomic situation, for example, affectscitizen well-being in a way that no subnational policies can adequately com-pensate. While some attention to subnational forms of democratic representa-tion is important in the Andes, what transpires at the subnational level cannotadequately compensate for failures of democratic representation at the nationallevel. Moreover, because major political parties function at both the nationaland subnational level, a crisis of democratic representation at the national levelinevitably adversely affects the legitimacy of democratic representation at thesubnational level. If decentralization weakens the national state, it might evenexacerbate the perception of a deficiency of democratic representation.

Political Reform and the Crisis of Democratic Representation

It would be inaccurate to picture these political systems as immobile, static, or re-form averse. On the contrary, during the past two decades, the Andes have beena veritable laboratory for experimenting with institutional reforms. During the


1990s, all five countries undertook major constitutional reforms.29 All have ex-perimented widely with the laws governing elections and political parties, as thechapters by Pachano, Pizarro Leongómez, Tanaka, and Crisp make abundantlyclear. As O’Neill argues in her chapter, one of the most consequential reforms inthe region has been the trend toward decentralization. Indeed, one of the funda-mental themes of this volume is the seemingly endless effort to improve mecha-nisms of representation via political reform.

Neither the political elites nor the electorate have remained paralyzed in theface of the erosion of the relations of representation. In an atmosphere of dis-enchantment and tension, the parties and the political elites, sometimes with thesupport of and other times under intense pressure from significant actors in soci-ety (the indigenous movement, e.g.) have modified existing institutional arrange-ments, seeking to improve the representativeness of these political systems.

The results of these reforms have been mixed. Despite significant gains in termsof representation of previously excluded minorities, some problems can be tracedin part to these reforms: the erosion of parties and, in some cases, the added diffi-culties in achieving effective government. The mixed legacy of these reformistefforts stems in part from the fact that institutional reform produces some un-intended, and at times undesirable, effects. The mixed results also stem from theinevitability of trade offs, something that has not been adequately addressed inthe literature on institutional reform. The twin goals of representativeness andgovernability sometimes stand in tension. Efforts aimed at enhancing either oneof these dimensions of democracy may have a deleterious impact on the other.Given the recent nature of the reforms and the limited number of cases, we donot undertake a rigorous assessment of their global long-term effects.30

The Chapters That Follow

The key agents of democratic representation are political parties, politicians, as-semblies, and elected executives, and hence these agents and their relation tocitizens (the represented) are at the core of our purview. Among the agents ofdemocratic representation, in most countries political parties have historicallyoccupied a particularly prominent position. In principle, democratic representa-tion can occur with individual politicians rather than parties as the agents ofrepresentation.31 In practice, however, in most democracies parties are the pri-mary means of representing and structuring interests in mass democratic politics(Sartori 1976). A deep discrediting and weakening of parties is therefore closelyassociated with a crisis of democratic representation. For this reason, Part I hasthree chapters on parties and their failures as agents of democratic representationin the Andes.

The ways that parties represent, and the failures of parties as agents of repre-sentation, vary across the five Andean countries. The failure of parties as agents


of representation was most profound in the two cases of party-system collapse,Peru and Venezuela (see Tanaka’s chapter). This similarity, however, masks anequally important difference. In the aftermath of the collapse of the Peruvianparty system, the rebuilding of institutionalized mechanisms of democratic rep-resentation has been painfully slow. Fujimori’s disgraceful demise and his anti-party attitudes and practices left a greatly weakened party landscape. In contrast,Hugo Chávez has more successfully built a party in Venezuela. In Colombia, thetraditional parties gradually eroded in the 1990s after a century of electoraldominance, but the party system did not collapse. The Conservative and Lib-eral parties continue to be the country’s most powerful electoral contenders.Among the traditional (pre-1990s) parties, they are electorally the strongest ofall the parties in these five countries. In Ecuador, notwithstanding a widespreadperception that mechanisms of democratic representation have failed, the partysystem of the 1980s has so far remained in place (Pachano’s chapter). In 2002,however, for the first time, a political outsider won the presidency. In Bolivia,the three mainstays of the post-1982 party system remained competitive until2002, when the ADN experienced a sharp demise. Since 2002, political out-siders have flourished, traditional mechanisms of representation have been indisarray, and direct popular mobilization has surged.

The final chapter in Part I, by René Antonio Mayorga, addresses a closely re-lated theme: the emergence of outsider politicians in response to the crisis of par-ties and party systems in the Andes. As Mayorga demonstrates, in the Andes a cli-mate of neopopulism prevails, in which multiple social sectors are demandingincorporation into the political system. This climate of neopopulism is notheaded by the organized popular sectors, but instead by the urban unemployed,the indigenous communities, those selling their wares on the streets, other groupswithin the informal sectors, coca growers, and peasants. According to Mayorga,intense social and political pressures have confronted the party systems. In somecases, such as Peru and Venezuela, this pressure was more than parties couldhandle.

Part II focuses on two important institutional issues related to democraticrepresentation. Brian Crisp’s chapter analyzes National Congresses in the Andeanregion. Congresses are one of the most important bodies of representation, andthey suffer from the same low credibility as parties. He addresses reforms intendedto enhance democratic representation but concludes, as do Pachano (Chapter 4)and Mainwaring (Chapter 10), that such reforms have not countered a deepen-ing popular sense that the mechanisms of democratic representation are not func-tioning adequately.

Kathleen O’Neill analyzes changes in the intergovernmental distribution ofpower, which affects democratic representation by allowing for (or not) theelection of representatives at the local and state level. O’Neill’s chapter raises an interesting question: Can innovations and apparent improvements in repre-


sentation at the local level offset serious deficiencies at the national level? We aredeeply skeptical, especially in cases where decision-making authority and re-sources are still centralized.

In Part III, chapters by Daniel H. Levine and Catalina Romero and by Deborah Yashar address representation of popular groups under democracy. Pop-ular organizations and movements do not fit our definition of “democratic rep-resentation” because they can articulate interests under both democratic and authoritarian regimes. Moreover, some popular mobilizations involve directparticipation in politics rather than “representation.” Nevertheless, these twochapters are essential for understanding the crisis of democratic representation.There is no clearer manifestation of this crisis than the repudiation some popu-lar groups express for the agents of democratic representation. For this reason,these two chapters are clearly relevant to the subject of this book. An examina-tion of the crisis of democratic representation requires attention to populargroups that epitomize the rejection of parties and legislatures.

Levine and Romero focus on urban citizen demands and perceptions under de-mocracy in the era of discredited formal mechanisms of democratic represen-tation. Their chapter illuminates how poor citizens attempt to further theirinterests outside the formal channels of democratic representation when thesechannels are discredited. Some movements explicitly reject the traditional ve-hicles of democratic representation. Some prefer direct participation or “self-representation” (Warren and Jackson 2002) to traditional forms of representation.

Deborah Yashar examines an increasingly important and often mobilizedgroup in the Andes, especially in Ecuador and Bolivia: the indigenous popula-tion. In Ecuador, Colonel Lucio Gutiérrez’s surprising electoral victory in the2002 presidential election was due in part to support from CONAIE (Confed-eración de Nacionalidades Indígenas del Ecuador, or Confederation of Indige-nous Nationalities of Ecuador) and, above all, from its political branch, Pachaku-tik. In Bolivia, Evo Morales placed second in the 2002 presidential election andwon in 2005 by mobilizing substantial support from indigenous peoples. Thismobilization of indigenous and popular communities is simultaneously a meansof widening the democratic spectrum and, as Mayorga argues in his chapter, adeep source of political tension. An antagonism has grown between the streetsas an expression of social mobilization, and Congress as an expression of insti-tutionalized democratic representation.32 Often, social mobilization is not seenas a complement to or reinforcement of institutionalized political activity, butinstead as an alternative and, in many cases, an anti-systemic alternative—an in-strument with which to change elected leaders via unconstitutional means oreven change the system by extra-institutional de facto means. Yashar’s chapteralso calls attention to an aspect of representation that has surfaced anew in re-cent years: the desire of some indigenous groups that people of their shared eth-nicity represent them.


The concluding chapter by Scott Mainwaring addresses the causes of lowconfidence in parties and assemblies (excepting Venezuela since Chávez tookoffice). In an argument that shares similarities with Mayorga’s, Mainwaring ar-gues that state deficiencies and the politicization of these shortcomings by com-peting parties are the primary causes of the crisis of democratic representation.Although this concluding chapter focuses on the Andes, we believe that statedeficiencies are central to understanding failures of democratic representationwell beyond this region. State performance is key to understanding the vicissi-tudes of many struggling competitive regimes in the world.

Conclusion

Many authors have claimed that conventional mechanisms of democratic repre-sentation are undergoing questioning and face declining legitimacy in most con-temporary democracies (Pharr and Putnam 2000; Pizzorno 1981). This chapterrelativizes these claims. Parties may have faced some erosion in the advanced in-dustrial democracies, but—with the exception of Italy in the early 1990s—theyhave faced nothing resembling the profound questioning that they now do in allfive Andean countries. The Andes show what a real crisis of democratic repre-sentation is.

We hope in this chapter to have contributed to thinking about what a crisisof democratic representation is and how it can be measured. This subject is oneof the fundamental issues in the Andean region today, and indeed well beyondthe Andes. If citizens believe that they are not well represented for an extendedperiod under democracy, democracy itself is easily imperiled, as has occurred inPeru, Venezuela, Bolivia, and Ecuador at different moments in the 1990s andthe first decade of the twenty-first century. The growing dissatisfaction with de-mocracy in most of Latin America (UNDP 2004) suggests that what has tran-spired already in the Andean region may be a harbinger of things to come in therest of Latin America. Understanding why citizens believe that democratic rep-resentation is failing them and addressing these shortcomings is one of the hugeintellectual and political challenges of our day. The rest of the volume exploresthe failings and successes of democratic representation in the Andes, the causesof the failings, and the consequences of this crisis.

APPENDIX

Coding Rules for Outsider Presidential Candidates

1. After a democratic transition, we did not count the first presidential election.2. A new party is one that didn’t win more than 5 percent of lower-chamber votes in

the previous election and that did not present any candidates for the National Congressor presidency prior to that.


3. If a party changed its name from Election t to Election t � 1, we did not count itas a new party at t � 1.

4. We did not count an alliance (coalition) of previously existing parties as a newparty. A coalition whose basis is not preexisting parties, however, counts as a new party.For example, the Movimiento Sí Colombia (Colombia Yes Movement) led by NoemíSanín, which obtained 27 percent of the votes in the first round in Colombia in 1998,counts as a new party. Although her party was composed mainly of leaders and follow-ers from the traditional parties, Sanín created a new electoral vehicle for her campaign.

5. We did not count a merger of two previously existing parties as a new party.6. In cases of a party schism, neither of the resulting parties is counted as new.7. We count as independents candidates who do not have a party affiliation. For ex-

ample, we count Alvaro Uribe Vélez as an independent in Colombia in 2002. Althoughhe was still formally a member of the Liberal Party, he did not run on that party’s ticketin 2002 but rather as an independent. In a similar vein, Claudio Fermín is counted as anindependent after his expulsion from AD in Venezuela.

Notes

We are grateful to Michael Coppedge, Brian Crisp, Paul Drake, Frances Hagopian, EricHershberg, Mala Htun, Wendy Hunter, Herbert Kitschelt, Soledad Loaeza, GerryMackie, René Antonio Mayorga, Carlos Meléndez, Ken Roberts, Martín Tanaka,Matthew Shugart, Kurt Weyland, Deborah Yashar, and Edurne Zoco for comments. Wealso thank Edurne Zoco, Angel Alvarez, Bong-Jun Ko, and Kathleen Monticello for re-search assistance.

1. Manin et al. (1999a, 1999b) provide an important discussion of this problem at a gen-eral abstract level but without thinking about the great cross-national variance in the sat-isfaction with or repudiation of mechanisms of democratic representation. We reverse thatfocus: we provide cursory attention to the general reasons why democratic representationmight fail and instead examine why it fails particularly in the Andes.

2. The literature on political disaffection has some relevance to our work; it, too,understands that representation sometimes fails. For a recent example, see Torcal andMontero (2006).

3. To this list one might add the Universidad Simón Bolívar, the Business AdvisoryCouncil, the Labor Advisory Council, and many other regional institutions.

4. In Colombia, which has had a persistent history of civil governments, a conserva-tive restoration of civil institutions took place in 1958. In Venezuela, a process of estab-lishing democratic institutions took place in 1958. Other than the short interlude from1947 to 1948, Venezuela had no democratic tradition before 1958. Despite this difference,the agreement upon which Venezuelan democracy was founded (known as the Punto FijoAgreement) was more inclusive than that of the Colombian Frente Nacional and gave riseto a more open and participatory democracy (see Bejarano 2000; Levine 1992).

5. In Venezuela and Colombia, Rafael Caldera (1993) and Álvaro Uribe Vélez (2002),respectively, although not political outsiders, gained the presidency via dissident move-ments of their own political parties, the COPEI and the Partido Liberal (Liberal Party).

6. Venezuelan exceptionalism in some attitudinal questions, including the one inTable 1.1 gauging voter support for democracy, requires a brief comment. Before the rise


of Hugo Chávez in 1998, Venezuelans’ support for democracy and democratic represen-tation was low. After 1998, support for democracy enjoyed a notable surge (Table 1.1),probably reflecting the very different ways in which the pro- and anti-Chávez poles in-terpret the question. Pro-Chávez individuals see the president as democratic, and hencerespond that democracy is always the best form of government. Anti-Chávez individualssee him as authoritarian, and hence agree—albeit while understanding the question in acompletely different way—that democracy is always the best form of government.

7. Although we have focused in this section on convergence among these five coun-tries, they also continue to have important differences.

8. Brennan and Hamlin (1999), Christiano (1996, 207–24), Ferejohn (1999), Fearon(1999), and Maravall (1999) also explicitly view representation through the prism ofprincipal–agent relationships.

9. Although the notion of “the best interest of the public” is intuitively appealing, the social-choice tradition (e.g., Arrow 1954) presented implicit critiques thereof thatrendered this concept problematic.

10. Moreover, the two definitions provided by Manin et al. are not the same. Actingin the best interest of the public is not the same as acting in the interest of the repre-sented. Acting in the best interest of the public may entail curbing wage raises at a givenmoment, whereas workers’ parties and politicians who represent workers would act inthe interests of the represented (the workers) by pressing for wage increases.

11. Weissberg’s (1978) definition is particularly distant from ours because it com-pletely severs the electoral linkage between specific voters and their representatives: “Ouranalysis defines representation as agreement between legislative voting and citizen opin-ion” (535n4). In his conception, an elected representative from one district can “repre-sent” a voter from another even though there is no electoral connection between them;the only issue that matters is programmatic convergence between a voter and a memberof an assembly.

12. Electoral accountability requires that citizens have the opportunity to vote politi-cians out of office. This possibility is diminished where reelection is prohibited, as is thecase in presidential systems with no reelection or with presidents who cannot be furtherreelected (e.g., the U.S. president in his second term). If reelection is banned, it is stillpossible to punish or reward an incumbent’s party, but not a specific politician unless sheruns for another office.

13. On the relationship between representation and responsiveness, see Eulau andKarps (1977) and Powell (2000).

14. This is not to claim that electoral accountability is completely ineffectual. Fiorina(1981), Key (1966), Manin (1997), and Popkin (1991), among others, have underscoredthe potential for electoral accountability through retrospective voting. Stimson et al.(1995) emphasize both retrospective voting and anticipatory shifts in policy to respondto changes in public mood as mechanisms that produce electoral accountability. SeeMansbridge (2003) for a discussion of different kinds of representation and their rela-tionship to electoral accountability.

15. We italicize seemingly because it is difficult to measure the impact of the large con-tingent of poor voters on the way parties, politicians, and assemblies formulate publicpolicy.

16. The fact that a crisis of representation hinges proximately on citizen evaluationsdoes not imply that a crisis of representation is driven by purely subjective processes. As Mainwaring argues in Chapter 10, citizens form their judgments of the agents of representation partly on the basis of state performance. Citizens have bounded rational-ity in their assessment of the agents of representation. They form reasonably “rational”


judgments. When states perform badly over a protracted period of time, citizens are un-likely to believe that the agents of representation are fostering some public good or aredelivering goods to them, and hence they are more likely to repudiate these agents.

17. Compared to the countless analyses on trust in institutions in the advanced in-dustrial democracies, there is a paucity of work on this subject in Latin America, in-cluding the Andes. For exceptions, see Cleary and Stokes (2006); Power and Jamison(2005); Turner and Martz (1997).

18. The question in the World Values Survey was the same as in the Latinobarómetro,but the coding was different. The options were “a great deal of trust,” “quite a lot,” “notvery much,” and “none at all.”

19. See Mishler and Rose (2001, 42) for data on trust in institutions in ten post-Soviet countries.

20. As occurred, for example, in Colombia during the Frente Nacional period(1958–74), when the predominance of the two traditional parties, the Liberals and Con-servatives, coincided with majority support from the population for these two parties.

21. Appendix 1 gives details on how we coded whether candidates were outsiders or not.

22. There are two types of outsiders: individuals who have never held political officeand run against the establishment, such as Alberto Fujimori, Hugo Chávez, and LucioGutiérrez, and those outside the party system, such as Rafael Caldera after his defectionfrom COPEI and Álvaro Uribe Vélez after leaving the Liberal Party. The latter are dis-sidents from traditional parties but are well-known political figures.

23. Data from Carlos Meléndez (personal communication). In their chapters, Mayorga and Tanaka analyze some consequences of the rise of political outsiders.

24. Much more remains to be done on programmatic representation in Latin Amer-ica; see Luna and Zechmeister (2005) on this issue.

25. The 1996 Latinobarómetro used a 0 –10 scale rather than the conventional 1–10scale.

26. If the standardized coefficient for either party was less than 1, we inverted it; thatis, we divided 1 by the standardized coefficient. If the coefficient was not significant at p � .10, we adjusted the coefficient to 1.00 because it is not statistically different from 1.00.

27. For data that show the weakness of ideological voting in Latin America (exceptChile and Uruguay) compared to most of the advanced industrial democracies, seeMainwaring and Torcal (2006).

28. The distance for each voter from the elite position is measured as an absolutevalue. The elite survey question we used excluded deputies from placing their own partyon the left–right scale. The Latinobarómetro surveys asked citizens to place themselves ona 0 –10 scale, whereas the World Values Survey and the Proyecto de Elites Latinoameri-canas use the more common 1–10 scale. To make the Latinobarómetro scale commensu-rable with the other two, we used the formula .9LB � 1 � WVS, where LB is the voter’sscore on the 0 –10 scale and WVS is the voter’s score adjusted to the 1–10 scale.

29. Colombia in 1991, Peru in 1992–93, Bolivia in 1994, Ecuador in 1997–98, andVenezuela in 1999.

30. For an evaluation of the consequences of institutional reform in Colombia andVenezuela, see Bejarano (2002). On the impossibility of predicting with certainty theconsequences of major constitutional changes, see Elster (1988).

31. The extent to which parties rather than individual politicians are the more salientagents of democratic representation varies by country (Dalton 1985; Esaiasson and Holm-berg 1996).


32. The debate raging throughout the Andean region between representative de-mocracy and “participatory democracy” (which is a mislabel) is an expression of thisconflict between Congress and the street.

References

Archer, Ronald. 1995. “Party Strength and Weakness in Colombia’s Besieged Democ-racy.” In Building Democratic Institutions: Party Systems in Latin America, ed. Scott Mainwaring and Timothy R. Scully, 164 –99. Stanford, CA: Stanford UniversityPress.

Arrow, Kenneth Joseph. 1954. Social Choice and Individual Values. New York: Wiley.Barnes, Samuel. 1977. Representation in Italy: Institutionalized Tradition and Electoral Choice.

Chicago: University of Chicago Press.Bartolini, Stefano, and Peter Mair. 1990. Identity, Competition and Electoral Availability:

The Stabilisation of European Electorates, 1885–1985. Cambridge: Cambridge Univer-sity Press.

Beissinger, Mark R., and Crawford Young, eds. 2002. Beyond State Crisis: Postcolonial Africa and Post-Soviet Eurasia in Comparative Perspective. Washington, DC: Woodrow Wilson Center Press.

Bejarano, Ana María. 2000. “Perverse Democratization: Pacts, Institutions, and Problem-atic Consolidation in Colombia and Venezuela.” Ph.D. diss., Columbia University.

———. 2002. “Buenas intenciones y efectos perversos: Los límites del reformismo in-stitucional en Colombia y Venezuela.” Comentario Internacional: Revista del CentroAndino de Estudios Internacionales 4, II semester: 177–86.

Bejarano, Ana María, and Eduardo Pizarro Leongómez. 2005. “From ‘Restricted’ to ‘Be-sieged’: The Changing Nature of the Limits to Democracy in Colombia.” In TheThird Wave of Democratization in Latin America: Advances and Setbacks, ed. FrancesHagopian and Scott Mainwaring, 235– 60. Cambridge: Cambridge University Press.

Brennan, Geoffrey, and Alan Hamlin. 1999. “On Political Representation.” British Jour-nal of Political Science 29 ( January): 109–28.

Burt, Jo Marie, and Philip Mauceri, eds. 2004. Politics in the Andes: Identity, Conflict, Re-form. Pittsburgh: University of Pittsburgh Press.

Chalmers, Douglas, Scott Martin, and Kerianne Piester. 1997. “Associative Networks:New Structures of Representation for the Popular Sectors?” In The New Politics of Inequality in Latin America: Rethinking Participation and Representation, ed. DouglasChalmers et al., 543–82. Oxford: Oxford University Press.

Cheresky, Isidoro. 2003. “Las elecciones nacionales de 1999 y 2001: Fluctuación del votodebilitamiento de la Cohesión Partidaria y crisis de representación.” In De la ilusión reformista al descontento ciudadano: Las elecciones en Argentina, 1999–2001, ed. IsidoroCheresky and Jean-Michel Blanquer, 19–51. Rosario, Argentina: Homo Sapiens Ediciones.

Christiano, Thomas. 1996. The Rule of the Many: Fundamental Issues in Democratic Theory.Boulder, CO: Westview Press.

Cleary, Matthew, and Susan Stokes. 2006. Democracy and the Culture of Skepticism: Politi-cal Trust in Argentina and Mexico. New York: Russell Sage Foundation.


Conaghan, Catherine M. 2000. “The Irrelevant Right: Alberto Fujimori and the NewPolitics of Pragmatic Peru.” In Conservative Parties, the Right, and Democracy in LatinAmerica, ed. Kevin J. Middlebrook, 255–84. Baltimore: Johns Hopkins UniversityPress.

Conaghan, Catherine M., and James M. Malloy. 1994. Unsettling Statecraft: Democracy andNeoliberalism in the Central Andes. Pittsburgh: University of Pittsburgh Press.

Converse, Philip. 1969. “Of Time and Stability.” Comparative Political Studies 2, no. 2:139–71.

Converse, Philip, and Roy Pierce. 1986. Political Representation in France. Cambridge,MA: Harvard University Press.

Crandall, Russell, Guadalupe Paz, and Riordan Roett, eds. 2005. The Andes in Focus: Se-curity, Democracy, and Economic Reform. Boulder: Lynne Rienner.

Crisp, Brian, Daniel H. Levine, and José Molina. 2003. “The Rise and Decline ofCOPEI in Venezuela.” In Christian Democracy in Latin America: Electoral Competitionand Regime Conflicts, ed. Scott Mainwaring and Timothy R. Scully, 275–300. Stan-ford, CA: Stanford University Press.

Dahl, Robert A. 1970. After the Revolution? New Haven, CT: Yale University Press.———. 1971. Polyarchy: Participation and Opposition. New Haven, CT: Yale University

Press.Dalton, Russell J. 1985. “Political Parties and Political Representation: Party Supporters

and Party Elites in Nine Nations.” Comparative Political Studies 18, no. 3 (October):267–99.

———. 1988. Citizen Politics in Western Democracies: Public Opinion and Political Parties inthe United States, Great Britain, West Germany, and France. Chatham, NJ: ChathamHouse.

———. 1999. “Political Support in Advanced Industrial Democracies.” In Critical Citi-zens: Global Support for Democratic Governance, ed. Pippa Norris, 57–77. Oxford: Ox-ford University Press.

Diamond, Larry. 1999. Developing Democracy: Toward Consolidation. Baltimore: JohnsHopkins University Press.

Dogan, Mattei. 1997. “Erosion of Confidence in Advanced Democracies.” Studies inComparative International Development 32, no. 3 (Fall): 3 –29.

Downs, Anthony. 1957. An Economic Theory of Democracy. New York: Harper and Row.Drake, Paul, and Eric Hershberg, eds. Forthcoming. Crises in the Andes: Comparative Per-

spectives. Pittsburgh: University of Pittsburgh Press.Dunn, John. 1999. “Situating Political Accountability.” In Democracy, Accountability, and

Representation, ed. Adam Przeworski, Susan C. Stokes, and Bernard Manin, 329– 44.Cambridge: Cambridge University Press.

Eaton, Kent. 2004. Politics Beyond the Capital: The Design of Subnational Institutions in SouthAmerica. Stanford, CA: Stanford University Press.

Elster, Jon. 1988. “Arguments for Constitutional Choice: Reflections on the Transitionto Socialism.” In Constitutionalism and Democracy, ed. Jon Elster and Rune Slagstad,303–26. Cambridge: Cambridge University Press and Universitetsforlaget.

Esaiasson, Peter, and Sören Holmberg. 1996. “Looking for Political Representation.” InRepresentation from Above: Members of Parliament and Representative Democracy in Sweden.Aldershot, England: Dartmouth Publishing.

Eulau, Heinz, and Paul D. Karps. 1977. “The Puzzle of Representation: Specifying Com-ponents of Responsiveness.” Legislative Studies Quarterly 2, no. 3 (August): 233–54.

Fearon, James D. 1999. “Electoral Accountability and the Control of Politicians: Select-ing Good Types versus Sanctioning Poor Performance.” In Democracy, Accountability,


and Representation, ed. Adam Przeworski, Susan C. Stokes, and Bernard Manin,55–97. Cambridge: Cambridge University Press.

Ferejohn, John. 1999. “Accountability and Authority: Toward a Theory of Political Ac-countability.” In Democracy, Accountability, and Representation, ed. Adam Przeworski,Susan C. Stokes, and Bernard Manin, 131–53. Cambridge: Cambridge UniversityPress.

Fiorina, Morris P. 1981. Retrospective Voting in American National Elections. New Haven,CT: Yale University Press.

Gleditsch, Kristian Skrede. 2002. All International Politics Is Local: The Diffusion of Conflict,Integration, and Democratization. Ann Arbor: University of Michigan Press.

González, Luis Eduardo. 1991. Political Structures and Democracy in Uruguay. Notre Dame,IN: University of Notre Dame Press.

Grompone, Romeo. 1996. “Representación política, partidos y escenarios futuros.” Re-vista Socialismo y Participación 73 (March): 11–18.

Guevara Mann, Carlos. 2001. “Forsaken Virtue: An Analysis of the Political Behavior ofPanamanian Legislators, 1984 –1999.” Ph.D. diss., University of Notre Dame.

Hagopian, Frances. 1998. “Democracy and Political Representation in Latin America inthe 1990s: Pause, Reorganization, or Decline?” In Fault Lines of Democracy in Post-Transition Latin America, ed. Felipe Agüero and Jeffrey Stark, 99–143. Miami: North-South Center Press at the University of Miami.

Hetherington, Marc J. 1998. “The Political Relevance of Political Trust.” American Po-litical Science Review 92, no. 4 (December): 791–808.

Hill, Kim Quaile, and Patricia A. Hurley. 1999. “Dyadic Representation Reappraised.”American Journal of Political Science 43, no. 1 ( January): 109–37.

Hinich, Melvin, and Michael Munger. 1994. Ideology and the Theory of Political Choice.Ann Arbor: University of Michigan Press.

Huntington, Samuel P. 1991. The Third Wave: Democratization in the Late Twentieth Cen-tury. Norman: University of Oklahoma Press.

Jackisch, Carlota. 1997. “Representación democrática y sistemas electorales.” In Sistemaselectorales y sus consecuencias políticas, ed. Carlota Jackisch, 7–134. Buenos Aires: Cen-tro Interdisciplinario de Estudios sobre el Desarrollo Latinoamericano (CIEDLA).

Key, V. O. 1966. The Responsible Electorate: Rationality in Presidential Voting, 1936 –1960.Cambridge, MA: Harvard University Press.

Kitschelt, Herbert. 1990. “New Social Movements and the Decline of Party Organiza-tion.” In Challenging the Political Order: New Social Movements in Western Democracies,ed. Russell J. Dalton and Manfred Kuechler, 179–208. New York: Oxford UniversityPress.

———. 2000. Linkages between Citizens and Politicians in Democratic Politics. Com-parative Political Studies 33, no. 6/7 (August/September): 845–79.

Kitschelt, Herbert, Zdenka Mansfeldova, Radoslaw Markowski, and Gábor Tóka. 1999.Post Communist Party Systems: Competition, Representation, and Inter-Party Competition.Cambridge: Cambridge University Press.

Levine, Daniel H. 1992. Popular Voices in Latin American Catholicism. Princeton, NJ:Princeton University Press.

Levitsky, Steven. 2003. Transforming Labor-based Parties in Latin America: Argentine Pero-nism in Comparative Perspective. Cambridge: Cambridge University Press.

Lipset, Seymour Martin, and Stein Rokkan. 1967. “Cleavage Structures, Party Systems,and Voter Alignments: An Introduction.” In Party Systems and Voter Alignments: Cross-National Perspectives, ed. Seymour Martin Lipset and Stein Rokkan, 1– 64. New York:Free Press.


Lipset, Seymour Martin, and William Schneider. 1983. The Confidence Gap: Business, La-bor, and Government in the Public Mind. Baltimore: Johns Hopkins University Press.

Listhaug, Ola. 1995. “Confidence in Political and Private Institutions.” In Citizens andthe State, ed. Hans Dieter Klingemann and Dieter Fuchs, 298–322. Oxford: OxfordUniversity Press.

Luna, Juan P., and Elizabeth J. Zechmeister. 2005. “Political Representation in LatinAmerica: A Study of Elite-Mass Congruence in Nine Countries.” Comparative Politi-cal Studies 38, no. 4: 388– 416.

Mainwaring, Scott, and Aníbal Pérez-Liñán. 2005. “Latin American Democratizationsince 1978: Democratic Transitions, Breakdowns, and Erosions.” In The Third Waveof Democratization in Latin America: Advances and Setbacks, ed. Frances Hagopian andScott Mainwaring, 14 –59. Cambridge: Cambridge University Press.

———. Forthcoming. “Regional Effects and Region-Wide Diffusion of Democracy:Why Regions of the World are Important in Comparative Politics.” In Regimes and De-mocracy in Latin America, ed. Gerardo Munck. Oxford: Oxford University Press.

Mainwaring, Scott, and Mariano Torcal. 2006. “Party System Institutionalization andParty System Theory after the Third Wave of Democratization.” In Handbook of Po-litical Parties, ed. Richard S. Katz and William Crotty, 204 –27. London: Sage.

Manin, Bernard. 1997. Principles of Representative Government. Cambridge: CambridgeUniversity Press.

Manin, Bernard, Adam Przeworski, and Susan C. Stokes. 1999a. Introduction to De-mocracy, Accountability, and Representation, ed. Adam Przeworski, Susan C. Stokes, andBernard Manin, 1–26. Cambridge: Cambridge University Press.

———. 1999b. “Elections and Representation.” In Democracy, Accountability, and Repre-sentation, ed. Adam Przeworski, Susan C. Stokes, and Bernard Manin, 29–54. Cam-bridge: Cambridge University Press.

Mansbridge, Jane. 2003. “Rethinking Representation.” American Political Science Review97, no. 4 (November): 515–28.

Maravall, José María. 1999. “Accountability and Manipulation.” In Democracy, Account-ability, and Representation, ed. Adam Przeworski, Susan C. Stokes, and Bernard Manin,154 –96. Cambridge: Cambridge University Press.

Mayorga, René. 2005. “Bolivia’s Democracy at the Crossroads.” In The Third Wave of Democratization in Latin America: Advances and Setbacks, ed. Frances Hagopian and ScottMainwaring, 149–78. Cambridge: Cambridge University Press.

McCrone, Donald J., and James H. Kuklinski. 1979. “The Delegate Theory of Repre-sentation.” American Journal of Political Science 23, no. 2 (May): 278–300.

Miller, Warren E., and Donald E. Stokes. 1963. “Constituency Influence in Congress.”American Political Science Review 57, no. 1 (March): 45–56.

Mishler, William, and Richard Rose. 2001. “Where Are the Origins of Political Trust?Testing Institutional and Cultural Theories in Post-Communist Societies.” Compara-tive Political Studies 34, no. 1 (February): 30 – 62.

Newton, Kenneth, and Pippa Norris. 2000. “Confidence in Public Institutions: Faith,Culture, or Performance?” In Disaffected Democracies: What’s Troubling the TrilateralCountries? ed. Susan J. Pharr and Robert D. Putnam, 52–73. Princeton, NJ: Prince-ton University Press.

Nohlen, Dieter, ed. 1993. Enciclopedia electoral de América Latina y el Cáribe. San José,Costa Rica: Instituto Interamericano de Derechos Humanos.

Novaro, Marcos. 1994. Pilotos de tormentas: Crisis de representación y personalización de lapolítica en Argentina (1989–1993). Buenos Aires: Ediciones Letra Buena.


———. 1995. “Crisis de representación, neopopulismo y consolidación democrática.”Sociedad (April): 95–117.

O’Donnell, Guillermo. 1994. “Delegative Democracy?” The Journal of Democracy 5, no. 1( January): 55– 69.

———. 2003. “Democracia, desarrollo humano y derechos humanos.” In Democracia,Desarrollo Humano y Ciudadanía, ed. Guillermo O’Donnell, Osvaldo Iazzetta, andJorge Vargas Cullell, 25–147. Rosario, Argentina: Homo Sapiens Ediciones.

O’Donnell, Guillermo, and Philippe Schmitter. 1986. “Tentative Conclusions aboutUncertain Democracies.” Part 4 of Transitions from Authoritarian Rule: Prospects for De-mocracy, ed. Guillermo O’Donnell, Philippe Schmitter, and Laurence Whitehead. Bal-timore: Johns Hopkins University Press.

O’Neill, Kathleen. 2005. Decentralizing the State: Elections, Parties, and Local Power in theAndes. Cambridge: Cambridge University Press.

Pedersen, Mogens N. 1983. “Changing Patterns of Electoral Volatility in European Party Systems: Explorations in Explanation.” In Western European Party Systems: Continuity and Change, ed. Hans Daalder and Peter Mair, 29– 66. Beverly Hills, CA:Sage.

Peruzzotti, Enrique. 2004. “Crisis de representación política y sociedad civil en Ar-gentina.” Política y Gobierno 11, no. 1 (Spring): 43–72.

Pharr, Susan J., and Robert D. Putnam, eds. 2000. Disaffected Democracies: What’s Trou-bling the Trilateral Countries. Princeton, NJ: Princeton University Press.

Pierce, Roy. 1999. “Conclusion: Mixed Signals.” In Policy Representation in Western De-mocracies, ed. Warren E. Miller, Jacques Thomassen, Richard Herrera, Sören Holmberg,Peter Esaiasson, and Bernhard Wessels, 9–32. Oxford: Oxford University Press.

Pitkin, Hannah Fenichel. 1967. The Concept of Representation. Berkeley: University ofCalifornia Press.

Pizzorno, Alessandro. 1981. “Interests and Parties in Pluralism.” In Organizing Interests inWestern Europe: Pluralism, Corporatism, and the Transformation of Politics, ed. SuzanneBerger, 247–84. New York: Cambridge University Press.

Popkin, Samuel. 1991. The Reasoning Voter: Communication and Persuasion in PresidentialCampaigns. Chicago: University of Chicago Press.

Powell, G. Bingham, Jr. 2000. Elections as Instruments of Democracy. New Haven, CT: YaleUniversity.

Power, Timothy J., and Giselle D. Jamison. 2005. “Political Mistrust in Latin America.”Comparative Sociology 4, no. 1–2: 47–72.

Przeworski, Adam. 1975. “Institutionalization of Voting Patterns, or Is Mobilization theSource of Decay?” American Political Science Review 69, no. 1: 49– 67.

Przeworski, Adam, Susan Stokes, and Bernard Manin, eds. 1999. Democracy, Accountabil-ity, and Representation. Cambridge: Cambridge University Press.

Riker, William. 1982. Liberalism versus Populism. San Francisco: Freeman.Roberts, Kenneth M. Forthcoming. Changing Course: Parties, Populism, and Political Rep-

resentation in Latin America’s Neoliberal Era. Cambridge: Cambridge University Press.Roberts, Kenneth M., and Erik Wibbels. 1999. “Party Systems and Electoral Volatility

in Latin America: A Test of Economic, Institutional, and Structural Explanations.”American Political Science Review 93, no. 3 (September): 575–90.

Rogowski, Ronald. 1981. “Representation in Political Theory and in Law.” Ethics 91(April): 395– 430.

Samuels, David J., and Matthew Soberg Shugart. 2003. “Presidentialism, Elections, andRepresentation.” Journal of Theoretical Politics 15, no. 1: 33– 60.


Sani, Giacomo, and Giovanni Sartori. 1983. “Polarization, Fragmentation, and Compe-tition in Western Democracies.” In Western European Party Systems, ed. Hans Daalderand Peter Mair, 307– 40. Beverly Hills: Sage.

Sartori, Giovanni. 1976. Parties and Party Systems: A Framework for Analysis. New York:Cambridge University Press.

Schattschneider, Elmer E. 1942. Party Government. New York: Farrar and Rinehart.Schumpeter, Joseph. 1946. Capitalism, Socialism, and Democracy. New York: Harper

and Row.Schwartz, Nancy L. 1988. The Blue Guitar: Political Representation and Community. Chi-

cago: University of Chicago Press.Silveira, Flávio. 1998. A decisão do voto no Brasil. Porto Alegre: Edipucrs.Stimson, James, Michael B. MacKuen, and Robert S. Erikson. 1995. “Dynamic Repre-

sentation.” American Political Science Review 89, no. 3 (September): 543– 65.Stokes, Susan C. 1999. “What Do Policy Switches Tell Us about Democracy?” In De-

mocracy, Accountability, and Representation, ed. Adam Przeworski, Susan C. Stokes, andBernard Manin, 98–130. Cambridge: Cambridge University Press.

Thomassen, Jacques. 1999. “Political Communication between Political Elites and MassPublics: The Role of Belief Systems.” In Policy Representation in Western Democracies, ed.Warren E. Miller, Roy Pierce, Richard Herrera, Sören Holmberg, Peter Esaiasson, andBernhard Wessels, 33–58. Oxford: Oxford University Press.

Torcal, Mariano, and José Ramón Montero, eds. 2006. Political Disaffection in Contempo-rary Democracies: Social Capital, Institutions, and Politics. New York: Routledge.

Turner, Frederick C., and John D. Martz. 1997. “Institutional Confidence and Demo-cratic Consolidation in Latin America.” Studies in Comparative International Develop-ment 32, no. 3 (Fall): 65–84.

United Nations Development Programme. 2004. La democracia en América Latina: Haciauna democracia de ciudadanas y ciudadanos. Lima: United Nations Development Programme.

Warren, Kay B., and Jean E. Jackson, eds. 2002. Indigenous Movements, Self-Representation,and the State in Latin America. Austin: University of Texas Press.

Weissberg, Robert. 1978. “Collective vs. Dyadic Representation in Congress.” AmericanPolitical Science Review 72, no. 2 ( June): 535– 47.

Weyland, Kurt. 1999. “Neoliberal Populism in Latin America and Eastern Europe.”Comparative Politics 31, no. 4 ( July): 379– 401.

Williams, Melissa S. 1998. Voice, Trust, and Memory: Marginalized Groups and the Failingsof Liberal Representation. Princeton, NJ: Princeton University Press.

Wlezien, Christopher. 1996. “Dynamics of Representation: The Case of U.S. Spendingon Defence.” British Journal of Political Science 26: 81–103.

———. 2004. “Patterns of Representation: Dynamics of Public Preferences and Policy.”Journal of Politics 66, no. 1 (February): 1–24.

Wood, B. Dan, and Angela Hinton Andersson. 1998. “The Dynamics of Senatorial Rep-resentation, 1952–1991.” Journal of Politics 60, no. 3 (August): 705–36.

Zoco, Edurne. Forthcoming. “The Collapse of Party Systems in Italy, Peru, and Venezuela.” Ph.D. diss., University of Notre Dame.


Part I

❖

PARTY SYSTEMS, POLITICALOUTSIDERS, AND THE CRISIS OFDEMOCRATIC REPRESENTATION

Peru and Venezuela, countries with very different historical trajectories,shared similar political outcomes in the 1990s. Both their party systems collapsed,one result of exhausted statist economic policies and the implementation of first-generation market reforms, resulting in the subsequent establishment of au-thoritarian, though formally democratic, regimes under presidents Fujimori andChávez. Such an outcome was unusual. In the rest of Latin America, party sys-tems evolved and declining traditional parties coexisted with new parties, al-though not easily and in the middle of a crisis of representation. This gradualevolution of party systems, despite its limits, helped to maintain the checks andbalances inherent to democratic rule.

The explanation for the party system collapses is to be found less in structuralcauses or in the poor economic performance of both countries than in the re-sponses of political actors to challenges posed by crises of representation at criti-cal junctures, when the actors were especially vulnerable. The organization ofthe political parties was crucial in determining these responses; it explains the ap-pearance of sharp internal conflicts and divisions that accentuated problems ofrepresentation and enabled leaders outside the system to come to power throughthe electoral route. Both Alberto Fujimori (Peru, 1990 –2000) and Hugo Chávez(Venezuela, 1998–present) are personalistic leaders with neopopulist and anti-system discourses who expressed the popular dissatisfaction with traditional ac-tors. Both presidents were relatively effective in dismantling the preexisting po-litical order through institutional reforms that, although formally democratic, inpractice created authoritarian governments. These leaders represented some pre-viously excluded popular sectors, not under democratic, but under plebiscitaryschemes. The collapse of the party systems ended the political balance that hadexisted in both countries, and new state institutions were created under thehegemony of a single political actor, which led to a concentration of power that

2

From Crisis to Collapse of the PartySystems and Dilemmas of DemocraticRepresentation: Peru and Venezuela

❖

Martín Tanaka

48 Tanaka

ultimately ended the checks and balances inherent to democratic rule. These kindof regimes leave an onerous heritage that confronts both countries with enor-mous challenges: how to rebuild state institutions when the political and socialactors are greatly weakened; and how to construct a new system of democraticrepresentation that is pluralistic and participatory, while at the same time ensur-ing governability in crisis contexts in which the expectations and demands of thepopulation are high.

Peru and Venezuela: Different Trajectories, the Same Results, and Some Explanations

From very different trajectories, Peru and Venezuela confronted crises, markedby the exhaustion of statist economic policies and the implementation of the firstgeneration of market reforms, which devastated the entire region in the 1980sand 1990s. Venezuela, with a relatively long democratic history dating from theend of the 1950s, exhibited a stable party system that featured cooperative be-haviors and centripetal political competition. The parties were representative,firmly established in society, with links to various spheres of civil society. Theirrootedness had its source in a long period of economic growth, which fosteredsocial integration and the civic involvement of excluded sectors. Peru, on theother hand, revived an always shaky democratic experiment in 1980, after twelveyears of a military government that had carried out profound structural changes.The fragile party system that emerged from the transition faced great challenges.This was a highly ideologized system, interacting with social movements and or-ganized interest groups strong enough to apply political pressure. To complicatematters further, the democratic experiment coincided with the beginning of thearmed conflict unleashed by two terrorist groups, the Sendero Luminoso (Shin-ing Path) and the MRTA (Movimiento Revolucionario Tupac Amaru, or TupacAmaru Revolutionary Movement), the first of which was particularly dogmaticand bloodthirsty.1

Despite these differences, Peru and Venezuela ended up with the same result:the sudden collapse of their party systems and the subsequent establishment ofauthoritarian regimes. In Peru, the collapse of the party system occurred be-tween 1989 and 1992; in Venezuela, between 1998 and 2000. In Peru, the voteshare of the four political groups that had garnered more than 90 percent of thevote in most elections in the 1980s fell to 71.8 percent in the municipal electionsof 1989, and to 68.2 percent in the 1990 presidential contest, when Alberto Fu-jimori was elected. It continued to fall until the parties became virtually extinctpolitically (see Table 2.1).

In Venezuela, the breakdown was even more rapid and is more surprising,given the apparent consolidation of the party system. Democratic Action (Ac-ción Democrática, or AD) and the Christian Democratic Party (COPEI) were

From Crisis to Collapse: Peru and Venezuela 49

the hegemonic actors in presidential contests until the presidential election of1993, when the winner was Rafael Caldera, who headed the Convergencia Na-cional (National Convergence) coalition. In 1993, the combined vote total forthe AD and COPEI presidential candidates was 46 percent. The collapse in the1998 presidential contest won by Hugo Chávez was spectacular: AD and COPEIdid not even run their own candidates. Instead, they backed the independentcandidacy of Henrique Salas Römer; their contribution to his 40 percent totalwas only 11 percent. Nor did they run candidates in the 2000 presidential con-test, when Chávez won again, this time under a new constitution (see Table 2.2).

These are disconcerting outcomes. In Peru, it was generally expected that theconflicts and problems of the 1980s would give rise to increasing ideological po-larization, setting off another military intervention (a scenario like Chile’s for theperiod 1970 –73). Instead, an “outsider” brought an end to the existing politicalorder, a result anticipated by no one. The expected outcome was polarizationand ungovernability, but what actually occurred was a grave crisis of representa-tion. In Venezuela, the crisis in bipartism could have led to a scenario like theone in Colombia, where the traditional parties have undergone a progressive de-cline, with increasing internal fragmentation. New parties have emerged and de-veloped in this context, and the old two-party system has been replaced by amoderate multiparty system. In contrast, in Venezuela, an “outsider” quickly didaway with an order that had seemed consolidated. Clearly, Peru and Venezuela

Table 2.1

Peru: Vote Percentages for the Major Political Parties, 1978–2000

AP PPC AP�PPC APRA LEFT (IU) Total

1978 (C) 23.80 23.80 35.30 29.40 88.501980 (P) 45.40 9.60 55.00 27.40 14.40 96.801980 (M) 35.80 11.10 46.90 22.50 23.30 92.701983 (M) 17.50 13.90 31.40 33.10 29.00 93.501985 (P) 7.30 11.90 19.20 53.10 24.70 97.001986 (M) NP 14.80 14.80 47.60 30.80 93.201989 (M) – – 31.20 20.40 20.20 71.801990 (P) – – 32.60 22.60 13.00 68.201992 (C) NP 9.70 9.70 NP NP 9.701993 (M) 11.60 5.70 17.30 10.80 3.90 32.001995 (P) 1.64 NP 1.64 4.11 0.57 6.301998 (M) 5.00 NP 5.00 7.00 NP 12.002000 (P) 0.40 NP 0.40 1.40 NP 1.80

SOURCE: Tuesta (2001).notes: The 1978 and 1992 elections were for constitutional assemblies (C). The 1980, 1985, 1990, 1995,

and 2000 elections were for the presidency (P). The elections of 1980, 1983, 1986, 1993, and 1998 were mu-nicipal elections (M). The “Total” column is the combined vote for the Peruvian Aprista Party (APRA), thePopular Christian Party (PPC), Popular Action (AP), and the United Left (IU). For 1978, we are treating as IUvotes the total vote for the parties that formed the IU coalition in 1980. In 1989 and 1990, the AP and PPCvotes are part of the Democratic Front (FREDEMO) vote.

NP: No participation; did not participate in election.

50 Tanaka

constitute exceptions to the regional scenario of the 1980s and 1990s. In all othercases, party systems survived, despite problems of legitimacy and representation,with high electoral volatility and institutional instability (in some cases involvingthe removal of presidents by non-constitutional means, as in Ecuador) leading toa slow mutation toward a new system in which new parties exist side by side withtraditional ones.2 Why didn’t Peru and Venezuela go down this road? Why andhow did they pass from crisis to collapse of the party system and the prevailinginstitutional order? Why and how did two countries with such different politi-cal trajectories end up with the same result?

On the surface, it might appear easy to explain what happened in these coun-tries by pointing to structural and economic variables and to the obvious inca-pacity of the various political actors to deal successfully with the challenges theyfaced. In this perspective, the parties alternated in power, failed to solve theproblems, and were discredited, and hence voters sought options outside thesystem. In Peru, the election of Alberto Fujimori was preceded by a severe eco-nomic recession, high rates of inflation, and extremely high rates of politicalviolence (see Table 2.3), which damaged the legitimacy of all the principal par-ties. In Venezuela, the 1980s were very bad years, especially the year 1983 (un-der the administration of COPEI president Herrera), the period from 1988 to1990 (under the second administration of President Carlos Andrés Pérez), andalmost the entire second administration of President Rafael Caldera (1994 –98),who won election as the head of an independent movement, having left COPEIwhen he failed to secure its presidential nomination (see Table 2.3). Thisdifficult period is said to be the cause of a progressive crisis of representationleading to a steady drop in party identification and growing political disaffection

Table 2.2

Venezuela: Vote Percentages for Presidential Elections, 1973–1998

Party 1973 1978 1983 1988 1993 1998 2000

AD 48.7 43.3 58.4 52.9 23.6 – –COPEI 36.7 46.6 33.5 40.9 22.7 – –AD � COPEI 85.4 89.9 91.9 93.8 46.3 – –MAS 4.3 5.2 3.5 2.7 – – –La Causa R – – 0.1 0.3 22.0 0.1 –Convergenciaa – – – – 30.5 – –Polo Patriótico/MVRb – – – – – 56.2 59.5Proyecto Venezuelac – – – – – 40.0 –Others 10.3 4.9 4.5 3.7 1.2 3.8 40.5

SOURCE: Roberts 2003.a Electoral coalition for Rafael Caldera’s campaign; included MAS.b Electoral coalition for Hugo Chávez’s campaign; included MAS.c Electoral movement organized by independent candidate Henrique Salas Römer; AD, COPEI, and other

minor parties contributed to its vote totals.

(see Table 2.4). Supposedly, these factors explain why someone like Chávezcould come to power.

Although the economic crises narrowed the margins of possibility and optionsavailable to the political actors in both countries, economic performance alonecannot explain the collapse of these two party systems. Other countries in the re-gion passed through similar or worse economic experiences, which also createdcrises of representation, yet their party systems managed to survive. Crucial to anunderstanding of the collapse of the party systems are the political actors’ deci-sions, especially those made at junctures when the decision makers were at theirmost vulnerable (in Peru, the 1990 election; in Venezuela, the 1998 election). Ina situation of a crisis of representation, change, and high vulnerability, internalconflicts led the parties into processes of division, which enabled “outsiders,”anti-system and anti-party caudillos, to win power through the electoral route.The relative consolidation of these new leaders enabled them to overthrow theprevailing order and replace it with a new order with authoritarian tendencies(see Mayorga, this volume).


Table 2.3

Peru and Venezuela: Inflation, GNP, and Subversive Acts, 1980 –1999

Venezuela

Peru Subversiveacts: National GNP

Annual rate GNP growth police growthYear of inflation rate records rate

1980 60.8 4.4 219 �3.81981 72.2 4.3 715 �1.01982 72.9 0.3 891 �1.61983 125.1 �11.8 1,123 �5.51984 111.5 4.7 1,760 �1.51985 158.3 2.3 2,050 01986 62.9 8.7 2,549 6.61987 114.5 8.0 2,489 3.81988 1,722.3 �8.4 2,415 5.91989 2,775.3 �12.9 3,149 �8.81990 7,649.7 �5.4 2,779 5.51991 139.2 2.8 2,785 9.71992 56.7 �0.6 3,002 7.11993 39.5 6.0 1,918 �0.51994 15.4 13.6 1,195 �3.71995 10.2 8.6 1,232 5.91996 11.8 2.5 883 �0.41997 6.5 6.8 681 7.41998 6.0 �0.4 474 0.71999 3.2 1.4 168 �5.8

SOURCES: Peru: INEI (Instituto Nacional de Estadística e Informática); Venezuela: Anuario estadístico deAmérica Latina y el Caribe (Santiago: CEPAL, 2001).

Collapse of the Party System in Peru: Polarization, IntrapartyConflicts, and Crisis of Representation

In previous work on Peru (Tanaka 1998, 2005), I argue that the breakdown inthe party system resulted not so much from the performance of the political ac-tors throughout the 1980s as from their actions starting near the end of 1988,when inflation had accelerated and the country had entered into a dynamicmarked by the 1989–90 elections.3 Despite the complicated situation, nothingportended that in 1990 a grave crisis of representation would develop and that inthe succeeding years the party system would collapse. On the contrary, both theanalysts and the actors themselves perceived that the principal danger lay in thegrowing polarization of party members, the abandonment of the political centerthat accompanied the crisis of the ruling American Popular Revolutionary Al-liance (Alianza Popular Revolucionaria Americana, or APRA), and the strength-ening of extremist actors. These trends led to serious problems of governability.In the context of the threat posed by Sendero Luminoso, this situation couldhave led to a repressive military intervention. Until 1989, the parties seemed rel-atively strong, with possibilities for recovery in the not too distant future.

At the extremes were the United Left (Izquierda Unida, or IU) on the left andthe Democratic Front (FREDEMO) on the right. From its founding in 1980, theIU, a political front formed by seven leftist organizations, had steadily increasedits electoral clout, winning more than 30 percent of the vote in the 1986 munic-ipal elections. Several 1987 opinion polls indicated that Alfonso Barrantes, bythat time the most likely IU candidate, would be the top choice among voters inthe 1990 presidential race. In late 1987, the IU called its First National Conven-tion for September 1988 (although, in the end, it was held in January 1989), tofine-tune the organization and its strategy for coming to power through the elec-toral route. At this convention, the Front would adopt rules, policy, and platformguidelines, formulate a plan for immediate political action, and choose a unifiedpolitical leadership. As for the right, Popular Action (Acción Popular, or AP) andthe Popular Christian Party (Partido Popular Cristiano, or PPC) suffered harsh

52 Tanaka

Table 2.4

Venezuela: Trends in Party Identification

Polling Member/ Independent/organization Sympathizer Not interested

Baloyra 1973 48.6 51.2BATOBA 1983 38.1 61.3CIEPA 1993 29.4 66.4REDPOL 1998 14.2 61.0

SOURCE: González (2002).

setbacks in the April 1985 general elections, after the second administration ofFernando Belaúnde (1980 –85). But by August 1987, they had taken the politi-cal initiative once again, heading the opposition to President Alan García’s pro-posal to nationalize the banking system. The rightist block then underwent asignificant revitalization. August 1987 saw the birth of the Liberty Movement(Movimiento de Libertad, or ML), led by the writer Mario Vargas Llosa and theeconomist Hernando de Soto, which promoted market-oriented ideas and statemodernization. January 1988 saw the formation of a major alliance involving theML, AP, and the PPC—the Democratic Front (FREDEMO). In the November1989 municipal elections, FREDEMO emerged as the country’s main politicalgroup, and opinion polls showed that Mario Vargas Llosa was likely to be Peru’snext president.4

By 1989, the Peruvian political scene was highly polarized. On one pole wasa left with revolution in mind, with a kind of electoral path to socialism similarto the one followed by the Popular Unity Front (Unidad Popular, or UP) in Al-lende’s Chile (1970 –73). On the other pole, the right advocated a liberal ideol-ogy and a profound modernization of the economy and the state within theframework of a market economy. Given the ideological polarization of theseprograms, the triumph of either the left or the right would have created prob-lems of governability. What occurred, unexpectedly, was a crisis of representa-tion: radicalized political groups abandoned the political center formerly occu-pied by APRA, and the empty space was filled by an “outsider.” Such an unusualand unexpected outcome is understood by analyzing the 1990 election campaignand the conflicts within the parties.

The campaign was marked by a deep recession, hyperinflation, and high lev-els of political violence. In 1989, the Sendero Luminoso announced that it hadarrived at a “strategic balance” with the forces of order—the stage prior to a“strategic offensive” that would lead to the seizure of power—and it began a“siege” of Lima. In this context, internal conflicts within the major parties led toopen struggles and divisions, leading a sector of the electorate to seek other op-tions outside the system. How can we understand the actors’ behavior? The con-text of crisis and violence, coupled with the (correct) perception that here wasan extreme situation involving the end of one political cycle and the chance tostart another, led the actors to abandon risk-averse behaviors, to be audacious,and to make decisions marked by ideological reasoning rather than pragmatism.Such conduct intensified the contradictions and internal conflicts among theprincipal actors and produced the vacuum of representation that the hithertounknown Fujimori took advantage of.

Around 1987 the IU was in need of a profound reorganization. Until then, de-spite its electoral gains, the IU had functioned mainly as a coalition of parties,which were represented by the general secretaries of the various participatingparties in a National Executive Committee (CDN), where each party maintained


its own political line. The IU’s internal problems grew more acute during the ad-ministration of Alan García, whose populist and revolutionary rhetoric createdproblems of identity and strategy. Barrantes, IU chairman until May 1987, hadmaintained a stand of “critical collaboration” with the García administration. In1987 Barrantes resigned his post because he did not have the backing of the ma-jority of the parties’ general secretaries, who espoused a much firmer oppositionline toward the García government. A clear, unified course of conduct was ur-gently needed, and that is why the first national convention was called. After anintense and interesting period of preparation, which saw the enrollment of morethan 130,000 members, an extremely high figure by Peruvian standards, the con-vention was held. But far from fostering the consolidation of the IU, it initiateda tortuous process of division.

On one side of the debate, aligned with Alfonso Barrantes, were those who be-lieved that to win elections and fashion a minimally stable and successful govern-ment it was essential to exclude the IU’s radical sector. The radical sector had notclearly rejected armed struggle and thus would make it impossible to surmount aveto by the armed forces and conservative sectors. On the other side of the de-bate were the parties of the “Revolutionary Block” (Bloque Revolucionario),5

which believed that the seeds of revolution were already present, making it ap-propriate to prepare for a large-scale political and possibly military confrontation.Accordingly, the real objective was not to arrive at a government through elec-tions but to prepare for taking power through insurrection. So, on one side werethose outlining a reformist program, broad in scope and appealing to the averagevoter; on the other were those propounding a strengthening of the parties’ bases,of strategic sectors, and a digging-in to prepare for the coming confrontation. Inthe middle of this controversy were the Peruvian Communist Party (PCP) andthose independent IU activists who had no party allegiance. The breakup of theIU, amid mutual recriminations and accusations, unfolded between January andOctober 1989 (the month in which candidates for the 1990 elections had to for-mally declare their intentions) and ruined the left’s electoral chances.6 In the 1990elections the left divided, presenting two presidential candidates. The IU candi-date, Henry Pease, polled 8.2 percent of the vote, while Alfonso Barrantes, can-didate for the newly created Socialist Left (Izquierda Socialista, or IS), won only4.7 percent.

The crisis of the left increased the electoral chances of the right. Throughoutmost of 1989, with the collapse of the left, presidential opinion polls indicatedthat Mario Vargas Llosa was the favorite. In the second half of 1989 and the earlymonths of 1990, the question was whether or not Vargas Llosa would get themore than 50 percent of the vote needed to win in the first round. FREDEMO,however, had its own internal problems. The leadership of Vargas Llosa andMovimiento Libertad within the alliance generated jealousies and rivalries in APand PPC. This friction came to a head in June 1989 when the FREDEMO

54 Tanaka

strategy for the November municipal elections became the subject of so muchdebate that Vargas Llosa tendered his resignation as a presidential candidate, aresignation he later withdrew.7

In spite of these problems, FREDEMO had a fairly good showing in the No-vember 1989 municipal elections. Although those elections witnessed the ap-pearance of the first “independent” candidates, who made manifest a delegiti-mation of the major parties (see Table 2.1), most of these independents werealigned with the major parties.8 The polarization and sense of urgency in thecountry affected FREDEMO and its campaign strategy, which makes it easier tounderstand why Vargas Llosa did not come up with a more conclusive victoryin the first round of the 1990 election (he won only 32.6 percent of the vote).Vargas Llosa distanced himself from the median voter with a fairly ideologicalcampaign, seeking a “clear mandate” to go ahead with profound neoliberal re-form. This campaign did not inspire enthusiasm in the electorate, especially af-ter the popular mobilization against neoliberal reforms in Caracas in February1989, under the administration of Carlos Andrés Pérez.

The crisis and chaos into which the government plunged seriously damagedAPRA’s electoral chances, yet APRA could not be completely written off. In the1989 municipal elections, APRA remained the second largest party at the na-tional level, behind FREDEMO and slightly ahead of the left candidates (seeTable 2.1). APRA’s candidate, Luis Alva Castro, won 22.5 percent of the vote inthe 1990 presidential election. But APRA, too, had internal problems that de-creased its electoral chances. According to the 1979 Constitution, Alan Garcíacould not seek reelection, and his efforts between 1987 and 1988 to pass a con-stitutional reform allowing him to run ended in failure. As a result, the generalsecretary of the party, Luis Alva Castro, competed with García for control ofAPRA. García fought to maintain control, and he decided to maintain his dis-tance from Alva Castro. Throughout most of the campaign, García gambled onleading the opposition to Mario Vargas Llosa’s candidacy (once again, for ideo-logical reasons) and backed Alfonso Barrantes rather than the APRA candidate.

The division of the left, FREDEMO’s internal problems and the extreme ide-ologization of its campaign, and the weakness of an APRA candidate who hadto assume the costs of the failures of García’s administration without receiv-ing the benefits of support from the top—all coming at an especially criticalmoment—created a vacuum of representation. The political center, havingbeen left more or less vacant, was subsequently occupied by a candidate who hadnot even been mentioned in the surveys until a few weeks before the election.9

From among the group of “minor” candidates, Alberto Fujimori, the formerrector of the Universidad Agraria (Rural University), suddenly turned out to bean attractive option. Once Fujimori began to rise in the opinion polls, Garcíabegan to support him through his connections in the press, and his support wasdecisive. With Garcia’s support, just a few weeks before the election, Fujimori


ceased to be a minor candidate and he went on to place second in the contest.10

In that first round, Vargas Llosa won with 32.6 percent, and Fujimori, surpris-ingly, came in second with 29.1 percent. In round two, with the votes of APRAand the left, Fujimori won the presidency, with 62.4 percent, compared to Var-gas Llosa’s 37.6 percent.

Once in office, Fujimori found himself with a minority in Congress. In theelections for the Senate and the Chamber of Deputies, FREDEMO obtained32.3 percent and 30.1 percent, respectively; Cambio 90 (Change 90), only 21.7percent and 16.5 percent; APRA, 25.1 percent and 25 percent (that is, it toppedCambio 90 in both houses); IU, 9.8 percent and 10 percent; IS, 5.5 percent and5.3 percent. Cambio 90 won only 32 of the 180 seats in the Chamber of Deputies,and in the Senate, only 14 out of 62. FREDEMO had 63 deputies and 21 sen-ators; APRA, 53 deputies and 17 senators (again, more than Cambio 90); IU,16 deputies and 6 senators; IS, 4 and 3. Fujimori had no possibility of aspiringto reelection in 1995 because reelection was prohibited by the 1979 Con-stitution. Therefore, his presidency was perceived as a singular episode, certainlyephemeral, and once it was over, the parties would again occupy center stage.Things turned out quite differently.

Collapse of the Party System in Venezuela: Crisis, Intraparty Struggles, and a Crisis of Representation

In the Venezuelan case, internal struggles and processes of division within themajor parties, in a situation in which they were particularly vulnerable, again ex-plain the unexpected rise to power of an “outsider” who went on to destroy theprevailing political and institutional order. The behaviors within the parties werea consequence of internal structures so highly disciplined that losing factionshad no space to air their differences and had incentives to break away. Internalstruggles and divisions within the Venezuelan parties are a tradition.11 Suchconflicts were a consequence of the way the Venezuelan parties were structured.They were extremely hierarchical and disciplined (Crisp 2001). This structurecreated obstacles to airing factional disputes openly, and thereby stimulated partydivision. The factions that lose internal disputes are weakened, without majoraccess to top party posts or candidacies in popular elections. As a result, they aretempted to try their luck outside the party apparatus. Article 185 of the 1961Constitution, by allowing for the reelection of a president after ten years, createdthe figure of the great caudillo waiting for the chance to make a comeback. Thus,former leaders do not disappear, but grow weak and then reappear. This phe-nomenon helps explain the return to the presidency of Carlos Andrés Pérez in1989 and of Rafael Caldera in 1993, both of whom were greatly weakened po-litically. Such comebacks are crucial in understanding Hugo Chávez’s rise topower in 1998.

56 Tanaka

An interesting comparison can be made with Peru, where the parties weretraditionally organized around an unchallengeable caudillo, but where a certainspace for factional disputes also existed, as long as they did not challenge themaximum leader. Víctor Raúl Haya de la Torre and then Alan García in APRA,Fernando Belaúnde in AP, Luis Bedoya in the PPC, and to some extent AlfonsoBarrantes in IU between 1980 and 1986, were the indisputable leaders of theirpolitical groups. Under them, however, disputes could develop, and these wouldbe internally arbitrated by their parceling out of posts and benefits, which madefor a measure of internal equilibrium. Problems cropped up when, in the con-text of the 1990 election campaign, all the main actors were simultaneously leftwithout adequate mechanisms for handling internal difficulties. Internal con-flicts ripped apart the IU when Barrantes lost his role as the arbiter of differ-ences. The right, joining up with FREDEMO in an alliance among equals, hadno mechanisms for internal arbitration, although each component party hadways of dealing with its own internal conflicts. Lastly, APRA was also experi-encing serious conflicts. García could not stand for reelection, and presidentialcandidate Luis Alva competed against him for control of the party.

In Venezuela, despite repeated internal conflicts, the two-party system hadfunctioned until 1993, with AD and COPEI alternating in power. In the secondhalf of the 1980s, however, the parties had lost legitimacy and the capacity to rep-resent the citizenry, as a result of the nation’s poor economic performance and thegovernment’s difficulties in confronting problems associated with the debt crisis.Not that the parties did nothing to confront the situation. In 1984, the Presiden-tial Commission for the Reform of the State (COPRE) was created. COPRE un-dertook substantial institutional changes. Among them were the introduction ofdirect popular elections of state governors (1989) and the Organic Law of Mu-nicipal Regimes, which instituted direct elections for mayors (1989). Previously,governors and mayors had been appointed. These changes were in line with thedecentralization process that originated in the 1970s. Other changes included theestablishment in 1993 of the mixed formula (proportional personalized vote) forelecting deputies to the National Congress and to legislative assemblies. These re-forms sought to open up a political system perceived as tightly closed (character-ized as a “partyocracy”), renew the leadership, and loosen the control of partybosses. The reform efforts only partially fulfilled their objective of improvedrepresentation, and they increased internal party tensions and conflicts, whichin the long run were decisive factors in the collapse of the party system. Thechanges also coincided with an economic crisis and reform. The concomitanttensions produced by party reforms and economic crisis explain Chávez’s com-ing to power.12

The implementation of market-oriented reforms really began during the sec-ond administration (1989–93) of Carlos Andrés Pérez, who spoke eloquently ofa “great turnaround.” Pérez faced a difficult financial situation that demanded


adjustment measures, which had long been postponed throughout the 1980s.The public-sector deficit was 9.9 percent of GNP in 1988, with a current-account deficit of 4.9 billion dollars, a situation without precedent in the 1970sand 1980s. In addition, the price of oil had fallen to around 13 dollars a barrel thatyear, after averaging 33 dollars in 1985, and the drop brought much uncertaintyto a country whose income is largely based on oil (Naím 1993; Hidalgo 2000).Carlos Andrés Pérez also faced problems in his relations with his own party. Hehad won the nomination, two terms after his first administration, by prevailingin a tough struggle against Octavio Lepage, who was backed by the outgoingpresident, Jaime Lusinchi. Pérez had always based his power within AD on hischarisma and his ability to communicate with the citizenry at large, rather thanon his role as a bureaucratic executive. This situation makes it easier to under-stand the first decision of his second administration: the formation of a govern-ing team made up of independent figures so as to secure a margin of maneuver-ability in the face of pressure from the party and interest groups. Engaged inrebuilding his political leadership with a new political base, he undertook neo-liberal economic reforms, which entailed reorienting AD’s traditional politicalidentity as a party with a statist economic policy. Pérez’s gambit is not surprising:an interesting parallel exists with Mexico’s president Carlos Salinas de Gortari(1988–94) and his policy of modernization, market reform, and relative distanc-ing from the “dinosaurs” of the PRI (Partido Revolucionario Institucional, or In-stitutional Revolutionary Party). Pérez also had the example of President VictorPaz Estensoro’s “New Political Economy” in Bolivia, launched in 1985. AD wasa bureaucratized structure, and in many instances a corrupt one, accustomed toclientelistic and corporate relationships and to financial favors. Pérez had strongincentives to remake AD under a modernizing leadership more in tune with theneed for structural reform that was becoming evident in the region.

The policy of the “great turnaround” generated a massive wave of sponta-neous protests in February 1989, especially in Caracas (Kornblith 1998; LópezMaya 2000). There was a feeling of indignation over the new policies, launchedby someone who had won the election by exploiting the image of a return tothe prosperous days of the 1970s. As has been explained by several analysts ofthe adjustment process, the citizenry did not perceive that a severe adjustmentwas necessary and inevitable in Venezuela as it had been in countries withdeeper recessions and hyperinflation (Corrales 2000; Roberts 2003; Weyland2002). This popular resentment led to the increasing isolation of the Pérez gov-ernment, even within AD. This was expressed, for example, by the distancingof the Confederation of Venezuelan Workers (CTV), the main labor confeder-ation, which in May 1989 called for a work stoppage to protest the government’seconomic policies, marking the first time the CTV had taken action against anAD government. But that was just one expression of an even greater distancing.The traditional party apparatus, already dealt a blow when Pérez defeated Lepage

58 Tanaka

for the nomination, felt left out of government decisions. Pérez intended to pur-sue neoliberal reform, a radical about-face from AD’s traditional populist poli-cies. Pérez thus lost support within his party, including among AD legislators,and this ultimately explains his removal from office by Congress in 1993.

Many analysts have argued that the social costs of the “great turnaround” de-stroyed the Pérez government’s opportunities. There is not much evidence, how-ever, to support that position. As the data in Table 2.4 show, the economy fell intoa deep recession in 1989, a result of the adjustment program, but growth recov-ered during the remainder of Pérez’s term. Why was Pérez unable to make po-litical capital out of a recovery, as did Fujimori and other leaders of successfulstabilization efforts such as Paz Estensoro in Bolivia and Carlos Menem in Ar-gentina? In the Venezuelan case, the interparty and intraparty conflicts are just asmuch a key to understanding Carlos Andrés Pérez’s fall as is economic perfor-mance. The existing discontent and conflict among and between Venezuela’s po-litical parties had devastating political effects. Failed coup attempts in Februaryand November of 1992, nevertheless generated sympathy among the popular sec-tors. This sympathy grew and spread because of the stands taken by political lead-ers from AD and the opposition. The political space acquired by Hugo Chávezwas in large measure handed to him by the parties and their caudillos with theiropportunistic stances. After the coup attempts, eminent politicians, among themformer COPEI president Rafael Caldera, far from condemning the participantsand defending the constitutional order, declared support for the insurrectionists.Their support for the coups further weakened the party system and exacerbatedthe unrealistic expectations of the citizenry and the hope for a “redeeming” lead-ership that would finish off the “old order” and bring prosperity to the country.13

AD, too, kept its distance from the Pérez government instead of defending it.Congress sought to hamper the administration and ultimately to remove it. InMay 1993, barely three months before the end of his term, Carlos Andrés Pérezwas removed from office by Congress after impeachment proceedings based ona dubious accusation of improper use of public funds, and Ramón J. Velasquez be-came the interim president. These events discredited the political system, whichfell captive to particularistic, narrow interests.14

The 1993 elections caught AD considerably weakened after the scandals con-nected with the dismissal of President Pérez. Nevertheless, there was not theslightest foreshadowing of the party system collapse that occurred in 1998. Ascan be seen in Table 2.2, despite an enormous drop in AD’s vote as comparedto 1988, the party’s candidate, Claudio Fermín, came in second behind Caldera.Fermín’s rise to leadership was a result of the renewal generated by the de-centralization process and the COPRE reforms. Fermín built his reputation onhis good management as mayor of Caracas. This new type of leadership hadsome complications, and it faced opposition from the traditional party appara-tus. To secure the presidential nomination, Fermín competed against Luis Alfaro


Ucero and Héctor Alonso López (who was backed by Carlos Andrés Pérez). Inthe 1993 congressional elections, AD, although dropping significantly in com-parison with previous elections, remained Venezuela’s largest party, clearly top-ping Convergencia and Radical Cause (La Causa R, or LCR) (see Table 2.5).

AD’s internal conflicts had helped COPEI come to power in 1968 and 1978,in a system in which these two parties alternated in power. Something differentoccurred in the 1993 election because COPEI too was seriously damaged by in-ternal conflict. Rafael Caldera failed to capture his party’s presidential candidacyin 1988, and he distanced himself from the candidacy of Eduardo Fernández.Then, in 1992, Caldera did not condemn the coups; on the contrary, he rode thewave of sympathy aroused by the perpetrators and joined in the criticism of thetraditional order and the neoliberal economic policy, adopting a populistdiscourse. For the 1993 elections, Caldera formed a new political group, Con-vergencia Nacional, an alliance that elevated him to the presidency once again,although he had earlier encouraged, within COPEI, the candidacy of OswaldoAlvarez against that of Eduardo Fernández. The exit of Caldera, the partyfounder, was a harsh blow for COPEI, and it ultimately meant the end of the sys-tem of the two parties alternating in power.15 Despite all its problems, COPEI’sshowing in 1993, like AD’s, did not herald as inevitable its steep demise in 1998.Although COPEI suffered a sharp drop in comparison with its 1988 electoral re-sults, it remained the second largest party in Venezuela’s Congress. The party’spresidential candidate, Oswaldo Alvarez, came in third in the presidential elec-tion (see Tables 2.2 and 2.5). Oswaldo Alvarez had built his political leadershipby serving as governor of the state of Zulia. Just as with Fermín in AD, de-centralization resulted in a renewal in political leadership but also exacerbated in-ternal conflicts in a party with a hierarchical tradition.

60 Tanaka

Table 2.5

Venezuela: Percentage of Seats in the Chamber of Deputies, 1973–2000

Party 1973 1978 1983 1988 1993 1998 2000

AD 51.0 44.2 56.5 48.3 27.6 29.1 18.8COPEI 31.8 42.2 30.0 33.3 27.1 14.3 4.2AD � COPEI 82.8 86.4 86.5 81.6 54.7 43.4 23.0MASa 4.5 5.5 5.0 9.0 – 10.6 12.7LCR – – – 1.5 20.1 3.2 2.4Convergencia – – – – 25.1 1.6 .1MVR – – – – – 25.9 46.1Proyecto

Venezuela – – – – – 12.7 4.2Others 12.5 8.1 8.5 7.9 – 2.6 9.1

SOURCES: Roberts 2003, 253; data for 2000 elections taken from Payne et al. 2003.a MAS deputies are regarded as part of Convergencia.

Tables 2.2 and 2.5 show that the vote lost by AD and COPEI made pos-sible both the emergence of Convergencia and the growth of LCR, a leftistmovement of union origin with bases in various regions of the country (López Maya 1997, 2001). As of 1993, Venezuela seemed to be embarking on a path of evolution from a traditional two-party system toward a moderate multipartismin which AD and COPEI would coexist with new parties. That is why the col-lapse of the party system cannot be readily inferred from a crisis of representa-tion or problems of legitimacy. Although party identification was clearly on thedecline, paralleled by an increase in the number identifying themselves as inde-pendents or as disaffected with politics (see Table 2.4), citizen preferences couldhave followed a pattern similar to that of Colombia, Ecuador, or Bolivia.

That did not happen. To understand why, it is crucial to analyze the 1998election campaign and, again, to look at how intraparty conflicts simultaneouslyruined the chances of all the actors in the system. This is what allowed an “out-sider” to come to power. The emerging parties, Convergencia and LCR, did notconsolidate themselves between 1993 and 1998. Convergencia paid the price forbad governmental performance. As can be seen in Table 2.3, economic perfor-mance between 1993 and 1998 was poor.16 The administration’s shortcomingsalso hurt Movement to Socialism (Movimiento al Socialismo, or MAS), whichwas part of the governing coalition. In the 1998 presidential election, Conver-gencia did not nominate its own candidate but backed Irene Sáez, who polledbarely 2.82 percent of the vote; in Congress, only three Convergencia deputiesand two senators won seats. MAS suffered a schism in the 1998 election. On oneside was the sector close to the Caldera government and the minister of plan-ning, Teodoro Petkoff; on the other was the minister’s critics (Leopoldo Puchiand Felipe Mujica), who ended up backing Hugo Chávez. MAS contributednine points to Chavéz’s 56.2 percent total and elected twenty-two deputies andsix senators.

As a leftist movement that was not part of the “traditional” order, LCR mighthave been the most obvious contender to fill the space vacated by Convergen-cia and MAS. LCR, however, also split before the 1998 election. One sectorchallenged Andrés Velásquez’s leadership and exited in April 1997 to found anew movement, Fatherland for All (Patria Para Todos, or PPT), under the di-rection of Pablo Medina. LCR nominated a presidential candidate, AlfredoRamos, who polled barely 0.1 percent of the vote, and it elected one senator andsix deputies. PPT backed Chávez, bringing him 2.19 percent of the vote, and itelected seven deputies and one senator.

COPEI continued the decline that had begun with the departure of itsfounder, Rafael Caldera, in 1993. In 1998, it did not field a presidential candidate,although Luis Herrera and Donald Ramírez sought the nomination. The partyfirst backed Irene Sáez and then another independent, Henrique Salas Römer. Itcontributed barely 2.15 percent of the 40 percent he polled. In Congress, COPEI


obtained less than half the votes it had in 1993. AD came out better in Congress,but in the presidential field it suffered a disaster. For Congress, AD maintainedroughly the same vote as in 1993, enough to remain the biggest party in theNovember 1998 elections, just as before (Table 2.5). Governors were also electedat that time, and AD again won the most gubernatorial races (Table 2.6). In thepresidential election, perhaps the most logical approach would have been to re-build the party image around the leadership of Fermín, who had fared relativelywell in the 1993 elections and who figured as a favorite in presidential publicopinion polls taken in the first half of 1997. However, the visibility of Fermín’sleadership and the increasing presence of leaders coming out of mayorships andgovernorships generated a reaction from the traditional party apparatus led byLuis Alfaro Ucero. The end result was Claudio Fermín’s departure from AD, asUcero won the presidential nomination. The latter gathered so little support thatAD finally withdrew his candidacy and backed Salas Römer, contributing 9.05percent to his total. Ucero stayed in the race although AD no longer backed him,and he obtained only 0.42 percent of the vote. The discrepancy between the di-saster of the presidential vote and the relatively good performance in the field ofgovernorships, along with acceptable results for Congress, suggests that AD’sproblem lay in its extremely poor handling of the presidential contest.

In the end, the 1998 presidential election had two main protagonists: HugoChávez, who won 56.2 percent of the vote, and Henrique Salas Römer, with40.0 percent. Seemingly, Chávez’s victory grew out of an inexorable need, giventhe context of crisis, the discrediting of the traditional system with its internalconflicts, and the fragility of the forces that had emerged in 1993. Yet Chávezand MBR-200 (Movimiento Bolivariano Revolucionario 200, or BolivarianRevolutionary Movement 200) always hesitated to stand for election, to enterinto the game of “the system.” Only in April 1997 did the group decide to end

62 Tanaka

Table 2.6

Venezuela: Number of Governors Elected, by Party, 1989–2000

Party 1989 1992 1995 1998 2000

AD 11 8 12 7 2COPEI 6 9 3 3 1MAS 2 4 4 3 3La Causa R 1 1 1 – –Convergencia – – 1 1 1Independents – – 1 – 1MVR – – – 7 12PRZVL-PROCA – – – 1 1MERI-COPEI-AD – – – 1 –PPT – – – – 2Total 20 22 22 23 23

SOURCE: Maingón (2002).

its abstentionist stance (López Maya 2003). Chávez began to lead in the presi-dential preference polls only at the beginning of 1998, after the disasters sufferedby his competitors. Throughout 1997, first Claudio Fermín and then Irene Sáezwere favored to win. The 1998 election campaign was extremely volatile; itsoutcome cannot be readily explained through macrovariables, such as the dis-crediting of the system or a crisis of representation, alone. The mistakes of theparties in the system, and their internal conflicts—precisely at a moment whenthese parties were particularly vulnerable—explain how Chávez came to power.The Venezuelan parties suffered splits because of their hierarchical structure,which left no space for dissidence or pluralistic competition among factions. Asa result, conflicts were very hard to handle and often ended in fractures.

Destruction of the Old Order and Transition to a Competitive Authoritarianism

Fujimori and Chávez are definitely opposites as political leaders: the former cameto power based on an ambiguous campaign platform that promoted his indepen-dent character, beyond “traditional politics.” He opposed Vargas Llosa’s market-reform proposals, but, once in power, implemented them and came to personifyneoliberal reform. Chávez always identified with a revolutionary project, a“Bolivarian revolution” based on nationalist and socialist features. However,both men are personalistic and authoritarian leaders with a populist and anti-system discourse. Both men’s candidacies reflected a crisis of representation, andboth men exploited popular dissatisfaction with traditional politics. They repre-sented some previously excluded popular sectors, under plebiscitary, not demo-cratic, schemes. Despite their differences, both established authoritarian regimeswith striking similarities. This resemblance illustrates that deep institutional re-form, under the hegemony of a president who undermines political competi-tion, eliminates the logic of checks and balances inherent to democratic rule andpaves the road to authoritarianism.

Once Fujimori was in power, his success in stabilizing the economy throughmarket-oriented reforms and his later achievements in combating terrorism (seeTable 2.3) allowed him to build a coalition that supported his leadership, whichwas authoritarian, anti-political, and anti-institutional.17 In Venezuela, Chávezused the 1998 election campaign to promote the idea that transforming thecountry should start with institutional change, with dismantling the order of thePunto Fijo agreements of 1958 set down in the 1961 Constitution. The upheavalinvolved in his coming to power, the traditional parties’ internal crises thatreached unsuspected extremes, and the support he received from some membersof the power elite18—all help to explain how Chávez destroyed the preceding in-stitutional order so rapidly. The relative consolidation achieved by both leaderskept them from ending up like other leaders who had come to power by


challenging an establishment that eventually drove them out (Collor in Brazil,1992; Serrano Elías in Guatemala, 1993; Bucaram, 1997, and, recently, Gutiér-rez, 2005, in Ecuador). Through their anti-system discourse, both leaders em-bodied and represented traditionally excluded sectors, but did so within clien-telistic and populist schema.19

Both Fujimori and Chávez constructed new institutional orders, under polit-ical hegemonies that ended those that had preceded through the adoption ofnew constitutions and the takeover of all public authorities. Fujimori proceededby first staging a “self-coup” (April 1992) and then calling for the election ofthe Democratic Constituent Congress (November 1992). Chávez acted by call-ing for a referendum to approve a Constituent Assembly (April 1999), whichthen ignored the authority of Congress and other institutions. Even thoughboth presidents stayed roughly within the margins of legality and enjoyed broadelectoral support, they substantially undermined pluralism, competition, andthe balance of power (horizontal accountability), thereby creating what can becharacterized as “competitive authoritarianisms” in which democratic repre-sentation was replaced by a plebiscitarian legitimacy.20 The disequilibrium wasfurther accentuated by the inability of the opposition in both cases to consoli-date viable alternatives and overcome obstacles to collective action.

In Peru, after the April 1992 coup, Fujimori called for a new Congress thatwould also serve as a constituent assembly. In November 1992, the ConstituentDemocratic Congress (CCD) was installed, with a Fujimorist majority, unlikethe 1990 –92 Congress in which Fujimori’s supporters had been a minority.21

The return to constitutional order was established by the October 1993 referen-dum, which approved the new constitution by a scant margin,22 and by the 1995general election for president and Congress, in which Fujimori easily won re-election in the first round. Fujimori won 64.4 percent of the vote, as well as a ma-jority in the unicameral Congress; his movement won 52.1 percent of the con-gressional vote. His closest competitor, Javier Pérez de Cuéllar, got 21.8 percentof the presidential vote, and his Union for Peru movement (Unión por el Perú,or UPP) captured 14 percent of the congressional vote. In the Congress electedin 1995, there were two main blocks: one linked to the government, Cambio90 –Nueva Mayoría (Change 90 –New Majority) with 67 out of 120 seats, andthe opposition block, led by UPP, with 17 seats. After these two, the party withthe greatest number of seats was APRA, which had only 8 representatives.

Problems appeared soon after the president’s reelection. Since Fujimorismwas a highly personalized movement, it required that Fujimori the person stayin power more than Fujimorism itself. The movement had no significant exis-tence beyond its leader.23

The path followed by Fujimori after 1992 is an interesting illustration of howto construct an authoritarian order through “democratic” means when holdinga majority in Congress, and of how a democratic invocation of the majority can

64 Tanaka

be used to destroy republican balance and liberal principles.24 The path Fujimorifollowed to set up an authoritarian government may be summarized by givingan account of the reelection law and the maneuvers intended to impede anychallenge to his 2000 presidential candidacy.25 In August 1996, Congress, witha solid pro-Fujimori majority, passed the law of “Authentic Interpretation” ofthe 1993 Constitution. According to this law, Fujimori’s first presidential termhad not been from 1990 to 1995 but from 1995 to 2000, since his first term hadbeen governed by the 1979 Constitution, not that of 1993. This law allowed Fu-jimori to stand for his “first” reelection in 2000. Shortly thereafter, in Septem-ber 1996, several opposition leaders began collecting signatures to seek a refer-endum on the repeal of the “Authentic Interpretation” law. The response camein October 1996, when Congress passed legislation regulating the exercise of thereferendum. A referendum would now require not only citizen signatures butalso the approval of at least two-fifths of the members of Congress (that is, 48votes). The route to blocking the referendum subsequently involved a congres-sional confrontation with the judges of the Constitutional Court, which in Jan-uary 1997 declared by a simple majority that the law of “Authentic Interpreta-tion” was inapplicable. Congress responded in May of that year by dismissing thejustices who had voted for that interpretation. In July 1998, the promoters of thereferendum presented petitions with 1,441,535 citizen signatures to the NationalOffice of Election Processes (ONPE). In August, the ONPE enforced the ref-erendum law of October 1996. Instead of calling for the referendum, it sent therequest to Congress, where the opposition did not have the forty-eight votesnecessary to approve the referendum.

Fujimori needed to do more than block the referendum. He also had to keepthe National Elections Board ( JNE) from being able to declare that there was abasis for challenging his candidacy by invoking its unconstitutionality, so he hadto control the JNE. According to Article 179 of the 1993 Constitution, the JNEis made up of five members: one elected by the Supreme Court from among itsretired and active justices; one elected by the Board of Supreme Prosecutorsfrom among retired and active supreme prosecutors; one elected by the Bar As-sociation of Lima from among its members; one elected by the deans of the lawfaculties of public universities from among their former deans; and one electedby the deans of the law faculties of private universities from among their formerdeans. Fujimori’s strategy consisted of controlling the institutions with repre-sentatives on the JNE. Accordingly, in June 1996, the reorganization of theJudiciary and the Prosecutor’s Office was announced. The government tookaction to ensure that the two representatives of these institutions would notimpede Fujimori’s reelection plans. In November 1997, the government an-nounced the takeover of the public universities, a move through which it as-sumed control over the deans of the law schools. That accomplished, Fujimoricould now count on three out of five votes on the JNE. As further insurance, in


May 1998 Congress passed a law changing the kind of vote required for the JNEto declare that there was a basis for challenging a candidacy; the vote went froma simple majority (three votes) to a qualified majority of four out of five.

After these machinations, in December 1999, Alberto Fujimori’s candidacywas filed. The opposition challenged his candidacy, but ultimately the challengewas rejected by the JNE. The route to reelection involved near-absolute con-trol over all state institutions. This became even more evident during the 2000election campaign, when public resources were mobilized to promote Fujimori.Even the armed forces got into the act.26

The Venezuelan case had some elements in common with the Peruvian one.27

Chávez first did away with the Congress elected in November 1998, in whichAD was the largest party. He also dismissed the governors, in whose ranks ADstill figured prominently (see Tables 2.5 and 2.6). To accomplish these objectives,on the day he took office, February 2, 1999, Chávez called for a referendum onconvoking a National Constituent Assembly (ANC). The referendum tookplace on April 25, 1999. More than 80 percent of the valid vote among thosewho cast a ballot was in favor of holding an ANC, but absenteeism was over 60percent. On July 25, 1999, the election for the members of the ANC was held,also with a high rate of absenteeism (over 53 percent). Chávez organized thiselection with a majoritarian electoral system (a personalized one) that enabledhim, with 65.5 percent of the vote, to control 94.5 percent of the seats (121 outof 128). The opposition won 34.5 percent of the vote but just 5.5 percent of theseats.28

With this comfortable majority, Chavism could draw up a constitution with-out having to make any major concessions to the opposition.29 On December15, 1999, a referendum was held to approve the new constitution, which wouldreplace the 1961 Constitution. Although “yes” received 72 percent of the validvotes, 56 percent of eligible voters did not cast ballots. What really destroyed thebalance of power in Venezuela was the ANC’s action on December 23, 1999.Invoking its role as the incarnation of the sovereign will and the expression of anew institutional order, the assembly dissolved the other public authorities, in-cluding the Congress of the Republic, the Council of the Judiciary, the SupremeCourt of Justice, and the Legislative Assemblies.30 Most of the new authorities,whose mandates had already been established under the new constitution, wereelected in 2000. On July 30, elections were held for president of the republic,deputies to the National Assembly (formerly the National Congress), governorsof the twenty-three states, deputies to the State Legislative Councils (formerlyLegislative Assemblies), metropolitan mayors, mayors, councilors to the citycouncils, and delegates to the Latin American Parliament and the Andean Par-liament. These elections were carried out in the context of Chávez’s hegemony,allowing Chávez to use “constitutional” and majoritarian means to build an al-most absolute authority with no institutional counterweights. The result was an

66 Tanaka

authoritarian wielding of power backed by a legality largely devoid of content.Hugo Chávez was reelected with 59.75 percent of the vote (against 37.5 percentfor Francisco Arias), and the Movimiento Quinta República (Fifth RepublicMovement, or MVR) obtained more than 46 percent of the congressional seats.31

All these maneuvers against the rule of law in both countries generated antag-onism, and during long periods of time the majority of the public disapprovedof the performance of Fujimori and Chávez. Why, then, was the political oppo-sition not able to put effective limits on these authoritarian presidents? The de-feat of the opposition can in large measure be attributed to its internal weaknessesand its fragmented nature, the absence of a common strategy, and the lack of asufficiently supported clear alternative offering more than a simple return to apast the citizenry rejected. The opposition raised institutional banners and re-spect for the rule of law and had little to say in social and economic terms; thatwas precisely the strong point for both Fujimori and Chávez. These leaders ex-pressed the interests of previously excluded popular sectors and mobilized themin clientelistic schemes. They drew their legitimacy from plebiscitary schemesand significant popular support, not from democratic legitimacy and respect forthe rule of law. In this sense, the problems of democratic representation remainin both Peru and Venezuela.

In Peru, Fujimorism always enjoyed the support of the “winners” of the eco-nomic reform process: that sector of the business community which is linked tolarge-scale mining interests, finance, and commerce, who benefited from tradeliberalization, privatization, and foreign investment incentives (Tanaka 2003;Gonzales de Olarte 1998). Yet this sector, while strategic, is extremely small andthus was not able to deliver the votes needed by Fujimori to win the 2000 elec-tions. Electorally, the regime’s legitimacy depended on the support of the poor.Thanks to privatizations, increased tax revenues, and greater access to loans fromabroad, and despite its neoliberal character, the Peruvian state renewed its eco-nomic presence under Fujimori. The second Fujimori administration saw thehighest social expenditure levels in more than two decades, and this helps explainthe regime’s greater support among the poorest of the poor. This support was builtthrough effective clientelistic schemes which targeted social expenditure under acentralized structure, controlled by the presidency (Tanaka and Trivelli 2002).

In Venezuela, President Chávez has always enjoyed considerable popular sup-port, but his support has fluctuated and his authoritarian actions have also gener-ated significant opposition. Chávez began his administration with very high ap-proval ratings, and they allowed him to destroy the existing institutional orderand establish a new one under his political hegemony. However, his popularsupport declined from 2001 until at least 2004, and social and political po-larization increased. Chávez’s policies generated rejection by the upper andmiddle classes and, at the same time, created high expectations among the poor,which identified with his populist, nationalistic, and revolutionary rhetoric. This


polarization reached high levels after the Decree of 49 Laws in November 2001,under extraordinary legislative measures granted by the Congress, which in-cluded a land reform law and a new oil law, among others. The polarization cul-minated in a coup d’état in April 2002. After the coup failed, the opposition leda general strike, between December 2002 and the first months of 2003, callingfor the removal of the president (Medina and López Maya 2003). After the defeatof the general strike, a dialogue was established, in May 2003, between the gov-ernment and the opposition, with the mediation of the Organization of Ameri-can States (OAS). An institutional mechanism to resolve the political conflict wasinvoked, and it turned into the call for a referendum to remove President Chávez.This tortuous process began by August 2003 with the first campaign to collect thesignatures needed for a recall referendum. After many conflicts, the recall votefinally took place in August 2004. Chávez obtained 58.25 percent of the vote, andthe opposition 41.74 percent. What happened? Despite all the allegations raisedby the opposition, which denounced the government’s dirty maneuvers and al-leged electoral fraud, the Venezuelan case can be analyzed along the same lines asthe Peruvian one. As in Fujimori’s Peru, in Venezuela there is an authoritariangovernment and an active opposition, but one that lacks a common strategy andis unable to present a clear alternative. While the opposition emphasized an in-stitutional discourse, Chávez underscored redistributive measures and expandedsocial expenditures, benefiting from rising oil prices in 2003 and 2004.

Despite the considerable differences between Fujimori and Chávez, the Pe-ruvian and Venezuelan political regimes have remarkable similarities, as does therelationship between the government and the opposition. These presidentsheaded formally democratic governments. Based on their political hegemonyand their control of the Congress, they rebuilt the institutional order and de-stroyed the checks and balances inherent to democratic rule. Their measuresprovoked an important opposition, which focused on institutional banners thatappealed to the middle classes but not to the popular sectors, which were se-duced by populist rhetoric and increasing social expenditures.

Both countries are suffering the consequences of the non-existence of a partysystem, along with the consequences of the fragility of the new leaderships andmovements that have sprung up in recent years. Because of the precariousness ofthe groups in power and those in opposition, the problems of democratic repre-sentation are not going away. In the vacuum left by the collapse of the party sys-tems, authoritarian governments emerged, but they failed to establish new hege-monic parties. The new movements appearing in recent years are characterizedby personalism, precariousness, improvisation, and volatility. This helps explainwhy Fujimorism collapsed unexpectedly after the irregular reelection of 2000.Fujimorism fell apart because of its internal contradictions and its personalisticnature; more specifically, the main explanation for the fall lies in the conflicts be-tween the president and his intelligence advisor, Vladimiro Montesinos (Tanaka2005).

68 Tanaka

The Challenges: How to Build Democratic Representation without Parties

Peru and Venezuela are exceptions in Latin America and among the Andeancountries, where crises of representation have brought about a gradual evolutionof party systems. In these other countries, the traditional parties are competingwith new parties that represent new social sectors. In Peru and Venezuela, a sud-den collapse of the party systems occurred, opening the way for political “out-siders.” The key to understanding the different outcome here lies in the internaldynamics of the political parties at junctures when they were particularly vulner-able—the 1990 election in Peru and the 1998 election in Venezuela. The partieswere unable to handle their internal conflicts, which resulted in schisms. As a re-sult, the traditional parties were discredited in the eyes of the voters. Once inpower, Fujimori and Chávez destroyed the preceding institutional order and es-tablished competitive authoritarian regimes, in which democratic representationwas replaced by plebiscitarian mechanisms of legitimation.

For both countries, major challenges lie ahead. In Peru, the challenge now isto surmount the legacy of a decade of authoritarianism that destroyed state in-stitutions, which were run so as to keep Fujimori in power. Another negativelegacy is the fragility of social and political organizations. The party system ofthe 1980s was destroyed; the gap was filled by Fujimorism and independentmovements, but none of these have consolidated themselves. Fujimorism alsodamaged Peruvian society’s capacity for collective action, undermining the rep-resentativeness of the social actors and isolating them from society. Contrary tosome views, the fall of Fujimorism was not the result of the growth of the op-position or of social protests. That false notion has led to overestimating the ca-pabilities of the political groups now at the center of the political scene and incontrol of the Congress, and to underestimating the continuity of patterns char-acteristic of Fujimorism, which continue within the political culture, the me-dia, the judiciary, and other institutions. The campaign for the new presidentialand congressional elections in 2001 showed the great weakness of the partici-pating actors, which is also demonstrated by the precariousness of the currentgovernment led by President Alejandro Toledo (2001–2006).

In Venezuela, the challenge is to see that the Chávez government does not fol-low a path similar to that of Fujimorism—in other words, to prevent Chávez’scontrol of institutions and the absence of checks and balances from leading to acritical weakening of political competition and pluralism that generates increas-ing arbitrariness, authoritarianism, and corruption. Such a path may foster a po-larization greater than already exists and may open the way for a cycle of greaterinstability and violence. The opposition to Chávez fluctuates between strategiesof violence, with calls for a coup (such as the April 2002 coup), and a negoti-ated institutional solution (a referendum, for example, leading to a recall and new elections). The opposition to Chávez is fractured and offers no clear alter-


native, making it hard to visualize a way out of the current situation of profoundpolarization.

In both countries we are witnessing the difficulties of conducting politicswithout parties (Levitsky and Cameron 2001). Precarious and volatile move-ments and parties generate two perverse logics. First, these groups have limitedtime horizons; they give priority to short-term logics and are unable to deviselong-term policies or strategies. Second, since the movements are new and pre-carious, their expectations regarding political rewards are not great. It is enoughfor them to obtain a few mayorships and governorships or congressional seats,and from these positions strive for future growth and consolidation. This logichinders the formation of coalitions, generates fragmentation, and impedes re-solving problems of collective action. For these reasons, in Peru and Venezuelaboth the actors in power and the opposition are weak. In the Peruvian case, thisis the story of movements such as Unión por el Perú (Unity for Peru), SomosPerú (We Are Peru), Solidaridad Nacional (National Solidarity), and Perú Posi-ble (Possible Peru).32 In Venezuela, the same is true of movements such as Con-vergencia, Proyecto Venezuela (Project Venezuela), and new groups such asPrimero Justicia (First Justice) and others.

Amid the climate of questioning of politics and parties present throughout theAndean region, the cases of Peru and Venezuela furnish valuable lessons about theimportance of parties. Despite their shortcomings, their absence makes problemsworse rather than solving them. The cases of Peru and Venezuela offer a remark-able contrast, compared to the other Andean countries, with their party systemsin crisis but not in a state of collapse. Citizen discontent and crises of representa-tionhaveencouraged the traditionalparties topromote institutional reforms inor-der to refurbish their images and to compete successfully with emerging politicalforces. As limited as these initiatives may be, they open up the political system andcreate spaces that canbeoccupiedbynewsocial andpolitical actors. In recentyears,Colombia, Bolivia, and Ecuador have undertaken decentralization processes, in-stituted the popular election of local officials, introduced new constitutions orconstitutional changes that broadened recognition of social rights, and effectedpolitical reforms to enhance the accountability of electedofficials—allwith resultsthat are far from insignificant. In Bolivia and Ecuador, the political systems havemade great progress toward recognizing the rights of indigenous groups, achiev-ing substantive advances in greater representation for ethnic groups that were tra-ditionally ignored; in Colombia, the heritage of the National Front and its exclu-sionary nature is definitely gone. In short, despite all the problems, the existenceof a party system constitutes an advantage for democratization.

Between 1998 and 2002, the region faced a new economic recession, a con-sequence of the exhaustion of first-generation market reforms. The EconomicCommission for Latin America (ECLA) labeled this period the “lost half a decade,” and its effects are still impacting the political arena. We currently face

70 Tanaka

a new critical juncture that will surely have medium and long-term conse-quences. The institutional changes mentioned above have been exhausted, inthe sense that the political systems have been opened and new sectors haveemerged, but in the crisis context, this emergence has created renewed govern-ability problems. Bolivia may also face the collapse of its party system in the nextgeneral elections; Ecuador and Colombia face increasing instability. Democraticrepresentation is still a central issue in the Andes. Parties do not represent ade-quately. New anti-system leaders emerge, but they further weaken democraticinstitutions. How to achieve an equilibrium between representation and dem-ocratic governance is part of the pending agenda.

