Politicians and Politics in Latin America 9781626371545

The premise of this book is, simply, that politicians matter—that an understanding of the role played by politicians in

196 50 2MB

English Pages 323 [337] Year 2007

Report DMCA / Copyright

DOWNLOAD PDF FILE

Recommend Papers

Politicians and Politics in Latin America
 9781626371545

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

A-FM

10/2/07

3:32 PM

Page i

Politicians and Politics in Latin America

A-FM

10/2/07

3:32 PM

Page ii

A-FM

10/2/07

3:32 PM

Page iii

Politicians and Politics in Latin America edited by Manuel Alcántara Sáez

b o u l d e r l o n d o n

A-FM

10/2/07

3:32 PM

Page iv

Published in the United States of America in 2008 by Lynne Rienner Publishers, Inc. 1800 30th Street, Boulder, Colorado 80301 www.rienner.com and in the United Kingdom by Lynne Rienner Publishers, Inc. 3 Henrietta Street, Covent Garden, London WC2E 8LU © 2008 by Lynne Rienner Publishers, Inc. All rights reserved Library of Congress Cataloging-in-Publication Data Politicians and politics in Latin America / Manuel Alcántara Sáez, editor. p. cm. Includes bibliographical references and index. ISBN 978-1-58826-548-7 (hardcover : alk. paper) 1. Representative government and representation—Latin America. 2. Political parties—Latin America. 3. Legislative bodies—Latin America. 4. Politicians—Latin America. I. Alcántara Sáez, Manuel. JL963.P64 2007 320.98—dc22 2007014417 British Cataloguing in Publication Data A Cataloguing in Publication record for this book is available from the British Library. Printed and bound in the United States of America The paper used in this publication meets the requirements of the American National Standard for Permanence of Paper for Printed Library Materials Z39.48-1992. 5

4

3

2

1

A-FM

10/8/07

10:52 AM

Page v

Contents

List of Tables and Figures Acknowledgments

vii xiii

Introduction, Manuel Alcántara Sáez 1

Studying Parliamentary Elites in Latin America, 1994–2006 Fátima García Díez and Araceli Mateos Díaz

1

7

Part 1 Politics and Democracy

2 3 4

The Conundrum of Representation Patricia Marenghi and Mercedes García Montero

29

The Meanings of Democracy and Institutional Trust Araceli Mateos Díaz

65

Legislators and Political Agendas Manuel Alcántara Sáez

87

Part 2 Party Politics and Ideology

5 6 7

The Organization of Political Parties Leticia M. Ruiz Rodríguez

113

The Dimensions of Polarization in Parliaments Cristina Rivas Pérez

139

The Determinants of Intraparty Ideological Differences Flavia Freidenberg, Fátima García Díez, and Iván Llamazares Valduvieco

161

v

A-FM

10/2/07

3:32 PM

vi 8 9 10

Page vi

CONTENTS

Programmatic Coherence Leticia M. Ruiz Rodríguez

173

The Parliamentary Left Salvador Martí i Puig and Salvador Santiuste Cué

195

The Parliamentary Right Manuel Alcántara Sáez and Iván Llamazares Valduvieco

219

Part 3 Conclusion

11 12

Legislative Careers: Does Quality Matter? María del Mar Martínez Rosón

235

Politicians and Politics in Latin America After Twenty-Five Years of Democracy Manuel Alcántara Sáez

265

Political Parties, by Acronym and by Country Bibliography The Contributors Index About the Book

275 283 303 305 323

A-FM

10/2/07

3:32 PM

Page vii

Tables and Figures

Tables 1.1 Distribution of Interviews by Country and Year 1.2 Analysis and Variables Used in This Study 1.3 Examples of Questions That Were Modified During the Course of the Study 1.4 Distribution of Interviews by Round, Country, and Political Party 2.1 Constitutional Regulations on Representation 2.2 Mean Values and Standard Deviations in the Style and Focus Indexes 2.3 Countries in Which the Right to Participate in Elections Is Offered Only to Political Parties 2.4 Influence of Different Actors on the Political Decisions Made by Legislators 2.5 Explanatory Variables Introduced into the Analysis: Style Dimension 2.6 Explanatory Variables Introduced into the Analysis: Focus Dimension 2.7 Multiple Regression: Style and Focus Regressed on Several Explanatory Variables 2.8 Results of the Analysis, by Country 2.9 Summary of Explanatory Factors of the Style Dimension 2.10 Legislative Period Analyzed and Number of Interviews 3.1 Preferences Regarding Democracy and Authoritarianism 3.2 Components of Democracy: Parties and Elections, by Country 3.3 Relationship Between Political Parties and Society 3.4 Typology of Democracies 3.5 Main Advantage of Democracy 3.6 Main Advantage of Democracy, by Ideology 3.7 Key Characteristics of Democratic Consolidation 3.8 Elements That Can Be Considered a Serious Threat to Democratic Consolidation, by Country 3.9 Threats to Democratic Consolidation

vii

17 19 20 23 31 36 40 45 47 49 50 52 54 59 66 69 70 71 74 76 77 78 79

A-FM

10/2/07

3:32 PM

Page viii

viii

TABLES AND FIGURES

3.10 4.1

Factor Analysis: The Threats to Democratic Consolidation Main Problems Faced by the Country’s Government, Second Round Main Problems Faced by the Country’s Government, Third Round A Classification of Latin American Countries According to the Importance Legislators Assign to Economic, Social, and Political Problems Main Problems Faced by the Country’s Government Differences in Perceptions of Problems in Both the Second and Third Rounds Main Problems Faced by the Country’s Government and Economic Performance Main Problems Faced by the Country’s Government and Data on Social Characteristics and Corruption Main Problems Faced by the Country’s Government and Political Data Perceptions for the Analysis of Intraparty Life of Political Parties Perceived Militancy by Party Legislators Perceived Participation by Party Legislators Degree of Perceived Internal Democracy by Party Legislators Preferred Degree of Power for Party Leadership Groups Perceived as Very Important in Making Decisions Pearson Correlation Indexes Perceived Party Life by Legislators Intraparty Life by Party System Correlations Between Party Life and Opinion on Political Parties Classification of Latin American Parties by Ideology Variables in the Analysis Analysis of Rotated Factors Proportion of Cumulative Variability Absorbed Relative Contributions of the Factors to the Programmatic Dimensions Relative Contributions of the Factors to Political Parties Classification of Political Parties in the Religion and Values Dimension Classification of Political Parties in the State Intervention Dimension Classification of Political Parties in the Democracy Dimension Classification of Political Parties in the Social Background Dimension

4.2 4.3

4.4 4.5 4.6 4.7 4.8 5.1 5.2 5.3 5.4 5.5 5.6 5.7 5.8 5.9 5.10 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9 6.10

80 88 89

90 92 105 106 107 108 118 119 121 123 124 125 126 129 131 132 141 142 143 144 145 146 149 150 151 152

A-FM

10/2/07

3:32 PM

Page ix

TABLES AND FIGURES

6.11 6.12 6.13 6.14 7.1 7.2 7.3

7.4 7.5 8.1 8.2 8.3 8.4 8.5 8.6 8.7 9.1 9.2 9.3 9.4 9.5 9.6 9.7 9.8 9.9 9.10 9.11 9.12 9.13 10.1

Classification of Political Parties in the Armed Forces Dimension Party Placement and Individual Self-Placement in the Left-Right Dimension Results of Factor Analysis Questions Used Differences Between Ideological Self-Placement and Party Positions, by Country Differences Between Ideological Self-Placement and Party Positions, by Political Party Pearson Correlation Coefficients Between the Aggregate National Variables and the Average National Differences of the Ideological Positions of Parties and Legislators Fixed Effects Estimates Estimation of Random Components Coherence in the Evaluation Component Coherence in the Strategy Component Pearson Correlations in the Strategy Component Variation in Programmatic Coherence, by Component Total Programmatic Coherence Pearson Correlation Coefficients Between Several Key Variables in the Study Correlation of Ideological Coherence and Ideology with Programmatic Coherence and Its Components Preferences for Democracy in the Principal Parties of the Latin American Left Degree of Agreement with the Statement That Elections Are Always the Best Way to Express Political Preferences Means of Confidence in Elections Degree of Agreement with the Statement That Without Parties There Cannot Be Democracy Main Advantage of a Democratic Regime Preferences Between Statism and the Market Preferences with Type of Taxes Preferences in the Privatization of Public Services Level of State Intervention in Price Controls Level of State Intervention in Creating Employment Level of State Intervention to Cover the Basic Necessities of Citizens Degree of State Intervention in Covering Health Care Level of State Intervention in Guaranteeing Free Access to University Education Left-Right Self-Placement Among Legislators of Right Parties

ix

153 155 156 157 164 165

167 168 168 180 182 183 184 185 188 191 199 201 202 203 204 208 208 209 210 211 212 212 213 222

A-FM

10/2/07

x

3:32 PM

Page x

TABLES AND FIGURES

10.2

10.3 10.4

10.5

11.1 11.2 11.3 11.4 11.5

Logistic Regressions: Differentiation of Right Parties in the Left-Right Dimension, and on Socioeconomic, Cultural, and Political Issues Summary of the Levels of Differentiation Between Members of Right and Nonright Parties Pearson Correlation Coefficients Between the Cox and Snell R2 and Other Political and Economic National Characteristics Pearson Correlation Index Between Membership in Right Parties and the Importance Governments Should Assign to the US in Foreign Policy Sociodemographic and Sociopolitical Characteristics of Latin American Legislators Sociodemographic and Political Characteristics of Latin American Legislators Classification of Quality of Experience of Latin American Legislators Distribution of Categories by Country Biplot Outcomes

Figures 2.1 Focus and Style in the Representative Role 2.2 Spatial Distinction Among Latin American Countries in the Index of the Focus and Style Dimensions 2.3 Spatial Distinction Among Latin American Political Parties in the Index of the Focus and Style Dimensions 2.4 In Your Opinion, Whom Do You Represent During Your Parliamentary Term? 2.5 Importance of Getting Resources for Your Province/State or Department 2.6 In the Case of a Conflict, How Would You Vote? 2.7 The Degree of Importance Assigned to the Defense of Party and Territorial Interests 2.8 Influence of Political Parties, Militants vs. Leaders 3.1 Institutional Trust, by Country 3.2 Democratic Stability and Index of Institutional Trust 4.1 Main Problems Faced by the Country’s Government by Ideological Self-Placement, Second Round 4.2 Main Problems Faced by the Country’s Government by Ideological Self-Placement, Third Round 6.1 Representation of the Polarization Dimensions, HJ-Biplot Axes 1 and 2

224 226

228

229 239 240 242 243 249

34 35 39 41 42 43 44 46 82 83 93 99 147

A-FM

10/2/07

3:32 PM

Page xi

TABLES AND FIGURES

6.2 8.1 8.2 8.3 8.4 9.1 9.2 9.3 11.1 11.2 11.3 11.4 11.5 11.6 11.7 11.8

Representation of the Polarization Dimensions, HJ-Biplot Axes 1 and 3 Degree of Programmatic Coherence Among Party Legislators Distribution of Coherence of Party Legislators in the Evaluation and Strategy Components Degree of Public Spending on Statism vs. Market, by Party Degree of State Role Preferred in Statism vs. Market, by Party Ideological Self-Placement of Parliamentarians Ideological Self-Placement of Political Parties Classification Tree Showing Level of Statism by Political Party Level of Experience Among Latin American Legislators Relationship Between Different Parliamentary Careers and Sociodemographic Variables, Biplot Axes 1 and 2 Relationship Between Different Parliamentary Careers and Sociodemographic Variables, Biplot Axes 1 and 3 Relationship Between Different Parliamentary Careers and Sociodemographic Variables, Biplot Axes 2 and 3 Distribution of Elected Legislatures by Level of Experience and District Size, Latin America Distribution of Elected Legislatures by Level of Experience and District Size, Closed-List Systems Distribution of Elected Legislatures by Level of Experience and District Size, Mixed Systems Distribution of Elected Legislatures by Level of Experience and District Size, Open-List Systems

xi

148 179 184 189 190 197 198 214 244 246 247 248 254 256 259 260

A-FM

10/2/07

3:32 PM

Page xii

A-FM.REV2

10/11/07

11:19 AM

Page xiii

Acknowledgments

POLITICIANS AND POLITICS IN LATIN AMERICA IS THE COLLECTIVE WORK

of a group of professors and researchers at the University of Salamanca in the areas of political science and public administration and/or from the university’s Instituto Interuniversitario de Iberoamérica. However, this study would not have been possible without the thorough and detailed work undertaken by a large number of people who carried out the fieldwork in Latin American congresses by personally interviewing representatives. (Because of the length of the list, which includes more than fifty people, we abstain from mentioning individual names for fear of inadvertently omitting one of them.) Practically all of the interviewers form part of the doctoral program in Latin American Political Processes or the master’s program in Latin American Studies, both at the University of Salamanca, but also linked to such Latin American research and postgraduate institutions as the Universidad Nacional Autónoma de México, the Instituto de Opinión Pública de la Universidad de Lima, and the Instituto de Ciencia Política de la Universidad de la República de Uruguay. In addition, staff of the Instituto Interuniversitario de Iberoamérica have been of invaluable help in data processing. Also, I am grateful to Marina Elias, who translated the manuscript into English. At different moments, financing for research has been provided, first and foremost, by the Spanish Ministerio de Educación through its Programas Nacionales, but also through the generous help of the Fundación BBVA, the Agencia Española de Cooperación Iberoamericana, Caja España, and, at the present moment, the Fundación Carolina, without whose valuable support the publication of this book would not have been possible. Finally, we would also like to thank the foreign institutions that assisted us when we visited their countries, benefiting from their libraries and from the intellectual stimulation of their Departments of Political Science or their Centers for Latin American Studies. For this support, we are especially grateful to Boston University, Georgetown University, the University of North Carolina at Chapel Hill, the University of Chicago, the University of Michigan at Ann Arbor, and Notre Dame University, as

xiii

A-FM

10/2/07

xiv

3:32 PM

Page xiv

ACKNOWLEDGMENTS

well as to FLACSO in Quito, the Institute of Ibero American Studies of Hamburg, and what was then called the Institute of Latin American Studies in London (now the Institute for the Study of the Americas). —Manuel Alcántara Sáez

A-FM

10/2/07

3:32 PM

Page 1

Introduction Manuel Alcántara Sáez

THE POSSIBILITY OF APPROACHING POLITICAL ANALYSIS FROM THE PER-

spective of politicians is an extraordinary opportunity, one whose importance goes beyond a mere descriptive knowledge of the opinions that legislators have about the activity in which they are immersed. The recognition of the role played by politicians in the way politics is carried out in their own countries, far from constituting a resuscitation of the elitist theories of the first half of the twentieth century, is of vital importance in present-day Latin America. After having left behind oligarchic and/or military regimes of an excluding nature, the current situation of the region is much different than it was when Seymour Martin Lipset and Aldo Solari (1967) published their influential work on elites in Latin America. In the preface to their book, the authors pointed out that, independently of the differences of social systems, one of the requirements for development was a competent elite, motivated to modernize society; at that time, the role of politicians, at least those of a democratic origin, was practically irrelevant. In the year 2007, however, Latin American politics is in the hands of politicians whose power emanates directly from a competence that is basically party based, through free and fair periodic elections. For such a thing to have happened simultaneously in so many countries and to have lasted for a continuous period of, on average, close to a quarter of a century is unprecedented in the region. The great importance accorded to institutions in the study of politics over the past fifteen years can be perceived in the widespread use of the phrase “institutions matter.” This is certainly the case, but it is also true that we should pay attention to actors who move in the institutional arena on a day-to-day basis. Espousing this approach does not imply that we should cease to pay attention to the general rules that govern human interaction. Rather, focusing on politicians permits us to see these human interactions in a new light, for these interactions do not exist independently from individual people; institutions themselves are constantly modified by the actions of individuals. In addition, after a quarter century of uninterrupted democracy in the majority of Latin American countries, the daily practice of politics

1

A-FM

10/2/07

2

3:32 PM

Page 2

POLITICIANS AND POLITICS IN LATIN AMERICA

within a reasonably stable set of rules has facilitated an unmistakable learning process. The challenges posed by having frequent elections, the struggles within political parties to obtain important positions, and the intense life in the different levels of representation and of political management are factors that increasingly contribute to converting politicians into actors of vital importance in every aspect of politics. Politicians matter. Political actions are influenced by institutional frameworks, but there is no political process without political players. Politicians have families, social and cultural origins, and they are marked by socialization processes that influence their ideas. In turn, the ideas of politicians are important to understand their strategic behavior and, in the end, the outcomes of the political game. Their belonging to political parties and moving in a specific institutional framework can affect them, but politicians also affect the characteristics of their parties and the institutional environments they move in. Calling a selection of persons a “political elite” continues to be complicated, inviting disagreement over the selection process, the people, and the very term “elite”; nevertheless, contextualizing the problem within the limits of the legislative power helps simplify things. This is the focus adopted in this book. In the first place, the political elite, for purposes of this book, is synonymous with the parliamentary elite. Legislators are clearly identifiable, and their functions and roles in the political system are known and, in representative democracies, are judged to be relevant. In the second place, the use of sampling techniques in specific universes helps eliminate disagreement over the selection process. Finally, the term “elite” is used in one of its simplest senses, as “someone who occupies a relevant place,” which is undoubtedly the case of Latin American national legislators.1 The study of politicians requires, at the very least, the intellectual effort demanded to pay attention to five different kinds of questions: these politicians’ positions in the general framework of power elites; their own processes of education, socialization, and structuring in the development of their representative roles; their shared and differentiated values; their evaluations, from ideological standpoints, of different public policies as solutions to societal problems; and their positioning vis-à-vis the smallest group of which they form a part, in terms of common strategies in both organizational and symbolic aspects. Politicians and, by extension, legislators are merely a part of the socalled power elite.2 This book, however, chooses to focus on formal party political life and parliament. With that focus, four groups of subjects are analyzed in this book. The most traditional route leading to politics in Latin America is getting a university education, most often in law, and having ties to family members who are or were politicians. This last route points toward a strong

A-FM

10/2/07

3:32 PM

Page 3

INTRODUCTION

3

tendency to inbreeding in the parliamentary elite, half of whom have relatives who are dedicated to politics. It is also indicative of ways of fitting into politics, depending on the politician’s country and on the existing party system. In addition, it is important to emphasize the importance that local politics has in the profiles of Latin American legislators. The background of shared values is one of the primary elements that helps to identify politicians, giving them strong distinguishing traits at the same time as it directs their behavior by allowing them to interpret politics in the framework of interparty competition. In this way, politicians give structure to a purely ideological component, through which they can draw up answers to existing challenges. Their positions on different public policies in which economic factors dominate demonstrate this ideological component and, at the same time, make it possible to discover to what degree different programs are related and what the level of cohesiveness is within parties. Finally, in this sense, politicians play different roles within their own parties, depending on the type of party leadership, the existence of intraparty democratic norms, their jurisdictions, and their access to funds for financing campaigns and party life. Although participation in party life is not very low in Latin America, internal democracy and a more intense interparty life are far from having been achieved to the same degree by all parties. These issues can be approached using biographies, official declarations, or manuscripts belonging to the individuals who make up the object of our study. If the purpose of a study is to learn about the past, using this type of information is a research strategy of unquestionable validity. However, this kind of material is not always readily available and tends to be dispersed and difficult to standardize. On the other hand, using personal interviews (with a questionnaire that is mostly closed) allows systematization of information and facilitates adding more data at a later date; moreover, it lends itself to data comparison and analysis, using statistical techniques. The negative side, however, is that this approach can only be used to study the present. The raw material used to construct this book’s empirical evidence comes from a research project focused on the opinions of representative samples of Latin American legislators and carried out between 1994 and 2006. The characteristics of this project are analyzed in the first chapter, written by Fátima García Díez and Araceli Mateos Díaz. In addition to providing an introduction to the project, these researchers explain the problems and decisions that the group had to face throughout the research process. The book is divided into three parts. Part 1 addresses subjects related to politics and democracy by focusing on the representatives’ perspectives on the democracies and problems of everyday political life. Part 2 revolves

A-FM

10/2/07

4

3:32 PM

Page 4

POLITICIANS AND POLITICS IN LATIN AMERICA

around the role played by the representatives’ ideologies, not only in shaping the major topics of conflict and polarization in the party system, but also as factors that give coherence to individual parties. Part 3 concludes with an acknowledgment of politics as a profession. Patricia Marenghi and Mercedes García Montero examine, in Chapter 2, the attitudes and perceptions of parliamentary elites in order to study current conceptions of representation held by these representatives. They define some of the factors influencing these legislators’ roles. The chapter analyzes how Latin American representatives conceive of their roles, using the two dimensions of representation: focus (which interests they represent) and style (whether they follow instructions—whatever the source of these instructions may be—or their own criteria). In Chapter 3, Araceli Mateos Díaz incorporates the point of view of the parliamentary elite into the debate on the meaning of democracy. She holds that the parliamentary elite are especially relevant actors since their opinions and attitudes constitute a reference in the shaping of the political orientations of citizens. She first shows the preference of this elite for a democratic regime, as opposed to an authoritarian one, and then details the representatives’ specific preferences regarding two components of democracy: elections and political parties. Later, in an attempt to more closely study the legislators’ visions of democracy, she presents what they consider to be the main advantages of the political system and democratic consolidation and what they hold to be the problems that can put democratic consolidation at risk. In Chapter 4, I identify what the representatives feel to be the main problems of national political life, as well as the strength of these problems and their evolution. I develop typologies by country and party ideological family. My analysis concentrates principally on two areas. First, I confirm that representatives consistently emphasize different problems, depending on their ideological orientations. I then discuss the relationship between the problems they identify and the realities in their respective countries, as seen in socioeconomic statistics. Leticia M. Ruiz Rodríguez, in Chapter 5, focuses on the organization of Latin American parties. Because of its impact on the nature and functioning of parties, party organization would seem to be partially responsible for the present disregard into which the Latin American political class has fallen. This chapter analyzes the perceptions of representatives concerning aspects of interparty organization, which are fundamentally related to the intensity of party life. The chapter also considers the relationship of the organizational question to its environment; that is to say, to the party system and citizens’ evaluations of the parties. Cristina Rivas Pérez dedicates Chapter 6, which relates ideology to parties, to examining the cleavages that contribute to the polarization and divi-

A-FM

10/2/07

3:32 PM

Page 5

INTRODUCTION

5

sion of political parties in Latin America. The perspective she employs is transnational; her starting point is the marked differences between the ideological positioning of the selected political parties (parties located on the left or right of the ideological continuum). Through the use of different statistical techniques, she aims to corroborate the existing differences between the Latin American left and right, based on the cleavages identified previously in her chapter. She also studies the main programmatic dimensions of polarization in present-day Latin American parliaments. Using factor analysis to identify latent structures, her analysis points to five dimensions, or factors (state economic intervention, the military, religion and moral values, social origins, and democracy). This technique allows her to represent graphically the positions of political parties in these five dimensions, making it possible to discover both the main sources of differences between different political parties and what aspects allow us to clearly distinguish between left and right parties in the region. On this basis, Rivas Pérez is able to identify the main Latin American party families. In Chapter 7, Flavia Freidenberg, Fátima García Díez, and Iván Llamazares Valduvieco undertake the study of the ideological heterogeneity of Latin American parties. Using multilevel analysis, they examine the impact of several causal determinants, both contextual and individual, on the ideological differences between legislators and their respective parties. Leticia M. Ruiz Rodríguez’s second contribution to this book, Chapter 8, analyzes the programmatic dimension of Latin American political parties. To this end, she measures the programmatic coherence of a group of parties at the level of the parliamentary elite. The representatives are the study’s measurement unit and the political parties are the analysis unit; thus the calculation of coherence allows the classification of political parties according to their programmatic structuring. The programmatic coherence is operationalized as the degree of agreement between the members of a party in their evaluation of a series of problems and in their definition of the strategies that should be followed. She shows that the parties demonstrate variable levels of programmatic coherence. In some cases, the thresholds of programmatic consensus are low, but, even so, they suggest a certain confluence in their diagnoses and solutions. This is not to deny the strength of other dynamics in Latin America, such as personalism and patronage, which sometimes exert stronger influences on the play between parties. In Chapter 9, Salvador Martí i Puig and Salvador Santiuste Cué analyze the impact on Latin American left party parliamentarians of the geopolitical, economic, and institutional transformations that have taken place over the past three decades. To this end, they explain what transformations have occurred in that time period at the international level, in political regimes of the region, and in the governments’ economic performance in a globalized context. After justifying their choice of parties to study, they analyze the

A-FM

10/2/07

6

3:32 PM

Page 6

POLITICIANS AND POLITICS IN LATIN AMERICA

perceptions of the representatives of left parties in power with respect to democracy and institutions as well as to the market and the state. Finally, they come to some tentative conclusions on how political representatives of the Latin American left think and express themselves. Iván Llamazares Valduvieco and I examine, in Chapter 10, the degree to which left-right self-placements and other types of attitudinal variables (socioeconomic, cultural, and political) allow one to distinguish between members of conservative and nonconservative parties. Our data show strong variations in the levels of ideological and programmatic distinctiveness of Latin American right parties. Our chapter also explores, from a comparative perspective, what the main reasons are for the levels of ideological and attitudinal differentiation among Latin American right parties. In Chapter 11, María del Mar Martínez Rosón looks at legislators’ professional trajectories, one of the least studied subjects in the Latin American region. She first reviews the key theoretical aspects of this concept in order to apply it to Latin American representatives. She then relates the resulting typology of representatives to a group of sociopolitical and sociodemographic variables, such as the representatives’ genders, ages, political socialization, or political ideologies. The objective of this chapter is to emphasize the relationship between different types of political careers, defined in terms of quality and vote. The results show that quality is important, since the representatives elected tend to be of a better quality when citizens can take into account the characteristics of the candidate when voting. I conclude this volume, in Chapter 12, by pointing out the relevance of a study of the parliamentary elite within the context of the new paradigm that politicians matter. Their special significance is demonstrated by their roles in representative democracy and in the context of their increasing professionalization. The idea that the quality of politics depends on the quality of politicians stands out as one that could be the guiding principle of a research agenda in the immediate future.



Notes

1. However, it must be noted that when this study was presented to them as research on parliamentary elites, some legislators emphatically refused to be considered as such. 2. Term coined by C. Wright Mills (1981), who, influenced by Weber, Parsons, Pareto, Mosca, Michels, and, above all, Aron (1989:195), was very careful to distinguish among “elite,” “political class,” and “ruling class.”