Notes

My thanks to María Jesús Osorio for compiling some of the data presented here, andto Thais Maingón, Scott Mainwaring, and Eduardo Pizarro Leongómez for their intel-ligent comments on the first version of the text. Responsibility for its limitations is ofcourse entirely mine. Part of the research for this paper was made possible through assis-tance from the Consejo Latinoamericano de Ciencias Sociales (CLACSO), under its regional grant program (Programa Regional de Becas CLACSO–Asdi) for senior re-searchers on Latin America and the Caribbean 2001.

1. For an overview of Venezuela, see Caballero (2000) and Levine and Crisp (1999);for Peru, see McClintock (1999b).

2. For the Colombian case, see Pizarro Leongómez; for Ecuador, see Pachano; forBolivia, see Mayorga, all in this volume.

3. In November 1989 there were municipal elections, and in April 1990 elections forthe president of the republic and all congressional seats.

4. According to an October 1989 APOYO poll, 47 percent of voters intended to votefor Vargas Llosa in the April 1990 presidential election.

5. Made up of the Unified Mariateguista Party (PUM), the National Union of theRevolutionary Left (UNIR), and the Popular Front of Workers, Peasants, and Students(FOCEP).

6. Candidates for the municipal elections filed in August 1989; presidential hopefulsfiled in October 1989; congressional candidates in January 1990. During that entire pe-riod, the internal struggles of all the parties were daily topics in the news media.

7. Vargas Llosa wanted FREDEMO to run Front candidates in the municipal elec-tions, whereas AP and PPC (Partido Popular Cristiano, or the Popular Christian Party)wanted to run candidates from their own ranks, with Front candidates being nominatedonly for the presidential and congressional contests. This disagreement led to VargasLlosa’s decision to temporarily withdraw his presidential bid.

8. Ricardo Belmont, for example, elected mayor of Lima as an independent, cam-paigned openly for Vargas Llosa, even making a speech at the latter’s end-of-campaign rallyin Lima.

9. Some writers maintain that the “Fujimori phenomenon” was the expression of agrave crisis of political representation in Peruvian society, and that it expressed ethnic,


cultural, class, and other problems of representation. In my view, such positions illustratethe fallacy of “retrospective determinism.” Once an event has taken place, an argumentis constructed presenting that event as inevitable. Yet less than a month before the elec-tion, it was almost impossible to imagine such an outcome.

10. According to a survey firm, APOYO, Fujimori no longer appeared under theheading “Others” (for very minor candidates) in its poll taken between March 8 and 11,when he had 3 percent of popular preferences. In the March 16 –18 poll, he registered 6percent; in the March 24 –26 survey, 9 percent. According to IMASEN’s March 5–7poll, Fujimori had 2.5 percent; in the March 9–12 survey, 6.1 percent; and in the March14 –16 survey, 9.5 percent. These figures began to increase at a faster rate, and Fujimorireached 29.1 percent on April 8.

11. See Coppedge (1994); Benton (1997); Corrales (2000); Crisp et al. (2003);Molina and Alvarez (2004), among others.

12. On COPRE, see Jácome (1999). On the relationship between decentralizationand the breakdown of the party system, see Lalander (2004) and Penfold (2001). On theeffects of the proportional personalized vote, see Crisp and Rey (2001) and Kulisheck andCrisp (2001).

13. The former COPEI candidate, Eduardo Fernández, was more consistent and sup-ported the constitutional president. The reason for Caldera’s stance is clear. Unable towin the presidential nomination through COPEI, he needed to create a political spaceoutside the party. Once president, Caldera granted amnesty to Chávez, a decision thatallowed the latter to run in the 1998 election. On the “imaginary redeemer” who wouldfinally take bodily form as Chavism, see Arenas and Gómez Calcaño (2000).

14. For an extended discussion of attempts to remove Latin American presidents inrecent years, see Pérez-Liñán (2001).

15. It is revealing to compare the stance of Rómulo Betancourt, long-time leader ofAD, with that of Caldera. The former, after serving as president, did not run again. Con-versely, Caldera participated in every election he could. The effects of these two behav-iors on the destinies of their respective parties are quite different. Betancourt’s behaviorstrengthened his party; Caldera’s damaged his party and the party system. On Caldera,see Crisp et al. (2003); and Álvarez (2004).

16. For an overview of the economy in the period from 1989 to 1998, see Hidalgo(2000) and Kelly (2001).

17. On Fujimorism, see Cotler and Grompone (2000); Degregori (2000); Rospigliosi(2000); Marcus and Tanaka (2001), among others.

18. The media were relatively favorable to Chávez until the 1999 Constituent As-sembly, when a distancing that later turned into open confrontation began. An illustra-tive example is Alfredo Peña, former editor of the daily El Nacional. Peña was elected byMVR to the Constituent Assembly and then to the mayorship of Caracas. He is now oneof the leaders of the opposition to Chávez (see Petkoff 2002).

19. On this point, see Tanaka (2002). For a different perspective on the Venezuelancase, see López Maya (2001, 2003) and Medina and López Maya (2003).

20. I take the concept of “competitive authoritarianism” from Levitsky and Way(2002); see also Schedler (2002) and Diamond (2002). For the Peruvian case, see alsoMcClintock (1999a), Conaghan (2001), and Tanaka (1999, 2002). On Venezuela, seeCoppedge (2002).

21. In the April 1990 elections, Cambio 90 obtained 21.7 percent of the votes for theSenate and 16.5 percent of the votes for deputies. In November 1992, Cambio 90 – NuevaMayoría won 49.2 percent of the votes and won 44 out of a total of 80 congressionalseats.

72 Tanaka

22. “Yes” votes prevailed over “No” votes, 52 to 48 percent, amid accusations offraud.

23. In the 2000 election, with Fujimori as the candidate, the Peru 2000 movementgot 42 percent of the votes for Congress. Just one year later, the movements identifiedwith Fujimorism, Cambio 90 –Nueva Mayoría and Solución Popular, obtained a mea-ger 4.8 percent and 3.6 percent, respectively.

24. See O’Donnell (1998). This phenomenon should lead to an in-depth discussionof how to conceptualize democracy and how the international community ought to dealwith these matters. The OAS was lax with Fujimori, despite his movement’s authoritar-ianism, because Fujimori had won a popular mandate. Now the OAS is in a similar situ-ation with Chávez.

25. On these points, see Ames et al. (2001); Bernales (2000, 2001); Sanborn et al.(2000), among others.

26. On the 2000 election, see the many election-observation reports produced by theOAS mission, the Carter Center, the National Democratic Institute, the U.S. State De-partment, the International Federation for Human Rights, the Electoral Reform Inter-national Service, and the Washington Office on Latin America. Also see the reports byPeruvian groups such as Transparencia, Foro Democrático, Consejo por la Paz, and theDefensoría del Pueblo (the government ombudsman’s office).

27. On the subjects discussed in this section, see Gómez and Patruyo (2000) andMaingón et al. (2000), among others.

28. It is interesting to compare Chávez’s and Fujimori’s electoral systems. Given thatFujimori had the majority of the votes, it might have been better for Fujimori had he,too, elected his unicameral Congress by a majority system in 1993, 1995, and 2000. Onall three occasions, however, Congress was elected from a single countrywide district, ahighly proportional system with preferences for minorities. Why did he do this? Fuji-mori wanted to have total control over his party and did not want to negotiate with lo-cal and regional political bosses, something unavoidable with a system of single-memberdistricts, even if he could have controlled the nomination process. Fujimori proceededin this way because he never established a solid political movement. Chávez, conversely,by 1998 had built a minimal political organization, which he put together between hisrelease from prison and the 1998 election.

29. Neither the Venezuela Constitution of 1999 nor the Peruvian Constitution of1993 is explicitly designed to build an authoritarian regime. Both constitutions focusedon promoting “direct democracy” and maintained the basic formalities of a demo-cratic constitution. These constitutions later became significant obstacles for Fujimoriand Chávez, and served as useful tools for the opposition in both Peru and Venezuela.Both constitutions allowed for referendums as a way to limit presidential power. Bothpresidents repeatedly violated the constitutional order created under their political hegemonies.

30. In January 2000 the ANC dissolved itself and was replaced by a “mini-Congress”made up of some ANC members and others appointed by the ANC. This body operateduntil the new Congress began its duties.

31. On the 2000 elections, see Carrasquero et al. (2001); and Maingón (2002). Al-though AD did not nominate a presidential candidate, the party is still a significant pres-ence in Congress as the prime opposition force. La Causa Radical, despite being the mainpillar of the vote for Arias Cárdenas (contributing some 19 percent of his 37.5 percent),got a minuscule vote for Congress.

32. The only exception is APRA, but it is partial, given the accentuation of its caudi-llismo. APRA’s presidential candidate, Abel Salinas, obtained a meager 1.38 percent of the


vote in the 2000 election, when its congressional list won 5.5 percent. Alan García won25.8 percent in the first presidential round of 2001 and 46 percent in the second, and theparty’s congressional list polled 19.7 percent of the vote.

References

Álvarez, Ángel. 2004. “COPEI: La triste historia de un partido sin vocación de poder.” InLos partidos políticos venezolanos en el siglo XXI, ed. José Molina and Ángel Álvarez,159–94. Caracas: Vadell Hermanos.

Ames, Rolando, Enrique Bernales, Sinesio López, and Rafael Roncagliolo. 2001. Situaciónde la democracia en el Perú (2000 –2001). Lima: PUCP.

Arenas, Nelly, and Luis Gómez Calcaño. 2000. El imaginario redentor: De la revolución deoctubre a la Quinta República Bolivariana. Temas para la discusión, Serie Arbitrada, no. 6. Caracas: CENDES.

Benton, Allyson L. 1997. “Patronage Games: The Effects of Economic Reform on Intra-Party Politics in Venezuela.” Paper delivered at the 1997 meeting of the Latin Amer-ican Studies Association, Guadalajara, México, April.

Bernales, Enrique. 2000. “La ilegitimidad constitucional del tercer gobierno de AlbertoFujimori.” In Perú 2000: Un triunfo sin democracia, ed. Cecilia Anicama et al., 57–108.Lima: Comisión Andina de Juristas.

———. 2001. “Aspectos constitucionales de la transición democrática.” In Las tareas dela transición democrática, ed. Cecilia Anicama et al., 33–58. Lima: Comisión Andina deJuristas.

Caballero, Manuel. 2000. La gestación de Hugo Chávez. 40 años de luces y sombras en lademocracia venezolana. Madrid: Catarata.

Carrasquero, José Vicente, Thais Maingón, and Friedrich Welsch, eds. 2001. Venezuelaen transición: Elecciones y democracia, 1998–2000. Caracas: REDPOL.

Conaghan, Catherine. 2001. “Making and Unmaking Authoritarian Peru: Re-Election,Resistance, and Regime Transition.” The North-South Agenda, Paper no. 47, Uni-versity of Miami ( June).

Coppedge, Michael. 1994. Strong Parties and Lame Ducks. Presidential Partyarchy and Fac-tionalism in Venezuela. Stanford, CA: Stanford University Press.

———. 2002. “Venezuela: Popular Sovereignty versus Liberal Democracy.” WorkingPaper, no. 294. Helen Kellogg Institute for International Studies, University of NotreDame, Notre Dame, IN.

Corrales, Javier. 2000. “Presidents, Ruling Parties, and Party Rules: A Theory on thePolitics of Economic Reform in Latin America.” Comparative Politics 32, no. 2 ( Janu-ary): 127– 49.

Cotler, Julio, and Romeo Grompone. 2000. El fujimorismo: Ascenso y caída de un régimenautoritario. Lima: Instituto de Estudios Peruanos.

Crisp, Brian. 2001. “Candidate Selection in Venezuela (and Its Impact on Legislator Behavior).” Paper presented at the Latin American Studies Association 13th Interna-tional Congress, Washington, D.C., September.

Crisp, Brian, and Juan Carlos Rey. 2001. “The Sources of Electoral Reform in Venezu-ela.” In Mixed-Member Electoral Systems: The Best of Both Worlds? ed. Matthew S.Shugart and Martin Wattenberg, 173–93. Oxford: Oxford University Press.

74 Tanaka

Crisp, Brian, Daniel Levine, and José E. Molina. 2003. “The Rise and Decline of COPEIin Venezuela.” In Christian Democracy in Latin America, ed. Scott Mainwaring andTimothy R. Scully, 275–300. Stanford, CA: Stanford University Press.

Degregori, Carlos Iván. 2000. La década de la antipolítica: Auge y huida de Alberto Fujimoriy Vladimiro Montesinos. Lima: IEP.

Diamond, Larry. 2002. “Thinking about Hybrid Regimes.” Journal of Democracy 13, no. 2: 21–35.

Gómez Calcaño, Luis, and Thanalí Patruyo. 2000. “Entre la esperanza popular y la cri-sis económica: Transición política en Venezuela.” Cuadernos del CENDES (Caracas)17, no. 43, segunda época ( January–April).

González, Sonia. 2002. “La desconfianza en los partidos en Venezuela.” Unpublishedmanuscript.

Gonzales de Olarte, Efraín. 1998. El neoliberalismo a la peruana: Economía política del ajusteestructural, 1990 –1997. Lima: Instituto de Estudios Peruanos.

Hidalgo, Manuel. 2000. “Liderazgo político y reforma económica: El caso de Venezu-ela, 1989–1998.” Zona Abierta (Madrid: Ed. Pablo Iglesias), no. 90/91: 91–160.

Jácome, Francine. 1999. “Reformas políticas en Venezuela: Una evaluación preliminar.”Ciencias de Gobierno (Instituto Zuliano de Estudios Políticos, Económicos y Sociales),no. 6 ( July–December).

Kelly, Janet. 2001. “The Syndrome of Economic Decline and the Quest for Change.”Unpublished manuscript.

Kornblith, Miriam. 1998. Venezuela en los noventa: La crisis de la democracia. Caracas: IESA.Kulisheck, Michael, and Brian Crisp. 2001. “The Legislative Consequences of MMP

Electoral Rules in Venezuela.” In Mixed-Member Electoral Systems: The Best of BothWorlds? ed. Matthew S. Shugart and Martin Wattenberg, 404 –31. Oxford: OxfordUniversity Press.

Lalander, Rickard. 2004. Suicide of the Elephants? Venezuelan Decentralization between Party-archy and Chavismo. Helsinki: University of Helsinki.

Levine, Daniel, and Brian Crisp. 1999. “Venezuela: The Character, Crisis, and PossibleFuture of Democracy.” In Democracy in Developing Countries: Latin America, ed. LarryDiamond et al., 367– 428. 2nd. ed. Boulder: Lynne Rienner.

Levitsky, Steven, and Maxwell Cameron. 2001. “Democracy without Parties? PoliticalParties and Regime Change in Fujimori’s Peru.” Paper delivered at the 13th Congressof the Latin American Studies Association, Washington, DC, September.

Levitsky, Steven, and Lucan Way. 2002. “The Rise of Competitive Authoritarianism.”Journal of Democracy 13, no. 2: 51– 65.

López Maya, Margarita. 1997. “The Rise of Causa R in Venezuela.” In The New Politicsof Inequality in Latin America: Rethinking Participation and Representation, ed. DouglasChalmers et al., 117– 43. Oxford: Oxford University Press.

———. 2000. “La protesta popular venezolana entre 1989 y 1993 (en el umbral del neo-liberalismo).” In Lucha popular, democracia, neoliberalismo: Protesta popular en AméricaLatina en los años de ajuste, ed. Margarita López Maya, 211–35. Caracas: Nueva Sociedad.

———. 2001. “Partidos de vocación popular en la recomposición del sistema políticovenezolano: Fortalezas y debilidades.” Paper delivered at the 13th Congress of theLatin American Studies Association, Washington, DC, September.

———. 2003. “Hugo Chávez Frías: Su movimiento y presidencia.” In La política vene-zolana en la época de Chávez: Clases, polarización y conflicto, ed. Steve Hellner and DanielHellinger, 97–120. Caracas: Nueva Sociedad.

Maingón, Thais. 2002. “Comportamiento político-electoral del venezolano y construc-ción de tendencias: 1998 y 2000.” Cuadernos del Cendes 49: 49–78.


Maingón, Thais, Carmen Pérez, and Heinz Sonntag. 2000. “La batalla por una nuevaconstitución para Venezuela.” Revista Mexicana de Sociología 62, no. 4 (October–December): 91–124.

Marcus, Jane, and Martín Tanaka. 2001. Lecciones del final del fujimorismo: La legitimidadpresidencial y la acción política. Lima: Instituto de Estudios Peruanos.

McClintock, Cynthia. 1999a. “¿Es autoritario el gobierno de Fujimori?” In El juegopolítico: Fujimori, la oposición y las reglas, ed. Fernando Tuesta, 65–96. Lima: FundaciónFriedrich Ebert.

———. 1999b. “Peru: Precarious Regimes, Authoritarian and Democratic.” In Democ-racy in Developing Countries: Latin America, ed. Larry Diamond et al., 309– 65. 2nd ed.Boulder, CO: Lynne Rienner.

Medina, Medófilo, and Margarita López Maya. 2003. Venezuela: Confrontación social y po-larización política. Bogotá: Ed. Aurora.

Molina, José, and Ángel Álvarez, eds. 2004. Los partidos políticos venezolanos en el siglo XXI.Caracas: Vadell Hermanos Editores.

Naím, Moisés. 1993. Paper Tigers and Minotaurs: The Politics of Venezuela’s Economic Re-forms. Washington, DC: Carnegie Endowment for International Peace.

O’Donnell, Guillermo. 1998. “Accountability Horizontal.” Agora (Buenos Aires), no. 8:91–160.

Payne, Mark, et al. 2003. La política importa: Democracia y desarrollo en América Latina.Washington, DC: BID–IDEA.

Penfold, Michael. 2001. “El colapso del sistema de partidos en Venezuela: Explicación deuna muerte anunciada.” In Venezuela en transición: Elecciones y democracia, 1998–2000,ed. José Vicente Carrasquero, Thais Maingón, and Friedrich Welsch, 36 –51. Caracas:REDPOL.

Pérez-Liñán, Aníbal. 2001. “Public Opinion and Executive Accountability: The Politi-cal Economy of Impeachment Crisis.” Paper delivered at the 2001 meeting of theLatin American Studies Association, Washington, DC.

Petkoff, Teodoro. 2002. “El presidente acusa a los medios: La luna de miel ha termi-nado.” Etcétera (Mexico City) (March).

Roberts, Kenneth M. 2003. “Party System Collapse Amid Market Restructuring in Ven-ezuela.” In Post-Stabilization Politics in Latin America: Competition, Transition, Collapse,ed. Carol Wise, Riordan Roett, and Guadalupe Paz, 249–72. Washington, DC:Brookings Institution Press, 2003.

Rospigliosi, Fernando. 2000. Montesinos y las fuerzas armadas: Cómo controló durante una dé-cada las instituciones militares. Lima: Instituto de Estudios Peruanos.

Sanborn, Cynthia, Francisco Eguiguren, and Bruce Kay. 2000. “Democracy and Gover-nance in Peru: An Assessment.” Lima: Management Systems International (MSI), un-der contract to the U.S. Agency for International Development.

Schedler, Andreas. 2002. “The Menu of Manipulation.” Journal of Democracy 13, no. 2:36 –50.

Tanaka, Martín. 1998. Los espejismos de la democracia: El colapso del sistema de partidos en elPerú, 1980 –1995, en perspectiva comparada. Lima: Instituto de Estudios Peruanos.

———. 1999. “La consolidación democrática en América Latina y la importancia de lacompetencia política: Lecciones desde la experiencia peruana.” In El juego politico: Fu-jimori, la oposición y las reglas, ed. Fernando Tuesta, 43– 63. Lima: Fundación FriedrichEbert.

———. 2002. La situación de la democracia en Colombia, Perú y Venezuela a inicios de siglo.Lima: Comisión Andina de Juristas.

76 Tanaka

———. 2003. “The Political Constraints on Market Reform in Peru.” In Post-Stabilization Politics in Latin America: Competition, Transition, Collapse, ed. Carol Wise,Riordan Roett, and Guadalupe Paz, 221– 48. Washington, DC: Brookings Institu-tion Press.

———. 2005. “Peru, 1980 –2000: Chronicle of a Death Foretold? Determinism, Will,Actors, and De Facto Powers.” In The Third Wave of Democratization in Latin America:Advances and Setbacks, ed. Frances Hagopian and Scott Mainwaring, 261–88. Cam-bridge: Cambridge University Press.

Tanaka, Martín, and Carolina Trivelli. 2002. “Las trampas de la focalización y la partici-pación: Pobreza y políticas sociales en el Perú durante la década de Fujimori.” Docu-mento de Trabajo, no. 121. Instituto de Estudios Peruanos, Lima.

Tuesta, Fernando. 2001. Perú político en cifras, 1821–2001. 3rd ed., revised and expanded.Lima: Fundación Ebert.

Weyland, Kurt. 2002. The Politics of Market Reform in Fragile Democracies: Argentina, Brazil,Peru, and Venezuela. Princeton, NJ: Princeton University Press.


Until the late 1990s, Colombia’s two-party system was one of the oldestand most stable in the world. In 1849, Don Ezequiel Rojas created the pro-grammatic foundations for the Liberal Party (PL). The same year, Mariano Ospina Rodríguez and José Eusebio Caro wrote the founding doctrines of theConservative Party (PC). This was over 150 years ago. Neither England nor theUnited States—the two classic models for the two-party system—can claimsuch a long tradition. The two parties that dominated the nineteenth century inEngland (Conservative and Liberal) gave way to two others (Conservative andLabour) that prevailed throughout the twentieth century. In the United States,the two current parties (Republican and Democrat) emerged shortly before theCivil War (1861– 65).

Unlike the party systems in England and the United States, the two-party sys-tem in Colombia actually functioned as a multiparty system, owing to a longtradition of factional struggle. In Colombia, as in Uruguay, party factions werepolitical entities with a higher degree of discipline and cohesion than the partiesproper; the parties were, in fact, little more than two “political subcultures” be-hind which the real political machinery moved. For all practical purposes, fac-tions separate from the official Liberal or Conservative Parties—such as the Lib-eral Revolutionary Movement (Movimiento Revolucionario Liberal, or MRL)or the Conservative Union—were parties in themselves. Each had its nationaldirectorate, its departmental and municipal directorates, its parliamentary cau-cus, and its government platform.

This system has undergone a deep transformation in the last decade. The internal fragmentation of the Liberal and Conservative Parties has intensified,with both moving from an internal structure based on “institutionalized fac-tions” (Morgenstern 2001, 236)1 to one based on personalistic factions—or, touse the term popularized in Colombia, “electoral micro-enterprises” (Pizarro

3

Giants with Feet of Clay: Political Partiesin Colombia

❖

Eduardo Pizarro Leongómez

Giants with Feet of Clay: Colombia 79

Leongómez 2002). At the same time, there has been a progressive decline in bi-partisanship—from a system in which the two parties controlled over 90 percentof congressional seats toward a system with other political movements that aregaining greater access to representation in local, departmental, and national leg-islatures with each passing day.

The Colombian party system is undergoing a profound transformation, theresults of which are still uncertain. Along with Uruguay and Honduras, Colom-bia until the late 1990s had one of the three oldest and most stable two-partysystems in Latin America. A reconstruction of the two-party system is unlikely,especially given the apparently irreversible crisis in the PC, whose decline hasbeen steady over the past three decades.2 Is Colombia headed for a collapse ofthe party system, as in Venezuela, Peru, and Italy? Or toward a reorganizationof the party system based on political coalitions that have blossomed in recentelections? No one dares make any predictions. A climate of enormous uncer-tainty reigns.

In this context of party atomization, the forms of political representation arealso undergoing profound change. Two or three decades ago, the two traditionalparties, typical multiclass, “catch-all” parties, monopolized the immense major-ity of partisan support. For generations, Colombians of all social classes identifiedwith Liberals or Conservatives (Losada and Vélez 1981). Today, party identitysurvives only in a few rural or semi-urban zones and among elderly urban vot-ers. The electoral panorama is dominated by “electoral micro-enterprises.” Thisform of personalistic atomization is the manifestation of a phenomenon unpar-alleled elsewhere in the world, except probably in Israel: the political parties’ in-ability to select candidates for legislative assemblies. In Colombia, candidatesfrom all the political or social movements (with only a few exceptions) increas-ingly nominate themselves, design their own campaigns, and organize their ownfinances. The political parties simply give their labels so that the candidacies canrun for office. The political panorama is therefore dominated by hundreds of“micro-representations” of all sorts (political, corporative, regional, ethnic, reli-gious) and on all levels (national, regional, and local) in which each parliamen-tarian, deputy, or councilman covers a certain nucleus of the population.3

The following analysis of the personalistic and particularistic representationthat prevails in Colombia draws upon studies by Carey and Shugart (1995),Shugart (1999), and, more recently, the Inter-American Development Bank(Seddon et al. 2002) and Panizza (2001), which focus on the impact of electoralsystems on the level of particularism in a given political regime.4 In these stud-ies, particularism is defined as a political leader’s ability to further his or her ca-reer by supporting specific social groups rather than national platforms. As wewill see, increasingly there are few incentives in Colombia to build a political ca-reer based on a party platform. Several institutional factors motivate politiciansto cultivate careers based above all on their personal reputations, and therefore

80 Pizarro Leongómez

on the constitution of electoral micro-enterprises composed of “specific voters”(the Colombian expression for a clearly identifiable and narrow electoralconstituency).

This phenomenon has given rise to a model of political representation thatI refer to as “dual representation.”5 On the one hand, legislative power is gener-ally an expression of segmented and particularistic representation; on the other,executive power is the expression of more general interests. The same phenom-enon is at the root of the increasing “hypertrophy” of presidential power, uponwhich the state’s conduct and the very governability of the political system rest,especially in relation to the figure of the president.6

Institutionalization and De-institutionalization

In 1995 the well-known book Building Democratic Institutions: Party Systems inLatin America (1995), edited by Scott Mainwaring and Timothy Scully, focusedattention on the degree of institutionalization or de-institutionalization of thecontinent’s party systems. According to this conceptual framework, in a systemof institutionalized parties, party organization is generally solid, electoral volatil-ity is low, parties are deeply rooted in society, and their operations are a result ofinstitutional routines rather than charismatic personalities or leaders. And vice-versa: in a weakly institutionalized system, party organization is weak, electoralvolatility is high, parties are not deeply rooted in society, and personalities dom-inate party life. In this comparative analysis, the Colombian party system ap-peared to be one of Latin America’s most institutionalized. How, then, can oneexplain the rapid de-institutionalization process in the Colombian party system?

Randall and Svasand (2002) suggest that analyses of party systems are based ona supposition that may be true in certain contexts and false in others. This sup-position assumes that the institutionalization of a party system is equivalent tothe institutionalization of the parties comprising the system. This is not alwaysthe case. The four criteria that Mainwaring and Scully (1995, 4 –5) used to mea-sure the institutionalization of the party system placed Colombia on a high levelbut, in turn, did not mention the internal erosion of the Liberal and Conserva-tive Parties.7 The decline of the two historical parties generates many questionssince both parties had, for decades, been able to survive through even dramaticchanges.8 Their ability to adapt was considered one of their most specific traits.What happened?

The literature on parties suggests many hypotheses for the weakening of po-litical parties on a global scale. Pennings and Hazan (2001, 267) maintain that “inthe majority of modern democracies, the relationship between parties and vot-ers has weakened. The reasons for this are often related to growing levels of ed-ucation and social well-being that render citizens less dependent on parties,unions and other collective representational bodies.” This is indeed probable in


countries with a high level of social well-being.9 But in nations where povertyand income inequality have increased, the explanation is most likely different. InLatin America, and particularly in Colombia, different explanations exist. Insti-tutionalist explanations have shown that different political and electoral reforms,beginning in the 1980s, undermined the traditional forms of political action inColombia and created a climate that favored the current “party atomization”(Cox and Shugart 1995). Authors, including Kenneth Roberts (2001, 183–84),have emphasized political economy factors, affirming that “the crisis of import-substitution industrialization and the neo-liberalism that took its place in the1980s and 1990s meant a new ‘critical juncture’ for Latin America’s political de-velopment.” The retreat of the state, the fiscal crisis, and austere economic poli-cies reduced the flow of resources needed to sustain both parties founded onclient-based networks (Colombia) and those founded on corporatist networks(Venezuela). According to Roberts, the new economic model brought with it theneed to construct a new matrix of political representation. Some parties in theregion adapted (e.g., in Chile, Uruguay, and Costa Rica); others did not. Finally,there are cultural explanations for rapidly declining party loyalty, a consequenceof the weakening of the old channels of political socialization such as family andregion (Pinzón 1998).

An analysis of the steady decay of the two-party system in Colombia can andshould be based on a number of explanations, including institutional as well ascultural and economic factors. This brief discussion, however, will be limited tothe institutional dimension—with an emphasis on the impact of electoral lawsand rules regulating party operations on the growing trend toward personalisticatomization.

Personalism and Particularism

The growing atomization of the traditional parties (the Liberals and Conserva-tives) and the profound fragmentation of the so-called “third forces” (the minorparties) in Colombia have caused two phenomena that feed off one another: onthe one hand, an extreme personalization of political life and, on the other, an in-creasingly particularistic political representation, oriented more toward specificsectors of the population.

In a pioneering work, Carey and Shugart (1995) proposed a method ofestimating the relative value of personal reputation versus party reputation forelected parliamentarians or congressional candidates seeking to further their po-litical careers. Recently, a team from the Inter-American Development Bank re-vived this proposal to measure the impact of electoral systems on the high or lowlevel of particularism in the world (Seddon et al. 2002; Panizza 2001).10 The cen-tral thesis of these studies is that levels of particularism are closely related topoliticians’ predominant reputation. If the party label is central to politicians’


electoral prospects, politicians tend to value projects of collective interest thatbenefit the parties as channels for aggregation and articulation of social interests.If, on the contrary, a personalistic reputation is central to politicians’ election,politicians tend to favor their personal reputation and, therefore, the interests ofspecific social segments in which their political influence is rooted. The most im-portant conclusion of the IDB study is that both excessively party-dominant sys-tems (in which parties develop corporative interests and pay little attention to thedemands of voters) and excessively personalistic systems (in which politicians’subordination to local or circumscribed interests prevents them from generatingnational agendas) have the most negative effect on good legislative governancethat can lead to economic development. Conversely, intermediate systems inwhich stable party systems exist—whether strong (Great Britain) or weak (theUnited States), but coupled with channels for integration and accountabilityamong the voters and their representatives—are ideal. This might be achievedin the one-member districts found in some countries.11

To determine the impact of electoral institutions on the type of reputationand, consequently, the model of representation (national or particularist), theauthors use three variables: the nomination of candidates (or ballot), whethervotes are pooled so that votes for one candidate of a party can benefit other can-didates of the same party, and the vote. Each receives a rating of 0, 1, or 2, wherethe lowest number indicates that the system favors parties, while the highestnumber indicates that the system tends to favor personalistic attitudes amongcandidates and legislators.12 The “Index of Political Particularism” measures theexisting incentives for politicians to build political support based on particularelectoral districts rather than adherence to party platforms. A high score on thisindex indicates that the system is candidate-centric with strong incentives forpoliticians to cultivate, above all, circumscribed interests. A low score is associ-ated with party-centric electoral systems.

In Colombia, the electoral and nomination systems motivate politicians tocultivate personal reputation, therefore favoring particularist representation. Ini-tially, this would seem a contradiction, given the fact that the system of propor-tional representation with closed lists tends to favor—as proven in other inter-national experiences—the role of the party elite in selecting candidates andranking them on the ballot. Proportional representation (PR) with closed listsshould in theory encourage a highly party-centric system. However, due to anumber of factors, PR in Colombia has transformed into what Carey andShugart (1995, 429) refer to as a “Personal-List Formula.” The origins of thissituation, or its institutional dimension, lie in several factors: first, the total free-dom of legally established parties to present a limitless number of lists of candi-dates for public offices; second, the ridiculous nature of the current CNE (Na-tional Electoral Council) standards for forming a political party or movement;13

and third, the almost total absence of entrance barriers to the electoral arena.


According to Article 108 of the 1991 Constitution, political actors are thoseparties, political movements, social movements, or significant groups of citizensable to participate in elections through endorsement, the creation of a party ormovement recognized by the CNE, or payment of a registration fee known as acaución mínima.14

Let us analyze each of the three variables (ballot, pool, and vote) separately.

Nomination of Candidates

The selection of political candidates by parties is fundamental to a democraticsystem. As Gallagher and Marsh (1988, 1) affirm, “The quality of the selectionof candidates determines the quality of representatives elected to parliament, often of the members of government and, by extension, national politics. Achange in the selection of party procedures in any country has profound impli-cations on the way politics function.”

A 1999 issue of the journal Party Politics reflected on the current function ofpolitical parties, recognizing that party functions are not what they once werebut that, nonetheless, the nomination of candidates continues to be an essentialfunction. “The modern party is a voluntary association whose declared aim is tobe represented, and to lead the institutions of government, in a given state orpolitical community. To this end, the party regularly engages in fulfilling threecritical functions: (1) Nominating candidates for public office; (2) Adoptingstatements of public policy, primarily in an election platform; (3) Mobilizingsupport for each of the above—candidates (public officials) and policies” (Yanai1999, 7).

The selection of political candidates is undergoing changes on an interna-tional level. In Western Europe, for example, parties are more open to having agreater number of their members involved in choosing parliamentary candi-dates. What is the reason for this change? Why, in spite of the fear aroused bythe weakening of parties in the United States due to the role of the primaries,are European parties opening, however timidly, the floodgates for popular par-ticipation? Many analysts relate this change to the weakening of linkages be-tween parties and their voters. Many parties are experiencing a reduction in thenumber of members or activists (Mair and van Biezen 2001) and are adoptingdifferent strategies for recovering their ties to voters. One such strategy is “thedemocratization of the selection of candidates” (Pennings and Hazan 2001,268). By increasing the number of people involved in selecting the candidates,parties seek to strengthen the feeling of belonging to the party.

In Colombia, the process by which candidates are selected for legislative bod-ies was not democratized: it was disjointed. There is no selection of candidates“from above” (i.e., from the party leaders). In fact, there is no selection of can-didates at all. Candidates are self-chosen and the parties simply grant them their


respective endorsements to participate in the elections on their behalf—in theso-called “Carnival of Endorsements.”15 Add to this indiscriminate concessionof endorsements the near absence of CNE requirements for recognition of a po-litical party or movement, or the simple chance to participate by paying a smallfee (caución mínima), or signing a responsibility policy (póliza de seriedad) (Art. 9,Law 130 of 1994), and, for all practical purposes, there are no entrance barriersto the electoral system.16

Table 3.1 illustrates the degree to which the electoral system has been frag-mented—in other words, its degree of personalism. This is the main conclusionof a report by a commission of international consultants—comprised of ArturoValenzuela, Joseph Colomer, Arend Lijphart, and Matthew Shugart—undercontract to Andrés Pastrana’s government: “Colombia’s current electoral systemis the most ‘personalistic’ in the world” (Lijphart et al. 1999, 237).

Although political parties currently control the party labels, which can beused only with the party’s endorsement,17 in practice, given the lack of entrancebarriers and the fact any candidate who wishes to run can do so through themechanisms described earlier (endorsement, fee, or simply a new label), the tra-ditional parties, as well as many new parties, make indiscriminate endorsements.In the literature on Brazil these are called “rental parties”—that is, parties thatare nothing more than flexible umbrellas designed to shelter all kinds of per-sonalist factions.

Table 3.1

Number of Lists that Registered for Senate and Lower-Chamber Elections, 1958–2002

Year Senate House

1958 67 831960 — 1131962 97 1431964 — 1921966 147 2151968 — 2211970 206 3161974a 176 2531978 210 3081982 225 3431986 202 3301990 213 3511994 251 6281998 319 6922002 321 883

SOURCE: Registraduría Nacional del Estado Civil.a Members of the House were elected, from 1958 until 1970, for a

period of two years. Beginning in 1970, terms were changed to cor-respond with those of the Senate (four years).


The system that allows a legislator to resign and turn his or her seat over tosomeone else who ran for office under the same party label has reinforced fac-tionalism. The Constitutional Assembly of 1991 put an end to the old, muchcriticized system of suplentes that had existed until that time, whereby the prin-cipal and the suplente rotated in the seat.18 When Article 134 of the 1991 Con-stitution (“Vacancies due to complete absences of members of Congress will befilled by unelected candidates, according to the order of inscription on the cor-responding list”) became law, it opened the way for a completely disjointed sys-tem known as the carrousel in Colombian parliamentary jargon. Legislative Act 03of 1993 perversely replaced the figure of the personal suplente: now all memberson the list are potential “substitutes who can assume the seat of the legislator.”According to this legislative act, a member of Congress can request an unpaidleave of absence for three months, renewable. Leaves are granted in cases of tem-porary absence, suspension of investiture by virtue of a judicial decision, inca-pacity certified by an official measure, domestic calamity duly proven, and casesof force majeure. Owing to this lax regulation, members of Congress plan inadvance within their “electoral micro-enterprise” the amount of time eachmember on the list will spend occupying a particular congressional seat. This per-verse mechanism is completely functional for the purposes of electoral micro-enterprises. These are constructed from an agglomeration of votes from differ-ent political leaders in a region, with the goal of building a viable electoralenterprise—that is, one capable of capturing a seat based on electoral remain-ders (residuos) (i.e., winning a seat after the first seats in an electoral district are al-located by the electoral quotient for that district) in the House or Senate.

A study by the Visible Congress Project (Congreso Visible) of the Universityof the Andes revealed that, by the end of 2001, 247 substitute legislators hadserved in the Senate and the Chamber of Deputies between 1998 and 2001without being directly elected. These individuals had occupied the second,third, fourth, and even lower positions on the list that had originally won thecongressional seat.19 This is one of the greatest aberrations of the Colombian po-litical system, allowing persons not elected by popular vote to assume legislativeoffice.

Substitutes who failed to win election routinely become members of Congresswhen the elected legislator takes a leave. The phenomenon not only reinforcesthe large-scale atomization of the current Colombian Congress—since the po-litical aims of the substitute member of Congress may differ from those of theelected parliamentarian—but also inhibits the formation of a professional polit-ical class. With regard to the subject at hand—political representation—its im-pact is to reinforce particularism in Congress. Whether their stay in Congress islong or short, legislators are interested only in successfully promoting those lawsthat directly affect their “electoral fiefdom.”


“War of the Leftovers”

The disjointing of the Colombian electoral system, in which only occasionallydo single-party lists or candidates exist, generates a formidable intraparty waramong multiple lists and candidates in each electoral district. This system has noparallel elsewhere in Latin America. As Rodríguez maintains (2003, 6), “Theelectoral formula—i.e., the specific way votes are translated into seats—rein-forces this tendency. The Colombian electoral system allocates votes using theHare quota, without vote pooling at the party level. This formula contains nostructural incentives to enhance internal party cohesion. It disproportionally re-wards small lists, increasing intraparty competition.”

With this organizational weakening of the parties and their institutionalizedfactions, it is impossible to speak of “vote pooling at the sub-party level.” Thevotes that one list wins cannot be transferred to another list, even within thesame party in the same electoral district. Therefore, there is zero-sum competi-tion among the many lists of the same party in the same district. If vote poolingoccurs at all, it is within each individual list.

Seats are allocated first according to how many electoral quotients a partyreaches in a district. The electoral quotient equals the number of seats dividedby the number of votes in a district. After this, lists win seats according to theorder determined by the Hare quota of electoral remainders.

As Table 3.2 shows and Cox and Shugart (1995) argued, lists that win morethan one seat are disappearing. The Colombian electoral system has come to re-semble, in spirit if not in letter, the pre-1993 Japanese system of the single “non-transferable vote.” As seen in Table 3.2, in the 1991 elections, nine lists electedmore than one senator, obtaining a total of 34 of the 100 elected senatorial seats.In contrast, in subsequent elections only three lists have elected more than onesenator. In 1991, sixty-six lists were able to elect only the head of their list,ninety-four lists did so in 1994 and 1998, and ninety-three elected only the headof their list in 2002.

Table 3.2

Electoral Performance of Lists That Elected More Than One Senator, 1991–2002

Seats won by Seats won by Total seats Number electoral quotients Hare remainders elected by lists

of lists that (lists that elected (lists that elected that elected elected more than more than more than more than

Year one senator one senator) one senator) one senator

1991 9 26 8 341994 3 3 3 61998 3 3 3 62002 3 4 3 7

SOURCE: Registraduría Nacional del Estado Civil.


At all levels, from the municipal to national, the overwhelming majority ofColombian electoral lists are unable to elect more than one candidate. This phe-nomenon is even more pronounced at the local (municipal council members)and regional (departmental assemblies) levels, or in lower-house elections (withdepartmental constituencies), than in the Senate, due to the high costs of cam-paigning for a list with a national constituency.

The Ballot

Due to the large number of electoral lists and candidates that represent the samepolitical party or movement in each district, except in special situations a votercannot support a political party’s or movement’s official candidate or list. Instead,she must support the candidate or list of one of the factions competing in thatdistrict. For example, during the 2002 senatorial elections, the National LiberalDirectorate (Directorio Nacional Liberal) did not present just a single list, butinstead endorsed 148 lists, each representing the Colombian Liberal Party. TheCNE recognized not only an assortment of lists representing the different liberalfactions, but also 26 lists of autonomous “parties or movements.” The NationalConservative Directorate presented 25 official lists and 31 dissident lists fromamong its different internal factions. The result has been brutal intraparty com-petition in which personal reputation, above all, is used to distinguish oneselffrom the dozens of “official” candidates within the same party. In this respect,the Colombian electoral system, according to Carey and Shugart’s criteria, re-ceives the highest possible score of two (“each voter casts one vote, either for acandidate or a party faction”).

The electoral ballot reinforces this phenomenon. Distributed by the statethrough the National Registry of Civil Status (Registraduría Nacional del EstadoCivil), the ballot contains a photograph of the person heading the list. In the2002 congressional elections, the ballot was an enormous poster with 322 pho-tographs. Voters were thus able to see the head of each respective list, but noneof the other candidates appeared on the ballot.

National Constituency and Particularism

One of the principal political-electoral reforms of 1991 was the creation of a na-tionwide electoral district for the Senate. Since then, all 100 senators have beenelected in a single nationwide district. Members of the Constituent Assemblysustained two motivations to justify this reform: first, they argued that politicalminorities would have easier access to Congress through a larger district giventhat, by way of a “disperse strategy” (Botero 1998), these groups would be ableto win votes throughout the entire country and gain one or more seats; and sec-ond, they felt that the creation of a nationwide district could serve as an incen-tive for new national leadership and projects.


The first objective was fulfilled. The scope of political and social representa-tion has widened, as seen in Table 3.3. Many parties have gained representationin the Senate. The second objective, to the contrary, has generally speaking beenan enormous failure, due to the spreading phenomenon of “personalist factions.”What happened?

In general, small districts tend to favor personalism and particularism. For ex-ample, Thomas Lancaster (1986, 70) has stated that “due to the fact client-based

Table 3.3

Senate Seats Won by Political Parties and Movements, 1991–2002

Party or Movement 1991 1994 1998 2002

Partido Liberala 59 59 61 53Partido Conservadora 26 32 27 26Unión Patriótica–Partido Comunista Colombiano 1 1Acción Democrática M-19 9Movimiento Unitario Metapolítico 1 1Partido Nacional Cristiano 1 1Unión Cristiana 1 1Laicos por Colombia 1 1Movimiento Compromiso Cívico Cristiano 1 1 1Alianza Nacional Popular 1 1 1Movimiento Autoridades Indígenas de Colombia 1Movimiento Obrero Independiente y Revolucionario 1Educación, Trabajo y Cambio Social 1 1Movimiento Cívico Independiente 1 1Movimiento Convergencia Popular Cívica 1 1Movimiento Ciudadano 1 1Alianza Social Indígena–MCI–Confiar Antanas 1Movimiento Bolivariano 1Vamos Colombia 1Movimiento de Defensa Ciudadana 2Movimiento Independiente Frente de Esperanza 1Alianza Social Indígena 1 1Movimiento Fuerza Independiente 2Movimiento Popular Unido 1Movimiento Ciudadanos por Boyacá 1Movimiento Convergencia Ciudadana 1Movimiento Dejen Jugar al Moreno 1Movimiento Huella Ciudadana 1Movimiento Independiente de Renovación Absoluta 1Movimiento Progresismo Democrático 1Movimiento Somos Colombia 1Movimiento Unionista 1Partido Social Demócrata Colombiano 1Partido Unidad Democrática 1Movimiento Vanguardia Social y Moral 1

Total 100 100 100 100

SOURCE: Registraduría Nacional del Estado Civil.aThe figures for the two traditional parties include all the factions that belong to those parties.


(pork barrel ) projects are distribution policies directed at geographically definedvoters, a geographically unlimited electoral system must be related to some de-gree of client-based activity.” Conversely, large districts are better suited to thecreation of national agendas supported by nationwide parties.

Carey and Shugart share this point of view, but only in cases where intrapartycompetition does not exist. According to them, with closed-list proportionalrepresentation, as the size of an electoral district grows, and therefore as thenumber of candidates on a list grows, party reputation will be the dominant fac-tor. With other electoral formulae, on the contrary, increasing district size canheighten the importance of individual reputation since each candidate mustcompete with candidates from his or her own party. “As the number of co-partymembers from which a candidate must distinguish himself increases, so does theimportance of establishing a personal reputation which is distinct from that ofthe party” (Carey and Shugart 1995, 430). As we have shown, intraparty com-petition in Colombian elections at all levels is fierce, owing to the dozens of lists(or candidates) that represent all the political parties or movements. This intra-party fighting is even fiercer in the Senate due to the number of seats (100) andcandidates involved in the countrywide district. In the Senate, candidates havea great need to differentiate themselves from others within the party by meansof a defined personal image. This need to emphasize one’s personal reputationin turn accentuates the need to build one’s own “electoral niche.” As demon-strated by Felipe Botero (1998), the weight of the local vote continues to be adetermining factor in electing senators. On average, 66.72 percent of senatorswere elected by votes coming from just one department in elections between1991 and 1998. In sum, not only do the three variables (ballot, pool, and vote)tend to generate an extremely personalistic system, but national constituencies,far from favoring the formation of national leadership and reinforcing nationalparties, have instead deepened the extreme atomization of the political partiesin Colombia.

In summary, Colombia scores a 2–2–2 with regard to Carey and Shugart’sthree variables. This is the highest possible level of personalism, comparable onlyto the single non-transferable vote (SNTV) system that existed in Japan and theSNTV open-endorsement system used in the Philippines. Carey and Shugart(1995) gave Colombia a score of 2–1–2, presupposing the existence of poolingat the individual list level. As we have seen, although electoral norms foresee thispossibility if a list wins more than one seat, in fact, few lists win more than oneseat in the Senate, and almost no lists win more than one seat in other legisla-tive bodies (the Chamber of Deputies, departmental assemblies, and municipalcouncils). Furthermore, the national constituency for the Senate reinforces theimpact of the electoral system, owing to the high degree of intraparty competi-tion generated by elections to this body.


“Electoral Niches” and Lost Votes

The transition from institutionalized factions to personalistic factions in the tra-ditional parties, as well as the severe fragmentation among the minor partiesknown in Colombia as “third forces,” has driven each parliamentary group, andeven each individual member of Congress, to seek a determined “electoralniche”—that is, a specific social sector as a foundation for electoral support. This“specific voter,” whether through client-based, corporatist, regional, ethnic, orreligious networks, is the object of countless diverse “micro-representations.”20

There are many examples, among them the Liberal Party’s Social SecurityMovement (Movimiento por la Seguridad Social). Led by Liberal senator Al-fonso Angarita Baracaldo, the movement provides congressional representationto retired people who receive state pensions, and it monopolizes parliamentaryinitiatives that favor this social sector. Another example is the Education, Work,and Social Change Movement (Movimiento Educación, Trabajo y Cambio Social) led by a senator from the independent left, Jaime Dussán, who is thespokesman for public school teachers. The Christian churches are dividedamong those who follow the Christian Union (Unión Cristiana) and support-ers of the National Christian Party (Partido Nacional Cristiano). Indigenouscommunities are split between the Social Indigenous Alliance (Alianza SocialIndígena) and the Indigenous Authority Movement of Colombia (MovimientoAutoridades Indígenas de Colombia).

These personalist factions of traditional parties, micro-parties, or micro-movements founded to exploit the wishes of “specific voters” do not make upthe bulk of the members of Parliament. In numerical terms, they are a small par-liamentary minority, but they have enormous political significance. They aretypically elected by independent urban voters, generally well-educated individ-uals with higher than average incomes (Pinzón 1998, 407). These members ofCongress, whose electoral support is based on programmatic issues (in Colom-bia referred to as voto de opinión, or opinion vote), do not require huge sums ofmoney for their political campaigns because they are all well known and highlyregarded. These legislators therefore tend to be more critical of political cor-ruption. The absence of a “specific electorate” allows them greater autonomythan most members of Congress have, and in general they are more apt to pro-pose and defend national agendas. This means that the existence of national andparticularist agendas as a dividing line between the executive and legislativebranches contains one notable exception, in that these members tend to be themain leaders in Congress for both the government and the opposition. To a largeextent, if the Colombian Congress occasionally produces some worthwhile leg-islation, it is because of presidential leadership and the leadership of these moreprogrammatically oriented members of Congress.21


One of the most negative consequences of this extreme atomization of polit-ical representation—beyond the disproportionate growth of lists and candidatesunder the rubric of dozens of political parties, micro-parties, and movements—is the increase in wasted votes. As political labels proliferate along with the num-ber of lists and candidates, the number of unrepresented sectors has also in-creased. Cumulatively, a large number of electoral lists fail to win any seats inCongress but still win a substantial share of the vote; the citizens who vote forthese lists are effectively unrepresented. In 1998, 222 lists for the Senate—70 percent of all the lists—won a total of 2,540,000 votes without winning anyseats. In the Chamber of Deputies, the situation is even more worrisome: of the696 lists that competed in elections that same year, more than 500 were leftwithout representation. At the level of departmental and municipal legislatures,the situation is even more dramatic. Particularist representation can contributeto improving political representation for some social sectors (e.g., indigenouscommunities),22 but on the whole, there are more losers than winners. Facedwith the parties’ inability to aggregate and represent collective interests, the cor-poratist, regional, or other types of micro-representations come together in agame with few winners and many losers.

Parliamentary Administration

How does this atomization and personalization of party life affect Congress? Thecentral hypothesis of many studies on this subject is that “the more personalis-tic the vote, the more individualistic the legislator’s conduct. And conversely,the more partisan the vote, the more partisan parliamentary conduct will be”(Amorim Neto and Santos 2001, 214). In Colombia, the predominance of thepersonalistic vote results in severe internal fragmentation of all popularly electedbodies, especially Congress, which generates party indiscipline.23 This correla-tion between personalization of the vote, fragmented representation, and con-gressional indiscipline should surprise no one since, as recent studies demon-strate, the more fragmented the Congress the greater the likelihood of strategicconduct and manipulation of the legislative agenda by individual legislators.24

Such constant negotiation between the government and individual members ofCongress in order to sway the vote drastically increases the transaction cost ofmoving legislation forward.25

A study by the Center for Socio-Legal Investigations (Centro de Investiga-ciones Sociojurídicas, or CIJUS) at the Universidad de los Andes (Bejarano et al.2001) illustrates this relationship. This study analyzed the legislative period fromJuly 20, 1998, through July 20, 1999. Immediately striking is the number ofproposed laws (354) and legislative acts (35) processed. According to Brian Crisp(1999), “legislative inflation” in Colombia contrasts sharply with the moderate


level of legislative activity in Venezuela, where in forty years only some 750 lawswere proposed in the Chamber of Deputies, among which “almost none werelocal or regional.”

Various factors explain this “legislative inflation.” First is the lack of congres-sional parties or caucuses as a source of coherent parliamentary agendas, whichgenerates a lack of filters to control the quality, financial sustainability, and legalrigor of proposed legislation.26 Second, as noted earlier, is the phenomenon of“personalist factions” as political and electoral units. The authors of the CIJUSstudy examined whether proposed bills were local, regional, or national. Theydiscovered that 78 percent of the laws proposed by legislators had a strictly lo-cal or regional impact. Only 22 percent of the bills addressed national issues.

In addition, the authors of the study also noted a proliferation of local or re-gional proposals presented by the suplentes on the list. “The suplente phenome-non . . . allows persons who were not popularly elected to use their time in Con-gress to improve their political pull in certain provinces by proposing legislativeprojects” (Bejarano et al. 2001, 232). The suplentes propose, above all, projectsthat honor regional personalities or commemorate dates of local interest, or lawsthat authorize the issuance of postal stamps as a way to raise departmental or mu-nicipal revenue. The proposed honors, commemorations, and stamps are alwaysaccompanied by local investment projects.

While a significant amount of proposed legislation in Congress is targeted at a“specific beneficiary,” proposals originated by the government are usually ori-ented toward addressing national problems. This is what I call “dual representa-tion.” The aforementioned CIJUS study of the origins of proposed laws or leg-islative action made an important discovery: “In no other case were regional orlocal proposals presented by the rest of the state. The government and otherbranches of public power are in charge of looking out for national interests, whilein Congress, these national interests are frequently mixed in with regional orlocal proposals” (Bejarano et al. 2001, 233). As is the case for Brazil (see AmorimNeto and Santos 2001), in Colombia it is because of the “executive connection”that the governability of the political system does not disintegrate into extremeparticularism. The undisciplined vote is not prevalent in parliamentary voting onproposals of governmental interest.27

In Colombia, the executive makes use of a wealth of resources (constitutionaland informal) for gaining support for proposed laws. In the constitutional arena,the president enjoys the privileges of parliamentary initiative (Art. 154), design ofthe national budget, total or partial veto of laws approved by Congress (Art. 167),exclusive legal initiative on certain subjects (Art. 154), and legislative emergency(Art. 163), among other prerogatives. In addition, as is true in many countries,the president can also use informal resources, especially pork barrel legislation, toinfluence the conduct of members of Congress. Needless to say, most membersof Congress are anxious to obtain these resources. Members of Congress with


“specific voters” are, in practice, a kind of lobby working for their fiefs or cor-porative interests. As such, they simply negotiate their vote on behalf of cer-tain government proposals in exchange for pork barrel resources. But with fewexceptions, they never commit to unconditional and lasting support to thegovernment. Support must be negotiated on each occasion. In this way, inColombia today, exchanging pork barrel resources for parliamentary support isone of the components of the legislative/executive relationship and congressionalconduct.

Conclusion

Colombia has not escaped the party crisis affecting the Andean region. Therehas been no collapse of the party system such as happened in Peru and Venezu-ela. Yet the level of disintegration of the traditional Liberal and ConservativeParties and the severe splintering of the “third forces” seriously affects the gov-ernability of the state and the very legitimacy of democratic institutions. If werely on the results of the 2002 Latinobarómetro, the level of support for and trustin democracy, political parties, and the Congress is very worrisome. Only 39percent of Colombians support democracy, as opposed to an average of 53.2percent throughout the Andean region and 56 percent in Latin America gener-ally. Satisfaction with democracy is even lower: 11 percent compared with 21.8percent among the Andean nations and 32 percent for the entire subcontinent.Only 10 percent of Colombians trust the parties, and 14 percent the Congress.

The roots of this deterioration of citizen support for and trust in democraticinstitutions are complex. However, the level of organizational deterioration inthe parties and the party system as a whole has had a pronounced influence. Ina system splintered into hundreds of particularist micro-representations, thenumber of losers in the electoral game increases exponentially. The electoral andparty systems in Colombia since 1991 have not only aggravated the work of gov-erning bodies on all levels, but have helped to heighten the overall crisis of rep-resentation in the political system.

Notes

I am extremely grateful for the generous critical comments of Michael Coppedge,Scott Mainwaring, and Ana María Bejarano.

1. Morgenstern (2001) differentiates between institutionalized factions ( Japan, Italy,Uruguay) and the ephemeral factions—in English politics, for example—centering ona specific topic or election. In Uruguay, the principal role of party factions, in both elec-tions and political decision making, has generated debate as to whether parties or factions


should be counted to characterize the system (see, e.g., Solari 1964; González 1991). Thesame debate is present in Colombia (Posada-Carbó 1997).

2. Conservative Party votes fell from 40 percent in the 1982 parliamentary electionsto 23 percent in 1998, which represents a permanent decline. Another manifestation ofthis decline is the absence, for the first time since 1942, of an official Conservative Partycandidate in the 2002 presidential elections. Unlike the PC, the PL has made an effortto renew itself ideologically—through adherence to the Socialist International—and or-ganizationally. Time will tell.

3. I later refer to a small parliamentary nucleus with enormous political influence,whose election by the urban populace grants them greater autonomy and, more impor-tantly, a real incentive to build less particularistic parliamentary agendas.

4. These studies are based on an ever-increasing literature founded on comparativestudies, which shows that there is a clear relationship between electoral institutions (rulesfor nominations, electoral formulae, and scope of districts), electoral strategies, and theconduct of members of Congress (Lancaster 1986; Cain et al. 1987; Mainwaring 1991;Shugart and Carey 1992).

5. The phenomenon of “dual representation” is present at every level of the state. Itis also visible at departmental and municipal levels, where the public policy of governorsand mayors contrasts with the more markedly particularistic vision of the deputies andcouncilmen.

6. Interestingly enough, the Constitution of 1991 sought to strengthen the role ofparliament and to improve the balance between the branches of public power. In practice,the atomization of the party system and, therefore, the absence of parliamentary caucuses,rendered the constitutional text on this issue ineffectual (Pizarro Leongómez 1996, 2001).

7. Mainwaring and Scully were, however, well aware of the dark cloud hanging overthe two traditional parties: “the party system may be entering a dissolution phase afterdecades of considerable stability. . . . Pronounced factionalism is simply a manifestationof the erosion of party organization in recent decades. Factions can introduce their owngroup of candidates . . . ; the loss of organizational control over the selection of candi-dates is extreme” (Mainwaring and Scully 1995, 18).

8. For example, military governments (1953–58) and, above all, the attempt by Pres-ident Gustavo Rojas Pinilla to create a political/military alternative to the parties, the“People–Armed Forces duo.”

9. Or even in certain urban sectors of Latin America. In Colombia, without a doubt,a primary source of the volatile vote in favor of independent candidates comes, above all,from the well-off and well educated (Pinzón 1998).

10. The IDB database covers 155 countries between 1978 and 1997.11. In the words of Shugart (1999, 319): “Personalistic systems fail to provide elec-

tions that turn primarily on collective policy, because of the incentives individual mem-bers have to collect personal votes, which are better captured through small-scale serviceprovision (clientelism). Camarillian systems fail to connect parties with collective policypreferences because individual members have no incentive to find out what voters want.Efficient systems, on the other hand, place members in a position where they must bal-ance the interests of both voters and party leaders.”

12. This is an ordinal scoring system. The nomination (ballot) measures the degree ofcontrol party leaders have over access to party nominations:

0: Party leaders present only one closed list that cannot be altered by voters;1: Leaders present a list that can be altered by voters;2: Leaders have no control over the lists.


Vote pooling measures whether or not the votes received by a party candidate con-tributed to the number of seats won by the party altogether in any given district:

0: A general pooling in favor of the party as a whole takes place;1: Pooling takes place at the party faction level;2: Pooling does not take place.

The vote variable indicates:

0: Voters can cast only one vote in favor of the party.1: They can vote for multiple candidates.2: They can cast only one vote in favor of a candidate or party faction.

13. According to Article 108 of the current constitution, “The National ElectoralCouncil will grant legal status to those political parties or movements . . . proving theirexistence with at least 50,000 signatures, or that obtained at least this same number ofvotes in the last elections, or that attained representation in the National Congress.” Thisconstitutional norm is developed in Article 3 of Law 130 of 1994, “Basic Statutes of thePolitical Parties and Movements.”

14. The caución mínima allowed a candidate or a list that did not have the support of aparty recognized by the National Electoral Council to pay a small fee to register for elec-tions. In the Constitutional Assembly of 1991, a “logic of incorporation” was imposed asa supposedly suitable mechanism for overcoming the two-party system that was perceivedas one of the sources of the national crisis (Pizarro Leongómez and Bejarano, forthcom-ing). Its effects have been contradictory: on the one hand, it opened the floodgates for new political forces; on the other, it accelerated the tendency toward the disintegra-tion of the party system that began with the National Front of 1958–74 (PizarroLeongómez 2001).

15. “La feria de los avales” (The carnival of endorsements), El Tiempo, January 29,2002. In this editorial, Colombia’s most influential daily newspaper denounced the wayin which uncontrolled endorsements from some 75 political parties or movements legallyrecognized by the National Registry of Civil Status was opening the door to candidaciesfrom armed groups (guerrillas and paramilitaries) and drug traffickers. The Liberal Partyalone endorsed 148 lists for the national Senate.

16. As a way of explaining the erosion of the two largest parties from the 1968–88period in Venezuela, some authors have used approaches based on rational choice orgame theory models to explain political behavior, arguing that the decentralizing reformsof 1989 weakened the AD and the COPEI by lowering entrance barriers for new par-ties and encouraging enterprising politicians to abandon or declare their autonomy frompolitical parties (Penfold 2001; Benton 1997). In Colombia, both the 1991 Constitutionand the 1994 Law of Parties created perverse incentives that ended up fueling the cur-rent party atomization (Pizarro Leongómez and Bejarano, forthcoming).

17. Carey and Shugart (1995, 429) erroneously state that candidates do not require“party endorsement to use the party label.” According to Article 5 of Law 130 of 1994,“Political parties and movements are owners of their names and of the symbol they haveregistered with the National Electoral Council.” Article 9 of the same law states that theparty label may be used only with the endorsement of a party’s legal representative.

18. Suplente can be roughly translated as “substitute” or “replacement.” Until the1991 Constitution, both senators and representatives were elected along with a suplente,who could replace the elected member of Congress. Article 261 of the 1991 Constitu-tion abolished the old system of suplentes.


19. Some examples are remarkable: the seat held by the well-known legislator SergioCabrera was finally occupied by the seventh candidate on his list, a complete unknown,Edgar Antonio Ruiz. In the Senate, Gabriel Ignacio Zapata temporarily ceded his seat tothe eleventh candidate on his list, Julio Acosta. “The system is anarchic. The cure provedworse than the disease,” affirmed Senator Luis Guillermo Vélez (El Tiempo, Novem-ber 25, 2001).

20. Japan’s electoral and party systems have been the subject of interesting comparativestudies with the Colombian case. Bouissou (1994, 385) speaks of “mini-parties with asingle cause” (mini seito). In Colombia, the analysis of the political campaigns of clientelis-tic senators and deputies is enlightening. They state their intentions to work on behalf ofeither certain social sectors (actors, athletes, retirees, teachers), specific social movements(community movements, cooperatives, unions), particular ethnic groups (black commu-nities from the Pacific region, Paez indigenous people), or particular regions (northernValle or southern Bolívar provinces). Generally, in Colombia the only caucuses that oper-ate relatively efficiently are the regional caucuses, especially the Atlantic Coast caucus—that is, members of Congress of all political persuasions who were elected from the Atlan-tic coastal region.

21. This is the case, for example, with senators such as Antonio Navarro, GermánVargas, Luis Alberto Ramos, Ingrid Betancourt, Carlos Gaviria, Claudia Blum, RafaelPardo, and a few others.

22. The indigenous communities are one of the social sectors that has most benefitedfrom the new electoral institutions. This is not merely because of the special electoral dis-trict for the Senate (two senators), but also because of the national scope of the electoraldistrict, which has allowed indigenous candidates to amass scattered votes and in this waydouble their chances of widespread political representation (Peñaranda 2002). This hasalso been the case for the black communities of the Pacific coast, the Christian churches,and other sectors. In other words, not everyone is a loser.

23. Tsebelis (1995) defines party discipline as the degree of party unity in congres-sional voting. The “Rice Index” is an operational indicator: RI � percentage of votes infavor minus percentage of votes against.

24. This type of conduct from highly fragmented and undisciplined congressionalrepresentatives in turn leads to high rates of party switching (or transfugismo, as it is knownin Brazil), as well as to the formation of unstable coalitions. Coalitions form with eachnew issue, and every coalition has a different configuration (Amorim Neto and Santos2001).

25. In the United States, it is said, no doubt exaggeratedly, that 535 members of Con-gress equals 535 parties, given the extensive pork barrel legislation that is required to en-sure efficient legislative operations. In Colombia, the 1991 Constitution expressly pro-hibits legislators’ slush funds (called auxilios parlamentarios) for clientelistic purposes. (Theseslush funds originated with the constitutional reform of 1968, which institutionalizedthem as a way to grease congressional approval of constitutional reforms.) Nevertheless,they are constantly revived under different forms: co-financing funds, co-financing fundsfor rural investment (DRI), and special interministerial or interparty funds for the prov-inces, as occurred with the latest National Development Plan (Plan Nacional de Desa-rrollo, 2002–2004). The reason is simple: in an atomized parliament, pork barrel spend-ing is the fuel that feeds the legislative dynamic given the absence of caucuses founded ona certain party or ideology. The 1968 auxilios parlamentarios allowed individual membersof Congress to select their favorite recipients for pork barrel projects.

26. In Colombia, the media have created a perverse form of evaluating congressionalconduct: the number of proposed laws presented by members of Congress, regardless of


their quality, coherence, and significance. With this kind of media pressure, and given thelack of party or other quality “filters,” hundreds of laws are proposed, generating enor-mous legislative gridlock.

27. One noteworthy—though not surprising—exception is the attempts by recentgovernments at political reforms aimed at changing the current rules of the game regarding political parties and legislation. These reform proposals have met with open resistance.

References

Amorim Neto, Octavio, and Fabiano Santos. 2001. “The Executive Connection: Presi-dentially Defined Factions and Party Discipline in Brazil.” Party Politics 7, no. 2:213–34.

Bejarano, Ana María, Laura Zambrano Robledo, Felipe Botero Jaramillo, Laura WillsOtero, and Francisco José Quiroz. 2001. “¿Qué hace funcionar al Congreso? Unaaproximación inicial a las fallas y los aciertos de la institución legislativa.” Estudios Oca-sionales (CIJUS–Universidad de los Andes, Bogotá).

Benton, Allyson. 1997. “Patronage Games: The Effects of Economic Reform on Inter-nal Party Politics and Party System Stability in Latin America.” Paper presented at the1997 meeting of the American Political Science Association, Washington, DC, Au-gust 28–31.

Botero, Felipe. 1998. “El Senado que nunca fue: La circunscripción nacional después detres elecciones.” In Elecciones y democracia en Colombia 1997–1998, ed. Ana María Be-jarano and Andrés Dávila, 285–335. Bogotá: Universidad de los Andes.

Boussiou, Jean-Marie. 1994. “Les élections législatives japonaises du 18 de juillet 1993.”Revue Française de Science Politique 44, no. 3: 379– 423.

Cain, Bruce, John Ferejohn, and Morris Fiorina. 1987. The Personal Vote: ConstituencyService and Electoral Independence. Cambridge, MA: Harvard University Press.

Carey, John, and Matthew Shugart. 1995. “Incentives to Cultivate a Personal Vote: ARank Ordering of Electoral Formulas.” Electoral Studies 14, no. 4: 417– 40.

Cox, Gary, and Matthew Shugart. 1995. “In the Absence of Vote Pooling: Nominationand Vote Allocation Errors in Colombia.” Electoral Studies 14, no. 4: 441– 60.

Crisp, Brian. 1999. “El comportamiento de los congresistas en América Latina.” Paperpresented at the Universidad de los Andes, July 13.

Gallagher, Michael, and Michael Marsh. 1988. Candidate Selection in Comparative Perspec-tive: The Secret Garden of Politics. London: Sage.

González, Luis Eduardo. 1991. Political Structures and Democracy in Uruguay. Notre Dame,IN: Notre Dame University Press.

Lancaster, Thomas. 1986. “Electoral Structures and Pork Barrel Politics.” InternationalPolitical Science Review 7, no. 1: 67–81.

Lijphart, Arend, et al. 1999. “Sobre la reforma política en Colombia. Informe de la Con-sultoría Internacional.” In Reforma Política: Un propósito de nación. Memorias. Serie Doc-umentos, no. 17. Bogotá: Ministerio del Interior.

Losada, Rodrigo, and Eduardo Vélez. 1981. Identificación y participación política en Colom-bia. Bogotá: Fedesarrollo.


Mainwaring, Scott. 1991. “Politicians, Parties, and Electoral Systems: Brazil in Compar-ative Perspective.” Comparative Politics 24: 21– 43.

Mainwaring, Scott, and Timothy Scully. 1995. Building Democratic Institutions: Party Sys-tems in Latin America. Stanford, CA: Stanford University Press.

Mair, Peter, and Ingrid van Biezen. 2001. “Party Membership in Twenty European Democracies, 1980 –2000.” Party Politics 7, no. 1: 5–21.

Morgenstern, Scott. 2001. “Organized Factions and Disorganized Parties: Electoral In-centives in Uruguay.” Party Politics 7, no. 2: 235–56.

Panizza, Ugo. 2001. “Electoral Rules, Political Systems, and Institutional Quality.” Eco-nomics and Politics 13, no. 3: 311– 42.

Peñaranda, Ricardo. 2002. “Los Nuevos Ciudadanos: Las Organizaciones Indígenas en elSistema Político.” In Degradación o cambio: Evolución del sistema político colombiano, ed.Francisco Gutiérrez Sanín, 131–81. Bogotá: Grupo Editorial Norma.

Penfold, Michael. 2001. “El colapso del sistema de partidos en Venezuela: Explicación deuna muerte anunciada.” In Venezuela en transición: elecciones y democracia 1998–2000, ed.José Vicente Carrasquero et al., 36 –51. Caracas: RedPol–CDB Publicaciones.

Pennings, Paul, and Reuven Hazan. 2001. “Democratizing Candidate Selection: Causesand Consequences.” Party Politics 7, no. 3: 267–75.

Pinzón, Patricia. 1998. “Una aproximación al voto urbano: El voto en las ciudadescolombianas.” In Ana Elecciones y democracia en Colombia 1997–1998, María Bejaranoand Andrés Dávila, eds., 401–32. Bogotá: Universidad de los Andes.

Pizarro Leongómez, Eduardo. 1996. “La crisis de los partidos y los partidos en la crisis.”In Tras las huellas de la crisis política, ed. Francisco Leal, 205–34. Bogotá: Tercer MundoEditores/IEPRI.

———. 2001. “Colombia: ¿Renovación o colapso del sistema de partidos?” In Colombiaante los restos del siglo XXI: Desarrollo, democracia y paz, ed. Juan Ibeas and Manuel Al-cántara. Salamanca: Ediciones de la Universidad de Salamanca.

———. 2002. “La atomización partidista en Colombia: El fenómeno de las micro-empresas electorales.” In Degradación o cambio: Evolución del sistema político colombiano,ed. Francisco Gutiérrez, 357– 401. Bogotá: Editorial Norma-IEPRI.

Pizarro Leongómez, Eduardo, and Ana María Bejarano. Forthcoming. “Political Reformin Colombia after 1991: Is There Anything Left to Reform?” In Democracy, Peace, andHuman Rights in Colombia, ed. Christopher Welna and Gustavo Gallón, Notre Dame,IN: Notre Dame University Press.

Posada-Carbó, Eduardo. 1997. “Limits of Power: Elections under the Conservative He-gemony in Colombia, 1886 –1930.” Hispanic American Historical Review 77, no. 2:245–79.

Randall, Vicky, and Lars Svasand. 2002. “Party Institutionalization in New Democracies.”Party Politics 8, no. 1: 5–29.

Roberts, Kenneth. 2001. “La descomposición del sistema de partidos en Venezuela vistadesde un análisis comparativo.” Revista Venezolana de Economía y Ciencias Sociales 2:183–200.

Rodríguez, Juan Carlos. 2003. “The Regional Diversity of Electoral Competition inColombia.” Paper presented at the 24th LASA Congress, Dallas, March 27–29.

Seddon, Jessica, Alejandro Gaviria, Ugo Panizza, and Ernesto Stein. 2002. “Political Par-ticularism around the World.” Working Paper, no. 463, Research Department, Inter-American Development Bank, Washington, DC.

Solari, Aldo. 1964. Uruguay: Partidos políticos y sistema electoral. Montevideo: Fundaciónde la Cultura.

Shugart, Matthew. 1999. “Efficiency and Reform: A New Index of Government Re-sponsiveness and the Conjunction of Electoral and Economic Reform.” University ofCalifornia, Irvine (mimeo).


Shugart, Matthew, and John Carey. 1992. Presidents and Assemblies: Constitutional Designand Electoral Dynamics. Cambridge: Cambridge University Press.

Tsebelis, George. 1995. “Decision Making in Political Systems: Veto Players in Presi-dentialism, Parliamentarism, Multicameralism, and Multipartism.” British Journal ofPolitical Science 25, no. 3: 289–325.

Yanai, Nathan. 1999. “Why Do Political Parties Survive?” Party Politics 5, no. 1: 5–17.

Among the many causes alluded to when explaining the problems of the An-dean countries, and especially those of Ecuador, the crisis of representation hasgrown in importance in recent years. Scholarly analyses as well as politicians re-fer to a “crisis of representation” as an unquestionable fact that obstructs policymaking and implementation (Barrera 2001; F. Bustamante 2000). Allegedly,deficiencies in representativeness result in problems of governability and condi-tions unfavorable to the consolidation of democracy. This perspective suggeststhat those deficiencies derive from the voters’ dissatisfaction with the meager re-sults of politicians’ actions in their role as authorities of popular representation,and that, at the same time, this dissatisfaction leads to mistrust not only of thepeople involved but of the institutions and the system as a whole. As a result, fol-lowing a period of exploration as voters experiment with different options, theyfinally reject representative democracy and focus on alternative options, rangingfrom seemingly innovative proposals to the election of anti-system leaders.

The validity of this analysis hinges on the relationship between the expecta-tions and the results of political representation. How representation is evaluateddepends on the returns derived from the representatives’ actions, which also sup-poses that voters expect those actions to bring about specific results. This analy-sis is therefore based on voters’ expectations on the one hand, and the results ofauthorities’ actions on the other. The crux of the analysis of representation liesin the relationship between voters and their representatives, not in an isolatedanalysis of each. What needs to be addressed is the bond between voters and rep-resentatives, or, in other words, the mandate emerging from voter expectations.An investigation of this relationship has been one of the weak points in Ecuador-ian studies of representation. Most such studies have focused on either voters orrepresentatives but not on both at the same time, and even less so on the re-lationship between the two.

4

Ecuador: The Provincialization of Representation

❖

Simón Pachano

Ecuador: Provincialization of Representation 101

The role of the political parties as fundamental actors in this relationship hasgarnered much attention in recent years and provided important clues withwhich to ascertain the nature of representation in Ecuador. Most studies haveemphasized the conditions under which representation is carried out (León1994; F. Bustamante 2000), especially representatives’ role in the clientelisticand corporatist practices that characterize politics in Ecuador. Recent studieshave concentrated on parties’ ideological orientation (Freidenberg 2000) andpolitical culture (Burbano 1998), as well as internal organization (Freidenbergand Alcántara 2001).

Party dispersion, fragmentation, atomization, uncertainty, and volatility arehighlighted in these studies. Most allude to negative or problematic aspects ofparties that impede their capacity to carry out their responsibilities (Conaghan1994; Arias 1995; Mejía 1998; F. Bustamante 2000). The limited capacity forrepresentation—regardless of how one understands it—particularly stands out,generally with respect to the predominance of clientelistic and corporatist prac-tices, as well as personalism. These analyses focus on the parties’ problems or in-ability to carry out their responsibilities; few point to the parties’ ability to sur-vive in a hostile environment. This is a good starting point, but it is necessaryto go further and explain this capacity to survive. There is no doubt that a cri-sis of representation exists, but it is important to know what that means.

Despite negative public opinion and even contradicting actions taken to un-dermine their weight and influence, the parties have secured a role as vehicles forpolitical representation.1 In the post-1979 period, four parties—Partido SocialCristiano (Social Christian Party, or PSC), Partido Roldosista Ecuatoriano(Ecuadorian Roldosist Party, or PRE), Izquierda Democrática (Democratic Left,or ID), and Democracia Popular (Popular Democracy, or DP)—have consoli-dated and together have won about three-fourths of the vote. This has occurredwithin the framework of a highly fragmented and atomized system. Just as im-portant, however, is the increasing share of the vote that these parties have man-aged to accumulate over time. One of the prominent characteristics of theEcuadorian party system is this apparently contradictory combination of frag-mentation and concentration. The large number of parties that win seats inCongress and gain access to representational positions in provincial and local as-semblies is offset by the predominance of a relatively small number of parties.Generally speaking, the parties have demonstrated a greater ability than inde-pendents to secure voters’ support.

This chapter uses as a starting point this ability of the Ecuadorian parties to sur-vive in a hostile environment in order to propose an alternative understanding ofthe problems concerning representation. I argue that there are other forms andmechanisms of representation and that all the political actors, especially the par-ties, can adapt to them. The survival of the parties is due to their ability to adaptto conditions that are not necessarily part of the institutional design of the

102 Pachano

political system. This adaptation clears the way for complex situations affectingparty consolidation since it requires renunciation of many of the classic functionsallegedly performed by parties in a democratic regime. Because of the type of re-lationship parties have with their electorates, they are forced to produce resultsthat do not satisfy the expectations of the population as a whole. The parties mustbe rooted in territorially and socially defined groups of voters in order to survive.This situation transforms parties into expressions of partiality and not of a pub-lic good, and leads them to develop a great ability to represent specific interestsand local arenas, but also leaves them with an enormous deficiency in represent-ing national interests. The main argument of this chapter is that the provincial-ization of parties—that is, their predominant focus on provincial issues—has fa-cilitated their survival but has also caused their main deficiencies.

I do not deny that there is a crisis of representation, but I try to identify thenature of this crisis in Ecuador, where the term seems too broad and may causeconfusion. As it has been applied to Ecuador, the expression “a crisis of repre-sentation” confuses several different levels, and it treats different kinds of prob-lems indiscriminately. Most analyses of a crisis of representation focus on threecentral themes: the political system’s outputs, the structure or formation of rep-resentative institutions, and the concept of political representation.

When analysts refer to the outputs yielded by the political system, they em-phasize the poor social and economic performance throughout the post-1979period.2 From this perspective, problems of representation derive from the in-ability to satisfy the demands of society. However, it is not clear to what extentthis failure to produce better results is a consequence of the system of represen-tation rather than of non-political factors. The government’s—and, in general,democratic institutions’—problems of efficiency cannot be attributed entirelyto the forms, mechanisms, and procedures of political representation. Certainly,representation has an effect on government because it contributes to the forma-tion of governments and establishes limits for governments and assemblies. Inthis sense, the forms of representation are one of the means of attaining the goalsof formation of both decision-making instances and operative institutions, butthe degree to which they are successful or the importance of their role in theseinstances is precisely what ought to constitute the focus of our analysis. This willbe the object of the first part of this chapter, which analyzes the main charac-teristics of the electoral system and especially the structure and formation of theNational Congress and its relationship to the executive.3

The representativeness of the popularly elected bodies and officials—the second theme in this chapter—focuses on one of the political system’s mainproblems, not only in Ecuador, but also in conceptual terms. Institutional architecture, design, and procedures are fundamental factors in the study of representation. A lesser or greater capacity for inclusion of the different socialactors, interests, and conflicts depends significantly on the design of represen-tative institutions. The degree of satisfaction with representation itself—not


necessarily with the results of the political system, which is another matter—depends to a great extent on institutional design. With a few exceptions (Mejía2002; Freidenberg and Alcántara 2001), this subject has received little attentionin the Ecuadorian case. Few analyses have been concerned with institutional is-sues; most have been oriented more toward sociological or anthropological ex-planations of representation.

My analysis emphasizes the cleavages in national politics, focusing on the in-stitutional structure’s ability to reflect and process these cleavages. I argue thatthere is a divide between these two (national cleavages and institutional structure)that clearly causes problems in representation. On the other hand, the general-ization of certain political practices has created alternative forms of representa-tion and of satisfying the demands of particularistic actors (through clientelismand corporatism), which have allowed parties to survive as mechanisms ofrepresentation. By substituting the formally established channels and mecha-nisms of representation, these forms of particularism have eroded them; however,they have also, simultaneously, been able to respond—however partially—todemands and also to constitute an alternative arena for representation.

The problems arising from this situation are related to the temporal dimen-sion of this coexistence. The main question for political science and for partiesand politicians is: For how long and in what conditions can this balance betweeninstitutions and everyday practices be maintained if these practices erode the institutions? Particularistic practices such as clientelism and corporatism ensureimmediate results, but they corrode the institutions by draining them of con-tent. Therefore, a basic question in this chapter is the relationship between for-mal institutions and political practice. This is the primary focus of the secondpart of the chapter.

Finally, it is important to consider the conception of political representationunderlying the claim that there is a crisis of representation. In most Ecuadorianstudies on this subject, representation is seen as an expression of a binding man-date or at least as a direct channeling of interests. Most of these studies highlightthe limited ability of institutions to process conflict, clearly one of the basicfunctions of representative mechanisms. Other observations focus on the limitedcapacity of the political system to adequately represent diverse social interests.Apparently, each social sector is expected to get a quota in representative bod-ies in order to ensure not only the processing of their respective demands butalso participation in decision making. This contradicts three basic principles ofa representative regime: majority rule, autonomy of the representatives, and, de-rived from this last one, the non-binding mandate. To a great extent in Ecuador,the argument that there is a crisis of representation is based on this erroneousperception of representation, leading to a demand for results that cannot be produced. I discuss this perception in an attempt to demonstrate that in orderto tackle the problems of representativeness, we need an adequate concept ofrepresentation.

104 Pachano

Endless Reform and a Contradictory Institutional Framework

The Ecuadorian normative and institutional framework has been continually al-tered by both Congress and the executive and by a Constitutional Assembly thatissued a new constitution in 1998. In 1983, before the end of the first post-transition presidential and congressional terms, the first constitutional reformswere introduced. This was the beginning of a litany of institutional reforms thatapparently will continue to be an integral part of Ecuadorian political practice.Political reform—legal, constitutional, procedural—has been used as a tool forsolving political conflict. Even small problems, those constituting the custom-ary practices of political actors that must be processed politically, lead to ques-tions concerning the institutional and normative framework, leaving this frame-work constantly up for grabs. As a result, the possibilities of consolidating validreference points for the actors involved are very limited.

Basic aspects of the electoral system have been constantly altered: the repre-sentational formula, the electoral calendar, district size, and the way in whichvotes are cast (see Table 4.1). This has been one of the greatest obstacles to theinstitutionalization of the party system.4 Constant change has made it impossiblefor two consecutive elections to take place under the same set of rules, and nei-ther the voters nor the political parties have enjoyed certainty concerning therules of the game.

These problems are due not only to the frequency and number of reforms,but also to the contradictions from one reform to the next. Competing partic-ularistic interests and pressure from diverse social groups that function withshort-term logic has produced a complex institutional system rife with contra-dictions (Conaghan 1995; Mejía 2002). Many of the components of the elec-toral system contradict one another. For example, some aspects of the systemwere intended to strengthen parties. In contrast, the personal vote system in-troduced in 1998 worked against parties. Even if the 1978 Constitution and theparty and electoral laws of 1979 had some birth defects, they have only becomeworse over time, mainly due to successive changes brought about by particular-istic interests and the need to respond to specific situations.

Although the three main objectives in the return to democracy were tostrengthen parties, attenuate the personalistic character of Ecuadorian politics,and prevent party system fragmentation, the new institutional rules had the op-posite effect. Parties have had serious problems with consolidation, and in the2002 elections the solid electoral performance displayed by the four largest par-ties was reversed.5 The parties without exception have been electorally success-ful in limited geographic spaces. In national politics, personalism continues to bea salient characteristic. Finally, the dispersion of representation has increased no-ticeably in Congress; more parties win seats with a small number of votes.


Table 4.1Ecuador: Main Political-Electoral Reforms, 1983–2003

Year Main reforms

1983 Reduction of the presidential and legislative terms (from 5 to 4 years)Introduction of intermediate election (every 2 years) for provincial deputiesGeneral election for deputies coincides with the first runoff presidential

election (instead of the second)Name of legislature is changed from Cámara Nacional de Representantes to

Congreso NacionalMechanism for budget approval is simplifiedExecutive is given special powers to propose laws in situations of economic

emergency

1985 Majority system replaces proportional representation systemElimination of the electoral threshold as a requisite for the permanent

registration of parties

1986 Return to proportional representation (with Hare formula)

1994 Immediate reelection is approved for all elected offices, with the exception ofthe president

Deputies are prohibited from managing or lobbying for budget appropriations

1996 The prohibition on alliances is eliminated1997 Introduction of the system of personalized voting with open lists

Seats are allocated according to individual votes by simple majority, regardlessof list total votes; proportional formula is eliminated

National deputies increase from 12 to 20

1998 National deputies are eliminatedNumber of provincial deputies is increased (with a base of two per province

instead of one as was formerly the case)Presidential and legislative elections are separated from local and provincial

elections (electoral calendar is diversified)Return to proportional representation (D’Hondt formula)Congress loses the ability to promote a vote of no-confidence against cabinet

ministersMechanisms for the appointment of legislative authorities introduced

(president and two vice-presidents are appointed according to size of partybenches but with a vote of the entire legislative body)

Change in executive-legislative relations (powers of Congress are restrictedin aspects such as budget approval, appointment of accountability authorities, among others)

New conditions for the runoff presidential election: absolute majority or 40 percent threshold plus 10 percentage points above next candidate

2000 Return to the allocation of seats by lists (D’Hondt formula), keeping the personalized voting system

2003 Elimination of D’Hondt formula ( January)Introduction of Imperiali formula (September)

106 Pachano

Table 4.2

Share of Congressional Vote Won by Four Major Parties, 1979–2002(% of valid votes, provincial deputies)

Parties 1979 1984 1988 1992 1996 1998 2002

PSC 6.4 11.5 12.4 23.2 27.9 20.3 26.4PRE a 5.1 16.3 16.0 21.3 17.5 11.9ID 14.8 20.0 22.6 9.5 7.1 11.9 11.9DP b 7.3 10.9 7.2 11.9 24.1 3.1Others 78.8 56.1 37.8 44.1 31.8 26.2 46.7

Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0

SOURCE: Supreme Electoral Court.a Formed in 1982.b Not officially recognized; participated under the auspices of the CFP.

This complex, contradictory, and constantly changing institutional design hasoperated in an environment that is socially, economically, and culturally hostileto the consolidation of parties and to representative institutions in general. Thepolitical practices and behaviors of the social and political actors have obstructedthe achievement of the objectives proposed at the beginning of the transition.Because of its importance, this subject has been frequently discussed from manyperspectives (Menéndez 1986; F. Bustamante 1997; Burbano 1998; De la Torre1996; CORDES, n.d.). Most authors have concentrated on practices and be-havior, without paying sufficient attention to the institutional aspects. Most an-alysts have posited a cause-effect relationship whereby institutions are deter-mined by the social structure and political culture. Such analyses express thesociological and cultural bias of Ecuadorian political studies.

Diffuse Multipartism: Rules and Their Implications

One of the outstanding characteristics of Ecuadorian democracy has been thedispersion and fragmentation of the party system. Since the return to a demo-cratic regime in 1979, at least ten parties have secured congressional representa-tion. All of them—even the largest ones that have maintained the most stableshare of votes—have experienced erratic electoral fortunes (see Table 4.2,which includes only the four largest parties). Many parties have failed to survivemore than two elections and have been replaced by new parties that are gener-ally as small as those that disappeared.6

Several components of the electoral system have fostered party proliferation:the use of the province as an electoral district, proportional representation, theprohibition of local or subnational parties, the “no immediate reelection” rule (ineffect from 1979 until 1994), and the implementation of two-round presidentialelections.


The Province as an Electoral District

The use of provinces, the country’s administrative-political divisions, as electoraldistricts causes five problems in representation. First, the effect of their size rangeis translated, at the electoral level, in the coexistence of districts of different mag-nitudes, with results characteristic of this situation (Taagepera and Shugart 1989;Snyder 2001). Parties can win seats with very few votes, especially if they areconcentrated in small provinces. This has been the strategy followed by partiesthat obtain a very limited percentage of the vote on a national level but that winseats in the National Congress with votes obtained in provinces with smallpopulations.7

The most common size for electoral districts is two seats, with seven prov-inces electing that number. In 2002, four provinces elected three deputies each, six provinces elected four, two provinces elected five, one province electedfourteen, and one district elected eighteen deputies (see Table 4.3). If small dis-tricts are defined as those that elect less than 4.0 percent of the members of Con-gress (the median is 3.5 percent), then half of the districts fit into this category.

Table 4.3

Share of National Electorate and Number of Deputies per Province, 2002

Share of national Number of Percent of totalProvince electorate deputies deputies

Galápagos 0.1 2 2.0Zamora 0.3 2 2.0Orellana 0.4 2 2.0Pastaza 0.4 2 2.0Morona 0.6 2 2.0Sucumbíos 0.6 2 2.0Napo 0.9 2 2.0Bolívar 1.5 3 3.0Carchi 1.5 3 3.0Cañar 1.6 3 3.0Esmeraldas 2.6 4 4.0Imbabura 3.0 3 3.0Cotopaxi 3.1 4 4.0Loja 3.3 4 4.0Chimborazo 3.8 4 4.0Tungurahua 4.5 4 4.0El Oro 4.5 4 4.0Azuay 4.8 5 5.0Los Ríos 5.0 5 5.0Manabí 10.1 8 8.0Pichincha 20.5 14 14.0Guayas 27.0 18 18.0

Total 100.0 100 100.0

SOURCE: Supreme Electoral Court.

108 Pachano

Parties with little national presence can concentrate their efforts in one or sev-eral of these provinces and win seats in Congress. If we add to district size theeffects of the use of proportional representation or the system of personalized,open-list voting (introduced in 1997), it is clear that the doors have been wideopen to the dispersion of voting and the fragmentation of the party system.

Secondly, as it stands, the system creates imbalances and distortions amongprovinces in terms of the relationship between representatives and represented.The proportion of votes needed to elect a deputy varies significantly from dis-trict to district; voters from different districts do not have the same weight. Aspointed out by Taagepera and Shugart (1989, 14) and Snyder (2001, 146ff.), thisviolates the “one person, one vote” rule since the weight of each individual voteis not the same in all districts.8 The representativeness of the deputies is affectedby malapportionment. Each deputy represents a very unequal proportion of thepopulation, and the deputies’ possibilities of establishing a relationship with vot-ers varies substantially, depending on district size. In the smaller districts, thepossibility of establishing direct, practically face-to-face relationships is greater,which may create a fertile ground for binding mandates (mandatos vinculantes),which in turn may form the basis for clientelistic and corporatist forms of representation.

The number of members each province has in Congress depends on one oftwo rules: a minimum of two seats per province, or one seat for every 300,000inhabitants. These rules create some malapportionment. The smallest provincesbenefit and the largest are adversely affected. The rule that a province gets oneseat for every 300,000 citizens clashes with the idea of not increasing the num-ber of deputies and restricting parliament to a reasonable size, and it meets theresistance of the small and mid-sized provinces that view any increase in the num-ber of seats for the large provinces as a threat to their interests. Malapportionmentalso has regional effects.9 As the country’s most populated region, with 50.5 per-cent of the national population, the Coastal region, comprised of only five prov-inces, elects only 39 percent of the members of Congress. At the other end of thespectrum, the Amazon and Galápagos regions—with a total of seven provincesthat benefit from the minimum of two seats per province, and with only 3 per-cent of the country’s population—elect 14 percent of the seats. Comprised often provinces, the Sierra region is the only one to achieve representation that isproportional between its population (46.6 percent) and its share of seats (47 per-cent) in the Congress. This has been one of the few subjects related to the elec-toral system that has been on the political agenda and debated in terms of itsrepercussions on the representativeness of the various provinces and regions.

Third, with the use of the province as an electoral district the myth of terri-torial representation is created, whereby the deputy is the representative of pro-vincial interests rather than of a political movement. This is an alien and evencontradictory concept given the unitary character of the Ecuadorian state, yet it


is widely generalized in national politics and shapes the behavior of Ecuador’spolitical parties. Parties have to favor representation of provincial interests, evenif it means sacrificing their own positions and a vision for the country as a whole.The vindication of regional, provincial, and local interests has become almost anobligation since it is one of the ways to win electoral support. This logic in turnhas fueled the configuration of captive voters and electoral bastions, as part ofthe logic of a narrower and more particularistic political arena.

Fourth, the use of the province as an electoral district has contributed to partyindiscipline (Mejía 2002). Most deputies who have abandoned their parties al-lude to the parties’ nonexistent or meager concern with their province of ori-gin, which constitutes a tacit vindication of a binding mandate. Most deputieswho switch parties have been rewarded by resources or payoffs for the province,either through negotiations with the government or by a relatively powerfulparty boss. These agreements between deputies and government—the presi-dential connection referred to by Amorin Neto and Santos (2001, 221)—areobvious from the time of the elections and not only in the deputies’ perfor-mance. The deputy thus fulfills his/her commitment to his/her province.

Fifth, spurred by their quest to obtain the greatest possible number of seats, theparties (especially the largest ones, with electoral bastions in the most populatedprovinces) must seek votes in the small provinces, which leads them to seek outlocal candidates who can win votes. Generally, to do this they must sacrifice theirown principles and become catch-all parties, adapting their discourse and pro-posals to particularistic local realities. Deputies who win election have enormousnegotiating power and enjoy considerable autonomy with respect to the parties.

Making provinces the electoral districts has fostered the provincialization ofpolitics. The electoral rules do not favor the national distribution of party vot-ing (or nationalization, as referred to by Mainwaring and Jones 2003), but insteadtend to force parties into subnational arenas. This also contributes to the over-burdening of the national level by channeling demands to the upper levels (gov-ernment and Congress), a trend that is also spurred by the country’s adminis-trative centralization, which leaves little space for decision making at the lowerlevels of municipalities and provincial councils. Lastly, the electoral system is anincentive for the corporatist and clientelistic practices that characterize Ecuador-ian politics. The elimination of the national deputies in 199810 heightened thenegative effects of the electoral system given that they were a push factor for theconfiguration of a national political arena.

Proportional Representation

The proportional representation (PR) electoral system fostered the fragmenta-tion of the party system by allowing minor parties to win seats in Congress. Theallocation of seats by means of a double quotient mechanism (using the Hare

110 Pachano

formula), useful for maintaining proportionality between votes and seats, becamean incentive for the proliferation of small parties that could gain representationwith few votes. This was particularly evident in provinces with the greatestnumber of voters (Guayas and Pichincha, both electoral bastions of the large par-ties) and in intermediate ones (Manabí, Los Ríos, Azuay, and El Oro), wherevotes are more dispersed. Parties gained representation with an insignificantnumber of votes as a result of PR with the Hare formula.

Small parties have used two strategies to gain representation in Congress: first,as seen in the preceding section, they can concentrate their efforts on provinceswith the fewest voters; or, they can compete in the large and intermediate prov-inces where the proportional formula favors them. Either way, parties can winseats with a minimal proportion of the national vote.

This system results in the consistent presence of legislative parties with onlya few seats. The Ecuadorian Congress has consistently had a significant numberof small parties, operationalized here as those with less than 5 percent of themembers of the national assembly. (This 5 percent maximum was equal to threedeputies in the legislature from 1979 to 1984, four from 1984 to 1996, and sixfrom 1998 to 2000.)11 The dispersion in the National Congress makes it difficultto assemble majorities in support of or in opposition to the government.

These small parties are important because no party has ever obtained themajority of deputies in the National Congress. Small parties have consistentlybeen necessary to pass laws and form opposition blocks. The small provinces (es-pecially the Amazonia provinces) have tended to bring together parliamentarycoalitions outside party lines, especially in situations where their votes can be ne-gotiated (Rowland 1998; Mejía 1998). They have acquired an importance dis-proportionate to the number of their legislators, giving them considerable nego-tiating power in important congressional votes and in electing congressionalleaders. Also contributing to the power of the small provinces is the relativelysmall size of the Ecuadorian Congress; a few votes can make the difference incrucial decisions.12

The effects of the proportional system are heightened by the lack of an elec-toral threshold that prevents parties from obtaining seats in Congress with lessthan a certain percentage of the vote. The threshold established by law (whichhas fluctuated between 4 percent and 5 percent for elections of deputies, provin-cial councilors, and municipal councilors and has not been in effect during theentire period under discussion) is for registration purposes only. Parties that failto meet the minimum share of votes in two consecutive elections lose the Su-preme Electoral Court’s official recognition. However, parties that fail to meetthe 4 –5 percent threshold may still win seats and function as parties during theirterm in office. Furthermore, because registration is forfeited only after the givenpercentage has not been achieved in two consecutive elections—they cannot runin the third election—those who win office with below-threshold percentages


may still keep their seats for up to two consecutive terms (which might mean asmany as eight years).

In addition, proportional representation has been an incentive for personalism.Many analysts have argued that PR with closed and blocked lists shouldstrengthen parties (Nohlen 1993). However, in conjunction with the use of prov-inces as electoral districts, the parties’ obligation to participate nationally, and theprohibition of alliances in the proportional elections, as well as the establishmentof PR within the context of reduced institutionalization and predominatingcaudillismo, it has produced the opposite effect. The parties have had to incorpo-rate candidates who can bring in votes.

In 1997, in response to a referendum, the Ecuadorian electoral system, in-cluding PR, underwent the greatest reform in its history.13 A majoritarian sys-tem based on personalized voting with open lists was introduced. However, theelectoral system introduced in 1997 was soon replaced, and for all practical pur-poses has reverted to the proportional system.

National and Subnational Parties

One of the main objectives of the 1979 Constitution was to strengthen politi-cal parties. The history of instability during the preceding half century was as-sociated with the absence of parties of national scope capable of aggregating in-terests and forming governments founded on popular legitimacy. For the firsttime in the country’s history, and together with the new constitution, electoraland party laws were approved, both with considerable regulatory content. Thenew provisions were intended to promote the formation of strong parties,whose stability would be assured in time by ample organizational support andtheir presence throughout the entire nation. The goal was the elimination or atleast reduction of the formation of electoral machines that might be capable ofwinning votes but that would have no real long-term life of their own, no rootsin society, and would be limited to certain regions or provinces.

The electoral and party laws had meticulous provisions that forced parties tocarry out a series of activities in order to obtain and maintain their registrationwith the Supreme Electoral Court. Parties were required to maintain organiza-tional structures in at least ten of the twenty provinces that existed at the time.Once they obtained legal recognition, they were required to present candidatesin at least ten provinces. Failure to comply with these two provisions resultedfirst in the cancellation of registration, and after a second election, in the loss oflegal recognition. These provisions have acted as more effective barriers to thefragmentation of the party system than the electoral threshold.14

These regulations did not achieve their main objective of promoting the for-mation of national political parties. Parties have concentrated their votes in cer-tain provinces or at most in one region. Except for a brief period in which ID

112 Pachano

and PSC maintained a national presence—in terms of their votes—the pre-dominance of provincial or regional parties has been the main characteristic ofthe Ecuadorian system. Electoral bastions, in which each party concentrates itsefforts and to which other parties find it difficult to gain access, have grownsteadily in strength. Even the dominant party finds it difficult to move beyondthese boundaries and compete in other provinces.

In addition to structural determinations—Ecuador is characterized by verydistinctive regional societies—some legal regulations, including the very onesdesigned to promote the formation of parties of national scope, have fosteredthe provincial focus of parties. The legal regulations force the parties to act ona national level and compete for seats in the National Congress. Otherwise, theaforementioned provisions would apply and parties would lose their registrationand be unable to run candidates. In this way, minor local or provincial politi-cians and parties have been shifted to the national level, and the particularisticconcerns of these politicians have found their way into the National Congress.

This has a double effect. On the one hand, it fills the national scene with smallparties, generally with local orientations that represent the interests of very nar-row sectors of society. Consequently, the overload of subnational concerns anddemands that might under different circumstances be resolved at the local levelbecomes more pronounced on a national level. On the other hand, the larger par-ties—which in Ecuador tend to be more ideological and to have a more nationalorientation—are forced to compete in elections with locally or provinciallyrooted parties. They sacrifice principles in order to win votes in these localities.In this manner, they contribute to the overload of subnational topics in nationalpolitics, thus reinforcing the regional cleavages that characterize Ecuadorian pol-itics. Both large and small parties, whether rooted in a certain sector or a certainideology, must adapt to the provincial or local orientation of politics.

The inflexibility of the provisions aimed at helping parties achieve a more na-tional scope has had a harmful effect on this same objective. Some degree offlexibility—allowing, for example, local or provincial parties to compete in mu-nicipal and provincial council elections—would have brought about positive re-sults and helped to strengthen national parties. Better results would have beenachieved if effective barriers to participation on a national level had accompa-nied this flexibility at the municipal and provincial level.

These problems have worsened since the 1998 Constitution eliminated the na-tional deputies. Until 1998, a minority of deputies was elected in a single nation-wide district that attenuated the provincialism of political life, whereas the ma-jority was elected using provinces as the electoral district. The national deputieswere seen—both by the voters and by themselves—as guardians of a nationalmandate that the provincial deputies lacked.15 Their elimination strengthened theperception of Congress as a forum of territorial representation that focuses on lo-cal problems.