A-1

10/2/07

3:33 PM

Page 7

1 Studying Parliamentary Elites in Latin America, 1994–2006 Fátima García Díez and Araceli Mateos Díaz

SINCE THE LANDMARK 1953 PUBLICATION OF DAVID EASTON’S THE

Political System: An Inquiry into the State of Political Science, political analysis has been inseparable from the analysis of political actors, under the conceptual umbrella of politics understood as a system. Politicians and political elites, among other relevant institutional actors, have taken on special importance. After all, in contemporary democracies elected representatives—“agents” of the “principals,” or citizens—are those who make the final political decisions. And since Weber and Parsons, political action is seen to be intrinsically linked to the perceptions, attitudes, and political orientations of its actors, as well as to the ways they construct their visions of politics and of their own roles as representatives. However, in this field there has been a clear imbalance between theoretical and empirical research, to the detriment of the latter. This lack of empirical research is even more striking when one considers the recent interest, in academic circles, in the evaluation and measurement of the quality of democratic life. The widespread dearth of analyses concerning political elites and their trajectories, and above all concerning their values, beliefs, and attitudes, is in striking contrast to the advances made on these topics in the field of public opinion. Issues such as these, so fundamental and theoretically relevant to understanding the political process of any system, are, in the end, empirical questions. The study of parliamentary elites in Latin America, conducted by a research group associated with the Instituto Interuniversitario de Iberoamerica (University of Salamanca), 1 helps fill this analytical gap. Although the study focuses on Latin American parliamentary elites, it is placed within the wider theoretical and methodological fields analyzing “politicians” as fundamental actors in political systems in general. Until 1994, the year in which this group elaborated its first project,

7

A-1

10/2/07

8

3:33 PM

Page 8

POLITICIANS AND POLITICS IN LATIN AMERICA

“Political Reform and Values of Parliamentary Representatives in Four National Cases in Latin America,”2 the only existing studies on this subject dealt with either specific national cases or specific topics. Thus, it was logical that this first project focused on case studies on Argentina, Paraguay, Costa Rica, and Venezuela. However, fieldwork made obvious the need both for a more in-depth thematic study and for the expansion of the geographical focus to include the whole continent, an insight that gave rise to the next two studies authored by the group.3 The accumulation of longitudinal information on representatives of successive legislatures for all Latin American countries provided a view of continuities and changes discovered in the attitudes and perceptions of the elite. This information has had a twopronged effect: it has enabled the search for explanations of the elite’s attitudes and perceptions and, at the same time, it has been crucial in evaluating the usefulness of the questions and response categories elaborated to measure the concepts investigated by this study. This chapter is divided into four sections. The first section is a description and theoretical justification of the study. The second concentrates on its methodology, as well as explaining the main approaches taken by the University of Salamanca team and others in analyzing and using the data. Also included are the indicators that have enabled the “operationalization” of the main concepts and analytical orientations of this study. The third section discusses why the repeated application of the opinion surveys made necessary the reconsideration of some indicators and categories relevant to the collection of information. Other decisions, related to the representativeness, reliability, and validity of the information, are also discussed in this section. The final section of this chapter is a recapitulation of the fundamental pillars on which the design and research strategy of this project rests.

■ Antecedents and the State of Research on Parliamentary Elites Taking into account the previous analyses of the study of political elites, the research by the University of Salamanca group has from its very beginning set out to expand and deepen the knowledge of the attitudes, opinions, and values of the Latin American parliamentary elite as determining factors in the consolidation and quality of democracies in this region. The centrality of parliament as an arena in which political representation is instituted and as a place crucial to drawing up public policies (Cohen, 1989) has made the study of political elites a central issue in political science. However, until the mid-90s, when the University of Salamanca research group began to function, there was a notable lack of research on Latin American political elites, in terms of production of both knowledge and publications about the political class that governed the recently

A-1

10/2/07

3:33 PM

Page 9

STUDYING PARLIAMENTARY ELITES IN LATIN AMERICA

9

(re)installed democracies of the region. The impact of the work of Robert Putnam (1973) and the analysis of the factors that determined the beliefs of the political classes of Great Britain and Italy made even more striking the lack of similar studies on Latin America, with the outstanding exception of the work of John Higley and Richard Gunther (1992). Regarding Eastern Europe, worth noting is the work of Higley, Jan Pakulsli, and Wlodzimierz Wesolowski (1998) on political elites in post-Communist societies. From a wider comparative interregional perspective, among studies worth citing are those of Mosheon Czudnoswski (1983), on the relationship between political elites and social change in Canada, Jamaica, Brazil, Kenya, Korea, and Turkey; and of Samuel Eldersveld (1989), on the relationship between democratic theory and empirical research on political elites in modern societies. Looking specifically at the United States, there is Eva Etzioni-Halevy’s 1993 study, The Elite Connection. The state of research on these issues in Spain is similar, for it was not until the second half of the 1990s that such studies—authored by Miguel Jerez (1997), Joan Botella (1997), Pilar Gangas (2000), Irene Delgado (2000), and Edurne Uriarte (2000)— appeared. The reasons for this lack are both academic and historical. The academic reasons have to do with the difficulty of access to the materials and human resources that fuel empirical studies of this nature, as well as with the evolution of academic interest itself, which was previously concerned with the processes of the transition to democracy and then with the institutional arrangements within which these democracies were constructed. From the historical point of view, until the most recent wave of democratization (Huntington, 1991), the fragile development of Latin American democracies made the legislative house a political arena of, at the very least, relative autonomy. The resurgence and strengthening of parliaments’ role in democratic political life, as parliaments gained the capacity to confront strong executives, thus become critical elements for the sustainability and quality of democracy (Close, 1995; Lujambio, 1993; Molinelli, 1991). An analysis of these processes can be found in Doh Chill Shin (1994), one of the most perceptive reviews of the literature on the third wave of democratization. When the empirical universe of democratic theory was expanded to include more geographical areas, it continued to include a number of unquestioned assumptions (O’Donnell, 1999b:304), one of them being the relationship established between the values, beliefs, and opinions of parliamentary elites and the quality of democracy. Even today, the study of the attitudinal and sociodemographic traits of parliamentary elites, as well as the techniques for measuring these traits, is an underdeveloped field (Diamond, 1999:66). The production of the Salamanca research team is one of the most substantial in this field for Latin America. Given the lack of theoretical-methodological references, the

A-1

10/2/07

3:33 PM

10

Page 10

POLITICIANS AND POLITICS IN LATIN AMERICA

first studies done by this team revolved primarily around case studies. Although these studies were done from a comparative perspective, they were concerned with four countries that had a more long-standing experience of democracy. The second set of studies widened the perspective to include the entire continent, incorporating recently established democracies. The collection of data became more systematized and the survey was refined. Thus, the project began with an initial round of data collection, conceived of as exploratory, which permitted the team to generate theoretically and empirically relevant hypotheses within an academic field characterized by limited production. This research process made it necessary to carry out case studies in which questionnaires were answered in personal interviews, in combination with a comparative perspective on a limited number of countries. The in-depth interviews made it possible to identify and incorporate new variables that had not been anticipated in the initial research design. The ideas and analytical framework that arose from this first stage were what permitted second and third stages to be carried out, this time on a regional scale. The observation of an increased number of countries, through a survey design, has made it possible for researchers to collect a great deal of information in a systematic fashion, as well as to generate the comparison and methodological description of a large number of cases. Moreover, the application of this questionnaire in different countries and at different moments has allowed the team to evaluate the reliability of the indicators and to make the reformulations necessary to collect more precise information. From the theoretical point of view, this research project takes as its starting point the concept of “political elite” itself, and this concept’s difference from the term “political class” in democratic contexts (Von Beyme, 1995), since outside of democratic contexts the idea of representation neither makes sense nor has any empirical referent. The study thus considers parliamentary representatives to be a group of actors belonging to the political elite, since the concept of political class diverges from that of those who hold the relevant positions of power. The theoretical frameworks that support the studies of political elites can be divided into three main groups (Alcántara and Llamazares, 1997:1): (1) those that are structural and concerned with the relationship between the socioeconomic origin of the elites and their access to channels of political influence; (2) those that are interactionist; that is to say, studies that relate the orientations and perceptions of the elites with the processes of interaction that take place between them; and (3) motivational points of view, which take as their base the vocations and attitudes that the representatives bring with them to the political arenas and those that they develop in the course of their political activity. It is for this reason that, to understand

A-1

10/2/07

3:33 PM

Page 11

STUDYING PARLIAMENTARY ELITES IN LATIN AMERICA

11

political action, it is indispensable to understand the way in which the elites see their own positions as well as how they perceive the instruments available to achieve their ends. There is, in addition, the more recent school of thought that recuperates the concern with institutions, combining the micro- and macrolevels of analysis. That is to say, it studies both elites and the political systems within which they function. 4 The two levels are seen as doubly connected. Importance is assigned to the impact that institutional arrangements and the historical experience of each country have on the political trajectories, ideological placements, values, and beliefs of its political representatives. At the same time, weight is given to the effects that the attitudes, ideologies, and opinions of political representatives have on the public policies, dynamics of the political system, and the nature and quality of democracy in each particular country. Together with the institutional and behavioral dimensions, the attitudinal dimension is crucial in the consolidation of democratic regimes. We cannot speak of “democratic consolidation” if public opinion in general and the political class in particular do not hold the belief that democratic procedures are the best form of government for collective social life (Linz and Stepan, 1996:6). Still less can we evaluate the kind of democracy and its quality without taking into consideration the attitudes of the political class and parliamentary representatives. It is through these representatives that society constitutes itself into a democratic system of government. The consolidation of democracy takes place on three levels: elites, parties, and organizations and citizenry. Due to the power and influence that they wield, the elites have an indisputable importance to the stability of democracy in the dimensions of beliefs and behavior (Diamond, 1999:66). The beliefs, values, and opinions of the elites are relevant first of all because they exercise more influence on political processes than do the beliefs, values, and opinions of other actors. In addition, as pointed out by Robert Dahl (1971:128), it is more probable that the elites have constructed systems of political beliefs and that their actions are ruled by those systems. Beyond the power that they exercise directly over events and decisions, the elites play an essential role in the formation of a country’s political culture as well as in the determination of what is appropriate behavior. In great part, the elites exercise leadership by the example they set; when they are in agreement with the rules and values of democracy, it is more probable that their followers will be in agreement as well. Generally, elites are carriers of distinctive values and beliefs, they invariably possess greater information on the system, and they often lead the process of large-scale changes in values (Diamond, 1999:163). The decisions of key political leaders during crises or in constructing the political system have a powerful influence on whether or not a democra-

A-1

10/2/07

3:33 PM

12

Page 12

POLITICIANS AND POLITICS IN LATIN AMERICA

cy will be stable, persistent, or weakened. Even when democracy is firmly consolidated, its quality can deteriorate; it is then that the adaptation and reinforcement of the elites can become crucial factors. This is not to contradict the idea that if a democracy is to become stable and effective, the majority of the citizenry must have an unwavering commitment to the system. Nevertheless, it is also true that the political culture of the elites is crucial to the consolidation of democracy. A democracy cannot work unless the elites accept the rules and limits of the constitutional system and the legitimacy of the opposition actors who, in turn, must also be committed to the democratic system of government. The concepts “consolidation of democracy” and “quality of democracy” are, therefore, intimately related, and factors used in determining what conditions favor the consolidation of democracy are also extremely useful for examining its quality. As Juan Linz and Alfred Stepan (1996) point out, within the category of consolidated democracies, there is a continuum from democracies of low quality to democracies of high quality. Consolidation can be presented as a dichotomy: that is to say, the attitudes and values of the elites do or do not contribute to the consolidation of democracy. However, this approach obscures a wide range of values that could be significantly relevant to the nature and quality of the democratic regime. The concept of consolidation avoids, to a certain extent, the problems of conceptualization and operationalization implicit in the concept of quality, but it also eliminates the possibility of explaining why some democratic regimes follow a substantially different course than others. The concept of democratic quality is a highly abstract one, and there is a great deal of disagreement about its capacity for operationalization. Sometimes the arguments follow opposite criteria: on one hand, some hold that the quality of democracy can be evaluated directly by its citizens (subjective). On the other end of the spectrum, some propose that the quality of democracy can be determined by institutional performance (objective). Guillermo O’Donnell’s most recent reflections on the subject are structured around this duality (1999a). What can be concluded is that the concept of quality is multidimensional, and that different qualities of democracy can exist within one political system. Among the most commonly cited dimensions are the following: liberty, the rule of law, accountability, responsiveness, equality, participation, competition (Diamond and Morlino, 2004). With respect to the objectives of research on political elites in general and parliamentary elites in particular, the dimensions with special relevance are those of quality of representation and quality of the parliamentary policymaking processes. The concept of quality is relational; a thing is seen to have quality with respect to another thing (Gutiérrez and Vargas, 1998); for this reason, the application of a research design must be comparative in spatial and chrono-

A-1

10/2/07

3:33 PM

Page 13

STUDYING PARLIAMENTARY ELITES IN LATIN AMERICA

13

logical terms. This concept implies, at least theoretically, a continuum of democratic content in the political life of the polyarchies that ranges from the everyday to the rare exercise of democratic values. In comparative studies of democracy, the focus has changed from “how” political transitions to democracy occur to “what” these recently (re)installed democracies are like (Diamond and Morlino, 2004). Within the context of this type of research, certain questions are posed only infrequently. These rarely addressed issues include who the governing political elites of these countries are, what they think, what the political implications of the differences between countries on this point are, and why these variations exist. The assessment of the quality of democracy from the point of view of elected representatives is thus a frequently ignored question in comparative studies of the quality of democracy (Alcántara, 2004c). The research design (survey) that this research team used, explained in the following section, is thus the ideal instrument for identifying the perceptions and attitudes of parliamentary elites at a regional level and their effects on the political practices chosen for study. This emphasis on the attitudes of elites diverges from other recent studies focusing on parliamentarians, such as those analyzing the strategic behavior of parliamentarians (Shepsle and Bonchek, 2005), the actors setting the agenda in the legislative process (Tsebelis, 2000), or the distribution of information among committee members (Krehbiel, 2004). The attitudes and perceptions of parliamentary elites can be considered as independent variables critically affecting the behavior of legislators. Studying the attitudes of individual parliamentarians from multiple angles allows a more thorough understanding of the characteristics of Latin American legislatures.

■ Objectives and Methodological Design In defining the study on parliamentary elites (University of Salamanca) in terms of its objectives, the team’s research is double-pronged, addressing both the explanatory and the descriptive, since it has two objects. On one hand, it aims to study the parliamentary elites in Latin America and the characteristics that differentiate or identify them (sociodemographic, ideological, party affiliations, etc.); on the other, it tries to explain possible causal relationships between representatives’ opinions and attitudes and their behaviors, as well as between their opinions and attitudes and the consolidation and quality of democracy. The team’s goal, after describing the characteristics of the parliamentary elite, is to show which characteristics affect its development and behaviors, as well as the implications of these characteristics for the development and nature of the democratic regime. As was previously mentioned, in the first stage5 the focus was on the

A-1

10/2/07

3:33 PM

14

Page 14

POLITICIANS AND POLITICS IN LATIN AMERICA

characteristics of parliamentary representatives and on causal relationships with a national or case study component. In the second and third stages,6 the goal was to obtain a more comparative (regional or global) view of the characteristics of parliamentary elites. The objectives of the use and analysis of these data vary, depending on the unit of analysis under consideration. In the first place, an intensive analysis strategy has been employed, since individual countries were used to carry out the case studies. In such studies, the objectives have to do with the characteristics and causal relationships within each country; the team attempts to explain the behaviors of parliamentary representatives in each country. In addition, thanks to its longitudinal nature, the information gathered in different legislatures is compared in order to study the change or continuity in opinions and attitudes, as well as their effects on the political system. Numerous articles focusing on single cases have been produced, as well as small “n” studies and regional comparisons.7 Some studies have taken as a unit of analysis the families of parties of the left8 or of the right,9 or have investigated whether political parties are unitary actors or if the factors affecting parliamentary representatives are motivations of an individual kind.10 In this way, within each country and throughout time, the data gathered have allowed conclusions to be drawn at the individual level (parliamentary elites), at the party level, and at the national level. This has made it possible to carry out comparisons that are cross-national and/or longitudinal over time. Another analytical strategy used has been to consider each country as a unit of observation within its regional context. A panoramic focus on Latin American parliamentary elites enables one to draw conclusions on a regional level11 and to establish possible differences between specific geographical areas: Central America, the Southern Cone, and the Andean region. The importance of region as a unit of analysis within the context of comparative politics cannot be stressed enough. As Mainwaring and Pérez-Liñán (2004b) have pointed out, regions are characterized by their own specific dynamics and political processes. Any empirical generalization that has a more overarching geographical framework must pay attention to regional specificities, for different regions will show different causal patterns. Mainwaring and Pérez-Liñán also argue for the existence of powerful international mechanisms of political diffusion and learning, which take on increased importance within these regions. The structure and strategy of the University of Salamanca’s research project, “Parliamentary Elites of Latin America,” not only allow the establishment of generalizations at the levels of individual, party, and country, but also the identification of traits specific to Latin America as a region in terms of the political trajectories, beliefs, attitudes, and values of its parliamentary elites. Finally, and combining the two strategies outlined above, the diachronic perspective is employed to

A-1

10/2/07

3:33 PM

Page 15

STUDYING PARLIAMENTARY ELITES IN LATIN AMERICA

15

find out which aspects remain constant and stable, and which undergo change both within individual countries and among them. Above all, the focus is on comparing countries and trying to explain why attitudes are stable in some and not in others. Although the unit of observation used is that of individual parliamentary representatives, in some cases the analysis focuses on representatives grouped by their political parties. The characteristics of the elites and of different countries (in terms of party systems and political-territorial structures) are some of the independent variables that determine the relations of causality to be studied. This study leaves no doubt as to the usefulness of the categories of right and left in Latin America. From the very beginning of this project, special attention has been paid both to the values and beliefs that serve as a basis for ideologies and to the possible existence of an ideological continuum that can be defined in terms of left-right. The results obtained up till now prove that one of the main poles around which party competition is articulated is left-right opposition, and that left-right placements, in turn, coincide with the positioning of elites on such other subjects as abortion and divorce, proor anticlerical stances, or state intervention. Recent research has corroborated these results on the level of the citizenry of Latin America (Colomer and Escatel, 2005). It has been possible to use this information to identify programmatic and referential ideological differences between parties,12 as well as to identify the polarizing potential of parliamentary systems (Carey, 2001). Empirical data on the left-right dimension (which until now did not exist on a continental level) have allowed the exploration of its repercussions on various political phenomena, such as parliamentary roll-call voting and the internal discipline or internal fractionalization of parties. Questionnaires have been considered the most appropriate method of data collection, since personal interviews would make it more difficult to compare and extend generalizations to all Latin American countries. Although the questionnaire includes some open and semi-structured questions, the majority are closed and blocked, in order to facilitate comparisons, and with scaled levels of measurement that allow for statistical analyses. In addition, using a questionnaire with a proportional sample design allows one to make inferences regarding political parties and the House of Representatives as a whole. In the case of bicameral legislatures, analysis is done exclusively of the lower chamber, for a more representative sample in proportional and party terms (this is not, however, the case of Colombia). As has been explained in this chapter, the fact that these data lend themselves to analyses at different levels has to do with a fundamental characteristic of the databases: the information contained in them has a hierarchical structure ideal for the application of multilevel analytical techniques.13 The universe considered for the sample design in each country is that of

A-1

10/2/07

3:33 PM

16

Page 16

POLITICIANS AND POLITICS IN LATIN AMERICA

the men and women elected as representatives in a legislature, grouped into political parties. The selection of the sample is based on criteria proportional to the total number of members of the house. The selection of specific individuals was random; the only considerations influencing the final selection were those of gender and age. In those cases in which the designed sample was not reached, data were weighted so that one may draw conclusions about the whole legislature; this weighting was based on criteria of representativeness by political party. Since the population that constitutes the object of study in each country is small, in spite of the fact that the sample includes a significant percentage of this population, error levels vary greatly from country to country (see Table 1.1).14 The design of the questionnaire is based on the operationalization of the main concepts of this study into variables capable of measuring the concepts’ most relevant dimensions. The main lines of the analysis in the questionnaire are the following: democracy, elections and forms of government, political parties, institutions and armed forces, the role of the state and political economy, national problems, integration and foreign policy, political trajectories and parliamentary activities, values, and social-demographic characteristics (see Table 1.2). The questionnaire consists of more than 80 questions and 320 variables, which include the indicators detailed previously. As of now, a total 3,967 personal interviews of Latin American representatives have been effected. The distribution of these interviews appears in Table 1.1.

■ Methodological Decisions Made in This Study In the University of Salamanca’s parliamentary elites research project, now in its tenth year, the team has had to make a number of important methodological decisions along the way. Some of these decisions have had to do with the questionnaire used to collect information. Others have been related to whether or not the object of analysis is individual countries or the region as a whole. There were changes made with respect to the indicators selected for the collection of data. In the second and third rounds, questions were added to address topics of current interest in the field of political science. Other questions were eliminated or replaced, either because they were not understood correctly by the interviewees or because they were not effective in eliciting the intended information. Some of the scales used were readjusted to achieve homogeneity with respect to other studies. These scales were shared with the participants so that they could take them into consideration in their evaluations, and to avoid misunderstandings when they were asked to proffer their opinions. Table 1.3 gives some examples of questions that were modified. When the object of analysis was that of individual countries, various

A-1

Distribution of Interviews by Country and Year

Argentina 1995–1997

68

Bolivia

1993–1997

74

Chile

1993–1997

93

Costa Rica

1994–1998

52

Colombiaa

% of the % of the % of the Chamber Number Chamber Number Chamber Number Second Legislative (sample of Third Legislative (sample of Fourth Legislative (sample of Round Term error) Interviews Round Term error) Interviews Round Term error) Interviews

26.5% Argentina 1997–2001 49.8% (±11.93) (±6.39) 56.9% Bolivia 1997–2002 75.4% (±9.05) (±5.18) 77.5% Chile 1997–2001 74.2% (±5.18) (±5.58)

128

Argentina 2003–2007

40.9% (±7.51) 2002–2007 61.5% (±7.45) 2001–2005 73.3% (±5.71)

105

98

Bolivia

89

Chile

91.2% Costa 1998–2002 86.0% (±4.22) Rica (±5.56) Colombia 1998–2002 54.7% (±10.47)

49

Costa 2002–2006 89.5% Rica (±4.80) Colombia 2002-2006 57.2% (±6.93)

51

88

80 88

95

Dominican Republic 1994–1998

62

51.7% Dominican 1998–2002 69.1% (±9.72) Republic (±5.64)

103

Dominican 2002–2006 78.7% Republic (±4.31)

118

El Salvador 1994–1997

46

58

72

El 2000–2003 76.2% Salvador (±6.50) Ecuador 2002–2006 98.0% (±1.50) Guatemala 2000–2004 69.9% (±6.44) Honduras 2001–2005 79.7% (±4.58) Mexico 2000–2003 24.8% (±7.92)

64

Ecuador

54.8% El 1997–2000 69.0% (±10.23) Salvador (±7.82) 87.8% Ecuador 1998–2002 92.6% (±4.43) (±2.65) Guatemala 1995–2000 78.8% (±5.96) 52.3% Honduras 1997–2001 55.5% (±8.65) (±8.09) 24.6% Mexico 1997–2000 25.2% (±8.09) (±7.84)

1996–1998

Guatemalaa Honduras 1994–1997 Mexico

1994–1997

67 123

112 63 71 126

El 2003–2006 95.2% Salvador (±2.5)

80

98 79

Guatemala 2004–2008

76.6% (±4.62)

121

Mexico

24.8% (±7.92)

124

102 124

2003–2006

(continues)

Page 17

% of the Number Chamber Legislative of (sample Term Interviews error)

3:33 PM

First Round

10/2/07

Table 1.1

A-1

continued % of the % of the % of the Chamber Number Chamber Number Chamber Number Second Legislative (sample of Third Legislative (sample of Fourth Legislative (sample of Round Term error) Interviews Round Term error) Interviews Round Term error) Interviews

Nicaraguaa

Nicaragua 1996–2001

Panamaa

Panamaa

Paraguay 1993–1998

47

Peru

1995–2000

87

Venezuela 1993–1998

69

Uruguay 1995–2000

73

Total

58.8% (±9.78) 72.5% (±5.96) 34.0% (±10.26) 73.7% (±6.22)

Paraguay 1998–2003

75.3% (±6.08)

81.3% (±5.46) Peru 2001–2006 69.17% (±6.70) Venezuela 2000–2005 60.6% (±6.58) Uruguay 2000–2005 68.7% (±7.08)

933

Source: Created by authors based on PELA 1994–2005. Note: a. It was not possible to obtain a representative sample.

70

65

Nicaragua 2002–2006

65.2% (±8.06) Panama 1999–2004 90.1% (±4.02) Paraguay 2003–2008 70.0% (±7.75)

60 64

Panama

2004–2009

87.2% (±4.4)

68

56

83 100 68 1,371

Uruguay 2005–2010

86.86% (±4.15)

86 1,270

393

Page 18

% of the Number Chamber Legislative of (sample Term Interviews error)

3:33 PM

First Round

10/2/07

Table 1.1

A-1

10/2/07

3:33 PM

Page 19

STUDYING PARLIAMENTARY ELITES IN LATIN AMERICA Table 1.2

19

Analysis and Variables Used in This Study

Analysis Democracy, elections, and form of government

Political parties

Armed forces Role of the state and public policies

Regional integration and foreign policy

Political trajectories and parliamentary activity

Left-right ideological axis Values and sociodemographic characteristics

Variables

Definition of democracy and evaluation of its advantages; factors linked to democratic consolidation and risks of democratic breakdown; satisfaction with democratic stability and trust of electoral processes; definition and preferences of form of government; democratic stability; assessment of majoritarian and proportional electoral systems; advantages and disadvantages of the presidential regime when compared to the parliamentarian one. Party identification, level of party activism, role of political parties, levels of popular mobilization, militancy levels, democracy and internal organization, trust in political parties, political participation in party life, legalization of political parties, vote discipline, opinion on party leaders. Opinions on the civilian control of the armed forces; assessment of the roles played by the armed forces. Identification and evaluation of problems, evaluation of economic policies: public expenditure, taxes, privatizations, social expenditure by sectors, level of state economic intervention, role of the state, direct and indirect taxes, privatizations. Foreign policy, regional integration and international arena, conditionality, origin of foreign investment, priorities in foreign policy, role of the European Union in Latin America, bilateral relationships between Spain and Latin America. Socioeconomic origins, political trajectories, party militancy, parliamentarian activity and experience, elected public offices held, problems faced in parliamentary life, reasons to be elected, identification of individuals and groups whom the parliamentarian believes she or he represents, decisionmaking evaluation of parliamentary services, political background of the family, compatibility with other personal economic activities, respondent’s evaluation of his/her income. Ideological self-placement, ideological distance among parties and leaders in the left-right continuum. Gender, age, income, profession, religion and religious practice, attitudes toward abortion and divorce.

decisions had to be made. The first one had to do with the stratification of the population, respecting the principle of representativeness. First of all, parties were selected that had come out of the most recent elections for the lower legislature with at least one elected representative. Within this group, it was judged that representation proportional to results obtained in the elections was best. This would ensure that the opinions and attitudes of all legislators would be included, even the ones that came from small groups. Parties that obtained fewer than five representatives were put into the cate-

A-1

10/2/07

3:33 PM

20

Page 20

POLITICIANS AND POLITICS IN LATIN AMERICA

Table 1.3

Examples of Questions That Were Modified During the Course of the Study Previous Question

Modified Question

In a context of crisis and political instability, how would you rate your agreement with the statement that democracy is preferable to any other form of government: very much in agreement, in agreement, somewhat in agreement, or not at all in agreement?