The provincialization of the parties is also fueled by the parties’ selection ofcandidates. In the process of candidate selection, parties constantly negotiatewith local leaders, who can usually impose their own conditions because theyhave captive voters. The local leaders are usually linked to local interest groups,so parties are forced to be responsive to those interests. For this reason, deputiestend to feel a greater connection and commitment to the local interests than tothe parties, which further promotes the idea that the deputy is a territorial rep-resentative with a binding mandate. Despite the problems derived from these ne-gotiations over candidate selection, the debate over the degree of democratiza-tion in selecting candidates is important. A more open candidate selection couldopen up the space for the participation of sectors that might not otherwise par-ticipate in the process;, however, it could also be deemed as a way of includinglocal oligarchic groups that in turn fuel the corporatist tendencies of Ecuadorianpolitics. In any case, candidates selected in this manner are the least likely to be-come disciplined party members on their legislative benches (Mejía 1998).

Alliances and Their Limitations

The prohibition of interparty electoral alliances that existed until the reforms of1996 created an obstacle to building coalitions in Congress (see Table 4.1). Theelectoral law established that for municipal council members, provincial coun-cilors, and both national and provincial deputies, each party needed to presentits own list. This provision sought to strengthen parties, assuming that partici-pation at all these levels of elections would require stable organizations and solidstructures. However, the regulations brought about unintended consequences.

The prohibition on electoral coalitions fostered party system fragmentationsince each party had to compete on its own. Pressured by the obligation to se-cure a minimum number of votes and present candidates in at least a minimumnumber of provinces, parties were forced to participate at each and every level.The inevitable result was the fragmentation of the system due to the enormousnumber of parties, most of them small, which under different circumstancesmight have formed alliances and thereby contributed to the formation of largeideological and electoral trends. Instead, parties competed with one another ina battle for access to government resources, and there were more incentives forinterparty confrontation than for reaching agreements. The confrontational ten-dencies of Ecuadorian politics are largely due to these provisions rather than tothe political culture.

Many local or provincial parties found that the regulations supported theirstrategy wherein seats are obtained via the proportional formula and the award-ing of seats by remainder. Forced to participate on their own on all electoral lev-els and in the greatest possible number of provinces, parties used this opportu-nity to their own advantage. Many local caudillos employed this provision tonegotiate successfully with small parties that, forced to compete on their own,

114 Pachano

needed a certain number of votes to guarantee their presence in Congress or, atthe very least, to comply with the required minimum number of votes.

Although the ban on coalitions was revoked in 1996, some barriers to coali-tions remain. Electoral alliances are now allowed, but the label of only one ofthe coalition partners is used to identify the coalition. The other party or par-ties are forced to give up their identity. Because of this, parties have incentivesto form alliances only when their chances of obtaining seats on their own arelimited or nonexistent. The formation of coalitions depends mainly on whethera party believes it would fare better by running on its own or as part of a coali-tion, and not on the political and programmatic orientation of the alliance orthe ideological principles guiding it. Coalitions are created for instrumentalelectoral purposes and not for the formation of large fronts identified by theirprinciples, objectives, or platforms.

Since 1997, national coalitions have been uncommon. Generally, coalitionshave been formed in provinces and municipalities, for elections held for provin-cial and municipal councils and for provincial deputies.16 The elimination ofnational deputies and the flexibility afforded by establishing coalitions in specificprovinces without compromising the respective parties in the rest of the countryhave motivated this pragmatic behavior. They have also heightened dispersion,since a cost-benefit calculation by a party can lead to infinite combinations, mostof them inexplicable in terms of the coalition partners’ programmatic positions.

The prohibition of alliances from 1979 to 1996 and later their liberalization andincreased flexibility have transformed parties into umbrellas that shelter a widerange of factions that enjoy great autonomy in selecting candidates. Althoughparties are formally national organizations, in electoral practice they operate likeprovincial organizations with relative autonomy in selecting candidates.17 A gameis set up, revolving around parties with more or less ability to represent localinterests, which is what really matters to the groups with which the parties haveto negotiate. An additional ingredient surfaces in the provincialization of the par-ties and their increased flexibility or loss of ideological-programmatic positions(in other words, in their transformation to catch-all parties).

Immediate Reelection

From 1979 until 1994, the immediate reelection of all authorities chosen in pop-ular elections, including deputies, was prohibited. Reelección cruzada, or crossoverreelection, was established, whereby a deputy could move from one type of postto another, either from national to provincial deputy or vice versa. However,since there were only twelve national deputies, the possibility of returning toCongress via this path was slim. At most, only twenty-four deputies (34.8 per-cent of the total members of Congress at that time) would be able to winreelection, and only if all the national deputies were reelected as provincial


deputies and, at the same time, their seats were taken by provincial deputies whowere elected as national deputies. This outcome was practically impossible, andit never occurred.

In 1983, when the Constitution underwent initial reforms, terms for all pro-vincial deputies were set at two years, while the term of a national deputy re-mained four years.18 Consequently, more than four-fifths of Congress had to bereplaced every two years, with no possibility of immediate reelection and min-imal hope of crossover reelection. The ban on immediate reelection brought in-stability to parliamentary activity. This instability in Congress was reinforced bythe annual election of congressional leaders and annual renewal of legislativecommittees. It became a substantial burden for parties to find candidates for allof these positions, given the ban on immediate reelection.

The negative effects of constant congressional turnover became apparent notonly in the instability in Congress—which assumed a short-term logic that af-fected legislative outcomes as well as its relationship with the executive branch—but also because political parties were forced to improvise to keep up with thesituation. None of the parties, not even those with the most solid structures,could respond to this challenge. Their reserves of leaders and militants capableof carrying out legislative functions were exhausted. Parties had to call on indi-viduals outside the party, generally local caudillos with popular electoral appealbut with no guarantee of loyalty or discipline to the party. This is one of theexplanations for the emergence of “floating politicians” (Conaghan 1994) withlimited loyalty to their parties.

Once again, the legal provisions resulted in outcomes radically contrary tothose desired. Instead of supporting the renewal of political leaders, encourag-ing greater participation in popular elections, and helping to reduce personal-ism, the prohibition of immediate reelection fostered improvisation, bred insta-bility, and accelerated the deterioration of the parties. It was an additionalincentive for the presence of local caudillos in national politics and for the grow-ing tendency toward the representation of local and corporatist interests.

Runoff Elections

In an attempt to strengthen the presidential mandate, the Constitution of 1979established runoff elections if no candidate wins more than 50 percent of the validvotes in the first round. The runoff system was intended to guarantee that a pres-ident’s legitimacy would be greater than that of presidents elected in the 1950sand 1960s, who were elected with a low percentage of the vote and only a smallmargin over their competitors. Allegedly, this lack of a clear popular mandate wasone of the reasons for governmental instability.19

The runoff system requires a number of conditions not present in the coun-try at the time it was established. As well, other components of the institutional

116 Pachano

arrangement stood in the way of achieving the necessary conditions. A basic re-quirement for the runoff system to operate adequately is the existence of strongparties, with stable electoral support and, above all, the ability to influence theway their constituencies vote so that the second round reflects organic politicaldecisions and not just the isolated electoral inclinations of each voter. In the ab-sence of parties that fulfill this requirement, the second round of presidential vot-ing represents the joint aggregation of separate wills, which does not generatestable and organic support for the government. These disparate wills have been,for the most part, channeled into negative votes against one of the final candi-dates rather than into votes in favor of another (Seligson and Córdova 2002).

For several reasons, including their inability to influence the way their follow-ers vote, Ecuadorian parties have consistently avoided publicly supporting pres-idential candidates (except of course their own) in the second round (Conaghan1995). As a result of the failure to forge electoral coalitions for the presidency,governing coalitions have not formed and sustained collaboration between theexecutive and the legislature. The entire post-1979 period has been character-ized by confrontation between these two branches of power. This so-called pugnade poderes—legislative/executive conflict—has on occasion placed regime stabil-ity at risk and has generally hampered governments (Sánchez-Parga 1998).

This destructive behavior by parties is due to several factors, among them for-mal institutional design, and in particular the lack of incentives for parties to de-velop collaborative practices. The cost of participating in governing coalitions,especially when parties hope to see governments rapidly erode, is much higherthan the cost of avoiding any electoral commitment in the second round.

The use of the two-round voting format in a system characterized by highfragmentation and volatility serves as an incentive for many parties to participatein presidential elections.20 Because of the dispersion of votes, small parties havean opportunity—unavailable under other circumstances—to reach the runoffround and even win presidential elections. Parties can go on to the second roundwith relatively few votes, as has occurred on several occasions (see Table 4.4).

Since 1984, congressional elections have taken place at the same time as thefirst round of the presidential election, creating an additional incentive for par-ties to present presidential candidates. With a presidential candidate, a party’sdeputies enjoy better prospects of getting elected. Without a presidential candi-date, parties have no way to offer future governmental benefits, so they are de-prived of one of the main attractions of congressional elections in an environ-ment where clientelism dominates. Therefore, parties generally present theirown presidential candidate even when their chances of winning might be greateras part of an interparty coalition.

The benefits obtained by parties in legislative elections come at the expenseof presidential elections. Party strategy is shaped by this context of great frag-mentation. Parties know they can obtain influence disproportionate to their


size, and the two-round system for electing the president creates this possibility.This strategy consists not only of gaining seats in Congress, but also of laying thefoundation for future relationships with the executive, regardless of who wins.As pointed out in the case of Brazil—quite similar to Ecuador in some ways—this strategy is generally linked to the pursuit and procurement of participationin the distribution of the national budget (patronage) (Amorin Neto and Santos 2001).

This subject cannot, therefore, be considered merely a question of the elec-toral timetable, or in other words, the election of deputies during the first roundof the presidential election. The main problem lies in the incentives generatedby the runoff system. This system creates an incentive for most parties to par-ticipate in presidential elections and lays the foundation for clientelistic rela-tionships between the president and the members of Congress. Congressionalelections have been held concurrently with the second round of presidentialvoting only once, in 1979, and afforded insufficient experience with which tojudge whether this might reduce the dispersion of presidential votes and thenumber of parties represented in Congress (see Table 4.5).21

Personalized Voting with Open Lists

In 1997, based on the results of a popular referendum, the Ecuadorian electoralsystem underwent a major reform that eliminated the system of proportionalrepresentation and replaced it with personalized voting with open lists. Underthis system, the parties’ lists of candidates become nothing more than a meansof presentation since voters cast their ballots for as many individual candidates asthere were seats to be filled, regardless of their party affiliation. The voter could

Table 4.4

Share of Presidential Vote in First Round, 1978–2002 (% of valid votes)

Candidates 1979 1984 1988 1992 1996 1998 2002

First place 27.7 28.7 24.7 31.9 27.2 34.9 20.6Second place 23.9 27.2 17.7 25.0 26.3 26.6 17.4Third place 22.7 13.5 14.7 21.9 20.6 16.1 15.4Fourth place 12.0 7.3 12.5 8.5 13.5 14.7 13.9Fifth place 8.0 6.8 11.5 3.2 4.9 5.1 12.1Sixth place 4.7 6.6 7.8 2.6 3.0 2.6 11.9Seventh place 4.7 5.0 1.9 2.4 3.7Eighth place 4.3 3.3 1.9 1.2 1.7Ninth place 0.8 1.6 1.4 0.9 1.2Tenth place 1.2 0.9 1.1Eleventh place 0.5 0.9Twelfth place 0.3

SOURCE: Supreme Electoral Court.

Ta

ble

4.5

Size

of L

egisl

ativ

e D

eleg

atio

ns,

1979

–200

2 (p

erce

ntag

es o

f par

ties

with

diff

eren

t-siz

ed d

eleg

atio

ns)

No.

of d

eput

ies

elec

ted

1979

1984

1986

1988

1990

1992

1994

1996

1998

2002

130

.0%

26.7

%28

.6%

25.0

%9.

1%30

.8%

28.6

%9.

1%0.

0%0.

0%2

20.0

13.3

0.0

25.0

18.2

15.4

28.6

36.4

22.2

16.6

310

.020

.021

.40.

027

.37.

70.

09.

111

.125

.04

10.0

6.7

14.3

8.3

0.0

0.0

0.0

9.1

11.1

8.3

50.

06.

77.

10.

00.

07.

70.

00.

00.

08.

36

0.0

6.7

7.1

8.3

0.0

7.7

14.3

0.0

0.0

0.0

7 to

10

10.0

13.3

7.1

25.0

18.2

7.7

14.3

9.1

11.1

0.0

11 t

o 20

10.0

0.0

14.3

0.0

27.3

15.4

7.1

18.2

11.1

25.0

21 a

nd o

ver

10.0

6.7

0.0

8.3

0.0

7.7

7.1

9.1

33.3

16.8

SO

UR

CE:

Supr

eme

Ele

ctor

al C

ourt

; And

rés

Mej

ía’s

data

base

.


vote a straight party line, but this option did not eliminate the personalized char-acter of voting because it was but one of the multiple ways of accruing votes.

Contrary to a proportional representation system with a personal vote (alsocalled a preference vote) in which the voter chooses one candidate from a givenparty or coalition, in Ecuador’s 1997 system, each voter could chose as manycandidates as there were seats in each province, from various lists. Dispersioncould occur in the very act of voting, since the voter had several votes or frac-tions of votes, something that does not happen under most systems. The vote it-self carried the seeds of dispersion. Therefore, a single person’s vote could pro-duce the same effect that would take the votes of several persons to accomplishin other electoral systems. This system provided maximum flexibility in choos-ing parties or, if one prefers, ideologies. In a large district, voters could cast votesfor candidates from all over the political spectrum, causing the spatial model forvoting (Downs 1957) to lose its power and the relationship between voters andparties to be almost completely annihilated.

The system’s most notorious effects were seen in the large districts where thepossibilities of selecting from different parties were greatest. The largest par-ties—those that underwent a process of consolidation throughout the post-1978 period and that helped support the stability of the party system—were themost affected, mostly because their electoral strongholds are in the largest dis-tricts. Because it adversely affected the main parties, this electoral system dealt ablow to the institutionalization of the party system. The open-list system alsoproduced a dispersion of votes in the smaller districts. In most small districts,22

the majority of voters distributed their votes widely.The open-list system weakened parties and furthered the extreme personal-

ism of politics (Pachano 1998). It is difficult to find a system that does a betterjob of fostering personalism and fragmentation. This electoral system fosteredthe floating character of both voters and politicians (Conaghan 1994, 1995). Thedisplacement of votes from one party to another—the very foundation of dis-persion and fragmentation—need not be put off until a later election since itcould be accomplished in a single act of voting. And with it also came reducedpossibilities of interpreting electoral results as a sanction or reward for differentparties since no unified party preference was expressed when a voter chose can-didates from several parties. In this way, the role of elections as a mechanism forassessing party performance (accountability) was significantly reduced. Althoughat the national level general tendencies could be discerned, they did not neces-sarily reflect the voters’ positions since multiple positions were expressed in asingle act of voting.

In conjunction with the ample opening awarded independents—establishedin 1994 as a result of another referendum—this electoral system left the partysystem vulnerable to deepening problems. It contributed to personalism, alreadya clear tendency in previous elections and one of the main factors contributing

120 Pachano

to the weakening of parties. It was introduced in a context of animosity towardparties, arising fundamentally from the poor performance of governments—invariably identified with certain parties—since the mid-1990s.

Diffuse Multipartism: Interests and Practices

The institutional framework described in the previous pages unfolds within a so-cial context characterized by diversity. Ecuadorian society is plural in social, eth-nic, linguistic, cultural, and regional terms (Almeida 1999; T. Bustamante 1992;Ibarra 1992; Deler 1987; Handelsman, n.d.; Rivera 1998; Pachano 1985). Socialscientists have identified the ethnic aspects of this diversity and regional differ-ences as important political factors, cleavages that define behavior and identities.Considerable scholarly attention has centered on ethnicity, understandably, giventhe impact of the indigenous mobilizations beginning in 1990, as well as on theformation of Pachakutik, the first party of ethnic origin, and its participation innational politics. A constant and active presence has made the indigenous a vis-ible actor on the national scene, although Pachakutik is a small party confined toa few provinces.23 The presence of an ethnic party has generated widespread in-terest in ethnicity and politics in the social sciences (Van Cott 2003, 2004).

Ecuador’s regional differences have been studied at length (Quintero 1991;Pachano 1985; León 1994; Deler 1987). Diversity is expressed in the form of regional societies differentiated along economic, social, cultural, and politicallines. The sources of this differentiation are structural, by which I mean that itderives from those factors that constitute a society and therefore greatly impactits formation and behavior. Each of these regional societies takes the shape ofrelatively differentiated spaces in which specific social relationships are estab-lished and build their own power structures, giving rise to strong regional iden-tities as well as unique behavior and habits. Social and political actors play thenational political game more with their own regional demands than as actors ona national level. Political parties and national instances of representation are al-ways heavily loaded with subnational demands and interests. This problem is ag-gravated by long-standing administrative centralization.

The existence of regional societies means that politics takes place on two lev-els. First, a political game in the regions—or in subnational arenas—revolvesaround controlling provincial and cantonal institutions. In this arena, local issuesare salient, politicians proceed through an important stage in their careers, andcollective actors are formed and battle for representation of regional interests.These political actors must also operate as expressions of national forces or atleast establish a close relationship with them. The prohibition of the formationof subnational parties creates a mandatory relationship between local issues andthe parties that, because of constitutional and legal provisions, must be national.Although this relationship has grown more flexible with the introduction of


independent political candidates—since independents are not required to main-tain a national organization and can limit themselves to local levels—it is still aburdensome imposition for parties.

Subnational issues are very present in national politics. Regional demands andthe social groups that represent them have an enormous effect on national issues.The power of subnational identities and regional issues in Ecuadorian politics isclearly visible in the constant presence of these regional problems, needs, anddemands at the national level. National political actors are forced to take a standon subnational issues, thus completing a circle that inhibits the identification,processing, and solution of national problems.

This interaction between the national and the subnational is at the heart ofpolitical representation in Ecuador. What is represented, who represents, andhow they are represented are the fundamental questions. In this game, powerfulsubnational actors are forced to act as emissaries of a binding mandate issuedfrom their regions in order to ensure their own survival, while weak national ac-tors, attempting to ensure their own survival, are forced to embrace subnationaldemands. The subnational actors do not prioritize the interests of the countryas a whole, even though they act in national fora such as the National Congress.To abandon this provincially oriented behavior would be political suicide forpoliticians from the provinces since they would be giving up their reason for ex-isting, as well as for those coming out of the national arena since they would nolonger have access to the subnational levels. If a political movement emerges atthe subnational level, it must make the transition to the national level, not onlybecause of legal determinations but also because that is where decisions are madeand resources are distributed. Conversely, if a political movement emerges at thenational level, it must penetrate the subnational levels because that is where theinterests that motivate the voters lie.24 The decisive factors in this two-sidedgame lie for the most part in the institutional/legal framework, especially in el-ements of the electoral system outlined above.

That is the problem that confronts the political parties. Their dilemma lies inthe necessity of either consolidating into national parties capable of working forthe general interest and structuring government proposals, or remaining subna-tional parties with loyal constituencies but continually dependent upon sociallyand spatially limited interests. In the light of the last twenty years, the latter isclearly the stronger tendency. To ensure their permanency, parties have strength-ened their links to regional or provincial interests and secured positions in elec-toral strongholds, even at the risk of sacrificing proposals of national scope andgiving up the possibility of producing positive results during their terms in gov-ernment. A result of this dynamic has been the provincialization of parties andof politics in general. Provincialization can be understood in two ways. First, itrefers to the electoral reclusion of parties in the strictly defined arenas in whichthey obtain their votes. Second, it also refers to the predominance of subnational

122 Pachano

issues in national politics, which in turn has a negative effect on policies and gov-ernability. Provincialization is one of the main characteristics of the party system,and in Ecuador it contributes to others such as fragmentation and atomization.

To appreciate the magnitude of provincialization in Ecuadorian parties, con-sider the parties’ electoral behavior in terms of territorial origin and respectivenumber of votes. As Table 4.6 shows, the parties with the most seats in Congress(PSC, ID, PRE, and DP) have won a high percentage of their votes in only oneprovince, clearly out of proportion with that province’s importance within thenational electorate. Guayas, Pichincha, Manabí, and Azuay are the provinceswith the greatest population. But the figures that the parties win in their strong-holds greatly exceed the proportion of voters that these provinces representcountrywide. While during the post-1978 period Guayas has fluctuated between24.0 percent and 27.5 percent of the national electorate, and Pichincha between18.0 percent and 20.0 percent, the parties with electoral strongholds in theseprovinces exceed these figures by amounts that have grown in recent years.25

Some parties fare well in the Coastal provinces (Guayas, Manabí) and othersfare well in the Sierra provinces (Pichincha, Azuay). The pronounced regionalelectoral differences have been a constant in Ecuador’s political and electoral his-tory. Electoral strength in one region automatically equals weakness in another,which explains the formation of impenetrable electoral strongholds. The partiesare severely limited in achieving a proportional distribution of votes throughout

Table 4.6

Electoral Strongholds of the Main Political Parties, 1979–2002 (share of national party vote won in First and Second provinces, provincial deputy elections)

Party

PSC PREa ID DPa

Year Province % Province % Province % Province %

1979 Guayas 30.72 Pichincha 34.11Pichincha 30.05 Guayas 16.23

1984 Pichincha 30.55 Guayas 76.37 Pichincha 26.51 Manabí 16.39Guayas 27.60 Pichincha 10.92 Guayas 11.02 Pichincha 13.45

1988 Guayas 38.05 Guayas 65.82 Pichincha 27.03 Pichincha 33.22Pichincha 19.15 Pichincha 5.37 Guayas 14.59 Guayas 11.04

1992 Guayas 51.25 Guayas 38.97 Pichincha 26.06 Pichincha 30.37Pichincha 6.35 Manabí 9.24 Guayas 10.92 Azuay 12.77

1996 Guayas 44.78 Guayas 34.34 Pichincha 29.55 Pichincha 42.39Pichincha 13.53 Manabí 15.24 El Oro 15.74 Guayas 9.25

1998 Guayas 44.80 Guayas 40.03 Pichincha 47.38 Pichincha 27.86Pichincha 15.55 Manabí 14.47 Azuay 9.59 Guayas 26.34

2002 Guayas 78.51 Guayas 59.96 Pichincha 66.02 Pichincha 39.45Pichincha 6.98 Manabí 12.88 Guayas 14.14 Manabí 29.22

SOURCE: Supreme Electoral Court.a Did not compete in 1979.


the nation. Given the relatively balanced distribution of population between theCoast and the Sierra, and given the absence of a third region capable of offsettingthis balance (due to the small population of the Amazon and Galápagos prov-inces), no party is likely to win a majority at the national level, something thatin fact has not occurred during the entire post-1979 period. In this sense, theprovincialization of the parties is one of the main reasons for party weakness andthe fragmentation of the system as a whole.

Regional discord is one of the most visible characteristics of Ecuador’s politi-cal system. Region tends to overshadow other political cleavages, so that theEcuadorian political game is defined more by the conflict between territoriallyconstructed identities than by economic or ideological cleavages. Its influence isobvious in the actors’ behavior and in the content of the national political agenda,and it forces political parties to act accordingly. The parties must represent spa-tially defined interests. The possibility of obtaining an even distribution of thevote for the different parties across the whole of the national territory is mini-mal, as is the space in which to build a national agenda.

This regionalization is clearly seen in the work by Mainwaring and Jones(2003), who document that Ecuador had the least nationalized distribution ofthe vote among seventeen countries in the western hemisphere. The authors cre-ated an index of party system nationalization. Between 1979 and 1996, Ecuadorattained an average coefficient of 0.57 on a scale of 0 to 1.26 Only Brazil (0.58)approximated Ecuador’s low level of nationalization. Bolivia scored 0.77, Chileand Uruguay 0.87, Costa Rica 0.90, and Honduras 0.92.

Another indicator of nationalization, the territorial distribution index (TDI)measures the distance between the number of votes a party wins in each provinceand the proportion of the national electorate in that province. A party is nationalin character if its votes are distributed by province in approximately the same pro-portion as the province’s share of the national electorate. This indicator is con-structed by adding together the differences between the proportion representedby each province in the national electorate and the proportion of votes eachprovince has in the party’s total number of party votes. It compares the weight ofeach province in the nationwide electorate with that of the provincial votes in thetotal number of party votes. Each party is assigned a figure for each election(Table 4.7).27 A lower number indicates less distance from the nationwide distri-bution of the electorate and consequently a party’s greater national presence.

Regardless of the indicator used, the distribution of each party’s votes differsmarkedly across different provinces. Based on the TDI, national distribution hasdeteriorated throughout the post-1979 period. Smaller parties (below 10 per-cent of the vote) show the most uneven electoral performance across differentprovinces. This means there is a relationship between the fragmentation of theparty system and the uneven distribution of party votes across provinces. This isan expression of the relationship between small parties and local interests, and

124 Pachano

of the fact that their presence in the national arena is due to legal imperativesand that the national arena is the only real space in which important decisionmaking occurs. Remaining on the fringes of the national institutions, specificallyCongress, would cost the local parties dearly.

The provincialization of the parties is directly expressed in parliamentary rep-resentation. As seen in Table 4.8, the configuration of largest parties from theCoastal and Sierra provinces is clearer at this level.28

In sum, the regulations designed to foster the formation of parties with nationalscope have turned out to be useless. Probably, the explanation is that these mea-sures were not meant to regulate already existing behavior, but instead to gener-ate new behavior designed to consolidate a modern political system. Therefore,instead of being measures aimed at channeling the demands and the representa-tion of regional or local interests, they were a way of denying or hiding these in-terests. These regulations were intended to impose certain behaviors, and they ig-nored the concrete conditions of the provinces and of the regional arenas ingeneral. For this reason, from the outset there was a risk that actors would useother channels to articulate their provincial or local demands. And when theseother ways failed to materialize—which could have been prevented through aprocess of decentralization of and increased flexibility in party and electorallaws—regional and local demands quickly found their way into the mechanismsdesigned specifically to evade them.

Due to the legal impossibility of forming parties with strictly local or regionalscope, the national parties—rather, those forced to be national—had to take on

Table 4.7

Territorial Distribution Index (TDI) of Main Parties, 1979–2002

1979 1984 1988 1990 1992 1996 1998 2002 Average

PSC 22.1 18.1 14.0 29.1 26.5 20.4 21.1 52.4 25.45ID 22.8 20.1 18.5 19.2 25.6 33.9 23.9 48.1 26.50MPDa 18.0 24.6 30.9 26.7 31.5 25.4 24.1 34.2 26.93DP 24.1 26.1 36.9 32.2 28.6 21.2 50.0 31.30FRA 23.7 27.9 39.0 29.1 27.1 44.4 31.86UDP- 22.6 24.3 57.8 26.6 33.6 26.4 31.87

FADIa

PRE 51.4 39.0 27.4 22.4 23.1 27.1 37.3 32.53CFPa 17.9 32.3 22.9 36.6 35.8 37.8 46.3 35.9 33.20PCEa 45.3 41.6 23.5 24.1 48.6 44.9 33.4 37.32APREa 46.3 43.6 22.9 31.9 30.7 36.4 60.8 38.93PSEa 41.2 50.4 35.9 38.4 34.4 45.5 39.5 42.5 40.99PLREa 28.6 26.5 38.5 48.4 53.3 27.1 52.0 72.1 43.31MUPP- 54.2 87.7 80.8 74.22

NPa

Average 29.4 31.7 29.8 32.02 33.6 33.7 39.1 50.4

SOURCE: Supreme Electoral Court.NOTE: Empty cells indicate party did not compete that year.a Parties with an average number of votes below 10 percent in that period.


the demands and the representation that would have been the province of theformer. Conceivably, this might not have occurred within a flexible frameworkthat allowed regional or local parties to coexist with national parties, providedthat the functions and scope of action for each of these had previously beenclearly defined. However, by applying general laws to diverse situations and,above all, by giving these laws the power to transform practices and to generatebehavior that turned out to be artificial, the local parties were checked but theirfunctions were transferred to parties expected to be national in character. Thus,national parties were forced to adapt to this distortion or run the risk of isolat-ing themselves from voters and losing their ability to represent them. This forcedthem into a situation of dependency with regard to local or regional interests,and the effort to respond to local interests overshadowed ideological and pro-grammatic considerations. Thus was completed a circle comprised of (a) thepresence of regional identities; (b) inflexible laws that sought to deny or mini-mize them; (c) the absence of adequate mechanisms to express these local andprovincial interests; and (d) parties forced to meet the interests of regional elec-torates. The main consequence was the shifting of local and regional issues to thenational arena, especially Congress, where debate can no longer be separatedfrom territorial determinations and the game described earlier between the na-tional and the subnational must be played.

Forced to act as representatives of particularistic local interests, parties act asvoices for narrow social and economic groups. The corporatist nature of politicsin Ecuador can be explained to a great extent by this relationship between re-gionally defined interests and political representation since pressure groupsachieve a dominant presence in local arenas and dominate representation. Polit-ical operations become tremendously complex, especially with regard to the pur-suit of agreements and the fostering of national politics, which takes place in anarena where particularistic and directly represented interests battle one another.

The indigenous peoples’ presence in Ecuadorian politics is emblematic of thislink between localized interests and the provincialization of the parties. These

table 4.8

Regional Distribution of Origin of Deputies, by Party, 1979–2002

Regional origin

Costal Amazonia-Party (Coastal) Sierra Galápagos Total

PSC 66.7% 31.8% 1.6% 100%PRE 75.3% 23.4% 1.4% 100%ID 30.2% 65.5% 4.3% 100%DP 28.3% 66.4% 5.3% 100%

SOURCES: Supreme Electoral Court; Freidenberg 2000.

126 Pachano

indigenous parties have stronger regional roots than other parties because thegeographic location of the indigenous populations creates a regional bias. Theindigenous population is located almost exclusively in the Sierra and Amazoniaprovinces. Therefore Pachakutik, the principle partisan voice of these groups,wins votes almost exclusively in these regions. It is an important player in theSierra and Amazon, but faces enormous difficulties in winning votes in theCoastal provinces. Its electoral shortcomings in the Coastal provinces have pre-vented it from developing a broader base, not only in electoral terms, but alsowith regard to the possible structuring of proposals of national scope that go beyond the particularistic interests of the indigenous peoples. Pachakutik hasadopted the same logic as the system as a whole, forced to take refuge in localbastions in order to build up its electoral strength at the cost of not having apresence in other regions.

One can extend what has been said about Pachakutik to all Ecuadorian po-litical parties. Even the largest parties have adopted this strategy of representinggroup interests in order to win a large number of votes in some provinces. Thisis the dilemma facing the parties and giving rise to the problems of representa-tion that, paradoxically, are not the ones most analysts emphasize when they re-fer to the crisis of representation.

Crisis of Representation or Crisis of Regulation?

The problems facing the party system originated basically in the rules that reg-ulate them. Their inorganic character—the fact they do not all point in the samedirection—the contradictions between their separate components, and the re-forms constantly introduced in response to short-term interests prevent the sys-tem from attenuating the structural conditions surrounding it. These structuralconditions give rise to actors, orientations, and behaviors that are ill suitedto the construction and consolidation of a political forum of national scope orpolitics built around an arena wherein the general interest can take shape. Thesestructural conditions would have had a less negative impact if another institu-tional design—specifically, a different electoral system—were in place. Struc-tural heterogeneity is not necessarily an obstacle to the elaboration of nationalproposals and, consequently, to the consolidation of national parties. The expe-riences of countries as diverse or more diverse than Ecuador (Spain, UnitedStates, Germany, Switzerland) have proven the power of institutional design toforge national parties and interests.

When they speak of the Ecuadorian crisis of representation, most analysts al-lude to aspects other than institutional design and refer instead to social, eco-nomic, and cultural factors (Arias 1995; Burbano 1998; Dávalos 2001). Theseanalyses emphasize the results produced by the system, and they question the ca-pacity of parties and democratic institutions more generally to represent interests.


They usually claim that links between the represented and the representativesare weak. In turn, such weak linkages are considered a threat to the smooth op-eration of democratic institutions and even to the system’s stability. In this man-ner these analysts finally arrive at problems of governability, through a forcedidentification with the problems of representation or representativeness.

A connection does exist between problems of representation and governabil-ity, but not the kind of connection that has usually been suggested in Ecuador.The political system fails to yield satisfactory results not because of a rupture be-tween the represented and the representatives—such a rupture does not exist ordoes not exist as acutely as the analysts claim. Nor is the main problem a lim-ited ability to represent interests—this ability is actually excessive given the par-ticularistic nature of representation in Ecuador. Rather, the problems of gov-ernability that stem from the system of representation arise because of the gamethat emerges out of a defective institutional design. The impossibility of foster-ing policies with national scope, the short-term focus of political action, and thepredominance of local and group interests impose a logic that leads to the im-mobilization of governments and Congress. The ongoing game between pow-erful local actors and weak national actors, driven and fostered by the institu-tional design, goes a long way in explaining the political system’s low capacity.The provincialization of the parties, a result of the electoral system describedabove, largely explains problems that have not been treated frequently enoughby Ecuadorian social scientists and, on the contrary, have remained hidden be-hind ideas such as the crisis of representation.

Notes

1. Until the 2002 elections, the four biggest parties of the post-1979 period (PSC,PRE, ID, and DP) displayed a tendency toward an increased share of the vote notwith-standing cyclical oscillations. Ecuador is halfway between the collapse of the political par-ties of Peru and Venezuela and the stability of Colombia and Bolivia.

2. Since the transition to democracy, a decline is visible in indicators such as the grossdomestic product, poverty indexes, distribution of income, the proportion of the bud-get assigned to social expenditures, and the population’s buying power. From 1980 to2000, there was zero growth in the gross domestic product; per capita income fell by 0.3percent between 1981 and 1991 and by 0.6 percent between 1991 and 2001; poverty in-creased from 34 percent of the population in 1990 to 56 percent in 2002.

3. An analysis of the substantive outputs of the Ecuadorian political system is beyondthe scope of this chapter. I only make general references to this subject without ignoringits importance in any analysis of the problems of representation.

4. According to Mainwaring and Scully (1995, 4), one of the criteria for the institu-tionalization of party systems is the permanency of electoral rules, together with the

128 Pachano

solidity of the organizations, reduced electoral volatility, the existence of roots in the society, and operations dependent upon bureaucratic routines more than on personali-ties or charismatic leadership.

5. The Social Cristiano, Roldosista Ecuatoriano, Izquierda Democrática, and Democ-racia Popular parties have won as much as 80 percent of the valid vote. In the 2002 elec-tions this vote share dropped noticeably, although this is not an indication of a party col-lapse of the magnitude experienced in Peru and Venezuela.

6. Small parties have disappeared as a result of a legal provision that requires that theywin a minimum share of the vote in two consecutive elections in order to maintain le-gal recognition. This legal barrier has fluctuated between 4 percent and 5 percent, andhas not remained continuously in effect during this period. This is a barrier only to reg-istration and not to representation since parties that do not meet the minimum maintaintheir seats in parliament and in other elective offices even after the second consecutivefailure to meet the threshold.

7. The Partido Socialista Ecuatoriano, the Frente Radical Alfarista, the Partido Lib-eral Radical, and the Movimiento Pachakutik have repeatedly done just this. Supportedby the absence of any true barrier to representation, they have survived several elections.

8. Snyder (2001, 149) considers the problems of malapportionment between votersand seats to be one of the causes of unjust elections, on a par with the buying of votes,the altering of outcomes, and electoral fraud.

9. Although regions do not constitute an official administrative-political division andare not a part of the electoral design, in the country’s political and social life they carryconsiderable weight.

10. This was one of the reforms introduced by the National Constituent Assemblyduring the constitution-making episode of 1998.

11. The number of members of the National Congress has fluctuated constantly. Thenumber of deputies increased from 69 in 1979 to 123 in 2000, with 71 between 1984 and1988, 72 in 1990, 77 in 1992, 72 in 1994, 82 in 1996, 123 in 1998, and 100 since 2000.

12. The most notorious example of the influence of small parties was the Frente Rad-ical Alfarista (Radical Alfarista Front, or FRA). Although it never had more than threedeputies, it gained the presidency of the Congress on two occasions. When Congress un-seated President Abdalá Bucaram in 1997 and appointed an interim president—in clearviolation of the Constitution—it elected the supreme leader of the FRA, Fabián Alarcón.

13. The Ecuadorian political system has been constantly reformed since 1979. Thishas become a source of instability since the country lacks a stable normative frameworkto guide the behavior of political actors. A summary of the many reforms introducedsince 1979 can be found in Table 4.1.

14. Other provisions regulate various aspects of internal party life and express the ori-entation of the new regulations and the parties’ role. The obligation to participate in aminimum number of provinces refers to multimember elections: elections for municipalcouncils, provincial councils, and the National Congress.

15. There was always a differentiation between national and provincial deputies, withregard not only to their electoral districts but also to their functions. When in 1983 theprovincial deputy’s term was reduced to two years, the national deputy’s term remainedat four. The minimum age requirement for provincial deputies is 25, while it was 30 fornational deputies. And although not in the end adopted, a proposal was made that wouldrequire candidates for the presidency of the Congress to be limited to national deputies.

16. Only twice, in 1996 and 2002, have national alliances been formed for presiden-tial and legislative elections. But even so, in 2002, the parties that formed this nationalalliance entered into different coalitions in the provinces.


17. The newspaper El Comercio drew attention to the importance of the parties’ pro-vincial politics in a series of reports published between February and August 2003. Eachparty’s selection of candidates responds to the specificities of a certain province.

18. The change in the electoral calendar was more profound. The presidential andlegislative terms were cut from five to four years and the term of a provincial deputy totwo years. The goal of increased stability and continuity through longer terms was there-fore subordinated.

19. This perception was wrong. There is no correlation between presidents electedwith a low percentage of the votes and instability of their governments.

20. The Ecuadorian party system is one of the most fragmented and volatile in LatinAmerica (Conaghan 1995; Mainwaring and Scully 1995; Mejía 2002). There are fewstudies on this subject. For example, there has been little exploration of the relationshipbetween party system fragmentation or electoral volatility and particularistic practicessuch as clientelism and corporatism, or between the provincialization of politics and parties.

21. The scheduling of parliamentary elections to coincide with the second round ofthe presidential election may affect the percentage of votes won by the party of the win-ning candidate; in 1979 this candidate’s party achieved the highest percentage of votesfor Congress during the entire period. But this too can be questioned, since it appliedonly to the winner and not to the other second-round presidential candidate, whoseparty did not fare well in the congressional election.

22. In the 1997 election, in seven of the nine districts that elected two deputies, can-didates from two different parties won. In five of the seven districts that elected threedeputies, three different parties elected one candidate each (Pachano 1998).

23. Pachakutik has taken part in elections since 1996. Its share of the vote (based onthe average number of deputies and provincial and municipal councilors) peaked in 1998at less than 5 percent of valid votes. In 2002, although it backed the winning presidentialcandidate, Pachakutik barely surpassed that percentage. The party has achieved significantresults in local elections, especially mayoral elections in cantons with a large indigenouspopulation, but it has been unable to penetrate several provinces, especially the Coastalones. Certain actions, such as Pachakutik’s support of the January 2000 coup that oustedPresident Mahuad, have led to greater renown but have at the same time limited theparty’s electoral growth.

24. Political parties have pushed this tendency to the limit by granting privileges tothe municipalities and provincial councils, where they have strengthened themselves elec-torally and where at the same time they have been able to develop successful administra-tions. The cases of the Partido Social Cristiano (PSC) in the mayor’s office in Guayaquiland the Izquierda Democrática (ID) in Quito are examples.

25. The only exceptions—the Partido Social Cristiano and the Izquierda Democráticabetween 1979 and 1986—illustrate the provincialization of parties that had a nationalscope during the first elections in the post-1979 period.

26. The indicator uses the Gini coefficient to measure inequality of distribution, inthis case the votes obtained by each party in electoral districts or subnational units. In thisapplication it has been inverted (1/Gini): a higher score equals a more nationalized dis-tribution of votes (Mainwaring and Jones 2003, 142).

27. The indicator is the product of the sum total of absolute values taken from thedifference between the weight of the province in the census (padrón) and the party’s pro-vincial votes, multiplied by the weight the province carries. The following formula canbe used to express this: TDI � (�|Pn � VPn|P)/2, where Pn is the weight carried byeach province in the electoral census and VPn is the weight of provincial votes over the

130 Pachano

party’s national voting. This is similar to the procedure used by Taagepera and Shugart(1989, 104ff.) to measure deviation from proportionality. Thanks to Andrés Mejía forhelp in arriving at this indicator.

28. The Coastal provinces are underrepresented as a result of using provinces as elec-toral districts and because only parties with the greatest number of votes during the pe-riod are included. The small parties are local or provincial groups and including themwould mean working with a constant and not a variable.

References

Almeida, José. 1999. “Identidades en el Ecuador: Un balance antropológico.”Antropología (Universidad Católica, Quito), no. 4: 14 –32.

Amorin Neto, Octavio, and Fabiano Santos. 2001. “The Executive Connection.” PartyPolitics 7, no. 2: 213–34.

Arias, Natalia. 1995. “Partidos políticos: ¿Héroes o villanos?” Ecuador Debate 36: 49– 61.Barrera, Augusto. 2001. Acción colectiva y crisis política: El levantamiento indígena ecuatoriano

en la década de los noventa. Quito: Abya-Yala.Burbano, Felipe. 1998. Cultura política y democracia en Ecuador: Una aproximación a nuestros

vacíos. Quito: CORDES.Bustamante, Fernando. 1997. “La cultura política y ciudadana en el Ecuador.” In

Ecuador: un problema de gobernabilidad, ed. Esteban Vega, 103–57. Quito: CORDES.———. 2000. “Los partidos como orientaciones culturales.” Iconos 9: 88–97.Bustamante, Teodoro. 1992. “Identidad, democracia y ciudadanía.” In Identidades y

sociedad, by various authors, 43–76. Quito: CELA.Conaghan, Catherine. 1994. “Loose Parties, Floating Politicians, and Institutional Stress:

Presidentialism in Ecuador, 1979–1988.” In The Failure of Presidential Democracy, ed. Juan Linz and Arturo Valenzuela, 254 –85. Baltimore: Johns Hopkins UniversityPress.

———. 1995. “Politician against Parties: Discord and Disconnection in Ecuador’s PartySystem.” In Building Democratic Institutions, ed. Scott Mainwaring and Timothy Scully,434 –58. Stanford, CA: Stanford University Press.

CORDES. n.d. La ruta de la gobernabilidad. Quito: CORDES.Dávalos, Pablo. 2001. “Movimiento indígena ecuatoriano: La construcción de un actor

político.” Revista Ciencias Sociales 20: 111–38.De la Torre, Carlos. 1996. Un solo toque. Quito: CAAP.Deler, Jean-Paul. 1987. Ecuador: Del espacio al estado nacional. Quito: Banco Central del

Ecuador.Downs, Anthony. 1957. An Economic Theory of Democracy. New York: Harper Collins.Freidenberg, Flavia. 2000. “Las posiciones ideológicas programáticas en los partidos ecua-

torianos.” Paper presented at the seminar “Political Parties in Latin America,” Insti-tute of Latin American Studies (ILAS), London.

Freidenberg, Flavia, and Manuel Alcántara. 2001. Los dueños del poder: Partidos políticos enEcuador, 1978–2000. Quito: FLACSO.

Handelsman, Michael. n.d. “La globalización y la construcción de nuevas expresiones deidentidad: El caso plurinacional del Ecuador.” Unpublished paper.

Ibarra, Hernán. 1992. “El laberinto del mestizaje.” In Identidades y sociedad, by variousauthors, 95–123. Quito: CELA.

León, Jorge. 1994. “El fin de un ciclo político electoral: El regreso de las elites tradi-cionales, apatía y cambio.” Ecuador Debate 32: 76 –89.

Mainwaring, Scott, and Mark P. Jones. 2003. “The Nationalization of Parties and PartySystems: An Empirical Measure and Application to the Americas.” Party Politics 9, no. 2: 139– 66.

Mainwaring, Scott, and Timothy Scully, eds. 1995. Building Democratic Institutions: PartySystems in Latin America. Stanford, CA: Stanford University Press.

Mejía, Andrés. 1998. Partidos políticos: El eslabón perdido de la representación. Quito:CORDES.

———. 2002. Gobernabilidad democrática. Quito: Honrad Adenauer.Menéndez, Amparo. 1986. La conquista del voto: De Velasco a Roldós. Quito: CEN.Nohlen, Dieter. 1993. Sistemas electorales de América Latina. Lima: Fundación Friedrich

Ebert.Pachano, Simón. 1985. “Movimientos sociales regionales.” In Movimientos sociales en el

Ecuador, ed. Luis Verdesoto, 151–81. Quito: ILDIS.———. 1998. La representación caótica. Quito: FLACSO.Quintero, Rafael. 1991. La cuestión regional y el poder. Quito: CEN–Flacso–York

University.Rivera, Fredy. 1998. “Los indigenismos en Ecuador: De paternalismos y otras repre-

sentaciones.” América Latina Hoy, no. 19: 57– 63.Rowland, Michel. 1998. “La crisis de representatividad del régimen político ecuatoriano:

Una aproximación institucional.” In Representación política y democracia, ed. CarlotaJacksich, 9– 43. Buenos Aires: Konrad Adenauer.

Sánchez-Parga, José. 1998. La pugna de poderes: Análisis crítico del sistema político ecuatoriano.Quito: Abya-Yala.

Seligson, Mitchell, and Polibio Córdova. 2002. Auditoría de la democracia. Quito:CEDATOS.

Snyder, Richard. 2001. “Devaluing the Vote in Latin America.” Journal of Democracy 12,no. 1: 146 –59.

Taagepera, Rein, and Matthew Shugart. 1989. Seats and Votes: The Effects and Determinantsof Electoral Systems. New Haven, CT: Yale University Press.

Van Cott, Donna Lee. 2003. “Institutional Change and Ethnic Parties in South Amer-ica.” Latin American Politics and Society 45, no. 2: 1–39.

———. 2004. “Los movimientos indígenas y sus logros: La representación y el re-conocimiento jurídico en Los Andes.” America Latina Hoy, no. 36: 141–59.


Democracy in the Andean countries is in a dismal situation. A process ofdecline and even reversal has been under way for some time and reached a crit-ical stage by the end of the 1990s. Representative democracy has followed con-tradictory and regressive paths, leading in some countries to authoritarian re-gimes, delegative democracies, or semi-democracies. In the last decade, mostcountries in the region have undergone devastating political turmoil, with major implications for the region’s stability, political foundations, and futureprospects.

Throughout the Andes, the signs of strain are manifest. Between 1997 and2005, Ecuador underwent a chaotic period of instability, witnessing the electionand overthrow of five presidents. Since the coup d’état of paratrooper HugoChávez in February 1992, Venezuela has suffered a mounting crisis of the stateand the economy, compounded by a dramatic collapse of its party system. Therise to power of an autocratic outsider has pushed the country into a politicalstalemate and an even deeper crisis of ingovernability. Chávez’s government hastriggered the polarization of society, in which a widespread opposition haslaunched four general strikes but failed to topple Chávez in either the farcicalcoup d’état of April 2002 or the protracted general strike between December2002 and January 2003. A decade earlier, Peru became the first Latin Americancountry to see its party system collapse, giving rise to a decade-long dictatorshipthat dismantled state institutions and degenerated into a mafia-type regime.Colombia’s party system is undergoing a process of dangerous atomization, andthe multifaceted threat of guerrilla violence, drug trafficking, and terrorism gripsthe state. Bolivia now faces serious strains after more than a decade of institutionbuilding (1985–97) that was not backed by sufficient economic growth and pov-erty reduction. The whole region’s democratic system is at stake and faces atwofold crisis of political representation and governability.

5

Outsiders and Neopopulism: The Road to Plebiscitary Democracy

❖

René Antonio Mayorga

Outsiders and Neopopulism 133

One of the most troubling aspects of the crisis of democracy has been the emer-gence of outsiders—that is, neopopulist and anti-political actors—in almost thewhole region. The rise of outsiders is relevant to the subject of this book for twomain reasons. First, the rise has been the dramatic outcome of the crisis of demo-cratic representation, and particularly of the collapse of parties, as argued in Chap-ter 1. Second, it has had disruptive consequences for representative democracy.I argue that the crisis of party systems stemmed from the failure of predominantparties as governing parties. Ultimately, problems of governability were the un-derlying cause of the crisis of democratic representation, that is, of the increasinginability of parties to reflect and articulate electoral preferences that became ap-parent in the deep distrust of citizens and the sharp decline of electoral support forparties. In Peru and Venezuela, outsiders sprang onto the scene with overwhelm-ing success. Both Chávez and Alberto Fujimori in Peru seized power demo-cratically and established political regimes that do not fit the category of liberaldemocracies.1 In Bolivia, two outsiders, Carlos Palenque and Max Fernández,created neopopulist parties with relative success and took an ambiguous stance to-ward representative democracy. Yet they could not prevail completely, given theirintegration into the party system in which they played a significant role. Bolivia’sdemocratic regime has faced, however, a different threat, from indigenous move-ments that seek to destroy democratic institutions and replace them with utopian,ethnic-based, direct democracy and nationalist populism.

Highlighting common patterns and differences among outsiders’ politics inPeru, Venezuela, and Bolivia from a comparative perspective, this chapter willaddress three main issues. First, it examines the causes for the emergence of out-siders and their rise to power. Second, it looks at the sequences and patterns ofparty system collapse. Third, it analyzes the far-reaching destructive conse-quences of outsiders’ politics on liberal democracy and democratic institutions.The main purpose is both to find common ground explaining neopopulism andthe emergence of outsiders in the Andean countries and to dwell on politicallyrooted differences between them. Why did successful outsiders—successful inthe sense that they rose to power—emerge in Peru and Venezuela? Why wereoutsiders in Bolivia only partially successful? What accounts for the rapid rise ofradical indigenous political movements?

Theoretical Approach

The concept of neopopulism is useful for addressing the nature of the politicscarried out by outsiders. This contention first calls for a definition of the con-ceptual differences between “neopopulism” and the classical term “populism,”widely used in Latin American social sciences since the 1960s. Such diverse na-tionalist, anti-imperialist political movements as the PRI (Partido RevolucionarioInstitucional, or Institutional Revolutionary Party) in Mexico, Peronism in

134 Mayorga

Argentina, the APRA (Alianza Popular Revolucionaria Americana, or AmericanRevolutionary Popular Alliance) in Peru, and the MNR (Movimiento Nacional-ista Revolucionario, or Nationalist Revolutionary Movement) in Bolivia have allbeen labeled as populist. As a widespread and crucial historical phenomenon inLatin America, populism has spawned competing theoretical perspectives that re-veal strong disagreement about the meaning of populism as a concept (Weyland2001, 13).

The concept of populism has been anchored in four theoretical perspectives:(1) a historical-sociological perspective, which stresses social mobilizations andsociopolitical coalitions arising in the context of the crisis of oligarchic domi-nation, the early stages of industrialization, and the transition from a traditionalto a modern society; (2) an economic perspective, which draws attention to pop-ulism as a type of redistributive policy and state interventionism responding toeconomic elites’ weakness and inability to develop class hegemony; (3) an ideo-logical perspective, which identifies populism with a specific discourse articulat-ing the constitution of a “popular actor” and the contradiction between this actor and the dominant classes, and (4) a political perspective, which explainspopulism as a pattern of mobilization of subaltern and/or excluded masses bypersonalistic leaders that is not based on institutional structures of political me-diation (Pécaut 1987, 245–54; Roberts 1995, 84 –85).

Mostly embedded in modernization and dependency theories, theoreticalwork on populism since the 1960s led to diverse meanings and a wide dispersioninto social, economic, ideological, and political domains. As Weyland (2001) as-serts, this theoretical work produced divergent cumulative and radial concepts.The result was broad dissemination and fuzzy meanings of populism as a con-cept. In contrast, theoretical efforts to understand and explain the paradoxicalresilience of populism in the last decade—including the cases I am dealing within this chapter—aim to reshape the concept as a theoretical tool by delimitingit in terms of a key domain and a predominant meaning, thus making it usefulfor empirical research. The recent attempts to build a theoretical approach basedon a political-institutional perspective provide a synthetic construction of theconcept of populism, integrating phenomena associated with “classical” pop-ulism that diverge from contemporary, neopopulist features (Weyland 2001;Mayorga 1995; Novaro 1996; de la Torre 2000; Martucelli and Svampa 1992;Pécaut 1987).

Given the flexibility and diversity of contemporary populism and the emer-gence of populist leaders in contexts that are a far cry from past nationalist and sta-tist populism, a mainstream in the current theoretical work restricts the concept’skey domain to the political realm and defines it as a predominantly political phe-nomenon. From this perspective, present forms of populism are no longer linkedto specific economic policies or social constituencies. Against the backdrop of“paradigmatic” cases such as Peru and Venezuela, I argue that contemporary


populism should be conceptualized fundamentally as a pattern of personalistic andanti-institutionalist politics, rooted mainly in the appeal to and/or mobilizationof marginalized masses. In this specific pattern of politics, the charismatic leaderexploits an ideological discourse of defending the poor and excluded, throughwhich he garners electoral support and democratically legitimizes the quest forand exercise of power.

In this regard, unlike historical populism, neopopulism is involved in the dem-ocratic game. It accepts the rules of political competition, but at the same timeresorts to the higher quality and legitimacy of the leader, who presents himselfas redeemer and embodiment of the people and the nation. As an ideology, neo-populism is therefore a pattern of ideological legitimation that is not at odds withrepresentative democracy. In fact, it takes advantage of the resources and incen-tives that representative democracy and its electoral mechanisms provide. Yet theCaesaristic conception of politics, the leader’s central role, and the lack of an in-stitutionalized party and support inevitably combine to undermine democraticinstitutions and to concentrate state power in the hands of the leader once hecomes to power. As anti-institutionalist practice, neopopulism therefore is a per-vasive form of anti-politics, that is, of politics carried out against parties, demo-cratic institutions, andestablishedpolitical andeconomic elites (Schedler 1994, 4).

Neopopulist discourse is basically anti-political insofar as it questions the es-tablished political parties as corrupt institutions and blames the political class andthe economic elites for the problems facing the country. Thus, the outsiders’ dis-course assumes fundamentally not only a radical rejection of the existing partysystems as such, but also the idea that parties are useless and pernicious organi-zations. This discourse is equivocal insofar as it is mostly an appeal to excludedpeople and simultaneously a commitment to neoliberal economic policies (Ma-yorga 1995). Neopopulist discourse has fostered an extreme neoliberal economicmodel of structural adjustment characterized by deregulation of markets, priva-tization of state enterprises, foreign-trade liberalization, and the absence ofsocial policies (González de Olarte 1998; Roberts 1995, 101–8). As the Peruvianexperience under Fujimori showed, nonetheless, these policies were not incom-patible with economic populism. Fujimori managed social policies through anextraordinary concentration of power in the executive and by relying upon di-rect paternalistic relationships that were conducive to the micro-level exchangeof material benefits for political support, even in a context of macroeconomicausterity (Roberts 1995, 106).2

This is paradoxical, given the link of classical populism with nationalism, anti-imperialism, and state intervention. For precisely both of these reasons—the useof populist rhetoric supporting neoliberal economic and political strategies, andthe type of ideological legitimation—I use the term “neopopulism” in order todifferentiate “classical” populism from its contemporary forms. Weyland definespopulism in a similar way, as “a political strategy through which a personalistic

136 Mayorga

leader seeks or exercises government power based on direct, unmediated, unin-stitutionalized support from large numbers of mostly unorganized followers”(Weyland 2001, 20 –21). This definition better fits neopopulism. It does not ex-plain, however, past forms of populism, nor, more importantly, why and howpopulist leaders like Juan Perón, Raúl Haya de la Torre, and Victor Paz Es-tenssoro, who were not democrats in the liberal sense, had a well-organized andstructured mass support and engaged in institution building.

Most importantly, neopopulism differs from historical populism in that its fun-damental characteristics include not only Caesaristic politics but the phenome-non of outsiders springing up from outside the established party system. At firstglance, the emergence of outsiders seems akin to “thunder in a clear blue sky.”But outsiders become key players essentially because of an auspicious context: acrisis of governability and a profound decay and breakdown of party systems.Thus, to understand the rise of outsiders and its disastrous consequences for dem-ocratic development, it is critical to draw attention to the fundamental fact thatwhen parties as government agents fail to perform reasonably in tackling the fun-damental problems and needs of citizens, they lose their capacity for political rep-resentation. Both phenomena engender a power vacuum that outsiders can ex-ploit for their benefit.

Neopopulism in Peru has turned out primarily to be somewhat of an odd mar-riage between anti-politics and neoliberalism, aimed at reducing the state and es-tablishing a market-centered economy. In contrast, in Venezuela, neopopulismhas been linked to statist economic policies that are more compatible with clas-sical populism. Despite the predominant economic policy of orthodox controlof fiscal deficits, Chávez has made a contradictory attempt to return to state-ledcapitalism with the 49 decrees of December 2001, which quickly triggered thewidespread opposition of powerful business groups.3 Neopopulism is an am-biguous and flexible phenomenon that has gained ground by assuming two ide-ologically different stances that—using traditional categories—could be labeledas “left-wing” and “right-wing.” Contrary to past populism, however, the right-wing tendency has linked neopopulism and anti-politics with neoliberal adjust-ment policies, as the cases of Fujimori in Peru and Bucaram in Ecuador demon-strate (Mayorga 1995; Weyland 1996, 2001; Novaro 1996; Knight 1998).4

Drawing upon recent research and my previous work (Mayorga 1995; Wey-land 1996, 2001; Roberts 1995), this chapter posits two key theses. I contend, first,that the emergence of neopopulism and anti-system actors has been the outcomeof two main processes: the decomposition of party systems and a deep crisis ofthe state—in fact, a crisis of governability. Second, I argue that in order to fleshout these reflections on common patterns and qualitative differences in thecases of Peru, Venezuela, and Bolivia, it is critical to pay attention to political-institutional contexts and processes. An institutional approach provides satisfac-tory theoretical tools for addressing these issues, focusing on the relevance of


institutional variables such as party systems, electoral systems, state structures, andgovernability problems.

As Linz has stressed, structural characteristics like class and economic struc-tures constitute a series of opportunities and constraints for both social and polit-ical actors and political institutions. Yet structural variables are not laws thatcausally determine historical and political development. Within a given institu-tional context, actors adopt choices and make decisions that substantially affectpolitical outcomes. Since these decisions are not structurally predetermined (or“structured contingencies”), outcomes are probabilistic and influenced by con-tingencies and chance—that is, more than one outcome is possible. For this rea-son, the variable of leadership can be decisive and cannot be predicted by anymodel (Linz 1978, 4 –5). Therefore, my historical-institutional approach focuseson the interaction between political institutions and contexts, on the one hand,and processes and decisions of political actors, on the other.

Causes of Emergence: Crises of Governability and Political Representation

Several key questions underlie the emergence of neopopulist outsiders. Why didpolitical parties fall prey to a structural crisis? Why did a collapse of the party sys-tems take place in Peru and Venezuela? Why and how did outsiders seize powerin these countries, and why did this not happen in Bolivia? Turning to the firstquestion, political parties and party-based governments in Peru and Venezuelaput the sustainability of democracy in jeopardy. Parties and their leaders were un-able to respond with effective policies to aggravating social and political problemscaused by socioeconomic decline and state crisis in a period of collapse of thestate-centered economy, requiring a shift to a market-centered economy. Theseproblems were not only structural but also a result of political decisions and badperformance of governing parties, thereby creating a providential scenario forthe rise of outsiders. In this sense, the key problems provoking the decline of par-ties were fundamentally problems of governability and not of political represen-tation in terms of the reflection of societal interests and demands.5

The crisis of political representation unfolded in several stages as an outcomeof an underlying and deepening crisis of governability that caused a growing gapbetween society and the parties, and consequently a crisis of political represen-tation. Over time, the dominant political parties in Peru and Venezuela suffereda significant loss of votes and seats because a majority of voters no longer trustedthem due to their failure as governing parties. These parties were still able to wina significant share of votes and seats, in the 1990 elections in Peru and the 1998elections in Venezuela, yet they fell short of winning presidential elections. In apresidential system, losing the presidential contest to outsiders amounts to a dra-matic loss of political representation and power.

138 Mayorga

Analyzing the crumbling of political representation as a backdrop for theoutsiders’ rise, it is necessary to dwell on the double dimension of politicalrepresentation. In a democratic system, political representation consists funda-mentally of a transfer of power to party representatives by citizens through fairand transparent electoral processes (Manin 1997; Sartori 1999). The fundamen-tal assumption of political representation is that representatives must articulate theinterests of society by “acting in the best interest of the public,” both in the ex-ecutive and in parliament.6 Furthermore, as Sartori contends, two souls and de-mands coexist in representative government: to govern and to represent (Sartori1999, 269). Since modern democracy is a system of representative governmentbased on parties, and elections have the paramount aim of leading to the forma-tion of a legitimate government with the responsibility to govern, it is a mistaketo address problems of political representation solely in terms of parties’ ability toreflect electoral preferences. Although the main political parties in Peru and Ven-ezuela did not lose political representation all at once, they nevertheless failed asrepresentative actors after performing poorly in government. As a consequence,they lost presidential power and subsequently were unable to survive as opposi-tion parties.7

A context of worsening socioeconomic crisis and the decline of political par-ties as governmental actors brought about a crisis of governability, providingauspicious conditions for the rise of outsiders and the anti-political logic ofneopopulist discourse. Prompted by the traditional ruling parties’ failure andloss of credibility, outsiders could present themselves as a radical alternative tothe party system and political elites, and as charismatic leaders claiming to carryout a mission of national redemption. Fujimori and Chávez played the politicalgame by established electoral rules; however, they claimed their authority notfrom democratic principles and rules, but from a higher legitimacy as charis-matic leaders.8 For this purpose, they used a radical, anti-political discourse as aneffective tool for identifying themselves with the needs of excluded people,playing the role of paternalistic leaders who embodied—more effectively thandemocratic institutions—the unity of the state and the people.

My second thesis is that the neopopulist politics of outsiders are not only apolitical strategy and an anti-institutional style of politics—as Weyland holds—but a strategy leading to the weakening and breakdown of liberal representativedemocracy and, particularly, to its transformation into a plebiscitary democracy.The so-called “return of the leader” has meant the destruction of democratic in-stitutions and the rise of authoritarian political regimes. When Fujimori wassworn in for his second term in July 1995, the question arose as to whether hewas an exceptional case or the spearhead of a new type of dictatorship that couldspread to other countries in Latin America (Rospigliosi 1995, 314). As events inVenezuela later demonstrated, the tendency toward authoritarian neopopulismhas not been an exception.


The breeding ground for outsiders and neopopulism has been complex andmultifaceted. Political-institutional and leadership-mediated causes are the cru-cial explanatory factors, while socioeconomic problems (which were both causesand effects) have constituted a critical context. Social cleavages, inequalities, andfragmentation deepened greatly during the 1980s. The rise of the informal econ-omy and the atomization of social groups impaired social organizations—and es-pecially labor unions. At the same time, economic decline generated unemploy-ment, hyperinflation, dissatisfaction with political parties, and, eventually, thedemise of populist policies (González de Olarte 1998; Naim 2001). These socio-economic factors put a strain on the whole political system, and particularly onthe party system and the state’s ability to cope with socioeconomic crisis. Par-ties—both in a polarized party system such as Peru’s and a moderate party sys-tem such as Venezuela’s—could not adequately respond to this crisis. Due to aprogressive weakening of the basic functions of representation and governancethat they exercise in a democratic system, parties lost their linkages to socialorganizations.

Thus, the key factor explaining the emergence of outsiders is the dramatic cri-sis of party systems resulting from a failure of democratic governability. From theonset of the transition to democracy in Peru, political parties failed as govern-mental actors to carry out policies that could solve the population’s grim socio-economic problems (See Table 10.1 in Mainwaring’s concluding chapter in thisvolume). Furthermore, they did not modernize their patterns of action, persist-ing in a zero-sum game of confrontation. Democracy implies not only dissentand confrontation but also consent and agreement on fundamental issues. But thepopulist governments of AP (Acción Popular, or Popular Action) (1980 –85) andAPRA that preceded Fujimori’s government used their electoral victories as carteblanche for a vertical style of leadership, refusing negotiation and agreementswith the opposition (Lynch 1999, 260). The presidential system and the tradi-tion of personalistic politics embodied in caudillismo nurtured this tendency ofpresidential power, with the executive excluding the opposition and exertingcontrol over all state institutions. In Venezuela, “the populist system of eliteconciliation” and the dominant party system of AD (Acción Democrática, orDemocratic Action) and COPEI (Comité de Organización Política Electoral In-dependiente, or the Committee for an Independent Electoral Political Organi-zation) were for decades the lynchpin of democratic stability and a strong presi-dentialist system (Rey 1991; Kornblith 1998). Venezuelan democracy was mainlybased on negotiation and consent. However, “partyarchy” had perverse effects,leading to a “pathological kind of political control” (Coppedge 1994, 2) andto pragmatism as the predominant political style: “Ironically, the same charac-teristics of parties that had promoted democratic governance in the first twodecades of the regime worked to undermine it in the last two decades”(Coppedge 2002, 10 –11).

140 Mayorga

Because political parties in government failed to ameliorate economic and so-cial problems, they lost their capacity to represent and channel social interests. Atthe end of García’s government, between 1987 and 1989, Peru’s GDP had de-creased by 20 percent, poverty had grown, and salaries had lost almost 60 percentof their purchasing power. By 1988, the inflation rate had reached 1,722 percent,peaking at 7,649 percent in 1990. In this context, the war waged by Shining Pathhad produced almost seventy thousand victims and economic losses of aboutUS $20 billion, equivalent to Peru’s external debt at that time (Ferrero 1993;McClintock 1989; Comisión de la Verdad y Reconciliación 2003). In Venezuela,poverty had increased dramatically in the last two decades. Beginning in 1983, thesuccessive governments of AD and COPEI could not stem the economic decline,although they had the opportunity to take advantage of two oil booms in the1970s. At the time of Chávez’s rise to power, 68 percent of the population livedbelow the poverty line. Moreover, per capita income in real terms was back atthe level of 1962, unemployment had reached 15 percent, and 45 percent ofthe workforce was employed in the informal economy (Naim 2001, 21). Tosummarize, at the end of the 1980s parties became targets of discredit and distrust.In both Peru and Venezuela, a crisis of party legitimacy erupted, mostly as a con-sequence of bad performance, inefficiency, and, last but not least, corruption. Asa consequence, people affected by the socioeconomic crisis turned to neopopulistoutsiders who promised to overcome poverty, corruption, and social inequalities.

Meanwhile, parties became hermetic organizations that were increasingly aliento society. In Peru, parties detached themselves from underlying developments insociety, thereby losing their constituencies by sticking to the traditional gameof movement-like (movimientista) politics while unions were losing their grip.Political representation and electoral politics were further eroded by a media-dominated logic of political competition that had begun to hold sway over poli-tics (Tanaka 1998, 180 –82). While crucial transformations of the social structureand the political arena were taking place, at the end of García’s government, self-sufficient parties were not able to change their practices and strategies. Thus, thethesis that political parties in Peru committed suicide and were not victims ofmurder sounds quite adequate (Lynch 1999, 257). In Venezuela, although thepresidential system was often prone to stalemate following the 1958 pact of PuntoFijo that restored democracy, partyarchy was instrumental and successful in mod-erating political parties’ conflict and guaranteeing democratic stability and gov-ernance. Nonetheless, the overwhelming control over the political system andcivil society generated a dangerous lack of horizontal accountability, abuses ofpower, widespread corruption, and impunity. Thus, as the economy went down-hill and poverty spread in the 1980s, the parties could not switch entrenched po-litical practices. They instead became “accomplices to their own destruction” and“accomplices also in the sense that they stubbornly and tragically resisted pressureto reform themselves” (Coppedge 2002, 10 –11).


The Process of Decline and Collapse of Party Systems

A useful theoretical approach is to analyze the breakdown of predominant partysystems and political leadership in a context of economic, societal, and state cri-sis as a sequential and perhaps also patterned process that will allow generaliza-tions in the Andean region.9 In Peru and Venezuela, the process of the declineand breakdown of representative democracy and party systems evolved in threedistinct phases: (1) a phase of detachment and estrangement of parties vis-à-vissociety, and their weakening due to internal struggles and inability to change;(2) a phase of electoral defeat, party breakdown, and democratic takeover ofpower by outsiders; and (3) a phase of final destruction of parties, wrought byFujimori’s autogolpe and an ensuing constituent assembly in Peru, and in Vene-zuela by a constituent assembly, after which Chávez achieved full control ofpower, dominating the executive, Congress, and the judiciary.

In Peru, the first stage was a protracted process spanning a decade, from 1980to 1990, in which socioeconomic and political problems worsened (Cotler2000). Democratization converged with acute problems of governability, in-cluding economic crisis and hyperinflation, a wave of strikes, the fierce offensiveby Shining Path, and the crisis and inability of parties to respond to these prob-lems. By the end of the decade, most parties found themselves in a deep internalcrisis, characterized by internal struggles within APRA and FREDEMO (FrenteDemocrático, or Democratic Front) and a fragmentation of the left (Tanaka1998, 170 –73; Lynch 1999, 254 –57). At the time of the November 1989 mu-nicipal elections and the April 1990 presidential election, “The political class asa whole was alien to citizens’ worries and locked in intra-party struggles”(Tanaka 1998, 173). The terrorist war that Shining Path had waged against thestate and society had reached Lima, sparking uncertainty and fear. Thus, in theprocess of democratization beginning in 1979, Peru had not addressed majorproblems in its society and political system. Unlike Bolivia, the democratizationprocess led to the decomposition of the party system and traditional leadership.

Two problems emerged together. On the one hand, parties could solve nei-ther socioeconomic problems nor the problems caused by guerrilla violence andterrorism. On the other, party organizations were unable to adapt to structur-ally rooted changes in political representation and to the new logic of interpartycompetition. Party strength was no longer determined by the mobilization ofinterest groups and corporate social actors, but by a logic of media-structuredlinkages to a diffuse, fragmented public (Tanaka 1998, 92–93, 168). Partiescould not and were not willing to reshuffle their structures and change boththeir political styles and policy orientations.

The second stage in the process of decline was the electoral defeat of tra-ditional parties and Fujimori’s democratic election in April 1990. At the end of 1989, popular perceptions about parties began to undergo a dramatic change.

142 Mayorga

A first sign that the scenario for the emergence of outsiders had been set was the November 1989 election of Ricardo Belmont, an outsider, as mayor of Lima. The trend was reinforced when the famous writer Mario Vargas Llosa, also an outsider, became the leading candidate of a new political front,FREDEMO. During the 1980s, high electoral volatility and a high degree ofpolarization and confrontation characterized the Peruvian party system. Theelectoral campaign of 1990 was no exception. Vargas Llosa polarized the elec-toral contest mainly because of his proposed economic shock program, prompt-ing the APRA government and leftist parties to make him the main target oftheir attacks. President García turned his back on his own party’s candidate, AlvaCastro, deciding two months before the election to support the unknown Fu-jimori instead. Thus, Fujimori obtained strong government backing. AfterMarch 11, 1990, when polls gave Fujimori only 3 percent of voter preference,government-linked media gave him a decisive boost. Regional developmentcorporations also offered the tractors Fujimori used so effectively in his cam-paign across the country.10 By mid-March, his support had reached 9.5 percent,and on the day of the first-round election, April 8, 1990, he obtained 29.1 per-cent of the vote. Finally, in the May 1990 runoff election, Fujimori achieved aresounding victory with 56.7 percent of the vote. Were both results really a sur-prise, then? Did Fujimori rise to power by accident and chance, as Tanaka(1998, 164) suggests?

The collapse of the party system—the final stage—occurred between 1992and 1995. Why did this collapse take place? According to Tanaka, who takesissue with “retrospective determinism,” the collapse of the party system andthe ensuing breakdown of democracy were not inevitable (Tanaka 1998, 200).Events could have transpired differently. The opposition parties could havepressed Fujimori to step down, since the electoral results in 1990 had not led toa catastrophic defeat of FREDEMO and the traditional parties. FREDEMO ob-tained 32.3 percent in the first round of the presidential election, and 30.1 per-cent of the seats in the Senate and the Chamber of Deputies. APRA obtained25.1 (presidency) and 21.5 percent (Congress), respectively, while the UnitedLeft (Izquierda Unida, or IU) obtained 9.8 and 10 percent, and IS (IzquierdaSocialista, or Socialist Left) 5.5 and 5.3 percent, respectively. Cambio 90 (Change90), Fujimori’s movement, became a minority faction in Congress, obtainingonly 21.7 percent of Senate seats and 16.5 percent of the seats in the Chamber ofDeputies. The outcome was a Congress in which opposition parties had an over-whelming majority and, consequently, a minority government, leaving Fujimoriin a very weak position. Moreover, Fujimori had neither organized party sup-port nor a government team with which to govern.

Fujimori’s politics and the choices made by the opposition parties broughtabout a showdown, eventually causing the breakdown of the party system. First,with the crucial aid of Vladimir Montesinos, Fujimori built a coalition with the


military. Their pact was based on the “Green Plan” that the armed forces haddeveloped with the aim of establishing a regime under military control. Fujimoriresorted to an authoritarian project anchored in the Green Plan to establish amarket-oriented economy within the context of a “controlled democracy”(Rospigliosi 1995, 311–12; Reyna 2000, 141). Second, he forged a coalitionwith de facto powers—business groups, foreign investors, international organi-zations—in the framework of his neoliberal economic policies. Third, by 1991,successful economic policies provided Fujimori with great legitimacy. Finally,based on the authoritarian project, he decided to override congressional oppo-sition and subordinate other democratic institutions.

Because he was the head of a minority government, Fujimori asked Congressfor emergency powers in order to cope with the economic crisis. In November1991, he deliberately presented 124 bills to Congress at once, but Congress de-manded a partial revision. Fujimori’s underlying motive was to provoke a show-down with Congress by accusing it of incompetence and obstructive opposi-tion to government policies. A stalemate between the legislative and executivebranches ensued, and Fujimori reacted by threatening to close the Congress. Bythat time, Fujimori and the parties in Congress were adversaries, engaged in a“war of attrition” (Tanaka 1998, 213) that culminated with the April 5, 1992,autogolpe that shut down Congress, the judiciary, and other state institutions.Due to the success of economic policies to reduce inflation and create stability,Fujimori won a decisive battle against the parties and other institutions with thebacking of the armed forces, big business and, most notably, overwhelming pop-ular support. Politically, Fujimori justified his move by arguing that the decom-position of prevailing institutions, widespread chaos and corruption, and theobstacles posed by Congress and the judiciary made effective governance im-possible. In addition, Fujimori contended that prevailing democratic institu-tions were “deceptive and false,” and that it was necessary to take an “excep-tional approach” to advance the process of national reconstruction (Manifiesto ala Nación, April 5, 1992). Thus, the autogolpe was one part of the strategy to con-centrate power. The Constituent Assembly was the other part.

In response to pressure from the Organization of American States (OAS), Fu-jimori decided to redemocratize the political regime by calling for a Constitu-ent Assembly in November 1992. The Constituent Assembly introduced a newconstitution that entrenched executive powers and established a provision thatallowed for the reelection of the president. The main parties of the traditionalparty system—APRA, AP, and IU—refused to participate in the assembly. Onlythe PPC (Partido Popular Cristiano, or the Popular Christian Party) and a newleftist front, MID (Movimiento de la Izquierda Democrática, or DemocraticLeft Movement), took part, but they suffered a grave defeat in the election forthe Constituent Assembly, whereas Fujimori’s candidates (Cambio 90 –NuevaMayoría, or Change 90 –New Majority) pulled in 49.2 percent of the vote.

144 Mayorga

A contested referendum in October 1993 approved the new constitution with aslight majority of 52.3 percent of the vote. A historic defeat of the traditionalparties occurred in the April 1995 presidential election, in which Fujimori won62.3 percent of the vote while all the traditional parties combined did not ob-tain even 5 percent. It was the first time ever in the contemporary history of de-mocracy in Latin America that a general election demonstrated the completebreakdown of the traditional party system.

In Venezuela, the process of party system decay and decomposition was evenmore protracted than in Peru. The process spanned fifteen years, from 1983 to1998. It began slowly, with an economic crisis in 1983 stemming from a severefall in oil prices and a massive flight of capital. That led to the first devaluation ofthe bolivar (the Venezuelan currency) in twenty-five years. The 1980s witnessedthe increase of the external debt, the instability of oil markets, capital flight, fiscalcrisis, administrative corruption, and, most notably, inflation (Kornblith 1998,1–32). An attempt by the government of Carlos Andrés Pérez (1989–92) to ad-dress the crisis with a neoliberal program failed, provoking a violent upheaval inthe nation’s capital, the Caracazo, in February 1989. In hindsight, the Caracazoturned out to be the turning point in the crisis of Venezuelan democracy. An endhad come to the economic stability and prosperity generated by the rent econ-omy based on oil wealth. From 1978 to 1989, per capita GDP shrank 29 percent;from 1985 to 1998, declining oil prices hit the country hard. The democraticstability stemming from the partyarchy established by AD and COPEI since1958 also came to an end, as the two coups d’état in 1992 and the impeachmentof President Pérez in the same year made evident. Electoral abstention grew from12 percent in 1983 to 39.8 percent in 1993, an indication of increasing disaffec-tion with the party system (Coppedge 2002, 11). The electoral victory of RafaelCaldera in the December 1993 presidential election signaled a second turningpoint in the process of party system decline. Caldera won 30.46 percent of thevote as the candidate of an alliance of small parties—MAS (Movimiento al So-cialismo, or Movement toward Socialism), Convergencia Nacional (NationalConvergence), the Communist Party, and others. His thoroughly anti-politicalcampaign triggered a serious delegitimization and breakup of the two-party sys-tem and its transformation into an unstable and weak multipartism (Molina andPérez 1994, 74 –75). The AD candidate, Claudio Fermin, and the COPEI can-didate, Oswaldo Alvarez, obtained 23.6 and 22.73 percent, respectively; in thecase of AD, this was half the percentage won in 1988. This weakened, fragileparty system broke down completely in the December 1998 election.

Another key factor in the process of decomposition was the failure of con-stitutional reforms and attempts to change political parties (Kornblith 1998,165–82). Paradoxically, the dominant parties undertook a process of constitu-tional reforms starting with the proposals made by the Presidential Commissionfor the Reform of the State (COPRE) at the end of the 1980s. Their goal was to


stave off the weakening of the political system and to restore the party system’slegitimacy. COPRE proposed sweeping reforms including decentralization, di-rect election of governors and mayors, the mixed-member electoral system, ref-erendums at the state and national level, and internal democratization of parties.With the exception of the last two reforms, Congress enacted decentralizationand the reform of the electoral system in 1989. The two coups d’état in Febru-ary and November 1992 produced a critical situation, however, that furthercrippled constitutional reforms (Kornblith 1998, 61–114). The suspension ofthe constitutional reform process damaged Congress’s reputation and weakenedthe parties and political leadership.

Attempts to curb the political system’s deterioration through state reformssuch as decentralization and electoral reforms were unsuccessful. The question,then, is why the reforms were not enough to restore Venezuelans’ faith in theirpolitical system and why they did not prevent the progressive delegitimizationof democracy (Coppedge 1994, 164). Arguably, institutional reforms were car-ried out too late and only partially. They did not have an impact on solving cru-cial problems. According to Kornblith, the leading parties, AD and COPEI, didnot share common criteria and were not really committed to political reforms(1998, 11–114). They lacked both the necessary ability to innovate and the po-litical will to carry out reforms. In a similar vein, Coppedge (1994, 164) con-tends that the basic problem has been that while political parties were the onlyactors in a position to adopt reforms, they were unwilling to make such reforms.This turned out to be one of the key problems. By the mid-1990s, after thefailed coups d’état and the impeachment of Carlos Andrés Pérez, doubts aboutthe stability of the democratic regime had become evident.

The final stage of breakdown of the Venezuelan party system took place overa two-year period that included the 1998 presidential election, the ConstituentAssembly, the 1999 referendum, and the July 2000 general election. The No-vember 1998 presidential election was the turning point in a fifteen-year processof decline of the dominant parties. Accordingly, Chávez did not destroy the oldparties; he rather filled a political vacuum (Coppedge 2002, 14). Chávez won thispresidential election in a landslide, with 56.2 percent of the vote, while the can-didate of the governing party, Convergencia Nacional (Salas Römer), those ofthe AD and COPEI, and the independent, Irene Salas, obtained low percentagesof the vote. AD fared relatively well in the congressional election that took placeone month before the presidential election, winning 24.1 percent of the vote,which was similar to the percentage the party had obtained in 1993. COPEI wasthe big loser, garnering only 12 percent of the vote. In comparison, Chávez’sparty won 19.9 percent of the vote, and his ally, MAS, obtained 8.9 percent.

Thus, separate congressional and gubernatorial elections in 1998 broughtabout a minority government and the danger of a deadlock between the execu-tive and Congress. The prevailing constitution ensured that Chávez’s adversaries

146 Mayorga

would control Congress and other institutions. Chávez controlled only one-thirdof the seats in the two chambers. This fact, and not the alleged shortcomings ofthe 1961 Constitution, was what prompted Chávez to convene the ConstituentAssembly. His primary motivation was not to tinker with the constitution, but touse the Constituent Assembly strategically as a mechanism to concentrate powerand to neutralize Congress, the courts, and all other guarantors of horizontal ac-countability (Coppedge 2002, 17–18). After another landslide in the Constitu-ent Assembly election, in which his alliance won 122 out of 131 seats, Chávezreinforced his strategy of forging a strong, personalized power base. He substan-tiated this in the 2000 presidential election, winning 59.75 percent of the vote,while the AD candidate obtained only 2.72 percent (Molina 2000, 34).

In conclusion, from a comparative perspective the rise to power of Fujimoriand Chávez was characterized by similarities and differences in both origins andprocesses. The key similarities lie in the decomposition of the party system, thefailure of political elites, and the ensuing governability problems that set the stagefor the emergence of outsiders. Important similarities also relate to the incentivesof presidentialism for outsiders and institutional arrangements such as the elec-toral runoff system. As Linz argues, there are structural reasons for the candidacyof outsiders. While an institutionalized party system makes it difficult for out-siders to enter a presidential competition, “the personalized character of a presi-dential election makes possible, especially in the absence of a strong party system,the access to power of outsiders” (Linz 1994, 26 –27). The differences are rootedmainly in contingent political causes and certain contextual factors, such as guer-rilla terrorism in Peru, the coups d’état in Venezuela, and self-destructive deci-sions of political actors.

Contingent political causes—associated with the rationality of actors, andspecific choices and decisions that cannot be explained as a predetermined, logi-cal result of the political crisis—also had a major impact on the triumph of out-siders. Key political actors such as Alan García in Peru (president from 1985 to1990) and Rafael Caldera in Venezuela (president from 1993 to 1998, and ear-lier from 1968 to 1973) helped anti-system actors like Fujimori and Chávez toseize power. Caldera himself came to power in 1994 with an anti-establishmentelectoral campaign. Once he took office, he pardoned Chávez, who was in jailafter the failed 1992 coup d’état. Nonetheless, these contingencies do not justifyTanaka’s (1998) conclusion that the emergence of outsiders was thoroughly con-tingent in Peru. Tanaka’s (1998, 197) claim that “Fujimori’s rise to power is verymuch a random product” understates structural causes, favorable strategic con-texts, and the rationality of actors prompting Fujimori’s success. As Tanaka him-self asserts, the crisis of the party system and the electoral runoff system paved theway for and benefited Fujimori. In a plurality system, Fujimori never would havecome to power (Tanaka 1998, 197; Mayorga 1995, 57). In an extremely polarizedcampaign, he took advantage of the strategic vote of anti–Vargas Llosa voters.


From the standpoint of rational choice, it might have been rational for the Gar-cía government, APRA, and leftist voters to vote for Fujimori in the runoff. Al-though in the end this decision proved to be self-destructive for them and otherparties, their disenchantment with the traditional parties, as well as the polariza-tion of the campaign between a candidate linked to the traditional elites and anunknown candidate associated with the Cholo and Indian population, made it ra-tional for anti–Vargas Llosa voters to prefer the unknown demagogue. Besides,given the alleged weakness of a future Fujimori government, APRA was count-ing on the possibility of controlling Fujimori.

Consequences: Neopopulism as a Political Regime

Both in Peru and Venezuela, the politics of outsiders have led to the same far-reaching consequences: full control of the state with the support of a majorityof the population and—as a key outcome—a plebiscitary democracy leading tothe weakening and demise of liberal, constitutional democracy.11 Neopopulistregimes in Peru and Venezuela emerged as electoral democracies, but they havemainly been regimes based on plebiscitary mechanisms, restricted pluralism,concentration of power in the head of government, the elimination of mecha-nisms of horizontal accountability, and popular demobilization.12 These regimesderive their legitimacy not only from democratic elections but also from thehigher or deeper legitimacy of the leader himself, owing to the plebiscitary char-acter of leadership. As Weber argues, the leader attains the confidence and trustof the people through mass-demagogic means. This kind of leadership is Cae-saristic, and its main tool is the plebiscite (Weber 1964, 1094). Thus, the corre-lation between neopopulism, deinstitutionalization, and autocratic rule has be-come manifest (Roberts 1995, 116; Weyland 2001, 25).

From the outset, neopopulist regimes under the sway of outsiders have beenmarked by a tension between democracy based on popular sovereignty (i.e., elec-toral majorities) and liberal democracy (based on constraints on presidential ac-tion, horizontal accountability, and checks and balances). Chávez’s regime is bestdefined as illiberal because his political movement controlled the executive, thecourts, and the legislature, and handpicked all the members of supposedly inde-pendent agencies (Coppedge 2002, 15–16, 36). In the context of the decline andcollapse of the party systems and of the frailty of state institutions, Fujimori andChávez carried out a politics of tabula rasa, or a deliberate politics of deinstitu-tionalization. Although democratic-liberal institutions were not suppressed(except for the period from April to December 1992 in Peru), they were sub-verted and eroded, while loose political movements and the growing politicalintervention of the armed forces filled the vacuum left by parties. Furthermore,Fujimori’s regime—and, increasingly, Chávez’s—has clearly shown that, in theabsence of parties, neopopulist regimes can turn into outright dictatorial regimes,

148 Mayorga

not only by eroding civil liberties and concentrating power, but also by substi-tuting the armed forces for parties. The breakdown of parties as mediation struc-tures prompted a tendency toward autocratic regimes, proving that no workingdemocracy is possible without parties.13

Is the concept of delegative democracy useful for defining and explainingneopopulist regimes? As O’Donnell depicts it, delegative democracy rests on thepremise that a candidate winning the presidency is entitled to govern as he orshe sees fit. Elected presidents present themselves as above political parties andorganized interests (O’Donnell 1994, 59– 60). This concept stresses both thecrucial democratic element of electoral legitimacy and the absolute predomi-nance of the presidency. Yet the concentration of power and the destruction ofindependent democratic institutions have been so extreme in Peru and Venezu-ela that the concept of delegative democracy does not sufficiently encompass theextent to which representative democracy has been undermined. The categoryof semi-democratic regime is more pertinent because it implies a significant ortotal removal of checks and balances, an absence of horizontal accountability,human rights abuses, and wide autonomy for the armed forces (Levitzky 1999,80; Mainwaring 1999, 102).

Since it has led to pervasive authoritarianism, neopopulism ought to be as-sessed not only as a political strategy of outsiders but, when successful, also andforemost as a strategy breeding an authoritarian, dictatorial regime, or at least anilliberal regime. Fujimori’s regime, in particular, turned out to be an extreme caseof an authoritarian regime with a “varnish of legitimacy” that degenerated intoa government of corrupt cliques that made political decisions as if they were statesecrets (Grompone 2000, 109). After the autogolpe in April 1992, Fujimori’s gov-ernment depended upon an extended spoils system and not on a pact of domina-tion or political hegemony. Although he lacked his own political organizationand power structure, Fujimori built a broad power coalition consisting of him-self and the de facto powers—the military, business, and international organiza-tions (Lynch 1999, 244 –52). An inner circle made up of Fujimori, VladimirMontesinos, and the military eventually dominated this alliance. It quickly be-came a criminal mafia and a corrupt gang of cronies engaged in embezzlement ofpublic funds, blackmail and corruption of media and business groups, influencepeddling, illicit enrichment, and arms dealing. The most disturbing aspect ofFujimori’s government was that Montesinos, who directed the SIN (NationalIntelligence Service), became the all-powerful executor of the pact with themilitary that grew out of the strategic Green Plan (Rospigliosi 1995, 329–31;Grompone 2000, 95–97; Reyna 2000, 138– 41). Through the undergroundleadership of Montesinos, Fujimori carried out a strategy of toma de casilleros14

within the armed forces and business groups (Grompone 2000, 95). The pactwith the military was not made with the armed forces as an institution but withthe highest-ranking officers within a system of personal mutual loyalties and


favors, which Montesinos firmly controlled. From the offices of the SIN, Mon-tesinos forged the pact by dismantling the armed forces’ institutional structuresand hierarchies and by arbitrarily planning promotions and passing over officerswho were due for promotion (Rospigliosi 1995; Obando 2001).

The SIN extended its responsibilities into the armed forces and thoroughlypenetrated them through a wide range of activities. The service became a watch-dog over the military, infiltrated the ministries and the state administration, gath-ered information about members of the opposition by tapping telephones andbribing members of Congress, exerted control over media information, andmanipulated electoral campaigns (Grompone 2000, 101; Reyna 2000, 138– 43).Due to the absence of institutionalized power structures, Fujimori’s autocraticregime depended heavily on the secret service led by Montesinos. As the SIN be-came the core of Fujimori’s power, his political fate was closely linked with Mon-tesinos’s scheming. In the end, Montesinos became not only the guarantor of Fu-jimori’s power but also the source of his ruin and downfall, which took placeshortly after the start of his third presidential term in July 2000. The collapse ofFujimori’s regime was not the result of the resurgence of a vigorous political op-position but of the scandal over a video that led to Fujimori’s resignation and re-vealed how dependent he had become on the network of corruption that Mon-tesinos managed.

Similarly, because he lacked his own political organization and power base,Chávez created a political movement in Venezuela, the MVR (Movimiento V.República, or Fifth Republic Movement), for the 1998 election. The MVR wasa loose organization that joined civilian and military members of diverse leftistorigins who had brought together the popular sectors supporting AD andCOPEI in the past. Nevertheless, the MVR was more than an electoral deviceand façade; it could have transformed itself into a single, hegemonic party, per-haps, and into an effective instrument of power. For the election, the MVR builtan alliance with small leftist parties such as the MAS and the PPT (Patria para To-dos, or Fatherland for All). Fourteen months after Chávez seized power, how-ever, the MVR split up when Chávez himself decided to get rid of it by reorga-nizing his original movement, the golpista MBR-200 (Movimiento BolivarianoRevolucionario, or Bolivarian Revolutionary Movement). Apparently, Chávezdisbanded the MVR because he saw it as an obstacle to his strategy of establish-ing a personalistic dictatorship.

Neither movement has been a political organization on which Chávez hasbased his power. He does not rely nor does he intend to rely upon a broad na-tional political structure, that is, on a party capable of organizing the masses as wellas mediating and channeling interests and conflicts. As a popular leader, Chávezhas attempted instead to make up for this lack of political structure in two ways:by creating direct, plebiscitary links with his constituency through the so-called“Bolivarian circles,” and by turning the armed forces into his political instrument

150 Mayorga

for state administration (Gómez Calcaño and Arenas 2001; Gómez Calcaño2002). Chávez has engaged both active and retired military officers in the execu-tive at the national and regional levels, in state enterprises (especially PDVSA, thehuge state-owned oil company),15 and particularly in the Plan Bolívar, which hasput the armed forces in charge of repairing schools, building infrastructure, pro-viding health care for the poor, and selling basic goods (Trinkunas 2002, 68– 69).

In short, Chávez’s strategy has been to transform the armed forces into his ownpolitical instrument, engaging them directly in state administration and in an ar-ray of activities that go beyond the responsibilities defined for them in the con-stitution (Trinkunas 2002, 65– 66; Manrique 2001, 325–26). According to someanalyses, a new model of military intervention in politics through the leader hasemerged. The model is characterized by the politicization of the armed forcesand, according to Manrique (2001, 327), even by their transformation into a“military party,” since they have taken on functions inherent to a political partyand hold key posts in the system of decision making. This thesis is debatable, how-ever. The armed forces are not organized as a party; they have no need to legit-imize their existence or to participate in elections, which is the fundamental func-tion of a party. Instead, the armed forces play a political role reluctantly, subjectto the political will and imposition of the charismatic leader on whom they de-pend. They are internally split into three factions: pro-Chávez officers (so-calledrevolutionaries by the government), institutionalists, and opponents. The lattertwo constitute the biggest factions, according to a classification of the Military In-telligence Unit (DIM, División de Inteligencia Militar) (Manrique 2001, 330).The military has become an “ersatz” or “surrogate” party in Chávez’s strategy.16

Unlike Fujimori, Chávez’s nationalist, statist orientation has prevented himfrom forging an alliance with business groups; and he has not sought the supportof international organizations. From the outset, he opted for confrontational pol-itics against private business, the media, unions, and even the Catholic Church,triggering a dangerous polarization between them and his own followers.17 Bykeeping political control in his own hands, he has attempted to militarize thestate apparatus and rely on military support to create an inner circle of followers.But apparently, he has neither an operator like Montesinos nor an SIN to assisthim. The DIM does not seem to play a similar role. Chávez’s policies have causeda deep internal rupture, the defection of several officers, and growing oppositionwithin the armed forces to the expansion of the military’s role, to the militariza-tion of the state, and to Chávez’s authoritarian populist policies (Coppedge 2002,27). The failed coup d’état of April 11, 2002, revealed that Chávez could not rallythe armed forces behind him, although he has subsequently achieved significantcontrol over the military.

Chávez feels committed neither to his own constitution nor to representativedemocracy.18 Despite the introduction of the new constitution in 1999, whichstrengthened presidential powers, Chávez claimed that Venezuela lives under a


“regime of constitutional transition” and that he seeks extraordinary powers be-yond those already conferred to him by the ley habilitante (empowering law). Healso stressed that he was making a superhuman effort to carry out a peaceful rev-olution, but that should this fail an armed revolution would be the only alterna-tive (El Nacional, May 5, 2001). This statement was a clear indication that Chávezdoes not respect the constitutional order he helped to establish, and would ratherpursue the path of strengthening his own personal power. Hence, a deep con-tradiction between the constitutional order and Chávez’s personal power projecthas become apparent—a contradiction that does not create an adequate foun-dation for a long-term, institutionalized, neo-populist regime but rather rendersit an oxymoron. From a historical perspective, their reliance on personalisticpower structures makes neopopulist regimes inherently less stable than institu-tionalized democratic regimes, and they tend, in fact, to be short-lived.

Since his rise to power, Chávez’s political project—grounded in a militaris-tic vision of politics—has left little room for doubt about his aim of dismantlingthe previous political regime and imposing a “peaceful revolution” againstmiddle-class, business, and labor interests.19 To achieve this, he resorted to “con-stitutional means” introduced by a constitutional reform, expanding presiden-tial powers under the guise of “participatory democracy” and allowing himselfabsolute legislative and decree powers in any matter.20 The project gained mo-mentum at the end of 2001, when the government emitted 49 presidential de-crees that signaled its course toward stronger state intervention in the economy,particularly in the agrarian sector. At the same time, Chávez’s legislative major-ity—61.2 percent of the seats after the 2000 election—had dwindled to littlemore than 50 percent due to defections in his coalition. Chávez thus sparked awidespread and radical confrontation with his political and social opponents,bringing about a historical rupture with the procedures of negotiation that haddominated party politics from 1958 until 1998.

Both the refusal to allow political bargaining and the instrumental use of “par-ticipatory democracy” have shed light on the authoritarian character of Chávez’sregime (Molina 2003). The 49 presidential decrees were a turning point in therelationship between government and the opposition since they provoked therebellion of the middle classes, organized labor, and business associations, andas a consequence an even deeper rupture within the armed forces.21 In April2002, the first general strike against Chávez’s regime set the stage for an uprisingby high-ranking officers and a coup d’état led by conservative business groups,which intended to establish absolute powers by abolishing the constitution. Thefailure of the coup d’état aggravated the conflict between the government and theopposition, however, given the fact that after his restoration to power, Chávezmaintained his strategy of imposing his “peaceful revolution” and thereby in-creasing the high-handed, autocratic concentration of power. Chávez’s firstreaction to the coup was to start a dialogue of reconciliation with the opposition

152 Mayorga

and to attain a modicum of governability. Yet he also took advantage of a weak-ened and leaderless opposition and the split in the armed forces to strengthenhis political power by purging the armed forces and by putting loyal officers intohigh commands. Simultaneously, the opposition forces came together in the so-called Democratic Coordinator—a loose coalition of eighteen parties and fortynon-governmental organizations, business associations, and labor unions—andswitched their strategy in November 2002 by calling a consultative, non-bindingreferendum that the National Electoral Council declared and supported in orderto cut Chávez’s mandate. When the government refused this demand, the Dem-ocratic Coordinator hardened its stance by calling a general strike in December2002. Private business, trade unions, the media, the Church, and PVDSA work-ers and management all supported the strike as a means to force either arecall referendum on Chávez’s government, his immediate resignation, or earlyelections.

Thanks mainly to the loyalty of the armed forces, the government was able towithstand this assault. As a result, the clash between government and oppositionturned into protracted trench warfare, which weakened the opposition evenfurther. After two months, the general strike failed to achieve its political aims.Once again, Chávez succeeded in clinging to power, while business interests andthe economy as a whole bore the brunt. Between 2000 and 2002 the country’seconomic decline was impressive. While the GDP declined 8.9 percent in 2002,in 2003 it slumped 10 percent, with the inflation rate soaring to 27 percent andthe unemployment rate to 15 percent (“A Tale of Two Years,” Latin AmericanRegional Report: Andean Group, January 6, 2004, 3). Exchange controls, pricecontrols of production costs, the massive import of consumer goods for subsi-dized sale, and an absence of investment marked the economy. Factories thatstopped producing were taken over by the military.

Unexpectedly, the government survived a pervasive and seemingly permanentpolitical crisis, but the price was mounting polarization and ungovernability. Inhindsight, the political crisis led to the defeat of the opposition’s strategy, puttingat least a temporary end to the stalemate in Venezuelan politics in Chávez’s favor.Chávez decided to burn bridges by escalating the pace of his revolution. Aftergaining control over the state oil company in February 2003, he embarked upona “revolutionary offensive” against the media, private business, and the legal sys-tem. He has shown his determination to carry out the policies envisaged in the49 decrees of December 2001 by detaining opposition political leaders, harassingthe media, manipulating the judiciary, and dismissing PDVSA staff. EmulatingJuan Velasco Alvarado’s military regime in Peru (1968–75), Chávez’s govern-ment is apparently heading toward establishing a “mixed economy,” fosteringthe development of a powerful state sector and a sector of agrarian cooperativesunder state control (Molina 2003).


Following its bungled attempt at a general strike, the opposition found itselfwithout a coherent strategy, in a state of confusion and internal struggle. TheDemocratic Coordinator eventually achieved consensus for a much-neededchange of political strategy by accepting the mediation of the OAS and JimmyCarter’s proposal to seek an electoral, democratic solution based on the consti-tution, which seemed to be the only way out. In the end, the opposition yieldedto Chávez’s proposal for solving the political crisis: the recall referendum allowedfor in Article 72 of the Bolivarian Constitution. The government played a doublegame by simultaneously participating in the OAS-backed negotiations with theopposition and embarking on the radicalization of the “Bolivarian Revolution.”Its strategy has been to control potential sources of destabilization politically, tomuzzle the media, to put state institutions and enterprises under tight politicalcontrol, and to boost state planning in the agrarian sector through cooperativesbased on the “model of endogenous development” launched in March 2003. Af-ter seven months of thwarting any electoral solution whatsoever, the governmentfinally accepted a recall referendum in an OAS-sponsored agreement—signedwith the opposition on May 29, 2003—which seemed to integrate both sides’demands. The agreement reflected Chávez’s conviction that he had strengthenedhis power enough to win a recall referendum.

Chávez achieved, in fact, an astonishing victory, with 59 percent of the votes.Four factors help to explain this outcome: the successful strategy of delayingthe recall referendum, assuming that economic recovery and massive publicspending for social programs—the misiones targeting the poor—would bolsterChávez’s prospects for triumph; the tight control over state institutions, particu-larly over the National Electoral Council; the enduring support of Chávez’s con-stituency to his government; and the opposition’s inability to put forward a cred-ible political alternative.22

Chávez’s overwhelming victory in the August 2004 recall referendum engen-dered vast political consequences. First, by shifting the balance of power to hisfavor, his victory apparently put an end to the regime’s instability and temperedthe high degree of polarization that beset the country. Second, the Demo-cratic Coordinator, which claimed that Chávez had committed fraud, broke upimmediately after its defeat.23 Several parties like AD, Justice First, and RadicalCause withdrew from the opposition’s umbrella. The business federation did thesame and acknowledged the new situation by signaling their willingness to seekagreements with the government. Third, to further consolidate his grip onpower, Chávez decided to step up the “Bolivarian Revolution” through variousinitiatives. A law passed in April 2004 permitted Chávez to gain total controlover the Supreme Court by expanding its members from twenty to thirty-twojustices. Moreover, a so-called social responsibility law, approved in January2005, gave him the tools to regulate the media and to restrict freedom of

154 Mayorga

expression. Finally, the collapse of the opposition in the recall referendum set thestage for another Chávez tour de force in the state and municipal elections of Oc-tober 2004, in which he won 20 of the 23 states—and 193 out of 332 munici-palities, among them the most important—while the opposition could only re-tain the oil-rich state of Zulia.