With which of the following statements are you most in agreement: • Democracy is preferable to any other form of government. • In a context of economic crisis or political instability, an authoritarian government can be preferable to a democratic one.

At the present time, the respective advantages and disadvantages of proportional electoral systems as opposed to majoritarian ones are very much under discussion. Despite the difficulty of settling these questions, in general terms, which of the following sentences best expresses your opinion: • The electoral system should guarantee equal representation for all political options. • The electoral system should guarantee equal representation for all political options that have received more than a certain number of votes. • The electoral system should guarantee the representation of minorities, while at the same time guaranteeing the possibility of creating majority governments. • The majoritarian electoral system is the only one capable of guaranteeing an efficient government.

At the present time, the respective advantages and disadvantages of different electoral systems are very much under discussion. Think of parliamentary elections, and taking into account the following two scales, please tell me what you believe would be the ideal positions for your country. Proportional system that guarantees equal representation of all political forces

Majoritarian system that guarantees strong and effective governments

01 02 03 04 05 06 07 08 09 10 Personalized voting system that guarantees a close relationship between the voter and his/her representatives

Closed-list voting system that favors the formation of strong, cohesive parties

01 02 03 04 05 06 07 08 09 10 If it were necessary to raise taxes, how do you think this raise should be effected: through direct or indirect taxes? And in the event that your country should decide to increase direct taxes, which ones do you think should be raised: direct taxes on capital or direct taxes on work?

And taking into account that in order to spend more, it could be necessary to raise taxes, how do you think that this increase should be financed: through taxes on capital and work or through taxes on consumption?

gory “others.” The reason for this division was to guarantee the anonymity of representatives from small groups when data were analyzed. Nevertheless, it should be kept in mind that in countries with a multiparty system and a large number of small parties, the category of “others” can be very heterogeneous, and thus data from this category can sometimes be incongruent and difficult to interpret.

A-1

10/2/07

3:33 PM

Page 21

STUDYING PARLIAMENTARY ELITES IN LATIN AMERICA

21

Representativeness is often limited by changes in the party affiliations of the representatives immediately after elections have been celebrated. A sample design based on electoral results can require modification only a few days after elections, when coalitions are being formed. The phenomenon of transfuguismo (i.e., changes in party affiliations) makes criteria of party representativeness difficult to set. The panorama is further complicated by the fact that, in some countries, although political parties go into elections as a united group, they may have internal divisions, or fractions. These difficulties make it necessary to consider homogeneous criteria for all countries. The research team decided that the composition of the lower chamber used to elaborate the sample design would be based on the results established the day of elections. If a representative changed parties after the election, for the purposes of the sample design this legislator would be considered a member of the party that he or she had belonged to at the time of elections. The second important decision taking into account the individual country as a unit of analysis is related to the longitudinal character of the study and the selection of measurement indicators in the questionnaire. On occasion, using the same tool to compare the answers of legislators from the same country makes it more difficult to include contextual elements relevant at the time of the questionnaire, reducing the chance to expand the collection of information or explicative variables. On the other hand, opinions on some topics could be heavily influenced by events; thus, these differences are not so much party-related or ideological as contextual. As pointed out previously, the research group has tried to correct errors to do with measurement, questions, or values that cause comprehension problems or problems eliciting appropriate information. This reformulation of questions and values has advantages and disadvantages. On one hand, the validity and reliability of the questions are increased. On the other, however, the comparative capacity of some indicators is lost. The research team decided that it was much more important to have reliable and valid measurement indicators than to continue to reproduce indicators whose relationship with the original theoretical concept was questionable. Data have been grouped by rounds to help one make inferences about the region(s) under discussion. The global comparison of data obtained in this study allows us to divide the data into three phases, corresponding with the times information was collected. The comparison is established, up to 2005, among three rounds of data collection; the fourth has not been completed in all countries. It should be pointed out that the dates of each phase are not identical in every country; this is because election dates vary from country to country, as do legislative terms (which range from three years to five years). Thus, for example, if the first round took place between 1993

A-1

10/2/07

3:33 PM

22

Page 22

POLITICIANS AND POLITICS IN LATIN AMERICA

and 1997, the dates of data collection could vary by as much as three or four years between countries. This time gap can condition legislators’ answers concerning topics that were relevant at one point in time, including at the regional level, but were not so relevant two or three years later. Thus, some opinions expressed in one round of interviews may be more similar to those expressed in the following round of data collection.

■ General Diagnosis Over the past decade, the study on parliamentary elites in Latin America, carried out by the Instituto Interuniversitario de Iberoamérica (University of Salamanca) under the direction of Manuel Alcántara Sáez, has conducted approximately four thousand personal interviews with congressional representatives. The resulting database has more than three hundred and twenty variables, structured around axes of analysis that include representatives’ beliefs and attitudes about democracy, elections and parties, institutions, the role of the state and public policy, and integration and foreign policy, as well as their views on their own political trajectories and their jobs as representatives. The representativeness of the design samples and the hierarchical structure of the data allow empirical generalizations to be made concerning the majority of Latin American countries, on four levels of analysis: individual, party, country, and region. As well as the potential of cross-country research, this ongoing study contributes the potential of longitudinal comparison at three different points of time—four, in the cases of El Salvador and Mexico. This research group’s contribution takes on special relevance in the context of the current interest in the quality of the recently (re)installed democracies of the region. The concept “democratic quality” requires designs of information collection that are comparative in both multidimensional and relational terms; this is a special feature of the project presented in this chapter. The information made available by this study goes beyond enabling indepth descriptions of the attitudes, beliefs, and values of Latin American parliamentary representatives. It offers, in addition, a great variety of options for the formulation of research problems of undeniable theoretical relevance. These problems include the degree to which the preferences of citizens and their representatives coincide; the variations in representatives’ perceptions, when compared between countries and over time; the fundamental determinants of these variations and their concrete implications in the carrying out of high-quality democratic political practices. Until the present moment, there had been no possibility of submitting such problems to a rigorous, valid, and reliable empirical examination.

A-1

10/2/07

3:33 PM

Page 23

■ Appendix Table 1.4

Distribution of Interviews by Round, Country, and Political Party First Round

Argentina PJ UCR FREPASO Provincial parties ARI PS Other partiesa Total Bolivia MNR ADN MIR UCS CONDEPA NFR MAS MIP Other partiesa Total Chile PDC RN UDI PPD PS PRSD Other partiesa Total Costa Rica PLN PUSC PAC Mov. Costarricense RN Other partiesa Total Colombia PL PC Coalición Other partiesa Total

Second Round

Third Round

Fourth Round

Universe Sample Universe Sample Universe Sample Universe Sample 131 68 22 12 — — 24 257

23 17 8 5 — — 15 68

119 68 38 22 — — 10 257

59 34 19 11 — — 5 128

131 46 — — 12 6 62 257

51 24 — — 6 2 22 105

Universe Sample Universe Sample Universe Sample Universe Sample 52 15 20 20 13 — — — 10 130

23 9 7 19 7 — — — 9 74

26 32 23 21 19 — — — 9 130

20 24 17 16 14 — — — 7 98

36 — 26 5 — 25 27 6 5 130

19 — 16 4 — 16 17 3 5 80

Universe Sample Universe Sample Universe Sample Universe Sample 37 29 15 15 16 — 8 120

30 22 10 11 14 — 6 93

39 23 23 16 11 — 8 120

29 17 17 12 8 — 6 89

24 21 36 21 12 6 — 120

18 16 25 15 9 5 — 88

Universe Sample Universe Sample Universe Sample Universe Sample 28 25 — —

25 23 — —

23 27 — —

20 22 — —

4 57

4 52

7 57

7 49

16 19 15 6 1 — 57

14 17 14 5 1 — 51

Universe Sample Universe Sample Universe Sample Universe Sample 83 43 21 14 161

48 29 3 8 88

109 48 — 9 166

60 27 — 8 95 (continues)

A-1

10/2/07

3:33 PM

Table 1.4

Page 24

continued First Round

Second Round

Third Round

Fourth Round

Dominican Republic Universe Sample Universe Sample Universe Sample Universe Sample PRD PRSC PLD BIP-PPC Other partiesa Total El Salvador ARENA FMLN PCN PDC CDU Other partiesa Total Ecuador PSC PRE DP MUPP-NP ID PRIAN PSP Other partiesa Total Guatemala

57 50 13 — — 120

29 20 13 — — 62

PL PN PINU PUD Other partiesa Total

47 11 34 7 4 103

72 36 42 — — 150

57 29 32 — — 118

Universe Sample Universe Sample Universe Sample Universe Sample 39 21 4 18 — 2 84

18 15 2 11 — 0 46

28 27 11 8 — 10 84

16 20 8 6 — 8 58

31 29 14 — — 10 84

20 25 11 — — 8 64

27 31 16 5 5 — 84

26 29 15 5 5 — 80

Universe Sample Universe Sample Universe Sample Universe Sample 26 20 12 8 — — — 16 82

24 15 10 8 — — — 15 72

28 24 34 7 18 — — 10 121

26 22 31 7 17 — — 9 112

24 15 — 11 15 10 6 19 100

23 14 — 11 15 10 6 19 98

Universe Sample Universe Sample Universe Sample Universe Sample

PAN FRG URNG UNIONISTA UNE GANA PSN PATRIOTA Integracionistas Other partiesa Total Honduras

67 17 49 10 6 149

44 18 — — — — — — — 18 80

35 14 — — — — — — — 14 63

19 63 5 10 6 — — — — 10 113

13 44 4 7 4 — — — — 7 79

14 31 — 5 29 34 7 9 8 21 158

11 23 — 4 22 26 6 7 5 17 121

Universe Sample Universe Sample Universe Sample Universe Sample 71 55 2 — — 128

35 30 2 — — 67

67 55 — — 6 128

37 30 — — 4 71

55 61 — 5 7 128

42 48 — 5 7 102 (continues)

A-1

10/2/07

3:33 PM

Page 25

STUDYING PARLIAMENTARY ELITES IN LATIN AMERICA Table 1.4

25

continued First Round

Mexico PRI PRD PAN Other partiesa Total Nicaragua

300 71 119 10 500

64 23 35 1 123

239 125 121 15 500

60 31 31 4 126

ANR PLRA Alianza UNACE MPQ PPS Other partiesa Total Peru Cambio 90/NM UPP APRA (PAP) Perú Posible UN FIM Other partiesa Total

Fourth Round

210 52 207 31 500

52 13 51 8 124

224 95 153 28 500

57 22 38 7 124

Universe Sample Universe Sample Universe Sample Universe Sample 42 36 15 93

32 27 11 70

48 38 6 92

27 29 4 60

Universe Sample Universe Sample Universe Sample Universe Sample

PRD PA SOLIDARIDAD Other partiesa Total Paraguay

Third Round

Universe Sample Universe Sample Universe Sample Universe Sample

PL FSLN Other partiesa Total Panama

Second Round

34 22 — 15 71

30 20 — 14 64

42 16 8 12 78

36 14 7 11 68

Universe Sample Universe Sample Universe Sample Universe Sample 38 33 — — — — 9 80

20 21 — — — — 6 47

45 — 35 — — — — 80

36 — 29 — — — — 65

37 21 — 10 10 2 — 80

23 16 — 7 8 2 — 56

Universe Sample Universe Sample Universe Sample Universe Sample 67 17 8 — — — 28 120

52 12 7 — — — 16 87

— 6 28 45 17 11 13 120

— 3 22 28 13 6 11 83

Source: Created by authors based on PELA 1994–2005. Note: a. “Other parties” refers to parties with fewer than five legislators in the lower legislative chamber.

A-1

10/2/07

3:33 PM

26

Page 26

POLITICIANS AND POLITICS IN LATIN AMERICA

■ Notes 1. The research group is directed by Manuel Alcántara Sáez and is linked to the Instituto Interuniversitario de Iberoamérica as well as to the University of Salamanca’s Department of Political Science. From the beginning, there has been participation by researchers from these institutions’ master’s programs in Latin American Studies and their doctoral programs in Contemporary Political Processes, as well as by other researchers from different universities, both in Spanish and elsewhere. 2. SEC94/0284. 3. SEC95/0845 and SEC02/3484. A fourth round of information has been collected for the cases of Mexico and El Salvador. 4. Thelen and Steinmo (1992); Rohrschneider (1994); Colomer (2001). 5. SEC94/0284. 6. SEC95/0845 and SEC02/3484. 7. Crespo (1997 and 1998); Freidenberg (1999); Ajenjo (1999); Corral (2004); Otero (2004); Díaz y Linares (2005). 8. Martínez Barahona (2000). 9. Del Campo and Martínez Barahona (2000). 10. García Díez (2000); Ruiz Rodríguez and García Montero (2003); Morgenstern (2003); Alcántara and Luna (2004). 11. Martínez (1997); Alcántara and Freidenberg (1997); Alcántara and Llamazares (1997); del Campo and Ramos (1997); Alcántara (1999, 2000b, and 2004); Alcántara and Ramos (2000); Morgenstern (2000); Martínez Rosón (2004). 12. Crespo (1997); Freidenberg (2000). 13. Steenbergen and Jones (2002). 14. We have not included the Brazilian case. Interviews in Brazil are now being processed; the practical challenges prevented us from conducting field research there previously.

A-2

10/8/07

10:56 AM

Page 27

PART 1 Politics and Democracy

A-2

10/8/07

10:56 AM

Page 28

A-2

10/8/07

10:56 AM

Page 29

2 The Conundrum of Representation Patricia Marenghi and Mercedes García Montero

THE PROBLEM OF REPRESENTATION HAS BEEN CENTRAL IN DISCUSSIONS

on the duties that parliaments carry out and the way in which legislators behave. In representative democracies, the matter of the tie established between politicians and citizens has been the source of strong controversies that are difficult to resolve from a theoretical point of view as well as within the framework of empirical studies. In theoretical terms, the conceptions of representation and the different definitions of democracy do not have a harmonious coexistence.1 Difficulties in the empirical approach arise for two main reasons. One of these is that most studies have concentrated on the way in which electoral systems distribute quotas and spaces of power, thus generating different situations of proportionality. The second reason is that so few analyses exist on the attitudes, opinions, and behaviors of representatives that one can draw only modest conclusions from them. Research on representation carried out in the last half of the twentieth century studies the ties established between voters and politicians, but the diversity of perspectives employed makes a true comparative approach impossible. This type of study can be divided into two major bodies of research: those that deal with procedural representation and those that concentrate on substantive representation.2 Most studies of this type have approached this problem from a perspective based either on districts (Miller and Stokes, 1963) or on parties (Italy: Barnes, 1977; Germany: Farah, 1980; Holland: Irwin and Thomassen, 1975; France: Converse and Pierce, 1986; Australia: McAllister, 1991; Sweden: Holmberg, 1989, 1999, 2000; Esaiasson and Holmberg, 1996; Norway: Matthews and Valen, 1999). The first wave of research of this type consisted of case studies, followed (in the past few decades) by comparative studies (Dalton, 1985; Miller, 1999; Schmitt and Thomassen, 1999; Kitschelt, Mansfeldova, Markowski, and Toka, 1999; Esaiasson and Heidar, 2000).3 There was also occasional early research on the collective perspective, which looks at legislative representa-

29

A-2

10/8/07

10:56 AM

30

Page 30

POLITICS AND DEMOCRACY

tion as a whole (Wiessberg, 1978), but this vein was not continued in later research.4 Other lines of research that could be categorized as belonging to the perspective of substantive representation are those that aim to decipher the ties between party commitments—electoral promises or party manifestos— and voters (Klingemann, Hofferbert, and Budge, 1993; Stokes, 1999 and 2001) or to explain situations of accountability and retrospective vote and the conditions in which voters punish governments for their performance (Powell and Whitten, 1993; Cheibub and Przeworski, 1999; Przeworski, Stokes, and Manin, 1999; Maravall, 2003). Other types of studies that can be included here are those that, from a different perspective of political representation, emphasize the connection between the sociodemographic characteristics of elites and those of voters, such as gender, race, or social class (Norris, 1996; Vowles, Aimer, Banducci, and Karp, 1998). As can be seen from this brief summary, most studies on political representation are on European and US democracies. Political representation among Latin American parliamentary elites is still practically unexplored, both in theoretical and empirical terms. There are a few exceptions. One is Guillermo O’Donnell’s work (1994) and his description of delegative democracies. Another is Jorge Domínguez’s analysis (1998), in which the author states that although damage can be caused to democracies by politicians’ not keeping their electoral promises, this damage is mitigated by the fact that citizens can judge politicians at the end of their terms (accountability). In addition, there are Susan Stokes’s studies (1999 and 2001), in which both approaches are combined. Last of all, the more recent work of Juan Pablo Luna and Elizabeth Zechmeister (2005) is noteworthy for its attempt to systematize information on citizens’ preferences and Latin American representatives. They created indicators that make the positions of citizens and legislators comparable on different issues, then measured the degree of agreement between these positions. Aside from these studies, few scholars have manifested interest in this area.5 Establishing the degree of congruence or incongruence that exists between the preferences of the representative and those of the people they represent is important in determining how close these two groups are. However, in order to do this, one must first understand the way in which representatives perceive their functions and roles.6 In order to measure the type of relationship generated, it is fundamental first to find out whom representatives feel they represent, what interests they are willing to defend, and what instructions, if any, they follow. In almost all Latin American countries, legislators, although elected by a particular electoral district, are considered, explicitly or implicitly, to have a national mandate, one that is independent and formally excludes imperative representation (see Table 2.1).7 However, underneath the shared formal institutions are differences so great they cannot

A-2

10/8/07

10:56 AM

Page 31

31 Table 2.1

Country

Constitutional Regulations on Representation

National Mandate

Constitutional Article

Argentina



Article 44: The legislative power of the nation shall be vested in a Congress composed of two Houses, one of deputies of the nation and the other of senators for the provinces and for the city of Buenos Aires.

Bolivia Brazil



Article 45: The House of Representatives is composed by the representatives of the people, elected by proportional representation in each State and Territory and in the Federal District.

Chile Colombia



Costa Rica



Article 133: The members of collegiate bodies of direct election represent the people, and must act with consideration for justice and the common good. Article 106: The representatives have that character for the nation and shall be elected by provinces.

Dominican Republic Ecuador



El Salvador



Guatemala Honduras



Mexico



Nicaragua



Article 135: Legislators will act based on national considerations, and will be politically responsible to society for the fulfillment of the duties invested by their office. Article 125: Legislators represent the people as a whole.

Prohibition of Imperative Mandate



Constitutional Article

Article 125: Representatives . . . are not bound by an imperative mandate.

Article 202: The Legislators will represent the people; their distribution across departments will be based on the quotient indicated by the National Electoral Court. Article 51: The House of Representatives will be formed by the representatives of the Nation. Article 2: Political power is exercised by the people through its freely elected representatives. Article 132: The legislative power is exercised by the National Assembly by delegation and mandate of the people. (continues)

A-2

10/8/07

10:56 AM

32

Page 32

POLITICS AND DEMOCRACY

Table 2.1

continued

Country

National Mandate

Panama



Paraguay



Constitutional Article Article144: Legislators will act in the interest of the Nation, and represent in the Legislative Assembly their respective political parties and the voters of their Electoral Districts.a Article 221: The House of Representatives is the Chamber of departmental representation.a

Prohibition of Imperative Mandate



Peru



Article 93: Representatives of Congress represent the Nation.



Uruguay Venezuela



Article 201: Legislators are the representatives of the people and of the states as a whole.



Constitutional Article

Article 201: Senators and Representatives are not subject to imperative mandates. Article 93: Members of congress . . . are subject neither to an imperative mandate nor to interpellation. Article 201: Legislators . . . are subject neither to mandates nor to instructions but rather to their consciences alone. Their vote in the National Assembly is personal.

Source: Based on the constitutions of each country. Note: a. The cases of Paraguay (with representation of departments) and Panama (with national mandate but party and district representation) are shown in italics.

go unnoticed. Among the factors that condition legislators’ perceptions of their representative roles are formal institutions, historical legacies, social cleavages, incentives that motivate actors, and the existence of certain types of informal practices—to name only a few. Along these lines, this chapter addresses two general areas. First, it aims to describe how legislators perceive their roles as representatives and, second, to explain, in each case, what factors influence the configuration of these roles. These factors, in turn, can result in different forms of behavior in each representative.8

■ The Representative Role: Focus and Style This chapter looks at the representative role of legislators, which is defined on the basis of two key dimensions: what interests legislators feel they should defend (focus of the representation) and the way they believe they

A-2

10/8/07

10:56 AM

Page 33

THE CONUNDRUM OF REPRESENTATION

33

should behave (style of the representation) (Eulau, Wahlke, Buchanan, and Ferguson, 1959; Thomassen, 1994).9 The dimension of focus has to do with the interests that representatives prioritize while carrying out their jobs. We can start out by dividing these into general interests (those of all citizens) and specific interests (those of a particular group). If we analyze the relationship “voters-representatives” in terms of the principal-agent theory and in the context of a particularized conception, the agent would represent and respond to a concrete principal (political party, territorial subunit, social group, ethnic group, interest group, or group of interest groups). Thus, to refine our analysis, this chapter will examine two types of focus: territorial and party. The dimension of style has to do with the traditional debate over representatives’ behaviors. This debate originates in the transition from the old regime—the imperative mandate, class instructions, or representation—to the modern one. This dimension seeks to analyze, first and foremost, if the legislators are in favor of following instructions (regardless of their source) or if they believe in following their own individual preferences. The second phase of this analysis enumerates the attitudes of those who tend to follow instructions; this phase seeks to determine what groups these representatives answer to, and to observe if their relationships with their constituents follow mandated patterns. Both of these dimensions have been measured according to indexes describing the perceptions, attitudes, and motivations of Latin American representatives. The information used is drawn from the study Parliamentary Elites in Latin America (see the chapter appendix). These indicators range from 0 to 6, where 0 indicates standpoints in favor of defending particular interests (in focus) and attitudes favorable toward following instructions (in style). Values closer to 6 correspond to legislators who declare themselves in favor of defending general interests (in focus) and of following their own criteria (in style). See Figure 2.1. Focus and Style: The Interests Legislators Defend and to Whom They Listen

Before we began the analysis and showed the position of Latin American representatives on these two indexes, we carried out a Pearson correlation analysis. This analysis supports the validity of separating the two dimensions, for it demonstrates that their relationship is not statistically significant and, thus, they portray two different and independent aspects of the representative role. In turn, the indicators used to construct each dimension are related to each other, with statistically significant correlations.10 Focus and style in Latin American countries. Once they have been transformed, the indicators allow us to see legislators’ positions, by country as

10/8/07

10:56 AM

34

Page 34

POLITICS AND DEMOCRACY Figure 2.1 Focus and Style in the Representative Role Own criteria

6

Focus

0 Particular interests

Following instructions

6

Style

A-2

General interests

0

Source: Created by the authors based on Thomassen 1994.

well as by party, in terms of the focus and style indexes. With respect to the focus index, Latin American countries occupy a middle ground between the defense of particular and general interests, with the means falling between 2.40 and 3.97. It is striking that, despite the fact that legislators in almost all Latin American countries have a national mandate, the means in the index fall in the middle. The countries with the highest average scores in this dimension are Mexico, Ecuador,11 the Dominican Republic, and Costa Rica. On the opposite end are Chile, Bolivia, Uruguay, and Colombia, whose legislators are close to the defense of particular interests (whether they be party or territorial) (see Figure 2.2). Despite the fact that most of the countries demonstrate average values on the index, a large variety of opinions exists within country. If the standard deviations shown are analyzed, the legislators of Argentina, Colombia, and Chile show the widest divergence of opinions, whereas the legislators from Uruguay, Mexico, and Nicaragua have more homogeneous views. The same phenomenon observed above, in the focus dimension, can be noted in that of style, as well: the countries are located in an intermediate position—between 1.64 and 3.2—on the index that measures whether the legislators lean more toward following instructions or their own criteria. In this case, legislators of the Dominican Republic, Ecuador, and Uruguay are more disposed to follow instructions; those most inclined to follow their own criteria are the legislators of Chile, Peru,12 and Bolivia. With respect to homogeneity of opinions within the group, the Dominican representatives display the greatest divergence of opinions, followed by the Venezuelans and the Nicaraguans. The representatives whose opinions are the most similar, on the other hand, are the Costa Ricans, the Salvadorans, the Uruguayans, the Chileans, and the Mexicans.

10/8/07

10:56 AM

Page 35

THE CONUNDRUM OF REPRESENTATION

35

Figure 2.2 Spatial Distinction Among Latin American Countries in the Index of the Focus and Style Dimensions

Chile

Bolivia

Focus

Nicaragua Colombia Argentina Uruguay

Honduras

Mexico Venezuela Paraguay

El Salvador Costa Rica

Ecuador

Style

A-2

Dominican Republic

Source: Created by the authors based on PELA 1994–2005.

Two interesting conclusions arise from reading the national aggregate data. The first is that, although they have a national mandate, Latin American representatives show a strong inclination toward particular interests. Although the mean for focus in the whole region is 3.37 (in the middle, but leaning toward the defense of the nation as a whole), a great number of representatives conceive of their role within territorial or party frameworks (whether these frameworks be their parties’ voters or the platforms on which they have presented themselves for elections). The second conclusion we can draw is that the majority of the representatives put more emphasis on representation by mandate than on independent representation (the average for style in the whole region is 2.76). Acting according to the voters’ opinions, political parties, public opinion, or any other social or interest group—or at least taking one of these factors into account—seems to be the trend among legislators, more than following their own opinions on what is best for society. Focus and style in political parties. The values for the focus index by political party demonstrate a wider diversity than that shown by country, with positions ranging between 2.05 and 4.16 (see Table 2.2). The parties whose legislators are more in favor of following particular interests are the

A-2

10/8/07

10:56 AM

Page 36

36 Table 2.2

Mean Values and Standard Deviations in the Style and Focus Indexes Style

Country/Party

Mean

Standard Deviation

Argentina PJ UCR FREPASO Bolivia ADN MNR MIR UCS CONDEPA Chile PDC RN UDI PPD PS Colombia PL PC Costa Rica PUSC PLN Dominican Republic PRD PLD PRSC BIP Ecuador PSC DP PRE ID MUPP-PP El Salvador ARENA FMLN PCN Honduras PLH PNH Mexico PAN PRD PRI Nicaragua PL FSLN

2.86 2.67 2.89 3.38 3.18 2.9 3.06 3.49 3.14 3.51 3.2 2.9 3.24 3.68 3.21 3.21 2.9 2.9 2.91 2.83 2.86 2.79 1.64 1.17 2.12 1.43 2.69 2.3 1.67 2.85 2.23 2.29 2.49 2.82 3.11 2.7 2.62 2.77 2.94 2.56 3.05 3.09 3.03 3.01 3 3.49 2.49

0.99 1.11 0.75 0.79 0.98 1.14 0.88 0.67 1.24 0.61 0.77 0.49 0.8 1.00 0.78 0.78 1.11 1.13 1.10 0.67 0.75 0.6 1.31 0.99 1.37 1.55 1.24 1.09 1.07 0.79 1.26 0.83 1.22 0.69 0.94 0.80 0.99 0.97 0.97 0.96 0.80 0.71 0.82 0.89 1.12 0.94 1.07

Focus

n

Mean

Standard Deviation

105 55 32 18 89 24 20 16 16 13 78 28 16 15 11 8 71 45 26 41 21 20 96 45 33 11 7 93 24 27 20 15 7 49 16 23 10 67 37 30 104 44 11 49 49 25 24

3.37 3.44 3.03 3.79 2.62 2.65 2.42 2.49 2.72 2.95 2.4 2.69 2.31 2.12 2.5 2.05 3.03 3.02 3.05 3.74 3.89 3.57 3.77 3.71 3.83 3.87 3.71 3.82 3.66 3.94 4.16 3.97 2.85 3.67 3.81 3.72 3.29 3.31 3.02 3.66 3.97 4.04 3.67 3.97 3.39 3.73 3.01

1.61 1.54 1.71 1.56 1.59 1.62 1.63 1.67 1.51 1.66 1.55 1.54 1.57 1.57 1.65 1.56 1.55 1.57 1.56 1.17 0.75 1.52 1.17 1.28 0.98 1.16 1.37 1.39 1.48 1.43 0.75 1.49 1.88 1.23 1.05 1.33 1.30 1.35 1.48 1.09 1.06 1.07 1.37 1.00 1.04 1.20 1.52

n 105 53 34 18 90 24 20 17 16 13 79 27 16 17 11 8 71 45 26 40 21 19 96 46 33 11 6 81 23 24 15 13 6 47 16 22 9 67 37 30 108 47 10 51 51 27 24 (continues)

A-2

10/8/07

10:56 AM

Page 37

THE CONUNDRUM OF REPRESENTATION Table 2.2

37

continued Style

Focus

Country/Party

Mean

Standard Deviation

n

Mean

Standard Deviation

n

Paraguay ANR PLRA PEN Peru UN FIM PP PAP Uruguay EP/FA PC PN Venezuela MVR ADN MAS COPEI

2.9 2.92 2.96 2.71 3.2 3.96 3.17 3.3 2.68 2.64 2.58 2.72 2.67 2.92 2.93 2.71 3.17 2.88

0.93 0.98 1.00 0.41 1.09 0.78 1.14 1.07 1.03 0.71 0.81 0.64 0.62 1.15 1.16 1.18 1.20 0.96

61 35 19 7 64 11 6 26 21 62 27 21 14 73 43 15 11 4

3.61 3.64 3.49 3.83

1.13 1.10 1.20 1.21

65 35 21 8

2.92 2.96 3.35 2.22 3.58 3.71 3.46 3.44 3.27

0.66 1.51 1.80 1.54 1.14 1.04 1.19 1.26 1.58

59 24 21 14 76 43 15 12 6

Source: Data based on PELA 1994–2005.

PS, the UDI, the RN, and the PPD of Chile; the PN of Uruguay; and the MNR and MIR of Bolivia. On the other end of the spectrum are the parties whose legislators more strongly support acting as representatives of general interests. The three highest places in this index are occupied by three Ecuadorian parties (the PRE, the ID, and the DP)13 and two from Mexico (the PRI and the PAN). However, an examination of this table also shows that there are small interparty differences within the Chilean, Bolivian, Colombian, and Dominican parliaments. As regards homogeneity shown by parties in this index, the Costa Rican PUSC and the Ecuadorian PRE have the lowest standard deviations; the opinions of their representatives on what interests they represent, therefore, are the most homogeneous of all parties. These two parties are followed by the Dominican PLD, the Mexican PRI, and the Venezuelan MVR, with similarly low standard deviations. The party whose representatives show the most divergence among themselves is the MUPP-NP of Ecuador, followed by the PC of Uruguay and the UCR of Argentina. The values of the political parties in the style index vary between 1.17 and 3.68. As is the case in the focus index, various parties within the same party system share similar values in the style index. Three Dominican parties (PRD, PRSC, and PLD) and three Ecuadorian parties (PSC, PRE, and

A-2

10/8/07

10:56 AM

38

Page 38

POLITICS AND DEMOCRACY

ID) stated that they follow instructions in their representative role. Within the parties whose legislators say that they are guided more by their own criteria, the ones that stand out are the Chilean UDI, the Bolivian CONDEPA and MIR, and the Nicaraguan PL. The representatives within the same party who have the greatest similarity of opinions on style are those of the Paraguayan PEN, the Chilean PDC, the Costa Rican PLN, and the Bolivian CONDEPA. Those who display the widest divergence in their answers, on the other hand, belong to the Dominican parties (PLD and PRSC), followed by the Ecuadorian PRE and the UCS of Bolivia. This “still photo” of the style and focus of representation in Latin American countries, measured at the aggregate level by political party, allows us to extract some models that will be explored at the end of the chapter, as we look for explanations for each case of representative patterns. Two observations are relevant concerning the focus of the representation: on one hand, the means of party organizations within national political systems are very similar in almost all cases; on the other hand, the standard deviations are high (that is, within the parties, representatives’ opinions are relatively varied). Regarding the style of the representation, the means in the parties’ positions within national political systems do not differ considerably, just as is true in the case of focus; however, the standard deviations are low (that is to say, opinions within the parties are more homogeneous than not). These considerations suggest the presence of different causal patterns in each dimension. See Figure 2.3. Focus: parties or territories. The long tradition of studies on the political

preferences of citizens and their link with political parties is based on the presumption that party organizations are largely in charge of selecting and “presenting” the demands of the voters. Among the functions of political parties is that of uniting apparently different demands in relatively coherent packages and creating predetermined offers with which it is possible to compete in elections, offering voters a political agenda that is more or less crystallized. This function appears explicitly in the institutional designs of some Latin American countries, as can be seen in Table 2.3, thus establishing that political parties are the only actors capable of competing in elections. This scenario indicates the interest of institutions in ordering political competition by distinguishing clearly between actors and avoiding high fragmentation so as to maximize consent and guarantee governability.14 It also presupposes the existence of organizations capable of processing their differences internally and of acting in the public arena in a disciplined manner. Moreover, this scenario presumes the existence of party loyalties between political elites and the desire of parties to collaborate in governing

10/8/07

10:56 AM

Page 39

THE CONUNDRUM OF REPRESENTATION

39

Figure 2.3 Spatial Distinction Among Latin American Political Parties in the Index of the Focus and Style Dimensions

Focus

Style

A-2

Source: Created by the authors based on PELA 1994–2005.

the nation. However, an examination of the answers contained in the focus index and in response to other questions in the PELA questionnaire (1994–2005) reveals that the party conception of “representation” is very weak in almost all the countries of the region. Figure 2.4 shows responses to the question “In your opinion, whom do you represent during your parliamentary term?” In only two cases is the answer “The party and its voters—to which the deputy belongs” chosen over “The voters of the district/department/province—in which the deputy was elected.” These countries are Uruguay and Costa Rica. The case of Costa Rica does not carry as much weight in this context, since its legislators generally tend to conceive of representation in terms of the nation as a whole. However, the case of Uruguay stands out because it is one of the countries in which particular interests have a crucial relevance. Legislators’ views of the importance of getting resources for the subnational unit that elects them are further indicators that most legislators who defend particular interests conceive of their representative role within a territorial framework. As can be seen in Figure 2.5, almost all the representa-

A-2

10/8/07

10:56 AM

40

Page 40

POLITICS AND DEMOCRACY

Table 2.3

Country

Countries in Which the Right to Participate in Elections Is Offered Only to Political Parties Constitution

Electoral Law

Bolivia

To be a legislator, one of the prerequisites Articles 86 and 105. is to be “nominated by a political party or by civic associations that are representative of the powers that be of the country, that have legal personality and that form blocks or alliances with the political parties” (Articles 60, 61, and 64). Brazil Affiliation with a political party is among the conditions for eligibility (Article14). Ecuador A member of the National Congress must be affiliated with one of the legally recognized political parties. The candidate must be sponsored by the same political party (Article 47). Honduras “The political parties and independent candidates constitute the forms of organization and the means of political participation of the citizens” (Article 4). Panama “Only political parties may nominate candidates for Legislator” (Article 141). Source: Elaborated by the authors based on the norms of each country.

tives of a region are “very” or “quite” interested in getting resources for their provinces or departments (82.9 percent). This indicator by itself does not confirm the hypothesis of the absence of a partisan logic of representation, since both questions (party interest vs. departmental/regional interest) are not mutually exclusive. However, if the legislators who answered “very important” or “quite important” to the previous question (on the importance of getting resources for the subnational unit that elected them) are asked how they would vote if there were a conflict between the needs of their state and the party position, we can determine the importance of these options in the stance of the representative (see Figure 2.6). The results show that, for all Latin American countries, the representative’s department is more important than his or her party (72.4 percent for the whole region).15 The only exception is the case of Uruguay, where the party is a more relevant actor than the region when legislators are deciding what position to take. The information contained in Figure 2.7 reinforces the above argument. The figure refers to the question “How important are the following for you as a representative: to defend the interest of your party? to defend the interest of your state?” Taking into account only those who answer “very important,” a greater tendency toward territorial representation than toward the

A-2

10/8/07

10:56 AM

Page 41

THE CONUNDRUM OF REPRESENTATION

41

Figure 2.4 In Your Opinion, Whom Do You Represent During Your Parliamentary Term? Question: Of the following options that I am now going to read to you, please indicate whom in your opinion you represent during your parliamentary term: all voters of your party, all of your district/department/province, the political party you belong to, all citizens of your country?

Dominican Republic

Source: Created by the authors based on PELA 1994–2005.

defense of party interests can be observed (except in the case of Uruguay). This question, however, allows us to see that representatives hold both objectives simultaneously. The coexistence of these two systems of logic (among others) reveals the complex nature of political decisions that legislators must make. Style: instructions. This section analyzes what kind of instructions repre-

sentatives follow, namely: to what extent do legislators take into account the opinion or interests of certain groups, persons, or institutions when they make these decisions, and from whom or from where do these instructions come?16 As previously pointed out, there are important cross-national variations in legislators’ answers. This makes it necessary to construct a table reporting legislators’ views on the actors they take into consideration when making political decisions (see Table 2.4).

A-2

10/8/07

10:56 AM

42

Page 42

POLITICS AND DEMOCRACY Figure 2.5 Importance of Getting Resources for Your Province/State or Department

Question: In your opinion, how important is it to get resources for your province/state/department during your parliamentary term: very important, quite important, not very important, or not at all important?

Source: Created by the authors based on PELA 1994–2005.

A general examination of the answers obtained clearly shows that representatives claim to take into account all of the actors listed, except for interest groups, to a high or very high degree. The groups most taken into account are the voters of the representative’s electoral district (in all countries, this is the group most taken into account by legislators when making political decisions), the party’s voters, public opinion in general, and party leaders. The Dominican Republic shows the highest ratings for most of the groups and institutions asked about; Honduras and El Salvador share second place in this respect. The representatives from Honduras accord the most weight to the party and to interest groups, while those from El Salvador are the most open to the opinions of party leadership, the media, and public opinion in general. In Argentina and El Salvador, legislators assign the least importance to the opinions of voters of their electoral districts. Peruvian and Colombian

A-2

10/8/07

10:56 AM

Page 43

THE CONUNDRUM OF REPRESENTATION

43

Figure 2.6 In the Case of a Conflict, How Would You Vote? Question: When there is a conflict between the needs of your state/province/department and the positions of your political party, how do you usually vote: always with your party or according to the needs of your state/province/department?

It depends on the issue

Always according to the needs of your state/province/department

Always along party lines

Source: Created by the authors based on PELA 1994–2005.

representatives are the least open to opinions of party legislators and, along with Chile, to those of party members. Chilean legislators pay the least amount of attention to public opinion in general (followed by Costa Rica, where the mean is also low on the importance given to the media). Finally, interest groups are almost irrelevant in the cases of Uruguay, Bolivia, and Chile. Since, in general, representatives are the most attentive to voters of their own district, to discover more subtle differences among the styles of each country’s legislators we will concentrate our analysis on the relevance of party actors as an influence on representatives’ decisions. Instructions by political parties are most taken into account in the Dominican Republic, Honduras, El Salvador, and Uruguay. On the opposite end of the scale— party instructions have the least amount of influence—are Chile, Peru, and Colombia (see Figure 2.8). With the exception of Chile, in all countries, party members have more influence than party leadership. In Ecuador and Costa Rica, the weight of both groups is similar.

A-2

10/8/07

10:56 AM

44

Page 44

POLITICS AND DEMOCRACY Figure 2.7 The Degree of Importance Assigned to the Defense of Party and Territorial Interests (percentagea)

Question: With respect to your job as a legislator, how important are the following aspects to you: very important, quite important, not very important, or not at all important?

Source: Created by the authors based on PELA 1994–2005. Note: a. Percentages are the sum of the “very important” answers.

■ Toward an Explanation of the Representative Role We have examined, by political party and by country, how Latin American legislators position themselves in the focus and style dimensions of representation. However, further questions should be addressed. What factors make legislators decide to follow their own criteria or, conversely, to follow instructions? What variables influence them to declare themselves in favor of defending particular interests or general ones? Are these variables the same in every Latin American country? And, finally, do the same factors condition the style and the focus of representation? Almost all political phenomena result from the interaction of multiple variables that, because of their changing nature and because they do not easily lend themselves to quantification, are quite often difficult to pin down. This is especially true when what one is attempting to explain is related to something as imprecise and as symbolic as political attitudes.

A-2

10/8/07

10:56 AM

Page 45

THE CONUNDRUM OF REPRESENTATION Table 2.4

45

Influence of Different Actors on the Political Decisions Made by Legislators (percentagea)

District Voters Dominican Republic Chile

Party Leaders

100.0 98.8

Dominican Republic Honduras

97.0 92.5

Public Opinion in General El Salvador

96.4 86.9 86.6 86.2 84.1 83.1 82.4 77.7 76.9 76.1 74.1 72.7 69.0 65.4 55.4

Honduras Paraguay Ecuador Peru

98.5 95.4 94.2 92.8

Uruguay Costa Rica El Salvador Nicaragua

90.8 88.1 80.4 79.7

Dominican Republic Argentina Paraguay Peru Nicaragua

Venezuela Colombia Nicaragua Costa Rica Mexico Uruguay Bolivia Argentina El Salvador

91.4 89.6 88.1 88.1 87.9 86.2 84.6 83.0 80.4

Bolivia Argentina Paraguay Ecuador Chile Mexico Peru Colombia Venezuela

79.1 74.1 72.3 71.8 71.1 69.8 63.8 61.0 50.6

Bolivia Ecuador Uruguay Honduras Mexico Colombia Costa Rica Venezuela Chile

Mass Media

Party Legislators

Dominican Republic El Salvador Bolivia

81.8 71.4 65.9

Ecuador Paraguay Nicaragua Peru Honduras

65.0 49.2 47.5 46.4 44.8

Argentina Mexico Venezuela Colombia Uruguay Costa Rica Chile

44.6 41.4 40.7 37.7 21.5 19.0 18.1

Honduras

97.0

Paraguay

86.2

El Salvador Uruguay Nicaragua Dominican Republic Argentina Costa Rica Bolivia Mexico Venezuela Ecuador Chile Colombia Peru

85.7 81.5 81.4

Interest Groups

El Salvador

80.4

Costa Rica

76.2

Costa Rica Venezuela

69.0 66.7

74.6

Argentina Chile Nicaragua Uruguay Dominican Republic Mexico Honduras Paraguay Ecuador Bolivia Colombia Peru

64.3 63.9 61.0 60.0

Honduras Dominican Republic Nicaragua Mexico El Salvador Venezuela Argentina Paraguay Ecuador Colombia Peru Uruguay Bolivia Chile

58.6 56.9 55.2 49.2 45.6 39.6 35.1 30.4

Party Militants

80.8 75.9 71.4 67.0 62.1 59.3 52.4 47.0 46.8 43.5

Voters Dominican Republic Honduras Uruguay

98.0 97.1 92.3

73.7 72.9 49.1 48.2 45.7 44.6

El Salvador Argentina Paraguay Mexico Bolivia

87.5 83.9 83.1 82.8 80.2

36.9 28.2 27.3 26.1 21.5 18.7 16.9

Nicaragua Venezuela Costa Rica Colombia Peru Ecuador Chile

79.7 76.5 69.0 67.5 63.8 63.1 49.4

Source: Data based on PELA 1994–2005. Note: a. Percentages are the sum of the “very” and “quite important” categories.

Thus, the following explanations, rather than proposing single causes, attempt to consider an inclusive gamut of explanatory factors (see Tables 2.5 and 2.6). In spite of our attempt to be inclusive, we have omitted a wide spectrum of relevant factors. These factors are mainly variables tied to historical development, to the particular territorial development of each case,

A-2

10/8/07

10:56 AM

46

Page 46

POLITICS AND DEMOCRACY Figure 2.8 Influence of Political Parties, Militants vs. Leaders

Question: When making political decisions, to what degree—very high, high, not very high, or not at all—do you take into account the opinions of the following groups, people, or institutions: party leaders and other representatives of your party (party leadership) and party members and voters (party membership)?

Source: Created by the authors based on PELA 1994–2005.

and to the presence of specific social and historical cleavages. The analysis of these factors requires case studies that should delve deeper into the special characteristics of national political systems. In this chapter, as seen in Table 2.5, the analysis of style is concentrated on the relevance of four groups of explanatory factors: (1) perceptions of the importance, functioning, and organization of political parties; (2) party ideology; (3) political career; and (4) systemic characteristics. For this reason, the explanatory potential of the independent variables is uneven when examined country by country. In the case of focus, the predictors are fewer and are grouped into three categories: party organization, political system, and political situation. In order to contrast these factors’ relationship and weight in the representative role of Latin American legislators, stepwise multiple regressions have be carried out. The focus and style dimensions for the whole region are

A-2 10/8/07

Table 2.5

Explanatory Variables Introduced into the Analysis: Style Dimension

About the political parties of the country in general

Opinion on the degree to which party leadership should influence legislatorsa Necessity of parties to democracya Whether the seat belongs to the party or to the legislatora Level of popular participation in partiesa Level of activism of militants within partiesa

About the legislator’s own political party

Relationship between parties and societya Internal democracya Level of activism of legislators’ own partiesa Level of voter participation in legislators’ own partiesb The existence of different party factionsb

Hypothesis Legislators who believe that party leaders should have less influence on representatives have a representative style that tends toward following their own criteria.

Page 47

Perceptions of the importance, the functioning, and the organization of political parties

Variable

10:56 AM

Type of Variable

Legislators who believe that parties are necessary to democracies are more open to following instructions (from the party). Legislators who believe that the seat belongs to the political party and that therefore when a representative leaves the party s/he should give up his/her seat have a style which is favorable to following instructions. The lower the popular participation of voters in party life, the higher the legislator’s inclination will be to follow his or her own criteria in his or her representative style. The level of activism of party militants of a country determines the inclination of its legislators to follow either instructions or their own criteria; the lower the level of activism of the party, the higher will be the inclination of legislators to follow their own criteria. Legislators who see parties as out of touch with society have more of an inclination to follow their own criteria and not follow instructions (of the party). Legislators who feel the level of democracy within their political party is low have more of an inclination to follow their own criteria. The higher the level of activism of party militants, the higher the inclination of its legislators to follow instructions. Legislators who feel that popular participation in their political party is high will tend to be more open to following instructions. Legislators who belong to political parties within which there are different party factions are more inclined to follow their own criteria. (continues)

47

A-2

Ideological self-placementa

Political career

Extreme ideological selfplacementa Experience (reelection)a Reason for election as a legislatora Years of experience in politicsa

Political system

Experience in other political officesa Effective number of parliamentary parties Opinion on the stability of democracya

Legislators who identify themselves ideologically as on the left are more independent than those on the right, who are more open to following instructions. Legislators who belong to parties that locate themselves ideologically on the left are more independent than those on the right. Legislators who locate themselves on ideological extremes, whether it be right or left, have a style which is favorable to following instructions. Legislators who have been elected more than once and who therefore have more experience have more of a tendency to follow their own criteria. Legislators who feel that they have been elected for individual reasons, whether because of their experience or because of their personal characteristics, have a more independent style than those who feel that they have been elected for reasons to do with the party to which they belong. The more years of experience the legislator has had, measured in terms of the years s/he has been in politics, the more of a tendency s/he will have to follow his or her own criteria in terms of style. The more experience the legislator has in other electoral offices, the more inclined s/he will be to follow instructions. In countries where the number of parties present in congress is greater, legislators show more of an inclination to follow their own criteria. Legislators who have a negative opinion on the stability of their country’s democracy have a style characterized by following their own criteria.

Notes: a. Data from PELA. b. Data from the research project Political Parties and Governability in Latin America (PPAL).

Page 48

Party ideological placementa

Hypothesis

10:56 AM

Party ideology

Variable

10/8/07

continued

Type of Variable

48

Table 2.5

A-2 10/8/07 10:56 AM

Explanatory Variables Introduced into the Analysis: Focus Dimension

Type of Variable Organization of the political party

Political system

Variable Importance of assemblies of activists or delegates in the nomination of candidatesa Level of autonomy of regional politiciansa Role of national leaders in the nomination of candidatesa Effective number of parliamentary parties Political system: federal vs. nonfederal Number of Houses Existence of provincial elections

Political situation

Coalition government

Page 49

Table 2.6

Hypothesis In parties where assemblies of activists or delegates are of great importance in nominating candidates, legislators are more inclined to defend particular interests. In parties where regional politicians are accorded a high level of autonomy, legislators show a greater inclination to defend particular interests. In parties where national leaders are those who make the decisions in the nomination of electoral candidates, legislators are more inclined to defend general interests. In countries where there is a small number of parties in congress, legislators have a higher tendency to defend general interests. Legislators from federal countries will have more of a tendency to defend particular interests than legislators in nonfederal countries. In bicameral countries, legislators show more of a tendency to defend general interests. In countries where there are provincial elections, legislators will tend to defend particular interests more. Legislators who belong to political parties who are in power or form part of a coalition government have a higher tendency to defend general interests than those who are in the opposition.

Note: a. Data from the research project Political Parties and Governability in Latin America (PPAL).

49

A-2

10/8/07

10:56 AM

50

Page 50

POLITICS AND DEMOCRACY

considered dependent variables in order to find possible predictors to explain variations in each of these dimensions. However, as previously pointed out, there is no one Latin American reality; each case has its own characteristics, shaped by its history and institutional design. It is therefore very difficult to make generalizations that will be valid for a great number of this region’s countries. Thus, in order to discover more specific relations, we feel it is necessary to carry out an analysis for each individual country. As Table 2.7 indicates, the style dimension has a relationship to five independent variables that, taken together, explain 16 percent of the inclination to follow personal criteria or to follow instructions. Not all of these five variables have the same level of influence on style. The variables that show the greatest predictive power are opinion on the level of activism of the country’s political parties and perception of the country’s political parties. This confirms the initial hypothesis that legislators who feel that activism is low and that party leadership should exercise less control over reprentatives will be more inclined to have an independent style. The third variable is that of political career; it indicates that legislators with more years of political experience follow their own criteria more than

Table 2.7

Dimensions

Multiple Regression: Style and Focus Regressed on Several Explanatory Variables Predictors

β

Style of representation Constant 4.503 Adjusted R2=0.158 Opinion on the influence F=25.674 that party leaders should Sign=0.000 exert on legislators –0.230 n= 745 Level of party militancy –0.223 Ownership of the seat 0.290 Years of political trajectory –0.014 Level of internal party democracy –0.145 Focus of representation Constant 2.304 Adjusted R2=0.081 Effective number of ANOVA F= 17.172 parliamentary parties -0.260 Sign=0.000 Number of chambers -0.448 n=833 Federalism 0.665 Importance of assemblies of party militants or party delegates in the nomination of candidates 0.489 Role of national leaders in the nomination of candidates 0.453

β

t

Sign

22.702

0.000

–0.172 –0.211 0.130 –0.147

–4.730 –5.818 3.612 –4.083

0.000 0.000 0.000 0.000

0.122

–3.329

0.001

4.770

0.000

-0.175 -0.139 0.190

-4.879 -3.969 5.206

0.000 0.000 0.000

0.177

4.305

0.000

0.140

3.257

0.001

Sources: Data based on PELA 1994–2005 and PPAL 1999–2001.