Chávez accomplished his objective to secure an almost total control over thestate with great success, mainly by establishing through electoral and mass-demagogic means a populist, plebiscitary democracy that subverted the under-pinnings of constitutional democracy. But he was also successful in taking fulladvantage of the opposition’s shortcomings and mistakes. Arguably the mostpowerful president in Venezuela’s history, Chávez is determined to hold on topower until 2030 in order to create an endogenous socialist model, which isnow his proclaimed goal. Consequently, he stepped up the “Bolivarian Revolu-tion” by expropriating several agro-industrial farms and by establishing “socialproduction companies” and worker’s co-management schemes (“Cowing thePrivate Sector” 2005). Moreover, he intends a new change of the constitutionthat would allow him an unlimited reelection as president. Given his far-reaching control of the state and backed by windfall profits from oil exports, thelong-term viability of his regime therefore looks brighter than ever. Yet it seemscontingent upon three key factors: maintaining the support of his constituencyby delivering on his promises, that is, by effectively reducing poverty and im-proving the livelihood of the poor; sustaining the welfare programs, which, inturn, hinge on a high and steady oil revenue; and last but not least, diversifyingthe economy to alleviate the dependence on oil exports.

The Waning of Neopopulist Parties and the Politicization of Indigenous Movements in Bolivia

Neopopulist outsiders in Bolivia constituted a different case because they didnot become a threat to the democratic system.24 At the end of the 1980s, twoneopopulist parties emerged. One was CONDEPA (Conciencia de Patria, orConscience of the Fatherland), founded by Carlos Palenque, the owner of TVand radio networks in La Paz, and the other was UCS (Unidad Cívica Solidari-dad, or Civic Solidarity Unity), a political movement built by Max Fernández,then the most powerful stockholder of the country’s most important brewery.Both largely represented and channeled the demands of informal and marginal-ized sectors of the population. CONDEPA’s constituency was the rural and mi-grant population of the department of La Paz, a population that was affected by adjustment policies and unrepresented by the established parties. Both CONDEPA and UCS, which had support from urban popular sectors, wereneopopulist parties with authoritarian leaders who developed an anti-politicaldiscourse against traditional elites and parties. They blended a personalistic,


clientelistic style of politics with—especially in the case of the UCS—a pleb-iscitary appeal to the masses and a commitment to market-oriented policies.Thus, the upsurge of neopopulist parties occurred in the contradictory contextof reinforcement of democratic institutions and a relative delegitimization ofgoverning parties arising from the negative social impact of adjustment policies.

Despite their initial strong anti-systemic bias, these parties did not underminethe legitimacy of the democratic system. On the contrary, they became pre-dominantly systemic parties that played a significant institutional role by inte-grating their constituencies, participating in the management of a few importantmunicipalities, and forging interparty agreements aimed at institutional reforms.Most importantly, they became coalition partners in the governments led byMNR and ADN (Acción Democrática Nacionalista). They did not achieve pres-idential power, but their impact on the party system was important for a decade.UCS won a notable percentage of seats in the 1993 and 1997 elections (15.4and 14.6 percent, respectively). It became a minor coalition partner in the gov-ernments of Sánchez de Lozada (1993–97) and Banzer (1997–2001), holdingthe Ministries of Sustainable Development and Labor. CONDEPA also won asignificant share of seats in the 1989, 1993, and 1997 elections, and was a coali-tion partner in Banzer’s government for one year. Thus, although these partiesemerged outside of and against the established party system, they soon becameincorporated through the integrative capacities of the moderate multiparty sys-tem. Nevertheless, both parties have also been “dual” parties: on the one hand,given the neopopulist, plebiscitary bias of their democratic ideology, they havenot been fully committed to democratic institutions; on the other, they haveparticipated in electoral processes and became relevant political actors. Neo-populist elements affected their discourse, yet they aimed at political integration.

The reasons for this political dualism were threefold. First, after 1985 a processof institutionalization strengthened the party system and transformed it into amoderate one. Party fragmentation and polarization were reduced, and a patternof consensual politics superseded traditional confrontational politics. Second, inthe framework of the prevailing constitution, consensual politics became thedriving force for crafting coalition governments, which became the bedrock ofBolivia’s democratic system. Such a system provided strong incentives for coop-eration among parties, so that even small parties could participate in buildingcoalition governments. Third, the crisis of populism and of the state-led econ-omy leading to the failure of the first democratic government between 1982 and1985 was overcome through successful structural adjustment policies, which formore than a decade and a half legitimated the democratic system (Mayorga 1995).

The death of their leaders and, more importantly, their inefficient and corruptparticipation in state administration weakened these personalistic political move-ments greatly. CONDEPA suffered a catastrophic defeat in the 1999 municipalelections and lost its stronghold in La Paz. This failure was the initial step toward

156 Mayorga

the destruction of the party in the June 2002 general election, when CONDEPAobtained only 0.36 percent of the vote. UCS also experienced a substantial loss ofpolitical power, gaining just 5.5 percent of the vote and five seats in parliament inthe same election (Table 5.1). This party also seems doomed to disappear.

In Bolivia, neopopulist parties were powerful actors, but as political organi-zations they ended up being an ephemeral phenomenon that failed to alter themain features of a surprisingly stable party and government system. Neverthe-less, due to the persistent problems of poverty and social exclusion, the poten-tial source for neopopulist and anti-systemic actors has remained. The anti-systemic pressures coming from neopopulist parties that emerged in the 1980shave withered away. At the same time, in recent years the widening ofdemocracy in a context of economic depression and deepening social conflicthas resulted, paradoxically, in a new polarization of the political system, stem-ming mainly from the politicization of indigenous social movements. Thetaming and ultimately the demise of neopopulist parties left a vacuum thatpeasant and indigenous movements have filled, while at the same time consti-tuting a different political trend and cleavage.25 This new paradox calls for ex-planation: while Bolivian democracy developed the capacity to include previ-ously excluded social groups politically, the dynamics of the inclusion ofindigenous movements spawned contradictions and tendencies that have putrepresentative democracy in jeopardy. New ethnic-political cleavages andshort-term problems have led to the politicization of indigenous social move-ments that evince some neopopulist tendencies.

What are the factors explaining the politicization of the Chapare and North-ern Altiplano indigenous movements, the most important to emerge since thedemocratic transition in 1982? Structural factors provide important background.

Table 5.1

Bolivia: Party Votes and Seats, Lower-Chamber Elections, June 30, 2002

Number oflower-chamber

Party % of votes % of seats seats

MNR 22.46 29.93 47MAS 20.94 22.29 35NFR 20.91 17.19 27MIR 16.32 19.74 31MIP 6.09 3.82 6UCS 5.51 3.18 5AND 3.40 3.18 5PS 0.65 0.63 1LJ 2.72 — —MCC 0.63 — —CONDEPA 0.37 — —

SOURCE: Corte Nacional Electoral (National Electoral Court).


Bolivia is going through an intense period of politicization of economic andsocial conflicts, mainly due to economic decline, to the incapacity of the state totackle problems, and to the party system’s lingering inability to mediate and chan-nel social interests and demands. This latter aspect is critical because it is linked,paradoxically, with an institutional reform—the 1994 Law of Popular Participa-tion. The law created new political opportunities for social movements, helpingto strengthen participation in local politics and thereby differentiate patterns ofaction and interest intermediation.26 The law’s implementation fostered the ero-sion of traditional linkages between parties and social organizations, and partic-ularly linkages with indigenous movements. Moreover, the law failed to relieveeconomic and social conflicts or to alleviate the stagnation of economic devel-opment in rural areas and the worsening of living conditions. It also exacerbatedethnic cleavages instead of strengthening a national identity and consciousness.

Short-term factors also lie at the root of the politicization of indigenous move-ments. The poor and marginalized sectors’ growing unrest and disenchantmentwith the performance of the economy and with the social consequences of lim-ited government policies have put serious strains on the democratic system, aftermore than a decade of institution building with insufficient economic growthand poverty reduction. During the fragile Banzer government, political mis-management, insufficient economic growth, and social conflicts intensified thestruggle of contentious social movements. Like other Latin American countries,since 1999 Bolivia has experienced deep economic stagnation and deteriorationof living conditions. Real GDP grew only 0.4 percent in 1999 and only about2 percent between 2000 and 2002, while informal self-employment amounted toabout 65 percent of the labor force. Under pressure from entrenched U.S. inter-ests, Banzer and Quiroga’s government (1997–2002) aggravated economic andsocial conflicts by sticking to a radical policy of coca eradication and the prohi-bition of coca trade in the Chapare without taking into account socioeconomicand political costs. This policy—which was not backed by a comprehensive al-ternative agrarian development program—adversely affected the economy, pro-voking a GDP decrease of about 5 percent. It had the concomitant political ef-fect of sparking the mobilization and radicalization of the coca growers’ unions,led by Evo Morales. The conflict spread to include other contentious groups suchas labor and teachers’ unions and, in particular, informal-sector organizations,which expressed their discontent not only with specific policies but with thepost-1985 economic model.

The immediate background for the social and political crisis in April andSeptember 2000 was the massive rejection of arbitrary increases in the water tar-iff in Cochabamba, where an international company had signed a contract withthe municipality and the central government to provide water and energy tothe city. Mass mobilizations and road blockages erupted in the so-called “waterwar” organized by peasant unions, teachers, labor unions, and informal-sector

158 Mayorga

organizations. Paralyzing key regions of the country, they issued a vast array ofover a hundred demands that were basically economic in character, including thecancellation of the water contract in Cochabamba, salary increases, an end to thecoca eradication policy, repeal of the agrarian reform law, and abolition of the pri-vatization of state enterprises. This social crisis was a watershed for the politicalsystem. It revealed both a great accumulation of social conflicts and a multiplic-ity of old and new actors that overwhelmed the country’s established political in-stitutions. The crisis also demonstrated a great potential for mobilization againsta weak national government and fragile regional state structures. Above all, itreflected an increasing gap between political parties and social movements, duenot mainly to a dramatic loss of political representation but to the governingparties’ inability to channel conflicts and to carry out effective public policies(Mayorga 2005).

Social and economic tensions began to erode political stability as never beforeby turning into political conflicts. The most far-reaching result of the 2000 so-cial and political crisis was to extend and enhance the politicization of indige-nous movements—a process that the Law of Popular Participation and the 1995and 1999 municipal elections had already fostered. The Law of Popular Partici-pation provided incentives and opportunities for the political inclusion of locallybased social organizations and leaders into municipal governments. The mixed-member electoral system established in 1994 helped bolster a locality-centered,constituency-serving political representation (Mayorga 2001). In the 1995 and1999 municipal elections, the Chapare peasant unions running on the IU andMAS tickets had already won a significant number of rural municipalities. In1995, their candidates obtained 3.7 percent of all seats (60 seats total) in munic-ipal councils, although not all belonged to the Chapare social movement. In the1999 municipal elections, the MAS was the only indigenous party participating,and it obtained 4.7 percent of all seats (80 seats total) in municipal councils.27

From 1995 onward, the Chapare peasant unions became the driving force in thelocal municipal arena.

This political advance at the local level extended to the national level with the1997 presidential election. Morales and three leaders of the coca growers’ unionssuccessfully ran on the IU ticket, gaining four seats in the Chamber of Deputies.While the government expected to cripple peasant coca unions by cracking downon the coca economy, the peasant unions had already changed the thrust of theirresistance into a political struggle and institutional presence. Relying upon thecoca unions as key networks and organizational resources, they achieved a highdegree of politicization by linking their mobilizations against the U.S.-backederadication policy with the struggle for national autonomy. They also interwovetheir defense of coca growing with the principles and values of Indian culturalidentity (Albó 2003).28 The indigenous movements became identity-based polit-ical movements. They framed their political action in a double strategy favoring


the preservation of the coca economy and the restoration of community tradi-tions and rights, claiming jurisdictional control over indigenous territories. Inthis way, persistent historical ethnic cleavages were successfully transformed intopolitical issues. They focused not only on ethnic political representation but on aradical restructuring of the state, a task that small, short-lived Indian parties hadundertaken in vain at the outset of the democratic transition. Thus, the Indianmovements developed a political struggle whose logic of action can be definedas “the ethnic demarcation of limits” (Eder 2001, 202). According to this classi-fication, social and political conflicts are constructed as ethnic-cultural conflictsof identity, so that existing economic and social interests turn out to be mediatedby identity conflicts. In this way they are less negotiable, at least for the leaders ofthese movements.

The extent of the hitherto unprecedented politicization of indigenous move-ments became apparent in the June 2002 general election, which triggered themost far-reaching political consequences since the emergence of a moderatemultiparty system and the collapse of the workers’ movement. The two indige-nous movements strengthened in the aftermath of the 2000 crisis and establishedthemselves as political movements or parties by participating in these elections.Serious political mistakes by Congress and Quiroga’s government bolstered theMAS, which obtained a stunning triumph with 20.94 percent of the vote andthirty-five seats in the lower chamber.29 According to the constitutional provi-sion that established that Congress elects the president from among the two can-didates with the largest number of votes if no candidate wins an outright ma-jority of the electorate, Morales, as the leader of MAS, the party receiving thesecond-highest number of votes, was entitled to participate as a presidential can-didate in the decisive congressional arena. He lost to Sánchez de Lozada, theMNR candidate, who built a majority coalition in order to be elected president.Since Morales already had political experience and had served as a deputy in thelegislature between 1997 and 2002, he cannot be considered an outsider. More-over, he did not run in the 2002 election as an independent or with a new partylabel. The other indigenous movement participating was the MIP (MovimientoIndígena Pachakuti, or Pachakuti Indigenous Movement), a party founded inNovember 2001 by Aymara peasant leader Felipe Quispe. As executive secre-tary of the national peasant confederation, Quispe was a political outsider whoalso became a powerful leader by building a stronghold in the Aymara commu-nities of some provinces in La Paz. His party obtained 6.09 percent of the voteand six lower-chamber seats (Table 5.1). The noteworthy electoral performanceof both parties reflected the dissatisfaction and distrust of rural and poor urbanvoters in the country’s Andean region vis-à-vis the established parties.

The second consequence of major political significance was the widening ofpolitical inclusion and participation. For the first time in Bolivia’s democratic his-tory, the June 2002 elections permitted the political inclusion and autonomous

160 Mayorga

political representation of indigenous social movements, thereby demonstratingthe adaptability of the political system (Mayorga 2001, 2002). Because both In-dian parties combined obtained 27.03 percent of the vote and 26 percent of theseats in Congress, the elections brought about a historical shift in political repre-sentation. They qualitatively advanced ethnic representation and confirmed themutation of these indigenous movements into strong political actors. But do theyhave the capacity to transform themselves into stable political movements and,above all, into political parties? Given their meteoric rise and growth, the MASand the MIP are more social and political movements than structured parties. In-deed, they are mainly social protest movements that articulate both the utopianand substantive demands of peasant sectors, which now face the challenge ofbuilding political parties in order to extend their social and political base. In thisregard, the key question is whether these movements will be able to consolidatethemselves as strictly ethnic-based parties or whether they will manage to de-velop as parties with a broad national constituency and political program. SinceBolivia is currently undergoing economic and urban changes that erode thestructural, social, and economic foundations for the development of ethnic-indigenous political parties, my hypothesis is that indigenous movements cannotbecome national political parties if they remain identity-based and bound bytheir Indian social constituencies.30 First, identity-based indigenous movementsdo not appeal to broad social groups in the increasingly mestizo and culturally di-verse Bolivian society; and secondly, these movements are constrained by partic-ularistic, ethnic-corporatist issues and have been unable hitherto to develop po-litical programs and strategies involving relevant national issues. The MAS hastaken on this challenge by championing the nationalization of natural resources,mainly of the huge natural gas resources, and an ethnic-based constituent assem-bly aimed at a radical restructuring of the state. This strategy, based predomi-nantly on direct forms of contention, has apparently strengthened electoral sup-port for the MAS in rural and urban indigenous sectors in the country’s westernregion. It has not, however, significantly broadened the party’s constituency toeither middle-class sectors or to the eastern lowlands that claim regional auton-omy and are hostile to the indigenous movement.31

The third consequence of indigenous politicization has been renewed polar-ization of the party system, demonstrating the extent to which ethnic-culturalcleavages were politicized. Since 1985, Bolivia has had a moderate multiparty sys-tem with three relevant parties at its axis: MNR, MIR (Movimiento de IzquierdaRevolucionario, or Revolutionary Left Movement), and ADN. Yet the indige-nous movements’ conquest of parliamentary power has polarized the party sys-tem. They are hostile to representative democracy and to the market economy,advocating instead a utopian model of ethnic identity-based, participatory de-mocracy and the return to a state-led economy. Both the MAS and the MIP putforward a political program with a strong anti-systemic bias drawing on an


ethnicist and fundamentalist ideology and upholding the utopian idea that theayllu—the traditional indigenous community—should serve as the mainstay ofa profound reorganization of the state and society. These political movements andthe established parties have had deep normative and political disagreements ondemocratic principles and rules of the game. The MAS rejected outright the ba-sic tenets of representative democracy and the market economy on the groundsthat they are alien to Indian cultures. Accordingly, it attempted a radical, stronglyanti-institutional strategy, dubbed “siege strategy,” aimed at blocking and de-stabilizing the government and the state by using both the tactics of mobilizationand its veto power against government initiatives in Congress, which require atwo-thirds majority. After the overthrow of Sánchez de Lozada’s government byan urban indigenous uprising in El Alto that claimed the right to nationalize gasresources, the MAS veered from confrontational politics to an electoral strategyand to supporting Mesa’s government, which assumed the radical goals of theMAS and the indigenous movement. The main reason for this change seemed tobe that Morales aims at transforming his political movement into a national partycapable of contending for state power.

Because of the polarization of the party system and the rise of a radical oppo-sition with veto power, three crucial tensions have arisen that affect the stabilityand future prospects of the democratic system. First, since the MAS and the MIPare basically social movements, their political practices respond to the logic of so-cial protest and contention, applying anti-institutional tools of pressure on po-litical institutions. Both parties are, in fact, extra-parliamentary movements thatdo not differentiate social and political styles of action and instead subordinatethe logic of politics to the logic of social protest movements. Consequently, tothe extent that indigenous movements entered the political system conceiving ofthemselves as anti-systemic social movements, the MAS and the MIP are dualpolitical movements.32 As a political movement, the MAS faces the dilemma ofcontinuing a politics of confrontational opposition to the democratic system orshifting to the role of a responsible opposition. To do the latter, it must turn intoan institutionalized party that abides by the rules of the game and acts as aninstitutional catalyst of social and political change. With its weak leadership andlocally restricted social base, the MIP does not appear to have the capacity totransform itself into an organized party. As a strong movement, the MAS, on theother hand, could probably rise to the challenge of developing national politicalstructures, thereby overcoming its origins as a network of peasant unions.

Second, the fundamentalist ideology of the peasant movements gives rise tocontradictions with their own pragmatic and concrete demands. The identity-oriented framing of the coca peasants’ struggles, for example, distorts the inter-ests of broad sectors of the rural population who are more interested in eco-nomic integration as a means to improve their living conditions than in autopian revival of the pre-Columbian past. Conflicting ideological principles

162 Mayorga

and strategic guidelines that try to blend issue mobilizations with identity mo-bilizations lie at the root of the MAS’s political action. Third, the paradoxicalconfluence of political inclusion of indigenous movements and polarization ofthe party system has stirred up contradictory perspectives. The political inclu-sion coming out of the June 2002 election turned out to be a serious threat tothe democratic system, as polarization and existing political strains between thetraditional parties and the indigenous movements undermined democratic sta-bility, governmental capacity, and state unity. Yet the current situation can alsobe seen as a historic opportunity to enhance the quality of democracy by de-veloping an integrated, pluralistic, multicultural, and multiethnic democracy.The key challenges facing Bolivian democracy, therefore, are overcoming po-larization, catastrophic stalemate, and state crisis, on the one hand, and achiev-ing full political inclusion of indigenous political movements, on the other—atask that will be more complex and demanding than the successful integrationof neopopulist parties at the beginning of the 1990s.

Conclusion

Stressing common patterns and qualitative differences in Peru, Venezuela, andBolivia, this chapter set out to explain from a comparative perspective the emer-gence of neopopulism and anti-system actors as an outcome of two main pro-cesses: the decomposition of party systems, and a deep crisis of the state—in fact,a crisis of governability. The key causal factor for the breakdown of political par-ties was this crisis and not a crisis of political representation, which in any case wasa result of the former. This line of reasoning agrees with the conclusion thatMainwaring draws in the final chapter of this book: namely, that at the core of thecontemporary crisis of representation in the Andes there is a crisis of democraticgovernability, associated with grave deficiencies in state capacity. Political partiesand leadership in Peru and Venezuela put the sustainability of democracy in jeop-ardy, becoming obstacles and problems as a consequence of their failure as gov-ernmental actors. A context of poor state performance, socioeconomic crisis, andpolitical party decline brought about favorable conditions for the rise of outsiders.By virtue of the failure of democratic governability and the traditional parties’ lossof credibility, outsiders sprang up, claiming to be the only way out of the crisis.

Outsiders’ politics in Peru and Venezuela had destructive outcomes. The mostnegative consequences for the democratic system were the concentration ofpower in the hands of high-handed leaders and the erosion of democratic insti-tutions, leading to the breakdown of liberal-representative democracy and to itstransformation into plebiscitary, semi-democratic regimes. By contrast, in Bo-livia a moderate centripetal party system and a coalition-based government sys-tem absorbed neopopulist outsiders. Due to this political integration, they didnot become a threat to democracy. Yet anti-systemic threats have sprung morerecently from ethnic-fundamentalist and populist indigenous movements that


became powerful veto players and achieved—in the midst of an economicrecession and mounting social conflict—a vital political space in Congress. Thisfact demonstrated the political system’s capacity to respond positively to thesweeping politicization of these movements. Nevertheless, the confluence of thestrengthening of ethnic-political representation, the crumbling of the moderateparty system, and the emergence of an ethnicist populist alternative has prompteda dangerous political destabilization, raising another historical challenge for Bo-livian democracy. Again at stake is its capacity to reestablish democratic govern-ability by enhancing the political integration of indigenous movements and at thesame time fostering an efficient state, capable of meeting the urgent needs ofthe poor and of advancing integral citizenship. Whether Bolivian democracy willbe able to tackle this weighty task will to a great extent depend on overcomingpolarization and stalemate, which in turn hinges on rebuilding political leader-ship and the party system as well as on restoring consensus-based politics.

Notes

I am very grateful to Scott Mainwaring, Eduardo Pizarro Leongómez, Richard Snyder,and James Mahoney for their insightful comments on previous versions of this chapter.

1. Ecuador also experienced an outsider’s rise to power with the short-lived govern-ment of Colonel Lucio Gutiérrez from January 2003 until his downfall in April 2005.

2. Stokes (2001, 142– 48) also refers to reasons for social spending in support of the poor.

3. The political programs of Palenque and Fernandez in Bolivia, which proposed torestore state capitalism, were similar cases.

4. Chávez’s difficulties in carrying out his statist program seem to indicate that clas-sical populism encounters structural limits and is therefore not viable. In the end, re-gardless of his ideological objectives, Chávez’s populism seems doomed to resemble Fuji-mori’s “petty cash” populism. I owe this observation to Eduardo Pizarro Leongómez.

5. By governability I mean the capacity of government to tackle fundamental prob-lems of society through effective decision making and public policies.

6. I draw upon the idea that “defining representation as acting in the interest of therepresented, provides a minimal core conception” (Przeworski et al. 1999, 2).

7. Tanaka highlights the critical dimension of governability, but he explains the crisisof Peruvian political parties mainly as a result of flawed short-term political decisions(Tanaka 1998, 71–85).

8. As Weber stressed, charismatic leaders do not feel bound by institutional rules andconstraints and instead demand faith and strict adherence from their followers (Weber1964, 834).

9. These reflections are based on Linz’s (1978) ideas about the breakdown of democracy.

10. Based on this support and the contacts Fujimori had within the APRA, Planas(2000, 295–301) contends that Fujimori was neither an independent candidate nor anoutsider.

164 Mayorga

11. Some analysts characterize Fujimori’s regime as a delegative democracy. Otherterms used to define these regimes have been soft dictatorship, Caesarist democracy, andneopopulist democracy.

12. This closely fits Linz’s definition of an authoritarian regime: “authoritarianismwith limited, not responsible, political pluralism, without elaborate and guiding ideol-ogy, without extensive nor intensive structured social support and political mobilizationin which a leader or occasionally a small group exercise power within formally ill-definedlimits but actually quite predictable ones” (Linz 2000, 159).

13. The Peruvian experience offers stark evidence of the indispensability of parties asmechanisms of representation (Levitsky and Cameron 2003, 27).

14. Translated as “takeover of pigeonholes,” this phrase describes the gradual controlachieved by Fujimori over the hard cores of military and economic power.

15. See “Venezuela’s Crisis: Towards the Endgame,” The Economist, April 13–19,2002.

16. According to a survey conducted by Arturo Keller in May 2001, only 17 percentof respondents supported Chávez’s strategy of civilian-military government; 9 percentsupported an exclusively military government, and 68 percent supported civilian gov-ernment (El Universal, May 6, 2001).

17. This profound polarization was the backdrop for the coup d’état of April 11,2002. In May 2001, Chávez’s politics had already led to “saber rattling” within the armed forces and to the possibility that military adversaries might be influenced by the radical pro-coup discourse of some political actors and leaders of the old “Punto Fijo”democracy.

18. In the OAS meeting in Quebec in May 2001, Chávez disagreed with the “Democratic Charter” and did not endorse the basic principle of representative democ-racy. He ratified this on his visit to Russia in May 2001, asserting that he believed in de-mocracy, “but not in the forms of democracy imposed on us.” He also declared on histrip to various Asian countries: “I am the second Latin American Castro.”

19. Before his rise to power, Chávez declared in an interview that he nurtured theidea of a “new-style militarism,” which in his view had almost been established in Peruin 1968–75 (Blanco Muñoz 1998, 73).

20. As General Medina stated: “The vast majority realize that Chávez has taken hismandate for social change and used it for a revolution that takes the country down a roadit doesn’t want to go” (“A Tragic and Dangerous Stalemate,” The Economist, October 12,2002).

21. At the beginning of his mandate, polls indicated that Chávez had the support of89 percent of the country. Three years later this support decreased to about one third(survey of Data Análisis 2002).

22. Moises Naim stresses the opposition’s miscalculation that Chávez could not lastand its lack of a long-term strategy (Latin American Regional Report: Andean Group, Sep-tember 7, 2004, 2).

23. Fraud accusations could not be substantiated. (see, e.g., Jennifer McCoy, “WhatReally Happened in Venezuela?” The Economist, September 2, 2004).

24. Linz (1994, 29) assessed Fernández as a potential threat to democracy.25. NFR (New Republican Force), a neopopulist party with regional roots in

Cochabamba harking back to traditional populism, emerged in 1995 and participated inthe June 2002 election, obtaining 20.91 percent of the vote and twenty-seven seats in Congress.

26. Contentious politics is produced when political opportunities broaden (Tarrow2002, 23).


27. Strangely enough, the MAS was the result of a leftist split from FSB, a rightistparty, in 1988. Since his own party organization was not recognized by the NationalElectoral Court, Morales “rented” this label for the 1999 election. Thus, the MAS became the political umbrella for the six Chapare unions.

28. Morales became a political leader in the context of this struggle. He can be con-sidered the offspring of U.S. policies and Banzer’s government.

29. Morales’s electoral stance got a decisive push from the U.S. ambassador, ManuelRocha, when he declared shortly before election day that Bolivia would risk the suspen-sion of U.S. aid if its citizens voted for Morales.

30. Levitsky and Cameron (2003, 17) are skeptical about the chances for party-building in the Andean region, pointing out the persistence of exclusion from citizenrights and the enduring legacy of colonialism.

31. The MAS obtained the highest percentage of votes (18.4 percent) in the munic-ipal elections in December 2004. According to a survey done in October 2005 on voteintention in the presidential elections scheduled for December 2005, Morales would get28 percent and ex-president Quiroga 22 percent (La Razón, October 7, 2005).

32. There are analytical differences between a social movement, a political movement,and a political party. As Kitschelt contends, “Collective organizations have distinctiveprofiles of political practices (or action repertoires according to Tilly) which allow us ina given moment to associate them more with one of these ideal types—social move-ments, interest groups, parties” (Kitschelt 2001, 356).

References

Albó, Xavier. 2003. Pueblos Indios en la Política. La Paz: Plural.Ayo, Diego. 2000. Los Desafíos de la Participación Popular. La Paz: CEBEM.Blanco Muñoz, Agustín. 1998. Habla el Comandante. Caracas: Fundación Pío Tamayo,

Universidad Central de Venezuela.Comisión de la Verdad y Reconciliación. 2003. Informe Final.Coppedge, Michael. 1994. Strong Parties and Lame Ducks: Presidential Partyarchy and Fac-

tionalism in Venezuela. Stanford, CA: Stanford University Press.———. 2002. “Venezuela: Popular Sovereignty versus Liberal Democracy.” Working

Paper, no. 294, Helen Kellogg Institute for International Studies, University of NotreDame, Notre Dame, IN.

Cotler, Julio. 2000. “La gobernabilidad en el Perú: Entre el autoritarismo y la democra-cia.” In El fujimorismo: Ascenso y caída de un régimen autoritario, ed. Julio Cotler andRomeo Grompone, 311–36. Lima: IEP.

“Cowing the Private Sector.” 2005. The Economist, September 15.Eder, Klaus. 2001. “De los intereses a la identidad y de ésta de vuelta a los intereses: Los

límites de la movimientización moderna.” In Construcción de Europa: Democracia y glob-alización, vol. 1, ed. Ramón Mainz, 197–218. Santiago de Compostela: Universidadde Santiago de Compostela.

Ferrero Costa, Eduardo. 1993. “Peru’s Presidential Coup.” Journal of Democracy 1 ( January).

Gómez Calcaño, Luis. 2002. “Los círculos bolivarianos: El mito de la unidad delpueblo.” Unpublished paper, CENDES, Caracas.

166 Mayorga

Gómez Calcaño, Luis, and Nelly Arenas. 2001. “Modernización: Autoritaria o actual-ización del populismo? La transición política en Venezuela.” In Venezuela: Rupturas yContinuidades del Sistema Político (1999–2001), ed. Marisa Ramos, 37– 68. Salamanca:Ediciones Universidad de Salamanca.

González de Olarte, Efrain. 1998. El neoliberalismo a la peruana: Economía política del ajusteestructural. Lima: IEP-CIES.

Grompone, Romero. 2000. “Al Día Siguiente: El fujimorismo como proyecto incon-cluso de transformación política y social.” In El fujimorismo: Ascenso y caída de un régi-men Autoritario, ed. Julio Cotler and Romeo Grompone, 77–178. Lima: IEP.

Kitschelt, Herbert. 2001. “Panoramas de intermediación de intereses políticos:Movimientos sociales, grupos de interés y partidos políticos a comienzos del sigloXXI.” In Construcción de Europa: Democracia y globalización, vol. 1, ed. Ramón Maiz,361–86. Santiago de Compostela: Universidad de Santiago de Compostela.

Knight, Alan. 1998. “Populism and Neopopulism in Latin America.” Journal of LatinAmerican Studies 30, no. 2 (May): 223– 48.

Kornblith, Miriam. 1998. Venezuela en los 90: La crisis de la democracia. Caracas: IESA.Levitsky, Steven. 1999. “Fujimori and Post-Party Politics in Peru.” Journal of Democracy

10, no. 3 ( July): 78–92.Levistky, Steven, and Maxwell Cameron. 2003. “Democracy without Parties? Political

Parties and Regime Change in Fujimori’s Peru.” Latin American Politics and Society 45,no. 3 (Fall): 1–33.

Linz, Juan J. 1978. Crisis, Breakdown, and Reequilibration. Baltimore: Johns Hopkins Uni-versity Press.

———. 1994. “Presidential or Parliamentary Democracy: Does It Make a Difference?”In The Failure of Presidential Democracy, ed. Juan Linz and Arturo Valenzuela, 3–87.Baltimore: Johns Hopkins University Press.

———. 2000. Totalitarian and Authoritarian Regimes. Boulder, CO: Lynne Rienner.Lynch, Nicolás. 1999. Una tragedia sin héroes: La derrota de los partidos y el origen de los in-

dependientes. Perú, 1980 –1992. Lima: Universidad Nacional Mayor de San Marcos.Mainwaring, Scott. 1999. “The Surprising Resilience of Elected Governments.” Journal

of Democracy 10, no. 3: 101–14.Manin, Bernard. 1997. The Principles of Representative Government. Cambridge: Cam-

bridge University Press.Manrique, Miguel. 2001. “La participación política de las fuerzas armadas venezolanas

en el sistema político, 1998–2001.” In Las fuerzas armadas en la región andina: ¿No de-liberantes o actores andinos? ed. Martín Tanaka, 305–36. Lima: Comisión Andina de Juristas.

Martucelli, Danilo, and Maristella Svampa. 1992. “La doble legitimidad del populismo.”Pretextos (October–December): 63–79.

Mayorga, René Antonio. 1995. Antipolítica y neopopulismo. La Paz: CEBEM.———. 2001. “The Mixed-Member Proportional System and Its Consequences in

Bolivia.” In Mixed-Member Electoral Systems: The Best of Both Worlds? ed. MatthewShugart and Martin Wattenberg, 194 –208. Oxford: Oxford University Press.

———. 2002. “La metamorfosis del sistema de partidos.” Opiniones y Análisis: El Esce-nario Post-electoral en Bolivia 60 (October–December): 67–112.

———. 2005. “Bolivia’s Democracy at the Crossroads.” In The Third Wave of Democra-tization in Latin America, ed. Scott Mainwaring and Frances Hagopian. Cambridge:Cambridge University Press.

McClintock, Cynthia. 1989. “Peru: Precarious Regimes, Authoritarian and Demo-cratic.” In Democracy in Developing Countries: Latin America, ed. Larry Diamond, JuanLinz, and Seymour Lipset, 335–86. Boulder, CO: Lynne Rienner.


Molina, José Enrique. 2000. “Comportamiento electoral en Venezuela, 1998–2000:Cambio y continuidad.” Cuestiones Políticas 25: 27– 66.

———. 2003. “La revolución bolivariana en Venezuela: ¿Socialismo autoritario en unmar de contradicciones?” Paper presented at the Latin American Studies AssociationCongress, Dallas, March 27–29.

Molina, José Enrique, and Carmen Pérez. 1994. “Venezuela: ¿Un nuevo sistema de par-tidos.” Cuestiones Políticas 13: 63–90.

Naim, Moises. 2001. “The Real Story Behind Venezuela’s Woes.” Journal of Democracy 12,no. 3 (April): 17–31.

Novaro, Marcos. 1996. “Los populismos latinoamericanos transfigurados.” Nueva Sociedad 144: 90 –103.

Obando, Enrique. 2001. “Las relaciones cívico-militares en el Perú en la década del 90:Lecciones para el futuro.” In Las fuerzas armadas en la región andina: ¿No deliberantes oactores políticos? ed. Martin Tanaka, 247–304. Lima: Comisión Andina de Juristas.

O’Donnell, Guillermo. 1994. “Delegative Democracy?” Journal of Democracy 5, no. 1:55– 69.

Pécaut, Daniel. 1987. L’ordre et la violence. Paris: EHESS.Planas, Pedro. 2000. La democracia volátil. Lima: Fundación Ebert.Przeworski, Adam, Susan Stokes, and Bernard Manin, eds. 1999. Democracy, Accountabil-

ity, and Representation. Cambridge: Cambridge University Press.Rey, Juan Carlos. 1991. “La democracia venezolana y la crisis del sistema populista de

conciliación.” Revista de Estudios Políticos 74: 533–78.Reyna, Carlos. 2000. “Un régimen de espíritu militar.” In Perú 2000: Un triunfo sin democ-

racia, 133–59. Lima: Comisión Andina de Juristas.Roberts, Kenneth M. 1995. “Neo-liberalism and the Transformation of Populism in

Latin America.” World Politics 48, no. 1 (October): 82–116.Rospigliosi, Fernando. 1995. “La amenaza de la ‘fujimorización’: Gobernabilidad en

condiciones adversas, Perú y los países andinos.” In Partidos y clase política en américalatina en los 90, ed. Sonia Picado and Daniel Zovatto, 311–36. San José: IIDH-CAPEL.

Sartori, Giovanni. 1999. Elementos de teoría política. Madrid: Alianza Editorial.Schedler, Andreas. 1994. “Antipolitical Opposition: A Framework for Comparative

Analysis.” Paper presented at the Vienna Dialogue for Democracy on “The Politics ofAntipolitics,” Institute for Advanced Studies, Vienna, July 7–10, 1994.

Stokes, Susan. 2001. Mandates and Democracy: Neoliberalism by Surprise in Latin America.Cambridge: Cambridge University Press.

Tanaka, Martín. 1998. Los espejismos de la democracia: El colapso del sistema de partidos en elPerú. Lima: IEP.

Tarrow, Sidney. 2002. Power in Movement: Social Movements and Contentious Politics. Cambridge: Cambridge University Press.

Torre, Carlos de la. 2000. Populist Seduction in Latin America. Athens: Ohio State University.

Trinkunas, Harold. 2002. “The Crisis in Venezuelan Civil-Military Relations: FromPunto Fijo to the Fifth Republic.” Latin American Research Review 37, no.1 ( January):41–76.

Weber, Max. 1964. Wirtschaft und Gesellschaft. Vol. 2. Köln: Kiepenhauer and Witsch.Weyland, Kurt. 1996. “Neopopulism and Neo-Liberalism in Latin America: Unexpected

Affinities.” Studies in Comparative International Development 31 (Fall): 3 –31.———. 2001. “Clarifying a Contested Concept: ‘Populism’ in the Study of Latin

American Politics.” Comparative Politics 34, no. 4 (October): 1–22.

Part II

❖

DECENTRALIZATION,LEGISLATURES, AND

DEMOCRATIC REPRESENTATION

In the 1980s and 1990s, the Andean countries, along with most of the rest oftheir Latin American neighbors—and, indeed, a great deal of the developed anddeveloping world—experienced a series of decentralizing reforms. These re-forms generated a great deal of optimism, based on accumulated scholarship sug-gesting that decentralization would yield significant fiscal and political benefits.Economic theories indicated that decentralization (or fiscal federalism)1 wouldincrease efficiency (Tiebout 1956; Musgrave 1959; Rubinfeld 1987; Bird 1990;Oates 1998).2 On the political side, decentralization was linked to democraticconsolidation and improved democratic practice (Diamond 1999; Huther andShah 1998; Fox 1994; Dahl 1971).3

Nearly ten years after the most recent of the major reforms in the region, as-sessments of decentralization’s effects have delivered a mixed verdict on thesepredictions. This is particularly true on the economic side, where some studieshave found that decentralization increases the size of government (Stein 1998),impedes fiscal restraint (Alesina, Carrasquilla, and Echavarría 2002; Rodden2002; Alesina et al. 1999), or increases corruption (Treisman 1999; Tanzi 1994;but see also Fisman and Gatti 2000).

Assessments of decentralization’s effects on democracy and other politicaloutcomes have lagged well behind economic assessments.4 Instead, most politi-cal analyses have focused on the causes of decentralization (O’Neill 2005;O’Neill 2003; Garman, Haggard, and Willis 2001; Willis, Garman, and Haggard1999; Barr 2001; Grindle 2000). Those studies that do explore the democraticdividends of decentralization tend to discuss the relationship between decen-tralization and political outcomes as one of many parts of their work. In addi-tion, they tend to focus on particular cases that exhibit exceptionally good(Campbell 2003; Tendler 1997) or bad (Eisenstadt 1999) results, painting an in-complete picture of decentralization’s effects. In short, there is little focused, sys-tematic analysis of decentralization’s effects on political outcomes.

6

Decentralized Politics and PoliticalOutcomes in the Andes

❖

Kathleen O’Neill

Exploring decentralization in the Andes is particularly interesting, as the sameperiod in which this supposedly democracy-enhancing reform has blossomedhas coincided with a crisis of democratic representation. This suggests two al-ternative interpretations: either decentralization has contributed to this crisis(through the mechanisms Scott Mainwaring labels “the paradox of democraticrepresentation” in this book’s Conclusion—i.e., by opening up new avenues ofpolitical contestation through which competing parties criticize each other inorder to attract votes and, as a byproduct, fostering a pervasive sense of state cri-sis); or decentralization, though contributing positively to democratic represen-tation, has served only to dampen what would have been a far worse crisis ofdemocratic representation had decentralization not been adopted.

This chapter investigates the relationship between decentralizing reforms, po-litical parties, and political representation in Bolivia, Colombia, Ecuador, Peru,and Venezuela. A great deal of scholarship paints a rosy picture of decentraliza-tion’s likely contributions to improved democracy. Diamond and Tsalik (1999)summarize five mechanisms through which decentralization should improvedemocratic practice:

First, it helps to develop democratic values and skills among citizens. Second, it increasesaccountability and responsiveness to local interests and concerns. Third, it provides ad-ditional channels of access to power for historically marginalized groups and thus im-proves the representativeness of democracy. Fourth, it enhances checks and balances vis-à-vis power at the center. Fifth, it provides opportunities . . . for parties and factions inopposition at the center to exercise some measure of political power. (121–22)

This chapter looks at the ways in which decentralization has affected politicalrepresentation either through, or in spite of, political parties, examining many ofthe mechanisms identified by Diamond and Tsalik. In particular, I explore the ex-tent to which decentralization has affected participation and public opinion (bothrelated to the first mechanism outlined above). Next, I look at the ways politicalparties have responded to opportunities at subnational levels of government andthe different extent to which they have been successful (this touches on mecha-nisms three and five outlined above). Finally, I explore a consequence of decen-tralization not predicted above: the way in which decentralization has changedthe career paths of politicians at the national level and has, as a result, affected po-litical party organizations.5 These same three mechanisms—voter turnout, partyvolatility across elections, and the growth of outsider candidates for president—are three of the defining characteristics of a crisis of democratic representationdescribed in this book’s Introduction. To the extent that decentralization in-creases turnout, dampens party volatility across elections, or contributes to fewerindependent presidential candidacies, it might be said to enhance representation.

As this analysis will show, decentralization’s relationship to these categories is complex. I find that the relationship between decentralization and political

172 O’Neill

representation varies significantly across countries and over time within coun-tries. What is more, it varies quite significantly within countries at any particularpoint in time. Instead of uniformly improving political representation, in someareas decentralization has empowered capable, enthusiastic reformers,6 while inothers it has increased the power of landed oligarchs or, worse still, armed ene-mies of the state. In every Andean country, one can point to several subnationalunits in which political representation has been unambiguously improved by de-centralization, and to at least a handful of units that have been made unambigu-ously worse off. The effects of decentralization on political representation areexperienced locally. Given this wide range of variation (along the three axes ofcountry, subnational unit, and time), and the multiple ways that decentraliza-tion is hypothesized to affect political outcomes, this assessment can only scratchthe surface.

In this chapter, I explore variation at the level of countries and over time, leav-ing the majority of subnational variation unexamined. The chapter is organizedinto four sections. The first summarizes the region’s decentralizing reforms,highlighting the unique origins and features of each country’s particular experi-ence. The next three sections analyze the ways in which decentralization has af-fected political representation across the region, examining decentralization’s ef-fects on political representation from three vantage points. The first of theselooks at political participation through the eyes of citizens. Looking at rates ofelectoral participation in subnational contests and at scattered public opiniondata, this section finds that enthusiasm for decentralization is high in all coun-tries. The second section explores the link between decentralization and politi-cal representation from the perspective of political parties. How have traditionaland emerging parties taken advantage of the new local and regional arenas ofpower that have been created and expanded by decentralizing reforms? Here,variation exists not just at the national level but at the subnational level withincountries and party by party. In most cases, decentralization has led to a declinein the ability of traditional parties to control subnational positions. However, insome cases, the decline in traditional party support at the subnational level haslagged behind the decline in their support at the national level. In other cases,parties seem to have adapted or used institutional rules to maintain their strongshowing at subnational levels. A third section explores the consequences of de-centralization from the politician’s point of view. Here I examine the impact ofdecentralization on the career paths of presidential candidates. As subnationalgovernments have grown more powerful, ambitious politicians have increasinglysought out elected subnational positions. Using their record of subnational ruleas a significant credential, many former mayors and governors have launchedcampaigns for national office, including the presidency. This represents a majordeparture from more traditional career paths and may represent an unintendedconsequence of decentralization for political parties: a threat to their ability to

Decentralized Politics and Political Outcomes 173

maintain a monopoly on political advancement within the system. However,this development may also provide an opportunity for traditional parties, ifthis increased competition within parties revitalizes their connection to their constituents.

Decentralization in the Andes

Unfortunately, a clear and universally accepted definition of decentralizationdoes not exist.7 It is thus critically important to define decentralization as I willbe using it in this chapter: decentralization is a reform (or series of reforms) thatincreases political power through the election of subnational officials where theyhave been previously appointed and that also accords some level of autonomousfiscal power to those elected officials.8 For a very rough sense of comparativelevels of fiscal decentralization across the Andes, see Table 6.1.

To speak of decentralization in the Andes requires careful attention to the va-riety of experiences even within this subregion of Latin America. Each countryin the region decentralized at a different time and through a different process.As a preface to my exploration of decentralization’s effects on democratic rep-resentation in Bolivia, Colombia, Ecuador, Peru, and Venezuela, this chapterbegins by briefly surveying these experiences. For ease of exposition, this sec-tion proceeds chronologically through the cases (see Table 6.2 for dates of keyreforms in each country).

174 O’Neill

Table 6.1

Expenditure Decentralization in Latin America

Subnational spending Country as a % of total spending

Argentina 49.3Brazil 45.6Colombia 39.0Bolivia 26.7Mexico 25.4Venezuela 19.6Uruguay 14.2Chile 13.6Peru 10.5Ecuador 7.5Paraguay 6.2Latin American averagea 14.6OECD average 34.9

source: These figures are based on 1995 data and are takenfrom IDB (1997, 157).

note: Countries in italics are Andean countries.a Not all Latin American countries included in the average are

listed in the table.

Unlike neighbors such as Brazil, Argentina, or Mexico, the Andean countriesdo not have a long history of federalism (Venezuela’s nominal federal structurenotwithstanding). Decentralization in these five cases thus represented a majorchange from a history of rather centralized rule.

Peru began to decentralize before its most recent period of democracy, dur-ing Fernando Belaúnde Terry’s first presidential administration (1963– 68). Theelection of municipal governments was reintroduced in the constitutional as-sembly (1978), and the election of mayors began once again, in 1980, when Belaúnde was re-elected to the presidency. Despite the fact that subnationalofficials were elected in Peru, these officials suffered from a severe shortfall inresources. During Belaúnde’s second term (1980 –85), he strengthened theirbase of fiscal resources somewhat, creating a formulaic system that transferredslightly more (but still largely insufficient) resources to local governments.

Decentralization to local governments changed little during Alan García’sturbulent presidency (1985–90). As it became increasingly clear that García’sAlianza Popular Revolucionaria Americana (American Popular RevolutionaryParty, or APRA) would not be a strong competitor in 1990 national elections, hisgovernment began to push (successfully) for the election of regional governors(Thedieck and Buller 1995), and the first election of this type took place in 1990.While APRA did abysmally in national polls in 1990, its candidates won regionalexecutive elections in all but one province (Buller 1993, 151). When Fujimoridismissed the legislature and other elected officials of the government in his 1992autogolpe, these regional officials were also dismissed and the election of regionalofficials ceased. Fujimori did, however, allow new local elections in 1993.

Instead of increasing funding to municipal governments, Fujimori intro-duced Decree Law 776 in 1993; this law drastically cut back transfers to local


Table 6.2

Popular Election of Subnational Executives

Country Year Summary

Bolivia 1995 First countrywidea popular election of mayorsColombia 1988 First popular election of mayors

1991 First popular election of governorsEcuador 1980 First popular election of mayors and prefects in new

democratic periodb

Peru 1980 First popular election of mayors in new democratic periodc

1990 First popular election of governors1992 Postponement of mayoral elections (1 year); end of the

popular election of governorsd

Venezuela 1992 First popular election of mayors and governors

a Mayors had been elected in large municipalities before 1995.b Mayors and prefects had been elected prior to the authoritarian period (1972–79)c Mayors were first elected in 1963, but elections were later interrupted by authoritarianism (1968–79).d President Toledo promised gubernatorial elections in 2002.

governments. In fact, municipal budgets shrank by more than 75 percent in theyear following this reform (Kay 1995). In place of these funds, Fujimori createda local development program called FONCODES that provided direct central government funding for particular local projects; he maintained a great deal ofdiscretion in determining the recipients of this money. In addition, he eased therules for seeking local election, allowing a flood of independent candidates thatmade it hard for either the remnants of old political parties or nascent politicalparties to build a base through these contests. In this sense, many scholars pointto Fujimori’s Peru as an example of recentralization (Dammert Ego Aguire1999; Delgado Silva 1995). Currently, local government in Peru continues toreceive very little funding; however, support for democratic election of regionalofficials appears to be growing (Tanaka 2002, this volume).

One of the earliest decentralizers in the region, and one of those that has takenthe greatest strides toward extending fiscal and political devolution, is Colombia.Colombia decentralized power to subnational governments in two stages: duringthe presidency of Belisario Betancur (1982–86) and during its constitutional as-sembly (1990 –91). Betancur’s government initiated laws that introduced the di-rect election of local mayors (the first elections were held in 1988) and that pro-vided local governments with a significant percentage of the central government’sfiscal resources, to be allocated through a complex formula that included suchitems as population, percentage of the population whose basic needs went un-satisfied, and fiscal effort in raising local resources. A second round of decentral-ization during Colombia’s constitutional assembly in 1990 –91 introduced thepopular election of regional governors, created automatic transfers to regionalgovernments, and included provisions strengthening local governments still fur-ther. There is no question that, for better or worse, Colombia is the most de-centralized government in the Andean region.

After Colombia’s decentralizing moves in the early 1980s, the next attempt atdecentralization occurred in Venezuela, beginning with legislative acts passed in1989. These acts allowed for the popular election of mayors and governors andguaranteed these governments both financial resources and new responsibilities.The outline of the decentralization process in Venezuela differed significantlyfrom that in Colombia and—as we shall see—in Bolivia; instead of a majorchange toward increased funding, Venezuela’s decentralization reforms followeda more gradualist approach. Each regional government seeking more autonomyhas to petition the legislature for its approval of increased responsibilities and theincreased funding to go with them (de la Cruz 1992). This purportedly allowsthe legislature to ensure that the subnational unit has the proper capacity for car-rying out such important responsibilities as local educational policy; in practice,it also gives the legislature enormous power over the pace and extent of decen-tralization (Kraemer 1999).

176 O’Neill

In 1994, Bolivia began an ambitious experiment in decentralizing reform.Bolivia’s Law of Popular Participation (LPP) created 311 municipalities, largelyin communities where no official local government had ever existed. It allowedfor the election of local councils. If any party in the municipal race achievedgreater than 50 percent of the vote, its mayoral candidate was elected directly to become the mayor; otherwise, municipal councils chose one of their mem-bers to be the mayor. In addition, this law guaranteed that 20 percent of the national fiscal intake would be divided among the municipal governments, on apurely per capita basis. One of the innovative features of Bolivia’s decentraliza-tion is its official recognition of grassroots civil organizations (neighborhood as-sociations, peasant unions, and indigenous groups). The LPP encourages mem-bers of these groups to form civic oversight committees (Comités de Vigilancia)to oversee the allocation of funds within the municipality. These organizationscan petition the central government to freeze fiscal transfers to the municipalitypending a central government inquiry into municipal management. Anotherfeature calculated to increase the accountability of mayors to their constituentsallows the municipal council to remove mayors from their position at one-yearintervals and to replace the mayor with another of their number.9 Regional gov-ernors remain appointed by the president.

Finally, the laggard in the region on decentralizing reform has been Ecuador.Despite the election of local and provincial government officials, these officialsreceive very little money, and much of the money that does arrive is subject tothe discretion of the central government. Extensive debates over decentraliza-tion have raged in Ecuador in recent years, peaking in 1997 and 1998 with theLey de 15%, which promised a transfer of 15 percent of the central government’sfunds to the local and provincial governments. Implementation of this law lan-guished until after it was reformed in Ecuador’s 1998 Constitution.

The 1998 Constitution changed the basis on which central government trans-fers (including those transferred under the Law of 15%) would be distributed.The Law of 15% had called for 10 percent of the funds to be distributed in equalparts to each canton: 50 percent according to an index of basic needs unsatisfiedamong the population; and 40 percent based on the population of each can-ton. The new constitution dropped the first criterion and added three others:financial capacity, improvement in level of life betterment, and administrativeefficiency. Finally, the Law of 15% was criticized for transferring funds withoutresponsibilities, which led to a change in the wording of the constitution. Article226 now reads, “There cannot be a transfer of responsibilities without a transferof equivalent resources, nor a transfer of resources without (corresponding) re-sponsibilities.” The decentralization law also allows for the decentralization offunds and responsibilities to occur through a bargaining process between the cen-tral and subnational governments (similar to Venezuela’s system). In addition, the


constitution also allows for the direct election of mayors in all municipalities(previously this had occurred only in provincial capitals and in cities with popu-lations greater than 50,000).

The political innovations called for in the new constitution have been imple-mented, but the financial aspects have lagged behind as the country grapples withthe worst financial crisis in its history. In 1999, although fiscal decentralizationhad only reached about 9 percent10 (roughly half of the 15 percent required bylaw), critics argued that the government simply could not “afford” to transferthese funds to subnational levels during a period of such national crisis. The eco-nomic problems contributed not only to the removal of Ecuador’s president, butto the first ever default on Brady bonds and to the adoption of the U.S. dollar asthe nation’s currency. In the government’s attempts to deal with these problems,decentralization has been set aside as a major issue.11 In addition, the focus of thedecentralization debate has shifted away from the decentralization of politicaland economic resources to municipal governments. Instead, the debate has beenrefocused on whether or not the government should grant autonomy to ethno-linguistic groups in specifically defined indigenous territories and to increasingthe power of the provinces (Cameron 2000).

Clearly, the Andean reforms reflect a wide range of experience with decen-tralization (see Table 6.1, above, for a sense of how the Andean countries com-pare to other Latin American countries—and to each other—in terms of howmuch subnational governments spend relative to what the national governmentsspend). Colombia has gone the furthest toward giving both mayors and gover-nors a democratic basis and a significant flow of fiscal resources, allocated ac-cording to transparent criteria. In addition, each level of government has its ownsources for raising revenue and the criteria for intergovernmental transfers givesubnational governments incentives to raise their own resources in addition toreceiving transfers from above.

Bolivia has also taken major steps toward decentralizing its government, buthas focused its efforts on mayors and local councils, stopping short of empower-ing regional governments. Its innovative features include several provisions forincorporating civil society into local government decision making.

Both Venezuela and Ecuador have recently taken significant steps toward de-centralization, but have adopted a more gradualist approach than either Colom-bia or Bolivia, allowing individual subnational governments the opportunityto petition for increased resources and responsibilities. On the political side,Venezuela (especially) and Ecuador (to a lesser extent) have made some significantstrides toward increasing the electoral accountability of subnational officials totheir communities, even where they do not control significant fiscal resources.

Peru’s experience is an important counterpoint to the stories of its neighbors:here decentralization proceeded tentatively during its early years of democracy,but many decentralizing reforms were either nullified or significantly rolled

178 O’Neill

back during Fujimori’s presidency, leaving elected local governments with fewfiscal resources. With such a wide range of experiences, the Andean region pro-vides an exciting set of cases for exploring the effects of decentralization ondemocratic representation. The next section begins this exploration by investi-gating patterns of participation in local elections.

Participation

One of the clearest ways that decentralization contributes to democratic prac-tice is by creating a wider variety of opportunities for citizens to participate inthe democratic process. Voter turnout is one way to judge the involvement ofcitizens in decentralized contests. Rather than compare the turnout of citizensacross the countries in the region, this section compares local turnout to nationalturnout within each country, in an attempt to control for various national lawsand cultural factors that might affect an individual’s interest in casting a ballot.For example, in Venezuela, where voting is mandatory, one would expect higherturnout than in Colombia, where it is not. The point of comparing national withsubnational turnout within a country is to control for these kinds of factors. Theevidence that follows suggests that Andean citizens are quite excited about theopportunity to participate in a greater range of democratic contests. Through-out the region, turnout in local contests relative to national contests comparesfavorably with more established democracies, suggesting that individuals are in-terested in their local governments and eager to participate in choosing localofficials. Of course, there is important variation within the region; in some cases,after an initial increase, turnout in subnational contests declines over time rela-tive to turnout in national contests. The gap between turnout in subnational andnational contests is also affected by the timing of elections and the actual powerof subnational officials across countries.

One of the major political development arguments for decentralization is thatthe opportunity for citizens to participate in elections and government at the lo-cal level gives them an education in democracy more generally. Perhaps this willhave the effect of conditioning citizens to expect the delays and compromises in-herent in democratic debate and decision making at all levels of government,making democracy more robust in the face of economic downturns and moreresistant to populism and demagoguery. While not fully exploring this claim,looking at electoral turnout figures does show a strong tendency for citizens whoare given the chance to participate in local elections to take that opportunity.

Before delving into the Andean electoral turnout results, I want to includesome data on turnout in local versus national elections in the United States. Al-though the United States is widely noted as a case with low electoral turnout innational elections within the developed world, its turnout rates in local contestsare even more abysmal. Most scholars of elections and voting behavior have re-


marked on a trend in which voters turn out in much smaller numbers for elec-tions at more local levels of government. Looking at the turnout rates in theUnited States, there appears to be a linear decline in participation rates as thepower invested in the office for which elections are being held declines. Inthe 1996 presidential elections, national voter turnout in the United Statesreached 48.91 percent, while for congressional elections, the number reached45.6 percent. Looking at one county,12 gubernatorial turnout in 1996 reachedjust shy of 45 percent, while voting in local elections reached only 25 percent ina year when the mayor was being elected (1997), and did not even reach 15 per-cent in 1999, when the mayor’s position was not open to election. This gap be-tween voter turnout in national and subnational elections in the United Statesis illustrative of a trend throughout the developed democracies. In the Andes,local electoral turnout also lags behind turnout in national elections, but thedivergence is not nearly as high as it is in the United States.

Looking at Figure 6.1, which plots national and subnational turnout in mil-lions of valid votes in each of the five Andean countries, the Colombian caseimmediately catches the eye. Here local elections stretch back to 1988, with na-tional elections continuing uninterrupted throughout this period.13

As Figure 6.1 demonstrates, local turnout in 1994, a year in which both na-tional and local contests occurred, was significantly higher in the national polling:local turnout reached only 63 percent of national turnout. What is more striking,however, are the sharp increases in local turnout in 1997 and 2000, when turnoutin local contests was more than double the turnout in national contests two yearsearlier. In the 2000 election, a “ballot for peace” was held concurrently with themunicipal elections; a great deal of voter turnout must therefore be seen as a re-sponse to this peace initiative rather than as an extraordinary increase in voter en-thusiasm for local contests. Still, the increased turnout in 1997—a year when lo-cal elections occurred independently of national elections (which should depressturnout)—does signal a strong level of interest in local politics by the electorate.

Bolivia’s local turnout follows a very different path than Colombia’s: herelocal turnout skyrockets and then declines significantly. Prior to the 1994 de-centralizing reforms, Bolivia held local contests only in its largest cities andtheir suburban neighborhoods. As a consequence, turnout in these contests wasrestricted. In 1995, the first nationwide round of local contests took place in311 municipalities. Turnout was extraordinarily high, with the total number ofvoters in the 1995 local elections exceeding the total number of voters whoturned out in 1993 for national legislative and presidential elections. After thispeak in the first local elections, turnout declined in Bolivia’s 1999 elections. Mosttroubling, the turnout in 1999 amounted to less than the number of voterswho took part in the 1991 local contests that were restricted to the largest citiesand towns in the country. What is more, the 1991 elections took place in thecontext of a system that devolved very few resources to local governments, in

180 O’Neill

Fig

ure

6.1

Tur

nout

in m

illio

ns o

f vot

ers,

by le

vel

20�

15�

10� 5� 0

Year

1980

1990

1985

1995

2000

Mun

icip

alLo

cal

Pres

iden

tG

over

nor

Nat

iona

l

20�

15�

10� 5� 0

Year

1980

1990

1985

1995

2000

Year

1980

1990

1985

1995

2000

Bol

ivia

Col

ombi

aC

hile

20�

15�

10� 5� 0

Year

1980

1990

1985

1995

2000

Ecu

ador

20�

15�

10� 5� 0

Year

1980

1990

1985

1995

2000

Ven

ezue

la

20�

15�

10� 5� 0

Year

1980

1990

1985

1995

2000

Peru

10 8 6 4 2 0

contrast to the much better funded system in place in 1999. Looking for an an-swer, one might concentrate on the institutional mechanisms governing localpolitics in Bolivia. Declining voter turnout may reflect voter frustration with theindirect nature of municipal elections for mayors or the furious pace at whichmayors were overturned and replaced by municipal councils between 1995 and1999 due to the censura procedures embedded in the earlier reforms. A third hy-pothesis might link declines in turnout to the restrictions that have kept indepen-dent candidates—those not affiliated with a recognized political party—fromrunning in local contests.14 In short, the result is overdetermined; certainly greaterinvestigation would be necessary to understand this drastic change in turnout.

A third case to explore is the Venezuelan system. Since the greatest degree ofdecentralization conferred resources on regional rather than local governments,I compare turnout in national elections for president with turnout for guberna-torial contests. The striking feature of the Venezuelan case is the huge increasein turnout between 1995 and 2000. While the 1995 gubernatorial elections metwith extremely low turnout (only 32 percent of 1993’s national turnout), turnoutincreased dramatically in the 2000 elections, when subnational turnout nearlyequaled national turnout. Given the gradualist nature of Venezuela’s decentral-ization reform, this increase in turnout could correspond to increases in real re-sources to regional governments. To investigate this hypothesis, one could lookat the degree of decentralization granted to each regional government and the in-crease in electoral turnout within that region between 1995 and 2000. Of course,as we will see in both the Ecuadorian and Peruvian cases, concurrent electionsgenerally lead to much higher turnout. In Venezuela, the 2000 elections wereparticularly exciting and elicited a high degree of turnout, making our task ofparsing out the differential contribution of decentralization to turnout impos-sible. To determine what is driving turnout in Venezuela, we would have to ob-serve at least one more non-concurrent set of regional elections; in addition, re-cent constitutional changes have changed electoral rules in new ways that mayalso be contributing to changes in voter turnout at both national and regionallevels. It seems too soon to draw any conclusions in the Venezuelan case.

Ecuador presents yet another pattern of national and municipal voter mobi-lization. When Ecuador returned to democratic rule in 1980, turnout in bothnational and local contests barely differed (municipal turnout was 95 percent ofnational turnout); however, as time went by, turnout increased at a higher ratefor national elections than for municipal elections. While I do not have the datato determine whether municipal turnout increased after 1998, when electoralrules changed to increase the number of local officials being popularly elected,the period 1980 –95 suggests a trajectory of growing divergence between na-tional and municipal electoral turnout. National turnout climbed at a faster ratethan municipal participation.

Given the patterns we have seen in Peru’s neighbors, we should perhaps expectto see very low turnout in local elections in Peru, since these local governments

182 O’Neill

have little fiscal power. Instead, Figure 6.1 shows strong turnout at the locallevel. When national and municipal elections are concurrent, local turnout issignificantly higher than in other years; still, in 1998, local turnout is considerablyhigher than national turnout in 1995 and it is 85 percent of national turnout in2000. Tanaka (2002) persuasively argues that local elections became an outlet forventing opposition to Fujimori in the later years of his presidency.

Comparing across these five countries, there does not appear to be a strongand direct relationship between the strength of municipal governments and therate at which citizens turn out to vote for subnational officials. Based on themost recent rounds of elections, turnout in municipal elections reached the fol-lowing percent of turnout in national elections in each country: 95 percent inColombia,15 88 percent in Venezuela,16 85 percent in Peru,17 74 percent inEcuador,18 and a dismal 54 percent in Bolivia.19 Extrapolating from the U.S.data noted above, the comparable local/national figure for the United Stateswould be just over 51 percent, lower than the figures in any of the five Andeancountries. It is fascinating, then, that subnational voter turnout as a percentageof turnout in the most proximate national elections in each of the Andean coun-tries is much higher than it is in the United States, even though subnational gov-ernments in the United States control far greater resources.

Within the Andean cases, however, there does appear to be some rough cor-relation between high voter turnout and the level of power controlled by sub-national officials. A rough classification of the relative strength of decentraliza-tion reforms across the five cases would probably rank the cases as follows:Colombia, Bolivia, Venezuela, Peru, and Ecuador. The real surprises here, then,are Bolivia (for its low turnout) and Peru (for its high turnout). A static view failsto capture the whole story, however. In Bolivia, local turnout skyrocketed andthen declined; in Colombia, local turnout has been steadily and sharply rising;in Ecuador, local turnout is slowly climbing; in Venezuela, it is rising sharply—but this may reflect a simple difference between concurrent and non-concurrentelections; and in Peru, local turnout is declining relative to national turnout.This pattern, except for the Bolivian anomaly, seems to fit more clearly with thescope and pace of decentralizing reforms across the five countries. This suggeststhat citizens are quite savvy in responding to devolution of power; where sub-national governments are not very strong, citizens do not exert themselves tovote at the same rate as do voters electing more powerful subnational officials.

Political Parties, Local Elections, and Public Opinion

One of the hypothesized benefits of decentralization is that it might make gov-ernment more responsive to its citizens. In particular, decentralization shouldgive political parties and leaders incentives to develop a closer relationship withtheir constituencies, thus reversing the crisis of representation. Alternatively, de-centralization may create the conditions for new parties to form at subnational


levels, gain credibility through good government performance, and then launchnational electoral campaigns. This section will examine these hypotheses usingpublic opinion data and election results by party. The two points investigatedhere are: First, how do citizens view their local elected officials and how doesthis compare with their views of national politicians? Second, how well have tra-ditional parties fared in local contests over time?

While turnout in local contests may be considered one measure of how citi-zens feel about their local politicians, there are many other factors that determinewhether or not an individual might vote. One might vote to get rid of corruptofficials; because it is a “civic duty”; in order to secure a position in the local bu-reaucracy; or because one happens to be walking by the polling place and findsthat all of one’s friends are there voting. To get a more direct feel for how citi-zens view their local governments and the political parties who contest elections,I turn to some scattered public opinion data.

The first thing to note about public opinion toward political parties in LatinAmerica is that it is exceptionally low. If parties pushed through decentraliz-ing reforms in order to increase the popularity of political parties, then it hasbeen a resounding failure. A 1998 poll conducted by the Wall Street Journal of theAmericas reports that the percentage of respondents who said they had some ormuch trust in political parties was: 10 percent in Bolivia, 15 percent in Colom-bia, 6 percent in Ecuador, 19 percent in Peru, and 6 percent in Venezuela. Com-pare this with 37 percent in the United States and 22 percent in Costa Rica. Po-litical parties ranked well below the press, the armed forces, large companies, thepolice, unions, and even the legislature, in every single Andean country. InColombia, where the traditional Liberal and Conservative Parties have faredbetter than most traditional parties in the region, trust in political parties hasconsistently ranked below trust in other institutions. In 1989, 15.9 percent ex-pressed trust in parties (Semana 1990, 89); by 1994 that number was 22 percent(Semana 1996); by 1995, it had returned to 16 percent (Semana 1996).

Are parties and governments at the local level doing better than parties andgovernments at the national level? Again, a brief look demonstrates a widevariety of opinions across and within countries. A Cedatos poll of citizens in pro-vincial capitals in Ecuador in 2000 —one year into mayoral terms—noted widevariation in mayoral approval ratings, with a high of 78 percent in Guayaquil anda low of 31 percent in Machala (Quito’s mayor, Paco Moncayo, won only a37 percent approval rating). Overall, however, the poll concluded that the “levelof approval of the mayors exceeds the President’s,” and that “the populationlooks at these authorities as actors close to their daily problems, their neighbor-hoods, and their communities.”

A 1998 poll of Bolivians residing in the capital city of La Paz found only 9.3percent who agreed with the statement that they had “much” confidence in themunicipal government; 46.9 percent claimed to have “little” confidence; while

184 O’Neill

nearly 42 percent responded that they had “no” confidence in their municipalgovernment (IINCIP 1998, 17). In the same poll, 13.3 percent claimed to have“much” confidence in the [national] government; only 7 percent had “much”confidence in political parties (IINCIP 1998, 16).

Lack of confidence in particular municipal administrations contrasts sharplywith support for the idea of decentralization. A 1994 Bolivian poll asked respon-dents if they were in agreement or in disagreement with the Law of Popular Par-ticipation (the decentralizing law); 44.8 percent said they were in agreement withit, while 26.4 percent disagreed with it and 28.8 percent either did not respondor said they did not know. At the same time, however, only 9.6 percent re-sponded that they knew much about the reform, while 62.4 percent said theyknew little about it and 15 percent said they knew nothing about it. Asked whythey were in favor or against the reform, 39.9 percent said they did not know; thiswas the most frequent response, with “other positive reasons” garnering 20.5percent, and the next highest response (with 11.7 percent) “community partici-pation” (Instituto de Encuestas 1994, 31–33).

A similar pattern emerges in the Peruvian case. When asked whether they ap-proved or disapproved of their district mayor’s management, respondents in fivemajor cities varied in their responses: approval was voiced by 55.9 percent inLima, 51 percent in Cusco, 40.1 percent in Junin, 37.9 percent in Loreto, and50.5 percent in Piura (Grupo Propuesta Ciudadana 1996, 135). In contrast, over80 percent of respondents in all five cities believed decentralization would im-prove agricultural development, employment, and education (Grupo PropuestaCiudadana 1996, 121). In a poll undertaken during Colombia’s 1991 constitu-tional convention, 82.2 percent of respondents said they believed that the revisedconstitution should allow for the popular election of governors (Semana 1991).Citizens in decentralizing countries appear enthusiastic about decentralizationbut only sometimes find their particular local governments to their liking.

How have particular parties done in decentralized contests? In particular, howhave traditional parties fared in subnational competition? It is conceivable that de-centralization might have the effect of strengthening political parties by bringingthem closer to their constituents and rebuilding the connections between themthrough the election process, which generates information at the local level aboutcitizen preferences. Another scenario suggests that decentralization might un-dermine traditional parties by creating the conditions under which minor partiesor popular independents could win power on a small scale. Launching an inde-pendent campaign or trying to create a nationally viable party can be prohibi-tively expensive. Local and regional elections create non-trivial seats of power thatmight be attainable for independents and incipient parties. Where traditional par-ties do not adequately represent their constituents and where barriers to entry arelow, decentralization may lead to the growth of new parties or encourage a bar-rage of independent candidacies. Some scholars view this as a major benefit for


democracy, while others caution that these fissures at the local and/or regionallevel may lead to a drop in party coherence and in policy coherence across levelsof government over time.

If decentralization is creating arenas of power where new parties are formingand building strength for a run at national office, we should expect to see a de-cline in traditional party vote share at the regional/local level first, followed bya fall in the traditional party vote share in national elections. The data in Figure6.2, which graphs the support for traditional parties at both the national andsubnational level in each country, do not show a clear pattern of this type in anyof the cases. The country that comes closest may be Bolivia, where a decline insupport for traditional parties 20 at both national and local levels between 1991and 1995 is followed by a decline in their support at the national level between1997 and 2002; however, there is an increase in the support for traditional par-ties in the 1999 local contests. In Colombia, prior to the 2002 presidential con-test, both national and subnational support for Liberals and Conservatives staysfairly stable. In Ecuador, traditional parties 21 tend to do better in subnationalthan in national elections; this is also true in Venezuela.22 In Peru prior to 1985,traditional parties 23 did equally well in national and subnational contests; the1989 municipal results foreshadowed the 1990 national results, with the tradi-tional parties losing support at all levels. Since 1990, the traditional parties re-mained equally weak at all levels, until Alan García’s respectable showing in the2001 presidential contest.

The experiences in Ecuador and Venezuela suggest that, to the extent thattraditional parties are losing vote shares, they are doing worse at the nationallevel than they are at the subnational level—exactly the opposite of what a the-ory of party building from local or regional roots would suggest. Instead ofbreeding grounds for new voices, subnational contests may be the last outpostfor traditional party politicians.

Changes in the electoral fortunes of traditional parties documented in Figure6.2 cannot be wholly linked either to the onset of decentralization in these coun-tries or to the particular contents of decentralizing reforms. In Peru and Vene-zuela, in particular, the precipitous fall in support for the traditional parties hadmuch more to do with failed economic policies and perceived corruption at thenational level. In addition, Fujimori changed the rules governing local electionregistration and candidacy to encourage the proliferation of independent candi-dates in local contests, perhaps in a bid to undermine the creation of rival partiesat this level of government. Given the collapse of traditional parties at the na-tional level in Peru and Venezuela, what is most surprising is the ability of theseparties to win a fair number of subnational victories. As previously noted, whenPeru’s APRA experienced a stunning defeat in the 1990 national elections, it stillmanaged to elect its partisans to all but one regional executive position. Likewise,as the AD (Acción Democrática, or Democratic Action) and COPEI (Comité

186 O’Neill

Fig

ure

6.2

Perc

ent

of t

he v

ote

won

by

trad

itio

nal p

arti

es, b

y le

vel

Reg

iona

lN

atio

nal

Loca

l

Year

1980

1990

1985

1995

2000

Bol

ivia

Col

ombi

aE

cuad

or

Ven

ezue

laPe

ru

Year

1980

1990

1985

1995

2000

Year

1980

1990

1985

1995

2000

Year

1980

1990

1985

1995

2000

Year

1980

1990

1985

1995

2000

10 8 6 4 2 0

10 8 6 4 2 0

10 8 6 4 2 0

10 8 6 4 2 0

10 8 6 4 2 0

de Organización Política Electoral Independiente, or the Social ChristianParty—literally, Committee of the Independent Political Electoral Organiza-tion) have fallen on hard times in national contests, they have been able to win afair number of gubernatorial and mayoral contests in Venezuela.

While it is tempting to focus on the cases where traditional party support hasdeclined precipitously, comparing the cases where support for traditional par-ties has remained high at the subnational level reveals some interesting insightsas well. It is particularly interesting to compare Colombia and Bolivia. In theBolivian case, institutions played a crucial role in allowing Bolivia’s major par-ties to retain their hegemony in local elections: electoral laws forced candidatesto run on established party platforms, disqualifying “independent” candidates.This law was changed early in Sánchez de Lozada’s second term. In the first mu-nicipal elections allowing non-party candidates to run (December 2004), polit-ical parties still won approximately 77 percent of the vote; however, traditionalparties did very poorly.24

In Colombia (and Peru, also), in contrast, independent candidates have mul-tiplied at a quick pace. The ability of Colombia’s parties to continue to domi-nate local elections may owe to their flexibility; it is not unusual to see severaldifferent candidates from the same party competing against each other. WhereasPeru’s system encourages candidates to run on independent platforms and Bo-livia’s forbids this, Colombia’s system encourages independent candidates to runwithin the major political parties, perhaps contributing to their vitality. It mustbe noted that this electoral vitality may come at the cost of jeopardizing the in-ternal coherence of parties and diluting the meaningfulness of party labels (seePachano and Pizarro Leongómez in this volume).

At the same time that national political trends have affected the ability of tra-ditional parties to dominate subnational politics, decentralization also creates theconditions under which subnational politics can affect politics at the nationallevel. The next section explores this direction of causality; in particular it ex-amines the extent to which decentralization has affected the career paths ofpoliticians.

Career Paths

The decentralization of real fiscal resources and the opening up of subnationalleadership positions to electoral contestation has a third plausible impact on po-litical practice in Latin America that is worth further exploration: it introducesnew career paths to ambitious politicians. Looking through history at the typi-cal political trajectory toward the presidency in Latin American countries, onewould find very few executive biographies listing gubernatorial or mayoral po-sitions. Instead, the political résumés of Latin American presidents typically in-cluded either prominent positions within the legislature or influential ministry

188 O’Neill

positions—or both. Today, the number of presidents and presidential hopefulswith a history of executive service at either the regional or the local level is mul-tiplying at a rapid rate. The turn-of-the-millennium round of presidential elec-tions in the Andes illustrates that candidates with experience in elected execu-tive positions at the regional or local level have won a significant percentage ofthe vote in many countries—and have won the presidency in some. Colombia’spresident, Álvaro Uribe, elected in 2002, served as the governor of Antioquia.Bolivia’s 2002 presidential elections saw the multi-term mayor of Cochabambacome in a close third in the voting. The winner of Ecuador’s 1998 presidentialelection had been the mayor of the capital, Quito. In Ecuador also, in 2002, oneex-president actually sought a subnational executive position after serving ashead of state: Leon Febres Cordero became mayor of Guayaquil after serving aspresident. Though Hugo Chávez in Venezuela had no previous governing ex-perience, his main competitor in the 2000 elections, Arias Cárdenas, was thetwo-term governor of Zulia. Of the Andean countries, only Peru did not havea major candidate in its 2001 presidential election with experience as a mayoror governor.

Both of these trends—governors and mayors running for higher office, aswell as national officeholders turning to regional and local executive positionsfor career advancement—have potentially profound consequences for the evo-lution of political parties in decentralized countries. Where independents canachieve power in a subnational contest and use their performance in those posi-tions to launch a bid for presidential power,25 traditional parties lose their abil-ity to control access to the national electoral arena. More importantly, wherepopular party members can use subnational elective positions to leverage theirpossibility of winning nominations within parties, traditional party elite losetheir influence over political succession and, perhaps, over party coherence. Onthe one hand, this trend has the positive effect of creating new avenues to powerand reining in the party elite. It may also have the effect of allowing more pe-ripheral parts of the country to play a greater role in national politics if they canelect one of their own to the presidency on a strong record of subnational gov-ernance. On the other hand, this diminishes a party’s ability to discipline itsmembers within legislatures once new executives have been selected. Further-more, this trend may shorten the time horizons of key policymakers as execu-tives become less accountable to party organizations in general.

To evaluate these claims, this section presents a great deal of empirical evi-dence, charting whether and how the road to the presidency has changed formajor presidential candidates between approximately 1980 and 2002. As in theother sections of this chapter, there is a wide range of experiences across thesefive countries. These differences tend to be strongly related to the ability of par-ties to control access to the executive and to subnational offices, and to thepower wielded by subnational officials. As one might expect, where mayors and


governors have become powerful, there has been an increase in the number offormer mayors and governors running for president and gathering a significantpercentage of the vote (I limit my analysis to presidential candidates who havewon at least 5 percent of the national vote). The ability of parties to play a strongrole in choosing candidates also plays a critical role, but it is a more uneven one:it is not always true that weak party control over nomination leads to an increasein the numbers of mayors and governors running for national office over time.In the ensuing paragraphs the record will be carefully examined.

Prior to decentralization, the political biographies of major presidential can-didates in Latin America tended to look much like that of Carlos Andrés Pérez,who was president of Venezuela between 1974 and 1979 and was voted in for asecond term in 1988. Carlos Andrés Pérez was first elected to a state legislature(1946); one year later, he was elected to serve in the national legislature. After amilitary coup interrupted his career, he returned to the national legislature(1959) and was soon (1960) named the first director general of the Ministry ofInterior Relations and later its head. A short while later, he became the party’scongressional leader (1964), then its national secretary (1968) and a member ofits powerful National Executive Committee. Finally, he was chosen as the party’scandidate (1973) and won the presidency after years of service to the party innational politics.

Contrast this with the political résumé of Colombia’s president, Álvaro UribeVélez. In 1976 he began his political career as the Benefits Chief of Medellín’spublic enterprises. Just one year later, he became the secretary-general of theMinistry of Labor and, from 1980 to 1982, he served as the director of CivilAeronautics. In 1982, he was appointed mayor of Medellín and was a council-man in that city between 1984 and 1986. From 1986 to 1994, he served as a sen-ator in the national legislature and was then elected governor of Antioquia in1994, a post he kept until 1997.

While these vignettes arise from different countries, they represent two verydifferent trajectories to the national executive. One begins with a short periodof political service at the subnational level and moves quickly into the nationallegislature and party service; the other begins with more extensive local service,moves on to national legislative service, and then returns to a subnational posi-tion before launching his candidacy for president. This comparison alone can-not support the contention that decentralization has fundamentally altered ca-reer paths. What is interesting is the extent to which career trajectories of thefirst type used to be much more common in the region, and the extent towhich, more recently, career trajectories of the second type have become thenorm throughout the region.

Venezuela provides an excellent example of this trend. In 1983 and 1988, be-fore decentralization took hold in the country (the laws were passed in 1989),none of the major candidates had served in a subnational executive position as

190 O’Neill

either a mayor or governor. In the 1993 presidential elections, while the winner(with 30.45 percent of the vote) had not served as a mayor or governor, the otherthree candidates (accounting for 68.28 percent of the vote) had all served as sub-national executives. This trend continued into the 1998 and 2000 elections,where the winner had not served in this role, but his major challengers (winning39.97 percent of the vote in 1998 and 37.5 percent of the vote in 2000) bothhad. In addition, it is interesting to note that the subnational positions held bymajor candidates in the 1993, 1998, and 2000 elections varied quite a bit. Twoof the major candidates in 1993 were former governors (Andrés Velásquez wasgovernor of Bolívar and polled 21.95 percent of the national vote; OswaldoAlvarez Paz was governor of Zulia and polled 22.73 percent). A third major can-didate in that race, Claudio Fermín, was a former mayor of the capital, Caracas.In 1998, the major contender to Hugo Chávez was Henrique Salas Römer, whowon 39.97 percent of the vote and had been the governor of Carabobo from1995 to 1998. Finally, in 2000, Chávez beat Francisco Arias Cárdenas (37.5 per-cent), the former governor of Zulia. Over time, it appears that most of the coun-try’s candidates emerge not from national political service in the legislature,ministries, or the party, but from subnational executive positions. In addition,these positions are spread throughout the country to some extent; there is notone key office that must be obtained to launch a plausible presidential campaign.

Although Colombia has one of the strongest records of decentralization in theAndean region, the rising importance of subnational executive positions as a pres-idential credential has not been as stark as it has been in Venezuela. In 1986, priorto the direct election of mayors and governors, the victorious presidential candi-date, Virgilio Barco, had served in state as well as local government. He began hispolitical career in 1937 as Secretary for Housing and Public Works in the state ofNorte de Santander. He moved on to become a departmental assembly member(1945– 47) in Norte de Santander, and then moved on to the Cúcuta city coun-cil (1947– 49). He then moved into the national legislature in 1949. Later, hewould work on several presidential campaigns; he served several times as a na-tional senator, was appointed mayor of the capital, Bogotá, served in several min-isterial positions, and as ambassador to Great Britain and the United States.

The result in 1990 was similar, with the victor, César Gaviria, having servedas mayor of Pereira (a post he was appointed to in 1970), in the course of a po-litical career that also involved service in the national legislature, various min-istries, earlier presidential campaigns, and service to the Liberal Party itself. Twoof his opponents (polling a combined 37.7 percent) had not served in such sub-national positions; however, a third opponent (Rodrigo Lloreda Caicedo, whowon 12.4 percent of the vote) had also served as an appointed governor (Valledel Cauca) early in his career (1968–79).

In 1994, the competition centered around two politicians, Ernesto Samper,who eventually won 44.98 percent in the first round of voting, and Andrés


Pastrana, who won 45.3 percent in the first round, but lost the second round.Samper’s résumé looks much like that of Barco, Gaviria, and other successfulLiberal presidential candidates: he served as the elected deputy to Cundina-marca’s departmental assembly and on the city of Bogotá’s council as well. Heserved in the national legislature, as Minister of Economic Development, and asan ambassador to Spain and to the UN. Pastrana, on the other hand, had begunhis political career in the city council of Bogotá and had become the city’s firstelected mayor, in 1988. At the end of his term, he won a Senate seat, but turnedit down to launch his first, ultimately unsuccessful, presidential campaign. Thispattern would become more common in the elections to come.

In 1998, Pastrana again ran for office and, this time, won. His main competi-tor, Horacio Serpa, has a more traditional résumé: he was mayor of Barran-cabermeja early in his career, but had gone on to such national posts as Ministerof Government (the highest cabinet position) and Minister of Interior. He hadbeen a senator and a member of the Constitutional Convention that rewroteColombia’s Constitution in 1991, and the National Director of the Liberal Party.While this is a very traditional background, two of Pastrana’s other competitorscame from less traditional backgrounds, including Noemí Sanín, who has largelya business background and no executive office holding at subnational levels, andAntanas Mockus, whose highest elected position was mayor of Bogotá. The2002 campaign brought back Serpa and Sanín, but was won in the first round byÁlvaro Uribe Vélez, a politician whose major experience was obtained in keysubnational executive positions.

Summing up the Colombian experience, it seems that subnational office-holding has long been a part of most presidential candidates’ credentials. Whatis new in recent years is that these positions are now seen as capping one’s ca-reer instead of as minor offices held to vault one into the national legislature, avariety of ministerial positions, and direct positions within the party leadership.Today’s most successful presidential candidates gain a great deal of their experi-ence in high-profile mayoral and gubernatorial positions and lack the extensiveexperience in legislative, ministerial, or party service that was once common.

In contrast to the Venezuelan and Colombian cases, Peru’s recent presiden-tial contests do not show a gradual increase in the number of major candidateswith experience as mayors and governors. This is not to say that political careertrajectories have remained stuck in a traditional mold; rather, the profile ofPeru’s major presidential candidates has changed drastically over time, but sub-national officeholding has not become an important milestone on the pathwayto the presidential campaign. The importance of mayoral experience in seekingthe presidency peaked in the 1985 presidential campaign, and then declined.Also, Peru’s experience differs notably from Colombia’s and Venezuela’s in thatall of its major candidates with subnational experience have come from the samesubnational position: mayor of Lima.

192 O’Neill

In 1980, the two major presidential candidates (Fernando Belaúnde and Armando Villanueva del Campo, who together polled 72.6 percent of the vote)had no experience as elected mayors or governors. Belaúnde had founded hisown political party and had been elected to the presidency in 1963; Villanuevahad held several ministry positions, and served in and been the president of bothhouses of the legislature. The third-place finisher, with 9.6 percent of the vote,Luis Bedoya Reyes, had been the mayor of Lima.

In 1985, Alan García won the presidency with 45.74 percent of the vote. Hebrought a very traditional record to the campaign: he had served in the consti-tutional assembly in 1978 and then as a national legislator from 1980 onward.His two major competitors, Luis Bedoya (21.3 percent) and Alfonso Barrantes(10.23 percent), had both been Lima mayors. Javier Orlandini (6.25 percent),the fourth-place candidate, had not. This was to be the height of subnationalofficeholder success in national contests.

The 1990 contest between Alberto Fujimori and Mario Vargas Llosa pittedtwo relative newcomers to politics against one another. Neither had subnationalgovernment experiences, nor did either have experience in national politics.Polling just 7 percent of the vote was Henry Pease García, another former Limamayor.

In 1995, the contest between Fujimori and Javier Pérez de Cuellar can also bedescribed as a contest of nontraditional politicians; neither major candidate in thiscontest had served in subnational government. Politicians with neither tradi-tional political career trajectories nor subnational government experience alsodominated contests in both 2000 and 2001. In 2001, the fourth-place candidate(9.85 percent), Fernando Olivera, chose as his running mate Ricardo Belmont,who had been mayor of Lima from 1990 to 1993 and again from 1993 to 1995.

In Peru, the political experience of presidential candidates changed drasticallyfrom 1980 to 2001; however, the decline of candidates with experience in thelegislature, government ministries, and high party positions was not combinedwith an increase in the number of candidates with subnational political experi-ence, as it was in several other cases. While several former mayors of Lima con-tested the presidency in 1980 and 1985, their presence in presidential contestsdeclined in the 1990s and beyond. What is also interesting in the Peruvian caseis the lack of variety in the kind of subnational political experience that variouspresidential candidates brought to their candidacies: all of those who had servedin subnational government served in the capital, Lima, as its mayor. Given theweakness of subnational governments—particularly after Fujimori came topower—the inability of former subnational officials to run on an impressiverecord does not seem too surprising.

Bolivia’s experience from 1980 to 2002 differs from all three of the cases justoutlined. In presidential contests from 1980 to 1997, no former mayors or gov-ernors launched presidential campaigns netting them more than 5 percent of


the national vote; however, in the 2002 elections, the third-place candidate hadbeen the mayor of Cochabamba—Bolivia’s third-largest city—for several terms.None of the four top candidates in the 1980 contest (polling a combined 84.4 per-cent of the vote) had served in subnational government; none of the four top can-didates in 1985’s contest (polling a combined 78.9 percent of the vote) had servedin subnational government. Similarly, 1989’s top five candidates (polling 93.6 per-cent) did not include any subnational officeholders; neither did 1993’s top fivecandidates (90.2 percent of the vote). Finally, none of 1997’s six major candidates(96.8 percent) were former mayors, although the vice-presidential running mateof fourth-place candidate Ivo Kuljis had been mayor of Cochabamba. The 2002contest suggests that career paths in Bolivia may also be changing, despite the rel-ative newness of subnational elections (mayors were first elected in 1995) and thetight control Bolivian parties retain over choosing candidates for both presiden-tial and mayoral positions. It is interesting to note that 2002 candidate ManfredReyes Villa did not run on a traditional party label—he headed the NFR (NuevaFuerza Republicana, or New Republican Force) list—and that polls consistentlyshowed him in the lead as the election neared.

Comparing across the four countries discussed here, a few trends become ev-ident. First, there appears to be a dramatic change in the career trajectory of ma-jor presidential candidates in most countries in the region. Major candidates inthe early 1980s tended to have a great deal of experience in the national legisla-ture; they often held several ministerial positions, ambassadorships, and positionsof leadership within their national political parties. To the extent that they servedin subnational levels of government, they did so early in their careers and oftenheld elected positions in subnational governments with few resources, or heldappointed positions at the subnational level. In the 1980 elections in Peru andBolivia, and in the 1983 election in Venezuela and the 1986 election in Colom-bia, nearly all of the major candidates’ biographies roughly fit this profile. Shortlyafter decentralization took hold in Peru, Colombia, and Venezuela, several oftheir most successful presidential candidates began to list election to subnationaloffice high on their résumés. In Venezuela and Colombia, major candidates camefrom a variety of governorships—not just from the states in which the capitalsresided. In very few cases did mayors come from cities other than the capitals; inPeru, the only candidates with subnational government experience had beenformer mayors of the capital city. Peru is distinct in two additional ways. First, itdemonstrates a decline in the number of candidates with subnational governingexperience as power is recentralized in the 1990s. Second, it demonstrates theimportance of real power in vaulting subnational officeholders to nationalprominence: in Peru, where decentralization extends to local governments andnot regional governments, only mayors (not governors) run for national office.In Venezuela and Colombia, where decentralization empowered both levels ofsubnational government, former mayors and former governors have both run

194 O’Neill

credible campaigns for president. Still, there is variation related to power and re-sources even here: in Venezuela, where regional governments are given manymore resources than local government, the number of governors running cred-ible national campaigns (two in 1993 and one each in 1998 and 2000) dwarfs thenumber of former mayors running equally credible campaigns (one in 1993).

A third consideration in trying to understand the rise in the numbers of may-ors and governors running for president is the role played by political parties. Inboth Peru and Venezuela, the ability of a strong, insulated party leadership tochoose national candidates has been eroding over time for reasons unrelated todecentralization. The extent to which this party decline has led to an increasein mayoral and gubernatorial candidacies for president differs starkly across thetwo cases: a weakening party system is not enough to vault subnational office-holders into national prominence. In addition, it appears that the strength ofthose officeholders—their access to sources of finance and their ability to builda strong record of performance at the subnational level—also plays a major role.Where subnational governments have been strengthened, candidacies of thistype are more likely than in cases where they have been weakened.

The Colombian case also sheds light on this issue. While Colombia’s major par-ties continue to play a key role in most major elections, these parties have neverhad strict control over choosing presidential candidates. Even without a falteringparty system, it is possible for politicians with a strong record of subnational gov-ernment management to rise to prominence within the ranks of presidential can-didates. Finally, in the Bolivian case, where political parties remained both strongand in control of the presidential (and mayoral) nomination process through2002, there was little change from the traditional career trajectory to presidentialcandidacy. Even with significantly strengthened local governments, popularmayors have not been able to make the jump to successful presidential candida-cies. This seems to suggest that strong party control over nomination plays a keyrole. It is too early to draw anything concrete from this experience, however.

At the same time that experience as a mayor or governor has become a cov-eted credential in a presidential run, the offices of mayor and/or governor havebecome more attractive in themselves. The increase in fiscal resources availableto officials serving in state and local government has made positions at these sub-national levels more attractive to ambitious politicians. According to a recentstudy (Campbell 2003, 12), “a new generation of leaders sought and won office”as a result of decentralization. Looking at a set of Latin American countries thatis broader than the Andean region, he notes that “mayors are four times morelikely to have professional backgrounds than a decade ago” (Campbell 2003, 3).In addition, while better educated and more accomplished individuals were be-ing drawn to local and regional government, they were also professionalizing thebureaucracies at this level. A World Bank study (1995) of Colombia documentsa steep decline in the ratio of employees to professionals in fifteen randomly


selected municipalities between 1988 (the first direct election of mayors) and1994. In almost all of the cases, this ratio dropped by more than half during thissix-year period; in dramatic cases, it fell from 62.3 to 7.3 and from 52 to 5.2(World Bank 1995, 19).

While a well-educated and professional workforce should improve local gov-ernment, the attraction of a more professional workforce to more powerful localand regional offices may also have made it more difficult for typically excludedgroups to win subnational offices. In Bolivia, one of the striking effects of in-creasing the power of local governments has been the decrease in the number ofwomen elected to serve in local governments. In the 1993 municipal elections,which occurred before the Law of Popular Participation and which elected a to-tal of 858 city councilors, 231 women were elected, representing roughly 27 per-cent of the total. In 1995, after Popular Participation, only 135 women wereelected out of a total of 1,625 councilors—representing only 8.3 percent of allcouncilors elected (SNPP 1996, 10).

This picture becomes even more complex if one examines the electoral suc-cess of indigenous and peasant candidates in Bolivia’s local elections. Based ondata from a series of questionnaires undertaken in 1996 and 1999, Xavier Albó(1999) notes that 464 indigenous and peasant candidates won election in 170municipalities in the 1995 elections, as either primary candidates or as alternates(suplentes). Indeed, 55 percent of municipalities elected at least one indigenousor peasant candidate as a member of the council or as an alternate (Albó 1999,16, 22).

Taken together, these signs point to the complex reality of representation atthe local level in decentralizing countries: access to powerful local offices leadsto an influx of talent and an immediate increase in the quality of the candidatesfrom the perspective of education and professional development. It may or maynot lead to an opening of public positions to a broader range of society. In Bo-livia, women have not fared well in the new environment, while indigenous andpeasant candidates appear to be doing quite well. To some extent, local govern-ments are becoming more professional and also more broadly representative ofthe wider public. These findings tend to support the idea that decentralizationimproves representation and also the ability of local government to provide bet-ter services than they had previously.

By Way of Conclusion

There is no simple answer to the question, Has decentralization improved dem-ocratic representation in the Andes? Looking across the three categories of dem-ocratic representation explored here—voter turnout, the ability of traditionalparties to win subnational elections over time, and the effect of decentralizationon independent presidential candidacies—decentralization’s effect on the crisis

196 O’Neill

of democratic representation appears mixed. However, mostly the crisis of dem-ocratic representation seems to have occurred despite decentralization, ratherthan because of it.

While a crisis of democratic representation is characterized by low voterturnout, the turnout in local and regional elections has been relatively highwhere subnational officials enjoy real access to power. Moreover, public opinionpolls show most Andeans optimistic in their outlook toward subnational gov-ernments. A crisis of democratic representation is also characterized by highparty volatility across elections; while this is true at the national and subnationallevels, traditional parties seem to do slightly better in subnational than in nationalelections, suggesting that decentralization is not the cause of eroding support fortraditional parties. Finally, independent presidential candidacies are hypothesizedto be a hallmark of a crisis of democratic representation. Since decentralizationprovides a base of experience and public exposure from which independentsmight launch a presidential bid, decentralization would seem to play a support-ing role in the crisis of democratic representation. The results in this area aremixed, however. While more candidates are running for president from sub-national executive positions in the Andean countries, many independent cam-paigns have not been borne from subnational experience and thus one might ex-pect independent candidacies from nontraditional sources even in the absence ofdecentralization. For example, it is notable that the nontraditional résumés ofseveral presidents and presidential candidates in Venezuela (Chávez), Peru (Fuji-mori, Toledo), and Ecuador (Gutiérrez) have not included subnational positions.This suggests that the trend toward nontraditional presidential candidacies, whileinfluenced by the availability of subnational positions within the public eye, hasoccurred despite decentralization in many countries.

Exculpating decentralization as the primary cause of a crisis of democraticrepresentation does not mean its effects on democratic representation have beenuniformly positive. On the contrary, its effects are mixed. In some places, wheresubnational governments are freely elected and have access to significant fiscalresources; where elections create a level of competition that leads to account-able public servants; or where well-educated professionals are drawn into therace and win, democracy has been improved. Elsewhere, where subnationalgovernments control few resources; where they must depend on the centralgovernment’s favor for the disbursement of those resources; where elections re-main controlled by strong, undemocratic forces (be they large landowners orarmed guerrilla movements); and where capable candidates are kept out of therace through institutional rules, the machinations of hegemonic parties, or byother means, democracy has not been improved by decentralization.

In general, the idea of decentralization remains popular in the region, regard-less of what the outcomes have been across countries. Polling evidence suggeststhat respondents make a distinction between their favor or disfavor of particular


subnational administrations (which varies widely over place and time) and theirsupport for decentralizing reforms (which is strong almost everywhere).

Not only do Andean citizens pay lip service to the concept of decentraliza-tion, they turn out to vote in large numbers, in many countries matching theparticipation in national contests. While subnational elections that are concur-rent with national elections elicit a greater turnout, the number of people thatturn out even in staggered subnational elections is often a high percentage of thenumber who turn out in national contests proximate in time. While this is gen-erally true, different countries experience different patterns in their subnationalturnout over time. Where subnational governments are weak, as in Ecuador,participation at this level grows at a declining rate. In Bolivia, where local gov-ernments are perceived to be unstable and only indirectly accountable, turnouthas also declined. Increasing subnational power seems to correlate with sharp in-creases in subnational turnout, as in Venezuela and Colombia.

At the same time that citizens are turning out to vote, they are often votingfor nontraditional parties in subnational contests, bringing new voices into thepolitical arena. The only exception to this trend is Bolivia, where independentcandidacies were not allowed in elections before 2002. In all other Andean coun-tries, traditional parties have seen their ability to win subnational contests declinesince decentralization was instituted. Where traditional parties have declinedprecipitously in national contests, the decline at subnational levels has often beenless severe than at the national level (Venezuela and Peru). This suggests a slowopening to new voices at the subnational level. Colombia stands out as a casewhere the traditional parties have maintained electoral strength in both nationaland subnational contests, perhaps suggesting that decentralization has had thedesired effect of generating information from base constituencies that has al-lowed political parties to change with the times and revitalize themselves.

Finally, decentralization has created a layer (sometimes two) of powerfulelected offices that has attracted a growing number of well-trained and ambitiouspoliticians. Not only has this increased the capacity of these governments and ar-guably improved public service delivery, it has also changed the career paths ofpoliticians who seek the presidency. In countries where governors and mayorswield significant resources, mayoral and gubernatorial positions allow politiciansthe opportunity to build a strong record of achievement that can be turned intoa powerful credential in seeking the presidency. As a result, the number of pres-idential hopefuls that have high-profile gubernatorial or mayoral experience ontheir résumés has climbed dramatically in recent years. This is particularly no-table in Venezuela and Colombia, but extends throughout the region. In Peru,where subnational governments have been somewhat weak and became weakerin the 1990s, the number of former mayors running for president and pollingmore than 5 percent of the vote peaked in 1985 and then declined. Furthermore,all of the candidates with subnational experience had been mayors of Lima; in

198 O’Neill

Colombia and Venezuela, presidential contenders came from a variety of gu-bernatorial and mayoral positions.

The relationship between decentralization and political representation iscomplex. Where decentralization has created strong subnational governments,it has largely improved democratic participation and representation; where it hasbeen weak, it has not done as much good. Decentralization has attracted morequalified managers to subnational government positions and given them incen-tives to serve their constituents. At the same time, these reforms have increasedthe ability of independents and dissident party members with high-profile sub-national government experience to launch presidential campaigns that weakenthe ability of national parties to create and promote policy coherence over time.This may be the largest unforeseen danger of decentralization to the operationof democratic politics.

Notes

1. There is a large literature in economics. For a good recent review article, see Ter-Minassian (1999).

2. Not all economists expected decentralization to yield uniformly good results onthe fiscal side. Notable criticisms include Prud’homme (1995) and Tanzi (1994).

3. Similarly, not all scholars of politics expected decentralization to be uniformly pos-itive. Dahl (1971), for instance, worried about the effects on democracy if subnationalidentities became an important cleavage in national politics.

4. Perhaps this is because political outcomes are harder to measure or because it is ex-pected that decentralization’s effects must be felt in the longer term.

5. This chapter will not explore the fiscal and economic consequences of decentral-izing reforms in this region. A great deal of scholarship has begun to focus on this area;many scholars find that fiscal decentralization as practiced in the region has adversely af-fected fiscal balance at the national level.

6. For an excellent discussion of “success” cases in Latin America, see Campbell (2003).7. Attempts to typologize types of decentralization through assigning particular

meanings to words such as “devolution,” “deconcentration” (Rondinelli 1981), and soforth have not been widely embraced.

8. Allowing for the popular election of mayors and/or governors and municipal, pro-vincial, or regional councils occurs as a discrete event and is therefore easy to locate tem-porally. Increases in the fiscal resources available to elected subnational governments area bit trickier to differentiate: How much money needs to be available to these govern-ments to make them meaningful? How would we define “autonomous” fiscal power? I include a fiscal component in this definition because electing officials who are com-pletely dependent upon fiscal resources controlled by the center effectively severs the ac-countability to local constituents created by the local election of these officials. Insteadof trying to define how many fiscal resources is “enough” to consider a reform as “de-centralizing,” I will simply refer to reforms that allow for greater resources to be


transferred to subnational governments as “more fiscally decentralized,” and those thatallow for fewer resources to be transferred as “less fiscally decentralized.” See IDB (1997)for a more careful treatment of these distinctions.

9. In practice, this has led to a high rate of turnover at the local level.10. This figure was obtained from a conversation with Jonas Frank, Ph.D. candidate

and worker at a Quito-based NGO, October 2000.11. In 1999–2000, debate over decentralization shifted from increasing transfers to

provincial and municipal governments to a debate over provincial autonomy, particularlyfocused on the potential autonomy of Guayas, the coastal province that includes the cityof Guayaquil. Led by former president and PSC member Febres-Cordero, the movementsucceeded in getting the issue on a national referendum held during the summer of 2000.Sufficient support was not gained for this move, which would have turned tax bases inGuayas that had formerly contributed to the national tax base into the sole ownership ofthe state, disadvantaging other states in the country that are net recipients of national taxredistribution.

12. Bernalillo County, which includes the city of Albuquerque, New Mexico.13. Beginning in 1994, presidential elections went to a two-round runoff system, but

up until then, they had occurred in a single round of balloting. My choice of looking at first- versus second-round turnout in national elections ends up making very little difference.

14. The indirect nature of elections and the lack of independent candidates have beencited as reasons for low turnout by Molina, García, and Landívar (2001, 14 –15). This law,barring independent candidacies, was changed early in Sánchez de Lozada’s second term.

15. Comparing 1998 national turnout and 1997 municipal turnout (instead of 2000municipal turnout, due to the unusual nature of those elections).

16. Comparing national and municipal results from 2000.17. Comparing 1998 municipal elections with 2000 national elections.18. Comparing 1993 municipal elections with 1995 national elections.19. Comparing 1997 national elections with 1999 municipal elections.20. MNR (Movimiento Nacional Revolucionario), MIR (Movimiento de Izquierda

Revolucionario), and the ADN (Acción Democrática Nacionalista).21. CFP (Concentración de Fuerzas Populares), PSC (Partido Social Cristiano), ID

(Izquierda Democrática), and the PRE (Partido Roldosista Ecuatoriano).22. AD (Acción Democrática) and COPEI (Comité de Organización Política Elec-

toral Independiente).23. APRA (Alianza Popular Revolucionaria Americana) and AP (Acción Popular).24. See “Partidos cuadruplicaron los votos de las agrupaciones,” December 23, 2004,

http://bolivia.com/noticias/AutoNoticias/DetalleNoticia24248.asp.25. My focus here is on the presidency, but this analysis could be extended to cover

career paths to key legislative or cabinet positions.

References

Albó, Xavier. 1999. Ojotas en el poder local: Cuatro años después. La Paz, Bolivia: CIPCA(Centro de Investigación y Promoción del Campesinado) and PADER (Proyecto deApoyo al Desarrollo Económico Rural).

200 O’Neill

Alesina, Alberto, Alberto Carrasquilla, and Juan José Echavarría. 2002. “Descentral-ización en Colombia.” In Reformas institucionales en Colombia: Una agenda reformista para los desafíos del nuevo siglo, ed. Alberto Alesina, 95–134. Bogotá: Fedesarrollo, Alfaomega.

Alesina, Alberto, Ricardo Hausmann, Rudolf Hommes, and Ernesto Stein. 1999. “Bud-get Institutions and Fiscal Performance in Latin America.” Journal of Development Economics 59, no. 2: 253–73.

Barr, Robert. 2001. “Parties, Legitimacy, and the Motivations for Reform: Devolutionand Concentration in Latin America.” Paper delivered at the 2001 Annual Meeting ofthe American Political Science Association, San Francisco, August 30 –September 2.

Bird, Richard M. 1990. “Fiscal Decentralization in Colombia.” In Decentralization, LocalGovernments, and Markets, ed. Robert J. Bennett, 394 – 410. Oxford: Clarendon Press.

Buller, Eduardo. 1993. “Regionalización y municipalización en el proceso de descen-tralización administrativa: El caso de Peru.” In Descentralización y gobiernos municipales,ed. CORDES, 135–77. Quito: Corporación de Estudios para el Desarrollo.

Cameron, John. 2000. “Municipal Decentralization and Peasant Organization inEcuador: A Political Opportunity for Democracy and Development.” Paper presentedat the Latin American Studies Association Conference, Miami, March.

Campbell, Timothy. 2003. The Quiet Revolution: Decentralization and the Rise of PoliticalParticipation in Latin American Cities. Pittsburgh: University of Pittsburgh Press.

Dahl, Robert. 1971. Polyarchy: Participation and Opposition. New Haven, CT: Yale University Press.

Dammert Ego Aguire, Manuel. 1999. Desborde territorial descentralista: Replanteando la reforma descentralista peruana: Territorios sociales, estado con regiones y municipios. Self-published.

de la Cruz, Rafael. 1992. “La estrategia de la descentralización en Venezuela.” In De-scentralización, gobernabilidad, democracia, ed. Rafael de la Cruz, 17–73. Caracas: NuevaSociedad.

Delgado Silva, Angel. 1995. Municipios, descentralización y democracia: Una propuestademocrática. Lima: Servicios Gráficos.


Diamond, Larry, with Svetlana Tsalik. 1999. “Size and Democracy: The Case for De-centralization.” In Developing Democracy: Toward Consolidation, by Larry Diamond,117– 60. Baltimore: Johns Hopkins University Press.

Eisenstadt, Todd. 1999. “Electoral Federalism or Abdication of Presidential Authority?Gubernatorial Elections in Tabasco.” In Subnational Politics and Democratization in Mex-ico, ed. Wayne Cornelius, Todd Eisenstadt, and Jane Hindley, 269–93. La Jolla, CA:Center for U.S.-Mexican Studies, University of California at San Diego.

Fisman, Raymond, and Roberta Gatti. 2000. “Decentralization and Corruption: Evi-dence across Countries.” World Bank Policy Research Working Paper, no. 2290.World Bank, Washington, DC.

Fox, Jonathan. 1994. “Latin America’s Emerging Local Politics.” Journal of Democracy 5,no. 2: 105–16.

Garman, Christopher, Stephan Haggard, and Eliza Willis. 2001. “Fiscal Decentraliza-tion: A Political Theory with Latin American Cases.” World Politics 53, no. 2: 205–36.

Grindle, Merilee. 2000. Audacious Reforms: Institutional Invention and Democracy in LatinAmerica. Baltimore: Johns Hopkins University Press.

Grupo Propuesta Ciudadana. 1996. Descentralización, participación ciudadana y reforma delestado: Encuesta en cinco departamentos del Perú. Lima: Grupo Propuesta Ciudadana.


Huther, John, and Anwar Shah. 1998. “Applying a Simple Measure of Good Governanceto the Debate on Fiscal Decentralization.” World Bank Policy Research Working Pa-per, no. 1894. World Bank, Washington, DC.

IDB. 1997. Economic and Social Progress of Latin America. Washington, DC: Inter-AmericanDevelopment Bank.

IINCIP. 1998. Encuesta de percepción política interna y externa en la ciudad de La Paz. La Paz:Instituto de Investigaciones en Ciencia Política, Universidad Mayor de San Andrés.

Instituto de Encuestas, Universidad Católica Boliviana. 1994. Encuesta de percepción políticaV: Ciudad de La Paz. La Paz: Fundación Hanns-Seidel.

Kay, Bruce. 1995. “Fujipopulism and the Liberal State in Peru, 1990 –1995.” ChapelHill: Duke University/ University of North Carolina Program on Latin AmericanStudies.

Kraemer, Moritz. 1999. “One Decade of Decentralization—An Assessment of theVenezuelan Experiment.” Unpublished paper, Inter-American Development Bank,Washington, DC.

Molina, Carlos Hugo, Martha García Ferrufino, and María Elisa Landívar. 2001. Evalua-ción de los consejos departamentales y provinciales: Estudios de caso. La Paz: Friedrich EbertStiftung–Instituto Latinoamericano de Investigaciones Sociales.

Musgrave, Richard A. 1959. The Theory of Public Finance: A Study of Public Economy. NewYork: McGraw-Hill.

Oates, Wallace E. 1998. The Economics of Fiscal Federalism and Local Finance. Northampton,MA: Edward Elgar Publishing.

O’Neill, Kathleen. 2003. “Decentralization as an Electoral Strategy.” Comparative Politi-cal Studies 36, no. 9: 1068–91.

———. 2005. Decentralizing the State: Elections, Parties, and Local Power in the Andes. NewYork: Cambridge University Press.

Prud’homme, Rémy. 1995. “The Dangers of Decentralization.” World Bank Research Ob-server 10: 201–20.

Rodden, Jonathan. 2002. “The Dilemma of Fiscal Federalism: Grants and Fiscal Perfor-mance around the World.” American Journal of Political Science 46, no. 3 ( July):670 –87.

Rondinelli, Dennis. 1981. “Government Decentralization in Comparative Perspective:Theory and Practice in Developing Countries.” International Review of AdministrativeScience 47: 133– 45.

Rubinfeld, Daniel L. 1987. “The Economics of the Local Public Sector.” In Handbook ofPublic Economics, vol. 2, ed. Alan A. Auerbach and Martin Feldstein, 571– 645. NewYork: North-Holland.

Semana. 1990. “La gran encuesta del 89.” Semana, no. 399– 400, December 26,1989–January 8, 1990, 83–97.

———.1991. “La gran encuesta.” Semana, no. 453, January 8–15, 1991, 20 –31.———. 1996. “La gran encuesta del 96.” Semana, no. 714, January 9–16, 34 –39.SNPP. 1996. Las primeras elecciones: Directorio de alcaldes y concejales de la participación popu-

lar. La Paz: Unidad de Investigación y Análisis, Secretaría Nacional de Participación Popular.

Stein, Ernesto. 1998. “Fiscal Decentralization and Government Size in Latin America.”Working Paper, no. 368. Inter-American Development Bank, Washington, DC.

Tanaka, Martín. 2002. “La dinámica de los actores regionales: El despertar del Letargo?”Unpublished paper, Instituto de Estudios Peruanos, Lima.

Tanzi, Vito. 1994. “Corruption, Governmental Activities, and Markets.” IMF WorkingPaper, no. 94/99, International Monetary Fund, Washington, DC.

202 O’Neill

Tendler, Judith. 1997. Good Government in the Tropics. Baltimore: Johns Hopkins Univer-sity Press.

Ter-Minassian, Teresa. 1999. Fiscal Federalism. Washington, DC: International MonetaryFund.

Thedieck, Franz, and Eduardo Buller. 1995. “Descentralización de la administración enel Peru.” In Descentralizar en América Latina? ed. Sociedad Alemana de CooperaciónTécnica /Programa de Gestión Urbana, 195–236. Quito: Sociedad Alemana de Co-operación Técnica /Programa de Gestión Urbana.

Tiebout, Charles M. 1956. “A Pure Theory of Local Expenditures.” Journal of PoliticalEconomy 5: 416 –24.

Treisman, Daniel. 1999. “Decentralization and Corruption: Why are Federal States Per-ceived to be More Corrupt?” Paper presented at the Annual Meeting of the Ameri-can Political Science Association, Atlanta, GA, August.

Willis, Eliza, Stephan Haggard, and Christopher Garman. 1999. “The Politics of De-centralization in Latin America.” Latin American Research Review 34: 7–56.

World Bank. 1995. Local Government Capacity in Colombia: A Country Study. Washington,DC: World Bank.


The Andean democracies share recent efforts to redesign their politicalinstitutions. Disappointment and frustration with existing institutional arrange-ments led Bolivia, Colombia, Ecuador, Peru, and Venezuela to recast the struc-tures governing their policymaking processes. Citizens perceived politicians to beself-serving and the policymaking process to be opaque and corrupt. What ismore, major issues confronting the countries, such as economic stagnation andeven open armed conflict, seemed to drag on without an effective governmentalresponse. Reformers sought to reshape the terms of the relationship between gov-ernments and citizens by adopting new constitutions. Legislatures have figuredprominently in these efforts to refine how representation works.

Despite fears of excessive power concentrated in the hands of presidents,legislatures have proven to be important in determining policy outcomes acrossLatin America (Mainwaring and Shugart 1997; Morgenstern and Nacif 2002;Johnson and Crisp 2003). However, legislators receive little trust from the pub-lic, and their behavior is often singled out as part of the justification for institu-tional reform. What role did legislatures play in provoking institutional over-hauls? Is there even preliminary evidence that the overhauls have changed the waylegislators carry out representation? If the way legislators behave in office is in parta function of the institutions through which they are nominated and elected,what does the experience of the Andes in the 1990s tell us about “engineering”representation?

The relationships between parties and among members of the same partisandelegation tend to take extreme forms in Andean legislatures. The way in whichlegislators went about representation helped motivate reformers. Rather thanbalancing national and parochial concerns, Andean legislators either blindlyfollowed national party elites but failed to establish any personal connectionwith the voters who elected them, or they acted as purveyors of particularistic

7

The Nature of Representation in Andean Legislatures and Attempts at

Institutional Reengineering

❖

Brian F. Crisp

Nature of Representation and Institutional Reengineering 205

rewards within their districts but failed to take positions on issues of national im-portance. Unfortunately, episodes of institutional reform often target the symp-toms of extreme behavior rather than its root cause—the electoral incentives oflegislators. As a result, new constitutions and electoral systems have done rela-tively little to change legislator behavior. Because of their failure to yield tan-gible results, episodes of constitutional reform have generated disillusionmentabout the efficacy of institutional engineering and the ability to modify the na-ture of representation.

Legislatures and Representation

Before we can adequately trace this chain of events, we need to consider morefully the concept of “representation” itself. When conceptualizing the link be-tween voters and their elected representatives, scholars have emphasized the ideaof responsiveness (Pitkin 1967). Eulau and Karps (1977) elaborated that respon-siveness to the substantive interests of constituents could take four forms: policyresponsiveness, allocations responsiveness, service responsiveness, and symbolicresponsiveness. What constitutes the substantive interests of constituents has beencaptured on many different dimensions, the left/right ideological continuum be-ing the most common. An important dimension of substantive interests—andrepresentation of them—that is often overlooked is what we might call the “na-tional/parochial continuum.” One of the most perplexing issues surroundingrepresentation is that legislators are almost always elected by a geographicallydefined sector of the population to represent its interests and yet their job as rep-resentatives is to govern the nation as a whole. This, according to Pitkin, is the“classical dilemma” of representation (1967, 215). Pitkin argues that the alterna-tives are not mutually exclusive: a representative should be responsive to bothparochial and national concerns (218). However, institutional designs are notneutral on the question of which strategy—focusing on parochial concerns ornational concerns—is the most efficient or rational manner for legislators to getreelected.

In the abstract, the median voter may have some mix of preferences for poli-cies targeted at the nation as a whole, at a particular geographic region or socio-economic sector, or at a particular locality or individual entity (Taylor-Robinsonand Diaz 1999; Crisp et al. 2004). Yet, the representation options offered by com-peting politicians may gravitate toward a single point on the national/parochialcontinuum or dimension. Where the electoral system is party-centered, partisanlegislative delegations will behave in a disciplined fashion in the pursuit of na-tional, programmatic goals that efficiently enhance the reputation of the party asa whole. On the other hand, where the system is more candidate-centered, indi-vidualistic members of Congress will focus on targeted or parochial issues that canenhance their individual reputations.

206 Crisp

The institutional incentives that determine whether a system is likely to berelatively more party- or candidate-centered—or the “intraparty” dimension oflegislative representation (Shugart 2001)—include the process by which oneachieves access to the ballot, the nature of the ballot itself, the degree of vote pool-ing, and/or the level at which votes are cast (Carey and Shugart 1995). The po-tential for variation even in the Andean cases is quite extensive when we considerthat “across cases” might mean not only across countries but also across cham-bers, across members of the same chamber elected under mixed-member rules orfrom districts of different magnitude, and across parties with varying candidateselection procedures competing under any of these rules.

The intraparty dimension of legislative politics is central to the concept of rep-resentation because it helps determine the policy goals—national to parochial—representatives will pursue. For example, if a voter wants a publicly fundedneighborhood clinic from her national government, the likelihood of obtaininga clinic in her particular neighborhood, and thus feeling well represented, is inlarge part a function of whether legislators are motivated to provide particular-istic goods to a well-defined constituency that can determine their prospects forrenomination and reelection. Where the electoral system is party-centered, thelegislature might take up serious questions related to healthcare policy, but it isrelatively less likely that individual legislators will feel beholden to delivering thebenefits of that policy to a particular constituency.

Even where elected representatives are disciplined members of a partisan leg-islative delegation, their ability to deliver on the programmatic, national cam-paign promises that enhanced the reputation of their party will obviously be afunction of that party’s size in Congress. The effective number of parties—or theinterparty dimension of legislative representation—and relatedly the size of anygiven party are conditioned by institutional characteristics including the seat al-location formula and district magnitudes. The timing of presidential electionscan also influence whether members of the successful candidate’s party can ridehis coattails into office. Thus, a party may have a clearly articulated ideologicalprogram but remain insufficient in size to implement it. In other words, it mayfaithfully represent the substantive interests of its constituents but not be able todeliver policy or allocation responsiveness, given its less than majority status(though legislators may still be able to deliver effectively on service and symbolicresponsiveness while in the minority). The size of partisan delegations seems lessimportant where intraparty institutional incentives encourage legislators to focuson their individual reputations. Where individual legislators behave in an entre-preneurial fashion, the number of party labels is unlikely to effect how they carryout the task of representation because members of the same delegation would beless likely to behave in a homogenous fashion—diminishing the usefulness ofconsidering “party” as a unit of analysis.


In order to evaluate the importance of legislative politics and the national/parochial continuum of representation on the demand for and effectiveness ofinstitutional overhaul, I will proceed as follows. I will operationalize the intra-party and interparty dimensions of legislative politics and describe the pre-reform Andean legislatures in these terms. In a third section, I will characterizethe constitutional reforms undertaken in the five countries, focusing partic-ularly on the institutional characteristics that promote party-centered or candidate-centered behavior among legislators. Next, I will evaluate whetherthe reforms undertaken led legislators to carry out representation differently inthe earliest case of constitutional reform—Colombia. After showing that legis-lator behavior varied only modestly as a result of political reform, I look at thescant evidence available on the impact of institutional reform on public opinionregarding support for democracy. I will conclude by drawing some lessons re-garding the efficacy of institutional engineering in the Andes and the potentialfor disillusionment that failed reform attempts may generate.

Pre-Reform Andean Legislatures

In carrying out representation, legislators can tend exclusively to the particularneeds of their individual districts; they can obediently support the party line re-garding major policy initiatives; or they can combine these two sets of prioritiesto offer a more balanced form of representation. Assuming that most politicianswant to perpetuate their careers, the way in which they engage in represen-tation—or balance national and parochial concerns—will in large part be a func-tion of the institutional rules that structure their prospects for reelection, elec-tion to another office, or even appointment to a political post. These rules governthe relations within parties—the intraparty dimension of representation—andthe relations among parties—the interparty dimension of representation.

The Intraparty Dimension of Representation

Intraparty politics is central to the way representation is conducted. When par-tisan delegations are relatively unified, they can promise broader programmaticpolicies and expect to be able to deliver on those promises if voters give them alegislative majority. However, where partisan delegations are only loose federa-tions of reelection-seeking individuals, they cannot consistently deliver on prom-ises that require unity on a single or limited number of issues.1 Instead, legislatorsfrom all parties are more likely to “logroll” and deliver particularistic rewards totheir individual constituencies. The cooperation that does occur in such legisla-tures will assure that individual constituencies get served (and that legislators de-liver legislative “pork” for which they get credit), but this form of representationdoes not revolve around intraparty cooperation and interparty conflict.

208 Crisp

Candidate selection procedures and general election rules serve as incentivestructures for legislators seeking reelection (see Table 7.1). Where party leaderscontrol access to the ballot under their party’s label, legislators are more likely tobehave as members of a cohesive unit. They must be responsive to the leaderswho exercise control over their future opportunities to run. On the other hand,where decentralized procedures, such as primaries or collecting a minimal num-ber of signatures, are involved, the incentive to be a disciplined member of aparty faction is diminished. While parties vary within a nation, major parties of-ten share candidate selection procedures. Most parties in Bolivia, Peru, and Ven-ezuela used elite-dominated nomination procedures, while in Colombia andEcuador party leaders did not restrict the use of their labels. In terms of generalelection rules, we must examine at what level votes are cast and to what levelthey pool. In Bolivia, Ecuador, and Venezuela voters could cast only one ballotfor a closed party list, and the votes were pooled at the level of the party. InColombia and Peru, on the other hand, it was possible to distinguish among can-didates from the same party. In Peru this was accomplished with open lists andin Colombia with sub-party lists. In Peru, votes were pooled to the level of theparty, while in Colombia they were only pooled to the level of the sub-party list.

Looking at Table 7.1, we see that there were three systems that did virtuallynothing to encourage legislators to balance national and parochial concerns whenconceiving of how best to represent. Bolivia and Venezuela were characterizedby intraparty centralization—party leaders dominated the candidate selectionprocess in most parties and voters could not disturb those party-prepared slates.Venezuelan parties in particular were known for their high levels of party disci-pline (Rey 1972; Coppedge 1994; Crisp 2000). Party leaders in Congress couldnegotiate among themselves and with the executive with every confidence theycould deliver the votes of their copartisans. Personalism and pork barrel politicsdid not plague budgetary decision making. On the other hand, citizens feltvirtually no connection to individual representatives. The move to a mixed-member system in 1993 was heralded as “personalization” because the single-member districts (there were a few multi-member districts in the nominal tier)

Table 7.1

Pre-Reform Intraparty Characteristics

Centralized Party-line Party-level Intrapartynomination? voting? vote pooling? incentives

Bolivia yes yes yes hyper-centralizedColombia no no no hyper-personalisticEcuador no yes yes not extremePeru yes no yes not extremeVenezuela yes yes yes hyper-centralized

SOURCE: Shugart, Crisp, and Moreno (2002).


were meant to more directly connect voters to legislators. However, the movewas too little and/or too late to stave off a major revamping of the governmentalinstitutions and party system.

Colombia anchored the other end of the spectrum, giving legislators little in-centive to think about the reputation of their parties when engaging in repre-sentation. Use of party labels went virtually unrestricted, and voters could choosefrom among multiple sub-party lists in every district. Partisan delegations inCongress were notably undisciplined, and patron-client links with voters werecultivated through the distribution of targeted pork-barrel programs. Both rep-resentatives and senators spent much more of their time traveling to maintain re-lations with the constituents (Ingall and Crisp 2001; Crisp and Desposato 2004).In Eulau and Karps’s terms, where parties were weak, legislators emphasized al-locations, service, and symbolic responsiveness relative to policy responsiveness.Presidents had to cobble together coalitions by negotiating with sub-party fac-tions and individual legislators. This coalition building came at the expense ofdiluting programmatic goals.

The Interparty Dimension of Representation

When we look at relations among partisan delegations in Andean legislatures,Ecuador stands out for its relatively high effective number of parties—the num-ber of parties weighted by their size. Nonconcurrent presidential and legislativeraces and the Hare allocation formula contributed to a single chamber withmore partisan actors. Colombia and Peru, on the other hand, had an average ef-fective number of parties less than 2.5 (see Table 7.2). All other things beingequal, it should be easier to construct legislative majorities where there are fewerparties, though this might come at the expense of “representativeness”—theproportionality between votes and seats.

Polarization, the dispersion of the vote away from the relative center of theparty system on a left–right continuum, was greatest in Ecuador, with Boliviaand Venezuela also showing a good deal of ideological diversity. The index rangesfrom �1 to 1 and “can reach its maximum only when half of the vote goes to theright and half to the left; if all of the vote went to just one extreme, polarizationwould be zero because the relative center would be the extreme as well and therewould be no dispersion” (Coppedge 1998, 557). Not surprisingly, there wasclearly a strong relationship between the effective number of parties, size of thepresident’s party, and ideological polarization. As the effective number of partiesincreases, the size of the president’s party (or any single party) generally decreases(R2 � �.93). Where the effective number of parties is high, so is the degree ofpolarization (R2 � .78). Thus, legislatures like the ones in Colombia and Peruwere characterized by a relatively low effective number of parties, little ideolog-ical diversity, and relatively great support for the executive. Ecuador’s legislature

210 Crisp

held down the other end of the interparty spectrum. Venezuela and Boliviafell in between, with Venezuela more closely resembling the concentrated casesand Bolivia more closely approximating the dispersion in Ecuador. Bolivia andColombia, on average, had the most conservative legislatures, while Peru andVenezuela had relatively leftward-leaning mean ideology scores (Coppedge1997, 1998). In sum, in strictly interparty terms the Andean legislatures werequite diverse. Institutional rules interacted with underlying preference structuresto generate variations in the partisan composition of legislative chambers acrossthe region.

Extreme Electoral Systems and Representation

Shugart has juxtaposed the interparty and intraparty dimensions of electoralsystems to capture their “efficiency”—the extent to which they “permit the ar-ticulation of policy-based electoral majorities” (2001, 28). In order to be efficient,

Table 7.2

Partisan Composition of Andean Legislatures Prior to Constitutional Reform

Effective number Mean Ideological Size of the of parties ideology polarization president’s party

Bolivia 3.98 .30 .44 32.831985–1989 4.32 .36 .54 33.101989–1993 3.92 .24 .58 25.401993–1997 3.71 .31 .19 40.00

Colombia 2.21 .18 .14 50.071982–1986 1.99 .20 .13 41.201986 –1990 2.45 .17 .17 49.201990 –1992 2.18 .16 .12 59.80

Ecuador 5.62 .02 .55 25.601979–1984 3.63 .13 .51 43.301984 –1986 5.77 �.05 .56 12.901986 –1988 7.39 �.07 .62 20.401988–1990 3.79 �.23 .43 43.101990 –1992 6.55 .00 .61 18.301992–1996 6.61 .35 .54 15.60

Peru 2.39 �.22 .39 56.901980 –1985 2.46 .07 .51 54.401985–1990 2.31 �.50 .26 59.40

Venezuela 3.16 �.18 .45 40.081979–1984 2.65 �.11 .51 42.201984 –1989 2.42 �.23 .44 56.501989–1994 2.83 �.17 .50 48.301994 –1999 4.74 �.22 .36 13.30

Rest of Latin 3.25 .02 .35 42.20America

SOURCES: Johnson and Crisp 2003; my own calculations based on data available at http://www.electionworld.org/.


electoral systems must institutionalize incentives that avoid the extremes of being hyper-representative and pluralitarian on the interparty dimension andhyper-centralized and hyper-personalistic on the intraparty dimension. In intra-party terms, hyper-centralized systems give individual legislators virtually no in-centive to cultivate a personal connection to constituents and instead leave themexcessively obedient to party leaders. Party labels rather than individual deedsand reputation constitute the core of the representative connection. A hyper-personalistic system does just the opposite. A hyper-representative system is onethat encourages a high effective number of parties and therefore relies on post-election bargaining to form governing coalitions. A pluralitarian set of institu-tions, on the other hand, frequently translates a plurality of popular support intoa clear governing majority.

Using a very simplified version of Shugart’s scheme, I depict the pre-reformincentives of Andean legislators on both the interparty and intraparty dimen-sions. These are very rough placements, but they capture some of the diversityjust discussed (see Figure 7.1).

While Peru and Colombia on average had the lowest effective number of par-ties in the Andes, they did not turn a plurality of support into a majority of seats—avoiding an extreme classification on the interparty dimension. In interparty

Figure 7.1Interparty and intraparty incentives of legislators in the pre-reform

Andean countries

Hyper-personalistic

Hyper-centralized

Venezuela

Hyp

er-r

epre

sent

ativ

e

Plur

alita

rian

Bolivia

Ecuador

Peru�Colombia

212 Crisp

terms, Ecuador was exceptional for never generating a majority party in the leg-islature, and therefore requiring post-electoral bargaining to generate policy. Thepresident’s party averaged less than 26 percent of the seats and never had morethan 43 percent. In intraparty terms, Colombia and Venezuela stand out foropposite reasons. Colombia was exceptionally personalistic, while Venezuela wasextremely party-centered. In sum, no legislature approached “efficient” repre-sentation at the center of Figure 7.1.

In intraparty terms, where party leaders exercised nearly ironclad disciplineover their legislative delegations, we would expect legislators to focus almost ex-clusively on national concerns that would enhance the reputation of the party asa whole. Conversely, where a lack of discipline results from legislators’ need tothink about their personal reputations, representation or responsiveness willgravitate toward the parochial end of the spectrum. Given the rather extremenature of representation as it was carried out in the Andes, we might expectconstitutional reforms to focus on the electoral incentives of legislators. Wherepartisan delegations ignored parochial concerns—Venezuela and Bolivia—in-stitutional changes that loosened the grip of party leaders would bring the sys-tems into greater balance. On the other hand, where legislators behaved as in-dividualistic entrepreneurs focusing on pork barrel and patronage—Colombiaand to a lesser extent Peru—we might expect reforms that would encouragelegislators to mix in concern for national, programmatic issues. In the next sec-tion, I will show that institutional reforms rarely, if ever, revamped the incen-tives at the heart of the intraparty dimension of representation.

Constitutional Reforms

Reformers often focused primarily on the constitutional allocation of powersacross branches.2 Rather than adjusting the incentives of legislators themselves,reformers chose to simply strengthen the president’s hand in dealing with Con-gress. For example, the president’s powers of veto and agenda control were sig-nificantly enhanced in Ecuador. If a president vetoes legislation on constitutionalgrounds, only the judicial branch can override his veto. In addition, legislatorslost their ability to amend the amount of expenditure proposed by the presidentin his budget. In Peru, agenda powers were likewise enhanced by limiting thelegislature’s ability to introduce new taxes or add to the proposed budget. Thenew Venezuelan constitution allows the legislature to delegate decree authorityon any matter—not just economic and financial matters as was previously thecase—and it expands the president’s authority over states of exception.

There were very limited changes to the electoral incentives that wouldinfluence the intraparty dimension of representation. In Venezuela, the upperhouse was eliminated, but the lower house is elected under mixed-member rules,as has been the case since 1993. The new constitution appears to mandate the use


of primaries or caucuses for nominating legislative candidates. Such a changewould pit copartisans against one another, heighten the importance of one’s per-sonal reputation, and perhaps diminish the party-centered behavior of legislatorsin office. However, thus far, political parties have chosen to ignore the provisionwithout repercussions. Bolivia adopted a mixed-member electoral system simi-lar to the one used in Venezuela. In an effort to get the best of both worlds, thegoal was to make Congress more accountable to local constituencies throughgeographic representation while maintaining the overall proportionality of thesystem (Shugart and Wattenberg 2001). This change could diminish the hyper-centralized nature of intraparty relations if legislators elected in the nominal tierhave the liberty to respond differently to constituents than to their copartisans onclosed proportional representation lists at the department level. Two factors donot bode well for dramatic changes in behavior. First, candidate selection proce-dures were left untouched, meaning that candidates who stray too far from thewishes of the party leaders may be denied access to the ballot in the future. Whatis more, the mixed-member proportional system was not adopted by a new re-gime devoid of common practices. It was grafted on to a highly centralized sys-tem with disciplined practices. Given the existing norms, anything short of aclear signal to dramatically change one’s behavior is likely to have muted effects.

In Peru, the new unicameral Congress was elected in a single nationwide dis-trict, abolishing the regional districts of the former lower chamber. A similar re-form was adopted for the Colombian Senate. Reformers appear to have reasonedthat a nationwide district would encourage legislators to focus on programmaticissues that would generate votes across the entire country. Peru maintained itsopen list system and Colombia its sub-party list system. Given the preferencevoting, candidates are still encouraged to gather the bulk of their votes region-ally, based on their personal reputations (on Colombia, in particular, see Crispand Ingall 2002; Crisp and Desposato 2004). In Ecuador, they moved fromclosed to open list rules. Open list rules could enhance the personal vote-seekingbehavior of legislators. The pre-reform system limited the influence of partyleaders by not centralizing control over ballot access, and the adoption of an openlist allowed voters to disturb any ranking of candidates established by party lead-ers. Thus, the one case that was relatively balanced between hyper-centralizedand hyper-personalistic incentives on the intraparty dimension adopted rules thatcould skew behavior in the personalistic direction. Thus, if we were to reviseTable 7.1 to reflect “post-reform intraparty characteristics,” only one cell wouldchange. The entry for Ecuador on party-line voting would not be “no.” So,rather than balancing the need to represent national and local concerns in thefour of the five cases that were extreme, five instances of constitutional reformmanaged to take the one efficient case and make it extreme.3

The effective number of parties represented in the legislature is influenced byseat allocation formulas, district magnitude, and the pull effect of presidential

214 Crisp

races. Increasingly proportional seat allocation formulas and relatively larger dis-trict magnitudes are associated with a higher effective number of parties. Con-current presidential races can exert a downward pull on the number of parties, es-pecially where the presidential race is decided by plurality rather than majorityrunoff. When legislative delegations are likely to behave in a relatively unitary andrational manner, the effective number of parties is critical for understanding thenumber of actors who must find a policy proposal acceptable (where parties arevery undisciplined, we must think about individual legislators as the mostrelevant unit of analysis within a legislature). Electoral systems that make itvery difficult for any party to achieve a majority of legislative seats are “hyper-representative,” while those that consistently translate a plurality of votes into amajority of seats are “hyper-majoritarian.” Recall from Figure 7.1 that of the An-dean cases, only Ecuador with its large effective number of parties tended towardan extreme on the interparty dimension of representation. Very few changes weremade to Andean constitutions and electoral laws that we should expect to have asystematic impact on the partisan fragmentation of legislatures (see Table 7.3).

Table 7.3

Partisan Composition of Andean Legislatures after Constitutional Reform

Effective number Mean Ideological Size of the of parties ideology polarization president’s party

Bolivia 5.62 .20 .47 25.401997–2001 5.62 .20 .47 25.40Pre-reform average 3.98 .30 .44 32.83

Colombia 2.57 .12 .14 56.071992–1994 3.00 .04 .25 54.001994 –1998 2.61 .16 .13 53.301998–2002 2.09 .16 .05 60.90Pre-reform average 2.21 .18 .14 50.07

Ecuador 5.23 .16 .45 25.601996 –1998 5.03 .28 .42 23.201998–2000 5.43 .04 .47 28.00Pre-reform average 5.62 .02 .55 25.60

Peru 4.78 .03 .14 41.051995–2000 2.91 .03 .14 55.802000 –2005 6.64 NAa NAa 26.30Pre-reform average 2.39 �.22 .39 56.90

Venezuela 3.77 NAb NAb 46.101999–2005 3.77 NAb NAb 46.10Pre-reform average 3.16 �.18 .45 40.08

SOURCES: Johnson and Crisp 2003; my own calculations based on data available at http://www.electionworld.org/ and in Coppedge 1997.

a Six of the thirteen parties that won seats are new parties, so Coppedge’s database includes no ideologyscores for them. Four others were not new but were not scored for the 1995 elections.

b The majority of the parties are new and therefore not classified by Coppedge (including the MVR).


The move to nonconcurrent elections in Venezuela should tend to increasethe effective number of parties—the number of parties weighted by their size—by shortening the president’s electoral “coattails,” but the only post-reformCongress elected thus far was elected concurrently with the president (comparefigures in Table 7.2 and Table 7.3 for pre- and post-reform measures, respec-tively). The abolition of the Senate, with its district magnitude of two, elimi-nates another source of pull that might have kept the effective number of partieslow prior to reform. The Bolivian Congress must now select the president fromonly the top two contenders, rather than the top three, when no candidate re-ceives a majority of the vote. There is some chance that this will encourage par-ties to coalesce for presidential campaigns and thus exert a downward pressureon legislative races, but any effect is likely to be very minimal, and there is noevidence of such an effect thus far. The adoption of a mixed-member electoralsystem for the legislature does reduce the average magnitude, but the statewideraces still exist in one tier of the voting. Finally, Colombia’s adoption of a single,nationwide district for the Senate with a magnitude of 100 is good for low votegetters, seemingly enhancing the prospects of small parties. However, given thatvotes are pooled at the level of the sub-party list, early evidence indicates thatlow vote-getting lists from the traditional Liberal and Conservative Parties havebeen the beneficiaries (Botero 1998). Given the limited amount of time thatmost reforms have been in place and their half-hearted nature, any differenceswe see in Table 7.3 between the pre-reform and post-reform effective numberof parties should not be attributed to institutional incentives.

Looking at the indicators of interparty relations summarized in Table 7.2 forthe pre-reform era and in Table 7.3 for the post-reform era, there is a decidedlynoticeable lack of change—indicating both the moderate nature of institutionalreforms and the importance of other explanatory factors, especially in periods of“crisis.” The institutionally determined patterns that characterize interparty andintraparty dimensions of representation identified in the extant literature areprobabilistic and based on repeated observations across time and space. Both timeand space are limited when the analysis is restricted to post-reform countries inthe Andean region, and periods of reform and crisis are hardly when we shouldexpect to observe “normal” causal connections. Unfortunately, despite the levelof frustration that provoked reform and the difficulty of actually carrying it outin many cases, reformers failed to address in a thoroughgoing manner the insti-tutional characteristics that would have most directly affected the incentives oflegislators to offer more balanced forms of representation. On the other hand,popular expectations that the effects of the new constitutions would be dramaticand far-ranging made it almost inevitable that disappointment or disillusionmentwould be the likely outcome. Given their limited and haphazard nature, shouldwe or the citizens of Bolivia, Colombia, Ecuador, Peru, and Venezuela have ex-pected reforms to change the way legislators engaged in representation?

216 Crisp

The Colombian Experience with Political Reform as a Cautionary Tale

The experience of Colombia seems particularly troubling for the possibility thatinstitutional reform is likely to lead to legislator behavior that garners confidenceand trust. Support for democracy averaged the lowest score in the Andes (see thenext section). Political overhaul did not translate into lasting public supportbecause the reforms implemented were watered down and off the mark, giventhe source of the woes. Colombia was the first Andean case to seek to reengineerrepresentation through constitutional reform, and it is the one country on whichI have sufficient data to look for pre- and post-reform changes in legislator be-havior. As the Andean case staking out the extreme of personalistic representa-tion, it seems like a prime candidate for being able to see some moderation inthe way elected representatives carry out their duties.

Virtually nothing was done to encourage different forms of behavior bymembers of the Chamber of Deputies who were and continue to be widely per-ceived as corrupt purveyors of particularistic rewards. Because deputies’ per-sonal vote-seeking incentives were not reduced, the nature of representation inColombia has changed less than citizens probably expected it to. Presidents con-tinue to bring forward national programs, only to spend most of their time try-ing to preserve even the remnants of a coherent set of policies. Building supportfor their policies, including a legislative majority, requires bargaining, compro-mises, and payoffs—including illegal payoffs in the form of corruption.

The incentive structure of members of the Senate was potentially changed bythe adoption of a single, nationwide district. Colombian reformers sought tocreate an upper house that was less clientelistic in its interaction with constitu-ents. By adopting a nationwide district for the Senate, the political reform wasintended to encourage senators to focus on large, programmatic concerns thatwould win them votes across the entire country. Returning to Eulau and Karps’s(1977) theoretical work on representation, the goal was to reward senators forpolicy responsiveness as opposed to allocation (including pork barrel rewards),service, and symbolic responsiveness. As Carey and Shugart (1995) reason, in-creasing magnitude in a system with intraparty competition (sub-party lists)should enhance personal vote-seeking incentives, and personal vote seeking isfrequently associated with pork barrel, particularist rewards. However, a nation-wide district should make enhancing one’s personal reputation by claiming creditfor large, programmatic proposals a viable means of winning reelection. For ex-ample, offering to defend the environment may not win you sufficient votes toget elected in any normal-sized district, but in a nationwide district such a can-didate could earn the votes of all the citizens throughout the country for whomthe environment was the single most important issue. The incentives were only“potentially” changed because a nationwide district allows voters from all over


the country to support candidates whose platforms are of broad, national appeal.However, it also allows candidates to seek all of their votes in a single department,just as they did prior to the reform. As a result, the Colombian Senate has a verydiverse set of members—traditional politicians dependent upon geographicallyconcentrated patron-client networks and a new breed of senator who receivessupport from geographically dispersed voters motivated by a policy issue of pri-mary concern to them.

Though most Colombians would probably conclude that the adoption of anew constitution generally, and a nationwide district for the Senate morespecifically, was unsuccessful, it has changed the nature of representation in theSenate. Much more of the chamber’s time is now spent considering bills targetedat the nation as a whole (see Table 7.4). What is more, it can be shown that a sen-ator’s probability of targeting a bill at the nation as a whole increases as his or herelectoral support base becomes more dispersed (Crisp and Ingall 2002). Is greaterattention to issues of national concern associated with increased party disciplinethat would facilitate adopting coherent programs of government?

In Colombia, roll call procedures are used, though not as frequently as onemight expect in a system with such purportedly low party discipline. The vastmajority of votes are cast by a show of hands, but only the number of “yeas” and“nays” are recorded—not who cast them. In the nearly twenty-five-year periodbetween the end of the power-sharing National Front and the constitutional re-placement of 1991, only sixteen roll call votes were taken in the Senate (twentywere taken in the Chamber of Representatives). During the first full legislativeterm after reform, lasting from July 1994 to June 1998, seventeen roll call voteswere cast in the Senate (only four in the Chamber of Representatives). This isrelatively scant information on which to base any judgments, and the mere factthat the very uncustomary roll call procedures were invoked leads to some doubtabout their representativeness (see Ames 2001 on the potential pitfalls when us-ing roll call votes to measure discipline). However, they are the only data avail-able on which to estimate party cohesiveness.

Table 7.4

Bill Targets in the Colombian Senate before and after Electoral Reform

Pre-reform Post-reformCongress Congress

(1986 –1990) (1994 –1998)

Nationally targeted bills 219 417Sectorally targeted bills 137 187Regionally targeted bills 58 32Locally targeted bills 75 75Individually targeted bills 18 18

Total number of bills 507 729

SOURCE: Crisp and Ingall 2002.

218 Crisp

Excluding highly consensual votes (where 90 percent of each major partyvoted the same way), Liberals and Conservatives showed virtually the same pro-pensity to break discipline as they did in the pre-reform era. When roll call pro-cedures were used when senators were elected in department-wide districts, 20 percent of Liberals and 19 percent of Conservatives were likely to dissentfrom the rest of their partisan delegations. In the post-reform era, with senatorshaving to make a name for themselves in a single, nationwide district where 100seats are up for grabs and sub-party slates are still competing against one another,31 percent of the Conservative delegation and 26 percent of the Liberal delega-tion were likely to defect on any given vote (during the 1994 –98 Congress).

Thus, institutional incentives can change behavior, but Colombian reformersfailed or were unable to carry out more far-reaching changes—including simi-lar changes for the Chamber of Deputies and the elimination of sub-party listsfor the election of either chamber (and thus intraparty competition). The adop-tion of a nationwide district for the Senate made it possible for candidates tofocus on programmatic concerns and win votes across the country from voterswith whom these positions resonated. However, increasing the magnitude of thedistrict without eliminating intraparty competition made it even more difficultfor partisan delegations to act as coherent units. Combined with little effort tochange the way deputies carry out representation in the lower chamber, themodest change in behavior by senators has not been enough to satisfy Colom-bians. In late 2002, political reform was once again at the top of the agenda.4

Public Perception of Legislatures (and Democracy More Generally)

Pre- and post-reform indicators of how legislators carry out representation arescant. As I will detail below, using indicators of citizens’ perceptions of legisla-tures to help us evaluate the effectiveness of institutional reforms in addressing the“crisis” in the Andes also poses several challenges. Thus far we have learned thatinstitutional incentives regarding how Andean legislators should carry out repre-sentation were diverse in terms of intraparty politics but tended to push legisla-tors toward one extreme or the other—thinking only in terms of their parties’reputation or only in terms of their personal reputations. Hyper-personalisticsystems should be associated with bills and laws focused on local and individualtargets. Party vote-seeking incentives on the other hand should be associated withbills and laws focused at broader targets including the nation as whole (Crispet al. 2004). Partisan delegations in Venezuela occupied the party-centered ex-treme, with Bolivian partisan delegations finishing a close second. Colombia helddown the opposite end of the spectrum, with very undisciplined partisan delega-tions, and Peruvian legislators were the second most personalistic. Despite newconstitutions in every Andean case, very little was done to change these incen-tives. In interparty terms, Ecuador stood out for its “representativeness,” with


Bolivia being the only other system to have an average effective number of par-ties greater than 3.6.

Tables 1.3 and 1.4 report public confidence and trust in Andean legislatures.Unfortunately, more complete time-serial data on these questions is not widelyavailable (or at least not at a price most individual academics can afford), and thesurveys were not begun until 1995—offering only pre- and post-reform obser-vations in one or two Andean cases. Before the surveys were administered, newconstitutions were adopted in Colombia in 1991, in Peru in 1993, and in Bo-livia in 1994. Ecuador was in the midst of reform and adopted a new constitu-tion in 1996, the first year for which I have data. Venezuela promulgated its newconstitution in 1999, midway between the two observations reported here.

It does not appear that incentives to pursue exclusively one extreme of repre-sentation or the other were a recipe for public support. Neither Colombiannor Venezuelan legislators generated much confidence from the general public.They were the only countries where fewer than 20 percent of respondents ex-pressed “much” or “some” confidence in Congress in 1996 (see Table 7.5). TheVenezuelan Congress generated significantly greater confidence in the post-reform 2002 surveys. It would be tempting to argue that institutional reformswere responsible for the boost, but this is only tenable in the most general terms.In Venezuela, the 1999 Constitution created a unicameral legislature and man-dated the use of participatory candidate selection procedures. The latter changewas blatantly ignored by all parties. Rather than confidence due to specific alter-ations in legislators’ incentives, it seems more likely that the increased confidencewas due to a more general sense that political, economic, and social changes wereunder way. Thus, the increased confidence would be more accurately attributedto President Chavez’s challenge—including, but certainly not limited to, leg-islative reform—to the traditional political and economic elites. Ecuador, theone case where intraparty incentives were relatively balanced between hyper-centralized and hyper-personalistic extremes, was the only case to generate“much” confidence among more than 6.75 percent of the population. The leg-

Table 7.5

Public Confidence in Congress, 1996

1996 2000

Bolivia 22 16Colombia 15 14Ecuador 27 9Peru 33 23Venezuela 19 37Andean average 23 20Rest of region 29 ??

SOURCE: Latinobarómetro 1996, 2000.

220 Crisp

islature received relatively high marks in 1996 when reform was underway. How-ever, by 2002 the legislature had the lowest level of confidence of any country inthe Andes. In my conclusion, I will return to the idea of institutional reforms cre-ating false expectations, generating only limited changes in behavior, and ulti-mately leading to disillusionment with the idea that institutions matter at all.

While the Andean countries’ average of “much” confidence was equal to orslightly greater than the rest of the region in 1995, the Andean average was sub-stantially lower than the rest of the region on “some” confidence in Congress.Any conclusions drawn from these figures must be considered very tentative, butone possible explanation for the differences in confidence is the extreme natureof the electoral system. The systems most extreme on the intraparty dimensiongenerally fare worse. People’s frustrations with the system are reflected in thelower than average levels of confidence and trust in the legislature.

Figure 7.2 traces the percentage of respondents who were “very satisfied” or“somewhat satisfied” with democracy in their country between 1995 and 2002.Clearly, a longer time series that included data predating all our cases of reformwould be preferable, but comparable annual data does not exist. In addition,these results report responses to a question about democracy as a form of

Figure 7.2Public satisfaction with democracy

Perc

enta

ge s

atisfi

ed

Years1996 19981995 1997 2000 20022001

Ecuador�Bolivia


60�

50�

40�

30�

20�

10�

0


government, not the legislature in particular. While the behavior of parties inthe legislature is undoubtedly key to most conceptualizations of representativedemocracy, we are making an additional assumption when we use feelings aboutdemocracy to discuss satisfaction with legislators’ behavior.

With those cautions in mind, it is the case that in three countries—Peru,Ecuador, and Venezuela—the first observation taken after political reformshowed the highest levels of satisfaction in the country’s time series. For ex-ample, satisfaction with Venezuelan democracy peaked with the 1999/2000 ob-servation5—the period when the new constitution was drafted and adopted.Satisfaction reached 55 percent in that year, while it averaged 19 percentagepoints less in the other six observations. Two years after the adoption of the newconstitution, the percentage of satisfied respondents had dropped by 15 percent.Similar to the trend for confidence in the Ecuadorian legislature cited above, ifthe bump in popularity was generated by the promise of change, it was fleeting.Colombia, the Andean country furthest from its experience of reform whenthese data were collected, had the lowest average level of satisfaction, dipping tojust 11 percent by 2002.6

In sum, Andean legislatures do not earn much confidence, and Andean de-mocracies, more generally, have trouble sustaining satisfaction. Figures for pub-lic confidence in the Congress suggest that legislators who focus either pre-dominantly on parochial concerns or predominantly on national concerns failto earn the respect of their citizens. Promising to use institutional reform toelicit better behavior can lead to a temporary spike in popular support, but whenthe expected new behavior fails to materialize, the public quickly becomes disillusioned.

Future Prospects of Andean Legislatures

Thus, in the end, the institutional reforms undertaken thus far have been ratherineffective. Striking a balance between programmatic and parochial forms ofrepresentation is not an easy task. Legislators in the Andes were prone to offerrather “extreme” forms of representation prior to reform, and the tentative ev-idence available indicates that their citizens have not been impressed with thedegree to which the legislators’ behavior changed as a result of adopting newconstitutional structures. The cynicism generated by the failure of one round ofreforms makes additional changes even more difficult to achieve. Because insti-tutions by definition involve the formalization of a set of practices or norms,they are not merely the reflection of underlying social or economic forces. Adisjuncture may emerge between voters’ preferences and the behavior encour-aged by a set of institutional structures. Unfortunately, just because the disjunc-ture is sufficiently severe to motivate reform, there is no guarantee that the re-forms undertaken will be optimal.

222 Crisp

This is not to say that institutional reform generally is not a viable route forchanging the nature of representation.7 The scholarly literature has identifiedcross-national differences in legislator behavior that follow the regular patternswe would expect given the nature of the electoral system. The evidence pre-sented here on the Colombian Senate indicates that legislators in a given systemwill change their behavior over time if the rules under which they are electedare changed. The lesson to be drawn from the Andes is that institutional changeshave not systematically revamped electoral systems to encourage legislators tooffer a mix of national and parochial policy promises. I suspect that reforms havefailed to dramatically change the nature of representation in Andean legislaturesfor a combination of two reasons. First, constitutional reformers may find it eas-ier to identify the symptoms of extreme representation than its root causes(Shugart, Crisp, and Moreno 2002). For example, where presidents have a hardtime getting their proposals through Congress, it may appear more obvious tostrengthen the president’s constitutionally allocated powers than to change theincentives of legislators. Second, the legislators and parties that are expected tochange their behavior are typically deeply involved in the reform process itself.They may be hesitant to implement sweeping changes to the rules of the polit-ical game because they are unsure whether they will prosper electorally under anew set of institutions. Continued comparative research on the motivations oflegislators as they provide alternative forms of representation can only furtherinform practitioners as they attempt to improve the relationship between electedofficials and those they govern.

Notes

1. Unfortunately, we do not have systematic information on the cohesiveness of par-tisan delegations in Andean legislatures. Data collection of this sort is underway for manycountries, but comparable data is still not available. Some legislatures simply do not usethe roll call procedures necessary to measure party discipline. Instead, we must, at leastfor the time being, rely on the incentives that encourage partisan or individualistic be-havior to evaluate intraparty politics in the Andes.

2. For a complete review of political reforms in Latin America during the 1990s, seeShugart, Crisp, and Moreno (2002).

3. In the Conclusion, Mainwaring argues that Andean cases “underwent importantelectoral system reforms in recent years in order to enhance the direct accountability ofrepresentatives to voters and to enhance mechanisms of representation.” He then reasonsthat if electoral reforms were extensive and yet the perception of crisis persists, then in-stitutions offer us little leverage on the causes of the crisis or its likely solutions. How-ever, as I noted in the text, changes to the incentives legislators face on the intraparty di-mension of representation were reformed very little, and/or in the wrong direction.


Thus, we have no evidence to evaluate whether reforms that would generate electoralefficiency (Shugart 2001) on the intraparty dimension would help ameliorate the crisis.No one should expect the creation of a few set-aside seats for indigenous candidates tosolve hyper-personalistic or hyper-centralized extremes. Certainly, they are not the onlyfactors to be considered, but without relevant reforms it seems hasty to conclude that in-stitutions hold no explanatory power.

4. Interestingly, in an effort to pre-empt President Uribe’s referendum on a wide array of political reforms—that would eventually go down to defeat due to a lack ofturnout—the Colombian Congress passed a bill mandating a single list per party in eachdistrict. Assuming no further reforms, the first congressional elections held under theserules are scheduled for early 2006.

5. The Latinobarómetro was not done in each country in both 1999 and 2000. Instead,some were surveyed in 1999 and others in 2000, and the results for the two years overwhich the region was completely surveyed are reported together.

6. By the end of the year, newly elected president Álvaro Uribe had placed politicalreform firmly back on the agenda, hoping to make the regime more efficient and lesscorrupt. His administration was received with a great deal of enthusiasm, and it seemedlikely that promises of change would generate optimism to be recorded in the 2003 sur-vey. If the boost in confidence was generated, would it be as fleeting as it had been inVenezuela?

7. See my earlier note taking exception to Mainwaring’s conclusion that electoral reforms were extensive.

References

Ames, Barry. 2001. The Deadlock of Democracy in Brazil. Ann Arbor: University of Michi-gan Press.

Botero, Felipe. 1998. “El Senado que nunca fue: La circunscripción nacional después detres elecciones.” In Elecciones y democracia en Colombia 1997–1998, ed. Ana María Be-jarano and Andrés Dávila, 285–335. Bogotá: Universidad de los Andes.

Carey, John M., and Matthew S. Shugart. 1995. “Incentives to Cultivate a Personal Vote:A Rank Ordering of Electoral Formulas.” Electoral Studies 14: 417–39.

Coppedge, Michael. 1994. Strong Parties and Lame Ducks: Presidential Partyarchy and Fac-tionalism in Venezuela. Stanford, CA: Stanford University Press.

———. 1997. “A Classification of Latin American Political Parties.” Kellogg InstituteWorking Paper, no. 244. Helen Kellogg Institute for International Studies, Universityof Notre Dame, Notre Dame, IN.

———. 1998. “The Dynamic Diversity of Latin American Party Systems.” Party Politics4, no. 4: 547– 68.

Crisp, Brian F. 2000. Democratic Institutional Design: The Powers and Incentives of Venezue-lan Politicians and Interest Groups. Stanford, CA: Stanford University Press.

Crisp, Brian F., and Scott W. Desposato. 2004. “Constituency Building in MultimemberDistricts: Collusion or Conflict?” Journal of Politics 66, no. 1: 136 –56.

Crisp, Brian F., Maria C. Escobar-Lemmon, Bradford S. Jones, Mark P. Jones, andMichelle M. Taylor-Robinson. 2004. “Electoral Incentives and Legislative Represen-tation in Six Presidential Democracies.” Journal of Politics 66, no. 3: 823– 46.

Crisp, Brian F., and Rachael E. Ingall. 2002. “Institutional Engineering and the Natureof Representation: Mapping the Effects of Electoral Reform in Colombia.” AmericanJournal of Political Science 46, no. 4 (October): 733– 48.

“Elections around the World.” http://www.electionworld.org/.Eulau, Heinz, and Paul D. Karps. 1977. “The Puzzle of Representation: Specifying the

Components of Responsiveness.” Legislative Studies Quarterly 2, no. 3: 233–54.Ingall, Rachael E., and Brian F. Crisp. 2001. “Determinants of Home Style: The Many

Incentives for Going Home in Colombia.” Legislative Studies Quarterly 26, no. 3:487–511.

Johnson, Gregg B., and Brian F. Crisp. 2003. “Mandates, Powers, and Policies.” Ameri-can Journal of Political Science 47, no. 1 ( January): 128– 42.

Latinobarómetro. Various years. “Press Releases” (Prensa). http://www.latinobarometro.org.

Mainwaring, Scott, and Matthew Soberg Shugart, eds. 1997. Presidentialism and Democ-racy in Latin America. New York: Cambridge University Press.

Morgenstern, Scott, and Benito Nacif, eds. 2002. Legislative Politics in Latin America. NewYork: Cambridge University Press.

Pitkin, Hanna Fenichel. 1967. The Concept of Representation. Berkeley: University of Cali-fornia Press.

Rey, Juan Carlos. 1972. “El sistema de partidos venezolanos.” Politeia 1: 175–230.Shugart, Matthew Soberg. 2001. “ ‘Extreme’ Electoral Systems and the Appeal of the

Mixed-Member Alternative.” In Mixed-Member Electoral Systems: The Best of BothWorlds? ed. Matthew Soberg Shugart and Martin P. Wattenberg, 25–51. New York:Oxford University Press.

Shugart, Matthew Soberg, Brian F. Crisp, and Erika Moreno. 2002. “Re-ConstitutingDemocracy: Institutional Patterns of Political Overhaul in Latin America.” Universityof California San Diego, University of Arizona, University of Iowa. Typescript.

Shugart, Matthew Soberg, and Martin P. Wattenberg, eds. 2001. Mixed-Member ElectoralSystems: The Best of Both Worlds? New York: Oxford University Press.

Taylor-Robinson, Michelle M., and Christopher Diaz. 1999. “Who Gets LegislationPassed in a Marginal Legislature and Is the Label Marginal Legislature Still Appropri-ate? A Study of the Honduran Congress.” Comparative Political Studies 32: 590 – 626.

224 Crisp

Part III

❖

POPULAR POLITICS AND THE CRISIS OF DEMOCRATIC

REPRESENTATION

This chapter addresses a core puzzle: Why is continued citizen mobilizationaccompanied by growing disempowerment of those same citizens? Why domovements fail, leaders burn out, and members disperse, and what are the im-plications of this organizational failure for democratic representation? Citizeninvolvement in such movements arises precisely because of the failure of con-ventional vehicles of representation to provide trustworthy and effective meansof connecting new citizen groups and their needs with public institutions. Thatthese new movements should also regularly fail raises important questions aboutthe quality and durability of democracy. Our consideration of the issues is rootedin a close examination of urban movements, mobilization, and empowermentand disempowerment in the recent experience of Venezuela and Peru. Thepuzzle that concerns us is of course not limited to these two countries: it is com-mon to all the Andean republics, and in different ways, to much recent ex-perience of urban mobilization in Latin America and beyond. After a briefaccount of urban citizen movements and politics in our two cases, we out-line general reflections on the nature of empowerment and disempowerment,on the peculiar combination of strengths and weaknesses that mark many con-temporary movements. A close examination of types of movements and theirlinks with political parties and protest follows. We close with analysis of twowaves of mobilization: in Peru (which sparked the ouster of President AlbertoFujimori) and in Venezuela (both for and against President Hugo Chávez Frías),and with reflections on the likely future of empowerment and disempower-ment for urban citizens and the implications of this perspective for democraticrepresentation.

8

Urban Citizen Movements andDisempowerment in Peru and Venezuela

❖

Daniel H. Levine and Catalina Romero

The Puzzle

The puzzle is set up by three key facts that situate Venezuela and Peru in a mean-ingful comparative perspective while providing grounds for comparison betweenthe social and political processes each country has experienced over the lasttwenty years. The first points to the decay, decline, and eventual disappearanceof once powerful political parties, and of the system of organizations and politi-cal norms built around them. The second addresses the creation, expansion, riseto prominence, and decay (often after specific goals were met) of networks com-prised of civic organizations, sometimes referred to as “civil society” or “popu-lar movements,” depending on the country and circumstances. The third is a tra-jectory of mobilization, activism, and sustained protest (associated with thetrajectory of new movements and networks) rising to peaks at moments of crisisand dissipating thereafter. In both countries, and through extended periods oftime, huge numbers of people were mobilized for sustained, repeated, and oftenrisky collective actions, including rallies, campaigns to collect and deliver signa-tures, marches, demonstrations, sit-ins, and the like. A profusion of new and of-ten short-lived groups combined with established organizations such as tradeunions or business federations, political parties, and professional groups to man-age and sustain the effort. Relevant moments of crisis, examined in more detailbelow, include the movement to reject Fujimori’s 2000 reelection, or the wavesof mobilizations and counter-mobilizations (centered on the Chávez govern-ment and its survival) that began in spring 2002 and culminated in the remark-able civic strike touched off in December 2002 and stretching into February ofthe next year.

The recent experience of Peru and Venezuela has sparked an extensive litera-ture.1 The key point to underscore, and the real value of comparing these other-wise very different societies, is how much the comparison sheds light on a com-mon effort to grapple with similar problems in the construction, defense, and“deepening” of democracy. Central to this effort has been a continuing, and notalways successful, struggle to enhance participation, broaden access to politicsthrough linked institutional reforms (including reforms in systems of representa-tion), and to strengthen the accountability of politicians and public institutions.In each case, the effort was spurred by the emergence of new capabilities andgroups outside the net of state and established political parties who have sought to openand energize politics. Facing institutions and leaders they rejected as corrupt andunresponsive, citizens in both countries turned to civil society as an arena for par-ticipation and a platform for demands about representation. In this light, the mo-bilization by citizen movements that we examine here involves more than simple“demand making” that then finds representation through established conven-tional channels: it is a claim to representation that politicizes new spaces andgroups in national life. The failure or short-circuiting of the movements raises

228 Levine and Romero

questions about the possibilities of constructing enduring representation thatstarts and ends in “civil society.”

Peru and Venezuela confront their common dilemma from very different start-ing points. Peru is a poor, ethnically divided country whose modern historydisplays a series of short-lived bouts with electoral democracy. The militarygovernment (1968–80) and the protests accompanying the reinstitution of civil-ian democratic politics marked an important new beginning on several critical di-mensions. Enormous numbers of people were brought to political action by thepopular movements of the late 1970s and enfranchised thereafter. This means thatstarting with the elections of 1980 there was perhaps for the first time in the his-tory of Peru a genuine mass public for electoral politics. The restoration of partyand electoral politics coincided with the beginnings of massive violence by Shin-ing Path. The popular movement would soon be caught between the violence oftwo armies: Shining Path and the repressive forces of the Peruvian state, bent oneliminating them. Throughout the same period, Peru experienced a rapid ur-banization that profoundly altered the character and dynamics of urban life. Ahost of new groups and spaces came into being, and movements and protestsbecame a part of the daily social and political scene (Dietz 1998; Stokes 1995).This was also a time of profound economic decay, which hit the popular move-ment hard, making collective action of any kind difficult. Democratic institutionswere undermined by the governments of Alberto Fujimori (1990 –2001), whosegrowing authoritarianism, isolation, and corruption ultimately triggered hisdownfall. The combination of violence, institutional and economic decay, andleadership betrayal was deadly for urban movements. Although the capacity andwill for mobilization remains, as visible in the massive protests against Fujimori’sfraudulent reelection in 2000, movements lack sustained organization and meansto ensure continuity and accountability.

Venezuela’s experience of civic movements starts later in time (mid-1980s) andarises within a well-established democratic system: the goal was to “democratizedemocracy” by broadening citizen access and loosening the grip of the country’spowerful political parties and dominant state apparatus. The existing democraticsystem was much stronger, richer, and more deeply established than was the casein Peru. From this position of strength, the fall is all the more notable. Beginningin the late 1980s, economic and institutional decay began to bite, popular dis-satisfaction with established institutions (especially the dominant political par-ties) grew sharply, and the political system entered an extended crisis that con-tinues to this day. As was the case in Peru, these developments were accompaniedby the emergence of a wide range of citizen movements, centered in the cities,that demanded more authentic and accountable representation. Venezuelanmovements differ from their Peruvian counterparts in many ways, not least thefact that from the beginning their membership base and agenda have been pre-dominantly middle class. The initial demands of movements found expression in

Urban Citizen Movements, Disempowerment in Peru, Venezuela 229

decentralizing reforms that devolved power to states and cities, expanded thenumber of offices open to election, and reduced barriers to participation. Butthese reforms were swept away by the continuing crisis of the country, and mar-ginalized by the victories of Hugo Chávez, who came to office with a whollydifferent agenda of total change (Kornblith 1999; Salamanca 2004; Levine 2002).The initial power of Chávez’s movement was enhanced by the collapse of olderpolitical structures; as these recovered ground and citizen movements began toemerge again, opposition mounted, mostly in the form of civil society mobiliza-tions, once again seeking political redress and accountability outside the formalstructures of the political system.

To summarize, in both countries the discredit and decay of established lead-ers and parties combined with institutional failure and sustained economic crisisopened the way, at different points and with country-specific nuances, for a widerange of movements to emerge and claim a voice as “civil society.” Participantsin these movements sought, by their activism, not only to satisfy immediate de-mands (say for housing or services), but also to express, by their action, a claimto citizenship and equal status apart from established, conventional structures ofrepresentation. They project not only their demands, but also their image ofthemselves as citizens, forcefully onto the public stage. Their activism politicizedurban spaces in the two countries in new ways: creating new forms of action andbuilding (often literally) new spaces for such activism. In both cases, the long-term results of such activism, in terms of sustained benefits, new policies, or ac-countable leaders, have been problematic. The weakness, reversibility, and oftenopen failure of the effort requires us to reconsider the possibilities and limits ofdemocratic representation, and to search for possible solutions in ways that gobeyond tinkering with electoral machinery.

The relation of empowerment and, by extension, disempowerment withdemocratic representation is central to our inquiry. Most discussions of empow-erment have a “people-friendly” character. They underscore the need to providepeople with the skills and capabilities that make access to power possible—toempower them and to enhance the “quality” and “authenticity” of representa-tion. As typically used in these discussions, “quality” and “authenticity” of rep-resentation involve more than simply assuring that electoral results reflect votesmore or less accurately and fairly (according to whatever electoral rules are inuse). Assuming universal suffrage and relatively free and open elections, repre-sentation that is authentic and of high quality entails lowering barriers to organi-zation, multiplying instances and arenas of political action and representation,making voting easier, and ensuring that representatives are more accountable andmore accessible to ordinary citizens. The goal of such reforms is to link new ur-ban spaces, groups, and networks to the institutional structures of the politicalsystem in ways that allow social energies to “bubble up” and find representation.Our concern is that the link has been problematic and the record at best mixed.


The line of analysis we follow here requires that the concept of representation(and democratic representation) be situated in a broad analytical and social con-text that extends beyond the details of electoral rules and institutional structur-ing to address the potential links of these public spaces with the new social spaces,movements, and claims to representation being advanced from society. In bothour cases, massive numbers of citizens have repeatedly joined together and soughtrepresentation of their interests through public, often risky, mobilizations of allkinds. Politics and systems of representation should be capable of linking to-gether these new networks and spaces, but with rare exceptions this has not hap-pened. Political leaders remain wedded to a top-down vision in which it is theywho know what to do and how to do it. They either do not make the links orthey use them for a time and move on. Citizen movements are too often leftstranded and divided, lacking enduring channels of contact or control into thepolitical sphere. The theoretical and practical challenge is to rethink the relationbetween social movements and political representation in ways that preserve theenergy and openness of both. That is our agenda here.

Movements and Politics in Peru and Venezuela

The decay of political parties and the rise of an explicitly anti-party politics iscommon to the recent experience of both Venezuela and Peru. In Venezuela, anentire political system built around powerful, permanent political parties weak-ened under long-term economic pressure, exacerbated by massive corruptionand ineffective leadership, and further undermined by reforms set in motion inthe mid-1980s. Although it is not easy to date the start of the decline with preci-sion, most observers agree that by the early 1990s, the two dominant parties—AD (Acción Democrática, or Democratic Action) and COPEI (Comité deOrganización Política Electoral Independiente, or Independent Political Organi-zing Committee—a term hardly ever used in subsequent years; the party has longbeen known only by the acronym)—were shadows of their former selves (Molinaand Pérez 2000; Crisp 2000; Crisp, Levine, and Molina 2003). Their weakenedcondition undermined the ability of leaders to respond effectively to the crisiscreated by the two attempted coups of 1992, and by the continued economic cri-sis. Once-legendary party discipline weakened, making secure interparty deals inthe legislature much more difficult to manage. COPEI divided, and its founder,Rafael Caldera, waged a brilliant anti-party campaign to win the presidency in afour-way race in 1993. This was the first presidential election since the restora-tion of democracy in 1958 that was not won by either AD or COPEI. Althoughthe two parties continued to do well in regional and local voting, the whole pe-riod was marked by continued intraparty divisions, by an explosion of citizen or-ganization (including insurgent unionism) escaping from party control, by risinglevels of abstention, and by growing anti-party sentiment.


The election of 1998 and subsequent national, regional, and local votes haveconfirmed the deathbed status of the established political parties and the entirepolitical system constructed around them. President Chávez has moved stronglyagainst the parties, and subsequent national voting has been dominated by personalist coalitions, both pro- and anti-Chávez. The voting system invented for elections to the Constituent Assembly, which wrote the country’s new “Bolivarian Constitution,” gave supporters and allies of President Chávez a dis-proportionate share of seats (95 percent of seats, with 66 percent of the vote).Subsequent legislative elections returned to the old system, with results (in termsof seats) that were more proportionate to votes received.

In Peru, political parties (with the sole exception of APRA—the Alianza Pop-ular Revolucionaria Americana, or American Revolutionary Popular Alliance)were not as powerfully structured or deeply organized as in Venezuela. A polit-ical system hinged on electoral competition between well-established parties,each with its affiliated movements and organizations, made a tentative appear-ance in 1955, and again, with the restoration of democracy and civilian politics,after 1980 (Tanaka, this volume). In this system, APRA was joined by AP (Ac-ción Popular, or Popular Action, founded several decades earlier by FernandoBelaúnde Terry), the PPC (a Christian Democratic Party) and by IzquierdaUnida (United Left), a loose coalition of leftist parties. The fortunes of theseparties rose and fell through the 1980s as the economic situation deterioratedand the insurgency (led by Shining Path) grew and extended its reach across thecountry. AP won the presidency with Fernando Belaúnde Terry in 1980, andthen plummeted in support; APRA won with Alan García in 1985, and then lostsupport; and the Izquierda Unida gained steadily in municipal elections throughto the mid- to late 1980s, only to collapse in division.

The 1990 election completed the decline of the parties as central political or-ganizations. This election ended up as a contest between two coalitions led byindependents—the writer Mario Vargas Llosa and the unknown Alberto Fuji-mori, the eventual winner. Only two years into his term, President Fujimoridissolved Congress and began rewriting the rules of the political game. He waselected for a second term in 1995, running against Javier Pérez de Cuellar, for-mer Secretary General of the United Nations. After Fujimori’s ouster—follow-ing his reelection in 2000, which was contested as fraudulent and boycotted bypossible opponents—the presidency was won by another independent, Alejan-dro Toledo, a leader of the anti-Fujimori movement. His opponent in the sec-ond round was former president Alan García, running again for APRA.

In both countries, the decay of parties and of a party system (strong or weak)was accompanied and pushed or pulled along by an explosion of citizen organi-zation and new movements of all kinds. We understand the emergence of newgroups and their sustained presence on the public scene as an effort to create spacesand connections where democracy can be practiced and interests aggregated


and pressed in political encounters of all kinds, from constitutional debates andlegislative discussions to petitions, referenda, marches, rallies, and demonstra-tions. Representation is clearly at issue, but the phenomena we address here arenot well captured by conventional “principal-agent” discussions of the matter.This is representation both as claim to voice and a legitimate place at the politicaltable and as a challenge to the terms of representation enshrined in existing pub-lic institutions. The process of social participation was visible earlier in Peru,where movements emerged in the 1970s in opposition to military rule and as anexpression, above all in the cities, of grassroots organizing to meet social andenvironmental needs. The Peruvian Catholic Church played a key role in pro-moting and protecting many such movements, training activists and providinginvaluable connections among them. By the end of the 1980s, and into theFujimori period, the combination of economic decline (which made collectiveaction of any kind more difficult) with increasing violence, both from ShiningPath and the government, undermined the ability of many groups to survive andrenew themselves.2 Mobilizations continued, of course (Dietz 1998; Stokes 1995;Tovar 1991; Levine and Stoll 1997), but became more short-lived and more lim-ited and specific in focus. The transition to a democratic regime combined withgenerational changes in the leadership of the Catholic Church also removed keyallies from the scene. Parties further weakened because urban organizations de-veloped a sense of autonomy, looking in a democratic way to their own collec-tive interests and goals, which seemed to be different from those of the politi-cal parties. The return to democracy in the 1980s reestablished elections at themunicipal level, providing public spaces for participation and the expression ofdemands.

In Venezuela, the power of party organizations and their ability to colonizecivil society and monopolize access to resources long inhibited the growth of in-dependent civic associations. As we have seen, these began to appear in the mid-1980s, with roots in movements in the business sector and, above all, groups ofurban property owners opposed to unrestricted development. Motives of neigh-borhood defense soon expanded into a broad agenda aimed at the creation ofmore autonomous urban governments with independent elected, not appointed,mayors and governor. This reform was put into effect in 1989, and combinedwith other decentralization measures, began to reshape the dynamics of partyleadership and campaigning in the 1990s. At the same time, the country’s long-term economic decline, which continued throughout the 1990s, underminedthe ability of party leaders to distribute patronage and thus hold loyalties. Au-tonomous professional groups appeared, private foundations and new businessgroups consolidated their position, and independent union movements began togain ground. The latter, most successful in the steel mills of Guyana, spawned asuccessful political movement, La Causa R (the Radical Cause). The term “civilsociety” appeared as a regular feature of Venezuelan political discourse, and


efforts to forge some kind of unified position were made (Gómez Calcaño 1998;Salamanca 2004; Levine 1998).

The political trajectory of Hugo Chávez Frías, his election to the presidencyin 1998 (affirmed in subsequent votes under a new constitution and with newelectoral rules), and his overall political project challenged the legitimacy of thecore political arrangements of the past four decades, and looked to build a newand supposedly more democratic society and political system. Fiery populist andclass-based rhetoric has been the daily bread of the “Bolivarian revolution” fromthe beginning, and mobilization of masses has been its core claim to legitimacy.Like Fujimori earlier in Peru, Chávez looked to destroy existing political par-ties (and associated groups, notably the trade unions), with the difference thatChávez wanted to rebuild politics in a “revolutionary” and “participatory” style,with a broad range of arenas and groups in direct contact with the leader and thestate. In practice, this has meant attacking and dismantling old structures, rest-lessly inventing and reinventing new ones, including, notably, the regime’s ownpolitical party, and diverting state resources into vaguely defined “Bolivariancircles.”

The decay and rout of the old system was so complete that it took several yearsfor opposition to begin to regroup. Early steps (2001) came with the defeat ofgovernment-sponsored efforts to “take” the Central University in Caracas for the“people,” and with the defeat of a government-sponsored referendum to “re-new” the leadership of the trade-union federation. These were followed by amassive series of work stoppages, strikes, and marches that became a regular fea-ture of the calendar in Caracas and, to a lesser extent, in other cities. Protest tech-niques common in other countries, such as cacerolazos (or banging of pots andpans, creating a truly deafening noise) and caravans of cars honking horns, wereput to use, and massive marches (long since abandoned in favor of television-centered campaigning) returned to center stage. A regime claiming legitimacyon the basis of its ability to mobilize was now running into massive counter-mobilizations. Fearing the appearance of weakness and the prospect of losingcontrol of the street, the regime began to put on its own massive marches. Therewas continuous escalation in this process from early December 2001 through tothe tragic events of April 11–13, 2002, when a huge march, heading throughdowntown Caracas to the presidential palace, was attacked by snipers. Many werekilled, and in the ensuing crisis the government was replaced and then retookpower as the military divided and different coalitions of citizens “took” and “re-took” the streets. All sides then pulled back from the brink for a while, but afterabout six weeks, the rhythm of marches and countermarches began again, accel-erating through the fall of 2002 and culminating in the remarkable civic strikeof late 2002 and early 2003.

It is instructive to compare the mobilizations that forced out Fujimori withthose competing to oust, support, or restore Chávez. The former were managed


by a loose coalition of groups from across the country, knit together by local- andnational-level activists with prior experience in mobilizations, energized by theOAS findings of fraud and irregularities in the 2000 reelection of Fujimori, andby growing revelations of corruption linked to Vladimiro Montesinos, Fujimori’sdirector of the National Intelligence Service. Mobilizations were sparked first bystudents and women’s groups, who began with symbolic acts such as sweeping theplaza of the Congress and regular washings of the national flag (to cleanse themof corruption). As protests expanded, they were joined by NGOs and then bypolitical parties, which added financing and organizational reach.3 The campaignitself combined enormous marches (such as the Marcha de los Cuatro Suyos4 inLima, on July 28, 2000) along with a series of sustained regional mobilizations,and innovations such as the previously mentioned weekly public washings of thenational flag. The organizations and political parties so prominent in the 1980shad disappeared from the political scene following the coup of 1992, losing theirlegal status after failing to win seats in elections for the Democratic ConstituentCongress in 1993 and later in the presidential elections of 1995. They resumeda role only when protests were well under way.

In Venezuela, by contrast, as the opposition to President Chávez recovered andbegan to gather force, the organizational backbone for sustained action did notrest on groups formed over the past ten or fifteen years. An unexpected but highlyeffective anti-government alliance was formed between the trade-union federa-tion, the business federation, the Catholic Church, and the mass media. The firsttwo provided organizational resources, the latter two, legitimacy and an amplifiedpublic voice. That this coalition was able to put so many people into the street onsuch a regular basis depended less on the groups’ own members than on the mo-tivation of a loosely linked net of neighborhood and human rights groups.5

Despite continuous reference to the role of “civil society,” in neither countrydid the specific membership organizations of the previous decade, once seenthemselves as the potential foundation for a new kind of politics, play a centralrole. Different kinds of organizations emerged to take the lead. Apart from hu-man rights groups, which have grown throughout the region in the last fifteenyears in response to dictatorship (Sikkink 1993), the key organizational playerswere either occasional coalitions gathered for a particular purpose around a spe-cific leader—for example, Alejandro Toledo and Perú Posible (Possible Peru)—or old-line organizations such as trade unions, business federations, or theChurch. Mobilization and commitment were sustained not so much by groupstructures themselves as by the presence of numbers of loose or “weak” tiesamong groups and individuals that facilitated connections and the exchange of in-formation, support, and resources across groups, social sectors, and physical spaces(Granovetter 1973; Smith 1996). If this is correct, mobilization—even massiveand sustained mobilization—is compatible with the absence of an organizationalunderpinning like that commonly provided by political parties. But at the same


time, the absence of a continuing organizational structure can undermine thepotential consolidation of gains and make it all but impossible for citizens to de-mand and achieve accountability from leaders without a new round of massive,institution-challenging mobilizations. This deserves closer examination.

Considerations on Empowerment, Disempowerment, and Representation

“Empowerment” is a notoriously plastic concept, often used in conjunction withequally protean terms such as “civil society” or “social capital.” Like “account-ability,” “empowerment” has no easy equivalent in Spanish, and neologisms suchas “empoderamiento” fill the linguistic gap. The elasticity of these concepts reflectstheir multidimensional character: they point to processes that involve organiza-tional growth, personal and collective identity, specific leadership skills, trust, theability to secure goods and services, and the like, and operate simultaneously ona range of social levels. Of these concepts, “empowerment” is perhaps the mostpeople-friendly. Empowerment denotes a kind of social and political process anda pattern of structure and organization that provides citizens with a growing rangeof arenas for access to the public sphere, reduces barriers to action, and createsconditions that enhance a sense of self worth and recognized personal as well ascollective identity.6

In this light, the relation between empowerment and a sense of citizenshipseems clear enough. Those women and men who come to see themselves as cit-izens with rights equal to others are in that measure set on the road to individualand collective action as normal and possible. The emphasis on identity, however,masks considerable ambiguity around the relation of empowerment to organiza-tion. Organization can further empowerment by linking individual and groupcapacities together and moving action to larger arenas. But at the same time, bysubordinating group efforts to leadership concerns and stifling independent de-cision, overarching organization can also disempower. In his work on religion inthe United States, Warner (1993, 1070) states, “It is to be expected that the em-powerment functions of religion are latent. At an individual level, those who seekwell-being in religion tend not to find it; those who gain well-being from reli-gion are not those who seek it.” The logic of Warner’s apparent paradox rests onan argument that locates empowerment (like social capital) in the long-term con-struction of community, trust, and the skills and disposition required for work-ing together—not just in creating or joining organizations, and much less in sim-ply “getting the goods.”

This is a lot for any social process to deliver, and many movements have notbeen able to fill the bill. A review of recent theoretical and empirical work on ur-ban social movements, empowerment, and representation in Latin America re-veals a slow recovery from a hangover brought on by exaggerated expectations,


laced with a heavy dose of idealization of the new movement. The autonomy ofmovements (vis-à-vis institutions such as political parties, state institutions, or theChurch) was overdone, and a romantic image of the “small is beautiful” kindmade many observers anticipate that a totally new kind of politics would arisefrom the seeds provided by these movements. This in turn would provide the ba-sis for a different pattern of representation with new kinds of political parties, andaltered institutions that would hopefully be more democratic and more fully em-powering of citizens than what had hitherto existed (Hellman 1992; Lander 1995;Levine and Stoll 1997; Lora 2002; Ortner 1995; Oxhorn 2001; Tovar 1991).

This did not happen: in case after case, the new politics was easily absorbedinto the old, and movements split or simply fell apart. That movements fail and“empowerment” does not endure should come as no surprise. Movements oftenfail or run out of steam: activism is costly and antinomian and the day-to-daypressures of economic and family survival make organization difficult to sustain(Piven and Cloward 1977, 1998). Anyway, as Stokes and others have shown forPeru, the development of supposedly more participatory (and therefore “em-powering”) styles of organization among the urban poor does not necessarily re-place older self-concepts and forms of action. People are practical, and new stylesof action take their place as an alternative to be weighed and perhaps used, as cir-cumstances seem to indicate.

What does disempowerment mean, and what is the path from empowermentto disempowerment? There is withdrawal from activism, often prompted byburnout, sometimes by family pressures (commonly gender specific and affectingwomen). There is also a failure of leadership replacement. Groups that campaignfor democracy may of course remain authoritarian within, and leaders may findit difficult to let new generations come to the fore. The problem is notorious ingroups linked to the Catholic Church (as many have been), where dependenceon clergy makes for enormous vulnerability if and when more conservative clergyarrive on the scene. Finally, of course, with the opening of new political spaces(through transitions to democracy or reforms within democratic systems)younger activists easily find other, perhaps more rewarding and less costly outletsfor their energies.

We do not suggest that empowerment is necessarily illusory. Many men andwomen have indeed acquired new skills and self-images and imparted these toothers in their communities. The central point here is that the concept is incom-plete, and the reality fragile. The difficulty lies more with links to organizationand the reliable construction of representation, which may undermine the con-solidation of gains. The linkage between the civic spaces of empowerment andthe public spaces of political representation and state power remains problematic.The absence of stable links to larger structures also undercuts the visibility ofgroups in the public sphere, which is essential to their gaining recognition aslegitimate actors and claimants of rights and goods.


The theoretical problem is to discern what there is about the way in which em-powerment was sought, representation constructed, or connections built by ur-ban movements that has self-limiting or perhaps self-destructive qualities. Ourworking concept of representation must be broad enough to encompass bothgroups and formal political structures. We also need to understand how the fateof groups and protest is related to the issue of formal, electoral mechanisms ofrepresentation—in other words, how elections, electoral mechanisms, and pre-election politics (candidate selection, district boundaries, voting systems) are re-lated to, and perhaps reinforce, patterns within groups. Other institutional mat-ters, most notably the impact of judicial and penal systems, are also vital, especiallyfor considerations of security of property and persons.

Our earlier review of movements and politics in our two cases showed thatalthough the party-focused model of organization was clearly stronger in Vene-zuela than Peru, in both countries, the decay (or, in Peru, the failed consolida-tion) of that model (and of its controlling norms) had contradictory effects. Thelong process of organizational deterioration in Venezuela set many potentialclients free from party controls while opening the field for new kinds of groupsoperating in newly created political spaces. Cases in point include the expansionof urban neighborhood movements, the impact of new electoral rules on the de-velopment of different styles of representation, and the emergence of a range ofgroups and federations self-consciously identified as civil society. In Peru, whereparties were never that strong to begin with, the surge of urban growth (fueledby internal war) overwhelmed older structures and spawned a proliferation ofurban groups of all kinds—unified by their common need to solve urgent andimmediate problems of housing, food, transport, education, and violence. Lack-ing reliable interlocutors and regular access to channels of influence and state re-sources, the connection between particular causes and concerns and more gen-eral political affiliations is hard for most people to identify, much less sustain.

In both countries, new urban citizen movements arose to address very specificneeds created by the urban context and the deteriorating economic situation. Sat-isfying needs required some rearrangement of the relevant institutions and polit-ical spaces, and led to campaigns for political and electoral reform. Building theseconnections and sustaining these campaigns requires allies and patrons: leadersand groups who can provide and manage access. There is a fine line here betweensustaining empowerment and falling into time-honored clientelist patterns, andthe line is easily blurred. One need not have the complex pattern of dependenceof PRI-controlled Mexico at its height (Eckstein 1977) to recognize that groupsand communities need allies in the state and the larger political arena and thatthese allies may and likely do have other agendas. State or party control of re-sources is critical here, hence the critical role often played by NGOs with auto-nomous resources in freeing groups from dependence on parties. The middle-class character of many of the neighborhood movements in Venezuela provides aroughly equivalent independence.


The institutional reforms put in place in Venezuela (in the 1990s) and Peru (inthe 1980s) opened new possibilities for organization, representation, and action.These possibilities were taken up with great vigor in both countries: localgovernments were energized, and a profound process of political de- and re-alignment got under way. But gains proved short-lived, and both Fujimori andChávez worked to recentralize politics, curtailing and limiting the reforms thathad gathered force in earlier periods. The post-1992 Fujimori governmentsmoved more and more into a populist mode, making citizen groups dependenton the state, and restricting independent access to resources. The Chávez regimemade an effort (successful for a while) to bypass formal processes of interest me-diation or representation in favor of a direct relation between the leader and thepeople (Kornblith 1993; Levine 2003; Salamanca 2004). This was a setback forthe autonomy of social movements, and for decentralization, which had providedthem with viable arenas for mobilization and action. In both countries the pres-ence of NGOs weakened, as many transnational groups turned their attentionand resources to newly opened fields of action in central and eastern Europe.

Much of the reevaluation of work on urban movements and empowermenthas been linked to the literature on transitions to democracy and democratiza-tion. But more is at issue than regime change. We believe that the issues can bemore effectively situated in a broad context of thinking about activism and socialmovements, and institutions. Notable cases of transitions to democracy presentthe following anomaly: citizen mobilization and new citizen groups that wereprominent in campaigning for democratization declined, split, and often simplydisappeared with the restoration of democracy. The anomaly lies not only in de-cline, which makes sense, given the availability of channels of action and of com-petition for resources and for supporters. Although decline was in all likelihoodinevitable, the process was accelerated in key cases by naive and unworkable un-derstandings of politics, and by untrustworthy and unreliable political allies. Withthe possible exception of Brazil, where the PT (Partido dos Trabalhadores, orWorkers’ Party) has clear roots in the popular movements and has grown steadilyat all levels, the common experience has been one of division and betrayal(cf. Blondet 1991; Lander 1995; Levine and Crisp 1998; Levine and Stoll 1997).

If we reframe the problem in terms of activism and social movements, theanomaly presented by activism with disempowerment is easier to understand.Two points are critical. First, movements commonly emerge, grow, succeed orfail, and decline, moving through what Tarrow (1994, 156) terms a “cycle ofprotest.” “What is distinctive about such periods,” he writes, “is not that entiresocieties ‘rise’ in the same direction at the same time [they seldom do]; or that par-ticular population groups act in the same way over and over, but that the demon-stration effects of collective action on the part of a small group of ‘early risers’ trig-gers a variety of processes of diffusion, extension, imitation, and reaction amonggroups that are normally quiescent.” In this light, the proper question is not somuch why groups do not survive, but what, if any, legacy they leave in new rules,


expectations, or capabilities. The second point is connected and has to do withthe opportunity structure that urban citizens face—resources and institutionalchannels available, accountability, and access. Writers like Castañeda have arguedthat a focus on local-level organization and the delivery of “good government”offers the most promising path to a rebirth of the left and sustained empowermentof popular sectors in Latin America. The record is mixed on both counts. We finda clear legacy of norms about rights and activism, but weakness at making endur-ing and representative connections. More often than not, surges of activism leaveactivists, at the end, at the mercy of a different charismatic leader—new face,same dependence. The record of institutional reform is promising but incom-plete, with a reversal of many reforms. Despite widespread attention to institu-tional design and institutional engineering, failures of accountability are more thenorm than the exception. The institution of provisions for referendum and recallof officials holds possibilities, but does little to address candidate selection, elec-toral rules, or the all too common impunity of police and lack of access to courts.

Urban Spaces and Urban Citizen Movements

The preceding considerations bring us to a closer look at urban spaces and urbancitizen movements: to the city as a stage or arena for action, and to its citizens asactors. Both dimensions are important. As in much of Latin America (and theThird World as a whole), the urban context in Venezuela and Peru is marked bydominant capital cities and explosive growth, with bigger cities growing fasterthan smaller ones and all cities faster than rural sectors. In recent years, regionalcities have experienced substantial expansion. Internal migration is the predom-inant motor of urban growth in both countries. In Venezuela, rural poverty, roadconstruction, and urban investment paid for by petroleum sparked a process ofmigration, beginning in the 1930s, that has substantially emptied the countryside.Explosive urban growth came later in Peru, but when it came, it was magnifiedby extreme rural poverty and internal war that drove refugees to seek safety in thecities. Rural-to-urban migration in Peru produced a mixing of ethnic groups ona scale unknown in the past: people of highland Indian culture came to Lima,bringing cultural expressions (such as Andean music or the Quechua language)with them.

In both countries, the new presence of migrants overwhelmed urban infra-structures (particularly in the capital cities of Lima and Caracas), creating urgentneeds for water, transport, education, and other services, and, of course, for rep-resentation.7 The spatial configuration of urban expansion, and the availabilityof transport within the city, has had notable impact on organization, citizenmovements, and empowerment. Our analysis of the emergence and problems ofurban movements is a structural one: following Eckstein’s pathbreaking work onMexico City (1977), we situate movements in a context created by the political


opportunity structure of nation and city, and given specific form and content bythe availability of resources and by those (NGOs, unions, political parties) pres-ent and competing to provide services, orientation, and leadership. This strategymakes theoretical sense. It also makes practical sense, given the difficulty of ar-riving at reliable estimates of the numbers of movements and associations exist-ing and active at any given time.8 We provide estimates where possible, but urgecaution in relying solely on the numbers.

In Peru, the career of urban movements did not follow the track of othersocial movements in the sense of a steady accumulation of forces. Rather, aftereach successful mobilization, the movements seemed to fade away. As one localleader said in despair, once electricity was obtained and public lighting was inplace, “They buy a TV set and stay at home.” The same thing happened afterstruggling to get water and sewage for the neighborhood and getting their housesconnected to the main service. Urban movements gained significance and pres-ence in Peru during the 1970s, when the public space was reduced by the pres-ence of a military regime, with elections possible only within the private spherewith voluntary organizations free to assemble and elect their leaders. The stateregulated these elections and acknowledged the right of elected leaders to nego-tiate for public services. These electoral practices and the experience of repre-sentation were important for the creation of an autonomous public space withinthe authoritarian regime. Toward the end of the decade, attempts were made tocentralize neighborhood organizations in Lima.

How can we best understand the empowerment of urban actors in the 1970sin the context of a changing political system? What was the meaning of the powerthat was being generated in these neighborhoods? From the perspective of classtheories of accumulation of power, this was clearly a process of gradual upwardsocial mobility, not a major transformation of power relations. But from the per-spective of building citizenship, there was indeed a significant change in terms ofpower: once subjects or clients, members of the movements became citizens withrights. The pursuit and exercise of political rights in the cities is conditioned onrefashioning the cities as political arenas not only for protest (claiming spaces) butalso as venues for classic kinds of representation, including the creation of rela-tively autonomous units of government. With the end of military rule and theadvent of democratic politics in 1980, municipal elections opened a public elec-toral space for movements. Many former movement leaders became mayors orcity council members.9

During the 1980s, the core agenda of major urban movements underwent anotable change in Peru. Housing and public services eroded, and former migrantshad new issues of concern, new demands to press. Tanaka (1998, 117) notes thatstruggle was centered increasingly within the private sector: “Achieving basicservices and the consolidation of the urban scene has changed in a radical waythe priorities of the urban settlers (pobladores), giving rise to a new pattern of


meaning regarding participation, collective action, and membership in organiza-tions. Attaining public goods lost its centrality and needs related to private goodsbecame more important.” Currently, most new land invasions lead to confronta-tions with groups of working-class landholders associated in cooperatives, insteadof the earlier struggles against big urban landholders or the state leading to face-to-face struggles more than initiating social movements sustained in collective be-liefs. In some sense, urban demobilization responds to urban development andto municipal administration of once self-managed neighborhoods. Cities growslowly into the margins of the old invasions or climb higher into the hills, in theprocess transforming single-family houses into multi-family dwellings.

In Lima, mobilizations have typically had very concrete goals (garbage collec-tion, security, housing and land titles, water, electricity, parks and green spaces)that were easily assumed by municipal governments. Candidates for municipalelections are now commonly seen as potential experts on city managementrather than as mobilizers or politicians, and provision of public services has be-come a core issue of campaigns. This has contributed to the multiplicationof candidates and the short lives of many local movements. Once the neighbor-hood is converted into a municipality, the local voluntary organizations resemblethose of any other part of the city: sport clubs, cultural associations, schoolparents’ associations, Christian communities, market vendors, teachers’ unions,and so on.

As political and economic crises became more acute throughout the 1980s, themore dynamic movements became those around survival: those dedicated toproviding food, resisting unemployment, and literally defending life from bothterrorism and an arbitrary, repressive state. Social actors that were organized andcould participate in public demands were mostly middle class: teachers, nurses,medical doctors, public employees, and public transportation workers. The sus-tained economic crisis, ties to foreign debt, and repeated structural adjustmentpackages weakened both businesses and unions. The result, throughout the1980s, was a growing demobilization of the masses, aided by a deadly mix of ter-ror and repression that began to rise sharply after 1980 with the sudden appear-ance of Shining Path on the national scene.

The particular character of this crisis helps explain the prominence of survivalorganizations in city life. The “Glass of Milk” municipal program, formed un-der Alfonso Barrantes, the Lima mayor elected as a member of United Left in1983, distributed a million glasses of milk every day nationwide to preschoolchildren, mothers with newborns, and later to tuberculosis patients. There werealso soup kitchens, known as “popular dining rooms” (comedores populares), someself-managed, and others sponsored by Catholic parishes and party-influencedorganizations such as the APRA-linked “Mothers’ Clubs” and the “popularkitchens” sponsored by AP. The most common pattern was that a group ofwomen got together, cooking in one of the members’ homes and selling meals


for a nominal price to their members, who would then pick up the meals andtake them home to eat with their families. Complementary aid came fromNGOs or the state, sometimes through the donation of cooking equipment suchas stoves or pots and pans, and also through the regular provision of food,including oil, rice, or wheat. Voluntary work by members, and their own con-tributions in financing the food, is central to the operation of this kind of organi-zation, and may be a reason why they often do not reach the poorest familiesin neighborhoods.

These and similar organizations changed how politics was conducted and rep-resentation conceived. In September 1988, after the first wave of structural ad-justment policies, they organized a huge mobilization under the slogan “Protestacon propuesta” (Protest with Proposals). They demanded support from the gov-ernment to buy food from local producers instead of importing it from abroad.This linked aid for the poor to rural development, joining the agenda of urbanmovements to peasant demands. Soon after, the Church started the EmergencySocial Program (PSE), offering resources to channel international support for thepoor. The program opened a public space where the leadership of the comedores,the entrepreneurial association CONFIEP, NGOs, and the Catholic Churchworked together to elaborate an emergency program. This program continuedunder the Fujimori government as the Social Emergency Program (PES).

In 1990, the leadership of the comedores decided to institutionalize their rightto receive public funding to feed the poor. The various organizations (indepen-dent and related to political parties) joined together toward this aim, lobbied leg-islators from different parties, and achieved their goal at the end of the year witha law, promulgated by Fujimori in 1991, that recognized the responsibility of thestate to feed the poor. A successful organization of comedores led the women intothe streets to oppose the terrorist movement Shining Path. In a mobilization atEl Agustino, they marched against the general strike called by Shining Path,chanting “ni con hambre ni con balas” (neither hunger nor bullets) to proclaimtheir autonomy and courage. Events like this cost many people their lives, eitherduring the protests or in their aftermath.

Once the distribution of food became centralized by the government in themid-1990s, the leadership, which was usually elected or rotated among the mem-bers of the associations, was replaced by personnel from the same organizationsbut loyal to the Fujimori regime. There were over two thousand self-managedcomedores, three thousand Mothers’ Clubs, and more than seven thousand Glassof Milk committees in Lima alone. Many political cadres from these organiza-tions joined the political movements Fujimori created for each new election as“independents.” Women candidates, leaders of the soup kitchens, were electedas council members in the municipalities, aiming to represent their organizations’own interests, but ending up as part of the political establishment, dragging theirformer constituency with them.


Throughout the 1990s, continued violence and economic crisis undercut thevitality of urban organizations, making it difficult to hold open assemblies and dis-cussions in the neighborhoods, and hard to elect new leaders. Urban citizenmovements were caught between terrorists, on the one hand, and a repressive andcontrolling government, on the other. The main goal of terrorist organizations,most notably Shining Path, was to control territory within the city, and neutral-ize, co-opt, or eliminate competing groups and leaders. For the same reasons, inthe name of national security, a central goal of the regime was to control theneighborhoods and establish secure ties with the population. Authorities weresuspicious of autonomous organizations; disempowerment and control were thedominant state strategies. Major mobilizations during the early 1990s were linkedto the killing of grassroots leaders and to massive demonstrations of solidarity, of-ten around funerals, as was the case following the public assassination of MaríaElena Moyano by Shining Path. Later, with the coming of peace, there was a no-table political vacuum since no political parties were working among grassrootsgroups and only the Catholic Church, the evangelicals, and NGOs remained toorganize what was left of civil society. It was only toward the end of the secondFujimori government (1995–2000) that the people recovered the streets andpublic squares as arenas for assembly and protest.