A-2

10/8/07

10:56 AM

Page 51

THE CONUNDRUM OF REPRESENTATION

51

those whose entry into politics is more recent. The fourth variable is the perception of whether or not the seat belongs to the party or to the legislator: when legislators believe that leaving their party does not mean giving up their seat, they are more inclined to let their own criteria guide their decisions.17 The last significant variable in the model is the level of internal democracy within the political party; it shows that legislators who perceive this level as low have more of a tendency to follow their own criteria. To sum up, it is possible to state that representatives who see their seat as belonging to the party demonstrate a higher tendency to follow instructions. On the other hand, representatives with a more independent style tend to be those with various years of political experience, who hold that political parties should not have much influence on legislators and who see as low both the level of militancy in their country’s parties as well as the degree of internal democracy in their own parties. The results obtained from the model used to explain the variation of the focus dimension have been very poor (see Table 2.7). Although there are five statistically significant variables, as a group these variables are not able to predict the variation of the focus dimension beyond a scant 8 percent. In addition, the coefficients for these variables point in a causal direction opposite to the one hypothesized by the theory. All this seems to indicate that the factors used are not the most relevant ones and that since the explanation of this dimension is rooted in territorial factors, it is necessary to search for causes connected to historical and nation-building processes, rather than to systemic, institutional, or party characteristics. This does not go to say, however, that advances cannot be made in the study of style, the other dimension of the representative role. Up to now, the explanatory factors observed for style indicate general tendencies. We now address the issue of what the dimension of style depends on in each of the Latin American countries included in the analysis. The results of the multiple regression analyses carried out for each country can be seen in Table 2.8. The countries in which the independent variables taken into consideration in the model show a greater predictive power are, in order: Nicaragua (in which the model explains 65 percent of the variability of style), Costa Rica (64 percent), Peru (55 percent), Chile (40 percent), the Dominican Republic (37 percent), Uruguay (36 percent), Mexico (33 percent), and El Salvador (28 percent). Those countries in which the model has less predictive power are Honduras (10 percent), Paraguay (14 percent), Bolivia (14 percent), Ecuador (16 percent), Colombia (19 percent), and Argentina (20 percent). Summarizing these results, it can be pointed out that perceptions of national political parties have great power to influence representative style in Latin American countries (see Table 2.9). Specifically, in six countries (Costa Rica, El Salvador, Mexico, Nicaragua, Paraguay, and the Dominican

A-2

10/8/07

10:56 AM

Table 2.8

Page 52

Results of the Analysis, by Country

Chile

Predictors

Adjusted R2=0.403 ANOVA F=7.758 Sign= 0.000 n=56

Constant Ownership of the seat Reason to be elected legislator Internal democracy Experience (reelection) Experience in other offices

Argentina

Predictors

Adjusted R2=0.205 ANOVA F=6.000 Sign=0.001 n=98

Constant Level of militancy in own party Reason to be elected legislator Internal democracy

Bolivia

Predictors

Adjusted R2=0.142 ANOVA F=5.401 Sign=0.007 n=87

Constant Level of popular participation in parties Level of militancy in political parties

Honduras

Predictors

Adjusted R2=0.104 ANOVA F=8.102 Sign=0.006 n=67

Constant Party ideological placement

Colombia

Predictors

Adjusted R2=0.199 ANOVA F=13.682 Sign=0.001 n=68

Constant Ownership of the seat

Costa Rica

Predictors

Adjusted R2=0.641 ANOVA F=9.050 Sign=0.000 n=35

Constant Years of political trajectory Relationship political parties–society Influence of party leaders Reason to be elected legislator Ideological self-placement Necessary role of parties in democracy

Nicaragua

Predictors

Adjusted R2=0.652 ANOVA F=12.602 Sign=0.000 n=45

Constant Experience (reelection) Influence of party leaders Experience in other offices Level of popular participation in parties Relationship political parties–society

β 3.540 0.407 –0.316 –0.363 0.398 –0.371 β 5.171 –0.243 -0.284 –0.292 β

β

0.273 –0.340 –0.359 0.267 –0.238 β

–0.250 –2.274 –0.258 β

4.734 –0.308 –0.273 β 4.035 –0.190

–0.284 –0.273 β

–0.345

t

Sign

8.001 2.365 –2.961 –3.048 2.383 –2.174

0.000 0.022 0.005 0.004 0.021 0.035

t

Sign

8.975 –2.031 –2.333 –2.104

0.000 0.047 0.023 0.040

t

Sign

10.396

0.000

–2.215 –2.126

0.031 0.038

t

Sign

8.646 –2.846

0.000 0.006

β

β

t

Sign

2.545 1.037

0.464

14.021 3.699

0.000 0.001

β

β

t

Sign

–0.440 2.526 2.705 –3.380 –2.399 3.020 –2.356

0.665 0.020 0.013 0.003 0.026 0.007 0.028

t

Sign

–0.338 0.019 0.625 –0.463 –0.219 0.233 –0.262 β

0.303 0.353 –0.421 –0.287 0.404 –0.304 β

0.601 2.074 –0.668 –1.812

0.483 –0.450 –0.508

0.759 4.283 –4.072 –4.140

0.455 0.000 0.000 0.000

–0.502 0.238

–0.340 0.238

–2.769 2.653

0.010 0.041 (continues)

A-2

10/8/07

10:56 AM

Table 2.8

Page 53

continued

Paraguay

Predictors

Adjusted R2=0.141 ANOVA F=4.871 Sign=0.012 n=60

Constant Influence of party leaders Years of political trajectory

Ecuador

Predictors

Adjusted R2=0.161 ANOVA F=5.911 Sign=0.005 n=70

Constant Relationship political parties–society Ownership of the seat

El Salvador

Predictors

Adjusted R2=0.28 ANOVA F=5.963 Sign=0.008 n=58

Constant Influence of party leaders Ideological self-placement

Dominican Republic Predictors Adjusted R2=0.374 ANOVA F=11.018 Sign=0.00 n=99

Constant Membership of the seat Level of militancy in own party Relationship political parties–society Influence of party leaders

Mexico

Predictors

Adjusted R2=0.336 ANOVA F=6.357 Sign=0.000 n=77

Constant Necessary role of parties in democracy Level of militancy in own party Democratic stability Relationship political parties–society Experience (reelection) Party ideological placement Influence of party leaders

Peru

Predictors

Adjusted R2=0.555 ANOVA F=14.739 Sign=0.003 n=49

Constant Experience (reelection)

Uruguay

Predictors

Adjusted R2=0.364 ANOVA F=12.163 Sign=0.000 n=50

Constant Level of militancy in political parties Years of political trajectory Extreme self-placement

β 4.352 –0.444 –0.027

β

–0.345 –0.286

t

Sign

9.187 –2.533 –2.100

0.000 0.015 0.041

β

β

t

Sign

0.769 0.507 0.831

0.344 0.333

1.627 2.649 2.569

0.110 0.011 0.013

β

β

t

Sign

2.014 –2.241 –2.144

0.055 0.035 0.042

t

Sign

3.518 2.795 –2.888 2.648 –2.415

0.001 0.007 0.005 0.010 0.019

t

Sign

8.974 3.257 –2.678 –3.208 –2.627 2.390 –2.075 –2.588

0.000 0.002 0.009 0.002 0.011 0.020 0.042 0.012

.969 –0.506 –0.677 β 2.526 0.731 –0.365 0.714 –0.345 β 7.049 0.245 –0.238 –0.358 –0.385 0.503 –0.216 –0.324

–0.381 –0.365 β

0.286 –0.282 0.266 –0.242 β

0.319 –0.277 –0.340 –0.283 0.236 –0.262 –0.217

β

β

t

Sign

2.253 1.369

0.722

9.787 3.839

0.000 0.003

β

β

t

Sign

10.789

0.000

–4.026 –2.628 2.717

0.000 0.012 0.010

4.695 –0.547 –0.021 0.795

Source: Data based on PELA 1994–2005 and PPAL 1999–2007.

–0.544 –0.435 0.337

A-2

10/8/07

10:56 AM

Table 2.9

Page 54

Summary of Explanatory Factors of the Style Dimension

Type of Variable

Variable

Countries in Which This Variable Serves as a Predictor

Perceptions of the Influence of the party leadership country’s political parties

Mexico El Salvador Nicaragua Paraguay Costa Rica Dominican Republic The necessity of parties to democracy Costa Rica Mexico Ownership of the seat Chile Colombia Ecuador El Salvador Dominican Republic Level of popular participation in Bolivia political parties Nicaragua Level of militancy in political parties Bolivia Uruguay Relationship between parties and society Costa Rica Ecuador Nicaragua Mexico Dominican Republic Perceptions of one’s Internal democracy Argentina own political party Chile Level of militancy in own political party Argentina Mexico Dominican Republic Political career Experience (reelection) Chile Nicaragua Mexico Peru Reason for having been elected as Argentina a representative Chile Costa Rica Years of political experience Costa Rica Paraguay Uruguay Experience in other offices Chile Nicaragua Political system Opinion on the stability of the Mexico democracy Ideology Ideological self-placement Costa Rica Ideological placement of the party Mexico Extreme ideological self-placement El Salvador Uruguay Source: Based on data in Table 2.8.

Republic), opinion on the influence that party leaders should exercise over legislators is significantly associated with our dependent variable. This confirms in all countries our initial hypothesis: those legislators that consider that the power of party leaders must be low are more likely to follow their

A-2

10/8/07

10:56 AM

Page 55

THE CONUNDRUM OF REPRESENTATION

55

own judgments. Thus, in Costa Rica, Ecuador, Nicaragua, Mexico, and the Dominican Republic, legislators who believe that the relationship between parties and society is increasingly weaker tend to follow their own criteria and not follow instructions. This same attitude is shared by legislators from Chile, Colombia, Ecuador, El Salvador, and the Dominican Republic, who feel that the seat belongs to the representative and not to the party. These results indicate that representative style depends on political parties, not so much in terms of party organization or program, but rather in how parties are perceived by legislators. If legislators consider that parties are necessary to democracy, that there are strong ties between parties and society, that the seat belongs to the party, and that there are high levels of popular participation and militancy, then they are more likely to follow instructions. Perceptions of political parties exert the strongest influence on style in various countries, such as Bolivia, Ecuador, the Dominican Republic, and Mexico.18 In the case of Mexico, the following are also greatly important: opinion on the stability of the country’s democracy, opinion on the need for political parties to maintain this stability, and opinion on the relationship between parties and society. Thus, the legitimacy accorded to the political system by the legislators themselves is crucial in their role as representatives. Opinions on their own political party also condition legislators’ style, although to a lesser degree. This is especially the case in Argentina, whose legislators’ representational style is conditioned by the degree of internal democracy (this variable is also significant in Chile) and by the level of popular participation in the political party itself (this level is important in Mexico and the Dominican Republic as well). Thus, as was hypothesized, Argentine legislators who feel that both of these factors are present to a high degree are positively disposed toward following instructions. The political career of the legislator also has an influence on style, although in fewer countries than does the perception of political parties. Specifically, in Chile, Nicaragua, Mexico, and Peru, if legislators have had previous representative experience, they are more inclined to follow their own criteria in their representational roles. Therefore, being reelected causes a representative to be more independent of his or her party, since s/he has accumulated political capital and can act without always following party guidelines. On the other hand, in Chile and Nicaragua, representatives with previous experience as elected officials show more of a tendency to follow instructions. Although this may seem contradictory, it is in fact not so, since being elected to public office various times indicates the confidence that a politician inspires in his or her party. Similarly, in Costa Rica, Uruguay, and Paraguay, representatives most likely to follow instructions are those with the most years in public office. One last variable related to political career, significant in Argentina, Chile, and Costa Rica, is opinion on why a legisla-

A-2

10/8/07

10:56 AM

56

Page 56

POLITICS AND DEMOCRACY

tor was elected. As initially hypothesized, legislators who believe that their election is due to their own personal merits are the most inclined to have an independent style. The countries in which variables related to legislators’ political careers are the most clearly related to style are Chile, Nicaragua, and, most especially, Peru. In this last country, previous experience as a member of congress conditions 55 percent of the variation of our dependent variable. Last of all, ideology is not very much of a determining factor in the style of Latin American representatives. Only in El Salvador and Uruguay does extreme ideological self-placement seem to have an influence; representatives who locate themselves on the extreme ends of the continuum are more open to following instructions. The ideological placement of parties plays a significant role in Mexico, and the ideological self-placement of legislators plays a significant role in Costa Rica. In both cases, the more to the left, the more favorable legislators are to following their personal judgments.19

■ Conclusion Each historical moment holds up ideal models of representational roles, products of the conception of the period’s political regimes and of the principles of representation put into practice. At the present point in time, when attempts are made to refine the role that legislators should play and the ways in which they should behave, there are two ways of conceiving of representation (consequences of different historical stages) that clearly occupy different poles. The first conceives of representatives as delegates with clear instructions from their districts.20 The second sees representatives as members of a deliberative body, of one nation, charged with defending the common good and general interest—thus free to act without ties to the designs of their districts.21 These two models, mandate and independence, manifest themselves in different combinations in legislators’ ideas about representation. This chapter has explored these conceptions about representation that exist among legislators in the young third-wave Latin American democracies. Through the analytical division of legislators’ representative role into two dimensions, focus and style, we have examined the predominant criteria in each of these countries (and their political parties) and have explained some of the factors that contribute to the formation of these criteria. The first conclusion that can be drawn from this comparative analysis is that, in spite of the existence of a national mandate—established in the constitution, in the majority of the cases—a great number of legislators have a particularist conception of the representative role. Mexico, the Dominican Republic, and Costa Rica stand out as countries in which the focus of representation is the closest to taking all citizens into account when making

A-2

10/8/07

10:56 AM

Page 57

THE CONUNDRUM OF REPRESENTATION

57

political decisions. The legislators of Chile, Bolivia, Colombia, and Uruguay, on the other hand, are the closest to the defense of special interests. In the case of the first three countries named in this second list, this is because of a strongly territorial conception of representation; in the case of Uruguay, it is because of a party-based conception of representation. In relation to this last issue, the general analysis, by country, of the legislators who state that they defend mainly special interests shows a weakly party-centered conception of political representation. In all cases, except those of Costa Rica and Uruguay, when making political decisions, legislators accord greater importance to the interests of the voters of their departments, provinces, or districts than to those of their parties and those parties’ voters. The second conclusion to be drawn is that the style of Latin American legislators is guided primarily by a conception of mandate more than by an independent conception of representation. Although there is a real division between legislators on this issue, the data suggest that representatives tend to guide their actions more by opinions of district voters, political parties, the public in general, or any other social group than by their personal criteria of what is most important for the country. This tendency is most pronounced in the cases of the Dominican Republic, Ecuador, and Uruguay. In this chapter, we categorized explanations of the style dimension (of the representative role) into four groups: perceptions of the importance, functioning, and organization of political parties; political ideology; political careers; and system characteristics. This analysis indicates that legislators who show a marked tendency to follow instructions are those who consider that their seat belongs to their political party and not to them as individuals. However, the profile of those with an independent style is of legislators with various years of political experience who hold that political parties should exercise very little influence over them and feel that both the level of militancy in their country’s parties and the internal democracy of their party organization are low. When these variables are examined by country, it is obvious that their explanatory role greatly varies from country to country. Factors of influence are different from one country to the next, as are their levels of influence on the dependent variables. The most commonly recurring factors were legislators’ opinions on the degree of influence that the party organization should exercise over them; their perceptions of the relationships between parties and society, to whom they felt their seats belonged; and the political careers of the legislators (most importantly, previous experience in the same office). In the case of focus, the predictive variables were fewer. We grouped them into three categories: party organization, political system, and political situation. The information provided by these variables does not account for the variation in the legislators’ opinions regarding the interests they claim to

A-2

10/8/07

10:56 AM

58

Page 58

POLITICS AND DEMOCRACY

represent. It is therefore necessary to identify the factors that help explain what interests legislators are willing to defend. Key among them can be variables connected to the electoral system (at both the national and subnational levels), to national historical developments, and to social and political cleavages. This analysis of the representative role of Latin American legislators is based exclusively on perceptions and opinions these representatives express on a theoretical level. Studies on party discipline in Latin American countries,22 which measure real behavior, seem to contradict many of the results obtained in our study. Thus, we could say that there is an obvious discrepancy between what legislators say and what they end up doing.23 In this sense, it is necessary to continue more in-depth studies of this kind in at least two directions. In one direction, studies should be done on the different kinds of representative logic that come into play in each case, depending on immediate contextual factors. Research should try to determine what situations cause legislators to act following their own criteria and in what situations they defend party or territorial interests. In addition, in another direction, research should be carried out on specific types of legislation that, because of their controversial nature, bring party interests into conflict with regional ones. Legislators’ behaviors in these cases should be analyzed to determine whether or not there are patterns that can be differentiated by political systems, party organizations, or other factors of categorization. The attitudinal dimension of representation—what representatives express on a theoretical level—must be complemented by a comparison with what these representatives do in practice. This would allow a clearer picture to be constructed of how Latin American legislators conceive of and carry out political representation in the region. At the same time, the explanatory factors used in this analysis should be examined more thoroughly in order to expand this study and to give a greater depth to the generalizations that have been obtained from it. Certain systemic variables included in this study have shown very little explanatory value—or none at all, such as the differences between federal and nonfederal, and unicameral and bicameral, political systems. These variables should be disaggregated in order to scrutinize them with an empirical analysis capable of determining if they truly contradict the relevance accorded them by theory. Other variables, such as party-related factors (the most used of all), have been shown to have an inconsistent influence on the dimensions of legislators’ representative role, depending on their country of origin. In this study, these variables are measured by the perceptions of the legislators and by their opinions on their own party organization. Further analyses should include other types of data on the functioning and organization of parties in order to enrich the picture that has been obtained from this general comparison.

A-2

10/8/07

10:56 AM

Page 59

THE CONUNDRUM OF REPRESENTATION Table 2.10

59

Legislative Period Analyzed and Number of Interviews

Country Argentina Bolivia Chile Colombia Costa Rica Dominican Republic Ecuador El Salvador Honduras Mexico Nicaragua Paraguay Peru Uruguay Venezuela Total

Legislative Period

Number of Interviews

1997–2001 1997–2002 1997–2001 1998–2002 1998–2002 1998–2002 1998–2002 2000–2003 1997–2001 2000–2003 1998–2002 1998–2003 2001–2006 2000–2005 2000–2005

112 91 83 77 42 99 103 56 67 116 59 65 69 65 81 1,185

Source: Data based on PELA 1994–2005.

■ Appendix Construction of the Dimensions: Focus Dimension

In order to construct the first dimension, focus, the following questions were used: Question 28: In your opinion, how important is it to get resources for your province/department/state during your parliamentary term: very important, quite important, not very important, not at all important. Question 56: With respect to your job as a Legislator, could you please tell me to which of the following three aspects you accord the most importance during your parliamentary activity: representing the nation, making laws, solving the country’s problems, watching over the actions of the government, defending the interests of your party, drawing up the national budget, representing the interests of your department. Very important, quite important, not very important, or not at all important. Question 58: Of the following options that I am now going to read, please indicate whom you believe you represent during your parliamentary activity: all the voters of your party, all the voters in your district/department/province, the political party you belong to, all the citizens of your country?

A-2

10/8/07

10:56 AM

60

Page 60

POLITICS AND DEMOCRACY

These indicators have been used to construct an index showing whether legislators are more in favor of defending the general interests of the whole population or whether, on the other hand, they are more in favor of defending special interests, whether these interests be party or territorial. The index has a scale of 0 to 6, with 0 indicating positions close to the defense of special interests and values close to 6 indicating positions in favor of defending general interests. In order to construct this index, the following transformations have been effected in the questions of the survey: Question 28 (importance accorded to getting resources for the legislator’s province) was recodified in the following way: the value 0 was assigned to legislators who felt that getting resources for their province was very important; 0.33 to legislators who saw this task as quite important; 0.66 to legislators who saw it as not very important; and those who saw it as not at all important were assigned the value of 1. Thus, the legislators interested in defending their territories would have values close to 0, and those who did not express very much interest in defending their province would have values close to 1. In the recodification, Question 56 (aspects to which legislators assign the most importance during their parliamentary activity) contributes a maximum of 2 points to the index. Legislators who placed a lot of importance on representing the nation were assigned 0.50 point, as were those who stated that they placed a lot of importance on solving the country’s problems. If they saw solving the country’s problems as quite important, they were assigned the value 0.33; if they saw it as not very important, they were given the value 0.17; and if they saw it as not at all important, they were assigned the value 0. On the other hand, the level of importance given to the defense of parties’ interests and the representation of departments’ interests was modified in the following way: those who stated that they saw these aspects as very important were assigned the value 0 (in each question); those who said they saw it as quite important were given the value 0.17; legislators who found each one of these aspects to be not very important were assigned 0.33 in each question; and those who found these aspects not at all important were assigned the value 0.5. Question 58 was assigned more weight than other questions because it most clearly captures legislators’ positions in the focus dimension. When a legislator responded that s/he believed s/he represented all the voters of the party, all the voters of the district/department orprovince, or the political party to which s/he belonged, a value of 0 was assigned. When the answer was all of the country’s citizens, the value was 3. Construction of the Dimensions: Style Dimension

In the case of the second dimension, style, the questions used were the following:

A-2

10/8/07

10:56 AM

Page 61

THE CONUNDRUM OF REPRESENTATION

61

Question 25: The issue of voting along party lines has traditionally given rise to very diverse opinions. With which of the following are you more in agreement? The legislator should always be required to vote along party lines, each legislator should be permitted to vote according to his or her own criteria, legislators should be required to vote along party lines in some subjects but not in others. Question 26: I would like to know how much in agreement you are with the following statement: “A political party should expel a representative who votes against certain political decisions of his or her party.” Are you very much in agreement, pretty much in agreement, not very much in agreement or not at all in agreement? Question 59: When making political decisions, to what degree— very high, high, fairly low or not at all—do you take into account the opinions of the following groups, people or institutions: district voters, party leaders, general public opinion, party members, media, other representatives of your party, interest groups and your party’s voters? Just as was done with the focus dimension, an index with a scale of 0 to 6 was constructed, based on these questions. In this case, the values close to 0 belong to legislators in favor of following instructions (independently of who issues these instructions, which can come from the party, from interest groups, from their own territory, etc.). Values close to 6 belong to legislators with attitudes favorable to following their own individual criteria. The transformations carried out to construct the index with the values from the questions were the following: Question 25 refers to the attitude shown with respect to party discipline. The maximum value that it contributes to our index is 2, assigned to those legislators who are in favor of representatives’ voting according to their own criteria. However, representatives who feel that legislators should follow party discipline in all cases were assigned the value of 0. In the middle, with 1 point, are those legislators who are of the opinion that party discipline should prevail in some subjects but not in others. Question 26, related to legislators’ switching their vote from one party to another, contributes a maximum value of 1, which corresponds to the value assigned legislators who completely disagree with expelling representatives who vote against certain political decisions of their party. Those who are not very much in agreement with expulsion are assigned the value 0.66; those who are pretty much in agreement receive the value 0.33; and those completely in agreement are assigned the value 0.

A-2

10/8/07

10:56 AM

62

Page 62

POLITICS AND DEMOCRACY

The last question used in the index contributes altogether the value of 3 points. The information elicited by this question refers to the institutions and actors that legislators take into account when making decisions. A total of eight actors have been considered (district voters, party leaders, general public opinion, party members, the media, other representatives of the party, interest groups, and party voters). For each actor not taken into account, legislators were assigned the value 0.37; while for actors taken into account to a fairly low degree, they were assigned the value 0.25. Legislators were assigned the value 0.12 for each actor taken into account to a high degree, and given 0 for actors taken into account to a very high degree. Those representatives who receive close to 0 are therefore in favor of following instructions.

■ Notes 1. As Manin points out (1998), representative governments include both democratic and oligarchic components. The functioning of representative governments requires a balanced combination of both types of elements. 2. “The first body of research, on ‘procedural’ representation, begins with citizens’ votes for parties in elections. Party voting is then linked to party representation in the legislature through aggregation of parties’ votes into victories. Democratic representation means that votes for parties should correspond to the seats those parties win in the legislature. Much of this research has focused on the way different election laws shape such representation. The second body of research begins with citizens’ preferences, rather than with their votes. Voter choices under electoral competition link citizens’ preferences to the preferences and behavior of legislative representatives. Parties’ candidates take positions on issues and the electoral commitments shape their policymaking after the election. Democratic representation means that citizens’ issue preferences should correspond to positions or behavior of their representatives” (Powell, 2001:1). 3. The studies of representation were influenced early on by the appearance of the responsible party model. See Schattschneider (1942), APSA (1950), and Ranney (1962). 4. For a summary of each one of these studies, as well as a complete overview of studies on representation, see Powell (2001). 5. There are also some case studies, such as for Mexico: Mascott (2004) and Moreno (2004). 6. These studies, which in some cases exist in the contexts of European countries (Searing, 1985 and 1994; Katz, 1997; Méndez-Lago and Martínez Rodríguez, 2002), do not exist for Latin America. 7. As shown in Table 2.1, in a few countries, the constitution does not explicitly establish the type of representation (those in which the space is blank). Paraguay and Panama are two exceptional cases. In Paraguay, the House of Representatives is assigned the representation of departments; in Panama, the two concepts are mixed: while legislators must act in the interests of the nation, they represent their parties and their districts. 8. Relevant to this approach are Searing’s study of the House of Commons in Great Britain (1985 and 1987) and Studlar and McAllister’s study of the Australian parliament (1996), which prove that legislators’ roles are related to their behaviors.