In Venezuela, as in Peru, urban space (above all, the streets, plazas, and neigh-borhoods of the capital city) is a prime arena for political action of all kinds: fromrallies, demonstrations, and marches, to street fighting. Urban mobilizationsplayed a central role in the overthrow of the country’s last dictator, Marcos PérezJiménez, in January 1958. Urban land invasions and the formation of vast newshantytowns remained a prominent feature of city life through the early 1960s,but have since faded. As noted earlier, a different kind of urban movement cameonto the national scene decades later with the emergence of “civil society” asan actor in national politics and the concerted drive to create spaces and vehiclesfor that action. Neighborhood associations (vecinos) were formed, with the initialgoal of urban development and defending property rights. Their agenda soonexpanded to include pressure for greater municipal autonomy, and the fiscaland electoral reforms this entailed. Early neighborhood associations began inthe 1970s in a series of middle-class areas of Caracas. FACUR (the Federación deAssociaciones de Comunidades Urbanas, or the Federation of Urban Commu-nity Associations) was established in 1971 as a coordinating body for these asso-ciations. FACUR provided a model for associations and similar regional federa-tions that soon began to spring up all across the country. By the early 1990s, therewere federations in every state, which together grouped an estimated total ofabout fifteen thousand associations. In 1987 the neighborhood movement suc-ceeded in gathering 140,000 signatures on petitions asking for a reform of thebasic law governing municipalities (Ley Orgánica del Regimen Municipal, orLORM). This was one of the most important nonviolent mobilizations to that


date in Venezuela. Changes to the law included the election of governors, theelection of mayors, the creation of parish councils, and the possibility of recallingofficials.

The impact of the movement was magnified by the school for neighborhoodgroups, the Escuela de Vecinos de Venezuela, or EVV. The EVV arose out ofclasses within FACUR, and consolidated on a national level in the mid-1980swith important support from business and from national and internationalNGOs. Since that time, the EVV has established regional offices, mounted regu-lar programs of courses for associations and local public officials, and maintaineda range of correspondence courses, periodic meetings, and media presentations.EVV leaders have generally resisted pressure to form a political party, preferringinstead to spin off a series of pressure groups, each devoted to a specific issue. Ex-amples include Queremos Elegir (We Want to Elect), a group devoted to elec-toral reform; Fiscales Electorales de Venezuela (Electoral Officials of Venezuela),dedicated to promoting citizen involvement in supervising voting sites; and Ven-ezuela 2020, an organization that promotes workshops and roundtables con-cerned with the shape of the country’s future. In other words, not a party butsomething more like “civil society” (García Guadilla and Silva Querales 1999;Gómez Calcaño 1998; Lander 1995; Levine 1994; Levine, Crisp, and Rey 1996;Salamanca 2004).

The term “civil society” came into wide use in Venezuela only in the 1990s.Until then, the political parties founded in the 1940s and the political systemconsolidated around them after 1958 encapsulated the expression of organizedsocial life through party-controlled networks. Much contemporary theorizing(Escobar and Alvarez 1992) depicts the “emergence of civil society” as, above all,defensive. The neighborhood movement, which began as uncoordinated effortsby urban middle-class citizens to resist unplanned city growth and to defendtheir neighborhoods, is a case in point. The emergence of the human rightsmovement is another. Human rights organizations began to appear in the 1980sin response to specific abuses and to challenge long-standing practices of official(especially police) impunity.10 They gained national stature and impact in thewake of the “Caracazo” (massive riots in Caracas) of February 27, 1989. Mount-ing violence throughout the 1990s kept them in the public spotlight. What thesegroups had in common was an effort to mobilize opinion (and people) outsidethe existing network of organizations controlled by the country’s political par-ties.11 Through the 1990s, as the political crisis grew and political parties wereblamed by many Venezuelans for all the country’s problems, “civil society” be-came a catchall banner for reform and right-thinking activism.

Once in power, the Chávez government made an effort to put its rhetoric ofparticipatory (as opposed to representative) democracy into practice througha series of provisions in the constitution that make a place, at least in theory,for the active participation of “civil society” in politics. The 1999 Bolivarian


Constitution, for example, provides that the legislative councils of the differentstates “consult with” civil society on matters of interest to the states (Art. 206)and that “civil society” nominate three members of the national Electoral Coun-cil, charged with managing elections (Art. 296). Similar provisions are scatteredelsewhere in the constitutional text. As a practical matter, disputes about how todefine “civil society” made it almost impossible to figure out who could andshould be recognized as speaking in its name. The steadily increasing polariza-tion of the country has also made “civil society” a highly contested term: bothpro- and anti-Chávez groups claim to speak in its name, denying legitimacy andauthenticity to the other. The results are occasionally anomalous: for example,although in theory civil society was to participate in evaluating candidates forthe “citizen power,” in practice the president took this task upon himself, on thegrounds that because a majority of the population had voted for him, he was theproper representative of civil society (Salamanca 2004). More often, lately, the re-sults are confrontational and too often deadly, with groups clashing in the streets.

It is not easy to come up with reliable estimates of the scale of the phenome-non. One review estimated that the total number of civic associations in Vene-zuela ranges from about twenty-five thousand to about fifty-four thousand (Sala-manca 2004). Of these, the largest proportion are neighborhood associations,with a substantial number of groups that specialize in promotion and develop-ment, working with government and international resources. There is also astrong but regionally concentrated cooperative movement, and a significantalthough numerically small network of human rights organizations.

Not all civic associations are mobilizational in character or intent. There aremusic groups, civic theaters, cooperatives, and sports clubs, and a host of relatedgroups whose logic and daily life need have little to do with mobilization andpolitical confrontation. But these and other groups are linked to politics (andthus to the state) in two important ways that draw them into the partisan arena.First, many if not most seek and receive resources from the state. Even in timesof economic decline, the Venezuelan state remains a powerful source of financ-ing and material resources for groups of all kinds. Second, the steadily increas-ing rhythm of mobilization and polarization since the late 1990s has made itdifficult for groups to keep apart from political division and on the margins ofconfrontation. Indeed, the years from 1989 to the present are arguably the mostprotest-filled period in the last one hundred years of Venezuelan history: onemassive urban uprising, two attempted coups, the impeachment and removal ofone president, and a rising tide of violent actions in the universities and on thestreets.

Protest surged following the Caracazo in February 1989, and after a shortrespite under the second government of Rafael Caldera (1993–98), the rhythmof demonstrations, marches, and street protests picked up again as the countryentered a new electoral cycle. To be sure, urban protests, often violent, had never


completely gone away. Student activism, sparked by regular violent actions un-der the leadership of encapuchados (literally, “hooded ones,” students with hoodsto shield their identities), regularly spilled over from campuses into the streets.Protest and the scope of confrontation broadened with the election of Chávezas president. López Maya (2002) shows that, among the kinds of protest, con-frontational actions showed the strongest increase in 1999. Her figures of coursedo not include events beginning in late 2001, when protest grew and mobiliza-tions, marches, and clashes became the daily bread of urban life, not only in thecapital city of Caracas, but throughout the country: mobilization and counter-mobilization, rally and counter-rally, with massive marches following one an-other at ever shorter intervals. Protests, occupations of buildings, and coordi-nated actions involving banging of pots and pans (cacerolazos) or blowing ofwhistles or car horns (bocinazos) became everyday occurrences.

Events reached their first crisis point in the bloody confrontations of April11–14, 2001, when snipers fired on a huge march making its way through Cara-cas to the presidential palace. The president was ousted and returned to office afew days later. After a brief respite while all sides stepped back from the brink,protests, marches, and countermarches—this time all over the country—beganagain, coming to a second crisis point in the civic strike that began at the endof 2002. The leadership and organizational backbone for the opposition evolvedquickly, starting with a pact between the trade-union movement, the businessfederation, and the Catholic Church. Union leaders, fresh from defeating thegovernment in a referendum, played a critical role in the day-to-day organizingof protest activities. They were soon joined by political party activists, humanrights groups, and others, as a range of new coordinating groups were put together (e.g., the opposition Coordinadora Democrática, or Democratic Co-ordinator, in the summer of 2002). For the present purposes, the most striking features of this whole process are the central roles played by old-line organiza-tions such as business and union federations, how efforts to resolve protest areundermined by the weakness of leaders on all sides, and the predominance ofextremists, free to act given the utter demise of a professional political class usedto negotiation and compromise.

Religion, Mobilization, and Public Space in Peru

The Catholic Church in a number of South American countries opened a pub-lic space to meet, associate, and participate under dictatorial regimes. When otherpublic spaces were closed and forbidden, assemblies or Christian prayer groupswere open for the faithful, and were used by citizens to exchange information,listen to others, form opinions, and circulate rumors, ironic stories, and hopes forthe future. In Peru, this role has been secondary. Even in the worst times ofthe dictatorship, between 1968 and 1980, freedom of association and of public


assembly were accepted, although state-sponsored organizations regularly com-peted with autonomous ones for members and public voice.

In this context, Christian communities offered a complementary space for association and critical reflection, contributing to the quality of leadershipand stimulating involvement in other organizations and in the political realm.There was a clear distinction between the public space of religion and the pub-lic space of politics. Committed Christians were, for the most part, careful to acton their own and not in the name of their particular church, Catholic or evan-gelical. This experience underscores the role religion can play in empowermentand disempowerment, but not necessarily in building representation.

Until 1980, Catholicism was the established church in Peru and the state wasconfessional. The Catholic Church expected to play a prominent public role.These expectations were not abandoned with the official separation of churchand state in the 1980 Constitution. The Church’s prominent role in the struggleagainst poverty, and later in the promotion and defense of human rights, openednew areas for common action with other organizations in civil society, as wellas with international agencies sharing the same goals. In 1988 the Church cre-ated a space for bringing together different actors in the Emergency Social Pro-gram, including international cooperation agencies, business entrepreneurs, andgrassroots leadership. Later, when repression and terrorism continued through-out the nation, the Church supported human rights organizations, putting itsnewly gained religious legitimacy behind its pastoral agents—clergy, nuns, andlaity—to care for the relatives of people missing and tortured and the innocentin prison. Drzewieniecki (2001, 4) writes:

In many parts of the country, clergy and Catholic lay workers developed new, more egal-itarian ways to work with the poor through parishes and the expanding network ofChristian base communities. These Catholic activists as well as CEAS (Comisión Epi-scopal de Acción Social or Episcopal Commission for Social Action), whose human rightsdepartment was founded in 1976, played a very important role in the development of hu-man rights work in many different areas of the country. CEAS became one of the mostimportant human rights organizations in the country and played an important role in thecreation and institutionalization of the Coordinadora Nacional de Derechos Humanos[National Human Rights Coordinator].

During the Jubileum campaign, called by Pope John Paul II and joined by anecumenical movement to lobby for forgiveness of the foreign debt to the poorestcountries, the network of base Christian communities in Peru under the leader-ship of CEAS (Comisión Episcopal de Acción Social) collected the largest num-ber of signatures among the participating countries around the world. In Peru,the same network provided volunteers for Transparency, an NGO formed to pro-mote fair elections through election observation. In the 2000 and 2001 elections,the volunteers played a key role.


Commitment and involvement come together in compromiso (commitment),a key word for Christians throughout the Catholic Church in Peru. The kindof compromiso at issue here works more for participation in general than for in-volvement in politics specifically. A clear distinction is drawn between the po-litical public sphere in which people can participate as citizens, and the socialpublic sphere created through the creation and practice of organizational life.This distinction is reinforced by the current experience of participation inchurch communities and parishes, and in social movements that remain alien-ated from institutional channels for representation.

The Catholic Church’s option for the poor has contributed to a growingawareness of identity embedded in common interests and culture that cuts acrossdifferent classes and ethnic divides, and even across different parties. But this re-ligious awareness has not had a similar intellectual elaboration in other fieldssuch as literature or politics. Institutional politics and modern culture remaindistant from the recently included citizens who often feel themselves marginal-ized or alienated from public agendas. Demands for cultural representation atthe institutional level—in Congress, government, the arts, and mass media—have been added to those of economic interests. And newly appointed bishopsin such important cities as Lima, Arequipa, and Trujillo are not helping to bridgethe gap or to link the elite with the citizenry.

Disempowerment as the Future of Empowerment

The combination of activism and massive citizen mobilization with disempow-erment joins the social and political trajectories of Venezuela and Peru in an un-expected convergence. From different starting points and the most varied social,organizational, and political traditions, these two nations have arrived at a sharedspace that does not augur well for citizen empowerment or representation. Thedecay of political institutions, including but not limited to political parties, hasleft Venezuelans and Peruvians with space for the creation of civil society—a space they have filled, as we have seen, with great energy and creativity. Butin the absence of reliable and trusted political intermediaries—either formal in-stitutions or political parties—these energies are rarely converted into sustainedand authentic representation. Is disempowerment the future of empowerment?Reading back from recent waves of massive, mostly urban mobilizations in eachcountry (anti-Fujimori in Peru, and pro- and anti-Chávez in Venezuela) mayprovide some clues.

The decay of institutions is arguably greater in Peru, where the all-encompassing corruption of the Fujimori-Montesinos system of rule spread dis-credit very widely. Taking a longer view, the very idea of citizenship remainsuncertain in Peru: although voting is obligatory, the regular rhythm of alterna-tion between authoritarian and democratic regimes (with change coming about


every ten years or so) has meant that as Peruvian citizens approach legal voting agethey cannot be sure if elections will be held at all, or what the electoral rules willbe. Individual or “civic” forms of citizenship began to take significant form inPeru as a result of the struggle against Shining Path and the MRTA (MovimientoRevolucionario Tupac Amaru, or Tupac Amaru Revolutionary Movement). Ear-lier movements have conquered citizen rights in the areas of economic and socialrights, and later in politics, in the struggle to guarantee rights of association andpublic demonstrations. Individual rights have been third-generation, reversingthe order outlined by Marshall (1983). With strong links to transnational humanrights networks, civic activists developed networks and actions within Peru thatsoon gained international impact (see Burgerman 2001 on El Salvador andGuatemala). Although twenty years of activism have earned these groups consid-erable social recognition, such that they are now considered an important part ofcivil society, such organizations remain geared to working with victims, and notto building a membership base.

The end of authoritarian rule and the return of democratic institutions andthe rule of law has not cleared the slate of human rights issues. Much remainson the agenda, as the creation in Peru of a Truth Commission to review the pasttwenty years demonstrates. Human rights groups have begun to broaden theiragendas to include social and economic rights as an integral component of hu-man rights. Activism and pressure has also continued, with specific concern forthe country’s political transition—the dismantling of the Fujimori regime andthe reconstruction of democracy and political rights. Explicitly “democratic”groups have been formed, first among university students in opposition to Fu-jimori’s reelection and in defense of the Constitutional Tribunal that rejectedhis (and the Congress’s) efforts to provide a basis for his reelection (Tanaka, thisvolume). These struggles were reinforced by the efforts of women’s collectives(Mujeres por la Democracia), as well as by artists and people from the media, insuch groups as Resistencia, with a creative adaptation of forms of protest fromother countries, such as the weekly washing of the national flag, sweeping thearea in front of the Congress, or the mounting of “Walls of Shame” in variousplaces, where passersby could post their ideas, photos, drawings, or commen-tary. There was also a series of street actions, rallies, and demonstrations in theplazas of Lima. Many of these elements came together under the direction ofopposition political parties in the massive Marcha de los 4 Suyos, held on thevery day on which Fujimori was sworn in for his third, and short-lived, presi-dential term.

To this point, the effort to construct democratic institutions in Peru has placedemphasis more on institutions than on resources or actors. The new cutting edgeof urban struggles and mobilizations is taking place outside Lima. In Iquitos, theRegional Front, strengthened during the border negotiations with Ecuador car-ried out under Fujimori, is advancing claims for resources for development and


demanding that benefits go to Peruvians before they go to Ecuadorans. There arealso active regional fronts in Tacna (on the frontier with Chile), in Puno, and inMadre de Dios (on the frontier with Bolivia). These are not urban movementsbecause they include peasants, but they are organized primarily in provincial citiesby businesspeople, academics (from the local universities), and local officials. Re-cently initiated processes of decentralization have added dynamism and resourcesto these organizations, and in this way have opened spaces for the emergence ofnew groups and leaders.

In Venezuela, new citizen movements and forms of protest, indeed the veryidea of “civil society” as an autonomous space for organization and action, ap-peared within an already established democracy. Their goal was not to challengeor overturn authoritarian rule, but rather to broaden or deepen that democracyby loosening the constraints imposed by a moribund but still all-controlling set of institutions and tacit rules centered on the political parties. The historicaltrack thus differs from Peru, but the resulting situation is surprisingly similar.The creation of a movement around Hugo Chávez Frías, his rise to power, andthe implementation of a “Bolivarian revolution” drew strength from the dis-credit of the old system and the implicit association of the movement with “civilsociety,” at least in rhetorical terms. The whole process makes sense as part of a general onslaught on the old system, its institutions, and its operative rules.The very word “representative” barely appears in the Bolivarian Constitutionof 1999. Instead, Venezuelan democracy is “and always will be democratic, par-ticipative, elective, decentralized, alternative, responsible, pluralist, and withrevocable mandates” (Art. 6). The results have been meager. New institutionshave either foundered or never made it off the drawing board.

In both countries steep economic decline and political deadlock followingclose on the heels of apparent euphoria (the resignation of Fujimori, the electionof Toledo, the victory of Chávez, his removal and restoration in 2002, surgingopposition, and growing violence) have combined to make sustained activismharder for many ordinary people. As a result of the extraordinary Venezuelancivic strike, individuals, businesses, movements, and national finances sufferedmajor costs. Although the decay of parties sets groups in both countries free, inthe same measure it sets them adrift and leaves them at the mercy of a supposedlydirect relation to the leader, whoever that may be. Civil society constructed inthis way is unlikely to yield enduring organization, and all too likely to be de-pendent on, and ultimately betrayed by, personalist elites as unaccountable astheir populist predecessors. Without strong and durable organization, “civil so-ciety” is unlikely to provide coherence and direction for a complex and conflict-ridden society.

The preceding observations underscore two points. First, it is clear that a con-ventional focus on representation is not adequate to capture the experience ofsocial movements, empowerment, and disempowerment that we have recounted


here. These movements operate on a different terrain, a terrain they help tocreate precisely because conventional vehicles of democratic representation aredifficult to access and fail to satisfy their expressed needs. At the same time, it isimportant to acknowledge that the failure of groups self-defined as “civil society”to negotiate the transition from mobilization to formal political representationweakens the groups, weakens the representation (including the formal system),and weakens democracy by disillusioning many who put their hopes in organi-zation as a way of creating something authentically democratic.

It will not be easy to solve the puzzle of mobilization with disempowerment.Part of the difficulty is practical: obstacles of all kinds litter the path of those whotry. There are also theoretical problems to address. Much thinking about em-powerment, citizenship, and representation is caught somewhere in betweenreflection on social movements and analysis of institutional design, on the onehand, and examination of efforts to expand citizen access and participation inalready existing arenas, on the other. We believe that the problem needs to berefocused on specific ways to provide enduring and legitimate form to this newparticipation—in other words, to institutionalize it. This is beginning to hap-pen in Peru through the Round Tables for the Fight against Poverty, establishedat district, provincial, and departmental levels. There has also been the NationalAccord, with participation of political parties, regional fronts, and the Mesas deConcertación (Negotiation Roundtables); and the Truth Commission. All thesecan be understood as kind of an end run around existing political spaces, insti-tutionalizing empowerment in new places. In Venezuela, despite provisions inthe 1999 Bolivarian Constitution for referenda and citizen-managed forums ofall kinds, none of this has been institutionalized. The mounting polarization andthe acute political crisis of the country has, if anything, exacerbated the frag-mentation of political forces to such a point that agreement on such spaces (letalone on who should participate in them) is very unlikely. As we noted earlier,the decentralization initiatives of the 1990s have been cut short, and some insti-tutional reforms (e.g., of the electoral system) have aggravated, rather than ame-liorated, problems of representation.

The future of urban citizen movements does not look as good as it used to,and it is not easy to be optimistic, at least not in the short or medium term. Itis important to be clear about the core problem. That organizations fail andleadership is unreliable or manipulative is itself nothing new. The difficulty formovements, and for the potential for democratic representation in and throughthem, is not so much in the survival of any given organization as such, but ratherin the creation of a kind of institutional safety net, something for groups to fallback on when times are hard. This is the role played, and played well, through-out Latin America by foreign-financed NGOs, whose monetary and organiza-tional autonomy provides an invaluable cushion. Such a role is long-standing inPeru and only now getting under way in Venezuela. Scholars also need to learn


what the members of movements know very well: that the agenda of urban citizen movements has changed. At issue is less the traditional range of urban demands: land, water, housing, transport, education, and security. Such con-cerns of course remain, but now and in the foreseeable future the pursuit andexercise of political rights in the cities are conditioned on refashioning the citiesnot only as political arenas for protest (claiming spaces), but also as venues forclassic kinds of representation. Only with these in place, and viable connectionsto other levels of government, can the puzzle begin to be solved.

Notes

1. On Peru, see, among others, Tanaka (1998, this volume); on Venezuela, amongothers, see Coppedge (2002), Molina (2002), Levine (2002, 2003), López Maya (2002),and the sources these authors cite.

2. The problem of violence and its aftermath is a central issue for Peruvian politicsover the last quarter century. On the scale of the violence and its aftermath, see the finalreport of the Commission of Truth and Reconciliation (Comisión de la Verdad y Re-conciliación 2004). On the scale of recent patterns of violence and protest in Venezuela,see Hernández (2002) and López Maya (2002).

3. The process is reminiscent of the rebirth of mobilization and protest in Chile,which Garretón calls the “invisible transition” that led up to the referendum that endedthe Pinochet regime (Garretón 1989).

4. The name comes from the four regions, or Suyos, of the Inca empire, or Tahuantinsuyo.

5. Unionized workers are only a small proportion of the total workforce, and the busi-ness federation is of course not a mass organization. The two made common cause, draw-ing on the union federation’s successful defeat of Chávez forces in the union referendumof fall 2001 and the business federation’s strong opposition to a package of decree laws an-nounced around the same time. Relations between the church and the government hadbeen tense for some time, inflamed by the president’s own erratic rhetoric (calling priests“devils in cassocks,” e.g.) and by his program for control and inspection of private edu-cation. The mass media have been a favorite target of the government since the begin-ning, and with rare exceptions, have responded in kind.

6. See Oxhorn (2001). “Efforts must be systematically undertaken at the grassrootslevel to begin to empower people by helping them to be proud of who they are—re-gardless of their social class, gender, ethnicity, religion, and so on. Studies have alreadydemonstrated the success of such efforts to overcome people’s symbolic exclusion”(14 –15).

7. The pressure was such in Peru that a well-known book by José Matos Mar (1984)is entitled Popular Overflow and the Crisis of the State in Peru (Desborde popular y crisis del estado en el Peru).

8. Contested definitions of what counts as civil society mean that such numbers remain in dispute (Oxhorn 2001; Salamanca 2004).

9. Representation was more effective when there were multiple electoral districts, aswas the case in 1985 and 1990, since a single nationwide electoral district (1992, 1995,


and 2000) restricts electoral possibilities to the elites, leaving aside the new leadershipemerging from recent movements.

10. Groups include PROVEA, COFAVIC, and Red de Apoyo Para la Justicia y la Paz;see Levine (1998) for details.

11. See the 1991 Annual Report of PROVEA, a major human rights group, whichstates: “In contrast to earlier years, and basically during and after the National Protests of February 1989, it was possible to confirm that the social spectrum participating inprotests is widening. Now participation in organized protests has opened fields of actionfor new groups: along with students and workers one finds a range of professional asso-ciations and social groups: doctors, nurses, peasants, Indians, firemen, police, culturalworkers, housewives, and neighborhood groups actively joining in movements in defenseof basic rights” (PROVEA 1992, 114 –15).

References

Blondet, Cecilia. 1991. Las mujeres y el poder: Una historia de Villa el Salvador. Lima: Insti-tuto de Estudios Peruanos.

Burgerman, Susan. 2001. Moral Victories: How Activists Provoke Multilateral Action. Ithaca,NY: Cornell University Press.

Comisión de la Verdad y Reconciliación. 2004. Hatun Willakuy: Versión abreviada del in-forme final de la Comisión de la Verdad y Reconciliación del Perú. Lima: Comisión de la Ver-dad y Reconciliación.

Coppedge, Michael. 2002. “Soberanía popular versus democracia liberal en Venezuela.”In Venezuela: Rupturas y continuidades del sistema político (1999–2001), ed. MarisaRamos Rollón, 69–96. Salamanca, Spain: Ediciones Universidad.

Crisp, Brian F. 2000. Democratic Institutional Design: The Powers and Incentives of Venezue-lan Politicians and Interest Groups. Stanford, CA: Stanford University Press.

Crisp, Brian F., Daniel H. Levine, and José E. Molina. 2003. “The Rise and Decline ofCOPEI in Venezuela.” In Christian Democracy in Latin America: Electoral Competitionand Regime Conflicts, ed. Scott Mainwaring and Timothy R. Scully, 275–300. Stan-ford, CA: Stanford University Press.

Dietz, Henry. 1998. Urban Poverty, Political Participation, and the State: Lima, 1970 –1990.Pittsburg: University of Pittsburg Press.

Drzewieniecki, Joanna. 2001. “Coordinadora nacional de derechos humanos: Un estu-dio de caso.” In Cuadernos de investigación social. Cuaderno no. 17. Departamento deCiencias Sociales, Pontificia Universidad Católica del Perú, Lima.

Eckstein, Susan. 1977. The Poverty of Revolution: The State and the Urban Poor in Mexico.Princeton, NJ: Princeton University Press.

Escobar, Arturo, and Sonia Alvarez, eds. 1992. The Making of Social Movements in LatinAmerica: Identity, Strategy, and Democracy. Boulder, CO: Westview Press.

García Guadilla, Maria P., and Nadeska Silva Querales. 1999. “De los movimientos sociales a las redes organizacionales en Venezuela: Estrategias, valores e identidades.”Politeia 23: 7–27.

Garretón, Manuel Antonio. 1989. “Popular Mobilization and the Military Regime inChile: The Complexities of the Invisible Transition.” In Power and Popular Protest:Latin American Social Movements, ed. Susan Eckstein, 259–77. Berkeley: University ofCalifornia Press.


Gómez Calcaño, Luis. 1998. “Civic Organization and Reconstruction of DemocraticLegitimacy in Venezuela.” In Reinventing Legitimacy: Democracy and Political Change inVenezuela, ed. Damarys Canache and Michael Kulisheck, 169–86. Westport, CT:Greenwood Press.

Granovetter, Mark S. 1973. “The Strength of Weak Ties.” American Journal of Sociology78, no. 6: 1360 –80.

Hellman, Judith Adler. 1992. “The Study of New Social Movements in Latin Americaand the Question of Autonomy.” In The Making of Social Movements in Latin America:Identity, Strategy, and Democracy, ed. Arturo Escobar and Sonia Alvarez, 52– 61. Boul-der, CO: Westview Press.

Hernández, Tosca. 2002. “El desafío de la violencia en el actual sistema político venezolano.” In Venezuela: Rupturas y continuidades del sistema político (1999–2001), ed. Marisa Ramos Rollón, 289–311. Salamanca, Spain: Ediciones Universidad.

Kornblith, Miriam. 1999. “Agenda de reformas y crisis sociopolítica en Venezuela: Unadifícil combinación.” Politeia 22: 83–120.

Lander, Edgardo. 1995. Neoliberalismo, sociedad civil, y democracia: Ensayos sobre AméricaLatina y Venezuela. Caracas: Universidad Central de Venezuela, Consejo de DesarrolloCientífico y Humanístico.

Levine, Daniel H. 1994. “Goodbye to Venezuelan Exceptionalism.” Journal of Inter Amer-ican Studies and World Affairs 36, no. 4: 145–82.

———. 1998. “Beyond the Exhaustion of the Model: Survival and Transformation ofDemocracy in Venezuela.” In Reinventing Legitimacy: Democracy and Political Change inVenezuela, ed. Damarys Canache and Michael Kulisheck, 187–214. Westport, CT:Greenwood Press.

———. 2002. “The Decline and Fall of Democracy in Venezuela: Ten Theses.” Bulletinof Latin American Research 21, no. 2 (April): 248– 69.

———. 2003. “El consenso democrático venezolano en dos tiempos, 1972–2002.” Po-liteia 30: 21– 40.

Levine, Daniel H., and Brian Crisp. 1998. “Democratizing the Democracy? Crisis andReform in Venezuela.” Journal of Interamerican Studies and World Affairs 30, no. 2:27– 62.

Levine, Daniel, Brian Crisp, and Juan Carlos Rey. 1996. “El problema de la legitimidaden Venezuela.” Cuestiones Políticas (Venezuela), no. 16: 5– 44.

Levine, Daniel, and David Stoll. 1997. “Bridging the Gap between Empowerment and Power in Latin America.” In Transnational Religion and Fading States, ed. SusanneHoeber Rudolph and James Piscatori, 63–103. Boulder, CO: Westview Press.

López Maya, Margarita. 2002. “Venezuela after the Caracazo: Forms of Protest in a Deinstitutionalized Context.” Bulletin of Latin American Research 21, no. 2 (April):199–218.

Lora, Carmen. 2002. “Sobre lo siniestro en el movimiento de mujeres.” Páginas (Lima),no. 173 (February): 55– 63.

Marshall, T. H. 1983. “Citizenship and Social Class.” In States and Societies, ed. DavidHeld et al., 248– 60. New York: New York University Press.

Matos Mar, José. 1984. Desborde popular y crisis del estado: El nuevo rostro del Perú en la década de 1980. Lima: IEP.

Molina, José Enrique. 2002. “The Presidential and Parliamentary Elections of the Boli-varian Revolution in Venezuela: Change and Continuity (1998–2000).” Bulletin ofLatin American Research 21, no. 2 (April): 219– 47.

Molina, José Enrique, and Carmen Pérez. 2000. “Venezuela ratifica el cambio: Elec-ciones de 2000.” In Venezuela: Rupturas y continuidades del sistema politico (1999–2001),ed. Marisa Ramos Rollón, 143–76. Salamanca, Spain: Ediciones Universidad.


Ortner, Sherry. 1995. “Resistance and the Problem of Ethnographic Refusal.” Compara-tive Studies in Society and History 37, no. 1: 173–93.

Oxhorn, Phillip. 2001. “When Democracy Isn’t All That Democratic: Social Exclusionand the Limits of the Public Sphere in Latin America.” Miami: North South Center.

Piven, Frances Fox, and Richard Cloward. 1977. Poor Peoples’ Movements: Why They Suc-ceed, How They Fail. New York: Vintage.

———. 1998. The Breaking of the American Social Compact. New York: New Press.PROVEA. 1992. Situación de los derechos humanos en Venezuela: Boletín de derechos humanos

y coyuntura. Caracas: Programa Venezolana de Educación y Acción en Derechos Humanos.

Salamanca, Luis. 2004. “Civil Society: Late Bloomers.” In The Unraveling of RepresentativeDemocracy in Venezuela: Toward a New Model of Participation, ed. Jennifer McCoy andDavid J. Myers, 93–115. Baltimore: Johns Hopkins University Press.

Sikkink, Kathryn. 1993. “Human Rights, Principled Issue Networks, and Sovereignty inLatin America.” International Organization 47 (Summer): 411– 41.

Smith, Christian. 1996. Resisting Reagan: The U.S. Central America Peace Movement. Chi-cago: University of Chicago Press.

Stokes, Susan C. 1995. Cultures in Conflict: Social Movements and the State in Peru. Berke-ley: University of California Press.

Tanaka, Martín. 1998. Los espejismos de la democracia: El colapso del sistema de partidos en elPerú, 1980 –1995, en perspectiva comparada. Lima: Instituto de Estudios Peruanos.

Tarrow, Sidney. 1994. Power in Movement: Social Movements, Collective Action, and Politics.Cambridge: Cambridge University Press.

Tovar, Teresa. 1991. “El discreto desencanto frente a los Actores.” Paginas (Lima) 111 (October): 25–39.

Warner, R. Stephen. 1993. “Work in Progress towards a New Paradigm in the Socio-logical Study of Religion in the United States.” American Journal of Sociology 98, no. 5(March): 1044 –93.


The third wave of democratization profoundly raised hopes and shaped op-portunities for political representation. Yet in the wake of authoritarian regimes,the creation of new electoral institutions, the revival (and formation) of politicalparties, and renewed respect for human rights, much of Latin America appearsto be suffering from a crisis of representation. This is evident not only in a di-verse set of new democracies (e.g., Ecuador, Bolivia, Guatemala, Peru, and Ar-gentina), but also in an older and smaller group of once stable, if limited democ-racies (Colombia and Venezuela) (Hagopian and Mainwaring 2005). As thisvolume has highlighted, this crisis of representation is particularly striking inthe Andean region.1 More established political party systems have collapsed (withdominant political parties suffering a decline in support, credibility, and legiti-macy) in Colombia, Venezuela, and Peru. Weak party systems have remainedweak in Ecuador and Bolivia. Coups have occurred in Peru (Fujimori’s 1992 au-togolpe) and Ecuador (2000), and attempted coups have taken place in Venezuela(2002)—although in each case civilians took office shortly thereafter. A formermilitary dictator was elected president in Bolivia (1997), and those affiliated with

9

Indigenous Politics in the Andes: Changing Patterns of Recognition,

Reform, and Representation

❖

Deborah J. Yashar

Latin America’s ruling classes, unable to wish Indians away, were quite happy tobuild nations without Indians, and this they have been trying to do for almost twocenturies. To their chagrin, as the new millennium dawns, not only are indigenouspeoples still present—and their numbers are rising, but they are actually challeng-ing the very model of the nation-state that ruling groups have tried so conscien-tiously to build up.

—Rodolfo Stavenhagen, “Indigenous Peoples and the State in Latin America”

258 Yashar

past military regimes have successfully won electoral office at subnational levelsin Bolivia, Venezuela, and Argentina. Popular mobilizations, moreover, havetoppled presidents in Ecuador (1997) and Bolivia (2003). State institutions areweakly and unevenly institutionalized, often subverting the norms, rules, andpractices that allow for participation and representation in formal political circles.And the fate of more “traditional” social movements in the workplace, shanty-towns, and countryside remains uncertain. Surveys of the region have high-lighted, unsurprisingly, the thin legitimacy of Latin America’s contemporarydemocratic institutions.2 In short, the institutional foundations, sociological or-ganization, and political imagination required for democratic representation ap-pear weak indeed.

This bleak political picture is tempered, however, by unprecedented organiz-ing, claim making, and even representation by and for indigenous people. Whilethe rest of the region appears stymied by weak political parties, emasculated so-cial movements, and corrupt political institutions, indigenous people have forgednational and international movements. Over the past three decades, these move-ments have proposed a set of reforms that include legal recognition, representa-tion, autonomy, and bicultural education, among other things. Increasingly, thesemovements have given life to indigenous politicians, political parties, and con-sultants that have shaped political debates and sought to push through political re-forms. In this regard, indigenous people have become a politically organized forceand emerged as new claim makers in the political arena. Identity politics anddemocratic participation have thus intersected recently in unexpected ways inLatin America—giving indigenous activists a powerful voice in civil society andindigenous politicians a new (although not always powerful) voice in politicalcircles.

This chapter discusses the changing terms and scope of ethnic representationin the Andes. While making reference to the region as a whole, it focuses partic-ularly on the three countries with the largest indigenous populations in the An-dean region: Ecuador, Bolivia, and Peru. The chapter is divided into four mainsections. The first section provides a brief comparative historical overview—highlighting how Latin American states wrote indigenous people out of formalpolitics only to be confronted at the end of the twentieth century with the riseof significant indigenous movements demanding recognition, representation,and reform. The second section steps back to problematize the concept of in-digenous representation. Next, contemporary institutional reforms that haveformally opened up new channels for indigenous people to seek representationare evaluated. The final major section discusses the challenges faced by indige-nous movements as they turn to partisan politics as a means to promote and rep-resent an indigenous agenda. These four sections collectively highlight the phe-nomenal advances in indigenous claim making and the significant obstacles toindigenous representation in the Andes.

Indigenous Politics in the Andes 259

Writing Indians Out of Politics, Bringing Indigenous People Right Back In

It is no surprise to argue that Latin American states vastly underrepresented in-digenous peoples just as politicians consciously misrepresented them.3 What isperhaps less well understood are the ways in which twentieth-century modern-izing projects did the same. Indeed, twentieth-century politicians attempted towrite indigenous people out of politics through nation-state projects designedto create (or at least project) homogeneous nation-states (Stavenhagen 1992).Latin American politicians presumed (as did most scholars) that ethnic identitiesand cleavages were secondary and ephemeral. As such, politicians enacted re-forms that not only sought to disguise the presence of indigenous people but ac-tively sought to assimilate them. Scholars in turn presumed that ethnic politicswere ultimately inconsequential for the study of Latin American politics in gen-eral, and political representation in particular. As such, Latin America was por-trayed as a region wracked more by class than ethnic cleavages. In this section,I look below the proverbial political surface and briefly highlight how corpo-ratist projects of the mid-twentieth century essentially wrote indigenous peopleout of politics, all the while providing indigenous communities with the foun-dations to subsequently become a political force.

At mid–twentieth century, Latin American governments promoted corpo-ratist projects of interest intermediation. As an entire literature has highlighted,these projects fundamentally set out to restructure society into functional groupsthat would access the state along class-based lines. While in some cases theseclass-based federations had ties to political parties that ostensibly representedthem during democratic periods—Movimiento Nacionalista Revolucionario(MNR, or National Revolutionary Movement) in Bolivia, Alianza Popular Rev-olucionaria Americana (APRA, or American Revolutionary Popular Alliance)in Peru, Partido Institucional Revolucionario (PRI, or Institutional Revolution-ary Party) in Mexico, and the Peronists in Argentina—in others (Ecuador,Chile, Brazil, etc.) these institutions were created without any clear links to ex-isting political parties.4 Studies of these corporatist projects provided lasting in-sight into the ways in which corporatism provided both inducements and con-straints for Latin America’s labor and peasant movements. However, these samestudies largely neglected the impact of these reforms on indigenous communi-ties and ethnic cleavages. This oversight is surprising.

Indeed, as part of these corporatist national projects, corporatist citizenship re-gimes attempted to turn Indians into national peasants in several of the Andeancases—in the 1950s and 1960s in Bolivia, in the late 1960s and 1970s in Peru,and in the 1970s in Ecuador. In Bolivia, indigenous people in the Andes weremobilized into peasant federations in the 1950s, with the presumption that they

would vote for (and be represented by) the MNR, the party that oversaw thecorporatist project. During the military period of the 1960s, the Bolivian mili-tary subsequently sought to displace the MNR and to tie the peasantry to them,in what became known as the Pacto Militar-Campesino (Military-Peasant Pact).In Peru, the populist military government of Juan Velasco Alvarado (1968–75)also set out to mobilize indigenous people into peasant federations and other cor-poratist organizations, such as the Sistema Nacional de Apoyo a la Movilización(SINAMOS, or National Support System for Mobilization). Accordingly, Ve-lasco declared the social death of Indians—referring to indigenous peoples fromthat point on as Peruvian peasants, renaming the Day of the Indian as the Day ofthe Peasant, and encouraging indigenous people to refer to themselves as peas-ants. In Ecuador, the populist military government of General Guillermo Ro-dríguez Lara (1972–76) in particular also set out to create these corporatist ties,although his efforts to forge a powerful peasant federation were less successfulthan in Bolivia and Peru.

Indeed, throughout Latin America, indigenous people gained access to landreform, social services, and other kinds of state-organized reforms only insofar asthey joined peasant organizations and channeled their demands through peasantfederations. Hence, indigenous people had strong incentives to publicly forsaketheir ethnic identities and to assume a class-based identity in union organizationsand exchanges with political officials. In exchange for mobilizing into these peas-ant federations, indigenous communities in fact did gain access to land (some ofwhich was communal), subsidies, services, and other benefits. These were im-portant resources that provided a modicum of political and material autonomyfor communities that often maintained indigenous networks and forms of gover-nance at the local level—beyond the gaze of state officials and peasant federations.

Corporatism, therefore, did advance material demands for many (althoughhardly all) indigenous people. However, it would be hard to declare that thesecorporatist projects represented indigenous people in any meaningful sense of theword. Indeed, given the rare and short-lived efforts to organize indigenous peopleinto political parties, as in Bolivia in the late 1960s, it became commonplaceto assume that indigenous identity was not an important political identity. Theassumption among politicians and scholars alike was that politically mobilizedpeople did so along non-ethnic lines—in populist parties, class-based socialmovements, and the like.

Latin American politicians complemented these corporatist measures with ed-ucational programs designed to promote assimilation alongside Indian institutes,which had been designed to study indigenous cultures and promote assimilation(although in many cases they simply languished).5 Such assimilationist programswere put into place in Peru and Bolivia, as well as Mexico and Guatemala, to in-corporate people perceived as backwards into the ranks of a new, and presumablymore civilized, nation.6 States encouraged indigenous men and women to discard

260 Yashar

any public display of indigenous identity, encouraged the adoption of mestizoidentities, and, consequently, publicly encouraged miscegenation to “whiten”the population. According to state officials and intellectuals, mestizaje would al-low for social mobility as one’s ethnic status changed from indigenous (other) tomestizo (one of us); this process would presumably depoliticize ethnic cleavages.7

What politicians and political scientists alike did not explicitly acknowledgewas that these corporatist and assimilationist projects did not do away with in-digenous identities. Rather they simply provided dual incentives—by encourag-ing indigenous people to assume a class-based identity in national public forumswhile at the same time providing the communal land base (via land reforms) tosustain ethnic ties and governance at the local level. Hence, class-based forms oforganizing and representation masked, and in some cases nurtured, the survivalof ethnic enclaves. Indigenous communities subsequently started to mobilizealong ethnic cleavages beginning in the 1970s and continuing through the 1990sin Ecuador, Bolivia, Colombia, and elsewhere. Indigenous mobilization in theAndes was complemented by indigenous mobilization in the Amazon.8

Elsewhere I have analyzed why, where, and how indigenous movementsdeveloped in the late twentieth century—with quite powerful organizationsemerging in Ecuador and Bolivia, and quite weak ones emerging in Peru (Yashar1998, 2005). In general, indigenous movements throughout the region initiallydeveloped to defend local autonomy. Local autonomy was increasingly placedin question as a result of state reforms that included the slowing down of land re-forms, colonization, the cut in social services, and the opening up of land marketsthat had previously recognized the inalienability and indivisibility of communallandholdings. With community spaces in question, indigenous movements be-gan mobilizing to defend their autonomy and to gain a stronger political voice.However, indigenous communities were not equally capable of mobilizing andwere successful in doing so only where two additional factors were present:transcommunity networks and political associational space.9 Transcommunitynetworks (those left in place by unions, churches, and non-governmental orga-nizations) enabled indigenous leaders to mobilize across disparate communi-ties—those separated by geographic distance and language; the first generationof indigenous movements emerged only where they could capitalize on theseexisting networks. In turn, political associational allowed indigenous move-ments to organize relatively free from repression. Both transcommunity net-works and political associational space were present in various forms in Ecuadorand Bolivia by the 1970s and 1980s; they were nearly absent in Peru by the1980s, as a result of the civil war that destroyed transcommunity networks andforeclosed spaces for political organizing.10 As a result, strong regional and na-tional movements emerged in Ecuador and Bolivia, and comparatively weak(some would argue largely nonexistent) movements emerged in Peru. In this re-gard, Peru’s weak and localized indigenous movements stand out in the 1980s


and 1990s as a continental anomaly, given the widespread pattern of indigenousorganizing in most other countries in the Americas.

Indeed, in the last third of the twentieth century, Latin American indigenousmovements emerged to defend local autonomy and to challenge the idea that in-digenous identity and culture are anachronisms. For while many indigenous menand women outwardly assimilated into mestizo culture (leading to an officialdecline in the absolute numbers of self-identified indigenous peoples), self-identified indigenous communities have survived—albeit, as all communities do,they have changed over time. Although the data on indigenous populations areproblematic, the following data (gathered from 1979 to 1991)11 on the Andeancountries are widely used to approximate the size of this population: Bolivia,60 –70 percent; Peru, 38– 40 percent; Ecuador, 30 –38 percent; Colombia, lessthan 2 percent; and Venezuela, less than 2 percent.12 The political challenge hasbecome more than a question of demography and numerical survival. It has be-come a question of setting an agenda for indigenous people.

By the mid-1990s, significant indigenous movements had formed through-out the Americas, with the strongest organizations in Ecuador and Bolivia. InEcuador, the most important group is the Confederación de Nacionalidades In-dígenas del Ecuador (CONAIE, or Confederation of Indigenous Nationalitiesof Ecuador). In Bolivia, there are several important groups: the ConfederaciónSindical Unica de Trabajadores Campesinos de Bolivia (CSUTCB, or UnifiedPeasant Workers Trade Union Confederation of Bolivia); the Confederación In-dígena del Oriente, Chaco, y Amazonía de Bolivia (CIDOB, or IndigenousConfederation of the East, Chaco, and the Amazon); and more recently, the co-calero, or coca growers movement. Moreover, there are prominent organizationsin Colombia as well, and (as noted) the weakest organizational forms exist in Peru.

Indigenous movements throughout the region have demanded formal recog-nition, local autonomy, legal pluralism, additional land reforms, and biculturaleducation, among other reforms. These movements have largely voiced theirdemands through classic social movement politics. They have organized un-precedented marches from the Amazon to the Andes (particularly in Bolivia andEcuador), staged highway disruptions, occupied government buildings, and or-ganized street protests against various political and economic reforms.

From a comparative perspective, indigenous mobilizational capacity stands inmarked contrast to the weakness of other social movements in Latin America,which have declined in strength since the heady days of anti-authoritarianprotests. The groups’ ability to mobilize explicitly on behalf of indigenous peopleand to take the lead in more general societal protests against various neoliberalreforms has undoubtedly led them to assume the mantle as the central and mostpowerful social movement actors in Ecuador and Bolivia. Like social movementactors before them, they have largely mobilized outside of the halls of the state—seeking to represent those that have been shunned by formal politics—and haveused the power of their numbers to voice their claims. As such, they have made

262 Yashar

the issue of indigenous people, organizations, and claims a part of political de-bate. And they have highlighted the inadequacy and injustice of efforts to eraseindigenous identity as a basis for political mobilization and representation.

While the movements and their demands as a whole are quite varied, thepoint to emphasize here is that indigenous people are organizing and articulat-ing ethnic-based agendas in unprecedented ways. As such, indigenous peoplehave increasingly emerged as a political force to be reckoned with. It is this con-temporary mobilization of indigenous people that is so striking. Not only haveindigenous people demanded recognition as such, but they have also demandedreform (including autonomy demands) and representation. This is happeningprecisely at the moment that the rest of the region appears to be undergoing acrisis in representation. So, the question becomes, to what extent has indigenousrepresentation advanced in recent years? The following two sections take up thatquestion.

Problematizing Indigenous Representation

This chapter has thus far highlighted the historic obstacles to indigenous repre-sentation in the Andes. Before turning to the question of if/how indigenousrepresentation has advanced in the contemporary period, we must first addressthe conceptual and analytical ambiguities that surround the question of in-digenous representation in the Andes. The editors of this volume have discusseddemocratic “representation” as the process by which an agent expresses the statedinterests of a principal within and before democratic institutions. With this as aconceptual base, one can conclude that “indigenous representation” refers to apolitical context in which indigenous people can elect officials and/or appointdelegates to act on their behalf.

While on the face of it, such a definition appears to be self-evident, it is in factinherently open to interpretation. In practice, this definition is better at indicat-ing negative cases. That is to say, it is easier to delineate when indigenous repre-sentation is not taking place than when it is occurring—or, rather, when it isoccurring in a meaningful sense. This conceptual ambiguity results because thedefinitions of the noun (representation) and the adjective (indigenous) them-selves beg important and interrelated questions about the principal (who isindigenous), interests (who defines them), agents (are they in fact representingthose in whose name they speak), and locales (where does representation takeplace). Let me elaborate on the conceptual problematic.13

Principals and Interests

When we speak about indigenous people, who in fact is the principal? Thisseemingly simple question elicits no simple answer. First, there are no easily ob-servable and agreed-upon measures for indicating who is in fact indigenous.


Ethnicity is a conceptually slippery category—all the more so now that schol-ars have come to recognize its constructed and changing boundaries. But ofequal importance, indigenous people in Latin America have embraced their in-digenous identities in some places while denying them in others. As such, wemust keep in mind that identity is an elective and fluid concept that is some-times, but not always, primary. While one cannot credibly claim to be indige-nous if one does not have indigenous ancestry, one can claim not to be indige-nous; as noted earlier, this latter position was, in fact, advocated by state policiesthroughout the region.

Given the fluid nature of ethnic identity in the region, “indigenous peo-ple” should be defined as those people who self-identify as such; such self-identification asserts shared ascriptive characteristics and a common history asthe original inhabitants of what we now call the Americas. But even this defini-tion confronts problems of measurement; indeed, basic demographic data on in-digenous populations remain highly disputed.

Even if we could definitively agree on who is indigenous and how many in-digenous people exist, the question of what is meant by the term “indigenous in-terests” remains unanswered. And how one identifies and measures indigenousinterests presents yet another methodological (and politically charged) chal-lenge.14 One might do so by aggregating individual indigenous preferences—thestated interests of indigenous elders, the findings of indigenous intellectuals, theagendas of indigenous movements, and/or the platforms of indigenous politi-cians. But definitions of indigenous interests are bound to vary, depending on themethodology used. For, as scholars of black politics and representation in theUnited States have debated for many years, there is no simple answer to this ques-tion (see, e.g., Swain 1995 and Tate 2003). Similarly, as Warren (1998) has so elo-quently shown for Guatemala, debates among indigenous spokespeople abouthow to define and pursue indigenous interests can be quite heated.

I do not intend to settle this debate here. Rather, I highlight these numerousquestions with two goals in mind. First, on a cautionary note, I do not wantto suggest that there is some mechanical and universal way of identifying indige-nous people, defining indigenous interests, and evaluating indigenous represen-tation. To the contrary, the topic is complex and would require many book-length treatments. Second, and following from this first point, it is incumbent onthe author, therefore, to indicate how she is using the term “indigenous repre-sentation.” Here, I use two (admittedly imprecise, narrow, conflict-laden, andnon-exhaustive) measures to discuss indigenous principals and interests. On theone hand, I have used a micro-conception of indigenous principals and interests:to analyze when individual indigenous people choose to identify as indige-nous and to select and elect representatives accordingly (without speaking towhether these representatives can and do defend indigenous interests). I implic-itly use this micro-measure in the previous section to show how indigenous

264 Yashar

representation was historically denied. In the next two sections, I use it to high-light how contemporary institutional reforms increase opportunities for indige-nous individuals to self-consciously take part as Indians in their countries’ dem-ocratic process, and to highlight the rise in the number of indigenous people whohave been elected to political office in recent years. On the other hand, I have alsoused a macro-conception of indigenous representation to analyze indigenous or-ganizations and social movements as they voice collective interests within the for-mal political process. I have largely used this macro-conception when analyzingthe challenges for indigenous representation, but it also has some relevance forthe subsequent discussion of institutional reforms and how they advance indige-nous representation. This macro-conception of indigenous principals and inter-ests is in some senses “thicker” than the micro-concept, insofar as the relevant in-digenous actors publicly claim their indigenous identity and self-consciously seekto promote some version of indigenous interests. While it is possible in thissecond scenario to identify a set of “indigenous interests” sanctioned by a par-ticular movement, it is important to also bear in mind that movements them-selves are not universally sanctioned, internally consistent, or made up of unifiedgroups.

Agents and Interests

In discussing indigenous representation, one must also problematize the agent—all the more so since it is often argued that elected and appointed officials are notthe real or true representatives of indigenous people. There is no agreement overwhether the agents/representatives of indigenous people must belong to thesame social group (however defined) as the principals; whether he or she will actin the best interests of the principals, even when the principals disagree; and/orwhether they are even accountable to the principals.15

Hanna Pitkin (1967), in particular, has discussed these issues in her classic workon representation, as has José Antonio Lucero (2002), who analyzes the com-plexity of indigenous movement representation in the Andes and puts forth apragmatic constructivist understanding of representation.16 Drawing on (andgreatly simplifying) the work of these scholars, I highlight two aspects of repre-sentation here: descriptive, or “mirror,” representation (which asks: Are the rep-resentatives from the same group as I am?); and functional, or “guardian,” repre-sentation (which asks: Are the representatives pursuing my interests?). While wecommonly discuss identity-based representation in terms of the former, we tendto equate ideological and partisan-based representation with the latter. Bothmeasures, of course, provide different sorts of insight.

In my discussion, I make use of both aspects of representation: I discuss the factthat more indigenous people are gaining office (although this measure does notnecessarily indicate whether these new political representatives are consciously


or actively pursuing the interests of indigenous people); and I discuss the fact thatmore elected officials and authorities (regardless of whether they are indigenousthemselves) are self-consciously pursuing agendas that have been articulated byindigenous social movements.17

Locale

Finally, in talking about indigenous representation, the question of the site ofrepresentation is particularly relevant. As discussed previously, indigenous peoplehave historically been denied formal access to the seats of state power. However,this does not mean that representation has not occurred in other ways. Indeed,as highlighted below, indigenous people have often operated simultaneously intwo spheres—in political spheres legally sanctioned by the state and in politicalspheres recognized by customary law. This chapter focuses on the advances andsetbacks of the former (in line with the focus of this volume), while recognizingthe centrality of the latter to indigenous claim making.

In sum, I cannot hope to adequately answer all of the complex questions as-sociated with indigenous representation in this chapter. Given the complexity ofdiscussing indigenous representation, I therefore delimit the scope of the ques-tions discussed in this chapter. The next section highlights the contemporarypolitical reforms that have created new opportunities for political participationby and for indigenous people. I largely discuss indigenous representation in itsmicro-analytic guise by looking at institutional changes that have promoted theopportunity for indigenous individuals to speak out and to be elected. The finalsection largely discusses the dilemma of indigenous representation in its macro-analytic guise, as indigenous movements seek to take part in formal electoralpolitics.

New Institutional Opportunities for Advancing Indigenous Representation

The third wave of democracy has forged new institutional opportunities for in-digenous people to participate in political debates, demand political inclusionand autonomy, and seek a greater voice and representation.18 This section out-lines some key institutional and constitutional reforms that have manifestlyopened up opportunities for indigenous people to seek political representa-tion.19 Indigenous interests have found greater voice—whether these interestsare defined in their “micro-analytic” or “macro-analytic” version. So, too, in-digenous representation has found more institutionalized spaces for securingpolitical office—whether these representatives are evaluated as “mirrors” or“guardians” of indigenous interests.

266 Yashar

An underlying theme for this section, however, is that such institutional opportunities are fundamentally constrained by weak states and by political institutions that work against sustained representation of indigenous individu-als, communities, and movements. Indeed, the very conditions that work againstrepresentation for all people in the Andes (the subject of this volume) also workagainst institutionalizing meaningful and sustained representation for those indigenous people and movements that seek to take part in the formal politi-cal arena.

Suffrage

The third wave of democracy obviously re-extended the rights of citizens tovote for their elected officials in Ecuador, Bolivia, and Peru. On the face of it,therefore, the current period of democratization has opened up channels for in-digenous people to take part in elections—a micro-foundation for delegatingauthority to representatives in national politics.

But the current period is more than a simple period of redemocratization.With the third wave, universal suffrage was extended in Peru (1979) and Ecuador(1980), a policy that overturned earlier literacy restrictions. Given the high ratesof illiteracy among indigenous populations—particularly those living in thecountryside—these measures had essentially prevented indigenous people fromvoting in the past. In this regard, the most recent round of democratizationliterally enfranchised indigenous people, thereby advancing opportunities forthem to voice their individual preferences for political representation. As in-digenous people started to organize as indigenous movements, political partiesincreasingly turned to indigenous communities to secure their votes. Indigenouspeople have found, therefore, that they have a greater political voice in electionsthan ever before.

Obstacles, however, remained. In Peru, the escalation of civil war in the 1980s,with Sendero Luminoso (Shining Path) and the Movimiento RevolucionarioTupak Amaru (Tupak Amaru Revolutionary Movement), placed serious limitson the ability of people to freely take part in electoral and partisan politics. Thewidespread violence in provinces with large populations of indigenous people(particularly Ayacucho) essentially foreclosed the ability of people to come to-gether independently, to publicly articulate political platforms, and to vote freelyin elections. Registration, moreover, was less than universal. Registration (and ac-cess to voting booths) has posed problems, particularly in the Amazonian regionsof the Andean countries discussed in this volume. In point of fact, registrationwas (and is) spotty and unreliable, most notably in the Amazonian basin—wherethe state is weak and some communities remain itinerant. Finally, voting for es-tablished political parties does not mean that once in office these elected officials


will “represent” those who voted for them. Indeed, in many cases there is con-siderable desencanto (disillusionment), as elected officials ignore communities be-tween elections or try to placate them with minor clientelistic rewards. Thisobservation is, of course, not peculiar to indigenous peoples.20 Rather, it is partof a larger dynamic whereby Latin Americans who cast their votes have dispro-portionately little faith that those elected (at the very least, those elected to na-tional office) will pursue their interests (Lagos 2003a, 2003b). How else to ex-plain efforts to force presidents out of office before they have completed theirterms in Bolivia (2003 and 2005) and Ecuador (1997, 2000, and 2005); theseanti-executive mobilizations are a testament to the weak faith in the electoral sys-tem as a vehicle for advancing interests and securing worthy representation.

Recognizing Multiethnic Citizenries

A second significant reform included the explicit recognition of multiethnic cit-izenries. Constitutions now recognize the multiethnic and multicultural makeupof Colombia (1991), Mexico (1992), Peru (1993), Bolivia (1994), Ecuador(1998), and Venezuela (1999), among others in the region.21 Indigenous peopleparticipated in the rewriting of these constitutions, albeit in single-digit num-bers, including seven indigenous representatives in Ecuador, three indigenousdelegates in Colombia, and three appointed delegates in Venezuela (Van Cott,forthcoming). In several countries, constitutions and/or corresponding legisla-tion recognize indigenous legal systems and authorities, although the actualterms of recognition remain ill-defined and largely unimplemented (Staven-hagen 2002, 32–33). These reforms are an important symbolic victory for in-digenous peoples, who have worked to undermine myths of national unity.22

As Yrigoyen Fajardo shows, constitutional recognition is more than a sym-bolic victory since it represents a fundamental shift away from the monoculturalpremise of Latin American judicial institutionalism: “The norms of such a sys-tem (indigenous customary law) were considered admissible only in the absenceof law, but never as an alternative to it, in which case it would be defined ascrime” (Yrigoyen Fajardo 2000, 197). Hence, the constitutional recognition ofethnic heterogeneity in some Latin American states has broadened the publicimagination and established a legal precedent for discussing ways to accommo-date and represent a diverse citizenry, as discussed next.

Indigenous Seats

In a few cases, Andean countries have apportioned seats for indigenous people.The most recent Colombian and Venezuelan constitutions have guaranteed twoand three indigenous seats in the national legislatures, respectively. Other re-forms have given the nod toward securing some indigenous representation inmunicipal and regional assemblies and councils. Venezuela now guarantees

268 Yashar

“a seat on state-level assemblies and municipal councils in districts where they [indigenous] have a significant presence”; and Peru’s 2002 Ley de Elecciones Regionales (Regional Elections Law) included a 15 percent indigenous quota onregional party lists for elections in the overwhelmingly indigenous Amazon, al-though the law did not indicate where those names would be placed on the lists(Van Cott, forthcoming). These are historically unprecedented changes in thesecountries, for they recognize indigenous people as political actors with guaran-teed rights to compete in elections and occupy political office. These reformsadvance the mirror image of indigenous representation.

These reforms, while revolutionary in concept, will not revolutionize the pol-icy output or the scope of indigenous representation. The set-aside seats are fewin number, making it hard to imagine that indigenous representatives can domore than join existing coalitions and voice their opposition to those policiesdeemed disadvantageous to indigenous peoples. Moreover, Van Cott (forth-coming) convincingly cautions against presuming that these seats will necessar-ily translate into the ability of indigenous people to forge their own organiza-tions and choose their own representatives. Indeed, these mechanisms couldprovide yet another venue for more established political parties to make theirmark at the expense of more local efforts to cultivate, vet, and propose indige-nous candidates. If significant and enduring indigenous representation in thestate is to occur, indigenous people will have to mobilize for more than a hand-ful of seats—which, on their own, can be seen as only concessionary tokens.23

Decentralization

As Kathleen O’Neill argues in her chapter in this volume, decentralization hasalso reconfigured the spaces and terms of participation and representation in theregion. Among the Andean countries, decentralization in Colombia, Bolivia,Ecuador, and most recently Peru has increased opportunities for electing officialsthat are responsible to local constituencies, thereby changing the location orspaces in which representation can take place. I will not repeat O’Neill’s argu-ment here, other than to underscore the following key points.

Clearly, decentralization has increased the opportunities for indigenous adultsto elect representatives that will be responsible to local constituencies. In manyplaces, the new reforms have increased the numbers of indigenous politicianselected to office. The first elections in Bolivia, in 1997, led to a marked rise inindigenous councilors and subsequently to the election of a number of indige-nous mayors (see Albó 2002, 82ff.). The same can be said for Colombia andEcuador (see Van Cott 2002, 51, 65). With the 2004 reform in Bolivia that nowallows candidates from social movements and indigenous movements to run forelected office (even when they do not have ties to a national political party), wewill perhaps see that number rise even more.


Also, it is worth noting that in all cases except Bolivia there is the legal possi-bility of creating some form of indigenous municipality (see Van Cott, forth-coming). However, in practice, these indigenous municipalities have largelyremained legal prospects rather than realized forms. In Bolivia, the Law of Pop-ular Participation does not allow for indigenous municipalities but does allow forindigenous communities to negotiate with existing municipalities to forge in-digenous districts; these districts, however, have no prior claim to resources orauthority—all of which must be negotiated with the corresponding municipal-ities. As such, there is a recognition of the right to choose indigenous authori-ties, but their purview is open to debate (Yashar 1998, 2005).

Indeed, as O’Neill, Gray Molina (2003), and others have shown, it is hard togeneralize about the impact of decentralization on democratic representation,given widespread variation across cases, within countries, and over time. In-deed, several cautionary flags have been raised. While decentralization has in-creased the opportunity to elect local officials, this has not necessarily translatedinto an increase in indigenous voice, participation, and representation. Indeed,localizing politics has meant, in some places, giving greater opportunities for lo-cal, non-indigenous elites to assert control over local regions. Moreover, evenin cases like Bolivia, where decentralization measures included oversight com-mittees (comités de vigilancia) that institutionalized a role for territorially basedcommunities, in practice, the committees have limited training and resources toeffectively perform that job. As O’Neill and Gray Molina show, more researchis needed to chart out and explain this variation in outcomes.

Finally, not all local governments have in fact received the kinds of financialresources that would enable indigenous communities to elect officials. O’Neillshows that there is a significant variation in resource bases for different sub-national governments. Citing 1995 IDB data, she notes that subnational spend-ing as a percentage of total spending in 1995 ranged from a comparative high inColombia (39 percent) and Bolivia (26.7 percent) to a middle range for Vene-zuela (19.6 percent), to a low in Peru (10.5 percent) and Ecuador (7.5 percent).In short, decentralization has not everywhere meant greater spaces for indige-nous people to assume office and/or elect those who will have the resources toeffectively act on their behalf.

Autonomy

Finally, several states have recognized some form of indigenous autonomy—thereby creating new spaces for indigenous people to seek ethnic representationand jurisdiction. In these autonomous spheres, indigenous people can chooseauthorities and governing systems outside of the jurisdictional purview of cen-tral states, local municipalities, and national law.

270 Yashar

New constitutions in Colombia (1991), Peru (1993), Bolivia (1994), andEcuador (1998) have gone a long way toward recognizing (although not neces-sarily implementing) indigenous laws and norms, indigenous authorities and au-thority systems, and jurisdictional functions (Yrigoyen Fajardo 2000; Van Cott2000, 2002). Moreover, national legislation has further institutionalized certainkinds of autonomy in several of these countries, particularly in the Amazon. InEcuador, for example, following the thirteen-day, two-thousand-person marchfrom Puyo to Quito organized by the Organización de Pueblos Indígenas delPastaza (OPIP, or Organization of Indigenous Peoples of Pastaza) in 1992, thegovernment eventually conceded nineteen different territorial blocs, encompass-ing 138 legally recognized communities and 1,115,475 hectares.24

In Bolivia, the state ultimately conceded four autonomous indigenous territo-ries following the 1990 March for Territory and Dignity, organized by CPIB(Central de Pueblos Indígenas del Beni, or Central for Indigenous Peoples of theBeni).25 In 1996 the Bolivian government passed a new agrarian reform that pro-vided indigenous communities with the legal basis for appealing for territorialrecognition—including the right to vast expanses of land and the political au-tonomy of indigenous authorities. By August 1997 the state had recognized sevendistinct territories totaling 2.6 million hectares, and it was processing thirty-fourmore demands totaling about 20 million hectares.26 And in 1998, the Boliviangovernment created a program to title 10 million hectares of indigenous lands(Plant 2002, 217). However, the Banzer administration (1997–2002) delayed theprocess of titling these lands (Van Cott 2002, 56), and third-party colonist andforestry concessions have also slowed down the process (Plant 2002, 209).

Demands for territorial autonomy in Bolivia have been complemented by ef-forts to establish indigenous districts (not municipalities); with the 1994 BolivianLaw of Popular Participation (largely a municipalization and decentralizationlaw), indigenous communities gained the right to request indigenous districts—albeit with mixed results. The terms of these indigenous districts in Bolivia areunderspecified; the law creates the possibility (but neither the obligation, terms,nor mechanisms) for establishing the districts. Consequently, indigenous dis-tricts are hard to negotiate and are not autonomous administrative units; theymight, but do not necessarily, institutionalize the right to customary law, bi-lingual education, communal property, state resources, etc. The future and fateof these municipal districts depends on the mayor, who has the power to recog-nize them and to determine resource allocation, but has no legal obligation to doone or the other.27 According to Luz María Calvo, the former head of theSubsecretaría de Asuntos Etnicos (SAE, or Subsecretariat of Ethnic Affairs), thesedistricts have not emerged unblemished. To the contrary, most have confrontedsignificant problems due to poorly delineated legal rights and limited skills withinthe community.28 Moreover, it should be noted that the law does not delineate


the precise terms of the relationship between the municipality and a hypotheti-cal indigenous district. Indeed, the financial and political relationship is largelyup for negotiation between those proposing the indigenous district and the in-dividual municipality. For example, it is up to the goodwill of the municipalgovernment to decide if it will transfer any financial and social resources (and ifso, how much) to the indigenous district; indeed, the municipal government hasno legal responsibility to distribute part of the co-participation funds to paylocal officials’ salaries, or to cover administrative costs (see Balslev 1997, 35– 41,53–58, 86). According to documents from the Secretaría de Participación Pop-ular (Dirección Nacional de Organización Territorial Administrativa), therewere 127 such districts by April 1997;29 by 2000, 138 had been formed (Velasco2001, 42).

In the 1991 Colombian Constituent Assembly, moreover, indigenous peoplesalso negotiated reforms that granted territorial autonomy (Clavero 1999, 187–89; Van Cott 2000). The 1991 Colombian Constitution referred to indigenouslands as territorial entities in Article 286; according to this article, existing po-litical authority structures assume governing capacity, including criminal andcivil jurisdiction, in these territories; moreover, the territories are responsiblefor determining their own development strategy and for administering publicresources as if they were municipalities.

We also find states recognizing some version of autonomous regimes (re-serves) in the Amazonian Basin in Brazil (Brysk 2000, 201). While concessionsof territorial autonomy in each country have confronted serious obstacles inimplementation (Plant 2002, 209), they constitute a significant symbolic and le-gal precedent for indigenous movements as a whole.

Beyond the Amazon, as well, the state has recognized indigenous communi-ties as politically autonomous units, which can choose their own representativesaccording to customary law. Bolivia’s 1997 agrarian reform law, for example,opened up the possibility to recognize the ayllus (communal kinship organiza-tions) that dot the Andean countryside. Ecuador’s 1998 Constitution also created“Circunscripciones Territoriales Indígenas y Negras” (Indigenous and BlackTerritorial Circumscriptions), although the legislature had not acted on this pos-sibility at the time of this writing. We have seen similar efforts beyond the An-dean region, including in Mexico, Guatemala, and Nicaragua.

Latin American states, therefore, have initiated reforms that recognize somedegree of political autonomy for indigenous people. Indigenous movementshave demanded this autonomy as a means of securing indigenous political juris-diction over that land, including the right of indigenous legal systems and au-thorities to process and adjudicate claims. In this regard, autonomy reforms haveprovided a legally recognized space in which indigenous people can choose al-ternative locales, forms, and modes of representation. State autonomy reforms,however, must be assessed cautiously. Stavenhagen (2002, 34 –35) concludes

272 Yashar

that many of the autonomous agreements remain ill defined and weakly imple-mented in Colombia and Nicaragua, among others.

Challenges of Implementation, Questions for Representation

So far in this section, I have highlighted significant and unprecedented consti-tutional and legislative reforms. Without a doubt, these reforms have createdgreater opportunities for indigenous individuals to vote for representatives, andhave designated new formal political entities (e.g., the municipality and the au-tonomous region) where they can do so. Moreover, these reforms have under-stood the need to officially recognize indigenous people as citizens and to se-cure some spaces for participation—both in the state and within indigenouscommunities.

These changes are without parallel, historically speaking. However, as I haveargued throughout this section, one must temper these observations with ahealthy dose of skepticism. There are serious obstacles to representation over-all—obstacles that indigenous people also confront. First, party systems remainweak at best, and absent at worst. In this context, the vote for indigenous rep-resentatives (however defined) is a tall order; for even where indigenous peopleoverwhelmingly choose who they want to represent them, these new represen-tatives face political battles, in which partisan lines, coalitions, and policy mak-ing remain nebulous, ever-changing, and often corrupt. While all is not bleak,the hurdles are high.

Second, and related, indigenous representatives have often inserted new itemsonto the policy agenda. However, their ability to push these initiatives throughis far from certain. As such, the national distrust of politicians is likely to rever-berate back to them, a point I will come back to in the final section. Finally,where states are weak (as they are to varying degrees throughout most of theAndes and Amazon), policy remains poorly implemented, political and civilrights are poorly upheld, and even the institutional reforms noted here areplaced in jeopardy. Indeed, progress on this institutional front has been slow anduneven, as noted by Stavenhagen.

While these legal advances are surely important in themselves, the open question is howthe new legislation will be implemented and how Indian communities will benefit. Theanswer is not at all clear. Complaints are increasingly heard that the new laws are not be-ing implemented as they should be, or that secondary legislation has not been adoptedafter general principles were laid down in the new constitutions. (2002, 33–34)

In short, while there are de jure opportunities, de facto obstacles remain, notleast of all because political parties remain weak in all of these countries and statesremain incapable of implementing many of the reforms upon which politiciansand citizens can agree. In this context, it is difficult to imagine how indigenouspeople can get their interests meaningfully and consistently represented in the


state and in government. For these reasons, indigenous people have largely mo-bilized outside of the state through social movements—although, as the nextsection explains, they have increasingly calculated that they must also enter intothe partisan political arena. This final section takes up the question of the chal-lenge of indigenous social movements as they seek to articulate indigenous agen-das and to push them, not only in the streets but also in the halls of the state.

The Democratic/Electoral Challenge: The Siren’s Call? From Street Politics to State Politics

To date, indigenous movements have primarily demanded political representa-tion and political change through social-movement politics—in the streets, inprotests, in documents, and in international fora.30 They have achieved somenotable political successes, including the negotiation of territorial autonomy, bi-cultural education, a chance to help run state offices, and a voice in public de-bates. In other words, key demands have found their way into policy (althoughrarely solely because of indigenous mobilization), even as most indigenousmovements initially eschewed party politics, electoral campaigns, and the like.31

With the advent of increasing political clout, increasing overtures from theexisting political parties (concerned to tap into a mobilized indigenous constit-uency), and an increasing desencanto with existing political parties (which fail topromote the agendas that indigenous communities and movements demand),many indigenous movements began to question their principled opposition toparticipating directly in electoral politics. By the end of the 1990s, many in-digenous leaders and movements began to run for office, form political parties,and engage in partisan alliances.

Indigenous movements in Bolivia and Ecuador increasingly turned to politicalparties. As Mainwaring and Scully (1995) have observed, these two countries(along with Peru and Brazil) have developed comparatively inchoate party sys-tems.32 In the 1990s, indigenous movements in Ecuador and Bolivia still con-fronted heavily patronage-driven political systems that appeared to offer little po-litical space for representation (however defined). Weak political institutionscontinued to foreclose the ability of new and less powerful groups to gain a polit-ical toehold in political debates; existing political parties often appeared to disre-gard their constituencies and to engage in personalistic and clientelistic behavior.

Despite the odds, indigenous activists and movements in Bolivia and Ecuadorhave successfully placed indigenous leaders into office. Indeed, by the end of the1990s, we find more indigenous people in elected and appointed office thanever before—although, as we shall see, the degree to which this notable changehas allowed them to represent indigenous people is an open question.

Bolivia’s Andean indigenous movements became a recognized electoral forcein the 1990s although they started flirting with party politics in the 1970s. In this

274 Yashar

earlier period, the Katarista and Indianista movements were largely unsuccess-ful at fielding candidates against a backdrop of internal and cross-movementdivision. While a few indigenous leaders were elected, they remained isolated—both from the broader indigenous movement and from formal party politics ingeneral. Collectively, these parties rarely achieved even 3 percent of the total na-tional vote (Van Cott 2005, chap. 3). Following Katarista Víctor Hugo Cárde-nas’s 1993 election to the vice presidency (an election that further divided theexisting Katarista parties), other indigenous activists started to organize moreforcefully—particularly in the aftermath of the reforms that President GonzaloSánchez de Lozada and Vice President Cárdenas oversaw in the areas of decen-tralization, municipalization, and agrarian reform. Cárdenas sought to promotegreater recognition, pluralism, dialogue, local autonomy, and bicultural educa-tion (so that class- and ethnic-based concerns would be addressed). And with thedecentralization and popular participation laws that municipalized the country,more indigenous people won electoral seats at the local level. In this regard, the1990s witnessed a radical increase in the mirror-version of indigenous represen-tation—with more indigenous people gaining elected office. Ironically, theKatarista movement that had spearheaded the indigenous movement and firstfielded indigenous candidates, including Cárdenas, did not remain a competitivepolitical party. The Amazonian movement, CIDOB, also fared miserably (Yashar2005, chap. 5). A new generation of Andean indigenous leaders, led by EvoMorales and Felipe Quispe, later took the lead—organizing new political move-ments behind them.

Both Evo Morales and Felipe Quispe have become prominent political actorswithin Bolivia, and both have mobilized within the CSUTCB and on behalf ofindigenous people. However, they mobilized different constituencies and artic-ulated different visions. Evo Morales came to represent the increasingly power-ful cocalero movement, largely composed of Quechua-speaking indigenouspeople. As a CSUTCB leader, he appropriated the nearly defunct Movimientoal Socialismo (MAS, or Movement toward Socialism) and refashioned it as a pro-cocalero, anti-neoliberal, and anti-imperialist party—one that has challengedneoliberal policies and U.S.-affiliated eradication programs that limit the rightsof indigenous people to grow coca (portrayed as central to an indigenous cos-movision). Morales has had a meteoric political career. Not only did he and sev-eral of his colleagues win national seats in the legislature in 1997, but he camein second in the 2002 presidential election and his party became a dominant ac-tor in the current legislature—claiming to speak both on behalf of indigenouspeople and the anti-neoliberal cause. MAS won 8 of 27 senate seats and 27 of130 seats in the lower chamber, where it assumed the second vice-presidencyand leadership of key committees (Van Cott 2005, chap. 3). Morales won the2005 presidential election in an unprecedented landslide; twelve MAS senatorsand seventy-two MAS deputies were also elected.


Morales’s vision of indigenous representation does not stand alone. Severalindigenous people have tried to run for office, although here I focus on thosewho come out of the indigenous movement itself.33 Notably, Felipe Quispe emerged as Morales’s rival, both within the CSUTCB and in electoral poli-tics.34 Quispe, a former politician, guerrilla leader, and secretary general of the CSUTCB, has articulated what appears to be a more millennial form of Aymaran nationalism, at times using what is perceived to be a militant discourse.He also ran for president in 2002 on the ticket of the Movimiento IndígenaPachakuti (MIP, or Pachakuti Indigenous Movement). In that election, he re-ceived 6 percent of the vote.35 He was seen as a formidable force and Morales’smost important rival—although Morales has thus far bested him in movementand electoral politics. Seen as a whole, the Bolivian case is remarkable, both be-cause numerous indigenous men and women have gained significant access tolocal office, and because prominent indigenous leaders have had an importantimpact on national-level electoral races. Indigenous leaders have come to oc-cupy the presidency and vice presidency, and they have figuratively and meta-phorically given the other candidates a run for their money. As such, one findseven traditional parties seeking to include indigenous running mates on theirtickets. In this context, the face of Bolivian politics has become more diverse.

However, while there has been a rise in ethnic-oriented politicians in thestate (both at the local and national level), it remains less certain whether ethnicpolitical parties as a whole are developing new ways of doing politics or whetherthey are largely replicating the clientelist politics of the broader party system. In-deed, the rise of prominent indigenous leaders has coincided with ongoing po-litical fragmentation and internal divisions among indigenous politicians, whichmakes it difficult to unambiguously evaluate indigenous representation in Bo-livia. It is true that the number of indigenous representatives in the national leg-islature has increased significantly (although not all of them necessarily see theirprimary role as representing indigenous people). Yet, given their competing vi-sions and their party’s organizational structure, their capacity to represent andpromote a common or broader “indigenous agenda” is less clear. Accordingly,the strength of political parties for the collective pursuit of indigenous interests(however defined) remains uncertain.

Ecuador’s indigenous movements did not have the same historic pattern of en-gaging in party politics as Bolivia’s. While individual leaders had some personalexperience, CONAIE, Ecuador’s largest indigenous organization, originally re-jected the electoral process, protesting during the 1990 election and encourag-ing followers to cast a null vote in 1992 (Andolina 1999, 210). In the 1996 elec-tions, following a complicated set of internal debates, CONAIE changed course.It decided to become part of a national coalition, the Movimiento de UnidadPlurinacional Pachakutik–Nuevo País (MUPP–NP, or Pachakutik Pluri-national Unity Movement–New Country). The coalition was comprised of

276 Yashar

Pachakutik (the recently formed indigenous party associated with CONAIE),the Coordinadora de Movimientos Sociales (Coordinator of Social Movements),and the Movimiento de Ciudadanía por un Nuevo País (Citizenship Movementfor a New Country).36 The coalition won a total of eight seats—including thosefor provincial and national deputies (Mijeski and Beck 1998, 4). Among thoseelected were Luis Macas, former president of CONAIE, and Miguel Lluco,former president of ECUARUNARI (Ecuador Runacunapac Riccharimui, orAwakening of the Ecuadorean Indian). The electoral success was striking, asMUPP–NP elected over seventy candidates at the local and national levels andwon seven of every ten races it entered.37 Former CONAIE leaders were alsoconsidered and chosen for national-level political appointments, during both theshort-lived Bucarám administration and the Alarcón administration. While thisentry into formal politics was not free of conflicts, it represented a noteworthyachievement.38 Moreover, as an elected member of the legislature, Miguel Llucooversaw the signing of International Labor Organization Convention 169, whichadvocates many of the indigenous collective demands proposed but not passed inthe Constituent Assembly. And former CONAIE leader Nina Pacari was a sec-ond vice president in the national legislature.

CONAIE also participated actively in the Constituent Assembly that was heldfrom December 1997 to May 1998. The party won 10 percent of the assemblyseats and was the third largest political force in the assembly (Andolina 1999,231, 313). While assembly members did not achieve all their goals, their partialsuccesses were striking. The resulting constitution included an article on thecollective rights for indigenous people, although the document did not outlineguarantees for these rights, nor did it recognize the country as a plurinationalstate (Andolina 1998, 1999; Mijeski and Beck 1998).

Despite these initial inroads, Pachakutik’s electoral performance has beenuneven since 1998. In 1998 it lost standing in the legislature, going from 10 to5 percent of the congressional seats (Mijeski and Beck 1998, 12). Later, whenCONAIE played a major role in the mobilization against then-president Mahuad,CONAIE and Pachakutik had to address internal divisions and to contend withcharges that CONAIE was not committed to representative democracy. The2002 presidential elections did not resolve these internal divisions. Pachakutikeventually allied with Colonel Lucio Gutiérrez, who then successfully ran for thepresidency (under the banner of the Sociedad Patriótica, or Patriotic Society,a newly formed political party). Gutiérrez had collaborated with CONAIE inthe 2000 coup against Mahuad. He assumed office in 2002 with support fromEcuador’s largest indigenous organization, as well as the presumed allegiance ofthe 14 indigenous candidates who had been elected to the Congress (100 seats intotal) and the various indigenous leaders elected to subnational office. Indeed,Gutiérrez appointed Luis Macas and Nina Pacari (two of CONAIE’s most visibleindigenous leaders) to two important ministerial positions. However, CONAIE’s


support soon turned to opposition after Gutiérrez implemented austerity mea-sures. In the ensuing months, divisions within CONAIE and with Pachakutikbecame even more pronounced. In April 2005, the Congress ousted Gutiérrezfrom office.

In other words, as indigenous people have assumed office in Bolivia andEcuador, less organizational and programmatic unity (rather than more) hasbeen the norm. For while all indigenous leaders would probably agree on a coreset of ideas—more political equality, inclusion, and respect; more social services;and more territorial autonomy—they do not necessarily share a vision of howto prioritize and achieve these goals. Insofar as there is now a recognition of theheterogeneity of indigenous interests, this recognition is a political advance;however, insofar as divisions make it difficult to push for political changes, thesedivisions are politically problematic. In this context, the type of indigenous rep-resentation that is being advanced remains somewhat ambiguous.

In each of the cases discussed, high-profile indigenous politicians presented amirror image of representation—literally changing the face of national politics.So, too, these same leaders advanced a guardian image of representation—ad-vancing many claims voiced by indigenous movements, including increased ac-cess to state resources, the creation of new state offices, the monitoring of landand autonomy reforms, and increased public accountability for legislation (not)passed. In many ways, the ability of these politicians to push these reforms restedon the recognition by larger national publics and politicians that there was alarge, mobilized indigenous force. In some cases, mainstream politicians sawvalue in indigenous demands for recognition, representation, and autonomy; inothers, they simply saw the need to capture votes and defuse protest. In eithercase, elected indigenous officials gained the ability to inform agendas (althoughmuch less frequently to pass legislation) because of the existence of a politicallyorganized indigenous movement and indigenous politicians able to mobilize ontheir behalf.

Ironically, perhaps, these successful electoral campaigns have not translatedinto the growth of indigenous movements and an ability to influence more fun-damental and sustained policymaking on behalf of a broader, more comprehen-sive indigenous agenda. For, despite early optimism and fanfare, electoral par-ticipation has posed some (perhaps short-term) challenges to the existingindigenous movements—a dynamic that is also identified in the broader litera-ture on social movements and democracy.39 Piven and Cloward (1979) and Tar-row (1998) have noted that the advent of political parties, more interaction withstate officials, and reform policies can undermine once-vibrant movement or-ganizations that were founded to protest and articulate new agendas—as theirstruggles are subsumed or displaced by these formal institutions and sites of po-litical negotiation. Accordingly, Latin American social movements have histor-ically voiced concern about the destructive impact that political parties and

278 Yashar

alliances with state officials can have on movement autonomy and integrity (seeHellman 1992; Foweraker 1995, chap. 4).40

At this early stage in the game, it would be foolhardy to draw conclusions aboutthe fate of Latin America’s indigenous movements, their decision to take part inelections, and the consequences for all forms of indigenous representation sincethe movements are just beginning to move into electoral and party politics.Moreover, given the varied national contexts—different histories with democ-racy, clientelism, party systems, electoral rules, and the like—we should be waryof simple generalizations for the Latin American indigenous movements as awhole. As Eckstein has stated,

The relationship between democratization and social movements is, in essence, histori-cally contingent. If and when political parties get the upper hand, social movements tendto lose their vitality; however, if they do not or before they do, political parties and so-cial movements may nurture each other. (Eckstein 2001, 398)

This contingency requires a greater span of time in which to observe these re-lationships. For democracy, in general, and electoral participation, in particular,can pose both opportunities and constraints for social movements, in ways aptlydelineated by Eckstein (2001, 398– 400).

That said, a few cautionary observations are in order with respect to the chal-lenges that electoral participation (taken to mean the decision to run in elec-tions) can pose for the unity and integrity of the movements themselves andtheir subsequent prospects for advancing some form of indigenous representa-tion. Four dynamics are highlighted here (drawn from Yashar 2005), the last ofwhich is likely to pose the most difficult challenge to Latin America’s indige-nous movements.

First, as indigenous leaders are elected and appointed to political office (bringing alongwith them an advisory staff ), they often leave indigenous movements with less experiencedleadership to take their place. For relatively young movements, this can be a partic-ular problem. While this would not necessarily be problematic if movementswere better institutionalized, in the short run it has challenged movements toidentify new actors who can assume leadership positions and to institutionalizethe mechanisms for doing so. This challenge was particularly noteworthy inEcuador, as the key executive leadership (which had visibly dominated themovement since the 1980s) chose to run for seats in the legislature beginning in1996. With the official departure from CONAIE of Luis Macas, Nina Pacari,Jose María Cabascango, Miguel Lluco, and others, Ecuador’s national indigenousconfederation had to quickly identify new leaders who could manage the con-federation, command the loyalty of its diverse constituency, and formulate agen-das that would speak to the demands and concerns of their mass base. AntonioVargas was elected in an extremely contentious process. While Vargas subse-quently made a name for himself as the new president of the movement, several


of his actions were severely questioned, and have, at least in the short term, weak-ened the power and influence of the movement. Vargas led CONAIE when it de-cided to form an alliance with the military to overthrow the constitutionallyelected government of Jamil Mahuad in January 2000. Not only did participationin the coup raise questions about the democratic credentials and credibility ofVargas’s leadership, but it also led to cuts in external aid to the movement as awhole (Lucero 2001). Moreover, the fact that CONAIE was quickly sidelined inthe aftermath of the coup raised questions about Vargas’s political skill. It is de-batable if a more experienced leader would have made similar choices, but it iscommonly suggested otherwise. In other words, movements that have beenidentified with the same leadership confront (short-term) obstacles when thatleadership chooses to move into party politics. These obstacles pose challengesfor indigenous movements to credibly and forcefully articulate indigenous agen-das (i.e., for them to serve as principals in the representative process) and to serveas desirable coalition partners for those newly elected indigenous leaders who arenow trying their hand in the game of electoral politics. Strikingly, Luis Macas re-turned to movement politics and was sworn in again as CONAIE’s president in2005, highlighting not only his immense popular support but also the generalcommitment to bring back more experienced leadership.

Second, those indigenous movement leaders who are elected to political office confront aHerculean task. Given their small numbers, it is nearly impossible for them to de-liver on the major demands they once made as movement leaders. Legislative ac-tion requires numbers of votes—it cannot just be mandated once in office. Inthis context, elected indigenous leaders are confronted with what appears to bea choice between (a) maintaining their ideological purity and hence appearingineffective (because they cannot achieve concrete goals), or (b) working to de-liver on some issues via legislative compromise, logrolling, and coalition build-ing—potentially appearing to betray the ideals of the movement. The ability tonavigate these two extremes is no easy task—particularly in a context of prevail-ing economic crisis, weak party systems, and patronage politics. This in turn canhave negative consequences for the movement from which these leaders emerged(Wade 1997, 17). The Bolivian case is telling in this regard. When Víctor HugoCárdenas was elected as Bolivia’s first indigenous vice president, he achieved na-tional and international kudos for his role in creating greater spaces for indige-nous voices and advances in important legislation. Nonetheless, he was widelycritiqued by the Kataristas in the Andes for betraying the ideals of the movement.While this criticism stemmed in part from older debates and divisions withinKatarismo, it was not limited to this; indeed, Cárdenas ended up working mostclosely with CIDOB, from the Amazon (with whom he developed cordial andproductive relations), rather than with CSUTCB, from the Andes (with whomrelations were conflictual). The Ecuadorean case is also suggestive on this score.

280 Yashar

As noted earlier, CONAIE had an impressive first showing in the 1996 elections.Just two years later, however, CONAIE witnessed a decline in electoral support,from 10 to 5 percent of congressional seats (Mijeski and Beck 1998, 12). The rea-sons for this decline in support are up for debate. But the data unequivocallyindicate that one cannot assume that indigenous electoral participation translatesinto constant and ongoing support. To date, we have no indication that this isthe case. To the contrary, indigenous officials confront even more difficult tasks:not only must they define their electoral constituency in a national context ofweak parties, apathetic electorates, and economic downturn, but they mustmaintain their image as political warriors for the indigenous movement fromwhich they emerged. As a case in point, in 2003, Ecuadorean president LucioGutiérrez forged a governing coalition in the legislature with Pachakutik and theMovimiento Popular Democrática (Popular Democratic Movement). After alittle more than two hundred days, Pachakutik pulled out (following CONAIE’scondemnation of the government), and tensions between Pachakutik andCONAIE ensued—with leaders from both groups denying that the rupture wasserious or noteworthy.

Third, highly respected leaders of indigenous movements are not necessarily embracedas ideal elected officials, particularly in better-established political party systems. We can-not assume that ethnic identification translates into votes for those who sharethat ethnic background; mirror representation is not always what people seek.In countries with weak party systems, we do find some notable examples whereindigenous leaders have won elections. As in Ecuador, nationally recognized in-digenous leaders in Guatemala were elected to the national legislature—for ex-ample, Rosalina Tuyuc, the indigenous spokeswomen of the Coordinadora Na-cional de Viudas de Guatemala (National Coordinator of Guatemalan Widows,or CONAVIGUA). However, in those cases where there existed a “compara-tively” stronger and older party system, as in Bolivia, we find a more checkeredhistory.41 Evo Morales emerged as an important presidential candidate in 2002and won the 2005 presidential election. Moreover, CSUTCB indigenous ac-tivists from Bolivia’s coca-growing region won legislative races in the 1990s.However, these electoral examples are the exceptions rather than the rule inBolivia. Indeed, Bolivia’s movements in the Andes (CSUTCB) and Amazon(CIDOB) suffered miserable failures when they initially entered the electoralarena, in races at both the executive and legislative levels. These electoral failuresfollowed on the heels of remarkable movement successes in mobilizing indige-nous people and negotiating favorable policy outcomes with the government inplace. In this context, CSUTCB leaders in the mid-1980s and CIDOB leaders inthe mid-1990s had expected to perform admirably in the electoral arena.42 Notonly did the leadership perform poorly in national elections in both cases (as waslargely expected), but perhaps even more surprising, they performed miserably at


the legislative level as well (which was unexpected). At the very least, indigenouscandidates thought that they would do well in their home districts or states. Yetin the 1997 elections, CIDOB (which allied with the Movimiento Bolivia Li-bre—MBL, or Free Bolivia Movement) did not manage to elect a single legisla-tive candidate. Hypotheses about these failed showings abound—including thefailure of the leadership to consult with their base and the internalized racism thatleads many indigenous people to question if their own leaders will and can per-form adequately in the formal, “white” world of electoral politics. Either way, thesimple fact is that one cannot predict indigenous electoral success from indige-nous movement success. The case of Bolivia is an important reminder of this.43 Inthe short-term, these disastrous electoral showings weakened indigenous move-ments and their indigenous movement leaders cum electoral candidates, asbroader constituencies came to question their political choice to enter electionsand the failed outcomes, having done so.

Fourth, as indigenous leaders engage in partisan politics, indigenous movements are morelikely to fall prey to partisan competition, thereby exposing themselves to the kinds of po-litical cleavages that can divide movements and weaken demands for recognition, represen-tation, and reform. As with the third observation, this is particularly problematic incountries with relatively stronger political party systems. This kind of partisancompetition is likely to happen anyway in a competitive electoral system. But asindigenous leaders search for partisan affiliations or coalitions, they accelerate la-tent partisan divisions within a given movement. As long as an indigenous move-ment does not take a formal political stance, it is possible for the same movementto house multiple partisan affiliations. Where and when movements formally de-cide to forge and/or ally with a given party, however, these political divides aremade manifest. This dynamic was illustrated by indigenous movements that con-sciously chose not to take a partisan stand to avoid divisions within the movement(as with CONAIE in Ecuador, prior to 1996), and by movements that engageddirectly in elections and confronted internal infighting (as with CSUTCB andCIDOB in Bolivia, and CONAIE following its decision to take part in electionsin 1996). Indeed, the CSUTCB has been wracked by partisan divides since the1980s. And CONAIE has confronted increasing internal conflict in the first de-cade of the twenty-first century over who to field for president and how to keepthe different regional organizations together under one confederational um-brella. Indeed, in the 2002 presidential election, CONAIE members divided overwhom to support—with some supporting former CONAIE president AntonioVargas and others supporting Auki Tituana. Moreover, the relationship betweenCONAIE and Pachakutik has become strained as the former has sought to pre-vent the latter from changing from a movement to a party. The ways in whichpartisan conflicts have played out within social movements have led scholars suchas Sieder (2002), Albó (2002), and Calla (2000) to observe that political partiesand partisan competition can further divide rural indigenous communities.

282 Yashar

The dangers of the co-optation of leaders and the fragmentation of indigenous move-ments in the post-constitutional phase of reform are high. In part this explains why manyindigenous activists have rejected political parties all together. (Sieder 2002, 9)

Conclusion

In sum, politics has been highly contested in recent years in the Andes. In a con-text of weak states, parties, and party systems, much of democratic politics in theAndes has taken a downward spiral—the focus of this volume. One of the few ar-eas where new voices are being heard and novel legislation is being passed is in therealm of ethnic politics. Rising and powerful indigenous movements have beeneffective claim makers and have compelled politicians to debate an emerging in-digenous agenda—although not always at a pace and with a content deemed ac-ceptable to indigenous people. In this regard, movements have forced public de-bate, although they have not always been able to select their representatives in thatprocess.

The strength of these new indigenous movements in tandem with new insti-tutional reforms (extending suffrage, decentralization, etc.) poses new and im-portant opportunities for indigenous social movements to try to translate theirstreet power into state power. As the previous section has shown, this is a tall or-der, posing challenges both for the integrity of the movements as well as for the newly elected indigenous leaders now in office. The last section highlightedthat in the short run this dual strategy has tested the unity and integrity of in-digenous movements, just as it has stymied most new indigenous politiciansconfronted by weak states, parties, budgets, and economies. The weakness ofAndean states is likely to be the greatest obstacle of all to political representa-tion—as voting is unevenly institutionalized, policies are poorly implemented,corruption is rampant, and the rule of law is spotty. Hence, while the crisis ofdemocratic representation is certainly a function of weak competitive party sys-tems in the countries discussed at greatest length in this book,44 it is more pro-foundly a function of the weak reach of the state (O’Donnell 1993; Yashar 2005;Mainwaring, this volume). Indeed, even laudable political reforms have beencompromised by the inability of the state to implement them, the resistance ofauthoritarian social forces, and the weak ability and presence of the state (in par-ticular, in the countryside).

Thus, indigenous movements and leaders confront particularly high hurdles torepresenting those in whose name they have been elected. For while the currentperiod has advanced micro-analytic and mirror conceptions of indigenous rep-resentation (getting more indigenous people to vote and be elected), it has dem-onstrated the ongoing weakness of macro-analytic conceptions of indigenousrepresentation (being able to sustain indigenous movements and parties, and passand implement legislation advancing collectively defined indigenous interests).


It is in this context, then, that efforts to organize within civil and political so-ciety are so striking and so necessary for advancing democratic representation—even if at times organizing parties and entering elections can marginalize socialmovements in the process. For while particular indigenous movements and indigenous political officials might not survive in the short-term, it is clear thatinteraction between movements and parties has fundamentally created a newpolitical imperative. Indigenous mobilization in both realms has solidified in-digenous peoples as political actors whose interests are, at least now, part of thenational dialogue; other political parties must at least take a stand on some of theissues associated with agendas articulated by indigenous social movements.While we cannot be so sanguine about the degree to which “indigenous inter-ests” will or will not be institutionalized, we can be certain that indigenous peo-ples are part of the citizenry and electorate and that their political issues (diverseas they may be) are part of the national dialogue. Latin America’s indigenousmovements thus have forged a fundamental but unresolved political debateabout how best to design political representation in multiethnic polities.

Notes

I thank the editors of this volume and José Antonio Lucero for their valuable sugges-tions and insights. Of course, all the normal caveats apply.

1. This chapter uses the terms “Andean region” in a very loose sociological, and onlypartially geographic, sense. I use the term to refer colloquially to those Latin Americancountries with portions of their national territories in the Andes, excluding Chile. Manyof these same countries also straddle the Amazon, including Ecuador, Bolivia, Peru,Colombia, and Venezuela.

2. Lagos (2003a, 2003b). She reports that while support for democracy remains rela-tively high, support for existing democratic institutions is weak.

3. This section draws on arguments in Yashar (1996, 1998, 1999, 2005).4. See Collier and Collier (1991) for a conceptual, theoretical, and empirical treat-

ment of the ties between corporatist projects, labor, political parties, and the state.5. See Stavenhagen (1988, 105; 2002, 27) and Maybury-Lewis (1991) for a discussion

of these institutes. While Brazil formed an Indian office in 1910, other Latin Americancountries largely founded these offices in the 1930s and 1940s.

6. See Stavenhagen (1992) and Wade (1997) for an overview of Latin America. SeeMallon (1992) for a discussion of the varied contexts and forms that this policy took inMexico, Peru, and Brazil. This attempt to create a more homogeneous population con-trasted with U.S. history, where more rigid social lines were drawn between Indians,blacks, and the colonial population.

7. Even if ethnic identity was understood as fluid, states and landlords often contin-ued to repress these same communities (when rebellious in the face of state colonization,development plans, and repressive rural labor relations) according to a rigid understand-ing of the appropriate class status of the heretofore “indigenous” population.

284 Yashar

8. Indigenous movements mobilized in defense of autonomy in the Andes and theAmazon. However, the process by which Andean and Amazonian indigenous movementsformed varied—with Andean communities seeking to defend autonomy in the contextof eroding corporatist citizenship regimes and Amazonian indigenous communities seek-ing to defend autonomy in the context of development reforms (some tied to corporatistland reforms) that promoted colonization of their lands. See Yashar (2005) for an expla-nation of why, where, and how indigenous movements have emerged in Latin America.

9. My theoretical thinking on social movements and the issue of networks and op-portunities draws from the social movement literature, in particular Tarrow (1998),McAdam, McCarthy, and Zald (1996), and McAdam, Tarrow, and Tilly (2001).

10. Indigenous organizing in Peru has been relatively weak in the contemporary pe-riod. However, subnational organizing did occur in Puno and in parts of the Amazon,where political associational space and transcommunity networks remained stronger thanin the rest of the country. In other words, the same three factors (state reforms that chal-lenge local autonomy, transcommunity networks, and political associational space) usedto explain the emergence of strong movements in Bolivia and Ecuador can be used toexplain subnational variation in Peru’s overall weak history of indigenous organizing.

11. Sources: 2001 Statistical Abstract of Latin America 37, Table 532: 104; Statistical Ab-stract of Latin America 30, Part 1: 150; Mayer and Masferreer (1979, 220 –21); and Varese(1991).

12. Demographic data about the numbers of indigenous people residing in eachcountry are based on (not terribly reliable) estimates rather than precise calculations. Thenumbers do not reveal the ways in which indigenous communities have changed withrespect to the meaning, content, scope, and form of identities, practices, or goals of in-digenous peoples. Nor do these figures intend to stipulate a shared identity among in-digenous peoples. That said, there is little dispute that large indigenous populations re-side in Bolivia, Peru, and Ecuador, with significantly fewer numbers in Colombia andVenezuela. Sources for data: 2001 Statistical Abstract of Latin America 37, Table 532: 104;Statistical Abstract of Latin America 30, Part I: 150; Mayer and Masferreer (1979, 220 –21);Varese (1991).

13. The following does not presume to be an exhaustive discussion of debates aboutrepresentation. Its goal is much less ambitious. Rather, it hopes to delineate some of theconceptual questions that arise as ones tackles questions about indigenous representation.

14. This observation recalls old Marxist debates about identifying a class in and for it-self. How does the analyst determine the interests of those who themselves do not agreeon their primary identification, goals, and priorities?

15. My understanding of “representation” has been most clearly shaped by Pitkin’sclassic book, The Concept of Representation (1967). I will not engage explicitly with herbook here, although my debt to her is great.

16. See Lucero (2002, chap. 2) for a provocative analysis of representation—high-lighting representation as mirror, filter, and producer of identities and interests.

17. This last indicator does not address if these representatives are accountable and, ifso, to whom.

18. This section on institutions draws heavily on the primary work of RodolfoStavenhagen, Xavier Albó, Ricardo Calla, Donna Lee Van Cott, José Antonio Lucero,Rachel Sieder, and Raquel Yrigoyen Fajardo, all listed in the References; and KathleenO’Neill, who contributed Chapter 6 to this volume.

19. This volume has primarily focused on representation via elected officials. In linewith the themes of the volume, this chapter focuses on the changing institutions andnorms that have facilitated the election of indigenous leaders and/or people. However,


it is worth noting that several states have attempted to respond to the question of “rep-resentation” and “inclusion” by forming indigenous offices within the executive branch.On the whole, these offices have been largely dependent on executive whim. Hence, inBolivia, former president Gonzalo Sánchez de Lozada (1993–97) decided in his first termto form super-ministries, one of which included an Office of Indigenous Affairs. WhenHugo Banzer became president of Bolivia in 1997, he created a new Ministry for Peas-ants, Native Peoples, and Indigenous Affairs.

20. See Stokes (2001), in particular, for her discussion of bait-and-switch politics.The observations made here also apply to societal opposition to Chávez in Venezuela—although indigenous people are not at the forefront of these protests and, therefore, thisdevelopment is not listed in the text.

21. Dandler (1996), Stavenhagen (1992, 2002), Van Cott (2000, 265– 68; 2002), andYrigoyen (2000). In a striking Guatemalan referendum in May 1999, the voting popu-lation (18 percent of the eligible electorate) rejected proposed reforms to amend the con-stitution and acknowledge the multiethnic composition of the country (as outlined inthe peace accords). For one of the few academic discussions of this surprising episode,see Warren (2002).

22. Indigenous movements have appealed to norms, laws, and organizations operat-ing in the international arena. As discussed by Brysk (1994, 1996, 2000) and Wilmer(1993), the international arena has provided a new discourse, funds, and forums that haveoften shaped debates about indigenous rights. In particular, indigenous movements havelobbied Latin American states to ratify the International Labour Organization (ILO)Convention 169 on Indigenous and Tribal Peoples in Independent Countries. Conven-tion 169 outlines the rights of indigenous peoples and the responsibilities of multiethnicstates toward them. At a minimum, it calls on states to recognize ethnic heterogeneitywhere states had advanced nationalist aspirations of mestizo homogeneity. The follow-ing Latin American states have ratified ILO Convention 169: Mexico (1990), Bolivia(1991), Colombia (1991), Costa Rica (1993), Paraguay (1993); Peru (1994), Honduras(1995), Guatemala (1996), Ecuador (1998), and Argentina (2000). Ratification providesa mechanism for advocating reforms to accommodate ethnically diverse populations.While these Latin American states have yet to live up to the terms of the convention, ithas provided a language, legitimacy, and set of transnational advocacy networks to con-tinue work on these issues at home.

23. Van Cott (2005) argues convincingly that where countries have relaxed the re-quirements for candidate registration (such that one does not need to be a member of aformal political party to be on the ballot), indigenous people and movements have greaterprospects of running in and winning elections. Of course, the same requirements that canincrease the chance of indigenous representation can simultaneously open the doors tomore easily elect catchall politicians with fewer institutional mechanisms for holdingthem accountable to their constituencies.

24. Interviews in 1997 by the author with Ecuadorean indigenous leaders Leonardo Viteri (March 6, 1997, Quito) and César Cerda (May 6, 1997, Quito); and Ecuadoreanpolitician and consultant Gonzalo Ortiz Crespo (February 27 and March 11, 1997,Quito). Also see Selverston (1994, 146) and Selverston-Scher (2001, 45).

25. Interviews in Bolivia by the author with indigenous leaders Marcial Fabricano ofCPIB and CIDOB ( June 13 and 20, La Paz) and Ernesto Noe of CPIB ( July 25, 1997,Trinidad); with researchers Zulehma Lehm (August 1, 1997, Trinidad) and WilderMolina ( July 29, 1997, Trinidad) at CIDDEBENI; and with lawyer Carlos RomeroBonifaz of CERES ( July 1, 1997, Santa Cruz, and July 29, 1997, Trinidad). See Liber-mann and Godínez (1992); Navia Ribera (1996); Molina (1997); Van Cott (2000).

286 Yashar

26. Interviews by the author in Bolivia with Isabel Lavadenz, former national direc-tor of the Bolivian National Institute of Agrarian Reform (August 4 and 5, 1997, La Paz),and Jorge Múñoz, researcher at UDAPSO (May 31, 1997, La Paz). See also Múñoz andLavadenz (1997).

27. Interview by the author with Bolivian lawyer Alcides Vadillo ( June 11, 1997, La Paz).

28. Interviews by the author in Bolivia with Luz María Calvo ( July 9, 1997, La Paz)and George Gray Molina (May 23, 1997, La Paz).

29. Also see Balslev (1997, particularly Annex 2, 117–21).30. This section draws on Yashar (2005, chap. 7).31. Of course some indigenous movements did forge political parties in an earlier pe-

riod. The Colombian indigenous movement has been engaging in party politics since the1990s, when it mobilized for the 1990 Constituent Assembly. In Bolivia, one finds evenearlier efforts to forge indigenous political parties in the 1970s, although these were over-whelmingly unsuccessful until the late 1990s.

32. While both Bolivia and Ecuador have comparatively weak, patronage-driven po-litical systems, Bolivia maintained a considerably more stable political party system thanEcuador. In Ecuador, political parties have largely unintelligible programmatic differ-ences, weak roots in society, little party discipline, and scant institutional endurance. Bo-livia, by contrast, has had a history of significant and enduring political parties. Most no-table among them is the MNR, dating back to the 1940s, and the once-socialist MAS,among others—although even in Bolivia most of these long-standing political partiesnow have little ideological coherence and unpredictable electoral support.

33. CONDEPA, a populist party that has reached out to Bolivia’s cholo community,is one such example. However, I will not comment on this party insofar as it is not anexample of an indigenous movement that then turned to party politics. Indeed, while in-cluding indigenous discourse and prominent indigenous figures, and while gaining asignificant electoral toehold among Bolivia’s cholo community, it does not parallel the ex-amples discussed in the text of indigenous movements turning to party politics.

34. In the 1997 Bolivian elections, the rivalry between Evo Morales and AlejoVeliz—both of whom wanted to assume the mantle of the cocalero movement—was alsopalpable.

35. Electoral data comes from “Bolivia: Elecciones Presidenciales de 2002,” PoliticalDatabase of the Americas (Georgetown University and the Organization of AmericanStates, 1999), http://www.georgetown.edu/pdba/Elecdata/Bolivia/pres02.htm.

36. Diario Hoy, August 24, 1996.37. Mijeski and Beck (1998, 4), and Pallares (1997, 544)—both cite the Washington

Post, July 23, 1996.38. With the entry into formal politics, several conflicts emerged over political al-

liances and appointments to ministries; charges of corruption and opportunism, particu-larly leveled against Amazonian leaders; tensions between Pachakutik, CONAIE, andNuevo País; etc. For a discussion of some of these conflicts, see Interpress Service, No-vember 1996; Diario Hoy, June 13, 1996, 3A; Mijeski and Beck (1998, 5); and Andolina(1999, 225–32).

39. For a discussion of social movements and democracy in Latin America, see, inparticular, Alvarez and Escobar (1992, particularly chapters by Hellman and Canel), andEckstein (2001). For a more general discussion of movement cycles (including their de-mise), see Piven and Cloward (1979) and Tarrow (1998).

40. There is also a significant literature that has analyzed what has become of anti-authoritarian movements in post-transition settings. Alvarez and Escobar (1992), Canel


(1992), and Schneider (1992) have noted that democratization (and the decline in human rights abuses) can take away the raison d’être of movements that once definedtheir mission as anti-authoritarian. Schneider also notes that different kinds of politicalparties can displace these movements once they regain the ability to negotiate in the po-litical sphere.

41. Bolivia has a weak party system, as noted by Mainwaring and Scully (1995).However, compared to Ecuador, Guatemala, and Peru, its system has been relativelymore institutionalized.

42. Several 1997 interviews by the author in Bolivia with indigenous leaders MarcialFabricano ( June 13 and 20, 1997, La Paz) and José Urañabi ( July 2, 1997), as well as col-lective interviews led by CIDOB ( June 27–30, 1997) in Camiri, Villamonte, and Mon-teagudo.

43. Importantly, indigenous representation at the municipal level did increase withdecentralization, as noted earlier.

44. Mainwaring and Scully (1995) observed that Ecuador, Bolivia, Peru (and I wouldadd Guatemala) have inchoate party systems. They are neither institutionalized norstable. Mexico, by contrast had (until recently) a hegemonic party system—preventingeffective and meaningful forms of competitive democratic participation. Scholars of thesecases commonly refer to a crisis of representation—particularly in the Andean cases.

References

Albó, Xavier. 2002. “Bolivia: From Indian and Campesino Leaders to Councillors andParliamentary Deputies.” In Multiculturalism in Latin America: Indigenous Rights, Diver-sity and Democracy, ed. Rachel Sieder, 74 –102. London: Institute of Latin AmericanStudies, Palgrave Macmillan.

Alvarez, Sonia, and Arturo Escobar. 1992. “Conclusion: Theoretical and Political Hori-zons of Change in Contemporary Latin American Social Movements.” In The Mak-ing of Social Movements in Latin America, ed. Arturo Escobar and Sonia Alvarez, 317–29.Boulder, CO: Westview Press.

Andolina, Robert James. 1998. “CONAIE (and Others) on the Ambiguous Spaces ofDemocracy: Positioning for the 1997–98 Asamblea Nacional Constituyente inEcuador.” Paper presented at the 21st International Congress of the Latin AmericanStudies Association, Chicago, September 24 –26.

———. 1999. “Colonial Legacies and Plurinational Imaginaries: Indigenous MovementPolitics in Ecuador and Bolivia.” Ph.D. diss., University of Minnesota.

Balslev, Anne. 1997. Distritos municipales indígenas en Bolivia: Las primeras experiencias en elChaco. La Paz: Subsecretaría de Asuntos Etnicos.

Brysk, Alison. 1994. “Acting Globally: Indian Rights and International Politics in LatinAmerica.” In Indigenous Peoples and Democracy in Latin America, ed. Donna Lee VanCott, 28–51. New York: St. Martin’s Press.

———. 1996. “Turning Weakness into Strength: The Internationalization of IndianRights.” Latin American Perspectives 23, no. 2: 38–57.

———. 2000. From Tribal Village to Global Village: Indian Rights and International Relationsin Latin America. Stanford, CA: Stanford University Press.

288 Yashar

Calla, Ricardo. 2000. “Indigenous Peoples, the Law of Popular Participation, and Changesin Government: Bolivia, 1994 –1998.” In The Challenge of Diversity: Indigenous Peoplesand Reform of the State in Latin America, ed. Willem Assies, Gemma Van der Haar, andAndré Hoekema, 77–94. Amsterdam: Thela Thesis.

Canel, Eduardo. 1992. “Democratization and the Decline of Urban Social Movementsin Uruguay: A Political-Institutional Account.” In The Making of Social Movements inLatin America, ed. Arturo Escobar and Sonia Alvarez, 276 –90. Boulder, CO: West-view Press.

Clavero, Bartolomé. 1999. Derecho indígena y cultura constitucional en América. MexicoCity: Siglo Veintiuno.

Collier, David, and Ruth Berins Collier. 1991. Shaping the Political Arena: Critical Junctures,the Labor Movement, and Regime Dynamics. Princeton, NJ: Princeton University Press.

Dandler, Jorge E. 1996. “Indigenous Peoples and the Rule of Law in Latin America: DoThey Have a Chance?” Paper presented at a workshop on the Rule of Law and theUnderprivileged in Latin America, Helen Kellogg Institute for International Studies,University of Notre Dame, November 9–11.

Eckstein, Susan. 2001. “Where Have All the Movements Gone: Latin American SocialMovements at the Millennium.” In Power and Popular Protest: Latin American SocialMovements, ed. Susan Eckstein, 351– 406. Updated and expanded version. Berkeley:University of California Press.

Foweraker, Joe. 1995. Theorizing Social Movements. London: Pluto Press.Gray Molina, George. 2003. “The Offspring of 1952: Poverty, Exclusion, and the Prom-

ise of Popular Participation.” In Proclaiming Revolution: Bolivia in Comparative Perspec-tive, ed. Merlee S. Grindle and Pilar Domingo, 345– 63. London: Institute of LatinAmerican Studies, University of London; Cambridge, MA: David Rockefeller Centerfor Latin American Studies, Harvard University.

Hagopian, Frances, and Scott Mainwaring, eds. 2005. The Third Wave of Democratizationin Latin America: Advances and Setbacks. Cambridge: Cambridge University Press.

Hellman, Judith. 1992. “The Study of New Social Movements in Latin America and theQuestion of Autonomy.” In The Making of Social Movements in Latin America, ed. Arturo Escobar and Sonia Alvarez, 51– 61. Boulder, CO: Westview Press.

Lagos, Marta. 2003a. “Public Opinion.” In Constructing Democratic Governance in LatinAmerica, ed. Jorge I. Domínguez and Michael Shifter, 137– 61. 2nd ed. Baltimore:Johns Hopkins University Press.

———. 2003b. “A Road with No Return?” Journal of Democracy 14, no.2 (April): 163–73.Libermann, Kitula, and Armando Godínez, eds. 1992. Territorio y dignidad: Pueblos indí-

genas y medio ambiente en Bolivia. La Paz: ILDIS; Caracas: Editorial Nueva Sociedad.Lucero, José Antonio. 2001. “High Anxiety in the Andes: Crisis and Contention in

Ecuador.” Journal of Democracy 12, no. 2 (April): 59–73.———. 2002. “Arts of Unification: Political Representation and Indigenous Movements

in Bolivia and Ecuador.” Ph.D. diss., Princeton University.Mainwaring, Scott, and Timothy R. Scully. 1995. “Introduction: Party Systems in Latin

America.” In Building Democratic Institutions: Party Systems in Latin America, ed. ScottMainwaring and Timothy R. Scully, 1–34. Stanford, CA: Stanford University Press.

Mallon, Florencia. 1992. “Indian Communities, Political Cultures, and the State in LatinAmerica, 1780 –1990.” Quincentenary Supplement, Journal of Latin American Studies24: 35–53.

Maybury-Lewis, David. 1991. “Becoming Indian in Lowland South America.” In Nation-States and Indians in Latin America, ed. Greg Urban and Joel Sherzer, 207–35. Austin:University of Texas Press.


Mayer, Enrique, and Elio Masferreer. 1979. “La población indígena de América en1978.” América Indígena 32, no. 1 (April–June): 220 –21.

McAdam, Doug, John D. McCarthy, and Mayer N. Zald, eds. 1996. Comparative Per-spectives on Social Movements: Political Opportunities, Mobilizing Structures, and CulturalFramings. Cambridge: Cambridge University Press.

McAdam, Doug, Sidney Tarrow, and Charles Tilly. 2001. Dynamics of Contention. NewYork: Cambridge University Press.

Mijeski, Kenneth J., and Scott H. Beck. 1998. “Mainstreaming the Indigenous Move-ment in Ecuador: The Electoral Strategy.” Paper presented at 21st International Con-gress of the Latin American Studies Association, Chicago, September 24 –26.

Molina, Wilder M. 1997. “El movimiento social indígena del Beni en el contexto delproceso de consolidación de la movilización intercomunal hasta la Marcha por el Ter-ritorio y la Dignidad (1987–1990).” Manuscript, Centro de Investigación y Docu-mentación para el Desarrollo del Beni (CIDDEBENI), Trinidad, El Beni, Bolivia.

Múñoz, Jorge A., and Isabel Lavadenz. 1997. “Reforming the Agrarian Reform in Bo-livia.” Development Discussion Paper for the Harvard Institute for International De-velopment, Harvard University, and UDAPSO, Bolivia. Presented in Cambridge,MA, and La Paz, Bolivia.

Navia Ribera, Carlos. 1996. “Reconocimiento, demarcación y control de territorios in-dígenas: Situación y experiencias en Bolivia.” Working Paper, no. 34. Centro de In-vestigación y Documentación para el Desarrollo del Beni (CIDDEBENI), Trinidad,El Beni, Bolivia.

O’Donnell, Guillermo. 1993. “On the State, Democratization, and Some ConceptualProblems: A Latin American View with Glances at Some Postcommunist Countries.”World Development 21, no. 8 (August): 1355– 69.

Pallares, Amalia Veronika. 1997. “From Peasant Struggles to Indian Resistance: PoliticalIdentity in Highland Ecuador, 1964 –1992.” Ph.D. diss., University of Texas at Austin.

Pitkin, Hanna Fenichel. 1967. The Concept of Representation. Berkeley: University of Cali-fornia Press.

Piven, Frances Fox, and Richard A. Cloward. 1979. Poor People’s Movements: Why TheySucceed, How They Fail. New York: Vintage Books.

Plant, Roger. 2002. “Latin America’s Multiculturalism: Economic and Agrarian Dimen-sions.” In Multiculturalism in Latin America: Indigenous Rights, Diversity, and Democracy,ed. Rachel Sieder, 208–26. London: Institute of Latin American Studies, PalgraveMacmillan.

Schneider, Cathy. 1992. “Radical Opposition Parties and Squatter Movements in Pin-ochet’s Chile.” In The Making of Social Movements in Latin America, ed. Arturo Escobarand Sonia Alvarez, 260 –75. Boulder, CO: Westview Press.

Selverston, Melina H. 1994. “The Politics of Culture: Indigenous Peoples and the Statein Ecuador.” In Indigenous Peoples and Democracy in Latin America, ed. Donna Lee VanCott, 131–52. New York: St. Martin’s Press, in association with the Inter-AmericanDialogue.

Selverston-Scher, Melina. 2001. Ethnopolitics in Ecuador: Indigenous Rights and the Strength-ening of Democracy. Coral Gables, FL: North-South Center Press.

Sieder, Rachel. 2002. Introduction to Multiculturalism in Latin America: Indigenous Rights,Diversity and Democracy, ed. Rachel Sieder, 1–23. London: Institute of Latin Ameri-can Studies, Palgrave Macmillan.

Stavenhagen, Rodolfo. 1988. Derecho indígena y derechos humanos en América Latina. Mex-ico City: El Colegio de Mexico, Instituto Interamericano de Derechos Humanos.

———. 1992. “Challenging the Nation-State in Latin America.” Journal of InternationalAffairs 45, no. 2 (Winter): 421– 40.

290 Yashar

———. 2002. “Indigenous Peoples and the State in Latin America: An Ongoing De-bate.” In Multiculturalism in Latin America: Indigenous Rights, Diversity, and Democracy,ed. Rachel Sieder, 24 – 44. London: Institute of Latin American Studies, PalgraveMacmillan.

Stokes, Susan. 2001. Mandates and Democracy: Neoliberalism by Surprise in Latin America.New York: Cambridge University Press.

Swain, Carol M. 1995. Black Faces, Black Interests: The Representation of African Americansin Congress. Cambridge, MA: Harvard University Press.

Tarrow, Sidney G. 1998. Power in Movement: Social Movements and Contentious Politics.Cambridge: Cambridge University Press.

Tate, Katherine. 2003. Black Faces in the Mirror: African Americans and Their Representativesin the U.S. Congress. Princeton, NJ: Princeton University Press.

Van Cott, Donna Lee. 2000. The Friendly Liquidation of the Past: The Politics of Diversity inLatin America. Pittsburgh: University of Pittsburgh Press.

———. 2002. “Constitutional Reform in the Andes: Redefining Indigenous-States Re-lations.” In Multiculturalism in Latin America: Indigenous Rights, Diversity, and Democracy,ed. Rachel Sieder, 45–73. London: Institute of Latin American Studies, PalgraveMacmillan.

———. 2005. Ethnic Political Parties in Latin America. Cambridge: Cambridge UniversityPress.

———. Forthcoming. “Turning Crisis into Opportunity: The Achievements of Ex-cluded Groups in the Andes.” In State and Society in Conflict: Comparative Perspectiveson Andean Crises, ed. Paul Drake and Eric Hershberg. Pittsburgh: University of Pitts-burgh Press.

Varese, Stefano. 1991. “Think Locally, Act Globally.” NACLA: Report on the Americas 25,no. 3.

Velasco, Luisa. 2001. “Areas de desarrollo indígena y distritos municipales indígenas: Eldesarrollo indígena desde las políticas de Chile y Bolivia.” Revista Madrid (Universidadde Chile), no. 4 (May): 1– 64.

Wade, Peter. 1997. Race and Ethnicity in Latin America. London: Pluto Press.Warren, Kay B. 1998. Indigenous Movements and Their Critics: Pan-Maya Activism in

Guatemala. Princeton, NJ: Princeton University Press.———. 2002. “Voting Against Indigenous Rights in Guatemala: Lessons from the 1999

Referendum.” In Indigenous Movements, Self-Representation, and the State in Latin Amer-ica, ed. Kay B. Warren and Jean E. Jackson, 149–80. Austin: University of Texas Press.

Wilmer, Franke. 1993. The Indigenous Voice in World Politics. Thousand Oaks, CA: SagePress.

Yashar, Deborah J. 1996. “Indigenous Protest and Democracy in Latin America.” InConstructing Democratic Governance: Latin America and the Caribbean in the 1990s, ed.Jorge I. Domínguez and Abraham Lowenthal, 87–105. Baltimore: Johns HopkinsUniversity Press.

———. 1998. “Contesting Citizenship: Indigenous Movements and Democracy inLatin America.” Comparative Politics 31, no. 1 (October): 23– 42.

———. 1999. “Democracy, Indigenous Movements, and the Postliberal Challenge inLatin America.” World Politics 52 (October): 76 –104.

———. 2005. Contesting Citizenship in Latin America: The Rise of Indigenous Movementsand the Postliberal Challenge. Cambridge: Cambridge University Press.

Yrigoyen Fajardo, Raquel. 2000. “The Constitutional Recognition of Indigenous Law inAndean Countries.” In The Challenge of Diversity: Indigenous Peoples and Reform of theState in Latin America, ed. Willem Assies, Gemma Van der Haar, and André Hoekema,197–222. Amsterdam: Thela Thesis.


Part IV

❖

CONCLUSION

In this chapter, I address the causes of the very low confidence in parties andlegislatures in the Andes. Low confidence in parties and legislatures is an im-portant manifestation of the crisis of democratic representation. Moreover, pre-sumably the same factors that explain the low confidence in parties and legisla-tures also help to account for the broader crisis of democratic representation.1

Therefore, at a broader theoretical level, I also attempt to contribute to under-standing why agents of democratic representation have become discredited inthe Andes.

A great deal has been written about institutional trust in the advanced indus-trial democracies. In contrast, with the exception of some excellent works byMishler and Rose (1997, 2001), work on institutional trust elsewhere is just start-ing to emerge.2 This is a striking omission considering that the crisis of trust inthe institutions of representative democracy is more profound in Latin Americathan in the advanced industrial democracies (see Chapter 1). Moreover, lookingat a wider sample of countries than the advanced industrial democracies canbroaden theoretical understanding of the causes of widespread disaffection withdemocratic representation.

My initial theoretical starting point was the literature on declining confidencein parties and legislatures in the advanced industrial democracies. This literaturehas generated interesting hypotheses that I initially presumed would be useful forunderstanding the Andes. Nevertheless, low confidence in parties and assembliesin the Andes has some specificities, such that the literature on the advanced in-dustrial democracies is not fully adequate to understand the Andes. Some of theliterature on the advanced industrial democracies, such as work that explains de-clining trust in institutions as a result of increasing skepticism among youngervoters or of a growing number of postmaterialist voters (Inglehart 1997b), is sim-ply misplaced for the Andes.

10

State Deficiencies, Party Competition, and Confidence in Democratic

Representation in the Andes

❖

Scott Mainwaring

296 Mainwaring

My argument about confidence in democratic representation in the Andes istwofold. First, I claim that the primary cause of the crisis of confidence in dem-ocratic representation has been state deficiencies. The term “state deficiencies”implies something more than merely poor governmental performance. By statedeficiencies, I mean that the state does not fulfill some of its basic governing,legal, and security functions. In this respect, the works by Beissinger and Young(2002), Bejarano and Pizarro Leongómez (2005), Linz and Stepan (1996), O’Donnell (1993, 2003), Rotberg (2004), Waldmann (2003), and Zartman(1995) on stateness and state failure provide more insight into the crisis of dem-ocratic representation in the Andes than the literature on trust in representativeinstitutions3 in the advanced industrial democracies.

Four of the five Andean countries, all but Colombia, have had poor eco-nomic and social performances since the early 1980s. In two other key areas, en-suring personal security for citizens and ensuring their legal rights, all five An-dean states have often failed. Poor state performance has negatively affectedcitizen evaluations of parties and assemblies. Citizens need states to devise poli-cies that address their salient concerns—jobs, income, housing, health, educa-tion, and personal security—and to enforce their rights as citizens. When statesfail to attend to these problems adequately, citizens tend to become disen-chanted with parties and assemblies. Widespread corruption has contributed tolow trust in parties and assemblies because rational citizens would be foolish totrust politicians who plunder from the public coffers.

The second part of my explanation is that political conflicts and citizen ex-pectations shape public evaluations of different institutions. Electoral competi-tion in an era of mass media has accentuated states’ performance problems. Parties and politicians have electoral incentives to politicize and publicize thefailures of competing parties and of governments they oppose. The messages thatbombard citizens regarding parties and Congress are overwhelmingly negative.They frame the way in which citizens perceive parties and assemblies. The pub-lic symbols and messages regarding other institutions are less consistently nega-tive. This fact helps explain why parties and assemblies generally (i.e., in mostdemocracies) have the worst evaluations of all the institutions that most surveysask about.

Whereas the argument about state deficiencies helps explain why parties andassemblies are particularly under assault in the Andean countries, the secondhelps explain why these representative institutions are particularly vulnerable topoor citizen evaluations in most democracies. The incentives to decry the fail-ures of other institutions are weaker because there is less reward attached to con-vincing citizens that they are failing. Whereas the first argument focuses on theperformance of institutions and citizens’ capacity to discern institutional perfor-mance, the second one underscores that citizen evaluations of institutions areconstructed through political battles and conflicts.

State Deficiencies, Party Competition, Confidence in Representation 297

In the literature on trust in institutions, both arguments have been under-developed. With the exception of Levi (1998), the literature on institutionaltrust has neglected the state. When we turn to the Andes, such neglect is un-justifiable. In a similar vein, the literature on trust has largely neglected a con-structivist emphasis on how party competition in an era of the mass media haseroded trust in parties and legislatures.

I also examine four alternative explanations of low confidence in institutions.I do not completely discard these alternative explanations, but they are second-ary to state deficiencies and to the politicization by competing parties and themedia coverage of poor state performance. A lack of trust could be the result ofdeficiencies in the institutional mechanisms of representation. Within this broadrubric, I identify three kinds of deficiencies: truncated representation, supply-side oligopolies in the party system, and limited accountability of representativesto voters due to the electoral and nomination systems. Representation is pro-foundly truncated if major sectors of society are not formally represented—theultimate and complete form of truncation—because of some exclusion (the il-literate, for example), or if they are formally represented but through traditionalface-to-face clientelistic or personalistic relationships that do not produce an ef-fective articulation of their interests. A second institutional problem that couldaccount for a perception of a crisis of representation is a supply-side oligopolyof representation. A supply-side oligopoly means that the diversity and numberof parties is limited, so citizens do not have much breadth of choice. A thirddeficiency in the institutional mechanisms of democratic representation couldstem from the electoral system. Voters might not be able to hold representativespersonally accountable.

A fourth possibility is that the low confidence in parties and assemblies couldreflect widespread distrust at the individual, interpersonal level. According tothis perspective, trusting individuals breed trust in institutions. Several promi-nent scholars have seen political trust as an expression of underlying interper-sonal trust (Almond and Verba 1963; Fukuyama 1995; Inglehart 1997a, 1999).In this theoretical approach, individuals who do not trust other members oftheir society will not trust representative institutions.

State Deficiencies and the Crisis of Democratic Representation

The main argument in this chapter is that state deficiency in the Andes has con-tributed to the crisis of democratic representation.4 O’Donnell (2003, 34) hasusefully defined the state as “a set of institutions . . . that usually penetrates andcontrols the territory and the inhabitants that the state claims to geographicallydelimit. These institutions have as a last resort, in order to implement decisions,control (supremacy) of the means of physical coercion that some specific stateagencies normally exercise over that territory. This control supports the state’s

claim to make state decisions binding for all members of its territory” (see alsoRotberg 2004).

The state includes a vast and complex array of institutions. Of primary con-cern in this chapter are the government, the judiciary, the police, and the armedforces. These four parts of the state have been conspicuously deficient in the fiveAndean countries, though with differences from one country to the next andfrom one government to the next. The police and the judiciary have the pri-mary responsibility for ensuring citizens’ security, and they have performed thistask poorly. The judiciary has the primary responsibility for upholding citizenrights, and it has been deficient in this responsibility, notwithstanding importantinnovations such as the tutela in Colombia and the ombudsman in Peru. Finally,the armed forces have the primary responsibility for ensuring the state’s physi-cal control of the country’s territory. In Peru from roughly 1980 until 1993,when Sendero Luminoso was defeated, the guerrillas controlled a large swath ofthe national territory in the Andes. In Colombia, the guerrilla insurgency andthe paramilitary have controlled major parts of the national territory (thoughthese regions are sparsely populated) since the 1980s.

State deficiency in the Andes has been very uneven across different parts ofthe state. It is also very uneven across regions within these countries; it is morepronounced in the poorer regions (Bejarano and Pizarro Leongómez 2005; O’Donnell 1993).5 Nevertheless, at the national level, state performance, on av-erage, has been sorely deficient. The Andean states are generally weak, notwith-standing variation across countries, governments, time periods, and specific statearenas.

The five states in the Andes, and most states in Latin America as a whole, haveperformed poorly since the onset of the debt crisis in 1982. Table 10.1 providesdata on basic performance indicators in the decade from 1996 to 2005 and onpublic perceptions regarding crime and corruption.6 Four of the five countries,all but Peru, had poor performances in economic growth from 1996 to 2005;Peru’s growth was barely adequate. Venezuela’s per capita income has declinedover a long period, beginning in the late 1970s. High unemployment and under-employment afflict the economies. Governments are not solely responsible foreconomic performance, but government policy and practice are important de-terminants of economic performance. Moreover, as the literature on economicvoting makes clear, citizens hold governments responsible for the economy(Lewis-Beck 1988).

It is not only in the economic realm that states have performed poorly. Trans-parency International’s evaluations of corruption are terrible for Bolivia, Ecua-dor, and Venezuela, and substandard in Colombia and Peru. Of 158 countriesranked in the Transparency International Corruption Perceptions Index in 2005,Colombia ranked 56th, Peru 68th, Bolivia 118th, Ecuador 119th, and Venezuela136th. In a region notorious for corruption, only two Latin American countries,

298 Mainwaring

Ta

ble

10.1

Stat

e Pe

rform

ance

and

Per

cept

ions

The

reof

in th

e A

ndes

, 19

96–2

005

% w

ho a

gree

% w

ho a

gree

% c

hang

e in

per

M

ean

infla

tion

,U

rban

TI

Cor

rupt

ion

that

cri

me

has

th

at c

orru

ptio

nca

pita

inco

me,

20

00–

2005

unem

ploy

men

t,Pe

rcep

tion

s In

dex,

incr

ease

d a

has

incr

ease

d19

96–

2005

(per

cent

)20

05 (

perc

ent)

Gin

i ind

ex, 1

990s

2005

lot,

1998

a lo

t, 19

98

Bol

ivia

93

9.5

2 (1

993)

2.5

7075

Col

ombi

a2

614

.57

(199

6)4.

083

83E

cuad

or11

911

.44

(199

5)2.

590

85Pe

ru17

210

.51

(199

7)3.

574

69V

enez

uela

�4

2112

.44

–.50

(19

97)

2.3

9494

SO

UR

CE

SA

ND

NO

TE

S:

For

chan

ge in

per

cap

ita in

com

e, m

ean

infla

tion

, and

urb

an u

nem

ploy

men

t, se

e E

CLA

C 2

005.

A

ll 20

05 fi

gure

s ar

e es

tim

ates

. Fig

ures

for

urba

n un

empl

oym

ent

use

diffe

rent

met

hodo

logi

es in

diff

eren

t co

untr

ies.

Figu

res

for

urba

n un

empl

oym

ent

for

Bol

ivia

are

for

2004

.D

ata

on in

com

e di

stri

buti

on c

ome

from

htt

p://

ww

w.w

ider

.unu

.edu

/wiid

/ww

ww

iid.h

tm. R

efer

ence

s fo

r in

com

e di

stri

buti

on

data

are

ava

ilabl

e at

htt

p://

ww

w.w

ider

.unu

.edu

/wiid

/ref

s.htm

. The

dat

a on

inco

me

dist

ribu

tion

are

bas

ed o

n di

ffere

nt

met

hodo

logi

es in

diff

eren

t co

untr

ies.

Sour

ce fo

r “%

who

agr

ee t

hat

crim

e ha

s in

crea

sed

a lo

t in

the

last

five

yea

rs, 1

998”

and

“%

who

agr

ee t

hat

corr

upti

on h

as

incr

ease

d a

lot

in t

he la

st fi

ve y

ears

, 199

8”: 1

998

Lat

inob

aróm

etro

.T

he T

I C

orru

ptio

n Pe

rcep

tion

s In

dex

is a

subj

ectiv

e ra

nkin

g th

at r

ange

s fr

om 1

(m

ost

corr

upt)

to

10 (

leas

t co

rrup

t).

Sour

ce: T

rans

pare

ncy

Inte

rnat

iona

l.

Paraguay and Haiti, had worse Transparency International scores than Bolivia andEcuador. In the 1996 and 1998 Latinobarómetro surveys,7 a large majority of citi-zens in all five countries agreed that corruption had increased greatly in the lastfive years. High crime rates in urban areas adversely affect the quality of life. In1998, on average, 82 percent of respondents in the five countries agreed thatcrime had increased a lot in the last five years. States have made little or no progressin addressing poverty, inequalities, corruption, crime, and poor public education.Even the regimes that have survived as democracies have been plagued by seriousdeficiencies. Citizens expect their elected representatives to resolve these prob-lems, or at least to take steps toward doing so.

Poor state performance could affect confidence in parties and legislatures inone of two ways (Mishler and Rose 2001). One is that objective, macro-levelperformance could directly affect confidence in institutions.8 Table 10.2 exam-ines this possibility, showing the bivariate Pearson correlation coefficients be-tween confidence in the national assembly (column 1) and political parties (col-umn 2) and seven macro indicators of state performance for the twenty-threecountries included in the 1995–97 wave of the World Values Survey that had acombined Freedom House score of 8 or less.9 These seven measures of state per-formance include some of the most salient problems that have affected citizensin the Andes. The first two measures assess long-term economic performance,while the others evaluate more recent performance.

300 Mainwaring

Table 10.2

Pearson Correlation Coefficients between State Performance and Confidence in Parties and Parliaments, Country-Level Indicators

Confidence in the Confidencenational assembly in parties

(N�23) (N�23)

Per capita income, 1996 .29 .35Per capita growth, 1987–96 .14 .13Log of inflation, 1992–96 �.32 �.32Secondary school enrollment, 1996 .33 .38Unemployment, 1996 �.34 �.55Transparency International score,

1996 (N � 19) .45 .36Homicide rate �.39 �.32

SOURCES: For confidence in parties and the national assembly, see the World Values Surveys, 1995–97. For per capita income, growth rates, inflation, and unemployment, seeWorld Bank, World Development Indicators, various years. For net secondary school enrollment,see World Bank, Human Development Report 1999. For homicide rates, see WHO 2002, basedon latest available data between 1990 and 2000.

NOTE: Countries included in the Transparency International correlations are UnitedStates, Finland, Norway, Australia, Sweden, Switzerland, Poland, Spain, Germany, Japan,South Korea, Chile, Argentina, Venezuela, Philippines, Brazil, Russia, Colombia, Mexico.Countries included in all other correlations include all the above plus Estonia, Croatia, Latvia, and Bulgaria.

Although the correlations are modest, all fourteen run in the expected direc-tion. Confidence in parties and assemblies was higher where per capita incomewas higher, where the growth performance was better in the previous decade,where inflation was lower, where more students of secondary school age were en-rolled, where unemployment was lower, where the Transparency Internationalscore was higher, and where the homicide rate was lower.

The other possibility is that the subjective perception by individuals of poorstate performance—somewhat independently of objective measures—coulderode confidence in parties and legislatures. If disgruntlement with democraticrepresentation is a product of state deficiencies, this should be reflected in surveydata. The individuals with the worst perception of the state’s performance shouldexpress the least confidence in the institutions of democratic representation.

To test this possibility, and also to examine the impact of interpersonal truston confidence in institutions in a multivariate analysis, I undertook an orderedlogistic regression.10 The independent variables related to perception of stateperformance include individuals’ evaluations of the country’s present economicsituation, of their household economic situation, of whether corruption hadgotten worse in the last five years, and of whether crime had gotten worse in thelast five years.11 Another independent variable assesses the impact of interper-sonal trust on confidence in parties and legislatures.12 Four demographic vari-ables serve as control variables: education, sex, socioeconomic status, and age.The results are shown in Table 10.3 for trust in political parties and Table 10.5for trust in the National Congress.13

The results show that the individual perception of poor state performancefueled low confidence in the core institutions of democratic representation.Those with the worst opinion of national economic performance had the leastconfidence in parties. This is the only independent variable that was statisticallysignificant in all five countries.14

Table 10.4 presents simulations to show the substantive effect of changes intwo independent variables, assessment of the national economic situation andinterpersonal trust, on confidence in parties. In all five countries, especially inBolivia and Ecuador, the simulation predicts that individuals with a better as-sessment of the national economic situation are much more likely to expressconfidence in parties. In the Bolivian survey, 35 percent of respondents whoviewed the country’s economic situation as good or very good expressed someor a lot of confidence in parties. None of the seventy-seven survey respondentswho saw the country’s economic situation as very poor expressed some or a lotof confidence in parties (these are actual survey results, whereas Table 10.4 isbased on a simulation and includes several control variables).

Assessments of the change in crime and corruption also powerfully impactedconfidence in parties. One or both (in Colombia) of these variables were significant. Consistent with both countries’ abysmal ratings in Transparency


302 Mainwaring

Table 10.3

Determinants of Confidence in Political Parties in the Andes (ordered logit)

Variables Bolivia Colombia Ecuador Peru Venezuela

Country’s economic situation .325b .187c .302a .176d .130d

(.110) (.074) (.073) (.092) (.067)Household economic situation .074 .493a .121 .154 .121

(.130) (.097) (.084) (.108) (.082)Interpersonal trust .299 .431b .332c .390c .034

(.195) (.139) (.145) (.170) (.174)Crime �.109 �.296a �.054 �.180b �.437b

(.114) (.080) (.105) (.067) (.132)Corruption �.267b �.318a �.342b �.042 �.040

(.092) (.079) (.104) (.066) (.120)Sex .256d .160 �.030 �.212d .114

(.151) (.120) (.114) (.121) (.112)Age .013c �.002 .010c .010c .011c

(.006) (.005) (.004) (.005) (.004)Education .047c �.012 .003 .000 .034c

(.021) (.019) (.018) (.018) (.017)Socioeconomic status .100 .005 �.129d �.192c .088

(.105) (.074) (.075) (.076) (.075)

LR chi2 (9) 53.97 122.92 66.22 52.71 35.75Prob. � chi2 .000 .000 .000 .000 .000Pseudo R2 .037 .054 .026 .023 .014N 703 1,144 1,114 1,003 1,393

SOURCE: 1996 Latinobarómetro.NOTE: Standard error appears within parentheses.a p � .001b p � .01c p � .05d p � .10

International evaluations, in Bolivia and Ecuador, individuals who believed thatcorruption had worsened had less trust in parties. In Colombia, both variableswere significant at p � .001. In Peru and Venezuela, the assessment of changein crime rates was statistically significant, while the assessment of change in cor-ruption was not.

In Table 10.5 the dependent variable is confidence in the national assembly,and the independent variables are the same as in Table 10.3. Results are highlyconsistent with those in Table 10.3.

These results suggest that state deficiencies are at the core of the contempo-rary crisis of representation.15 If states were more effective, confidence in thecore institutions of democratic representation would improve. If this analysis iscorrect, political reforms should focus first and foremost on creating more ef-fective states and only secondarily on creating more open systems of represen-tation. The formal systems of representation in these countries are already open.The grave deficiency is in state capacity.16

The state deficiency that probably has the most direct effect on citizenconfidence in parties and legislatures is corruption. Rational citizens would befoolish to trust corrupt politicians. They might accept politicians who fail to de-liver economic goods; they might, for example, understand that this is a resultof difficult economic circumstances (Linz and Stepan 1989). But it is quite another matter to accept corruption in public officials, especially in times ofeconomic hardship.17

Perceptions of corruption had a profound impact on confidence in democraticrepresentation in Bolivia, Colombia, and Ecuador. In all five Andean countries,since the 1990s there have been huge scandals involving public officials. InVenezuela, President Carlos Andrés Pérez was impeached in 1992, and thetriggering event was a corruption scandal. In Colombia, President Samper(1994 –98) was blemished by the widely known fact that he had accepted mil-lions of dollars of campaign finance money from a drug cartel. In Ecuador, Pres-ident Mahuad (1998–2000) fell victim to a coup, in part because of reports thathe had been involved in corruption. In Peru, throughout the 1990s, widespreadrumors circulated that ex-president Alan García (1985–90) had been involved incorruption. García was forced into exile for much of the decade. His successor,Alberto Fujimori, became implicated in massive corruption scandals after resign-ing in 2000. High-ranking officials other than the president have been implicated


Table 10.4

The Impact of Assessment of the National Economic Situation and Interpersonal Trust on Confidence in Parties

(simulation based on Table 10.3)

% who express a lot or some confidence in parties:

(1) (2) (4)Economic Economic (3) You can neversituation of situation of You can be too carefulcountry is country is trust most when dealingvery good very bad people with others

Bolivia 28 10 18 13Colombia 15 8 13 9Ecuador 31 12 21 16Peru 26 15 25 18Venezuela 15 10 11 11

SOURCE: 1996 Latinobarómetro.NOTES: The simulations show the effect of allowing one independent variable (the as-

sessment of the national economic situation in columns 1 and 2, and interpersonal trust incolumns 3 and 4) to vary while all other independent variables are held constant at theirmean values. Columns 1 and 2 show the percentage of respondents predicted to express a lotor some confidence in parties if the respondents view the country’s economic situation asvery good or very bad. Columns 3 and 4 show the percentage of respondents predicted toexpress a lot or some confidence in parties if the respondents believe that you can trust mostpeople or that you can never be too careful when dealing with others.

in corruption scandals in all five countries. In the absence of more effective sanc-tions against corruption, politicians and bureaucrats engage in widespread patri-monial practices with impunity (Guevara Mann 2001). Given that the mediahave a powerful presence in forming citizen perceptions of politics, and given thesalience of reports of corruption in the media, it would be astonishing if citizensexpressed high confidence in parties and assemblies.18

Was low confidence in parties and legislatures particularly pervasive amongsome social groups? This question contains key information for assessing whatunderpins low confidence. If confidence in Congress and parties were particu-larly low among the poor and uneducated, this could reflect political disaffec-tion among these sectors. Low confidence among the poor might be a productof disenchantment with political systems that generate bad outcomes for them.If confidence were particularly low among the educated, this might reflect akeener and more informed sense of the deficiencies of parties and legislatures,parallel to what Döring (1992) called rational skepticism in his analysis of theadvanced industrial democracies.

304 Mainwaring

Table 10.5

Determinants of Confidence in Congress in the Andes(ordered logit)

Variables Bolivia Colombia Ecuador Peru Venezuela

Country’s economic situation .403a .199b .210b .495a .193b

(.110) (.071) (.071) (.090) (.064)Household economic situation .036 .180c .203c �.005 .030

(.126) (.090) (.081) (.105) (.074)Interpersonal trust .315d .336b .340c .470b �.074

(.190) (.135) (.142) (.162) (.166)Crime �.013 �.118 �.007 �.127d �.301c

(.119) (.079) (.105) (.065) (.128)Corruption �.383a �.339a �.334b �.090 �.132

(.095) (.077) (.107) (.064) (.109)Sex .168 �.031 .000 �.002 �.132

(.148) (.115) (.112) (.119) (.102)Age .019b .001 .009c �.001 .012b

(.006) (.005) (.004) (.005) (.004)Education .038d .002 .025 �.023 .052b

(.020) (.019) (.017) (.017) (.016)Socioeconomic status .104 .057 �.038 �.188c .040

(.103) (.072) (.075) (.074) (.069)

LR chi2 (9) 73.02 63.07 54.38 77.00 35.75Prob. � chi2 .000 .000 .000 .000 .000Pseudo R2 .045 .026 .020 .030 .014N 687 1,140 1,104 1,010 1,393

SOURCE: 1996 Latinobarómetro.NOTE: Standard error appears within parentheses.a p � .001b p � .01c p � .05d p � .10

In terms of sex, education, and socioeconomic status, confidence in Congressand parties is fairly even across groups in all five countries. Except for age, thesedemographic variables have an inconsistent and generally weak effect on trust inparties. In Peru, men were more likely than women to trust parties; in Bolivia,the opposite was true at p � .10. In the other three countries, the difference be-tween men and women was not statistically significant. The sex variable had noimpact in Table 10.5 (Congress).

Inglehart (1997b) and Dalton (1996) have argued that younger voters are less likely than older voters to trust representative institutions.19 They claim thata younger generation with more postmaterialist voters helps account for decreasing confidence in institutions. The opposite pattern obtains in the An-dean countries except for Colombia. In Bolivia, Ecuador, Peru, and Venezuela,older voters expressed less confidence in parties.

Education had no effect on confidence in parties in Colombia, Ecuador, andPeru. Contrary to expectations, but consistent with findings of Cleary andStokes (2006) for Argentina and Mexico, in Bolivia and Venezuela, the moreeducated expressed less confidence in parties and in Congress. In Ecuador andPeru, but not in the other three countries, those with a lower socioeconomicstatus expressed less confidence in parties; in Ecuador, however, this variable wasbarely significant (p � .09). Socioeconomic status had a statistically significanteffect in only one country (Peru) for confidence in Congress.

In societies marked by egregious inequalities, the less privileged have reason tobe more skeptical about democratic institutions, yet the effect of low education,gender, and low socioeconomic status on confidence in institutions is weak (usu-ally not significant) and inconsistent (e.g., those with higher education expressless confidence in parties in two countries, while those with higher socioeco-nomic status express more confidence in two other countries). In contrast, theassessments of state performance have a profound effect. The key to understand-ing the very low confidence in parties and legislatures has to do with state per-formance and perceptions thereof.

State Deficiency: Specifying the Concept

In the literature on the advanced industrial democracies, much has been writ-ten on confidence in institutions. When I began to reflect about the causes ofthe crisis of democratic representation in the Andes, I initially borrowed fromthis literature, in particular from authors who argued that declining confidenceis a product of poor governmental performance. This literature, however, hastwo shortcomings for understanding the crisis of democratic representation inthe Andes. The first is conceptual. It is meaningful to speak of poor govern-mental performance when some governments underperform. However, whenthe problem is chronic, and when the institutions that fail include a panoply ofstate institutions and are not limited to the government, then the notion of


“state deficiencies” becomes more appropriate. The concept of state deficien-cies properly identifies a broader range of state institutions than just the gov-ernment as responsible for the shortcomings of public institutions, policies, andpolitical outcomes.

The government is the high command of the state—the officials at the top ofthe executive branch. Many parts of the state (e.g., the judiciary) are deliber-ately constructed to have considerable autonomy with respect to the govern-ment. In the Andes, the state deficiencies extend well beyond the governmentsand include other important state agencies—most notably the justice systemand the police. In Colombia, the armed forces have also failed for an extendedtime in defeating non-state armed combatants—in particular, the guerrilla andthe paramilitary forces.

Even with the exceptional good government such as the first Sánchez deLozada government in Bolivia (1993–97), other parts of the state such as thejustice system and the police continued to have serious deficiencies. A goodgovernment can effect significant change in some policy areas, but it has limitedability to make rapid, far-reaching changes in other parts of the state such as thejustice system and the police.

The second problem with the notion of poor government performance asthat term has been used for understanding the declining confidence in institu-tions in the advanced industrial societies is that the magnitude of state deficien-cies in the Andes (and most of Latin America) is vastly worse than in the ad-vanced industrial democracies.

The boundary between poor governmental performance and state deficien-cies has heuristic and conceptual value rather than a rigorous demarcation. Twokey issues—a temporal dimension to the failure and a broader array of state institutions that fail—distinguish state deficiencies from poor government performance.

If on the one side the concept of state deficiency differs from the notion ofpoor governmental performance, on the other side, it differs from a state collapseor a failed state. The notion of state deficiencies, while signifying a problem oflonger duration and greater scope than the concept of poor government perfor-mance, indicates a less profound problem than a failed state or a state collapse inwhich the state loses its ability to govern, provide physical security for citizens,and control the country’s territory (Beissinger and Young 2002; Rotberg 2004;Zartman 1995). State deficiencies means that the state fails to provide citizenswith an important array of public goods, but it can still function. Crime ratesmay be high, but the state still controls the territory it is supposed to govern.With a failed or collapsed state, this is not the case.

The distinction between state deficiencies and state collapse is easier to de-marcate than the boundary between state deficiencies and poor government performance. A state collapse means that the state loses the capacity to control

306 Mainwaring

the country’s territory and to govern. Nowhere in the Andes in recent decadeshas there been a state collapse. A failed state in the Andean region has been par-tial and limited to specific regions: the areas in Colombia controlled by guerrillagroups and to a lesser degree (because they are less destructive of the state andcooperate more with it) those controlled by the paramilitary,20 and the areas ofPeru where Sendero Luminoso had a pervasive presence between 1980 and1993. Otherwise, stateness problems in the Andes have been far less acute thanthose that have plagued Sierra Leone, Rwanda, Congo, Liberia, Bosnia, Sudan,Afghanistan, and some other countries in the past two decades.

Democratic Representation: Citizen Views

Other evidence also suggests that the core cause of the crisis of democratic rep-resentation is state performance (in part, the results of democratic representation)rather than the process of democratic representation. When confronted with achoice about representation in survey questions, many respondents opt for sub-stantive results of the political system over process. Many even evince ambiva-lence about retaining a system of democratic representation. This ambivalenceappears in responses to the three survey questions in Table 10.6. The first


Table 10.6

Attitudes about Democracy and Representation in the Andes

(1) (2) (3)

% who agreed % who agreed that having experts,that having a not government,

% who agreed strong leader who make decisions that democracy does not bother according to what they

is always the with parliament and think is best for the best form of elections is very country is very good government good or fairly good or fairly good

Bolivia 50 NA NAColombia 46 53 71Ecuador 46 NA NAPeru 45 39 63Venezuela 67 48 69Average for other Latin

American countries 53a 44b 60b

SOURCES AND NOTES: For column 1, see 2003 Latinobarómetro. For columns 2 and 3, see 1995–97 and 1999–2001 waves of the World Values Survey. Respondents had four choices: very good, fairly good, fairlybad, and very bad. Percentages are the share of valid answers (i.e., excluding those who did not respond and did not know).

a The twelve other Latin American countries are Argentina, Brazil, Chile, Costa Rica, El Salvador,Guatemala, Honduras, Mexico, Nicaragua, Panama, Paraguay, and Uruguay.

b The seven other Latin American countries are Argentina, Brazil, Chile, the Dominican Republic, El Salvador, Mexico, and Uruguay.

column shows the percentage of respondents who agreed that “Democracy is always the best form of government.” In all of the countries except Venezu-ela, one-half or more of respondents gave one of the three alternative non-democratic answers: “In certain situations, an authoritarian government can bepreferable to a democratic one”; “To people like me, it doesn’t matter whetherwe have a democratic or non-democratic government”; or no answer in a smallpercentage of cases.21 Citizens in these countries and in most of the rest of LatinAmerica are ambivalent about representative democracy and democratic repre-sentation. Under some circumstances, they are willing to forgo it.

The other two columns in Table 10.6 are taken from the last two waves of theWorld Values Surveys, which included Colombia, Peru, and Venezuela, but notBolivia and Ecuador. Even the first question, which is worded in a way thatmight load the dice against answering favorably, elicited large numbers of posi-tive responses. Between 39 percent (Peru) and 48 percent (Venezuela) agreedthat having a strong leader who did not bother with parliament and electionswas a good idea. Of course, doing away with parliament and elections wouldentail eliminating democratic representation. A large minority of citizens waswilling to curtail democratic representation.

A sizable majority in all three countries agreed that “having experts, not gov-ernment, make decisions according to what they think is best for the country”is a very good or fairly good way of governing the country (last column). Thisresponse implies acceptance of the technocratic ideal that experts rather thandemocratically elected representatives should govern. In this response, good pol-icy results mattered more than allowing democratically elected representativesto make policy. In a similar vein, in a 1999 Venezuelan survey, 50 percent of therespondents agreed that it would be a good idea to abolish state legislatures,while only 39.5 percent disagreed with this suggestion (González de Pacheco2001, 175). These data suggest that the chief complaint for large numbers ofLatin American citizens was not a deficiency in the mechanisms of representationbut rather in the results thereof; indeed, a majority was willing to make do withless democratic representation.

One of the most stunning examples of public approval for less democratic rep-resentation occurred with President Alberto Fujimori’s decision to close the Pe-ruvian Congress in April 1992. In the immediate aftermath of this coup, 88 per-cent of Peruvians approved of Fujimori’s decision to close Congress. Fujimori’spublic opinion approval ratings shot up to 79 percent shortly after his coup.22

Beyond State Deficiencies: The Effects of Party Competition and Television

In this section, I engage in more speculative analysis of why parties and Congresshave especially low levels of confidence. The answer is not immediately obvious.Because of state deficiencies, these institutions have been very tarnished. In a

308 Mainwaring

context of poor state performance, it would be surprising if citizens retained ahigh level of confidence in parties and Congress. But as Table 10.7 shows, otherstate institutions enjoy higher levels of confidence than parties (which are not, ofcourse, part of the state) and the national assembly. If the core problem is statedeficiency, why do some state institutions enjoy greater confidence than partiesand the national assembly? It is doubtful that the performance of parties andCongress is worse than that of the judiciary, the police, or some other state in-stitutions that enjoy higher levels of confidence.

Table 10.7 provides data on four state institutions: the judiciary, the police, themilitary, and the presidency. The core institutions of democratic representation,political parties and the National Congress, stand out as the least trusted institu-tions in the Andean region. In all five countries, parties and the Congress hadamong the lowest levels of confidence in the 1996 and 1998 Latinobarómetro sur-veys. Moreover, even compared to the rest of Latin America, the level ofconfidence in parties and Congress was distinctively low in the Andes.23

Confidence in the judiciary was generally greater than confidence in partiesand Congress. This fact is noteworthy because most Latin American judicial sys-tems are sorely deficient (Brinks 2003, 2004; Dodson and Jackson 2003;O’Donnell 1999; Pásara 2002). The judiciary is characterized by slowness, by un-evenness in the application of justice, and by an inability to prosecute criminals.

Confidence in the police was almost uniformly higher than confidence inparties or Congress. Yet inefficacy, violence, and corruption have plagued thepolice in most Latin American countries (Brinks 2003, 2004; Chevigny 1995,2003). The police in almost all of Latin America are notorious for violence and


Table 10.7

Citizen Confidence in Institutions in the Andes

% of respondents who express some or a lot of trust in the following institutions:

political the national the the the the press the the parties assembly police judiciary military (1996) Church presidency

Bolivia 20 28 27 27 34 63 85 36Colombia 17 23 32 29 40 44 80 35Ecuador 14 18 27 22 63 55 73 35Peru 17 19 25 17 38 42 88 25Venezuela 15 20 23 29 59 55 77 37Andean

average 17 22 27 25 47 52 81 34Average for

twelve other countriesa 24 30 35 35 34 49 78 40

SOURCES: See 1996 Latinobarómetro, q33, for data on the press (not included in the 1998 survey), and 1998 Latinobarómetro, q38, for all other data.

a The twelve other Latin American countries are Argentina, Brazil, Chile, Costa Rica, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, Panama, Paraguay, and Uruguay.

for using torture to extract confessions, and in many countries, for assassinatingsuspected criminals. Given the violence that has plagued the Andean countries,and the inefficacy, violence, and corruption of the police,24 it is remarkable thatthe police enjoy higher confidence than political parties do.

Across the Andes, confidence in the armed forces was higher than confidencein parties and Congress. Note particularly the confidence the military enjoyed inEcuador and Venezuela. This confidence in the armed forces almost surely helpsexplain that golpista military leaders—those who previously led coups—weredemocratically elected as president in Venezuela in 1998, in Ecuador in 2002,and to a lesser degree (because of the lower level of confidence in the armedforces) in Bolivia in 1997.

It is doubtful that the performance of the judiciary and police is better thanthat of parties and legislatures. The particularly low level of confidence in par-ties and Congress must reflect more than objective performance differences.

Citizen confidence in institutions is subjective and is not exclusively based ontheir objective performance.25 Rather, a combination of constructivist and ra-tionalist factors are at work.26 Politics (including the perceived trustworthinessof political institutions) is not only about objective realities, but also about theway political actors construct interpretations of those objective realities. Thediscourses that actors use to judge institutions and to frame politics affect citizentrust in institutions. Citizen assessments of political institutions are the result ofpolitical processes and battles. They cannot be taken strictly as an objective in-dicator of institutional performance.

Two factors related to the way in which political processes and conflicts shape citizen assessments of institutions help account for the particularly lowpublic trust in parties and Congress. First, parties politicize the failures of theircompetitors. Opposition parties may obstruct government policy because a fail-ure of the governing party or parties enhances the opposition’s chance of be-coming the government in the next elections. In some countries of the region(Bolivia and Ecuador, e.g.), the common impression is that opposition partiesfoster and thrive on the failure of government parties.

One of the ways for a new party to make inroads into the electoral market isto claim that all existing parties have failed. In this context, the greater diversityand number of competing parties in recent years meant more actors with avested interest in calling attention to the shortcomings of competing politicalparties. The incentives of the game foster persistent criticism of competing par-ties. For example, opposition political candidates typically exaggerate the in-crease in crime in most of Latin America, leading to a situation in which citi-zens believe that crime is even worse than it actually is (Basombrío 2003;Smulovitz 2003). The self-interested behavior of competing political parties isthe rationalist side of this argument; the fact that citizens develop their own in-terpretations of the trustworthiness of institutions is the constructivist part.

310 Mainwaring

No state institutions are subject to this kind of competition. The media andpoliticians may criticize the police and judiciary, but there is far less institution-alized competition among the police and judiciary than among political parties.Rivalry may exist among different police forces—for example, between theFederal Police and the statewide military police in Brazil—but the success ofone police force does not depend on the failure of another. In contrast, in onepowerful sense, competition among political parties is zero sum: the electoralgains of one party necessarily come at the expense of losses for another. More-over, no new police force is likely to gain by criticizing the performance of the existing police. In contrast, a new political party must criticize existing par-ties or else it will not be able to gain an electoral toehold. Whereas among com-peting parties enduring organizational solidarity is rare, among different policeforces the norm is organizational solidarity even when institutional performanceis poor (Brinks 2004). The entire institution is likely to suffer or to benefitjointly. Widespread complicity within the police and judiciary is the counter-point to the pervasive competition and criticism among parties.

The public’s proclivity to blame politicians and spare the police was apparentin an online survey conducted by one of Argentina’s oldest and most prestigiousnewspapers, La Nación. In June 2004, Argentine president Kirchner stated thata recent crime wave in greater Buenos Aires was the product of disgruntled former police officers who had been purged from the force. The survey (whichwas not random) asked respondents what the main cause of the crime wave was. Notwithstanding Kirchner’s statement and despite many well-publicizedepisodes of police involvement in criminal activities, far more respondents (49.7 percent) blamed public policy in citizen security than any other factor.Only 5.5 percent blamed police inefficiency; 16.5 percent of the respondentsbelieved that the primary cause was the socioeconomic crisis; and 15.7 percentblamed sectors of the police reacting to the purges within the police.27 In sum,far more people blamed politicians and public policy than the police for thecrime wave. Although this evidence comes from Argentina rather than an An-dean country, a similar phenomenon occurs in the Andes.

Changes in the supply side of the electoral market paradoxically helped fuelthe growing sense of a deficit of representation. The growing diversity and num-ber of parties meant that more parties could politicize failures of their competi-tors. Competing parties criticized the status quo from a broader range of per-spectives.28 In a context of widespread poverty, citizen insecurity, and inequality,it is easy for opposition parties and the media to politicize the shortcomings ofgovernment, and hence the failures of democratic representation to generatepublic goods and good policy.

Television has contributed to the cynical evaluations of Congress and parties.Television and radio are the primary vehicles by which competing parties politi-cize the failures of their opponents. They are also the primary vehicles by which


most people get their political information.29 Television has paid far more attention to scandals, corruption, and the shortcomings of politicians than totheir virtues. When citizens are bombarded with stories of political scandals and corruption, it is no wonder that confidence in politicians, parties, and Congressplummets.30 Several scholars have found that television has contributed to de-clining trust in institutions or political cynicism in the advanced industrial de-mocracies.31 Television and radio have likewise reinforced cynical attitudes re-garding parties and Congress in the Andean region.

Why has television regularly presented negative images of parties and legisla-tures? Research on the United States suggests an answer that probably holds forthe Andean countries: competitive market pressures. Stories about scandal,conflict, and corruption sell; stories about the quotidian processes of assembliesand parties do not. Ironically, freedom of the press, which is one of the mostnormatively valued aspects of democracy, may facilitate the denigration of dem-ocratic representation. In addition, competing parties deliberately politicize thefailures of their political adversaries, and the media report such criticisms.

The Expansion of Citizenship and Democratic Representation in the Andes

In the next four sections, I look at four alternative explanations of low confidencein democratic institutions. Some scholars who work on Latin America have ar-gued that the crisis of representation is a product of the exclusion of some citi-zens or the inadequacy of the mechanisms of representation. For example, Lynch(2000, 93) argued that “In Peru, in the last 30 years, there has been an importantgrowth in the number of citizens who cannot express themselves politically in ademocratic fashion and who are not represented by the existing state institutions.This difficulty of political expression and institutional representation is the fun-damental cause of the frustration with the democratic regime.”32

The argument in this concluding chapter is different and is consistent withthe claims of René Antonio Mayorga and Martín Tanaka in this volume. I showthat there has been a dramatic expansion of democratic representation in all five countries in the post-1978 period. I examine four forms of the expansionof citizenship and representation: the massive incorporation of new citizens,qualitative changes in citizenship, new opportunities for representation madepossible by the introduction of direct elections for governors and mayors and byother political reforms, and new opportunities for representation of indigenouspeoples.

Until the late 1970s, four of the five Andean countries (all but Venezuela) 33

suffered from profound institutional deficiencies in democratic representa-tion. Truncated representation and supply-side oligopolies were serious prob-lems. However, since 1978, especially in Bolivia, Ecuador, and Peru, democratic

312 Mainwaring

representation has expanded. Truncated citizenship is less widespread than in thepast, and the number and diversity of competing parties has with few exceptionsbeen substantial. In this respect, it is paradoxical that a crisis of representationhas roiled the Andean countries since the 1980s.

Truncated representation could create disaffection with representative insti-tutions because the excluded or those who are formally included but withoutan effective voice identify and react against their exclusion.34 If sectors of soci-ety are disenfranchised, they do not enjoy any democratic representation, muchless a truncated form, and in such circumstances the system-wide representationis truncated. Representation can be more or less truncated; it is profoundly trun-cated when representatives exercise personalistic control over the represented.

Truncated representation became less widespread in the post-1978 period ascitizenship expanded.35 A first expression of expanding citizenship is that dem-ocratic representation was restored in Ecuador in 1978 after eight years of mil-itary rule; in Peru in 1978 (a constituent assembly) after a decade of militaryrule; and in Bolivia in 1982 after eighteen years of military dictatorship, punc-tuated by a few months of civilian government in 1979 and 1980. The dicta-torships in all three countries banned elections at all levels of government. Theelection of assemblies and executives restored democratic representation.

The post-1978 regimes broadened citizenship relative to the past. The en-franchisement of the illiterate in Peru in 1978 and in Ecuador in 1979 led to amassive increase in the electorate. In Ecuador, the ratio of votes cast to totalpopulation jumped from 12.1 percent in the congressional election of 1958 to41.7 percent in 2002. In Peru, 14.9 percent of the population voted in the 1956presidential election; 45.3 percent voted in the 2001 presidential election. Be-cause these decades witnessed rapid population growth, these increases in per-centages meant that massive numbers of new citizens entered the ranks of therepresented. In Bolivia, the increase was also pronounced, from 27.4 percent ofthe total population in the presidential election of 1960 to 35.4 percent in 2002(see Figure 10.1).

Accompanying this numerical expansion in the ranks of the represented wereprofound qualitative changes that reflected dramatic increases in urbanizationand education, and the growth of the mass media. Table 10.8 shows the changesbetween 1960 and the 1990s in urbanization and education. The percentage ofthose living in urban areas increased sharply in all five countries. Urbanizationtransformed the kind of citizen who was represented. In the 1950s, Bolivia,Colombia, Ecuador, and Peru had rural majorities. Large numbers of poor, ru-ral residents faced a limited choice in the electoral market. The electoral mar-ket in rural regions was dominated by oligopolistic clientelistic exchanges, withlimited real competition because local patrons controlled the peasant vote(Archer 1995; López Jiménez 1997). In the urban setting of the 1990s, the per-sonalistic domination that characterized the countryside four decades earlier was


not possible. The share of voters subjected to direct personalistic domination fellsharply (López Jiménez 1997, 268–92).

Education levels rose dramatically in all five countries from the 1950s until the1990s. Secondary-school enrollment increased at a stunning rate in Colombia,Peru, and Ecuador. Even though the increase was less dramatic in Bolivia andVenezuela, the percentage nevertheless tripled and doubled, respectively.36

Higher levels of education changed the relationship between the represented andthe representatives. The represented now had access to more information aboutthe electoral market. The electoral influence of a younger, more educated gen-eration gradually rose.

Television also changed the relationship between voters and their representa-

314 Mainwaring

Figure 10.1Votes cast in presidential elections as a percentage of total population,

Andean countries, 1950s to 2004

Perc

enta

ge

Presidential Elections1970 19901960 1980 2000

Ecuador�Bolivia


60

50

40

30

20

10

0

Sources: International Institute for Democracy and Electoral Assistance (IDEA), http://www.idea.int; Oficina Nacional de Procesos Electorales (ONPE), http://www.onpe.gob.pe; InternationalFoundation for Election Systems, http://www.ifes.org; Latinamerica Press, http://www.latinamericapress.org, based on UNICEF 2001 and INEI 2002 reports; Nohlen (1993); ConsejoNacional Electoral (CNE); International Foundation for Election Systems, http://www.ifes.org/eguide/turnout2002.htm; World Factbook 2002, http://www.odci.gov/cia/publications/factbook/.

tives. In conjunction with urbanization, television meant a shift in campaignstyles, particularly for the presidency. If previously local politicians and party or-ganizations campaigned in a local fashion, by the 1990s, especially for the presi-dency, television campaigns were of paramount importance. Rather than relyingheavily on the party machinery, candidates for national office now transmit theirmessages and seek votes through the mass media.

A third form of expansion of democratic representation in the post-1978 pe-riod has been spatial—through political decentralization. Democratic represen-tation refers to not only the National Congress and president, but also to thosewho are elected at the local and intermediate level.

Prior to the 1980s, all five countries were governed in a centralized way, andmechanisms of political representation were concentrated at the national level.In all five countries, governors and most mayors were appointed rather thanelected before the 1980s. Bolivia, Colombia, and Venezuela undertook far-reaching decentralizing innovations in the 1980s and 1990s to enhance demo-cratic representation (see Kathleen O’Neill’s chapter in this volume).

In Bolivia, mayors of most cities were appointed until 1995, when direct elec-tions for mayors were introduced. In 1994, the Sánchez de Lozada governmentundertook a wide-ranging decentralization initiative known as the Law of Pop-ular Participation (Grindle 2000; Van Cott 2000b, 149–79). This law estab-lished the direct election of mayors and municipal councils, brought about fiscaldecentralization, and gave local communities greater control over social policy(Mayorga 2002a, 204 –7). In Peru, governors were appointed until 1990, whendirect popular elections were introduced. In Colombia, the first direct elections


Table 10.8

Secondary Education and Urbanization in the Andes

Percentages

SecondaryYear education Urbanization

Bolivia 1960 12.0% 39.3%1990-96 37.0 (1990) 61.4 (1996)

Colombia 1960 12.0 48.21996 66.7 73.1

Ecuador 1960 12.0 34.41994-96 50.0 (1994) 59.6 (1996)

Peru 1960 15.0 46.31997-96 72.5 (1997) 71.3 (1996)

Venezuela 1960 21.0 61.21996 39.5 86.1

SOURCE: World Bank (1998).NOTES: Secondary education is measured as total enrollment, regardless of age, as a percent-

age of the population of the age group that officially corresponds to the secondary level of ed-ucation. Urbanization is measured as the urban population as a percentage of the total population.

for mayors were held in 1988, followed by the first direct elections for gover-nors in 1991. The direct election of mayors and governors changed the systemof representation and transformed clientele networks (Pizarro Leongómez 2002,373). In Venezuela, the first direct elections for both mayors and governors wereheld in 1989 (González de Pacheco 2001). As O’Neill argues in this volume,these new instances of representation have altered political dynamics, especiallyin Bolivia, Colombia, and Venezuela.

Democratic representation expanded in a fourth way in the 1980s and 1990s.The indigenous peoples, who were historically marginalized both socially andpolitically in these five countries, have been formally incorporated into the po-litical system in recent decades and won special recognition and rights, especiallyin Bolivia (see Yashar’s chapter in this volume; Albó 1994). In Bolivia, the Lawof Popular Participation fostered an expansion of the indigenous movement(Van Cott 2000a, 2000b). It formally recognized the legal status of indigenouscommunities and the legitimacy of indigenous authority. It has facilitated thefuller incorporation and representation of the indigenous majority in Bolivianpolitics. It brought about a sharp increase in the number of elected indigenousofficials (Van Cott 2000b, 170). In 2002, a new leftist indigenous party, MAS(Movimiento al Socialismo, or Movement toward Socialism), headed by cocagrower Evo Morales, won 21 percent of the seats in the Chamber of Deputies.The party’s presidential candidate, Morales made it to the runoff round in 2002,and won in a landslide victory in 2005.

In Ecuador, the creation of the Commission of Indigenous Affairs during theBorja government (1988–92) marked the beginning of a gradual process of ex-pansion of indigenous representation. In the 1990s, the indigenous movementblossomed into a major social actor (Barrera Guarderas 2001). A new indige-nous party, Pachakutik, was created in 1995, and it gained nearly 10 percent ofthe congressional seats in 1996. In 2002, Pachakutik made new electoral inroadsand supported the winning presidential candidate, Lucio Gutiérrez. Even inColombia, where the indigenous represent a meager 1.3 percent of the totalpopulation, in the 1991 Constitution they nevertheless won some special rightsand have been afforded special mechanisms of representation (Peñaranda 2002).In a similar vein, the Venezuelan Constitution of 1999 contains a whole sectionon indigenous rights.

These sweeping transformations in the nature of citizenship and democraticrepresentation do not imply that citizenship has been extended evenly to all in-dividuals. The exercise of rights in the contemporary Andes is still uneven acrossindividuals of different classes, racial backgrounds, and sexes (López Jiménez1997, 379– 471; O’Donnell 1993, 1999; Van Cott 2000a, 2000b).37 Yet thesepersistent inequalities in the exercise of citizenship coexist with huge advancesin the extension and the exercise of citizenship (Mayorga 2002a; Tanaka 1998).

316 Mainwaring

If frustration with truncated representation were primarily responsible for thecrisis of representation, then the individuals most vulnerable to such truncationshould express the least confidence in parties and legislatures. The poor andpoorly educated are those most subjected to truncated representation.38 As shownabove in Tables 10.3 and 10.5, however, education and socioeconomic statushave a weak impact on confidence in parties and legislatures in the Andes. More-over, in the cases where education has a statistically significant impact, the less ed-ucated express more confidence in parties and Congress.

The contemporary crisis of representation is therefore not primarily a result oftruncated representation. Those who are subjected to truncated representationare little or no more likely than other individuals to express low confidence inparties and legislatures. The lack of confidence extends far beyond the poor, thepoorly educated, women, and other marginalized groups. Representation con-tinues to be unequal across classes, sexes, and ethnic groups, but today the issueis quality of access, not the simple exclusion of groups.

Changes in Democratic Representation: Growing Number and Diversity of Partisan Options

The previous section addressed the demand or consumer side of democraticrepresentation—that is, changes in the size and composition of the citizenshipthat is entitled to democratic representation. In this section, I address changes inthe supply side of democratic representation—in the number and diversity ofparty options available to citizens. An oligopolistic offer of different partisan al-ternatives could in principle be responsible for a crisis of democratic representa-tion. At the extreme end of a supply-side monopoly, under many authoritarianregimes, there are no competitive elections and no democratic representation.Even under democracies and semi-democracies, supply-side oligopolies some-times limit the range of options. In the Andes, this may have been the case in thepast, but it is not the problem today.

The options for democratic representation come mainly from the party sys-tem. Independent candidates may run for and win office, but they cannot sub-stitute for political parties. Parties enable citizens to identify what broad policypackages or ideological alternatives are being offered (Downs 1957; Hinich andMunger 1994). As such, they remain the fundamental mechanism of democraticrepresentation. For this reason, my discussion of the supply (offer) side of dem-ocratic representation focuses on the party systems. (For more details on theparty systems of the Andean countries, see the chapters by René Antonio Mayorga, Eduardo Pizarro Leongómez, Simón Pachano, and Martín Tanaka).

The discussion will focus on two dimensions of party systems, both of whichaffect the diversity of party options: the number of parties and their ideological


diversity. With fewer parties and with less diversity among them, the party sys-tem is more oligopolistic. With more parties and greater diversity among them,the electoral market is more open, and citizens have a wider range of options.

A supply-side oligopoly is identifiable by the number and diversity of partiesin the party system, not by citizen sentiment that parties are all the same. It is anobjective rather than a subjective phenomenon. In contexts of crises of repre-sentation, regardless of whether there is a supply-side oligopoly, citizens mayview parties that are quite diverse ideologically and programmatically as beingessentially the same.39

By the 1980s, in Bolivia, Ecuador, and Peru, citizens had a very diverse rangeof options for political representation. In this sense, the diversity of party optionsexpanded relative to earlier periods in all three countries. Venezuela had a nar-rower range of options until the 1993 elections, when both the number and the

318 Mainwaring

Figure 10.2Effective number of parties, lower chambers, in the Andean countries

EN

P

Elections1970 19901960 1980 2000

Ecuador�Bolivia


10

8

6

4

2

0

Sources: Nohlen (1993); Political Database of the Americas, http://www.georgetown.edu/pdba;Payne, Zovatto, Carrillo Flórez, and Allamand Zavala (2002); Colombian Senate website, http://www.senado.gov.col; Corte Nacional Electoral (Bolivia), http://www.cne.org.bo; Oficina Na-cional de Procesos Electorales (Peru), http://www.onpe.gob.pe; Registraduría Nacional del EstadoCivil, Estadísticas Electorales, 1990; Registraduría Nacional del Estado Civil. Estadísticas Electorales,1992; Registraduría Nacional del Estado Civil, Estadísticas Electorales, 1994; Registraduría Nacionaldel Estado Civil website (Colombia).

diversity of options grew. Colombia had a smaller number and narrower rangeof parties until the 1990s, when a duopoly eased as partisan options expanded.

Figures 10.2 and 10.3 show the changes over time in the effective number ofparties (ENP) in these five countries,40 demonstrating the growing fragmenta-tion of the party systems in the 1980s and 1990s. In four of the five countries, allbut Ecuador, the effective number of parties has reached a historical high sincethe mid-1990s. Since the inception of their current democratic regimes, Boliviaand Ecuador have had fragmented party systems, with a large supply of partisanoptions available to voters. Peru’s party system became highly fragmented in1990, as did Venezuela’s in 1993. Colombia and Venezuela were the only coun-tries of these five that did not have fragmented party systems by 1990. Colombia


Figure 10.3Effective number of parties in the Senate: Peru, Venezuela, Bolivia,

and Colombia

EN

P

Elections1985 19951980 1990 2000

Bolivia�Colombia

�Peru�Venezuela

8

7

6

5

4

3

2

1

0

Sources: Nohlen (1993); Political Database of the Americas, http://www.georgetown.edu/pdba;Payne, Zovatto, Carrillo Flórez, and Allamand Zavala (2002); Colombian Senate website, http://www.senado.gov.col; Corte Nacional Electoral (Bolivia), http://www.cne.org.bo; Oficina Na-cional de Procesos Electorales (Peru), http://www.onpe.gob.pe; Registraduría Nacional del EstadoCivil, Estadísticas Electorales, 1990; Registraduría Nacional del Estado Civil, Estadísticas Electorales,1992; Registraduría Nacional del Estado Civil, Estadísticas Electorales, 1994; Registraduría Nacionaldel Estado Civil website (Colombia).

joined the ranks of fragmented systems in 1998, with 3.7 effective parties (invotes) for the Chamber of Deputies and 3.9 for the Senate, followed by a sharpincrease to 5.4 in the lower chamber and 6.2 for the Senate in 2002. The data onthe number of parties demonstrate that the crisis of representation is not a prod-uct of an oligopolistic electoral market with few party options.

The ideological/programmatic diversity of representation also grew consid-erably relative to earlier decades, especially in Colombia and Venezuela. Untilits near demise in 2002, the ADN (Acción Democrática y Nacionalista, or Na-tionalist Democratic Action) represented the right pole of Bolivia’s post-1978party; the MIR (Movimiento de Izquierda Revolucionaria, or RevolutionaryLeft Movement) the left pole in 1978–85 and thereafter the center or center-left; and the MNR (Movimiento Nacionalista Revolucionario, or NationalistRevolutionary Movement) the center-left in 1978–85 and thereafter the center(Mayorga 2002a). Citizens could also opt for an indigenous party or for person-alistic vehicles such as CONDEPA (Conciencia de Patria, or Conscience of theFatherland) and the UCS (Unidad Cívica Solidaridad, or Civic SolidarityUnity). Between 1989 and 2002, the ideological diversity in the party system di-minished considerably, but in 2002, two new indigenous parties, the MAS andthe MIP, became major electoral contenders. By virtue of their positions as anti-system parties, the emergence of the MAS and the MIP (Movimiento IndígenaPachakuti, or Pachakuti Indigenous Movement) created a polarized party systemwith widely divergent ideological positions (Mayorga 2002b). Prior to 1978,Bolivia had never experienced full democracy, and it had never had anywherenear the range of party options that started to sprout that year.

In Colombia, the traditional party system that prevailed from the late nine-teenth century until 1991 was oligopolistic; it offered a limited number and di-versity of partisan options (Archer 1995; Hartlyn 1988; Rodríguez-Raga 2002).During this lengthy period, only the two traditional parties, the Conservativesand Liberals, were serious contenders for power.

From 1958 to 1974, under the National Front coalition that formally institu-tionalized curbs on electoral competition, Colombia was a quintessential ex-ample of representation limited by a supply-side oligopoly. Even after the disso-lution of the National Front, the range of options in Colombia’s electoral marketwas narrow. Potential options on the left were violently eliminated by repres-sion. Between 1958 and the late 1980s, radical opposition forces, rather thanworking within the party system, opted out and joined leftist guerrilla forces.From 1958 until 1991 no third party ever won more than 6 percent of the seatsin the Senate and 7 percent in the Chamber of Deputies. The two traditionalparties, which together almost always won at least 90 percent of the seats in bothchambers, had only modest ideological/programmatic differences after 1958.

The party system became more diverse after 1991, and the supply of party op-tions increased (Bejarano and Pizarro Leongómez 2005; Pizarro Leongómez

320 Mainwaring

2002). The leftist M19 won nineteen of the seventy seats in the 1991 Constitu-tional Assembly, marking a breakthrough for third parties. The ConstitutionalAssembly approved a new electoral system designed to make it easier for newparties to win representation in the Senate (Botero 1998; Rodríguez-Raga 2002;Ungar and Ruíz 1998). The Conservative Party suffered a serious electoral ero-sion in the 1990s (Pachón 2002); in 2002, it did not field a presidential can-didate for the first time since 1942. No third party has stepped up as a majorelectoral contender. Instead, an “atomization of the party system” (PizarroLeongómez 2002) has occurred, with a proliferation of very small parties and in-dependent candidates (Archer and Shugart 1997; Ungar and Ruíz 1998). The ab-sence of a major third party gives citizens fewer major choices at the nationallevel than in the other four Andean countries. At the local level in the majorcities, however, citizens have a wide choice of options among the two traditionalparties, minor third parties, and independent candidates (Querubín, Sánchez,and Kure 1998).

In Ecuador, the system has offered a wide number and range of parties in theentire post-1978 period. The PSC (Partido Social Cristiano, or Social ChristianParty) has anchored the conservative pole; the Izquierda Democrática (Demo-cratic Left), the center-left. Many populist, conservative, and center-left andleftist parties have also competed (Conaghan 1995; Friedenberg 2001; MejíaAcosta 2002). In 1996, the largest indigenous party, Pachakutik, also emerged asa relevant electoral player.

Before 1978, the main parties in Peru were personalistic vehicles such asUNO (Unión Nacional Odriísta, or National Odriísta Union), the party createdby Manuel Odría, who was president from 1948 until 1956; the centrist AcciónPopular (AP, or Popular Action), created in 1956; and the APRA (Alianza Pop-ular Revolucionaria Americana, or American Popular Revolutionary Alliance).The APRA was originally a left populist party when it was created in 1924. Bythe late 1950s, in order to avoid frightening the elites who had consistently re-pressed or proscribed APRA and to secure a place in the political system, theparty moved to the center (Cotler 1995; García Montero and Friedenberg 2001;Graham 1992, 23–36). Although AP and APRA had different social bases, inthe 1960s they did not diverge sharply along programmatic lines.

In contrast, from 1983, when Alfonso Barrantes was elected mayor of Limaon the Izquierda Unida ticket, until 1992, programmatic and ideological differ-ences in the party system were huge. The Izquierda Unida coalition anchoredthe left pole of the system from 1980 until its near collapse in 1990. Many of itsleaders favored revolutionary socialism (Lynch 1999, 199–220). AP and thecenter-right Partido Popular Cristiano (Popular Christian Party), a ChristianDemocratic party (Schmidt 2003), anchored the other pole, with APRA in thecenter-left during the 1980s. The party system collapsed between 1992 and 1995(Lynch 1999; Tanaka 1998; Tanaka, this volume). During the heyday of the


Izquierda Unida, 1983–90, voters could choose among a remarkably diverse setof ideological options. From 2001 until 2006, ideological breadth was more lim-ited in Peru than in Bolivia, Ecuador, and Venezuela.

In Venezuela, party competition between 1973 and 1988 revolved mainlyaround two hierarchical, centralized, and disciplined centrist parties, AcciónDemocrática (AD, or Democratic Action) and COPEI (Comité de Organi-zación Política Electoral Independiente, or Committee of Independent PoliticalElectoral Organization). These fifteen years were the apogee of what Coppedge(1994) called a partyarchy—a democracy dominated by centralized, disciplined,and hierarchical political parties. The partyarchy began to erode in 1988, andthe dominance of AD and COPEI was shattered in 1993. The party system ex-perienced a major change in 1993 with the presidential election of RafaelCaldera, the former COPEI leader who bolted from the party when he failedto win the presidential nomination (Crisp, Levine, and Molina 2003). Calderawas the first candidate from outside the AD and the COPEI to ever win a fairpresidential election, and the two traditional parties’ share of the legislative votealso declined that year. The old party system collapsed after the election of HugoChávez in 1998 (Penfold Becerra 2001; Tanaka, this volume).

In short, the crisis of democratic representation reflects neither primarily trun-cated representation nor a supply-side oligopoly. The primary problem is not aninstitutional blockage that impedes representation. Indeed, the proliferation ofdifferent partisan alternatives itself reflects a crisis of representation in the Andes.The growing number and diversity of party options stem from deep dissatisfac-tion with existing parties. This proliferation, however, has not brought about anemergence of well-institutionalized parties.

The large number and diversity of party options in all of these countries atsome times in recent history raises critical questions: Why, despite a wide arrayof party options, have none of the parties seemingly been able to resolve thepressing needs of the majority of citizens? Why do citizens look for options outside the party system when the options inside the system are so diverse? Why is there a crisis of democratic representation when the diversity of optionsis so great? The answer is the combination of state deficiencies and the politi-cization by parties and the media of those state deficiencies. These factors haveled large numbers of citizens to reject the established parties and seek somethingdifferent.

Electoral Systems, Accountability, and Democratic Representation

A different kind of institutional deficiency could stem from limited personal ac-countability of representatives to voters and oligarchic parties and limited inter-nal party democracy. Electoral accountability depends on many factors. I focuson two institutional factors, the electoral and nomination systems.

322 Mainwaring

Electoral systems and nomination systems establish widely different relation-ships between citizens and their elected representatives. Systems range fromparty-oriented to personal-oriented in their incentive structures for politicians(Carey and Shugart 1995; Crisp, this volume; Moreno, Crisp, and Shugart 2003;Shugart 2001), with corresponding differences in the nature of accountabilityand representation. In some competitive political regimes, voters choose a partyrather than specific candidates for the National Congress, and the parties some-times also control the nominations of candidates. Such systems are in prin-ciple better at fostering programmatic voting (Crisp, this volume), but they areweaker in terms of direct accountability of representatives to voters. These sys-tems are especially prone to oligarchic and hierarchical parties. They might bevitiated by not giving citizens enough personal voice over who represents them,and hence by establishing weak linkages of accountability between individualrepresentatives and citizens. Under these systems, the direct accountability of in-dividual representatives to voters is extremely limited and parties are very hier-archical. This is a key feature of what Coppedge (1994) called partyarchy. Inprinciple, widespread dissatisfaction with representation could result from suchlimited direct accountability of representatives to voters and from hierarchicalparties.

At the other extreme, citizens have considerable direct voice in choosing theirrepresentatives. These systems establish stronger personal accountability of rep-resentatives to their constituents.41 In such systems, voters cast their ballot for aspecific candidate rather than for a party, and if they become disgruntled withthis candidate they can choose a different one in the next election.

In most of the Andean countries, it is highly unlikely that the crux of the con-temporary crisis of democratic representation is limited personal accountabilityof representatives to voters and excessively centralized parties. The kinds of elec-toral systems used to elect the legislatures (except in Colombia) and presidentsin the Andes are common in other parts of the world. In many of these othercountries, either there is no subjective perception of a crisis of representation orelse it is a much weaker perception than exists in the Andes. It is difficult to ex-plain the distinctive crisis of representation in the Andes on the basis of electoralsystems that are common in other parts of the world.

Three Andean countries (Bolivia, Ecuador, and Venezuela) underwent im-portant institutional reforms in recent years in order to enhance the direct ac-countability of representatives to voters and to enhance representation. Thesewide-reaching reforms, however, did not ameliorate the widespread disgruntle-ment with representation.

These three countries undertook important reforms that moved parties fromhyper-centralized toward the center of Brian Crisp’s intraparty dimension (thisvolume, Figure 7.1). In Venezuela, from 1958 until 1993, the personal ac-countability of all individual members of Congress to voters was nonexistent.


This situation changed in 1993, when the electoral system was changed to amixed-proportional format for the Chamber of Deputies. With this mixed-proportional format, voters cast two ballots for the lower chamber: one for aparty list and the other for representatives in single-member districts. In thesingle-member districts, elected representatives are directly accountable to vot-ers (Crisp and Rey 2001).

In Bolivia prior to 1994, the electoral system fostered disciplined, hierarchi-cal, and centralized parties (Gamarra and Malloy 1995, 419; Mayorga 2002b),much as in Venezuela. This system allowed for no direct accountability of mem-bers of Congress to voters. In 1994, Bolivia adopted a mixed-proportionalsystem that gave voters more direct control over slightly more than half of themembers of the Chamber of Deputies (Mayorga 2001a; 2001b; 2002a, 182–86).Sixty-eight members of the Chamber of Deputies are elected in single-memberdistricts with a plurality vote, and sixty-two are chosen in nine department-widedistricts with a proportional system with closed lists. The single-member districtsentail a shift away from hyper-centralization in Crisp’s intraparty dimension.42

Some parties in Bolivia introduced primary elections, which also shifted themaway from the hyper-centralized pole.

In Ecuador, too, politicians embraced wide-reaching reforms of the electoralsystem to enhance democratic representation. In 1994, a reform eliminated themonopoly of representation enjoyed by political parties and allowed movementsand independents to run for office. In 1997, the Congress eliminated the previ-ous system of closed lists and instead implemented a system based on preferencevoting for individual deputies. With this new system, voters cast their ballots forspecific candidates. Whatever problems this new system had—and they wereconsiderable (Pachano 1998)—it allowed citizens to cast a personal vote fordeputies. The introduction of a rule allowing reelection also fostered greater per-sonal accountability of representatives to voters (Mejía Acosta 2003).

In Colombia, for both the Chamber of Deputies and the Senate, voters casttheir ballots for specific factions. Although the electoral system has perverse consequences (Archer and Shugart 1997; Pizarro Leongómez 2002; PizarroLeongómez, this volume), it offers some personal accountability of politicians tovoters. As Archer and Shugart (1997, 132– 40) noted, this accountability isbased on clientelism more than programmatic or ideological bases.

Electoral accountability in the Andes still suffers from deficiencies, and theColombian electoral system in particular generates perverse incentives (Archerand Shugart 1997; Pizarro Leongómez 2002; Pizarro Leongómez, this volume).Nevertheless, the core of the crisis of democratic representation in the Andesdoes not stem from the electoral systems. All electoral systems involve tradeoffsin the kinds of accountability and representation they foster. Many other coun-tries in the world use electoral systems without personal voting, in which indi-vidual elected politicians have no direct accountability to voters. In many of

324 Mainwaring

these democracies, the subjective sense of a crisis of confidence in democraticinstitutions is much less acute than it is in the Andes.

Institutional reforms can make a significant difference in how democraciesfunction. Nevertheless, I am skeptical that electoral-system reform can resolve thecrisis of democratic representation in the Andes. It will take more than changingthe electoral incentives between voters and representatives to fix this problem.

Political reforms in the Andes have focused primarily on addressing a per-ceived deficit of representation and on perceived deficiencies in representativedemocracy per se. The main problem, however, is one of results and not ofdeficiencies in the formal mechanisms of democratic representation. Poor resultshave led to a focus on the mechanisms of representation and on the perceiveddeficiencies of representative democracy.

This issue has important consequences not only for understanding the crisisof representation, but also for responding to it. If the core problem is a deficit ofresults more than one of representation, this has implications both for how to in-tellectually apprehend what is taking place and for how to address the problem.If the core problem were truncated representation, an oligopolistic electoralmarket, or deficiencies intrinsic to representative rather than direct democracy,institutional changes might remedy the situation. Institutional changes couldfoster more participation, stimulate decentralization, or tighten bonds betweencitizens and representatives through mechanisms such as preference voting inproportional elections, a mixed-proportional electoral system, referenda to en-courage more direct popular input, or mechanisms to encourage direct democ-racy at the local level.

If, however, the root of the crisis of democratic representation is statedeficiencies, then efforts to enhance popular participation and overcome elec-toral market oligopolies might not ameliorate the problem. Indeed, some insti-tutional reforms intended to enhance representation might exacerbate statedeficiencies. Bejarano and Pizarro Leongómez (2005) have argued that inColombia the efforts to enhance representation by overcoming truncated rep-resentation and by eliminating supply-side oligopolies led to major institutionalreforms in the second half of the 1980s and in the 1991 Constitution. But thesereforms exacerbated the atomization of the parties (Pizarro Leongómez 2002),made it harder for the government to win the congressional support needed totackle salient problems, and ultimately led to worse state performance. In turn,worse state performance fueled the subjective perception of a crisis of represen-tation. In this way, efforts to enhance representation intensified the crisis of rep-resentation.

In sum, the subjective sense of a crisis in representation has arisen more be-cause the state is not satisfying citizens’ needs (the results of representation) thanbecause of frustration with the process of representation (truncated representa-tion, supply-side oligopolies, or lack of accountability of elected politicians to


voters).43 Of course, dissatisfaction with the results of representation has fueleddissatisfaction with the process as well.

Interpersonal Trust and Institutional Confidence

Some scholars have claimed that interpersonal trust fosters stable democracy and economic development (Fukuyama 1995; Inglehart 1997a, 172–74; 1999).In a related vein, Putnam (1993, 167–76) argued that interpersonal trust is acomponent of social capital,44 which in turn produces effective political systems.Almond and Verba (1963, 266 –99) argued that high interpersonal trust was akey component of a civic culture, which was propitious for democracy. Diamond (1999, 208) wrote that “if trust is low and expectations of fellow cit-izens are pervasively cynical, institutions will be mere formalities, lacking com-pliance and effectiveness.” If these perspectives were correct, interpersonal trustmight explain confidence in institutions, and low interpersonal trust might be the key to low confidence in parties and legislatures.45 These perspectives,however, have a limited capacity to explain low trust in parties and assembliesin the Andes.

If institutional confidence were a product of interpersonal trust, then threeconsequences should result. First, at the aggregate level, in countries with greaterinterpersonal trust, there should be greater confidence in parties and legislatures.Second, within countries, individuals who exhibit greater trust in other peopleshould evince greater trust in these institutions. Third, confidence in institutionsshould be reasonably even across different institutions.

Based on the 1996 and 1998 Latinobarómetro surveys, the data for seventeenLatin American countries provide mixed support for the first hypothesis. TheLatinobarómetro survey asks a question about interpersonal trust: “Generallyspeaking, would you say that you can trust most people, or that you can never betoo careful when dealing with others?” The country-level correlation betweenthe percentage of respondents who expressed some or a lot of support in politi-cal parties and those who stated that one could trust most individuals was .46 in1996 (significant at p � .07 in a two-tailed test) and .67 in 1998 (significant atp � .01). The country-level correlation between confidence in Congress andinterpersonal trust, however, was weak at .31 in 1996 and .28 in 1998. Neithercorrelation approaches statistical significance at p � .10.

This interpersonal trust variable was included in the regressions in Tables 10.3and 10.5 above. In three of the five countries for parties and in four for Con-gress (though barely in Bolivia; p � .097), interpersonal trust was a statisticallysignificant predictor of confidence in institutions. Assuming for the momentthat confidence in institutions is a product of interpersonal trust rather than viceversa, the data therefore provide modest support for the hypothesis that institu-tional confidence is a product of interpersonal trust. But in all five countries except Peru, at least two of the four variables on perception of state performance

326 Mainwaring

have greater statistical significance than interpersonal trust. (In Peru, one vari-able on perception of state performance, the perception of crime, had greaterstatistical significance.) Moreover, in the simulations shown in Table 10.4 above,in no country did individuals who agreed that “You can trust most people”show vastly greater confidence in democratic representation than those who responded that “You can never be too careful when dealing with others.” Thesubstantive effect of interpersonal trust on confidence in parties and Congresswas not huge in any of the five countries, and it was null in Venezuela.

The third test provides even less support for the idea that low interpersonaltrust explains low confidence in parties and legislatures. Table 10.7 above enablesus to assess (a) whether the distrust in parties and legislatures was generalizableto religious and civic institutions; and (b) whether, even short of such a gener-alized distrust of institutions, there was a distinctive distrust of state institutions.These questions are important in light of the theoretical literature on politicaltrust. If interpersonal trust generated generalized confidence in institutions,there should not be great variance in confidence from one institution to thenext. A competing tradition sees institutional confidence as an expression ofjudgments about institutional performance (Mishler and Rose 2001). In this lineof analysis, there might be sharp differences in confidence according to citizenperceptions of institutional performance.

The highly negative assessment of parties and Congress did not characterize allinstitutions. In the Andes and in Latin America as a whole, the Church washighly trusted. Citizens also expressed much more confidence in the press thanin parties and Congress. The differences in confidence between the Church andthe press, on the one hand, and parties and Congress, on the other, are huge.Thus, there is no basis for concluding that low personal trust consistently pro-duces low confidence in institutions. Confidence in institutions depends on cit-izen evaluations of specific institutions, which in turn is a function of institutionalperformance and of the political construction of the assessments of institutionalperformance. In Hardin’s (2002) terms (although he reserves the concept to in-dividuals), some institutions are not trustworthy. They fail to give citizens rea-sons to believe that they will take the citizens’ interests seriously. The lowconfidence in Congress and political parties is not a result of a generalized lack ofconfidence in institutions, nor is it a product of low interpersonal trust.

The Andean and other Latin American cases pose doubts about Inglehart’s(1997a, 172) claim that “interpersonal trust plays a crucial role in democracy.”Inglehart argues that democracy depends on “the development of a culture oftrust” (172). In Latin America, in the post-1978 wave, democracy has beenstable as a political regime (notwithstanding considerable governmental instabil-ity in some countries) despite low interpersonal trust.

Interpersonal trust is probably a product of the effective functioning of stateinstitutions more than a prerequisite for it. This reverses the causal argumentmade by Inglehart (1999) and Putnam (1993), among others.46


Where democratic states cannot minimally enforce legality, and where cor-ruption and crime are rampant, it would be non-rational for citizens to trust oneanother. When citizens are fearful that they might be robbed or killed as a partof daily life, where they fear going out at night because citizen security is poor,and where they believe that the political system is predicated on widespreadgraft, they have no reason to trust people they do not know. Interpersonal trustsuffers in contexts of state deficiencies, and low interpersonal trust is not theprimary explanation for the lack of confidence in parties and legislatures in theAndes.

The literature that sees interpersonal trust as a way of creating more effectivedemocracy (Fukuyama 1995; Putnam 1993; Inglehart 1997a, 1999) is probablyright that states function more effectively in societies with high interpersonaltrust. Nevertheless, in the Andes, the stronger causal arrow goes in the oppositedirection, namely, that an effective state fosters interpersonal trust (Levi 1998).

Where state capacity erodes, such fundamental problems as generating jobs sothat people can secure their livelihoods, generating a good enough education sys-tem so that people have realistic opportunities to advance in their lives, and pro-viding citizen security remain unresolved. Citizen security is probably particu-larly key to interpersonal trust. Where states fail to curtail crime, citizens haveevery reason to be suspicious and fearful of people around them. Where crime isrampant and the justice system ineffective, as is true in the Andes, interpersonaltrust is likely to be low because rational people cannot trust others. Ineffectivejudiciaries, violent police forces, and effective impunity for most criminals havehampered interpersonal trust by fueling the perception that there is no effectivejustice.

The low interpersonal trust in the Andes, then, probably reflects more the re-alities of legal systems that have failed, police forces that are corrupt and inef-fective, public corruption, and states that have failed to address the needs of theircitizens more than a societal heritage of low interpersonal trust, as Inglehart(1997a, 1999) suggests. Moreover, when governmental leaders are not trust-worthy—in particular, when they are corrupt and place their own political ca-reers above the public good—this example generates a lack of trust in society.In short, the fundamental problem is creating a state that fosters interpersonaltrust, not generating interpersonal trust so that the institutions of representativedemocracy enjoy greater confidence.

The Paradox of Democratic Representation: More Representation,Greater Crisis of Representation

By many indicators, there was a boom in democratic representation in theAndes after 1978. The exclusions that prevented people from formal citizenshipand the practices that produced truncated citizenship became far less pervasive.

328 Mainwaring

More people became citizens in a formal sense, and fewer were subjected to tra-ditional forms of personalistic domination. Supply-side oligopolies weakened orvanished as a greater number and diversity of parties emerged. Yet greater repre-sentation coexisted with a deepening crisis of democratic representation.47 Thiscombination creates the paradox of democratic representation in the Andes.

In principle, the expansion of representation might have satisfied citizens,leading to a perception that the system of representation was open and legiti-mate even if other aspects of the political system were not working well. In prac-tice, the expansion of citizenship and of the supply side of representation prob-ably reinforced the subjective sense of a crisis of representation.

The massive incorporation of new citizens and their partial empowering, evenin countries that continue to be characterized by staggering social inequalities,has promoted an awareness of the ability and right to secure collective and par-ticularistic goods from the political system. The expansion of citizenship and theerosion of truncated forms of citizenship gave more people opportunities to ex-press disappointment with the political system and to place demands upon it.48

In the context of poor state performance, citizens understandably focus theirfrustration on suppliers of representation (parties and politicians) that fail toprovide more effective policy results. Parties and politicians are not solely re-sponsible for poor state performance—they share this responsibility with thepolice, judges, bureaucrats, and countless public-sector agencies and firms. Butbecause of elections and campaigns, they are the most visible among the actorsthat run the state, and ultimately they are supposed to have oversight of mostother state actors.

Through elections, citizens can directly take action by voting against partiesand politicians who seem to fail them. Except through unusual means (legal ac-tion or filing a complaint with the ombudsman), they cannot easily take actionagainst other state actors that contribute to state deficiencies. Seen from this per-spective, a crisis of representation has occurred not despite an increase in repre-sentation but rather partly because of it.

This paradox of democratic representation, by which greater representationfueled a crisis of representation, was not inevitable. If democratic governmentshad performed better, democratic representation could have expanded withouta crisis of representation. Poor state performance interacted with expandingdemocratic citizenship and a more competitive electoral market to form a caul-dron of frustration with democratic representation.