A-2

10/8/07

10:56 AM

Page 63

THE CONUNDRUM OF REPRESENTATION

63

9. Numerous studies on representation have not sufficiently emphasized this question and have mixed these dimensions, which must be kept separate for analytical purposes. It is probable that in some cases these two dimensions are related; nevertheless, they do capture two different aspects of political representation. 10. The Pearson correlation coefficient between both variables equals –0.07 and is significant at the 0.28 level. 11. The case of Ecuador should be revised and complemented with other studies since the political literature on Ecuador continually emphasizes the fact that the regional variable strongly influences the makeup of the party system and the relationships established between political actors. The argument is that the Ecuadorian party system is fragmented into two regional subsystems (Costa-Sierra) that permeate the politics of the whole country (Freidenberg and Alcántara, 2001; Pachano, 1996). For more extensive analyses of the regional problem in Ecuador, see Maiguashca (1994) and Quintero (1991). However, judging by the information collected in this study, Ecuador would be one of the cases in which legislators conceive of their role in national terms. The explanation for this apparent contradiction is that the information used to measure the opinions and attitudes of the representatives was collected in the field at a moment in which the country was experiencing abnormal political conditions triggered by the fall of President Bucaram and the reform of the constitution and the Code of Ethics. In these processes, the obligation of the representative to act in a national sense and to be held responsible in political terms by society was being debated. 12. Peru has the same number of points as Chile in the style index. It does not appear in Figure 2.2 since we do not have the information necessary to calculate the focus dimension. 13. We have already commented on the special situation in which the data on Ecuador were collected (see note 10 in this chapter). 14. An extreme case is that of Panama, whose constitution (Art. 145) grants to parties the right to revoke the mandate of the principal or substitute legislators the party may have nominated. The following requisites and formalities must be fulfilled: (1) The causes for revocation and the applicable process shall be present in the Party Statutes. (2) The causes shall refer to serious violations of the Statutes and of the ideological, political or programmatic platform of the party and must have been approved through a resolution of the Electoral Tribunal before the date of nomination. (3) The affected person shall have the right, within the party, to speak and defend himself before two authorities. (4) The decision of the Party to revoke the mandate shall be subject to appeal, the contents of which shall be known exclusively to the Electoral Tribunal, and which shall suspend the revocation. (5) Political Parties shall also be able to revoke the mandate of principal and substitute Legislators who have expressly and in writing resigned from their Party. 15. If we exclude from the analysis the answers “it depends on the issue,” “doesn’t know,” and “doesn’t answer,” and count only the answers “always along party lines” and “always according to the needs of the province/department/state,” it can be observed that 15.8 percent are in favor of the first option and 78.7 percent in favor of the second. 16. Question asked: “When making political decisions, to what degree do you

A-2

10/8/07

10:56 AM

64

Page 64

POLITICS AND DEMOCRACY

take into account the opinions of the following groups, people or institutions: very high, high, not very high or not at all: voters of your electoral district, your party’s leader, public opinion in general, party members, the media, other representatives from your party, interest groups, your party’s voters.” 17. Whether or not the seat belongs to the party or to the individual legislators is a determining factor in legislator behavior. As was pointed out in this chapter, most constitutions establish—implicitly or explicitly—that the seat belongs to the legislator. However, in practice most legislators consider that they owe their position to their party (mostly in cases in which they were elected in closed lists controlled by the party organization). Thus, they view the seat as belonging to the party more than to the individual who occupies it. In these cases, legislators almost always vote along party lines. In contrast, those who firmly believe that the seat belongs to them usually cast their votes more independently 18. The analysis reveals Mexico to be the least parsimonious model, with seven variables that are statistically significant. 19. This relationship is along the same lines as the contributions of Moreno (2004:52), who states that, in the case of Mexico, “[I]deology seems to play an important role in explaining the variation of voting patterns: the farther s/he is to the left, the more likely it is that the legislator will vote independently.” 20. This idea is a product of modern parliamentarians in the first stage in the history of representative government, which in many European kingdoms began in the thirteenth and fourteenth centuries, when representatives of different states or communities within a society were invited to give their consent to measures adopted by the king. 21. This idea results from parliaments’ taking over aspects of the king’s sovereignty and of the consolidation of the modern nation-state. 22. Carey (1999, 2001); Ugalde (2002); Jones (2001); Nacif (2002); Ames (2000); Limongi and Figueiredo (1995); Mainwaring and Pérez Liñán (1997); García Montero and Marenghi (2003). 23. In all cases, the raw material for research on party discipline in Latin America has been nominal vote, relatively infrequent in the region. This could be one of the causes for the inconsistencies that appear between opinions and behavior. On these issues, see García Montero and Marenghi (2003).

A-3

10/2/07

3:35 PM

Page 65

3 The Meanings of Democracy and Institutional Trust Araceli Mateos Díaz

■ Democracy: Definition, Components, and Models There are multiple definitions of “democracy,” which have led to many different classifications of democratic regimes. Thus, David Collier and Steven Levitsky (1997) find more than 550 subtypes of democracy—linked to different characteristics—in studies on this particular political regime. The most basic definition mentioned by most of these works includes four components: free elections, universal suffrage, full participation, and civil liberties. Any regime that does not meet all these criteria is considered “nondemocratic.” Along these lines, Scott Mainwaring and Aníbal PérezLiñán (2005:15) define as “democratic” every regime that meets four conditions and maintains them for a number of years: the government is elected in free and fair elections, civil liberties are protected, the electorate is formed by most adults, and no influence in exercised on elected authorities by the military or other nonelected actors. There have also been many disagreements on which indicators should be used to measure democracy and its characteristics. Mainwaring, Daniel Brinks, and Pérez-Liñán (2001) show that the results of studies on the development of democracy will depend on the indicators we use to measure this dependent variable. This chapter focuses on the perceptions that political elites1 have of democracy, and on the indicators we must examine to have a better grasp of what these political elites understand by “democracy.” The fact that most legislators prefer a democratic regime over other alternatives does not necessarily reveal a shared acceptance of the core elements of democracy. The majority of Latin American legislators favor democratic over nondemocratic regimes, but in some countries there is a significant percentage of representatives who agree with the idea that “in contexts of economic crisis and political instability, an authoritarian government can be preferable to

65

A-3

10/2/07

3:35 PM

66

Page 66

POLITICS AND DEMOCRACY

a democratic one” (see Table 3.1). Respondents agreeing with this view range, in the second round of this survey, from 15.7 percent in Nicaragua to 5.7 percent in Peru. The third round of the survey reveals important changes in Bolivia (where respondents favorable to an authoritarian regime rose to 6.3 percent), Ecuador (where favorable respondents reached 12.2 percent), and Venezuela (where they increased to 7 percent). By contrast, the percentage of people favoring nondemocratic regimes significantly decreased in Guatemala, Nicaragua, and Peru. In Chile, 9.1 percent of the respondents shared this positive view of nondemocratic regimes. Contemporary democratic regimes involve electoral processes in which citizens select the holders of political offices. Candidates are presented by or linked to political organizations. Usually, these political organizations are political parties. Political parties aim at representing and channeling social demands; they also legitimate the political system of which they are a part. Due to the importance of elections and political parties in the democratic process, an accurate representation of legislators’ democratic ideas requires the analysis of their views regarding these two crucial components of democracy. To assess their views of these two elements, this study focuses

Table 3.1

Preferences Regarding Democracy and Authoritarianism (row percentagea) Second Round

Argentina Bolivia Chile Colombia Costa Rica Dominican Republic Ecuador El Salvador Guatemala Honduras Mexico Nicaragua Paraguay Peru Uruguay Venezuela

Third Round

Democracy

Authoritarianism

Democracy

Authoritarianism

100.0 93.9 84.1 95.5 95.9 96.1 92.9 96.9 86.1 100.0 96.0 81.4 93.8 94.3 100.0 95.7

0.0 4.1 11.4 4.5 4.1 3.9 6.3 3.1 10.1 0.0 3.2 15.7 1.5 5.7 0.0 4.3

99.0 91.3 90.9 96.8 98.0 97.4 87.8 96.3 93.4 98.0 94.4 93.3 98.2 96.3 100.0 90.0

1.0 6.3 9.1 3.2 2.0 2.6 12.2 3.8 5.8 2.0 4.8 3.3 0.0 2.4 0.0 7.0

Source: Data based on PELA 1994–2005. Notes: The question read: “With which of the following sentences do you agree the most?—Democracy is preferable to any other form of government; In the context of an economic crisis and political instability, an authoritarian government may be preferable. a. Percentages do not always add up to 100 because of DK/DA responses.

A-3

10/2/07

3:35 PM

Page 67

THE MEANINGS OF DEMOCRACY AND INSTITUTIONAL TRUST

67

on the level of parliamentarians’ agreement with the following two assertions: “Elections are the best form to express political preferences,” and “Without political parties, there cannot be a democratic regime.” Overall, Latin American legislators are more in agreement with the crucial role of elections than with the necessity of political parties. Democratic theory includes elections as one of the basic criteria that must be met by democratic regimes. Elections—the mechanism through which citizens can choose their representatives and distribute power among competing political actors—must meet certain criteria to be democratic: they must be competitive, periodic, free, and fair. In fact, doubts about elections as the main channel to represent social demands might be rooted in the lack of one of these criteria. Interestingly, in the second round of surveys, the number of legislators disagreeing that elections are the best way to channel political preferences reaches 30 percent in El Salvador, 17.1 percent in Nicaragua (both countries lack a sustained democratic tradition), and about 10 percent in Bolivia, Colombia, and Mexico. Changes in the third round are particularly remarkable in Bolivia and Ecuador. In both cases, the number of legislators doubting that elections are a key component of democracy rises to 20 and 23.5 percent, respectively. These changes in Bolivia and Ecuador anticipate the institutional crises both countries experienced in the following years (see Table 3.2). These results give rise to two considerations. In the first place, it could be argued that some of the necessary characteristics of democratic elections are often not met. That is, in many cases elections are not totally free, competitive, egalitarian, and inclusive. This could be the reason why some legislators favor other mechanisms of representation. And, in the second place, it is surprising that some legislators express such strong doubts about elections in general, considering that they themselves became political representatives precisely through this procedure. Again, their doubts may be related to the insufficient fulfillment of democratic criteria in the elections of many Latin American countries over the last few decades. The degree to which national elections are fair can be measured by the confidence (or lack thereof) that legislators display in electoral processes. That explains why there is a significant correlation between the idea that elections are the best way to channel political preferences and the degree of trust in the last elections held in each country: .19 and .18 in the second and third rounds, respectively.2 On the other hand, these data may reflect the fact that the crucial importance of elections has not yet been internalized by some Latin American legislators. This failure to embrace certain criteria of a democratic regime might be linked to the characteristics of the transition processes experienced by their countries. In fact, this limited political learning has been used by some authors3 to characterize the problems of democratic consolidation and stabil-

A-3

10/2/07

3:35 PM

68

Page 68

POLITICS AND DEMOCRACY

ity, since the learning of democratic values entails adopting a specific type of frame in which political decisions and behavior must be placed. Political elites play a crucial role not only by virtue of their actions, but also insofar as their opinions and values exert a strong influence on citizens. According to data presented by the PNUD (2004:137–138), in 2002 about 57 percent of the people interviewed in Latin America preferred democracy over any other political regime. However, similar to some legislators, a significant segment of the public also displayed doubts regarding such important democratic institutions as the parliament and political parties. The second of the indicators this study uses to examine specific ideas legislators have of democracy is their view of the role of political parties. Political parties are a representative and participatory component of democracy; some authors point out that they are the institutional vehicle most adequate to maintaining the stability and legitimacy of democratic regimes.4 Parties are one of the main actors in democratic politics in Latin America, and their behavior has strong repercussions on the political system. For this reason, their institutionalization contributes to the stability and quality of democracy.5 Parties aggregate interests and channel citizen participation, thus influencing government agendas. However, the existence of political parties does not guarantee the representative character of governmental decisions, nor that a democratic regime will work well.6 Many analyses focusing on Latin American parties have underscored the deficiencies of these organizations.7 Their authors have linked the crisis of parties to the crisis of representation, not only as far as formal institutions are concerned, but also regarding the relationships between party structures and social demands. Francis Hagopian (2005:321) even holds that democracy is stable in those countries where the institutions of political representation are perceived in a more favorable light, by both citizens and political leaders, and where political parties are responsive and take into account popular demands. By contrast, in those countries in which these channels of representation are eroded, antisystem leaders can gather popular support, and democratic instability may substantially increase. The opinions of legislators reflect the presence of significant doubts in some Latin American countries regarding the necessary role of parties for democracy. In the second round, about 42 percent of Salvadoran legislators, 37.5 percent of Ecuadorian, and 33.3 percent of Nicaraguan are not very much, or not at all, in agreement with the necessary role of parties. Percentages for these countries are similar in the third round (see Table 3.2). These data reflect the fact that legislators identify problems in party life in several Latin American countries. Parties fail to represent the interests of significant segments of society, and are not the only actors in the political arena. Social movements and other forms of political expression try to fill a void in the representative and participatory roles of political parties.

A-3

10/2/07

3:35 PM

Page 69

THE MEANINGS OF DEMOCRACY AND INSTITUTIONAL TRUST Table 3.2

Components of Democracy: Parties and Elections, by Country (row percentagea for each round) “Elections are the best way to express specific political preferences” Second Round

Degree of agreement with: Argentina Bolivia Chile Colombia Costa Rica Dominican Republic Ecuador El Salvador Guatemala Honduras Mexico Nicaragua Paraguay Peru Uruguay Venezuela

69

Third Round

“Without political parties there cannot be democracy” Second Round

Third Round

Not at Not at Not at Very Not at Very all/not Very all/not Very all/not much/ all/not much/ very much/quite very much/ very quite very quite much in in much in quite in much in in much in in agreement agreement agreement agreement agreement agreement agreement agreement 1.6 11.2 4.5 10.2 4.1

98.4 88.8 95.5 89.8 95.9

1.9 20.0 4.5 5.3 9.8

98.1 77.5 94.3 94.7 90.2

11.7 31.6 20.7 23.9 16.3

88.3 68.4 79.3 76.1 83.7

19.0 36.3 20.5 24.2 27.5

81.0 62.5 79.5 75.8 72.5

4.9 8.9 29.7 6.3 1.4 12.7 17.1 1.5 3.4 0.0 2.9

95.1 91.1 70.3 93.7 98.6 87.3 82.9 98.5 96.6 100.0 97.1

1.7 23.5 16.3 10.7 4.9 6.5 10.0 7.1 2.4 2.9 12.0

98.3 75.5 83.8 89.3 95.1 93.5 90.0 92.9 97.6 97.1 88.0

3.9 37.5 42.2 15.2 15.5 42.1 33.3 3.1 31.0 1.4 8.7

96.1 61.6 57.8 84.8 84.5 57.9 65.2 96.9 67.8 98.6 91.3

5.9 24.7 36.3 12.4 9.8 35.5 21.7 5.5 20.5 14.7 31.0

94.1 74.2 63.8 87.6 90.2 64.5 78.3 94.5 79.5 85.3 68.0

Source: Data based on the PELA 1994–2005. Note: a. The remaining row percentage corresponds to DK/DA responses.

This crisis of interest representation by parties and of party identification among citizens becomes apparent when legislators are questioned about the relationship between parties and society (see Table 3.3), the most common perception being that a widening gap exists between society and political parties. In fact, those who unquestioningly identify themselves with political parties are in the minority in most countries, the exceptions being Paraguay and the Dominican Republic, followed by Honduras. Since party identification is one of the classic explanatory factors for political and electoral participation, the lack of such identification may partly explain the decrease in electoral participation of citizens, who are seeking alternative channels for participation. As stated previously, this questioning of political parties as an indispensable democratic element is not present across the board in Latin America. Nevertheless, one way of assessing the crisis of parties and the factors con-

A-3

10/2/07

3:35 PM

70

Page 70

POLITICS AND DEMOCRACY

Table 3.3

Relationship Between Political Parties and Society (row percentage per round) Second Round

There is a widening gap Most between citizens society identify and with the political parties parties Argentina Bolivia Chile Colombia Costa Rica Dominican Republic Ecuador El Salvador Guatemala Honduras Mexico Nicaragua Paraguay Peru Uruguay Venezuela

Third Round

Not very many people truly identify with the parties anymore

DK/DA

Not very There is a many widening people gap truly Most between identify citizens society with identify and political with the political parties parties parties anymore

DK/DA

4.7 12.2 0.0 4.5 8.2

64.8 62.2 64.0 64.8 75.5

30.5 25.5 34.8 30.7 14.3

0.0 0.0 1.1 0.0 2.0

1.9 5.0 4.5 3.2 17.6

61.5 70.0 52.3 58.9 56.9

36.5 22.5 43.2 37.9 25.5

0.0 2.5 0.0 0.0 0.0

61.2 10.7 15.6 7.6 23.9 6.3 18.6 66.2 3.4 13.7 1.4

35.9 53.6 59.4 35.4 59.2 71.4 61.4 32.3 34.5 78.1 72.5

1.9 35.7 25.0 54.4 16.9 16.7 18.6 1.5 59.8 8.2 26.1

1.0 0.0 0.0 2.5 0.0 5.6 1.4 0.0 2.2 0.0 0.0

29.7 12.2 20.0 19.0 51.0 8.1 76.7 58.9 7.2 14.7 11.1

62.7 53.1 56.3 39.7 39.2 64.5 21.7 35.7 54.2 77.9 53.5

5.1 34.7 20.0 39.7 9.8 25.0 1.7 5.4 36.1 7.4 33.3

2.5 0.0 3.8 1.7 0.0 2.4 0.0 0.0 2.4 0.0 2.0

Source: Data based on the PELA 1994–2005.

ditioning low citizen identification with parties is to examine the relationship between the percentage of legislators who feel that parties are crucial for democracy and the level of confidence that legislators have in their parties. Both in the second round as well as in the third, the correlation coefficient is statistically significant (.204 and .165, respectively).8 Thus, an association exists between the level of confidence in political parties and the percentage of legislators who feel that these parties are crucial for the existence and proper functioning of democracy. Based on the data contained in Table 3.1, among the Latin American legislators interviewed, there is a clear preference, in ideal terms, for a democratic regime over an authoritarian one. Nevertheless, this indicator is not sufficient to argue that this preference implies acceptance of all key components of the democratic political system. Some countries depart from this pattern, a fact that can be seen in the attitudes expressed toward political parties and elections. The following countries can be placed squarely in this

A-3 10/2/07

Table 3.4

Typology of Democracies (row percentagea) Direct Party Democracy

Direct Electoral Democracy

Democracy

No Disagree

Yes Disagree

Yes Agree

Yes Disagree

Yes Agree

Disagree

Disagree

Disagree

Agree

Agree

Second Round

Third Round

Second Round

Third Round

Third Round

Second Round

Third Round

Second Round

Third Round

Total for Latin America 0.6 0.6 3.6 3.0 4.7 Countries in which the percentage for the category of democracy is high and relatively constant

4.7

16.8

17.6

74.2

74.1

Paraguay Dominican Republic 1.0 Uruguay Argentina Honduras Guatemala 1.5 Chile Colombia 2.3 5.8 Countries in which the percentage for the category of democracy is increasing

5.6 1.8

3.3 2.1 1.4 11.8 15.5 13.2 20.8 15.1

3.7 5.3 11.8 16.3 9.0 10.6 15.0 20.7

95.1 93.8 91.8 86.6 83.1 82.4 77.8 74.4

90.7 93.0 85.3 81.7 86.0 78.8 80.0 73.9

2.9 1.9

2.0 1.0

2.7 3.5

4.3 1.5 5.4 4.8

1.6 3.1 6.8 1.6 1.4

2.7 2.5 3.3

1.4 2.3

5.0 8.0 2.5 2.2

1.3 3.6 2.5 7.7

15.5 7.0 3.3 11.3

1.3 5.4 3.4 7.7

16.7 28.1 32.8 22.6

20.3 19.6 31.4 28.2

65.5 54.4 54.1 46.8

77.2 71.4 61.9 55.1

3.9 4.5 12.2 1.2

2.9 9.1 5.4 3.8

3.9 5.6 5.4 19.8

12.8 6.1 23.9 31.7

21.6 23.6 23.0 20.9

83.0 83.3 65.2 58.7

68.6 66.3 56.8 54.7

Peru 2.4 Nicaragua 3.5 7.0 Mexico 0.8 0.8 9.0 El Salvador 1.3 19.4 Countries in which the percentage for the category of democracy is decreasing Costa Rica Venezuela Bolivia Ecuador

Second Round

71

Source: Data based on the PELA 1994–2005. Note: a. Countries placed in order of percentages in the category “democracy” from highest to lowest.

Page 71

Nonrepresentative Democracy

3:35 PM

Preference for democracy “Without political parties, democracy is not possible” “Elections are the best mechanism for expressing political preferences”

Authoritarianism

A-3

10/2/07

3:35 PM

72

Page 72

POLITICS AND DEMOCRACY

group: Bolivia, El Salvador, and Venezuela, followed by Guatemala and Mexico. Legislators’ answers to the following three indicators can be used to create a typology of democracy: the preference for a democratic regime over an authoritarian one, political parties as an essential element of democracy, and elections as the best form to express certain political preferences. Within the category of legislators who express a favorable attitude toward democracy, this typology can be used to discover the number of legislators who disagree with two of the elements—political parties and elections— that make a democracy. There are five categories that make up this typology: four of them correspond to the democratic regime, and the fifth, to the authoritarian one (absence of a democratic regime).9 (See Table 3.4.) The percentage of legislators who adopt an authoritarian stance—that is to say, who reflect authoritarianism in all three indicators—is very low. In the third round, some representatives in Bolivia, Ecuador, and Costa Rica take authoritarian positions. Since 2002, the first two of these countries have experienced political crises and the third, a generalized mistrust of the political class. The second category of this typology (which implies a general support of democracy, but without conceding importance to political parties or to elections within the political system) is also in a minority. It is, however, an opinion more widely expressed among the representatives of certain countries, especially El Salvador in the second round (19.4 percent) and Bolivia in the third (12.2 percent). The category “direct party democracy” implies both the questioning of the role of elections as an instrument to express political preferences and, at the same time, the acceptance of parties as channels for representation. This category does not have a clearly differentiated pattern by country, although it is clear that—especially in Peru in the second round and Ecuador in the third10—legislators have doubts about the electoral process. In the case of Ecuador, the coming into power of Lucio Gutiérrez may be at the bottom of this opinion, as well as the special characteristics of the 1995 elections in Peru. The most striking information in Table 3.4 appears under the categories “direct electoral democracy” and “democracy.” The first of these categories reflects the preference for a democratic regime and elections, but questions the role of political parties. This view may be due to the crises of political parties in, above all, Peru, Nicaragua, Mexico, El Salvador, Costa Rica, Venezuela, Bolivia, and Ecuador; these crises could have affected the conception of democracy as well as the transmission of democratic political orientations, which are quite different from those of other countries. Last of all, the category “democrats,” in accordance with the positions expressed in ideal terms by the parliamentary elites, is fundamentally present in Paraguay and the Dominican Republic. In these countries, the percentage of

A-3

10/2/07

3:35 PM

Page 73

THE MEANINGS OF DEMOCRACY AND INSTITUTIONAL TRUST

73

those who have a view of democracy that does not question these two components is constant over time. The grouping of the countries contained in Table 3.4 is based on the level of democratic attitudes among parliamentary elites. The countries of the first group have a high percentage of legislators showing democratic attitudes and positions throughout the two legislative periods analyzed. Legislators from the second group of countries show positive changes in attitude from one legislative period to the next, resulting in a considerable increase in the percentage of “democrats.” The countries of these two groups constitute examples of cases in which patterns of polyarchic actions have become institutionalized; legislators’ evaluations of this institutionalization are reflected clearly in the positions they manifest. The third group is composed of those countries where democratic attitudes decreased considerably, mainly due to crises in the party systems and in the political systems of these countries. In Costa Rica, these attitudes can be said to reflect the beginning of the reshaping of the political system and the increase in voter abstention. In Venezuela, legislators’ attitudes point to a serious questioning of the political system, which later led to the system’s being completely reformed. The decrease of democratic attitudes in Bolivia foreshadows the government crisis that came on the heels of Sánchez de Lozada’s resignation. Of these three countries, Ecuador has the lowest percentage of democratic orientations in the three indicators analyzed. The opinions of this country’s representatives during the past two legislative periods clearly reflect the recurring crises in the Ecuadorian political system, which resulted in the failure of the last two presidents to finish out their terms.

■ Advantages and Consolidation of Democracy The paradigms that can be used when analyzing the democratic political system can be divided into two groups: those whose purpose is to determine the essential dimensions that must be present in order to consider a political system democratic, and those that study elements that either help or hinder the creation and consolidation of democracy. These paradigms seek to establish what elements contribute to democratic stability or instability and allow comparisons between countries to be established;11 they can be differentiated on the basis of the conditions or requisites they emphasize. 12 Although most analyses try to draw conclusions that can be extended to a group of countries, the conditions or prerequisites that favor the existence of democracy tend to be specific to the case under analysis or centered in a specific stage in the process of democratic maturity. Debates on and analyses of democratization in Latin America and the characteristics of the regions’ democracies proliferated in the second half of

A-3

10/2/07

3:35 PM

74

Page 74

POLITICS AND DEMOCRACY

the 1990s.13 Elements crucial to this process—including, among others, electoral, economic, international, or cultural elements—were reviewed to discover in what countries the most or fewest obstacles to democratization were present. In the past twenty-five years, the transformations and processes of institutional, political, and economic reform in Latin America have been oriented toward installing representative democracies without neglecting certain participatory components. Democracy is associated with both goals and procedures helpful to achieve those goals. Responses to questions on democracy are thus conditioned by how respondents view not only democracy, but also its main components; the idea of democracy is thus linked to the system of beliefs and orientations that each individual holds. In this regard, it becomes particularly important to establish what the advantages are that, according to political elites, democratic systems have. It is also important to establish what goals these elites believe democracy must help to achieve. Table 3.5 allows us to identify the meanings of democracy for Latin American legislators by considering which advantages are most often mentioned by respondents. In this table, it can be observed that there are no important differences from one round to the other. The main advantage is, for 35 percent of respondents, that democracy protects individual rights and liberties. The second most important advantage lies in the possibility that this political system offers for electing governments. The third most highly valued advantage of democracy, according to about 10 percent of respondents, is that it favors economic growth. The least important advantages are

Table 3.5

Main Advantage of Democracy (column percentage)

Economic growth The protection of individual rights and liberties The possibility to elect governments Greater equality of opportunities Respect for human rights and for the rights of minorities The possibility to participate in decisions A better distribution of income Conflict resolution by peaceful means The possibility to participate in politics through political parties Other responses DK/DA Total

Second Round

Third Round

9.9 34.5 15.4 8.6 8.7 9.7 4.6 2.9 3.3 1.8 0.5 100

9.8 36.4 16.2 8.6 8.5 7.8 4.2 3.1 4.7 0.0 0.6 100

Source: Data based on the PELA 1994–2005. Note: The question read: “I am now going to show you a list of the possible advantages of a democratic regime; I would like to know which, in your opinion, is the main advantage.”

A-3

10/2/07

3:35 PM

Page 75

THE MEANINGS OF DEMOCRACY AND INSTITUTIONAL TRUST

75

peaceful conflict resolution (2.9 and 3 percent in the second and third rounds, respectively) and political participation through parties (3.3 and 4.7 percent, respectively). Therefore, for Latin American legislators, the key components of democracy are respect for individual rights and liberties and political participation.14 There are, however, significant differences among respondents when one considers national membership. Economic growth is considered an important advantage (in the second round) by 15 percent of Guatemalan, Honduran, Paraguayan, and Nicaraguan representatives; the possibility to elect governments is an important advantage of democracy for more than 20 percent of Mexican and Honduran legislators. Respect for human and minority rights is an important advantage of democracy for more than 10 percent of the representatives of Bolivia, Argentina, Ecuador, and Costa Rica. The ideal vision of democracy can also be distinguished along ideological lines. (See Table 3.6.) A larger percentage of those who place themselves on the right find economic growth an important advantage of democracy (15.5 and 16.4 percent in the second and third rounds, respectively), in contrast to 6.8 and 11.5 percent of respondents on the left in these two rounds. It is important to point out that there is more dispersion among the issues emphasized by leftist legislators than among the other legislators. Democratic consolidation entails meeting certain conditions; these conditions vary, depending on the studies to which one refers. Juan Linz and Alfred Stepan (1996:33) emphasize the importance of a combination of dimensions linked to behavior, attitudes, and constitutional frameworks. For them, democratic consolidation requires the routinization and internalization of democracy in social, institutional, and psychological life. These authors also indicate the importance of five interconnected conditions: a free and active civil society, a relatively autonomous political society, political actors subject to the law, a state bureaucracy acting under the authority of the democratic government, and an institutionalized economic society (see Linz and Stepan 1996). Schmitter (1995:5) points out that democratic consolidation is the process by which transformations and solutions adopted in an accidental manner become regulated, recognized, and accepted by those who take part in democratic government. It is also possible to analyze the characteristics that, according to legislators, are critical to the process of democratic consolidation. Table 3.7 shows the responses of parliamentarians to the question on the most important conditions of democratic consolidation. Legislators mention four elements: consensus on the constitutional framework and basic institutions (thus sharing Linz and Stepan’s view on the importance of constitutional agreements), free and fair electoral processes, decentralization and regional democratization, and the presence of democratic values among citizens.

A-3

Economic growth The protection of individual rights and liberties The possibility to elect governments Greater equality of opportunities The respect for human rights and the rights of minorities The possibility to participate in decisions A better income distribution Conflict resolution by peaceful means The possibility to participate in politics through parties Other responses DK/DA Total Source: Data based on the PELA 1994–2005.

Center-Left (3–4)

Center (5–6)

Center-Right (7–8)

Page 76

Left (1–2)

3:35 PM

Main Advantage of Democracy, by Ideology (column percentage)

10/2/07

76

Table 3.6

Right (9–10)

Second Round

Third Round

Second Round

Third Round

Second Round

Third Round

Second Round

Third Round

Second Round

Third Round

6.8

11.5

7.4

8.1

11.9

8.8

9.1

11.9

15.5

16.4

23.9 17.9 5.1

25.9 10.3 12.1

37.1 14.3 10.7

36.9 13.4 7.3

36.3 14.8 7.9

38.2 18.0 7.1

32.0 18.3 7.3

36.5 19.8 11.9

29.3 15.5 12.1

41.1 13.7 16.7

10.3 12.8 12.0 1.7

10.9 10.3 10.3 2.3

9.6 9.9 3.3 2.5

12.3 8.7 5.0 3.1

8.0 9.1 3.7 3.3

7.1 7.9 3.8 3.0

8.2 11.0 5.0 3.2

5.6 5.6 1.2 4.0

8.6 1.7 6.9 1.7

5.5 2.7 1.4 5.5

7.7 1.9 1.7 100

4.0 0.0 2.3 100

2.5 2.3 0.8 100

5.0 0.0 0.3 100

2.3 1.4 0.5 100

5.3 0.0 0.6 100

4.6 1.7 0.0 100

3.6 0.0 0.0 100

6.9 1.8 0.0 100

4.1 0.0 0.0 100

A-3

10/2/07

3:35 PM

Page 77

THE MEANINGS OF DEMOCRACY AND INSTITUTIONAL TRUST Table 3.7

77

Key Characteristics of Democratic Consolidation (column percentage)

Civil control of the military Consensus on the constitution and the main institutions Decentralization and regional democratization Moderation of extreme left and right parties Free and fair elections An independent constitutional court that supervises the constitutional character of legislation Economic agreements between the government, the unions, and employers Democratic values among citizens Other responses DK/DA Total

Second Round

Third Round

7.2 31.6 15.9 2.1 21.5

5.6 29.5 17.5 2.8 19.9

5.3

8.4

4.4 10.9 0.8 0.2 100

4.4 11.5 0.0 0.6 100

Source: Data based on the PELA 1994–2005. Note: The question read: “Of the following characteristics, which, in your opinion, is the most relevant for the consolidation of democracy in ____ [legislator’s country]?”

Some legislators also refer to the presence of an independent, constitutional court. By contrast, ideological moderation and economic agreements between the government, unions, and employers are not considered important for democratic consolidation. There are also important national differences in the responses to this question. Civil control of the military is considered a very important element in Guatemala (19 percent), Nicaragua (18.6 percent), Honduras (18.3 percent), Chile (17 percent), and, to a lesser degree, in Paraguay (13.8 percent) and El Salvador (10.8 percent) in the second round. However, with the exception of Chile (where 18.2 percent of the respondents continue to consider this element a key component of democratic consolidation), the importance assigned to this factor considerably declined once the first period of democratization was left behind. The precise moment in which each country started the democratization process and the circumstances that provoked democratic developments may have exerted some influence on opinions regarding the main problems for democratic consolidation. Tables 3.8 and 3.9 reveal which factors can, in the legislators’ views, pose a threat to democratic consolidation. Those most mentioned are poverty and discrimination, economic crisis, crime, a poorly functioning judicial branch, and terrorism and political violence. Relationships between the government and the military, or between the executive and the legislative branches, are not perceived as serious problems for democratic consolidation. Some of these threats are related to each other, or are seen by Latin

A-3 10/2/07

Elements That Can Be Considered a Serious Threat to Democratic Consolidation, by Country (percentage)

Argentina Bolivia

Relationship between the armed forces and the government Economic crisis Terrorism and political violence Poorly functioning judicial branch Drug trafficking External debt Crime Labor conflicts Poverty and discrimination Lack of interest in politics on the part of the citizenry Relationships between the executive and legislative branches

2

3

2

3

2

3

Colombia 2

3

Costa Rica 2

3

El Dominican Ecuador Salvador Guatemala Honduras Mexico Nicaragua Paraguay Republic 2

3

2

3

2

3

2

3

2

3

2

3

2

3

2

3

7.8 6.7 21.4 51.3 16.8 12.5 28.4 9.5 n/a n/a 33.0 37.8 3.1 8.8 41.8 24.0 4.2 4.9 25.4 26.6 27.1 15.0 29.2 3.6 40.8 10.2 50.0 76.2 71.4 96.3 25.8 51.1 81.8 78.9 71.4 74.5 84.8 91.8 79.7 77.5 87.3 82.6 87.3 90.2 88.9 79.8 81.4 98.3 96.9 98.2 92.2 94.1 16.4 26.7 38.8 60.0 28.1 29.5 94.3 95.8 16.3 45.1 33.0 48.0 43.7 67.5 68.3 57.0 25.3 50.0 61.1 56.5 35.7 48.3 16.9 39.3 27.2 39.8 64.8 35.2 34.4 58.6 20.3 73.4

72.4 42.9 66.7 78.1 27.6 83.8

71.4 81.6 53.1 55.1 46.9 89.8

73.8 73.8 62.5 81.3 61.3 91.3

38.2 50.6 1.1 44.9 14.6 61.8

35.2 45.5 12.5 54.5 27.3 68.2

75.0 79.5 47.7 69.3 37.5 92.0

72.6 85.3 72.6 73.7 32.6 90.5

61.2 77.6 42.9 67.3 22.4 87.8

74.5 74.5 70.6 82.4 52.9 84.3

67.0 33.0 68.8 77.7 43.8 90.2

75.5 48.0 81.6 79.6 38.8 91.8

73.4 73.4 39.1 89.1 56.3 84.4

73.8 73.8 53.8 90.0 53.8 88.8

86.1 83.5 53.2 81.0 34.2 83.5

86.0 85.1 43.8 90.9 43.8 90.1

78.9 74.6 84.5 97.2 29.6 91.5

76.5 79.4 73.5 88.2 49.0 91.2

69.8 91.3 65.9 88.1 36.5 94.4

78.2 85.5 54.8 78.2 41.1 89.5

82.9 74.3 62.9 85.7 31.4 88.6

73.3 56.7 81.7 55.0 43.3 98.3

70.8 63.1 44.6 92.3 36.9 87.7

96.4 82.1 58.9 91.1 55.4 92.9

58.3 89.3 79.6 66.0 55.3 95.1

72.9 83.9 86.4 82.2 25.4 79.7

50.8 62.9 43.9 36.3 47.2 51.1 43.2 62.1 73.5 76.5 44.6 60.2 71.9 72.5 53.2 57.9 40.8 53.9 56.3 55.6 57.1 31.7 60.0 67.9 31.1 66.9 16.4 26.7 33.7 57.5 15.7 14.8 37.5 40.0 22.4 25.5 42.9 43.9 28.1 50.0 29.1 27.3 16.9 15.7 46.8 56.5 48.6 78.3 50.8 35.7 58.3 37.3

Source: Data based on the PELA 1994–2005. Note: The question read: “At the present time, there are a number of issues that can put at risk or present a threat to the consolidation of democracy in _____ [legislator’s country]. I am now going to read a list of these issues. Please indicate to what degree you believe each of the following constitutes a threat at the present time to democracy in ______ [legislator’s country]: very much so, quite a lot, not very much or not at all.” n/a = not applicable in this country.

Page 78

Round

Chile

3:35 PM

Table 3.8

A-3

10/2/07

3:35 PM

Page 79

THE MEANINGS OF DEMOCRACY AND INSTITUTIONAL TRUST Table 3.9

79

Threats to Democratic Consolidation (row percentage for each round) Second Rounda

Relationships between the military and the government Economic crisis Terrorism and political violence Poorly functioning judicial branch Drug trafficking External debt Crime Labor conflicts Poverty and discrimination Lack of interest in politics on the part of the citizenry Relationships between the legislative and the executive branches

Third Round

Not Very Much/Not at All

Very Much So/Quite a Lot

Not Very Much/Not at All

Very Much So/Quite a Lot

60.6 19.8 50.0 26.1 25.9 38.9 21.7 53.0 11.2

18.8 63.3 33.0 57.0 57.0 43.9 61.2 30.0 71.7

76.2 16.9 50.6 27.3 32.7 40.7 23.8 58.9 11.5

19.9 82.8 48.9 72.0 66.8 58.5 75.8 40.3 87.1

41.0

41.8

43.3

55.7

53.1

29.1

60.5

38.5

Source: Data based on the PELA 1994–2005. Notes: The question read: “At the present time, there are a number of issues that can put at risk or present a threat to the consolidation of democracy in _____ [legislator’s country]. I am now going to read a list of these issues. Please indicate to what degree you believe each of the following constitutes a threat at the present time to democracy in ______ [legislator’s country]: very much so, quite a lot, not very much, or not at all.” a. The cases of Uruguay, Peru, and Venezuela were not included in the total number of cases because in these countries the scale consisted of five instead of four categories.

American legislators as problems of a similar type. In order to identify the logical structure that underlies these perceptions of threats, a factor analysis was carried out for each of these rounds (see Table 3.10). The number of threat groups extracted by this analysis in the second round is two, and the variation accounted for is 45.8 percent. In the third round, three components are extracted, and the percentage of variation accounted for equals 54.4. In the second round, the first component is identified as economic threats; it is constituted by such problems as poverty and discrimination, economic crisis, external debt, public insecurity, and drug trafficking. The second component relates to political-institutional risks, and includes such threats as terrorism and political violence, the relationship of the government with the military, the relationship between the executive and the legislative, a poorly functioning judicial branch, labor conflicts, and lack of citizen interest in politics. In the third round, labor conflicts and lack of public interest in politics are seen as important problems, together with socioeconomic risks. As in the second round, these two variables present the lowest correlation in their component. Factors 2 and 3 reveal the difference between political risks—

A-3

10/2/07

3:35 PM

80

Page 80

POLITICS AND DEMOCRACY

Table 3.10

Factor Analysis: The Threats to Democratic Consolidationa Second Round Components 1

Poverty and discrimination Economic crisis External debt Crime Drug trafficking Terrorism and political violence Relationships between the military and the government Relationships between the executive and the legislative branches Poorly functioning judicial branch Labor conflicts Lack of interest in politics on the part of the citizenry

Third Round Components

2

.769 .742 .738 .677 .465

1

2

3

.776 .662 .694 .674 .764 .821

.713 .622

.826

.614 .507 .475

.692 .428

.429

.448

.550

Source: Data based on the PELA 1994–2005. Note: a. Based on the Kaiser-Meyer-Olkin and Bartlett test statistics. The extraction method used is the principal component analysis; the rotation method used is the varimax with Kaiser normalization:

Kaiser-Meyer-Olkin measure of sample adequacy Bartlett test of sphericity

Rotation converged in

Chi-square Df Sign.

Second Round

Third Round

.871 2650.512 55 .000 3 iterations

.852 3257.060 55 .000 6 iterations

connected to the relationships between the executive, the legislative, and the military—and risks related to political management and control, which include terrorism and political violence, a poorly functioning judicial branch, and drug trafficking. In spite of the fact that economic growth is not one of the most mentioned advantages of democracy, factor analysis reveals that lack of economic growth can give rise to risks for democratic consolidation. Manuel Alcántara (1997:22) interprets the relationship between democracy and economic risks in light of the tension between legitimacy and efficacy. Economic growth has been used by many authors as a predictor of democracy.15 However, other works insist on the exceptional character of Latin America, where increases in economic development can lead to decreases in the probability of democracy.16

A-3

10/2/07

3:35 PM

Page 81

THE MEANINGS OF DEMOCRACY AND INSTITUTIONAL TRUST

81

■ Democratic Stability and Institutional Trust The analysis in this section turns to perceptions of the degree of stability of democracy in Latin American nations at the time these surveys were conducted. This analysis uses several indicators to construct an aggregated index of institutional trust. The main hypothesis of this analysis is that the perception of democratic stability is higher in those cases in which there are higher levels of institutional trust (and institutional legitimacy). Current research on attitudes toward democracy and institutional trust center around political disaffection.17 Mariano Torcal (2001:1167) defines political disaffection as “the subjective feeling of inefficacy, cynicism and lack of trust in the political process, politicians, and democratic institutions, which in turn generates distance and alienation regarding these political elements, lack of interest in politics, and low levels of participation in the main institutions of political representation without putting into question the political regime.” Institutional disaffection, or the lack of trust in the democratic political institutions, affects democratic functioning. Most of the data on political disaffection stem from public opinion surveys. By contrast, this chapter analyzes the attitudes of Latin American political elites, and focuses on the legislators’ trust in political institutions and on their assessment of the stability of democracy. The importance of this analysis derives from the role of political elites as key actors in the construction and definition of the political culture and attitudes of citizens, and also in the creation and social transmission of symbols, discourses, languages, and myths.18 Institutional trust is measured as the mean level of trust in the following institutions: the judicial branch, political parties, the military, the legislative branch, the president of the republic, public servants, and the police. Figure 3.1 shows the evolution of the values of institutional trust among legislators in each of the Latin American countries. In the second round, the highest levels of institutional trust are found in the Dominican Republic, Costa Rica, Chile, and Colombia; the lowest levels are those of Nicaragua, Bolivia, and Guatemala. Changes in institutional trust are quite remarkable in the third round, particularly in Bolivia and Ecuador. Data from these countries reveal—previous to the political events that would unfold in both cases19—doubts around several key institutions in the democratic process. The same can be said in the case of Venezuela, where the index of institutional trust is similarly low. However, in this case we lack data to conduct a diachronic comparative analysis. The relationships between the level of institutional trust and the perception of democratic stability are shown in Figure 3.2. Those countries in which legislators perceive that democracy is very stable, or quite stable, show a high level of institutional trust. These are the cases of Chile, Uruguay, Costa Rica, the Dominican Republic, and Honduras. These coun-

10/2/07

3:35 PM

82

Page 82

POLITICS AND DEMOCRACY Figure 3.1 Institutional Trust, by Country

Dominican Republic

A-3

Source: Created by the authors based on PELA 1994–2005. Note: There are no data for this question in the second round in Peru, Uruguay, and Venezuela.

tries display high levels for both variables in the two rounds. In the opposite situation are Nicaragua and Guatemala, where legislators perceive, in both rounds, some democratic instability and display lack of trust in the main political institutions. Those countries that occupy an intermediate position are Mexico, Colombia, El Salvador, and Paraguay; those that show a decreasing level of confidence in institutions and of perceptions of stability are Bolivia and Ecuador. In Argentina, both variables display constant values in both rounds, with a high perception of stability and middle levels of institutional trust.

■ Conclusion Validity and reliability problems have given rise to important debates on the indicators that should be used to measure support for and preferences regarding political regimes. Expressing preference for a democratic over an authoritarian regime is not a sufficient indicator to assess whether parliamentary elites agree with the core components of democracy. In order to assess the meanings of democracy more completely and specifically, this chapter analyzed other indicators: legislators’ evaluations of the role of

A-3

10/2/07

3:35 PM

Page 83

THE MEANINGS OF DEMOCRACY AND INSTITUTIONAL TRUST

83

Figure 3.2 Democratic Stability and Index of Institutional Trust

◆ 2nd Round ■ 3rd Round

Dominican Republic

Dominican Republic

Percentage Democracy Very/Quite Stable

Source: Created by the authors based on PELA 1994–2005.

political parties and electoral processes, and the characteristics and advantages of democracy over other types of political regimes. Legislators’ support for the democratic regime is not in doubt in the Latin American countries analyzed in this chapter. The legitimacy and advantages of democracy, when compared with authoritarian regimes, prevail in all rounds, even if a minority holds that in some situations an authoritarian regime can be preferable to a democratic one. However, this support for democracy cannot be extended to certain components of the democratic regime, such as political parties. A significant number of legislators consider that democracy could function in their countries without political parties. This perception is related to low levels of confidence in political parties among legislators, as well as to the belief that parties are losing importance as channels for the participation and representation of citizens. The fact that legislators express serious doubts about political parties triggers questions on what kind of influence parliamentarians and politicians can exert on public levels of political disaffection.20 The data presented in this chapter show that the cases in which legislators showed doubts and lack of trust in political parties were those that later experienced party system crises and severe destabilization of their political

A-3

10/2/07

3:35 PM

84

Page 84

POLITICS AND DEMOCRACY

systems. These are the cases of Ecuador and Bolivia, and, to a lesser degree, of Venezuela and Costa Rica. The first two cases display increasing levels of mistrust of political parties and a strong decline of confidence in other democratic political institutions. Perceptions of democratic stability transmit information on levels of democratic consolidation. However, democratic consolidation also entails the internalization, routinization, and institutionalization of democratic political structures. This chapter has shown how the internalization of democratic political structures and confidence in political institutions are key elements linked to democratic stability in Latin America. Legislators’ acceptance of the crucial role of political parties and elections in democracy, in combination with trust in political institutions, offers some guarantees regarding democratic stability. Lack of institutional and procedural trust by political elites can lead to democratic instability, and transmit similar opinions and perceptions to citizens. This is why the political orientations of the political elite are considered a necessary element to facilitate not only democratic transition, but also democratic consolidation and stability. Consensus on the components of democracy and “the rules of the game” is critical to overcoming difficult social and economic situations. In spite of the fact that countries are going through different democratic consolidation processes, there are shared views on some requisites of democratic consolidation: consensus among political forces on the constitutional framework and basic political institutions, decentralization and regional democratization, and democratic values among the public. In the same vein, although specific national characteristics condition different types of consolidation problems, legislators often refer to economic factors and political-institutional relationships as the main problems for the consolidation of democracy. The meanings of democracy for legislators have been inferred from the advantages that these representatives find in this political regime, the most important one being the protection of individual rights and liberties. The second advantage most referred to is that democracy is a political system that provides economic growth, allows citizens to elect governments, facilitates equality of opportunities, and favors human and minority rights. Except for economic growth, democratic advantages are linked by legislators to civil and political rights, which constitute the basis of their view of democracy. Political learning and the internalization of institutional elements that make the existence and functioning of democracy possible, in combination with institutional trust, are key elements in the assessment of democratic stability. The different political paths followed by, on one hand, Ecuador, Bolivia, and Venezuela (where legislators display declining levels of democratic attitudes and institutional trust) and, on the other, Chile, Uruguay, and the Dominican Republic (where democratic attitudes among legislators are

A-3

10/2/07

3:35 PM

Page 85

THE MEANINGS OF DEMOCRACY AND INSTITUTIONAL TRUST

85

on the upswing) underline the importance of examining democratic orientations in order to assess the prospects of democratic consolidation in Latin America.

■ Notes 1. Data for this chapter stem from the PELA (1994–2005). More specifically, this chapter analyzes data from interviews conducted in what we call the “second round” (which includes legislatures between 1997 and 2002) and the “third round” (which includes legislatures between 2002 and 2006); see Chapter 1, Table 1.1, in this volume to examine the distribution of interviews by country and round. The data gathered in the first round of interviews have not been used in the present chapter, because in the second and third rounds the questions relevant to our analysis and longitudinal comparison were put to legislators in a greater number of countries than in the first round. This analysis is mainly guided by a cross-national comparative perspective, but it also considers changes across time in individual countries. 2. In both cases, the Kendall Tau is significant at the .01 level. The question on elections reads as follows: “And what is your degree of confidence in the last electoral processes that took place in your country? To indicate your level of confidence, please use the following scale, which ranges from 1 to 5, where ‘1’ corresponds to minimum confidence and ‘5’ to maximum confidence.” 3. Brutonel, Gunther, and Higley (1992). 4. Mainwaring and Scully (1995). 5. Alcántara (2004a:226). 6. Leiras (2002:9). 7. Cavarozzi and Abal Medina (2002). 8. In both cases, the Kendall Tau is significant at the .01 level. 9. This typology aims to distinguish between legislators using exclusive categories, so that those who have a position will be defined by the three characteristics that result from the indicators used. 10. They correspond to the 1995–2000 legislature in Peru and the 2002–2006 legislature in Ecuador. 11. See Lipset (1959); Cutright (1963); Almond and Verba (1965); Rosenbaum (1975); Muller and Seligson (1994). 12. Some of these paradigms emphasize economic development or the existence of certain political and cultural orientations. 13. See O’Donnell (1994, 2002); Mainwaring and Shugart (1997); Mainwaring and Pérez-Liñán (2004a). 14. “One of the pillars of participatory democracy lies in the respect for rights considered basic to the population: civil rights and political rights. The respect for these rights implies the respect for, among other things, freedom of expression, freedom of the press, freedom of assembly, universal suffrage, individual votes and free elections” (IDD-Lat, 2004: 21). 15. Diamond (1992); Londegran and Poole (1996); Przeworski and Limongi (1997). 16. Mainwaring and Pérez-Liñán (2004a:221). 17. Pharr and Putnam (2000). 18. Morán (1997:201). 19. Field research was conducted in Bolivia and Ecuador in 2003. 20. Other analyses using interviews of politicians also reveal declining levels of support for political parties (PNUD, 2004).