Conclusion

With the exception of Levi (1998), the state has been absent from the discus-sion about confidence in institutions. It is time to remedy this lacuna. The lowconfidence in parties and assemblies in the Andes stems above all from state


deficiencies. States have failed to resolve the policy concerns of the vast major-ity of citizens. Income inequalities have increased; poverty has increased in somecountries; crime has increased; job generation has been poor; economic growthhas been sluggish or worse (notwithstanding robust growth for most of LatinAmerica in 2004 and 2005); and there is a widespread perception that govern-mental corruption is rampant. Within Latin America, the Andean states haveperformed particularly poorly on most issues that affect the quality of everydaylife for citizens. Consistent with the argument of this chapter, the crisis of dem-ocratic representation has been more acute in the Andes than in most of the restof Latin America.

An equally notable lacuna in the comparative literature is the absence of thenotion that confidence is the product of political construction of assessments ofthe trustworthiness of different institutions. Although my empirical evidence onthis point has been limited to the Andes, this argument has broader comparativerelevance. It is the key to understanding why parties and assemblies most oftenare at the bottom of the order of institutions that are trusted by citizens.

Although the empirical evidence in this chapter has been limited to confidencein parties and assemblies, the assumption underlying the chapter is that statedeficiencies and the poor public image of parties formed by party competitionand the media account for other manifestations of a crisis of democratic repre-sentation. I also suspect that the two main arguments apply well beyond theAndean cases, and beyond Latin America.

A radical critique of representative democracy, which is widespread amongthe contemporary left in the Andes—especially in Bolivia, Ecuador, and Vene-zuela—is that it is intrinsically limited. In this perception, what is needed ismore direct democracy. Representative democracy is indeed intrinsically lim-ited. It is not a panacea. It does not automatically produce good public policy,and in the five Andean countries, with a few exceptions, representative democ-racy has not produced good results for most citizens since 1978.

Nevertheless, I doubt that more participatory democracy would resolve theshortcomings of democracy in the Andes or elsewhere. Participatory democracyhas been successful in some cases, but it has failed in other places.49 More par-ticipatory democracy may in some cases improve the quality of government atthe local level, but it provides no answer to the problems of formulating goodpolicy at the national level. Although good local government can make a posi-tive difference, good national government is essential in addressing many prob-lems citizens face. Formulating good national policies rests above all with pres-idents and their ministers, national legislatures, and other parts of the centralstate. Civil society can contribute to formulating good national policies, but itcannot play the lead role.

These remarks open the nettlesome question of why state performance hasbeen poor. Unfortunately, the state has been poorly researched empirically in

330 Mainwaring

and beyond contemporary Latin America. As a result, much remains to be an-alyzed about why state performance has been deficient. To some degree, the an-swer is a long-term historical one; compared to Argentina, Chile, and Uruguay,the states in the Andean region (especially Bolivia, Ecuador, and Peru) have longbeen deficient. The long legacy of patrimonialism, social and political exclusion,and dependence on extractive industries did not foster successful state building.However, some more historically proximate causes also merit attention becausestate performance, at least in many important arenas ( job creation, citizen se-curity, and, in Colombia and Peru, state control of the national territory) haseroded in the last quarter century.

One important contributing factor is that the debt crisis of the 1980s bank-rupted the state in most of Latin America and spurred inflation in most of theregion during that decade. Three decades of economic growth for the regionground to a halt, and market-oriented policies replaced the state-led develop-ment of 1950 –80. States had fewer resources to work with.

While some neoliberal reforms of the 1980s and 1990s were necessary giventhe state’s bankruptcy, other reforms weakened the state. In some circles, statebashing was in fashion in the 1990s. Yet the state is important for regulating someaspects of the economy, formulating policies, correcting market deficiencies,providing and enforcing a legal framework, protecting property rights, protect-ing citizens from crime, ensuring rights, and educating most children and teen-agers, among many other functions. The state cannot resolve all citizen problems,but without a functional state, society and the market do not function well. Inthe 1990s in Latin America, some state bashers were so eager to shrink the statethat they failed to appreciate how important it is for democracy and for economicperformance (Foxley 2004). In the Andes, levels of tax collection are so low thatstates are severely handicapped in performing their functions. Some reforms dis-mantled the state rather than striving to build a more efficient and effective state.The patrimonial practices of parties and politicians have also weakened the state(Guevara Mann 2001; Mainwaring 1999, 175–218; Weyland 1996).

Some important factors that weakened the Andean states were idiosyncratic.In Bolivia, Peru, and especially Colombia, the growth of the cocaine industryfueled criminality and weakened the justice system. In Peru and Colombia, rev-olutionary guerrilla movements sowed widespread destruction that debilitatedthe state.

Better state performance is key to promoting greater confidence in the insti-tutions of representative democracy and greater satisfaction with democracy.When democratic governments fail to produce what citizens need for a longtime, most citizens will distrust the institutions of representative democracy.

Social scientists appropriately seek to generalize findings across different re-gions of the world. Nevertheless, it might be misplaced to seek a generalized ex-planation for low or declining confidence in institutions in the contemporary


world. The Andean evidence suggests that state performance and capacity havea profound impact on confidence in parties and assemblies. In contrast, somework on the advanced industrial democracies has argued that economic perfor-mance has little impact on confidence in representative institutions (Pharr 2000).The impact of state performance on confidence in institutions might be stron-ger in the Andes than it is in the advanced industrial democracies. The reason iseasy to detect. In the Andes, state deficiencies have been acute in most of thepost-1978 period; in none of the advanced industrial democracies has it beencomparable. Poor economic performance in the Andes has had a dimension un-known in the advanced industrial democracies since the Great Depression. Statedeficiencies of this magnitude eroded confidence in the institutions of represen-tative democracy. Rampant corruption and widespread crime also shatteredconfidence in democratic representation.

I close by reiterating what this volume has contributed to the contemporarysocial science literature. First, we have added to and challenged the existing lit-erature on representation by shifting the focus to a region where democraticrepresentation is failing. Rather than assuming that representation works, that itis programmatic or ideological, and that patterns of representation are usuallystable, we have argued that representation sometimes fails, that it often has aweak programmatic and ideological component, and that it is often not stable.Second, we have added to the empirical understanding of democratic represen-tation in the Andes. Third, we have begun a new debate about the causes of acrisis of democratic representation. Fourth, we addressed the consequences of sucha crisis. Finally, we hope to have contributed to a conceptual issue: understand-ing what a crisis of democratic representation is.

Normatively, we would prefer a world in which democratic representation issuccessful and stable, but much of the world does not live under such condi-tions. It is important to begin understanding the nature of democratic repre-sentation where it fails, and the causes and consequences of such failures.

Notes

I am grateful to Ana María Bejarano, Matthew Cleary, Brian Crisp, Eric Hershberg,Mala Htun, Andrés Mejía, Guillermo O’Donnell, Kathleen O’Neill, Patricia Rodríguez,Mitch Sanders, Richard Snyder, Susan Stokes, Saika Uno, Jorge Vargas Cullel, andEdurne Zoco for helpful comments. Angel Alvarez, Dan Brinks, Saika Uno, and EdurneZoco provided research assistance. Thanks to seminar participants at the Instituto de Estudios Peruanos, the Kellogg Institute for International Studies at the University ofNotre Dame, the Pontificia Universidad Católica del Peru, and the Universidad CatólicaAndrés Bello, Caracas, Venezuela, for helpful observations.

332 Mainwaring

1. I use the terms “confidence” and “trust” in institutions interchangeably. TheSpanish and Portuguese question in the surveys asks whether citizens have confianza ininstitutions; confianza can be translated as either trust or confidence.

2. See Cleary and Stokes (2006), Power and Jamison (2005), and Turner and Martz(1997) on confidence in institutions in Latin America.

3. Throughout this chapter, whenever I refer to “representative institutions,” Ispecifically mean the representative institutions of liberal democracy. Not all representa-tive institutions are specific to liberal democracy, but my sole concern is those that are—above all, parties and national assemblies.

4. Mayorga (this volume) makes a similar argument. Some scholars who work on theadvanced industrial democracies have argued against the idea that declining confidencein institutions is a result of poor institutional performance (see McAllister 1999; Pharr2000, 177–81; Putnam, Pharr, and Dalton 2000, 24). Others argue that institutional performance has affected confidence in specific institutions (Alesina and Wacziarg 2000;Hardin 2000; Mishler and Rose 2001; Newton and Norris 2000; Warren 1999, 333–35;Williams 1985).

5. Colombia is a poignant example. The state in Bogotá has functioned reasonablywell in the last decade. In contrast, in the rural areas of intense armed conflict, the stateis profoundly deficient.

6. The reason for using subjective data on crime and corruption is that no reliablecross-national data are available.

7. In early 2005, when I revised this chapter, 1998 was the most recent publicly avail-able Latinobarómetro survey.

8. The evidence regarding the impact of economic performance on confidence in in-stitutions in the advanced industrial democracies is mixed. Lawrence (1997) and Pharr(2000, 177–81) argue that it is not clear that poor economic performance has caused de-clining trust. However, the meaning of poor economic performance in the advanced in-dustrial democracies is profoundly different from its meaning in the Andes. In the Andes, inflation rates reached 7,650 percent (Peru in 1990) and 8,171 percent (Bolivia in1985). Hyperinflation has devastating consequences for many citizens. Moreover, noneof the advanced industrial democracies has experienced the kind of protracted and pro-found economic decline that Venezuela has. Per capita income in Venezuela today is sub-stantially lower than it was in 1960. It would be surprising if such poor economic per-formance did not cause citizens to become skeptical of the institutions of democraticrepresentation.

9. I restricted the analysis to the twenty-three countries that had a combined Free-dom House score of 8 or less because countries that had a higher Freedom House scorehad authoritarian political regimes.

10. This form of regression analysis is used when there are more than two but fewerthan about seven values for the dependent variable, and when the values of the depen-dent variable fit into a clear progression from the lowest value to the highest, or vice versa.

11. These are q1, q4, q12, q13c, and q13e of the 1996 Latinobarómetro. The questionfor q1 is: “In general, how would you describe the present economic situation of thecountry? Would you say that it is very good, good, about average, bad, or very bad?” Forq4: “In general how would you describe your present economic situation and that ofyour family? Would you say that it is very good, good, about average, bad, or very bad?”For q13c and q13e: “From the list of issues that I am going to read out, do you think thatthey have increased a lot or a little, or have decreased a lot or a little or remained the samein the last five years?” The question for q13c refers to crime, and for q13e, corruption.I treated q1, q4, q13c, and q13e as continuous variables.


12. This is q12 of the 1996 Latinobarómetro: “Generally speaking would you say that you can trust most people, or that you can never be too careful when dealing withothers?”

13. None of the correlations among independent variables is sufficiently high to pro-duce problems of colinearity. The strongest correlations among the independent vari-ables are .52 between perception of crime and perception of corruption in Peru, and .51for the same two variables in Ecuador.

14. The Latinobarómetro also asked about individual’s assessment of whether the econ-omy had improved over the past year—a retrospective evaluation in Fiorina’s (1981) ter-minology—and their judgment about whether it was likely to improve in the next year(a prospective judgment). Of these three questions about the national economy, the as-sessment of the current economic situation was the best predictor of confidence in par-ties. I used it for this reason.

15. The arguments in this chapter about the impact of state deficiency on confidencein institutions could be assessed more thoroughly if survey data were available for a longtime period. Unfortunately, these data are not available for the five Andean countries ina cross-national survey before 1996.

16. This argument is consistent with Camp’s (2001) claim that Latin American citi-zens view democracy differently than U.S. citizens. In particular, Latin Americans tendto view democracy more in terms of social and economic equality and progress than U.S.citizens, who focus more on procedural issues. Kornblith (1998, 39) has a converging ar-gument that poor performance in Venezuela is the primary culprit for the crisis of dem-ocratic representation.

17. In a similar vein, Coppedge (2005) argues that the combination of corruption andeconomic decline was key in the erosion of democracy in Venezuela.

18. My argument echoes Pharr (2000), who showed that corruption had a powerfulimpact on citizen mistrust in Japan. See also della Porta (2000) and Seligson (2002a).

19. For a contrary finding, see Magalhães (2006), who reports a weak causal impactof these socio-demographic variables on confidence in institutions.

20. Bejarano and Pizarro Leongómez (2005) used the concept “partial collapse of thestate” to describe the Colombian situation.

21. This widespread view that democracy is not always the best form of governmentprobably helps explain the popular support for the 2000 coup that ousted democraticallyelected president Mahuad in Ecuador and the 2002 election of President Lucio Gutiérrez,who led the coup against Mahuad.

22. These data come from Kenney (2004). The 88 percent approval rating was in anApril 9–11, 1992, nationwide survey conducted by CPI. The 79 percent approval rat-ing for Fujimori was in an APOYO nationwide survey conducted April 11, 1992. Ken-ney notes that support for Fujimori’s coup remained strong for two years after his actions.

23. The percentage of respondents who express a lot or some confidence in Congressand parties was statistically lower at p � .001 in the Andes, compared to the other twelveLatin American countries.

24. I was unable to find systematic cross-national data on police homicides. A 2001U.S. State Department report on Venezuela offers some data on police killings. This re-port states that, “According to the Government, over 2000 suspected criminals werekilled in shoot-outs with the police during the first eight months of the year. . . . Manypoliticians contributed to a climate of official acceptance of the excessive use of forcewhen, during the national election campaign, they employed slogans such as ‘bullets forthe underworld’ and ‘the only good criminal is a dead criminal.’ ” If we annualize thisfigure to 3,000 killings for a year, and divide by the population of Venezuela (23,542,600in 2002), there were 12.75 police killings per 100,000 people per year. By way of

334 Mainwaring

contrast, São Paulo, known for a violent police force, had an average rate of 1.9 policekillings per 100,000 individuals in the 1990s. These figures are from the Ouvidoria daPolicia de São Paulo. New York City reported 41 deaths at the hands of the police in1990, its worst year, for a rate of .23 per 100,000 inhabitants (Chevigny 1995, 67). Venezuela’s police homicide rate per 100,000 inhabitants was thus more than six timesgreater than São Paulo’s, and more than fifty times greater than New York’s.

25. My analysis diverges from scholars who argue that confidence is simply the prod-uct of a rational calculation. For such an argument, see Bianco (1994). I partly agree withthis line of interpretation, but confidence is also the product of a battle among politicalactors to construct or destroy trustworthiness. What citizens perceive as a rational calcu-lation, that is, whether or not they believe that an institution is trustworthy in rationalterms, is politically constructed.

26. In comparative politics, the most prominent constructivist scholarship is on na-tionalism and ethnic identity. Constructivist interpretations are largely absent in the lit-erature on confidence in institutions. For a partial exception, see Mansbridge (1997),who focuses on sociocultural changes that generated rising expectations for governmentaction and created government overload.

27. http://comunidad.lanacion.com.ar/encuestas, June 8, 2004.28. King (1997) argued that growing polarization in the United States contributed to

a greater mistrust in institutions. If this is true in the United States, where polarization islimited, in Bolivia and Venezuela, where polarization is acute, it could easily fuel mis-trust in the institutions of representative democracy. This is especially true because someof the left criticizes representative democracy as inadequate, as a system of governmentthat excludes mass involvement and reinforces elite domination.

29. According to the 1996 Latinobarómetro, in all five Andean countries, television andradio dwarfed other sources of political information. In Bolivia, for example, 83 percentof respondents said that they obtained political information from television, 65 percentfrom radio, 48 percent from newspapers, 31 percent from the family, 23 percent fromfriends, 17 percent from work colleagues, and 9 percent from school or university colleagues.

30. Della Porta (2000) and Pharr (2000) emphasize the role of official misconduct andcorruption in explaining declining confidence in Italy and Japan, respectively. It is highlylikely that the salience of stories of scandals, corruption, and official misconduct has alsocontributed to low trust in parties and Congress in the Andes. The statistical significanceof the corruption variable in Bolivia, Colombia, and Ecuador in Tables 10.3 and 10.5supports this interpretation.

31. See Cappella and Jamieson (1997), Fallows (1996), Lipset and Schneider (1983,403– 6), Mutz and Reeves (2005), and Putnam (2000). For a dissenting view, see Norris(2000).

32. Some scholars who detect a crisis of representation in the advanced industrial de-mocracies argue that it has resulted from deficiencies in the system of representation. Ad-vocates of more direct participation often hold this view (see Barber 1984).

33. In Venezuela, by the 1970s, truncated citizenship was much less pervasive than inthe other four countries (Levine 1973).

34. Lynch (2000, 93–116) understood the Peruvian crisis of representation in the1990s as fundamentally resulting from truncated representation.

35. This argument on the expansion of citizenship in the Andes echoes LópezJiménez’s (1997) arguments about the development of citizenship in Peru.

36. Secondary-school enrollment is measured as the ratio of total enrollment, regard-less of age, to the population of the age group that officially corresponds to the second-ary level of education (World Bank 1998).


37. See DaMatta (1985, 55–80) and Holston and Caldeira (1998) on the unevenquality of citizenship in Brazil. O’Donnell (1993) coined the suggestive concept “low in-tensity citizenship” to capture this phenomenon.

38. Indigenous peoples (Albó 1994; Van Cott 2000a, 2000b), blacks, and women havealso historically been more subjected to truncated representation. The 1996 Latino-barómetro did not ask about ethnicity, so I was not able to analyze the impact of ethnicityon confidence in parties and assemblies.

39. Citizens might view parties that are programmatically and ideologically diverse asessentially the same because they were all deficient in delivering results. This is one read-ing of what has occurred in Ecuador and Peru, where governing parties with sharply di-verging left-right positions have all failed (albeit to different degrees).

40. The effective number of parties is a mathematical formula designed to measurehow many parties there are in a party system in a given election. It can be measured invotes or seats. The formula is ENP � 1/sum(p2), where p is the proportion of seats (orvotes) obtained by each party (Laakso and Taagepera 1979).

41. There is a tradeoff between fostering personal accountability of representa-tives to voters and fostering programmatic politicians. Systems with stronger personalaccountability of politicians to voters typically are weaker in fostering programmatic vot-ing and linkages (see Crisp’s chapter in this volume; Moreno, Crisp, and Shugart 2003).

42. In his contribution to this volume, Crisp argues that these electoral reforms didnot significantly change mechanisms of accountability and representation. My argumenthere is different. Carey and Shugart (1995) argued, correctly in my view, that in systemswith no intraparty competition in the actual election a smaller district magnitude (thenumber of representatives elected per district) strengthens the need to cultivate a personalvote. By this logic, the fact that Bolivia and Venezuela moved from proportional elec-toral systems with moderate to high district magnitudes to mixed systems in which half(Venezuela) or more (Bolivia) of the lower chamber is elected in single-member districtsshould have created stronger incentives to cultivate personal reputations, and by implica-tion stronger direct accountability of individual deputies to voters.

43. In a similar vein, Hibbing and Theiss-Morse (2002) argue that U.S. voters valueeffective policies much more than greater political involvement.

44. See Newton (1999) for a critique of this perspective. He argues that political trustand interpersonal trust are different phenomena, as does Hardin (2002). Norris (2000,62– 63) and Katzenstein (2000, 122–29) report that the relationship between interper-sonal trust and trust in institutions is weak in advanced industrial democracies. This find-ing is consistent with the one here for the Andes. Their arguments run counter to theclaims of Inglehart (1999). See Fishman (2004, 93–109) for a critique of Putman’s socialcapital theory.

45. None of the authors cited in this paragraph makes precisely this claim, but it is aplausible extension of their arguments.

46. Inglehart (1999) argues that institutions function well when interpersonal trust ishigh. He further argues that democratic institutions do not necessarily produce interper-sonal trust. See Seligson (2002b) for a critique. Muller and Seligson (1994) argue that In-glehart reverses the causal relationship between interpersonal trust and democracy. Theyclaim that it is more likely that interpersonal trust is the product of democracy rather thanvice versa. Hardin (2002, 151–72) and Cleary and Stokes (2006) are also skeptical thatinterpersonal trust breeds confidence in government and institutions, respectively. On thecapacity of the state to create interpersonal trust, see Levi (1998, 83–87).

47. Considerable evidence suggests that dissatisfaction with democratic representa-tion intensified in the 1990s: the growing popularity of outsider presidential candidates,

336 Mainwaring

increased electoral volatility, the collapses of the party systems in Peru and Venezuela, anddecreasing confidence in parties and assemblies.

48. My argument has similarities to Huntington’s (1968), who emphasized that in-creasing participation can create problems of governability. But in contrast to Hunting-ton, my argument focuses on the effects of state deficiency rather than the effects of in-creasing participation without increasing institutionalization. Also, Huntington focusedmainly on the institutionalization of parties; my central focus is state capacity. In the con-text of state shortcomings in responding to citizens’ needs, the explosion of citizenshipmeant that more people could express dissatisfaction with the system. Before the 1980s,many democratic governments in Latin America functioned well, even in the context ofsharp increases in participation. The primary problem is not primarily one of low insti-tutionalization of parties, but rather of an erosion of state capacity, which in turn pro-voked deinstitutionalization of party systems.

49. For arguments that representative democracy is superior to direct democracy, seeBrennan and Hamlin (1999) and Manin (1997). These contemporary arguments harkback to the classic views of James Madison and—albeit without his elitism—Schumpeter(1946).

References

Albó, Xavier. 1994. “And From Cataristas to MNRistas? The Surprising and Bold Al-liance Between Aymaras and Neo-Liberals in Bolivia.” In Indigenous Peoples and De-mocracy in Latin America, ed. Donna Lee Van Cott, 55–81. New York: St. MartinsPress and the Inter-American Dialogue.

Alesina, Alberto, and Romain Wacziarg. 2000. “The Economics of Civic Trust.” In Dis-affected Democracies: What’s Troubling the Trilateral Countries? ed. Susan J. Pharr and Rob-ert D. Putnam, 149–70. Princeton, NJ: Princeton University Press.

Almond, Gabriel A., and Sidney Verba. 1963. The Civic Culture: Political Attitudes and De-mocracy in Five Nations. Princeton, NJ: Princeton University Press.

Archer, Ronald P. 1995. “Party Strength and Weakness in Colombia’s Besieged Democ-racy.” In Building Democratic Institutions: Party Systems in Latin America, ed. Scott Main-waring and Timothy R. Scully, 164 –99. Stanford, CA: Stanford University Press.

Archer, Ronald P., and Matthew Soberg Shugart. 1997. “The Unrealized Potential ofPresidential Dominance in Colombia.” In Presidentialism and Democracy in Latin Amer-ica, ed. Scott Mainwaring and Matthew Soberg Shugart, 110 –59. Cambridge: Cam-bridge University Press.

Barber, Benjamin R. 1984. Strong Democracy: Participatory Democracy for a New Age. Berke-ley: University of California Press.

Barrera Guarderas, Augusto. 2001. Acción colectiva y crisis política: El movimiento indígenaecuatoriano en la década de los noventa. Quito: Centro de Investigaciones CIUDAD.

Basombrío, Carlos. 2003. “The Militarization of Public Security in Peru.” In Crime andViolence in Latin America, ed. Hugo Frühling and Joseph S. Tulchin, with Heather A.Golding, 153–78. Washington, DC: Woodrow Wilson Center Press.

Beissinger, Mark R., and Crawford Young. 2002. “Introduction: Comparing State Crisesacross Two Continents.” In Beyond State Crisis: Postcolonial Africa and Post-Soviet Eur-asia in Comparative Perspective, ed. Mark R. Beissinger and Crawford Young, 3–50.Washington, DC: Woodrow Wilson Center Press.


Bejarano, Ana María, and Eduardo Pizarro Leongómez. 2005. “From ‘Restricted’ to ‘Be-sieged’: The Changing Nature of the Limits to Democracy in Colombia.” In TheThird Wave of Democratization in Latin America: Advances and Setbacks, ed. FrancesHagopian and Scott Mainwaring, 235– 60. Cambridge: Cambridge University Press.

Bianco, William T. 1994. Trust: Representatives and Constituents. Ann Arbor: University ofMichigan Press.

Botero, Felipe. 1998. “El senado que nunca fue: La circunscripción nacional después detres elecciones.” In Elecciones y democracia en Colombia, 1997–1998, ed. Ana María Be-jarano and Andrés Dávila, 191–214. Bogotá: Universidad de los Andes, Departamentode Ciencia Política, Fundación Social, Veeduría Ciudadana a la Elección Presidencial.

Brennan, Geoffrey, and Alan Hamlin. 1999. “On Political Representation.” British Jour-nal of Political Science 29 ( January): 109–28.

Brinks, Daniel. 2003. “Informal Institutions and the Rule of Law: The Judicial Responseto State Killings in São Paulo and Buenos Aires in the 1990s.” Comparative Politics 36,no. 1 (October): 1–17.

———. 2004. “Legal Tolls and the Rule of Law: The Judicial Response to Police Killingsin Latin America.” Ph.D. diss., University of Notre Dame.

Camp, Roderic Ai. 2001. “Democracy through Latin American Lenses: An Appraisal.”In Citizen Views of Democracy in Latin America, ed. Roderic Ai Camp, 3–23. Pittsburgh:University of Pittsburgh Press.

Cappella, Joseph N., and Kathleen Hall Jamieson. 1997. Spiral of Cynicism: The Press andthe Public Good. Oxford: Oxford University Press.

Carey, John M., and Matthew S. Shugart. 1995. “Incentives to Cultivate a Personal Vote:A Rank Ordering of Electoral Formulas.” Electoral Studies 14, no. 4: 417–39.

Chevigny, Paul. 1995. Edge of the Knife: Police Violence in the Americas. New York: NewPress.

———. 2003. “The Control of Police Misconduct in the Americas.” In Crime and Vio-lence in Latin America, ed. Hugo Frühling and Joseph S. Tulchin, with Heather A.Golding, 45– 67. Washington, DC: Woodrow Wilson Center Press.

Chinchilla, Laura. 2003. “Experiences with Citizen Participation in Crime Preventionin Central America.” In Crime and Violence in Latin America, ed. Hugo Frühling andJoseph S. Tulchin, with Heather A. Golding, 205–32. Washington, DC: WoodrowWilson Center Press.

Cleary, Matthew, and Susan Stokes. 2006. Democracy and the Culture of Skepticism: Politi-cal Trust in Argentina and Mexico. New York: Russell Sage Foundation.

Conaghan, Catherine M. 1995. “Politicians against Parties: Discord and Disconnectionin Ecuador’s Party System.” In Building Democratic Institutions: Party Systems in LatinAmerica, ed. Scott Mainwaring and Timothy R. Scully, 434 –58. Stanford, CA: Stan-ford University Press.

Coppedge, Michael. 1994. Strong Parties and Lame Ducks: Presidential Partyarchy and Fac-tionalism in Venezuela. Stanford, CA: Stanford University Press.

———. 2005. “Explaining Democratic Deterioration in Venezuela through Nested In-ference.” In The Third Wave of Democratization in Latin America: Advances and Setbacks,ed. Frances Hagopian and Scott Mainwaring, 289–316. Cambridge: Cambridge Uni-versity Press.

Cotler, Julio. 1995. “Political Parties and the Problems of Democratic Consolidation inPeru.” In Building Democratic Institutions: Party Systems in Latin America, ed. Scott Main-waring and Timothy R. Scully, 323–53. Stanford, CA: Stanford University Press.

Crisp, Brian F., Daniel H. Levine, and José E. Molina. 2003. “The Rise and Decline ofCOPEI in Venezuela.” In Christian Democracy in Latin America: Electoral Competition

338 Mainwaring

and Regime Conflicts, ed. Scott Mainwaring and Timothy R. Scully, 275–300. Stan-ford, CA: Stanford University Press.

Crisp, Brian F., and Juan Carlos Rey. 2001. “The Sources of Electoral Reform in Vene-zuela.” In Mixed Member Electoral Systems: The Best of Both Worlds? ed. Matthew SobergShugart and Martin P. Wattenberg, 173–93. Oxford: Oxford University Press.

Dalton, Russell J. 1996. Citizen Politics: Public Opinion and Political Parties in Advanced In-dustrial Democracies. Chatham, NJ: Chatham House.

———. 2000. “Value Change and Democracy.” In Disaffected Democracies: What’s Trou-bling the Trilateral Countries? ed. Susan J. Pharr and Robert D. Putnam, 252– 69.Princeton, NJ: Princeton University Press.

DaMatta, Roberto. 1985. A casa e a rua: Espaço, cidadania, mulher e morte no Brasil. SãoPaulo: Editora Brasiliense S.A.

della Porta, Donatella. 2000. “Social Capital, Beliefs in Government, and Political Cor-ruption.” In Disaffected Democracies: What’s Troubling the Trilateral Countries? ed. SusanJ. Pharr and Robert D. Putnam, 202–28. Princeton, NJ: Princeton University Press.


Dodson, Michael, and Donald W. Jackson. 2003. “Horizontal Accountability and theRule of Law in Central America.” In Democratic Accountability in Latin America, ed.Scott Mainwaring and Christopher Welna, 228– 65. Oxford: Oxford University Press.

Dogan, Mattei. 1997. “Erosion of Confidence in Advanced Democracies.” Studies inComparative International Development 32, no. 3 (Fall): 3 –29.

Döring, Herbert. 1992. “Higher Education and Confidence in Institutions: A SecondaryAnalysis of the European Values Survey, 1981–1983.” West European Politics 15, no. 2(April): 126 – 46.

Downs, Anthony. 1957. An Economic Theory of Democracy. New York: Harper and Row.ECLAC (Economic Commission for Latin America and the Caribbean). 2005. Balance

preliminar de las economías de América Latina y el Caribe 2005. Santiago, Chile: ECLAC.Fallows, James. 1996. Breaking the News. New York: Pantheon Books.Ferejohn, John. 1999. “Accountability and Authority: Toward a Theory of Political Ac-

countability.” In Democracy, Accountability, and Representation, ed. Adam Przeworski,Susan C. Stokes, and Bernard Manin, 131–53. Cambridge: Cambridge UniversityPress.

Fiorina, Morris. 1981. Retrospective Voting in American National Elections. New Haven,CT: Yale University Press.

Fishman, Robert. 2004. Democracy’s Voices: Social Ties and the Quality of Public Life in Spain.Ithaca, NY: Cornell University Press.

Foxley, Alejandro. 2004. “Más mercado o más estado para América Latina?” Paper pre-sented at the InterAmerican Development Bank seminar, Lima, Peru, March.

Friedenberg, Flavia. 2001. “Ecuador.” In Partidos políticos de América Latina: Países Andi-nos, ed. Manuel Alcántara Sáez and Flavia Friedenberg, 235– 406. Salamanca: Edi-ciones Universidad de Salamanca.

Fukuyama, Francis. 1995. Trust: The Social Virtues and the Creation of Prosperity. NewYork: The Free Press.

Gamarra, Eduardo A., and James M. Malloy. 1995. “The Patrimonial Dynamics of PartyPolitics in Bolivia.” In Building Democratic Institutions: Party Systems in Latin America,ed. Scott Mainwaring and Timothy R. Scully, 399– 433. Stanford, CA: Stanford Uni-versity Press.

García Montero, Mercedes, and Flavia Friedenberg. 2001. “Perú.” In Partidos políticos deAmérica Latina: Países Andinos, ed. Manuel Alcántara Sáez and Flavia Friedenberg,410 –84. Salamanca: Ediciones Universidad de Salamanca.


González de Pacheco, Rosa Amelia. 2001. “Reforma del estado y descentralización enVenezuela: Percepciones y comportamiento del votante.” In Venezuela en transición:Elecciones y democracia, 1998–2000, ed. José Vicente Carrasquero, Thais Maingon, andFriedrich Welsch, 164 –87. Caracas: CDB Publicaciones.

Graham, Carol. 1992. Peru’s APRA: Parties, Politics, and the Elusive Quest for Democracy.Boulder, CO: Lynne Rienner Publishers.

Grindle, Merilee S. 2000. Audacious Reforms: Institutional Invention and Democracy in LatinAmerica. Baltimore: Johns Hopkins University Press.

Guevara Mann, Carlos. 2001. Forsaken Virtue: An Analysis of the Political Behavior of Pana-manian Legislators, 1984 –1999. Ph.D. diss., University of Notre Dame.

Gutiérrez Sanín, Francisco. 2002. “Historias de democratización anómala: El PartidoLiberal en el sistema político colombiano desde el Frente Nacional hasta hoy.” InDegradación o cambio: Evolución del sistema político colombiano, ed. Francisco GutiérrezSanín, 25–78. Bogotá: Grupo Editorial Norma.

Hagopian, Frances. 1998. “Democracy and Political Representation in Latin America inthe 1990s: Pause, Reorganization, or Decline?” In Fault Lines of Democracy in Post-Transition Latin America, ed. Felipe Agüero and Jeffrey Stark, 99–143. Miami: North-South Center Press.

Hardin, Russell. 2000. “The Public Trust.” In Disaffected Democracies: What’s Troubling theTrilateral Countries? ed. Susan J. Pharr and Robert D. Putnam, 31–51. Princeton, NJ:Princeton University Press.

———. 2002. Trust and Trustworthiness. New York: Russell Sage Foundation.Hartlyn, Jonathan. 1988. The Politics of Coalition Rule in Colombia. Cambridge: Cam-

bridge University Press.Hibbing, John R., and Elizabeth Theiss-Morse. 2002. Stealth Democracy. Cambridge:

Cambridge University Press.Hinich, Melvin J., and Michael C. Munger. 1994. Ideology and the Theory of Political

Choice. Ann Arbor: University of Michigan Press.Holston, James, and Teresa P. R. Caldeira. 1998. “Democracy, Law, and Violence: Dis-

junctions of Brazilian Citizenship.” In Fault Lines of Democracy in Post-Transition LatinAmerica, ed. Felipe Agüero and Jeffrey Stark, 263–96. Miami: North-South CenterPress.

Huntington, Samuel P. 1968. Political Order in Changing Societies. New Haven, CT: YaleUniversity Press.

Inglehart, Ronald. 1997a. Modernization and the Post-Modernization: Cultural, Economic,and Political Change in 43 Societies. Princeton, NJ: Princeton University Press.

———. 1997b. “Post Materialist Values and the Erosion of Institutional Authority.” InWhy People Don’t Trust Government, ed. Joseph S. Nye Jr., Philip D. Zelikow, andDavid C. King, 217–36. Cambridge, MA: Harvard University Press.

———. 1999. “Trust, Well-being, and Democracy.” In Democracy and Trust, ed. Mark E.Warren, 88–120. Cambridge: Cambridge University Press.

Katzenstein, Peter J. 2000. “Confidence, Trust, International Relations, and Lessons fromSmaller Democracies.” In Disaffected Democracies: What’s Troubling the Trilateral Coun-tries? ed. Susan J. Pharr and Robert D. Putnam, 121– 48. Princeton, NJ: PrincetonUniversity Press.

Kenney, Charles D. 2004. Fujimori’s Coup and the Breakdown of Democracy in Latin Amer-ica. Notre Dame, IN: University of Notre Dame Press.

King, David C. 1997. “The Polarization of American Parties and Mistrust of Govern-ment.” In Why People Don’t Trust Government, ed. Joseph S. Nye Jr., Philip D. Zelikow,and David C. King, 155–78. Cambridge, MA: Harvard University Press.

340 Mainwaring

Kitschelt, Herbert. 2000. “Linkages between Citizens and Politicians in Democratic Pol-itics.” Comparative Political Studies 33, no. 6 –7 (August–September): 845–79.

Kornblith, Miriam. 1998. Venezuela en los noventa: Las crisis de la democracia. Caracas: Edi-ciones IESA.

Kornblith, Miriam, and Daniel H. Levine. 1995. “Venezuela: The Life and Times of theParty System.” In Building Democratic Institutions: Party Systems in Latin America, ed.Scott Mainwaring and Timothy R. Scully, 37–71. Stanford, CA: Stanford UniversityPress.

Laakso, Markku, and Rein Taagepera. 1979. “Effective Number of Parties: A Measurewith Application to Western Europe.” Comparative Political Studies 12, no. 1 (April):3 –27.

Lawrence, Robert Z. 1997. “Is It Really the Economy, Stupid?” In Why People Don’t TrustGovernment, ed. Joseph S. Nye Jr., Philip D. Zelikow, and David C. King, 111–32.Cambridge, MA: Harvard University Press.

Levi, Margaret. 1998. “A State of Trust.” In Trust and Governance, ed. Valerie Braithwaiteand Margaret Levi, 77–101. New York: Russell Sage Foundation.

Levine, Daniel H. 1973. Conflict and Political Change in Venezuela. Princeton, NJ: Prince-ton University Press.

Lewis-Beck, Michael S. 1988. Economics and Elections: The Major Western Democracies. AnnArbor: University of Michigan Press.

Linz, Juan J., and Alfred Stepan. 1989. “Political Crafting of Democratic Consolidationor Destruction: European and South American Comparisons.” In Democracy in theAmericas: Stopping the Pendulum, ed. Robert A. Pastor, 41– 61. New York: Holmes andMeier.

———. 1996. Problems of Democratic Transition and Consolidation: Southern Europe, SouthAmerica, and Post-Communist Europe. Baltimore: Johns Hopkins University Press.

Lipset, Seymour Martin. 1959. Political Man: The Social Bases of Politics. Garden City, NY:Anchor Books.

Lipset, Seymour Martin, and William Schneider. 1983. The Confidence Gap: Business, La-bor, and Government in the Public Mind. Baltimore: Johns Hopkins University Press.

López Jiménez, Sinesio. 1997. Ciudanos reales e imaginarios: Concepciones, desarrollo y ma-pas de la ciudadanía en el Perú. Lima: Instituto de Diálogo y Propuestas.

Lynch, Nicolás. 1999. Una tragedia sin héroes: La derrota de los partidos y el origen de los in-dependientes, Perú 1980 –1992. Lima: Fondo Editorial Universidad Nacional Mayor deSan Marcos.

———. 2000. Política y antipolítica en el Perú. Lima: DESCO, Centro de Estudios y Pro-moción del Desarrollo.

Magalhães, Pedro C. 2006. “Confidence in Parliaments: Performance, Representation,and Accountability.” In Political Disaffection in Contemporary Democracies: Social Capital,Institutions, and Politics, ed. Mariano Torcal and José Ramón Montero, 190 –214. Lon-don: Routledge.

Mainwaring, Scott. 1999. Rethinking Party Systems in the Third Wave of Democratization:The Case of Brazil. Stanford, CA: Stanford University Press.

Mainwaring, Scott, and Mariano Torcal. 2006. “Party System Institutionalization and Party System Theory after the Third Wave of Democratization.” In Handbook of Political Parties, ed. Richard S. Katz and William Crotty, 204 –27. London: SagePublications.

Manin, Bernard. 1997. Principles of Representative Government. Cambridge: CambridgeUniversity Press.


Manin, Bernard, Adam Przeworski, and Susan Stokes. 1999. “Elections and Representa-tion.” In Democracy, Accountability, and Representation, ed. Adam Przeworski, Susan C.Stokes, and Bernard Manin, 29–54. Cambridge: Cambridge University Press.

Mansbridge, Jane. 1997. “Social and Cultural Causes of Dissatisfaction with U.S. Gov-ernment.” In Why People Don’t Trust Government, ed. Joseph S. Nye Jr., Philip D. Ze-likow, and David C. King, 133–53. Cambridge, MA: Harvard University Press.

Maravall, José María. 1999. “Accountability and Manipulation.” In Democracy, Account-ability, and Representation, ed. Adam Przeworski, Susan C. Stokes, and Bernard Manin,29–54. Cambridge: Cambridge University Press.

Mayorga, René Antonio. 2001a. “Electoral Reform in Bolivia: Origins of the Mixed-Member Proportional System.” In Mixed Member Electoral Systems: The Best of BothWorlds? ed. Matthew Soberg Shugart and Martin P. Wattenberg, 194 –208. Oxford:Oxford University Press.

———. 2001b. “The Mixed-Member Proportional System and its Consequences in Bo-livia.” In Mixed Member Electoral Systems: The Best of Both Worlds? ed. Matthew SobergShugart and Martin P. Wattenberg, 432– 45. Oxford: Oxford University Press.

———. 2002a. “Democracia y liderazgo político en Bolivia.” In ‘Dadme un balcón y elpaís es mío’: Liderazgo político en América Latina, ed. Wilhelm Hofmeister, 167–212. Riode Janeiro: Fundação Konrad Adenauer.

———. 2002b. “La metamorfosis del sistema de partidos.” Opiniones y Análisis: El Esce-nario Post-electoral en Bolivia 60 (October 2002): 67–112.

McAllister, Ian. 1999. “The Economic Performance of Governments.” In Critical Citi-zens: Global Support for Democratic Governance, ed. Pippa Norris, 188–203. Oxford:Oxford University Press.

Mejía Acosta, Andrés. 2002. Gobernabilidad democrática: Sistema electoral, partidos políticos ypugna de poderes en Ecuador, 1978–1998. Quito: Konrad Adenauer Stiftung.

———. 2003. “La reelección legislativa en Ecuador: Conexión electoral, carreras leg-islativas y partidos políticos (1979–2003).” In El legislador a examen. El debate sobre lareelección legislativa en México: Una perspectiva histórica e institucional, ed. F. F. Dworak,206 –28. Mexico City: Fondo de Cultura Económica.

Mishler, William, and Richard Rose. 1997. “Trust, Distrust, and Skepticism: PopularEvaluations of Civil and Political Institutions in Post-Communist Societies.” Journalof Politics 59, no. 2 (May): 418–51.

———. 2001. “Where Are the Origins of Political Trust? Testing Institutional and Cul-tural Theories in Post-Communist Societies.” Comparative Political Studies 34, no. 1(February): 30 – 62.

Moreno, Erika, Brian F. Crisp, and Matthew Soberg Shugart. 2003. “The Accountabil-ity Deficit in Latin America.” In Democratic Accountability in Latin America, ed. ScottMainwaring and Christopher Welna, 79–131. Oxford: Oxford University Press.

Morrison, Andrew, Mayra Bivinic, and Michael Shifter. 2003. “The Violent Americas:Risk Factors, Consequences, and Policy Implications of Social and Domestic Vio-lence.” In Crime and Violence in Latin America, ed. Hugo Frühling and Joseph S.Tulchin, with Heather A. Golding, 93–122. Washington, DC: Woodrow WilsonCenter Press.

Muller, Edward N., and Mitchell A. Seligson. 1994. “Civic Culture and Democracy:The Question of Causal Relationships.” American Political Science Review 88, no. 3(September): 635–52.

Mutz, Diana C., and Byron Reeves. 2005. “The New Videomalaise: Effects of TelevisedIncivility on Political Trust.” American Political Science Review 99, no. 1 (February):1–16.

342 Mainwaring

Newton, Kenneth. 1999. “Social and Political Trust in Established Democracies.” InCritical Citizens: Global Support for Democratic Governance, ed. Pippa Norris, 169–87.Oxford: Oxford University Press.

Newton, Kenneth, and Pippa Norris. 2000. “Confidence in Public Institutions: Faith,Culture, or Performance?” In Disaffected Democracies: What’s Troubling the TrilateralCountries? ed. Susan J. Pharr and Robert D. Putnam, 52–73. Princeton, NJ: Prince-ton University Press.

Nohlen, Dieter, ed. 1993. Enciclopedia electoral de América Latina y el Cáribe. San José,Costa Rica: Instituto Interamericano de Derechos Humanos.

Norris, Pippa. 2000. “The Impact of Television on Civic Malaise.” In Disaffected Democra-cies: What’s Troubling the Trilateral Countries? ed. Susan J. Pharr and Robert D. Putnam,231–51. Princeton, NJ: Princeton University Press.

O’Donnell, Guillermo. 1993. “On the State, Democratization, and Some ConceptualProblems: A Latin American View with Glances at Some Post-Communist Coun-tries.” World Development 21, no. 8: 1345– 69.

———. 1999. “Poliarchies and the (Un)Rule of Law in Latin America: A Partial Con-clusion.” In The (Un)Rule of Law and the Underprivileged in Latin America, ed. Juan E.Méndez, Guillermo O’Donnell, and Paulo Sérgio Pinheiro 303–37. Notre Dame,IN: University of Notre Dame Press.

———. 2003. “Democracia, desarrollo humano y derechos humanos.” In Democracia,desarrollo humano y ciudadanía, ed. Guillermo O’Donnell, Osvaldo Iazzeta, and JorgeVargas Cullell, 25–147. Rosario, Argentina: Homo Sapiens.

Pachano, Simón. 1998. La representación caótica (Análisis del sistema electoral ecuatoriano).Quito: FLACSO, Sede Ecuador.

Pachón, Mónica. 2002. “Partido conservador y sus dinámicas políticas.” In Degradación ocambio: Evolución del sistema político colombiano, ed. Francisco Gutiérrez Sanín, 79–130.Bogotá: Grupo Editorial Norma.

Pásara, Luis. 2002. “Justica y ciudadania realmente existentes.” Política y gobierno 9, no. 2,2nd semester: 361– 402.

Payne, Mark, Daniel Zovatto G., Fernando Carrillo Flórez, and Andrés AllamandZavala. 2002. Democracies in Development: Politics and Reform in Latin America. Wash-ington, DC: Inter-American Development Bank and the International Institute forDemocracy and Electoral Assistance.

Peñaranda, Ricardo. 2002. “Los nuevos ciudadanos: Las organizaciones indígenas en elsistema político.” In Degradación o cambio: Evolución del sistema político colombiano, ed.Francisco Gutiérrez Sanín, 131–81. Bogotá: Grupo Editorial Norma.

Penfold Becerra, Michael. 2001. “El colapso del sistema de partidos en Venezuela: Ex-plicación de una muerte anunciada.” In Venezuela en transición: Elecciones y democracia,1998–2000, ed. José Vicente Carrasquero, Thais Maingon, and Friedrich Welsch,36 –51. Caracas: CDB Publicaciones.

Pharr, Susan J. 2000. “Officials’ Misconduct and Public Mistrust: Japan and the Tri-Lateral Democracies.” In Disaffected Democracies: What’s Troubling the Trilateral Coun-tries? ed. Susan J. Pharr and Robert D. Putnam, 173–201. Princeton, NJ: PrincetonUniversity Press.

Pizarro Leongómez, Eduardo. 2002. “La atomización partidista en Colombia: El fenó-meno de las microempresas electorales.” In Degradación o cambio: Evolución del sistemapolítico colombiano, ed. Francisco Gutiérrez Sanín, 357– 401. Bogotá: Grupo EditorialNorma.

Power, Timothy J., and Giselle D. Jamison. 2005. “Political Mistrust in Latin America.”Comparative Sociology 4, no. 1–2: 47–72.


Putnam, Robert D. 2000. Bowling Alone: The Collapse and Revival of American Community.New York: Touchstone.

Putnam, Robert D., with Robert Leonardi and Raffaella Y. Nanetti. 1993. Making Democracy Work: Civic Traditions in Modern Italy. Princeton, NJ: Princeton UniversityPress.

Putnam, Robert D., Susan J. Pharr, and Russell J. Dalton. 2000. “Introduction: What’sTroubling the Trilateral Democracies?” In Disaffected Democracies: What’s Troubling theTrilateral Countries? ed. Susan J. Pharr and Robert D. Putnam, 3–27. Princeton, NJ:Princeton University Press.

Querubín, Cristina, María Fernanda Sánchez, and Ileana Kure. 1998. “Dinámica de laselecciones populares.” In Elecciones y democracia en Colombia, 1997–1998, ed. AnaMaría Bejarano and Andrés Dávila, 117– 40. Bogotá: Universidad de los Andes, De-partamento de Ciencia Política, Fundación Social, Veeduría Ciudadana a la ElecciónPresidencial.

Rodríguez-Raga, Juan Carlos. 2002. “¿Cambiar todo para que nada cambie? Repre-sentación, sistema electoral y sistema de partido en Colombia: Capacidad deadaptación de las élites políticas a cambios en el entorno institucional.” In Degradacióno cambio: Evolución del sistema político colombiano, ed. Francisco Gutiérrez Sanín,221– 60. Bogotá: Grupo Editorial Norma.

Rotberg, Robert I. 2004. “The Failure and Collapse of Nation-States: Breakdown, Pre-vention, and Repair.” In When States Fail: Causes and Consequences, ed. Robert I. Rot-berg, 1– 49. Princeton, NJ: Princeton University Press.

Sartori, Giovanni. 1976. Parties and Party Systems: A Framework for Analysis. New York:Cambridge University Press.

Schmidt, Gregory D. 2003. “The Great Minority: Christian Democracy in Peru.” InChristian Democracy in Latin America: Electoral Competition and Regime Conflicts, ed. ScottMainwaring and Timothy R. Scully, 330 – 62. Stanford, CA: Stanford University Press.

Schumpeter, Joseph. 1946. Capitalism, Socialism, and Democracy. New York: Harper andRow.

Seligson, Mitchell A. 2002a. “The Impact of Corruption on Regime Legitimacy: AComparative Study of Four Latin American Countries.” Journal of Politics 64, no. 2(May): 408–33.

——— 2002b. “The Renaissance of Political Culture or the Renaissance of the Ecolog-ical Fallacy?” Comparative Politics 34, no. 3 (April): 273–92.

Shugart, Matthew Soberg. 2001. “‘Extreme’ Electoral Systems and the Appeal of theMixed Member Alternative.” In Mixed Member Electoral Systems: The Best of BothWorlds? ed. Matthew Soberg Shugart and Martin P. Wattenberg, 25–51. Oxford: Ox-ford University Press.

Smulovitz, Catalina. 2003. “Citizen Insecurity and Fear: Public and Private Responses inArgentina.” In Crime and Violence in Latin America, ed. Hugo Frühling and Joseph S.Tulchin, with Heather A. Golding, 125–52. Washington, DC: Woodrow WilsonCenter Press.

Stokes, Susan. 1999. “What Do Policy Switches Tell Us about Democracy?” In Democ-racy, Accountability, and Representation, ed. Adam Przeworski, Susan C. Stokes, andBernard Manin, 98–130. Cambridge: Cambridge University Press.

Tanaka, Martín. 1998. Los espejismos de la democracia: El colapso del sistema de partidos en elPerú, 1980 –1995, en perspectiva comparada. Lima: Insituto de Estudios Peruanos.

Turner, Frederick C., and John D. Martz. 1997. “Institutional Confidence and Demo-cratic Consolidation in Latin America.” Studies in Comparative International Develop-ment 32, no. 3 (Fall): 65–84.

344 Mainwaring

Ungar, Elisabeth, and Germán Ruiz. 1998. “¿Hacia la recuperación del Congreso?” InElecciones y Democracia en Colombia 1997–1998, ed. Ana María Bejarano and AndrésDávila, 191–214. Bogotá: Universidad de los Andes, Departamento de CienciaPolítica, Fundación Social, Veeduría Ciudadana a la Elección Presidencial.

Van Cott, Donna Lee. 2000a. The Friendly Liquidation of the Past: The Politics of Diversityin Latin America. Pittsburgh: University of Pittsburgh Press.

———. 2000b. “Party System Development and Indigenous Populations in LatinAmerica: The Bolivian Case.” Party Politics 6, no. 2: 155–74.

Waldmann, Peter. 2003. El estado anómico: Derecho, seguridad pública y vida cotidiana enAmérica Latina. Caracas: Nueva Sociedad.

Warren, Mark E. 1999. “Democratic Theory and Trust.” In Democracy and Trust, ed.Mark E. Warren, 310 – 45. Cambridge: Cambridge University Press.

Weyland, Kurt. 1996. Democracy without Equity: Failures of Reform in Brazil. Pittsburgh:University of Pittsburgh Press.

———. 1998. “The Politics of Corruption in Latin America.” Journal of Democracy 9, no.2: 108–21.

Williams, John T. 1985. “Systemic Influences on Political Trust: The Importance of Per-ceived Institutional Performance.” Political Methodology 11, no. 1–2: 125– 42.

World Bank. 1998. World Development Indicators. Washington, DC: World Bank.WHO (World Health Organization). 2002. 2002 World Report on Violence and Health.

http://www.who.int/violence_injury_prevention/violence/world_report/en/.Zartman, I .William. 1995. “Introduction: Posing the Problem of State Collapse.” In Col-

lapsed States, ed. I.William Zartman, 1–14. Boulder, CO: Lynne Rienner Publishers.


Acción Democrática. See ADAcción Democrática Nacionalista.

See ADNAcción Popular. See APAD (Venezuela, Democratic Action), 20,

22–23, 36n7, 57– 62, 139, 140, 153,186 –87, 322; collapse of, 48–50,73n31, 95n16, 144 – 46, 149, 231

ADN (Bolivia, Nationalist DemocraticAction), 20, 22, 33, 155, 160, 320

agent of democratic representation. See principal-agent theory

Alarcón, Fabián, 128, 277Alfaro Ucero, Luis, 59– 60, 62Alianza Popular Revolucionaria Ameri-

cana. See APRAAlianza Social Indígena, 90Alva Castro, Luis, 55, 142Alvarez Paz, Oswaldo, 60, 144, 191American Popular Revolutionary Alli-

ance. See APRAAndean Community, 7Andean Parliament, 7, 66Andean region, defined, 1, 6 –11, 284n1anti-party politics, 51, 231. See also anti-

politics; anti-system actorsanti-politics, 63, 133, 135–36, 138, 144,

154. See also anti-party politics; anti-system actors

anti-system actors, 5, 15, 34, 100, 320; inBolivia, 155–56, 160 – 63; in Peru andVenezuela, 47, 51, 64, 71, 136, 146.See also anti-party politics; anti-politics

AP (Peru, Popular Action), 20, 49, 52–54, 71n7, 139, 143, 232, 242, 321

APRA (Peru, American Popular Revolu-tionary Alliance), 20, 52–53, 55–57,64, 73n32, 134, 139, 141, 142, 143,147, 163n10, 175, 186, 232, 243, 259,321

Argentina, 7, 29, 59, 134, 175, 257, 258,259, 305, 311, 331

Arias Cárdenas, Francisco, 73n31, 189,191

armed forces. See militaryassemblies, 1–5, 12, 14, 30, 32–35, 79,

188–96, 204 –22, 323. See also con-fidence in institutions; constituent assemblies; legislative representation;individual countries

assimilation, indigenous, 259, 260, 261,262

authoritarian regimes, 8, 9, 132, 257,308, 317; definition of, 164n12; inpower (Peru and Venezuela), 69, 138,142– 43, 148, 150 –52, 229; socialmovements and, 13, 34, 154, 237, 241,262, 287n40, 288; rise of (Peru andVenezuela), 30, 37n6, 47, 48, 51, 63–68, 249–51. See also dictatorships

autogolpe (self-coup). See Fujimori, Al-berto: coups

Aymara, 159, 276

Banzer, Hugo, 155, 157, 165n28, 271,285–86n19

INDEX

Barco, Virgilio, 191, 192Barrantes, Alfonso, 52, 54, 55, 57, 193,

242, 321Bedoya Reyes, Luis, 57, 193Belaúnde Terry, Fernando, 53, 57, 175,

193, 232Belmont, Ricardo, 71n8, 142, 193blacks, representation of, 96nn19,22, 264,

284n6, 336n38Blancos (Uruguay), 29Bolívar, Simón, 7Bolivarian revolution, 63, 153–54, 234,

251Bolivarian Revolutionary Movement 200,

62, 149Bolivia: confidence in institutions, 16,

17(tab), 184 –85, 218–21, 301, 303,305, 309; Congress, 158– 63, 180,213, 215, 324; constitution, 268, 271;coups, 310; decentralization, 70, 172,177, 183, 185, 186 –88, 269–70, 275,315; democracy in, 8, 320; economicperformance, 8, 10 –11, 132, 157, 296,298; elections, 22, 155–56, 158– 60,162, 164 – 65nn25,27,31, 180, 183,186, 189, 193–96, 198, 257, 269,275–76, 281–82, 310, 313, 315; insti-tutional reform, 157, 215, 218–21,268–72, 313, 323–24; outsiders, 9, 22,33, 132–33, 154 – 63, 198; party de-cay, 4, 22, 33, 319; party system, 154 –56, 160 – 63, 320; subnational repre-sentation, 177, 178, 180, 182, 186 –88,193–96, 198–99, 269–71, 275; turn-out, 24 –25, 180 –83, 198. See alsoindigenous representation; individualparties

Brazil, 64, 92, 117, 123, 175, 239, 259,272, 274, 311, 336n37

Bucaram, Abdalá, 64, 128n12, 136, 277business sector: in Peru, 67, 143, 148,

228, 235, 242, 248; in Venezuela, 136,143, 148, 150 –52, 153, 228, 233, 235,245, 247, 253n5

Caldera, Rafael, 22–23, 36n5, 38n22, 49,50, 56, 59– 61, 72nn13,15, 144, 146,231, 322

Cambio 90 (Peru, Change 90), 56, 64,72n21, 73n23, 142, 143

Cambio 90 –Nueva Mayoría (Peru,Change 90 –New Majority), 64,72n21, 73n23, 143

Caracazo, 144, 245, 246Cárdenas, Víctor Hugo, 275, 280Cartagena Agreement, 7Catholic Church, 237, 247– 48, 327; in

Peru, 233, 235, 242, 243, 244, 247–49; in Venezuela, 150, 152, 235, 247

Change 90. See Cambio 90Change 90 –New Majority. See Cambio

90 –Nueva MayoríaChapare, 156, 158, 165n27Chávez Frías, Hugo: base, 11, 67, 138,

145– 46, 227; consolidation of rule,33, 149–54; 234; democracy and,164n18; coups (1992), 8, 23, 59, 60,132, 144 – 46, 150, 231, 246; coups(2002), 68, 69, 132, 146, 151, 164n17,246, 257, 310; economic policy, 136,154; media and, 72n18; opposition to,67– 68, 69, 73n29, 132, 136, 150 –54,164nn21,22, 227, 230, 234 –35, 247;presidency, 64, 66 – 69, 141, 145– 47,73n28, 232, 239; recall referendum,68, 152–54; rise, 2, 7, 8, 9, 23, 38n22,49, 51, 56 – 63, 64, 132–33, 140, 189,191, 230, 251. See also under military

checks and balances, 47– 48, 63, 69, 147, 148, 172. See also horizontal accountability

Chile, 7, 10, 28, 49, 53, 81, 123, 253n3,259, 331

Christian Union (Colombia), 90CIDOB (Bolivia, Indigenous Confedera-

tion of the East, Chaco, and the Ama-zon), 262, 275, 280, 281–82

citizen discontent. See disaffectioncitizen movements, 227–31, 238, 240,

244, 251; failure of, 227, 230, 237,239, 241; future of, 252–53; urbancontext for, 240 – 47; urban organiz-ing, 242– 46. See also civil society; col-lective action; indigenous movements;popular movements; social movements

citizen security. See securitycitizenship, 230, 236, 249–50; expansion

of, 163, 241, 312–17, 328–29, 337n48Citizenship Movement for a New Coun-

try (Ecuador), 277

348 Index

Civic Solidarity Unity. See UCScivil society, 5, 30, 228, 236, 258, 330;

in Bolivia, 177, 178; in Peru, 228, 230, 235, 238, 244, 248, 249, 251–52; in Venezuela, 48, 228, 230, 233,235, 238, 244 – 46, 249, 251–52. Seealso citizen movements; grassroots; indigenous movements; mobilization;popular movements; protest; socialmovements

clientelism, 12, 27, 29–30, 94n11, 103,238, 297, 313; in Bolivia, 155, 276; inColombia, 10, 29, 81, 88, 96nn20,25,216, 316, 324; in Ecuador, 101, 103,108, 109, 116 –17, 129n20; in indige-nous politics, 268, 274, 276, 279; inPeru, 64, 67; in Venezuela, 58, 64, 67.See also patronage

closed list, 82, 89, 94n12, 111, 208, 213,324. See also PR

CNE (Colombia, National ElectoralCouncil), 82–84, 87, 95n17

coca, 7, 281, 287n34, 316; eradication of, 157, 158; growers, 33, 157–59,161, 262, 275, 281

collective action, 13, 64, 69, 70, 228,229, 233, 236, 239, 242. See also civilsociety; citizen movements; indigenousmovements; indigenous representation;mobilization; popular movements;popular protests; social movements

Colombia: confidence in institutions, 16,17(tab), 21, 184, 218–21, 303, 305,309; Congress, 79–80, 85, 87–89, 91–93, 191–92, 213, 215–18, 222, 223n4,320; constitution of 1991, 83, 85, 94n6,95nn13,14,16,18, 96n25, 192, 217–19,268, 271, 272, 316, 321, 325; decen-tralization, 70, 172, 176, 178, 180,183, 185, 186 –88, 269–70, 315–16;democracy in, 8–9, 36n4; economicperformance, 8, 9–11, 298; elections,86 –87, 89, 91, 94n2, 176, 180, 183,186, 189, 191–92, 194 –96, 223n4,315–16; electoral system, 78–93, 324;institutional reform, 85, 87–88, 213,215, 216 –21, 222, 268–72, 325; out-siders, 9, 22–23, 38n22; party systemdecline, 33, 49, 78–93, 132, 319–21,325; subnational representation, 79,

87, 91–92, 94n5, 176, 180, 186 –88,191–92, 194 –96, 198–99; turnout,24 –25, 179, 180, 198. See also indige-nous representation; individual parties

Colombia Yes Movement, 36Colorados (Uruguay), 29Committee of Independent Political

Electoral Organization. See COPEIComité de Organización Política Elec-

toral Independiente. See COPEICommunist Party (Peru), 54Communist Party (Venezuela), 144CONAIE (Ecuador, Confederation of

Indigenous Nationalities of Ecuador),34, 262, 276 –82

Conciencia de Patria. See CONDEPACONDEPA (Bolivia, Conscience of the

Fatherland), 154 –56, 287n33, 320Confederación de Nacionalidades Indíge-

nas del Ecuador. See CONAIEConfederación Indígena del Oriente,

Chaco, y Amazonía de Bolivia. SeeCIDOB

Confederación Sindical Unica de Traba-jadores Campesinos de Bolivia. SeeCSUTCB

Confederation of Indigenous Nationali-ties of Ecuador. See CONAIE

Confederation of Venezuelan Workers(CTV), 58

confianza. See confidence in institutionsconfidence in institutions, 14, 16 –18, 21,

38n17, 268, 297, 312, 327; assemblies,5, 34, 184, 204, 218–21; in Bolivia, 16,17(tab), 184 –85, 218–21, 301, 303,305, 309; in Colombia, 16, 17(tab), 21,184, 218–21, 303, 305, 309; conceptof, 333n1; in Ecuador, 17(tab), 184,218–21, 301, 303, 305, 309, 310; in-terpersonal trust and, 326 –28; judici-ary, 309–10; military, 184, 309–10;parties, 3, 9, 21, 34, 184, 137– 40; inPeru, 17(tab), 21, 184 –85, 218–21,305, 309; parties and assemblies, 9, 21, 34, 295–97, 300 –306, 308, 312,317, 326 –32; state performance and,296, 300 –301, 305, 307, 309, 326 –27, 329, 331–32; in Venezuela, 16,17(tab), 21, 145, 184, 218–21, 305,309, 310. See also state performance

Index 349

Congress. See assemblies; confidence ininstitutions; individual countries

Conscience of the Fatherland. SeeCONDEPA

Conservative Party (Great Britain), 78Conservative Party (PC, Colombia), 20,

23, 29, 33, 38n20, 78–81, 87, 93,94n2, 184, 186, 215, 218, 320 –21

Conservative Union (Colombia), 78constituent assemblies: in Bolivia, 160; in

Colombia, 85, 87, 176, 272, 321; inEcuador, 104, 277; in Peru, 64, 141,143, 175, 193, 313; in Venezuela, 64,72n18, 141, 145– 46, 232

constitutional assemblies. See constituentassemblies

constitutions. See individual countriesconstructivism, 4, 265, 297, 310, 335n26Convergencia Nacional (Venezuela, Na-

tional Convergence), 49, 60 – 61, 70,144, 145

Coordinadora Democrática, 152, 153,247

Coordinadora de Movimientos Sociales(Coordinator of Social Movements),277

Coordinator of Social Movements(Ecuador), 277

COPEI (Venezuela, Committee of Inde-pendent Political Electoral Organiza-tion), 22–23, 59– 62, 139, 149, 186 –88, 322; collapse of, 20, 23, 48– 49,95n16, 140, 144, 145, 231; internalpolitics 36n5, 38n22, 50, 72n13

COPRE (Presidential Commission forthe Reform of the State), 57, 59, 144 –45

corporatism: in Bolivia, 160, 259– 61; in Colombia, 90, 91; in Ecuador, 101,103, 108–9, 113, 115, 125, 129n20;effect on indigenous representation,259– 61, 285n8; in Peru, 260 – 61; inVenezuela, 10, 81

corruption, 3, 135, 171, 184, 204, 312,330, 332; in Bolivia, 155, 298–304; inColombia, 90, 216, 223n6, 298–304;as context for indigenous movements,258, 279, 283, 287n38; in Ecuador,298–304; and low trust in institutions,296, 298–304, 299(tab), 302(tab), 328;

in Peru, 140, 143, 148– 49, 186, 228–29, 231, 235, 249, 298–304; of police309–10, 328; in Venezuela, 58, 69,140, 144, 186, 228, 231, 298–304

Costa Rica, 8, 10, 81, 123, 184, 286n22coups: 1, 8, 23. See also individual countriesCPIB (Central for Indigenous Peoples of

the Beni), 271crime, 299–302, 306, 310, 311, 327,

328, 330, 331, 332. See also security;violence

crisis of democratic representation: causesof, 3 – 4, 16 –35, 50 –51, 93, 133, 162,172, 197, 283, 295, 296, 297, 302,305, 307, 312–13, 317, 320, 322–25,329; consequences of, 4, 68–71, 133,230 –31, 252–53, 325; definition, 5,14 –16

CSUTCB (Unified Peasant WorkersTrade Union Confederation of Bo-livia), 262, 275–76, 280, 281–82

CTV (Confederation of VenezuelanWorkers), 58

debt crisis, 57, 298, 331decentralization, 31, 32, 70, 171–99,

315, 325; in Bolivia, 70, 172, 177, 183, 185, 186 –88, 269–70, 275, 315;in Colombia, 70, 172, 176, 178, 180,183, 185, 186 –88, 269–70, 315–16;definition, 174; in Ecuador, 70, 172,177–78, 182–83, 186 –88, 269–70;effect on politicians’ careers, 173–74,188–96; electoral participation, 173,179–83; indigenous representationand, 177, 269–72; parties and, 173,183–88; in Peru, 172, 175–76, 178–79, 182–83, 185, 186 –88, 315; inVenezuela, 57, 59, 60, 145, 172, 176,178, 182–83, 230, 233, 239, 252, 316.See also subnational representation

Democracia Popular. See DPdemocracy. See satisfaction with democ-

racy; support for democracyDemocratic Action. See ADdemocratic breakdown, 1–2, 4, 6, 8, 141,

142Democratic Coordinator (Venezuela),

152, 153, 247Democratic Front. See FREDEMO

350 Index

Democratic Left. See IDDemocratic Left Movement (MID), 143Democratic Party (U.S.), 78democratic representation, definition of,

11–13delegative democracy, 132, 148, 164n11dictatorships, 8, 9, 132, 138, 147– 49,

164n11, 235, 244, 247, 257, 313. Seealso authoritarian regimes

direct democracy, 73n29, 133, 325, 330,337n49

direct election, 57, 70, 145, 176, 178,191, 196, 233, 312, 315–16

disaffection (with democracy), 1– 6, 10, 11, 15–16, 22, 23, 220(fig), 295,304, 313, 322–23, 326, 331, 336 –37nn47,48; in Bolivia, 159; in Colom-bia, 93, 221; in Ecuador, 100, 221; inPeru, 50, 139, 221, 229; in Venezuela,50, 52(tab), 70, 144, 139, 221. See alsosatisfaction with democracy; supportfor democracy

disempowerment, 227, 236, 239, 244,248, 249–53; concept of, 237. See alsoempowerment

dissatisfaction with democracy. See disaf-fection; satisfaction with democracy;support for democracy

DP (Ecuador, Popular Democracy), 101,122, 127n1, 128n5

drug trafficking, 9, 11, 95n15, 132, 303,331

dual representation, 80, 94n5

economic performance, 3, 8, 9–11, 31, 47, 70, 298, 300 –301, 331, 332,333n8. See also individual countries

economic reform. See market reformEcuador: coalitions, 111, 113–14, 116,

128n16; confidence in institutions,17(tab), 184, 218–21, 301, 303, 305,309, 310; Congress, 102, 104 –18,121–25, 128n11, 277–78, 324; consti-tutions, 104, 111, 112, 115; 177–78,219, 268, 271, 272, 277; coups, 1, 8,23, 129n23, 257, 277, 280, 303, 310,334n21; decentralization, 70, 172,177–78, 182–83, 186 –88, 269–70;democracy in, 8; economic perfor-mance, 8, 10 –11, 102, 127n2, 178,

296, 298; elections, 104 – 6, 109–20,127n1, 128nn5,6,14,16, 129nn21,22,23,132, 182, 183, 186, 189, 276 –77, 281, 282, 310, 313, 316; electoral sys-tem, 100 –127, 323–24; institutionalreform,104 – 6, 105(tab), 111–20,128nn13,15, 129n18, 213, 218–21,268–72, 323–24; outsiders, 9, 22–23,33, 38n22, 163n1; party system frag-mentation, 101, 104, 106, 108, 109,111, 113, 116, 119, 123, 124, 319;provinces and provincialization, 102,107–14; subnational representation,101, 109, 110, 112, 113, 120 –26,177–78, 182–83, 186 –88, 194 –99,269–71; turnout, 24 –25, 182–83,198. See also indigenous representation;individual parties

Ecuadorian Roldosist Party. See PREeducation, 80, 253, 296, 300, 301, 328,

331; bicultural, 258, 260, 262, 274,275; decentralization and, 176, 185;voters’ levels of, 14, 305, 313–14, 317

Education, Work, and Social ChangeMovement (Colombia), 90

effective number of parties, 206, 209–11,213–15, 219, 318(fig), 319, 320,336n40

elections. See electoral rules; electoral systems; electoral volatility; runoffelections; turnout; individual countries

electoral accountability, 12, 13–14, 31,37nn12,14, 82, 119, 178, 297, 322–36;personal, 322–24, 336n41

electoral micro-enterprises, 79–80, 85,90, 91

electoral participation. See turnoutelectoral reform. See institutional reformelectoral rules, 3, 127n4, 230 –31, 240,

279; in Bolivia, 161, 208; in Colom-bia, 97n27, 208; in Ecuador, 104, 109,182, 208; in Peru, 138, 208, 238, 250;in Venezuela, 138, 208, 238. See alsodecentralization; electoral systems

electoral systems, 12, 73n28, 79, 81–82,119, 137; accountability and, 297,322–26; institutional reform and, 205, 210 –12, 214, 220, 222; protestand, 238. See also closed list; legisla-tive representation; mixed formula;

Index 351

electoral systems (continued )nomination rules; open list; personal-ized voting; PR; preference voting;single-member districts; turnout; indi-vidual countries

electoral volatility, 3– 4, 9, 18–20, 50,172, 197; in Bolivia, Peru, Venezuela,9, 19; in Colombia, 9, 19, 94n9; inEcuador, 9, 19, 101, 106, 116

electorate, expansion of. See citizenshipemployment, 10 –11, 29, 328, 330, 331.

See also unemploymentempowerment, 227, 240, 241, 244, 248;

concept of, 236 –37; future of, 249–53; institutionalization of, 252; rep-resentation and, 230, 236 – 40, 241; urban space and, 240, 241

England. See Great Britainevangelicals, 244, 248EVV (Venezuela, Escuela de Vecinos de

Venezuela), 245

FACUR (Venezuela, Federation of UrbanCommunity Associations), 244 – 45

Fatherland for All, 61, 149Febres Cordero, León, 189, 200n11Federation of Urban Community Associ-

ations, 244 – 45Fermín, Claudio, 36, 59, 62, 63, 144, 191Fernández, Eduardo, 60, 72n13Fernández, Max, 133, 154, 163n3,

164n24Fifth Republic Movement (MVR), 69,

149FRA (Ecuador, Radical Alfarista Front),

128nn7,12FREDEMO (Peru, Democratic Front),

52–57, 71n7, 141– 42Frente Democrático. See FREDEMOFrente Nacional. See National FrontFrente Radical Alfarista, 128nn7,12Fujimori, Alberto: anti-party tendencies,

33, 38n22; base, 67, 138; corruptionand, 303; coups, 64, 141, 143, 148,175, 235, 257, 308, 334n22; economicpolicies, 135–36, 176; opposition to,67– 68, 69, 73n29, 142, 149, 183, 227,234 –35, 250; presidency, 64 – 69,73n28, 142– 43, 148– 49, 164n11, 176,178–79, 186, 229, 232, 239; reelection

law change, 65– 66, 143, 229, 232,235, 250; rise, 22, 48, 50, 53, 55–56,63– 64, 133, 139, 141– 42, 146 – 47,193

Fujimorism, 64, 67– 69, 73n23

García, Alan, 53–55, 57, 74, 140, 142,146, 147, 175, 186, 193, 200n14, 232,303

Gaviria, César, 96n21, 191, 192Germany, 126governability, 32, 71, 80, 141, 146; com-

parison of Peru and Venezuela, 139–54; crisis of, 132, 133, 136 – 40, 162–63; definition of, 163n5; in Colombia,80, 93; in Ecuador, 100, 122, 127; in Peru, 48, 52, 53, 146, 163n7; inVenezuela, 48, 146, 152. See also statedeficiencies

grassroots, 177, 233, 244, 248, 253n6Great Britain, 78, 82, 93n1, 191“great turnaround,” 57–59Guatemala, 4, 250, 257, 260 – 61,

288nn41,44; indigenous politics, 260,272, 281, 286n21

guerrillas, 197; in Bolivia, 276; inColombia, 9, 95n15, 132, 298, 306 –7,320, 331; in Peru, 141, 146, 298, 331

Gutiérrez, Lucio, 7, 23, 34, 38n22, 64,163n1, 197, 277–78, 281, 316, 334n21

Haiti, 300Hare formula, 86, 109, 110, 209Haya de la Torre, Víctor Raúl, 57, 136Herrera Campins, Luis, 50, 61Honduras, 79, 123horizontal accountability, 64, 140, 147–

48. See also checks and balanceshuman rights, 12, 148, 248, 250, 257;

movement, 235, 245, 246, 247, 248,250

ID (Ecuador, Democratic Left), 101, 111,127n1, 128n5, 129nn24,25, 200n21,321

ideological convergence. See program-matic representation

ideological representation. See program-matic representation

ILO Convention 169, 277, 286n22

352 Index

Independent Political Organizing Com-mittee. See COPEI

Indianista movement (Bolivia), 275Indigenous Authority Movement of

Colombia, 90Indigenous Confederation of the East,

Chaco, and the Amazon. See CIDOBindigenous movements: autonomy and,

272; in Bolivia, 133, 177, 156 – 63,261– 62, 274, 276 –78; developmentof, 258, 261– 62, 267, 274, 285n8; in Ecuador, 261– 62, 274, 276 –78;electoral success and, 278–83; in Peru, 261– 62, 285n10. See also citizenmovements; civil society; indigenousrepresentation; popular movements; so-cial movements; individual organizations

indigenous parties. See individual partiesindigenous representation, 34; 257–84;

autonomy and, 270 –73; backgroundfor, 259– 63, 267; concept of, 263– 66;in Bolivia, 70, 133, 156 – 63, 177, 196,267–72, 275, 316; in Colombia, 91,96n22, 268–72, 316; in Ecuador, 70,120, 125–26, 178, 267–72, 277, 316;indigenous interests and, 263– 65, 266,276, 278, 23, 284; “mirror” conceptand, 265, 266, 269, 275, 278, 281, 283;obstacles to, 273–74; party politicsand, 274 –83; in Peru, 147, 267–71;reform toward, 5, 32, 155– 63, 258,265, 266 –74, 283; in Venezuela, 268,316. See also indigenous movements

inflation, 50 –53, 58, 139, 140 – 44, 152,301, 331, 333n8

informal sector, 33, 139, 140, 154, 157institutionalism, 3, 4, 81, 135, 163n8,

150, 268institutionalization, 147, 337n48; of party

systems, 80 –81, 104, 111, 119, 127n4,155, 337n48

institutional reform, 3, 5, 31–32, 33–34,81, 240, 325; leading to authoritariangovernment, 47, 63; of legislatures,204 –5, 207, 212–22, 222n3. See alsodecentralization; indigenous represen-tation; individual countries

Institutional Revolutionary Party. See PRIinstitutional trust. See confidence in

institutions

interpersonal trust, 297, 301, 302(tab),303(tab), 304(tab), 326 –28

IS (Peru, Socialist Left), 54, 56, 142Italy, 19, 21, 35, 79, 335n30IU (Bolivia, United Left), 158IU (Peru, United Left), 20, 52–54, 56,

57, 142, 143, 232, 321–22Izquierda Democrática. See IDIzquierda Socialista, 54, 56, 142Izquierda Unida (Bolivia), 158Izquierda Unida (Peru). See IU (Peru)

Japan, 86, 89, 93n1, 96n20, 334n18,335n30

judiciary, 69, 238, 268, 298, 306, 310,311, 328, 329, 331; in Colombia, 298;confidence in, 309–10; in Peru, 69,143, 298; in Venezuela, 141, 152

Justicialist Party (Argentina), 29

Katarista movement (Bolivia), 275, 280

labor, 58, 151–52, 157, 259. See alsounions

Labour Party (Great Britain), 78La Causa R (LCR), 60 – 61, 73n31, 233land reform, 68, 260, 261, 262, 271Law of Popular Participation (Bolivia),

157, 158, 177, 185, 196, 270, 271,275, 315, 316

LCR (Venezuela, Radical Cause), 60 – 61,73n31, 233

left-wing guerrillas. See guerrillaslegislative representation: in Bolivia, 208,

209–12; in Colombia, 209–12; con-cept of, 205–7; in Ecuador, 209–12;interparty dimension of, 206 –8, 209–12, 214, 215, 218; intraparty dimen-sion of, 206 –8, 210 –16, 218, 220,222–23nn1,3, 324, 336n42; nationalvs. parochial, 204 –8, 212, 221, 222; in Peru, 209–12; in Venezuela, 208,209–12. See also institutional reform oflegislatures

legislatures. See assemblies; confidence ininstitutions; legislative representation

legitimacy, 5, 6, 15, 31, 35, 50, 57, 67,93, 155, 258

Ley de 15% (Ecuador), 177Ley de Elecciones Regionales (Peru), 269

Index 353

Liberal Party (Great Britain), 78Liberal Party (PL, Colombia), 23, 29, 33,

36, 38n20, 78–81, 87, 93, 94n2, 184,186, 191, 215, 218, 320 –21

Liberty Movement, 53, 54local level reform. See direct election;

institutional reform; subnational representation

M19 (Colombia), 321Macas, Luis, 277, 279, 280Mahuad, Jamil, 23, 129n23, 277, 280,

303, 334n21malapportionment, 108, 128n8market-oriented economy. See market

reformmarket reform, 10, 29, 53, 47, 48, 53,

57–58, 70, 63, 135, 143, 155, 331. See also neoliberal reform

MAS (Bolivia, Movement toward Social-ism), 158– 62, 165nn27,31, 275,287n32, 316, 320

MAS (Venezuela, Movement to Social-ism), 61, 144, 145, 149

MBR-200 (Venezuela, Bolivarian Revolu-tionary Movement 200), 62, 149

media, 4, 296 –97, 304, 311–12, 313,315, 322, 330; in Colombia, 96 –97n26; in Peru, 69, 142, 148– 49, 249, 250; in Venezuela, 72n18, 150,152–54, 235, 245, 253n5. See alsotelevision

Mexico, 58, 133, 175, 238, 240, 259,288n44, 305; indigenous representa-tion in, 260, 268, 272

MID (Peru, Democratic Left Movement),143

middle-class sector, 67, 68, 151, 160,229, 242, 245

military: confidence in, 184, 309(tab),309–10; democracy and, 12, 148, 298,310; Chávez’s use of, 147, 149–52,164nn16,19; in Colombia, 306; Fuji-mori’s coalition with, 66, 142– 43,147– 49, 164n14

military coups. See coupsmilitary dictatorships. See authoritarian

regimes; dictatorshipsMIP (Bolivia, Pachakuti Indigenous

Movement), 159, 160, 161, 276, 320

MIR (Bolivia, Revolutionary Left Move-ment), 20, 160, 320

mixed formula (mixed-member electoralsystems), 57, 145, 158, 206, 212, 213,215

ML (Peru, Liberty Movement), 53, 54MNR (Bolivia, Nationalist Revolutionary

Movement), 20, 134, 155, 159, 160,259, 260, 287n32, 320

mobilization, popular, 15, 34, 134 –35,227–31; disempowerment and, 249–53; in Bolivia, 33, 34, 157–58, 161–62, 258, 268; indigenous, 34, 120,161– 62, 261– 63, 274, 284; in Ecua-dor, 120, 182, 258, 268, 277; in Peru,227–29, 233–35, 239, 241– 44, 247–49, 250 –51; in Venezuela, 227–28,230, 234 –35, 249, 239, 244 – 47; pop-ulism and, 134 –35; religion and, 247–49; urban, 227–31, 235–36, 239. Seealso civil society; protest

Montesinos, Vladimiro, 68, 142, 148–49, 150, 235, 249

Morales, Evo, 275–76; as leader of cocagrowers, 134, 157, 275; as presidentialcandidate, 7, 22, 28, 34, 158, 159,165nn28,29, 281, 316

Movement to Socialism. See MAS (Venezuela)

Movement toward Socialism. See MAS(Bolivia)

Movimiento al Socialismo. See MASMovimiento Autoridades Indígenas

de Colombia (Indigenous AuthorityMovement of Colombia), 90

Movimiento Bolivariano Revolucionario-200, 62, 149

Movimiento de Ciudadanía por unNuevo País, 277

Movimiento de Izquierda Revolucionario(MIR), 20, 160, 320

Movimiento de la Izquierda Democrática(MID), 143

Movimiento de Libertad (ML), 53, 54Movimiento de Unidad Plurinacional

Pachakutik Nuevo País (MUPP-NP),276 –77

Movimiento Educación, Trabajo y Cam-bio Social (Education, Work, and So-cial Change Movement), 90

354 Index

Movimiento Indígena Pachakuti. SeeMIP

Movimiento Nacionalista Revolucionario.See MNR

Movimiento Pachakutik, 128n7. See alsoMUPP-NP, Pachakutik

Movimiento por la Seguridad Social (Social Security Movement), 90

Movimiento Quinta República (MVR),67, 149

Movimiento Revolucionario Liberal. SeeMRL

Movimiento Revolucionario TupacAmaru. See MRTA

Movimiento Sí Colombia (Colombia YesMovement), 36

MRL (Colombia, Liberal RevolutionaryMovement, 78

MRTA (Peru, Tupac Amaru Revolution-ary Movement), 48, 250, 267

multipartism, 61, 106 –26MUPP-NP (Ecuador, Pachakutik

Plurinational Unity Movement-New Country), 276 –77

MVR (Venezuela, Fifth Republic Move-ment), 67, 149

National Christian Party (Colombia), 90National Congress. See assembliesNational Convergence. See Convergencia

NacionalNational Electoral Council (Colombia,

CNE), 82–84, 87, 95n17National Front (Colombia), 23, 36n4,

38n20, 70, 95n14, 217, 320Nationalist Democratic Action. See

ADNNationalist Revolutionary Movement.

See MNRneighborhood associations, 13, 177, 244,

246neoliberal reform, 55, 58– 60, 63, 67,

81, 135–36, 143, 144, 331; protestagainst, 55, 58, 262, 275. See also mar-ket reform

neopopulism, 33, 47, 132– 63; theory of,133–37. See also personalism; plebisci-tarian representation; populism

new parties, 5, 19–23, 47, 49–50, 61,69, 84, 106, 183, 185, 186, 321

New Republican Force (NFR), 164n25,194

NFR (Bolivia, New Republican Force),164n25, 194

NGOs (non-governmental organizations),13, 152, 261, 235, 238, 239, 241, 243,244, 245, 252

Nicaragua, 272, 273nomination rules, 94n4, 189–90, 195,

206, 212–13, 238– 40; accountabilityand, 297, 323–26; in Bolivia, 208,213; in Colombia, 82, 83–87, 94n12,208; in Ecuador, 113, 208, 213; in-digenous representation and, 286n23;in Peru, 208; in Venezuela, 208, 219

nomination systems. See nomination rulesNueva Fuerza Republicana (NFR),

164n25, 194

OAS (Organization of American States),68, 73nn24,26, 143, 153, 235

oil, 58, 68, 140, 144, 150, 152, 154, 243open list, 85, 108, 111, 117–20, 208,

213. See also PROPIP (Ecuador, Organization of Indige-

nous Peoples from Pastaza), 271outsiders, political, 1, 3, 4, 9, 15, 21–25,

33, 172, 197; in Bolivia, 9, 22, 33,132–33, 154 – 63, 198; in Colombia,9, 22–23, 38n22; definition of, 21,38n22, 133; in Ecuador, 9, 22–23, 33,38n22, 163n1; indigenous movementsand, 154 – 63; ingovernability and,132–33, 135–36, 137– 40; party sys-tem collapse and, 141– 47; in Peru, 4,9, 22–23, 38n22, 49–51, 53–56, 69,132–33, 137– 44, 146 – 49, 162– 63;regimes of, 147–54, 162– 63; in Vene-zuela, 4, 9, 22–23, 38n22, 49–51, 53–56, 69, 132–33, 137– 40, 141, 144 –54, 162– 63. See also neopopulism;populism

Pacari, Nina, 277, 279Pachakuti Indigenous Movement. See

MIPPachakutik, 34, 120, 126; in elections,

128n7, 129n23, 277–78, 316, 321; re-lationship with CONAIE and, 281–282, 287n38

Index 355

Pachakutik Plurinational Unity Movement-New Country (MUPP-NP), 276 –77

Palenque, Carlos, 133, 154, 163n3Paraguay, 300paramilitary forces, 9, 95n15, 298, 306,

307particularism, 204, 206, 207, 216, 217,

329; definition of, 79; in Bolivia, 160;in Colombia, 79, 80, 81–89, 91, 92,93, 94nn3,5, 216; in Ecuador, 103,104, 109, 112, 125–26, 127, 129n20;in Venezuela, 59

Partido Liberal. See Liberal Party(Colombia)

Partido Liberal Radical (Ecuador), 128n7Partido Nacional Cristiano, 90Partido Popular Cristiano. See PPCPartido Revolucionario Institucional.

See PRIPartido Roldosista Ecuatoriano. See PREPartido Social Cristiano. See PSCPartido Socialista Ecuatoriano, 128parties, political, 1– 6, 9, 10, 12–35,

38n31; absence of, 69–71; decentral-ization and, 183–88. See also new par-ties; party decline; party options; partysystem collapse; polarization; individualcountries; individual parties; and underconfidence in institutions

partyarchy, 139, 140, 144, 322, 323party competition, 4, 27, 282, 297, 310 –

12, 322, 330party decline, 2, 14, 20, 32, 47, 49, 79–

81, 173, 189, 195, 257. See also individ-ual countries

party-line voting, 119, 207, 208, 213party options, 317–22party system collapse: in region, 1, 20 –

21, 33, 59, 133; comparison of Peruand Venezuela, 47–51, 137, 141– 47;consequences of, 69–71. See also indi-vidual countries

party system institutionalization. Seeinstitutionalization

party system polarization. See polarizationPastrana, Andrés, 84, 191–92Patria Para Todos (PPT), 61, 49patronage, 117, 212, 233, 274, 280,

287n32. See also clientelism

Paz Estensoro, Victor, 58, 59, 136PC. See Conservative Party (Colombia)PCP (Peruvian Communist Party), 54peasant federations and unions, 157–58,

161, 177, 259– 60Pease García, Henry, 54, 193Pérez, Carlos Andrés, 8, 50, 55– 60,

144 – 45, 190, 303Pérez de Cuéllar, Javier, 64, 193, 232Pérez Jiménez, Marcos, 244Peronism, 133, 136, 259personal (personalized) voting, 57, 213,

216, 324, 336n42; with open lists, 104,108, 111, 117–20. See also preferencevoting

personalism, 29, 30, 94n11, 134 –35, 208,211(fig), 297, 313–14, 320, 329; inBolivia, 154 – 45, 274; in Colombia,78–79, 81–91, 92, 94n11, 212, 216,218; in Ecuador, 101, 104, 111, 115,119, 213, 274; in Peru, 47, 63, 68,139, 218, 251, 321; in Venezuela, 47,63, 68, 139, 151, 232, 251. See alsoneopopulism; plebiscitarian representa-tion; populism

personalistic representation. Seepersonalism

Peru: after Fujimori, 69–71, 250; con-fidence in institutions, 17(tab), 21,184 –85, 218–21, 305, 309; Congress,56, 64, 69, 142, 213, 232; constitution,64, 65, 73n29, 143, 219, 248, 268,271; coups, 64, 141, 143, 148, 175,235, 257, 308, 334n22; decentraliza-tion, 172, 175–76, 178–79, 182–83,185, 186 –88, 315; democracy in, 8,47, 48, 229; economic performance, 8,10 –11, 48, 50, 51, 53, 140, 186, 229,242; elections in, 22, 48– 49, 52–56,64 – 69, 71n6, 72n21, 73n23, 137,141– 44, 146 – 47, 175–76, 182–83,186, 189, 192–95, 228, 229, 232, 235,241– 43, 248, 250, 313, 315; institu-tional reform, 63– 68, 213, 218–21,239, 268–71; new parties, 20; out-siders, 4, 9, 22–23, 38n22, 49–51, 53–56, 69, 132–33, 137– 44, 146 – 49,162– 63; party system collapse, 1, 4,21, 33, 47–56, 57, 69–71, 140, 141–47, 228, 231, 232, 238, 251, 319, 321–

356 Index

22; subnational representation, 23,175–76, 182–83, 186 –88, 192–99;turnout, 24 –25, 182–83, 198. See alsoFujimori, Alberto; individual parties

Perú Possible (Possible Peru), 70, 325Peruvian Communist Party (PCP), 54Philippines, 89PL. See Liberal Party (Colombia)plebiscitarian representation, 30, 31, 132,

138; in Bolivia, 154 –55; in Peru andVenezuela, 47, 63, 64, 69, 147, 149,154, 162. See also neopopulism; per-sonalism; populism

polarization, party system, 28, 209–10,214(tab); in Bolivia, 155–56, 160 – 63,320, 335n28; in Peru, 49, 69–70, 142;in U.S., 335n28; in Venezuela, 67– 68,69–70, 132, 150, 152, 153, 164n17,246, 252, 335n28

police, 298, 311; confidence in, 184,309–10, 328; deficiency in, 306, 329;impunity of, 240, 245. See also corrup-tion; violence

political outsiders. See outsiderspolitical reform. See institutional reformpoliticians: as agents of democratic repre-

sentation, 5, 6, 12–14, 27, 32, 38n31;career paths of, 188–96

Popular Action. See APPopular Christian Party. See PPCPopular Democracy. See DPpopular election. See direct electionpopular movements, 11, 12, 34, 228, 229.

See also citizen movements; civil soci-ety, indigenous movements; socialmovements

popular sectors, 33, 240; in Bolivia, 154;in Peru and Venezuela, 47, 59, 63– 64,67, 68

Popular Unity Front (UP), 53populism, 29, 30, 64, 134 –36, 139,

163n4, 179, 260, 321; in Bolivia,164n25, 260, 287n33; in Peru andVenezuela, 54, 59, 60, 63, 67, 68, 135,136, 139, 163n4, 234, 239, 251. Seealso neopopulism; outsiders; personal-ism; plebiscitarian representation

pork barrel projects, 89, 92–93, 96n25,207–9, 212, 216. See also clientelism;particularism; patronage

Possible Peru, 70, 325poverty, 10 –11, 81, 300, 311, 330; in

Bolivia, 156, 157; in Ecuador, 127n2;in Peru, 140, 248; in Venezuela, 140,154, 240

PPC (Peru, Popular Christian Party), 20,52–54, 57, 71n4, 143, 232, 321

PPT (Venezuela, Fatherland for All), 61,149

PR (proportional representation), 82,213, 325; in Bolivia and Venezuela,324 –25; in Ecuador, 105(tab), 106,108–11, 113, 117, 119. See also closedlist; open list; preference voting; sub-party lists

PRE (Ecuadorian Roldosist Party), 101,122, 127n1, 128n5, 200n21

preference voting, 119, 213, 324, 325.See also open list; personal voting

Presidential Commission for the Reformof the State. See COPRE

PRI (Mexico, Institutional RevolutionaryParty), 58, 133, 238, 259

principal-agent theory, 11–14, 15, 30 –31, 37nn8,16, 263– 66, 295

programmatic representation (conver-gence), 2, 12, 27–30, 37n11, 38n24,320, 332

proportional personalized vote. See mixedformula

proportional representation. See PRprotest, popular, 8, 33, 238, 239, 253; in

Bolivia, 11, 160, 161, 262; in Ecuador,11, 262, 276; indigenous, 262, 274,278; against neoliberalism, 55, 58; inPeru, 227–29, 235, 241, 243– 44, 250;in Venezuela, 8, 58, 227–29, 234 –35,244, 246 – 48, 251, 254n11. See alsomobilization

PSC (Ecuador, Social Christian Party),101, 112, 122, 127n1, 129nn24,25,200nn11,21, 321

PT (Brazil, Workers’ Party), 239Punto Fijo Agreement, 36n4, 63, 140,

164n17

Quispe, Felipe, 159, 275, 276

Radical Alfarista Front, 128nn7,12Radical Cause (LCR), 60 – 61, 73n31, 233

Index 357

rationalist, 310reform. See institutional reformRepublican Party (U.S.), 78Revolutionary Left Movement (MIR), 20,

160, 320runoff elections, 115–17, 146 – 47, 214

Sáez, Irene, 61, 63Salas Römer, Henrique, 49, 61, 62, 145,

191Samper, Ernesto, 191–92, 303Sánchez de Lozada, Gonzalo, 4, 155, 159,

161, 188, 275, 286 –87n19, 306, 315Sanín, Noemí, 36n4, 192satisfaction with democracy, 1– 6, 15, 35,

93, 100, 102, 331; electoral participa-tion and, 23; electoral volatility and,18–19; empirical indicators of, 93,220(fig), 221; outsiders and, 21, 47, 63; poverty and, 10, 139. See also con-fidence in institutions; disaffection;support for democracy

security, 3, 238, 242, 296, 298, 306, 311,328, 331. See also crime; violence

semi-democracy, 1, 4, 8, 21, 22, 30, 132,148, 162, 317

Sendero Luminoso. See Shining PathSerpa Uribe, Horacio, 23, 192Shining Path (Peru, Sendero Luminoso),

48, 52, 53, 140, 141, 229, 232, 233,242– 44, 250, 267, 298, 307

shirking, 13–14single-member districts, 73n28, 324,

336n42Social Christian Party. See PSCSocial Indigenous Alliance (Colombia),

90Socialist Left (IS), 54, 56, 142social movements, 11, 13, 165n32, 239,

252, 258, 260, 278; in Bolivia, 156 –61; indigenous representation and,262, 265, 274, 278–79, 282, 283–84,269; in Peru, 48, 242, 249; lifecycle of,239– 41; political representation and,231, 236 –37. See also citizen move-ments; civil society; indigenous move-ments; popular movements

Social Security Movement (Colombia),90

Spain, 16, 27, 126, 192

state, 9, 31, 108, 136, 138, 163; capacity,12, 302, 328, 332, 337n48; collapse of,306 –7; institutions of, 298, 306. Seealso state deficiencies

state deficiencies, 3, 6, 35, 162, 296 –303,305–9, 322, 325, 328–30, 332; con-cept of, 296, 305–7. See also corrup-tion; governability

state performance, 35, 296 –98, 330; citi-zen evaluation of, 15, 16, 37n16, 296,305, 327; confidence and, 296, 300 –301, 305, 307, 309, 326 –27, 329,331–32; crisis of democratic represen-tation and, 307, 325, 329, 331–32;empirical indicators of, 300 –301,326 –27; outsiders and, 162. See alsoconfidence in institutions

statism, 47, 48, 58, 134, 136, 150, 163n4structural adjustment, 135, 242– 43subnational representation, 5, 31, 34, 87,

172–99. See also decentralization; di-rect election; individual countries

sub-party lists, 208, 209, 213, 215, 216,218

suffrage, 12, 267– 68, 283supply-side oligopoly, 19, 297, 312, 313,

317–20, 322, 325, 329support for democracy, 9(tab), 36 –37n6,

93, 207, 216, 284n2, 307–8. See alsoconfidence in institutions; disaffection;satisfaction with democracy

Switzerland, 126

television, 234, 311–12, 315, 335n29. See also media

terrorism, 48, 63, 132, 141, 146, 242–44, 248

Toledo, Alejandro, 69, 197, 232, 235, 251truncated representation, 297, 312–13,

317, 322, 325, 328–29, 336n38trust in institutions, parties, etc. See con-

fidence in institutionsTruth and Reconciliation Commission

(Peru), 250, 252turnout, 5, 23–25, 172, 173, 179–84,

196 –98, 200nn13,14,15, 223n4. Seealso individual countries

UCS (Bolivia, Civic Solidarity Unity),154, 155, 156, 320

358 Index

unemployment, 139, 298, 301; in Bolivia,157, in Peru, 242, in Venezuela, 140,152

Unidad Cívica Solidaridad. See UCSUnidad Popular (UP), 53Unified Peasant Workers Trade Union

Confederation of Bolivia. SeeCSUTCB

Unión Cristiana, 90Union for Peru (Unión por el Perú,

UPP), 64unions, 11, 80, 139, 184, 260, 261; in

Bolivia, 157, 158, 161, 177; in Colom-bia, 96n20; in Peru, 140, 228, 235,241, 242; in Venezuela, 150, 152, 228,231, 233–35, 241, 247, 253n5. See alsolabor

United Left (Bolivia), 158United Left (Peru). See IU (Peru)United States, 78, 83, 236, 264; com-

parison to, 22, 37n12, 78, 82, 96n25,126, 179–80, 183, 184, 284n6, 312,334n16, 335n28; interests, 11, 157;policy, 158, 165nn28,29, 275

UNO (Peru, National Odriísta Union),321

UP (Chile, Popular Unity Front), 53UPP (Union for Peru), 64urbanization, 229, 240, 313, 315urban poor, 5, 34, 237, 243, 248Uribe Vélez, Álvaro, 23, 36, 36n5,

38n22, 189, 190, 192, 223nn4,6Uruguay, 10, 29, 78, 79, 81, 93n1, 123,

331

Vargas, Antonio, 279–80, 282Vargas Llosa, Mario, 20, 53–56, 63,

71nn4,7,8, 142, 146 – 47, 193, 232Velasco Alvarado, Juan, 152, 260Velásquez, Andrés, 61, 191Venezuela: challenges, 69–71; Congress,

57, 64, 66, 68, 141, 144 – 46, 176,190 –91, 215, 219, 324; confidence ininstitutions, 16, 17(tab), 21, 145, 184,

218–21, 305, 309, 310; constitution,64, 66, 73n29, 212, 219, 221, 245– 46,268, 316; coups (1992), 8, 23, 59, 60,132, 144 – 46, 150, 231, 246; coups(2002), 68, 69, 132, 146, 151, 164n17,246, 257, 310; decentralization, 57, 59,60, 145, 172, 176, 178, 182–83, 230,233, 239, 252, 316; democracy in, 2,4, 8–9, 36n4, 47– 48, 229; economicperformance, 8, 9–11, 48, 51(tab), 57–59, 61, 140, 144, 229, 233, 296, 298;elections, 22, 48– 49, 56 –57, 59– 63,66 – 67, 69, 72n13, 73n28, 137, 144 –46, 149–54, 182, 183, 186 –88, 189,190 –91, 194 –95, 215, 230, 231–32,234, 245, 310, 316, 318–19, 322; gen-eral (civic) strike, 68, 132, 152–53,228, 234, 237, 251; institutional re-form, 57, 63– 68, 144 – 45, 212–13,215, 218–21, 233–34, 239, 245–54,252, 268–72, 323–24; new parties,20; outsiders in, 4, 9, 22–23, 38n22,49–51, 53–56, 69, 132–33, 137– 40,141, 144 –54, 162– 63; party systemcollapse, 1, 4, 21, 33, 47–51, 56 – 63,68, 69, 79, 95n16, 132, 137, 141– 47,186, 228, 231–32, 238, 251, 319, 322;subnational representation, 176, 182–83, 186 –88, 190 –91, 194 –99, 271;support for democracy in, 36 –37n6,turnout, 24 –25, 179, 182–83, 198.See also Chávez Frías, Hugo; individualparties

violence: police and, 309–10; political, in Colombia, 132; in Peru, 50, 53,51(tab), 141, 229, 233, 238, 244, 245,253n2; in Venezuela, 69, 251. See alsocrime; security

vote pooling, 82, 86, 89, 94 –95n12, 206,208, 215

Western Europe, 17, 19, 21, 83. See alsoindividual countries

Workers’ Party (Brazil, PT), 239

Index 359

Documents

0804752788 Crisis of Demo