A-3

10/2/07

3:35 PM

Page 86

A-4

10/2/07

3:35 PM

Page 87

4 Legislators and Political Agendas Manuel Alcántara Sáez

■ The Main Problems Governments Face The PELA surveys allow us to analyze legislators’ perceptions of the main challenges faced by their respective governments and the obstacles to democratic consolidation. That this survey provides us with data about most Latin American countries—and at different points in time—also makes it possible to analyze changes and continuities in legislators’ views and to compare across nations in search of shared patterns and distinctive trajectories. Responses to the open question on the main problems faced by the respondents’ countries are divided into six groups, as shown in Tables 4.1, 4.2, 4.3, and 4.4. Three groups name general problems: economic, social, and political. The economic group includes responses to questions about inflation, external debt, or unemployment; the social group includes responses concerning poverty, social inequality, and discrimination; and the political group, responses relating to governability, relationships among the state branches, and the role of political parties. The three remaining groups refer to specific and critical problems in Latin American democracies—violence and crime, corruption, and drug trafficking.1 Our data show that in spite of the presence of important cross-national differences, Latin American legislators tend to assign special importance to economic problems. Thus, in the second round (see Table 4.1), economic problems are considered as the most important by more than half of the legislators of Argentina, Peru, and Venezuela, and by more than two-thirds in Costa Rica, Nicaragua, Panama, and Uruguay. In Uruguay, more than 90 percent of respondents refer to economic problems as the most important; in Bolivia and the Dominican Republic, however, social problems are considered the most important. In Mexico, economic and social problems are identified as the most important by similar percentages (around 30); violence

87

A-4

10/2/07

3:35 PM

88 Table 4.1

Page 88

POLITICS AND DEMOCRACY Main Problems Faced by the Country’s Government, Second Round (row percentagea)

Country (Year the Survey Was Conducted) Argentina (1998) Bolivia (1998) Chile (1998) Colombia (1998) Costa Rica (1998) Dominican Republic (2000) Ecuador (1998) El Salvador (2000) Guatemala (2002) Honduras (1998) Mexico (2001) Nicaragua (1998) Panama (2002) Paraguay (1998) Peru (1995) Uruguay (2001) Venezuela (1995)

Economic Social Problems Problems

Violence and Political Public Drug Problems Insecurity Corruption Trafficking

55.4 28.6 37.2 12.1 67.3

18.7 32.7 23.2 9.4 6.1

9.4 1.0 25.7 4.2 16.2

29.5 44.6 34.0 29.1 49.3 31.7 67.1 73.4 46.2 52.7 91.2 59.9

38.2 17.9 8.8 8.9 8.5 30.1 10.0 1.6 16.4 37.9 5.6 7.2

4.0 8.1 17.2 15.2 17.0 15.7 6.2 18.4 3.6 3.2 7.5

1.6 4.7 54.1 1.0 36.5 7.6 31.0 7.3 3.1 7.7

7.0 4.1 1.2 6.0 8.2

24.5 14.2

1.9 11.6 17.7 1.4 4.9 2.9 15.6 10.8

2.4

1.1 12.7

n 128 98 86 88 49 102 112 64 79 71 123 70 64 65 87 68 69

Source: Data based on PELA 1994–2005. Notes: The question read: “In your opinion as a Legislator, what is the main problem presently faced by your government?” a. Percentages for less frequently mentioned problems are not reported.

and crime are identified as the most important problems in Colombia and El Salvador and are considered crucial by a large percentage of Honduran legislators as well. Other problems are thought far less important: Chile is the only case in which more than 20 percent of the respondents name political problems as the most important; in no country is corruption mentioned by 20 percent; and drug trafficking receives attention only in Bolivia and Colombia. Concern about economic problems increases in the third round (see Table 4.2). Nearly three-fourths of legislators in Panama, Costa Rica, and Chile consider economic problems to be the most important ones faced by their respective governments. Between 55 and 65 percent of the respondents in Argentina, Uruguay, Venezuela, the Dominican Republic, Honduras, and Bolivia also mention economic problems as the most important ones. That is to say, in eleven out of seventeen countries, legislators perceive economic problems to be of crucial importance. Legislators in Colombia, Guatemala, Mexico, and Nicaragua pay the least attention to economic problems. Social problems receive the most attention in Colombia, Peru, and Uruguay, while

A-4

10/2/07

3:35 PM

Page 89

LEGISLATORS AND POLITICAL AGENDAS Table 4.2

89

Main Problems Faced by the Country’s Government, Third Round (row percentagea)

Country (Year the Survey Was Conducted) Argentina (2004) Bolivia (2003) Chile (2002) Colombia (2003) Costa Rica (2002) Dominican Republic (2003) Ecuador (2003) El Salvador (2004) Guatemala (2004) Honduras (2002) Mexico (2004) Nicaragua (2002) Panama (2004) Paraguay (2003) Peru (2001) Uruguay (2005)

Economic Social Problems Problems

Violence and Political Public Drug Problems Insecurity Corruption Trafficking

56.5 64.9 72.6 19.4 74.5

17.4 7.1 6.8 21.9 9.9

10.3 9.1 16.0

57.5 43.9 46.3 29.1 57.9 36.8 37.3 75.0 50.9 53.0 55.6

15.1 10.2 21.3 12.4 7.6 18.8 17.0 7.4 17.9 22.8 31.2

4.3 31.6 17.6 11.6 2.0 28.1 41.4 1.5 8.0 10.7 7.1

10.0 1.2 1.1 45.4

.8 10.2 4.9 5.9

6.5

3.4 1.0 12.5 26.0 21.8 2.4

10.4 5.1 1.3

4.2

7.9

1.5 10.7 2.5

3.6 7.2 2.8 7.4 7.2 11.1

2.9

1.3 2.4

n 104 80 88 95 51 118 98 80 121 102 124 60 68 56 82 80

Source: Data based on PELA 1994–2005. Notes: The question read: “In your opinion as a Legislator, what is the main problem presently faced by the government of your country?” a. Percentages for less frequently mentioned problems are not reported.

political problems are the most important to 41 percent of Nicaraguan legislators. After the Nicaraguans, Ecuadorian and Mexican representatives assign social problems more importance than do all other Latin American legislators. Violence and public insecurity are the most important problems for Colombian and Guatemalan legislators (chosen by 45.5 percent and 26 percent of the legislators, respectively), and the second most important problem for 21.8 percent of Honduran legislators. Corruption is considered a far less important problem, except in Bolivia, Peru, and the Dominican Republic, where around 10 percent of respondents in each of those countries refer to it as the most critical. As for drug trafficking, its importance decreases from the second round to the third round in Colombia and Bolivia. Very small percentages of Salvadoran, Mexican, and Dominican legislators refer now to this problem as one of critical importance. The comparison between the second and third rounds underscores the importance of economic problems in Latin American countries.2 However, as Table 4.3 reveals, there are also important cross-national differences. Concern about economic issues is crucial and constant in Panama, Uruguay, and Costa Rica, closely followed by Argentina and Venezuela. In Bolivia,

A-4

10/2/07

3:35 PM

90

Page 90

POLITICS AND DEMOCRACY

Table 4.3

A Classification of Latin American Countries According to the Importance Legislators Assign to Economic, Social, and Political Problems (second and third rounds) Economic Problems

Very high importance High importance

Medium importance

Low importance

Very low importance

Panama Uruguay Costa Rica Venezuela Argentina Chile Honduras Peru Nicaragua Paraguay Bolivia Ecuador Dominican Republic El Salvador Mexico Guatemala Colombia

Social Problems

Political Problems

Peru

Uruguay Mexico Dominican Republic Argentina Bolivia Paraguay Colombia Chile El Salvador Ecuador Nicaragua Guatemala Venezuela Costa Rica Honduras Panama

Nicaragua Mexico Chile Ecuador El Salvador

Guatemala Paraguay Costa Rica Argentina Venezuela Peru Uruguay Bolivia Dominican Republic Panama Colombia Honduras

Source: Based on PELA 1994–2005. Note: Cases were classified according to the following criteria: average values between 60 percent and 75 percent, very high importance; average values between 45 percent and 60 percent, high importance; average values between 30 percent and 45 percent, medium importance; average values between 15 percent and 30 percent, low importance; average values below 15 percent, very low importance.

Chile, the Dominican Republic, and, to a lesser degree, El Salvador and Ecuador, the importance of economic problems increases as economic conditions deteriorate. In the remaining countries, with two exceptions, the importance of economic issues remains constant; in Nicaragua and

A-4

10/2/07

3:35 PM

Page 91

LEGISLATORS AND POLITICAL AGENDAS

91

Uruguay, other problems were considered more important by the country’s representatives. The perceived importance of social problems increases in Uruguay, Colombia, El Salvador, and, to a lesser degree, Nicaragua. By contrast, in Bolivia, the Dominican Republic, and to a less significant extent in Chile, Ecuador, Mexico, and Peru, the perceived importance of social problems decreases. The importance of political problems increases in Ecuador, Nicaragua, and, to a smaller extent, Mexico, while it decreases in Chile, Costa Rica, and Paraguay (see Table 4.4).

■ Perceptions of Country Problems and Ideology This section analyzes whether the legislators’ self-placement in the leftright dimension, where 1 stands for left and 10 stands for right,3 helps explain perceptions of countries’ main problems. This analysis focuses on those countries in which there are wide differences among ideological groups regarding perceptions of national problems.4 Figures 4.1 and 4.2 display those problems for which there are wider differences among ideological groups, provided that this difference is at least 20 percentage points. In the second round, as shown in Figure 4.1, the country that displays the greatest differences in perceptions of problems is El Salvador, where, in three out of six categories of problems, wide differences exist among the three ideological groups we are considering. These differences concern views of economic and social problems, as well as of violence and public insecurity. By contrast, in Argentina, Ecuador, Nicaragua, Paraguay, and Uruguay, there are no important ideological differences; in Bolivia, Colombia, Mexico, Panama, and the Dominican Republic, there are differences concerning only one group of problems. In seven out of twelve countries, economic problems are perceived quite differently depending on the legislators’ ideology: Chile, Costa Rica, El Salvador, Honduras, Peru, the Dominican Republic, and Venezuela. Ideological leanings condition wide differences in perceptions of social problems in El Salvador and Peru; of political problems in Chile, Costa Rica, Guatemala, and Mexico; of violence and public insecurity in Colombia, El Salvador, Honduras, and Venezuela; of corruption in Guatemala and Panama; and of drug trafficking in Bolivia. In the third round, the country that shows the widest differences among ideological groups concerning the largest number of problems is Nicaragua (see Figure 4.2). However, in this round there are hardly any differences by ideological orientation in the cases of Bolivia, Colombia, Honduras, the Dominican Republic, and Venezuela. In Argentina, Ecuador, El Salvador, Mexico, Paraguay, and Peru, differences concern only one group of problems. In ten out of twelve countries, economic problems are perceived differ-

A-4

Country Argentina Bolivia Chile Colombia Costa Rica Dominican Republic Ecuador El Salvador Guatemala Honduras Mexico Nicaragua Panama Paraguay Peru Uruguay Venezuela

Social Problems

Political Problems

Second

Third

Second

Third

Second

Third

55.4 28.6 37.2 12.1 67.3 29.5 44.6 34.0 29.1 49.3 31.7 67.1 73.4 46.2 52.7 91.2 59.9

56.5 64.9 72.6 19.4 74.5 57.5 43.9 46.3 29.1 57.9 36.8 37.3 75.0 50.9 53.0 55.6 57.0

18.7 32.7 23.2 9.4 6.1 38.2 17.9 8.8 8.9 8.5 30.1 10.0 1.6 16.9 37.9 10.3 7.2

17.4 7.1 6.8 21.9 9.9 15.1 10.2 21.3 12.4 7.6 18.8 17.0 7.4 17.9 22.8 31.2 14.7

7.0 4.1 1.2 6.0 8.2 1.9 11.6

10.3 9.1 16.0

17.7 1.4 4.9 2.9 15.6 10.8

7.9 4.3 31.6 17.6 11.6 2.0 28.1 41.4 1.5 8.0 10.7 7.1 8.9

Violence and Public Insecurity Second 1.6 4.7 54.1 1.0

Corruption

Drug Trafficking

Third

Second

Third

Second

Third

10.0 1.2 1.1 45.4

9.4 1.0 25.7 4.2 16.2 4.0 8.1 17.2 15.2

.8 10.2

24.5

2.9

14.2

6.5

36.5 7.6 31.0 7.3

3.4 1.0 12.5 26.0 21.8 2.4

3.1 7.7

1.5 10.7

12.7

2.5 12.6

17.0 15.7 6.2 18.4 3.6 4.1 7.5

4.9 5.9 10.4 5.1 1.3 3.6 7.2 2.8 7.4 7.2 11.1

4.2 1.3 2.4

1.1

1.0

Source: Data based on PELA 1994–2005. Note: The question read: “In your opinion as a Legislator, what is the main problem presently faced by your government?”

n Second

Third

128 98 86 88 49 102 112 64 79 71 123 70 64 65 87 68 69

104 80 88 95 51 118 98 80 121 102 124 60 68 56 82 80 100

Page 92

Economic Problems

3:35 PM

Main Problems Faced by the Country’s Government (percentage)

10/2/07

92

Table 4.4

A-4

10/2/07

3:35 PM

Page 93

Figure 4.1 Main Problems Faced by the Country’s Government (percentage) by Ideological Self-Placement, Second Round Bolivia, 1998

(N =

31

31

13)

Chile, 1998

(N =

28

32

24)

continues

A-4

10/2/07

3:35 PM

Page 94

Figure 4.1 Continued Colombia, 1998

(N =

33

41

18)

Costa Rica, 1998

76.6

75.0

60.0

30.0 25.0

13.3

12.5

10.0

10.0 6.6 0.0

(N =

0.0

10

30

0.0

0.0

0.0

0.0

0.0

0.0

8)

continues

A-4

10/2/07

3:35 PM

Page 95

Figure 4.1 Continued Dominican Republic, 2000

(N =

40

43

17)

El Salvador, 2004

(N =

26

16

22)

continues

A-4

10/2/07

3:35 PM

Page 96

Figure 4.1 Continued Guatemala, 2002

(N =

19

43

14)

Honduras, 1998

(N =

10

39

22)

continues

A-4

10/2/07

3:35 PM

Page 97

Figure 4.1 Continued Mexico, 2001

(N =

38

62

17)

Panama, 2002

(N =

9

32

23)

continues

A-4

10/2/07

3:35 PM

Page 98

Figure 4.1 Continued Peru, 1995

(N =

20

37

29)

Venezuela, 1995

(N =

32

27

Source: Created by the authors based on PELA 1994–2005.

5)

A-4

10/2/07

3:35 PM

Page 99

Figure 4.2 Main Problems Faced by the Country’s Government (percentage) by Ideological Self-Placement, Third Round Argentina, 2004

(N =

30

40

8)

Chile, 2002

(N =

28

29

27)

continues

A-4

10/2/07

3:35 PM

Page 100

Figure 4.2 Continued Costa Rica, 2002

77.8 71.4

28.5

14.3

14.3

0.0

(N =

14

27

7)

Ecuador, 2003

(N =

46

27

24)

continues

A-4

10/2/07

3:35 PM

Page 101

Figure 4.2 Continued El Salvador, 2004

(N =

32

14

34)

Guatemala, 2004

(N =

33

33

29)

continues

A-4

10/2/07

3:35 PM

Page 102

Figure 4.2 Continued Mexico, 2004

(N =

36

38

24)

Nicaragua, 2002

(N =

28

15

15)

continues

A-4

10/2/07

3:35 PM

Page 103

Figure 4.2 Continued Panama, 2004

(N =

10

27

28)

Paraguay, 2003

(N =

14

28

14)

continues

A-4

10/2/07

3:35 PM

Page 104

Figure 4.2 Continued Peru, 2001

(N =

32

36

10)

Uruguay, 2005

(N =

31

28

Source: Created by the authors based on PELA 1994–2005.

3)

A-4

10/2/07

3:35 PM

Page 105

105

LEGISLATORS AND POLITICAL AGENDAS

ently depending on the legislators’ ideology (Chile, Costa Rica, Ecuador, El Salvador, Guatemala, Mexico, Nicaragua, Panama, Paraguay, Peru); this is true for social problems in four countries (Costa Rica, Nicaragua, Panama, Uruguay); political problems in four countries (Argentina, Chile, Guatemala, Nicaragua); and violence and public insecurity in Uruguay. In the case of drug trafficking, there are no great differences in any of the countries in this sample. Figures 4.1 and 4.2 show graphically the magnitude of differences in perceptions of problems. Arrows in these figures single out cases in which ideology conditions very different perceptions of problems. These are, especially, Chile, El Salvador, Guatemala, and Mexico in the second round (see Figure 4.1), and Argentina, Chile, El Salvador, Guatemala, Panama, and Uruguay in the third round (see Figure 4.2). In several countries, the presence of important differences by ideological group is consistent across time: Chile and, to a lesser degree, Costa Rica, El Salvador, Guatemala, and Peru (see Table 4.5).

■ Legislators’ Perceptions and National Realities The persistent, crucial importance that legislators assign to economic problems (with the exception of Colombian legislators) suggests the need to analyze the relationships between national realities and the views of parliamentarians. As recent studies have shown (Hagopian, 2005), economic security, public insecurity, and corruption, in that order, are rated as the most important problems by Latin American citizens. These public perceptions most likely condition the legislators’ views of their nations’ problems. Table 4.6 reveals, however, that legislators’ perceptions are not mechanically determined by the objective economic problems in their countries. For instance, legislators in Panama, Uruguay, Costa Rica, Venezuela, and Argentina show the highest concern about economic problems; however, according to the economic figures reported in Table 4.6, these countries did not face especially intense economic problems during the second round.

Table 4.5

Economic Problems Chile Costa Rica El Salvador Peru

Differences in Perceptions of Problems in Both the Second and Third Rounds Social Problems

Political Problems

Violence and Insecurity

Corruption

Drug Trafficking



Chile Guatemala







A-4

10/2/07

3:35 PM

106

Page 106

POLITICS AND DEMOCRACY

Table 4.6

Main Problems Faced by the Country’s Government and Economic Performance (percentage) Economic Problems

Country Argentina Bolivia Chile Colombia Costa Rica Dominican Republic Ecuador El Salvador Guatemala Honduras Mexico Nicaragua Panama Paraguay Peru Uruguay Venezuela

Annual GDP Growtha

Inflationa

Unemploymentb

Second

Third

Second

Third

Second

Third

Second

Third

55.4 28.6 37.2 12.1 67.3 29.5 44.6 34.0 29.1 49.3 31.7 67.1 73.4 46.2 52.7 91.2 59.9

56.5 64.9 72.6 19.4 74.5 57.5 43.9 46.3 29.1 57.9 36.8 37.3 75.0 50.9 53.0 55.6 57.0

3.9 5.0 3.9 0.6 8.4 7.3 2.1 2.2 2.2 2.9 –0.1 3.7 2.2 –0.4 8.6 –3.4 4.0

8.8 2.5 2.2 3.9 2.9 –0.4 2.7 1.8 2.1 2.0 1.3 1.0 4.1 2.6 0.2 2.5 3.2

0.9 7.7 5.1 18.7 11.7 7.7 36.1 2.3 8.0 13.7 6.4 13.0 1.0 11.5 11.1 4.4 59.9

13.4 3.3 2.5 7.1 9.2 27.4 7.9 2.1 5.5 7.7 4.5 4.0 1.4 14.2 2.0 19.4 16.2

12.9 6.1 6.4 15.3 5.4 13.9 11.5 6.5

21.0 8.5 9.0 17.6 6.8 16.1 8.7 7.1

5.2 2.5 13.2 16.1 6.6 8.2 15.3 10.3

6.2 2.8 12.9 16.1 9.3 17.0 14.0

Sources: a. Data excerpted from World Bank 2005. b. Data from PNUD 2004. Notes: Data drawn from the second round correspond to the years the study was conducted in each of the countries. Data for the third round, when possible, correspond to the years the study was conducted. However, the most recent data are from 2003 for annual GDP growth and inflation, and from 2002 for the remaining variables.

Also remarkable is the fact that in Colombia only 12.1 percent of legislators in 1998 considered economic problems to be the most important ones—at that time GDP growth was lower than 0.6 percent and inflation had reached 18.7 percent. These figures are clearly worse than those of many countries in which economic problems are considered the most important ones faced by the government. Of course, the perceptions of Colombian legislators may be conditioned by the high levels of violence present in this country. In contrast to economic problems, Latin American legislators, with the only exception of Peru, do not consider social problems to have very high or high importance; this, despite the fact that, according to data in Table 4.7, in the majority of countries, more than 50 percent of the population is below the poverty line. Finally, neither of the two indexes of democratic development included in this analysis seems to be connected to perceptions of political problems (see Table 4.8). The legislators who assign the most importance to political problems are those of Nicaragua, Mexico, Chile, Ecuador, and El Salvador.

A-4

Main Problems Faced by the Country’s Government and Data on Social Characteristics and Corruption (percentage)

Argentina 18.7 Bolivia 32.7 Chile 23.2 Colombia 9.4 Costa Rica 6.1 Dominican Republic 38.2 Ecuador 17.9 El Salvador 8.8 Guatemala 8.9 Honduras 8.5 Mexico 30.1 Nicaragua 10.0 Panama 1.6 Paraguay 16.9 Peru 37.9 Uruguay 10.3 Venezuela 7.2

Violence and Social Insecurity

Homicides per 100,000 Inhabitantsb

Corruption Perception Indexc

Corruption

Third

Second

Third

Second

Third

Second

Third

Second

Third

Second

Third

17.4 7.1 6.8 21.9 9.9 15.1 10.2 21.3 12.4 7.6 18.8 17.0 7.4 17.9 22.8 31.2 14.7

17.8 62.1 23.2 50.9 22.5 46.9 56.2 49.8 59.9 79.1 41.1 69.9 25.3 46.3 47.6 9.4 48.0

41.5 33.6 20.6 50.6 20.3 44.9 49.0 48.9

1.6

10.0 1.2 1.1 45.4

5.3

7.0

.8 10.2

3.4 81.2 5.7 11.7 14.4 46.2

5.3 84.6 6.2

7.0 4.1 1.2 6.0 8.2 1.9 11.6

3.0 2.8 6.8 2.2 5.6 3.1 2.3 4.1 2.5 1.7 3.7 3.0 3.0 1.5 4.5 4.3 2.3

2.5 2.3 7.5 3.7 4.5 3.3 2.2 4.2 2.2 2.7 3.6 2.5 3.7 1.6 4.1 6.2 2.7

77.3 39.4 69.4 61.0 54.8 15.4 49.4

4.7 54.1 1.0 36.5 7.6 31.0 7.3

3.4 1.0 12.5 26.0 21.8 2.4

3.1 7.7

1.5 10.7

12.7

2.5 12.6

12.1 11.3 13.2 17.5 6.2 5.2 15.6

15.7 43.4 11.1 12.3 13.7 18.4 4.5 5.2 26.0

17.7 1.4 4.9 2.9 15.6 10.8

4.9 5.9 10.4 5.1 1.3 3.6 7.2 2.8 7.4 7.2 11.1 1.0

107

Sources: a. CEPAL 2003: 54–55. b. Pan American Health Organization 2004. Notes: Data drawn from the second round correspond to the years the study was conducted in each of the countries. Data for the third round correspond to the years the study was conducted when possible. However, the most recent data are from 2003 for annual GDP growth and inflation, and from 2002 for the remaining variables. c. Transparency International Latinoamérica y el Caribe (TILAC). This index is based on levels of corruption as perceived by entrepreneurs, risk analysts, and the public in general. This index ranges from “0” (“extremely corrupt”) to “10” (“extremely honest”).

Page 107

Second

% Population Below the Poverty Linea

3:35 PM

Poverty and Social Inequality Country

10/2/07

Table 4.7

A-4

10/2/07

3:35 PM

108

Page 108

POLITICS AND DEMOCRACY

Table 4.8

Main Problems Faced by the Country’s Government and Political Data (percentage)

Political Problems Country Argentina Bolivia Chile Colombia Costa Rica Dominican Republic Ecuador El Salvador Guatemala Honduras Mexico Nicaragua Panama Paraguay Peru Uruguay Venezuela

Index of Citizen Perceptions of Democracya

Index of Democratic Developmentb

Second

Third

2000–2004

2002–2005

7.0 4.1 1.2 6.0 8.2 1.9 11.6

10.3 9.1 16.0

59 40 53 41 59

3.918 2.883 8.757 4.218 8.575 2.631 2.376 4.452 3.884 3.107 6.136 2.963 6.914 3.214 4.352 8.366 2.243

17.7 1.4 4.9 2.9 15.6 10.8

7.9 4.3 31.6 17.6 11.6 2.0 28.1 41.4 1.5 8.0 10.7 7.1 8.9

38 42 34 51 47 47 53 38 41 66 53

Sources: a. Hagopian 2005:334. This index summarizes, for the 2000–2004 period, six average percentages relative to support for democracy, satisfaction with democracy, importance of voting, valid votes, trust in government, and democracy as the best political regime. b. Índice de Desarrollo Democrático de América Latina IDD-LAT 2005. Index calculated as the addition of three main dimensions: respect for political rights and civil liberties, index of institutional quality and political efficiency, and a third index of the ability to generate welfare policies and of economic efficiency. Note: Data referring to the second round correspond to the years the study was conducted in each of the countries.

However, this group of countries includes only one case of very low democratic quality (Ecuador). This group of countries also includes one of the countries of higher democratic quality (Chile).

■ Conclusion In a majority of Latin American countries, economic problems play the most prominent role in the agendas of the parliamentary elites. Colombia and Guatemala, where violence and drug trafficking are seen as the most important problems, are the only persistent exceptions to this pattern. In spite of their objective importance, social problems play a secondary role in legislators’ perceptions of the problems their government faces. Finally, political problems engage the interest of a very limited number of represen-

A-4

10/2/07

3:35 PM

Page 109

LEGISLATORS AND POLITICAL AGENDAS

109

tatives. This is so in spite of the fact that political and institutional reforms rank high in the programs of contemporary Latin American democracies. In general, comparison between legislators’ perceptions regarding national problems and the real characteristics of Latin American countries suggests that the views of legislators do not directly reflect the main characteristics of their respective countries; thus, much work is needed to identify the different factors that shape these politicians’ views. As this chapter has shown, in five of these countries ideology is one factor that clearly conditions the perceptions parliamentarians have regarding the main problems faced by their respective governments.

■ Notes 1. These data belong to the second and third rounds of this survey. On the characteristics of these rounds, see Chapter 1. 2. It may be useful to remember that, according to PNUD (2004), for 34.6 percent of Latin American citizens, the main problem was unemployment, followed by poverty, social inequality, and insufficient income for 26.3 percent; crime and drugs for 11.9 percent; corruption for 11.2 percent; political violence for 7.5 percent; and services and deficient infrastructures for 6.8 percent of respondents. 3. For this analysis, values have been recoded in the following way: left (values between 1 and 4), center (values between 5 and 6), and right (values between 7 and 10). 4. Differences are considered “wide” when they are 20 percent or more.

A-4

10/2/07

3:35 PM

Page 110

A-5

10/2/07

3:36 PM

Page 111

PART 2 Party Politics and Ideology

A-5

10/2/07

3:36 PM

Page 112

A-5

10/2/07

3:36 PM

Page 113

5 The Organization of Political Parties Leticia M. Ruiz Rodríguez

AT THE PRESENT TIME, THE LEGITIMACY OF POLITICAL PARTIES AND OF

their members’ behavior is at an all-time low. The factor of nostalgia does not enter into this case, since Latin American political parties have never, in general terms, had a golden age.1 However, various indicators suggest that levels of legitimacy have sunk so low that it is essential to discover the fundamental reasons for this phenomenon. Various authors have already addressed this task, some more successfully than others.2 Citizen mistrust of political parties is not unique to Latin America, but the phenomenon is especially strong in this region. Juan Linz (2002) states that political parties throughout the world are facing a paradoxical situation: recognition of their importance to the functioning of democracies exists side by side with high levels of dissatisfaction with and mistrust of parties. In the case of Latin America, there has never been such an uninterrupted stretch of elections in so many countries; at the same time, citizens’ confidence in parties has never been so low. In other words, parties, which— through their organization, presence, and participation in elections—have been central agents in achieving status for democracy throughout the region, are now going through an unprecedented crisis. This chapter studies legislators’ views of both the levels of participation in parties as well as the levels of involvement permitted to average members in making decisions. However, these two organizational elements are not the only ones that contribute to the crisis referred to above. Within the framework of a situation with multiple causes, ideological articulation and programmatic coherence are also significant. Moreover, other factors affect the perception of political parties as well, such as the inability to represent and aggregate interests. Other examples would be the behaviors of parties when in power, such as not keeping promises and party actions in the legislative branch. Lack of party discipline as well as corruption at all levels

113

A-5

10/2/07

3:36 PM

114

Page 114

PARTY POLITICS AND IDEOLOGY

are other possible causes of the mistrust of parties. The organizational element, however, is more basic than these causes. That ideology and organization are the two aspects that define a party justifies the attention to these elements in this chapter; nevertheless, it is important to also pay attention to other, interrelated party spheres and dynamics. For this reason, this chapter also looks at the intensity of party life. The elitist nature of party functioning, described as a deficit since 1915 (Michels, 1991), is inevitable in any party organization. We show that, on one hand, all parties go beyond a certain minimal level of participation; in some cases, this level is quite high. On the other hand, in few parties are average members able to influence decisionmaking processes. This situation is described in this chapter as one of participatory parties, but with responsibility distributed vertically and not horizontally. This chapter also examines the relationship of these aspects to the variable of party ideology, since, as stated above, ideology and organization are both definitional aspects of a political party. This study of party participation and distribution of decisionmaking responsibilities is based on party elites’ perceptions of these issues.3 Opinions were gathered in legislative periods which either have not yet finished or were immediately previous to the present legislature.4 How parties manage themselves as organizations helps reveal the kinds of parties they are. Thus, we offer a new model of perceived party life and an index that takes into account both of these organizational aspects (participation and responsibility).

■ The Study of Party Organization in Latin America Despite the importance of the organizational aspects of party life, it is only since the 1990s that this issue has been studied in a systematic way. Aspects analyzed include parties’ histories,5 their roles in transition and consolidation6 and in political systems,7 organizational questions (such as how candidates are selected),8 the level of party institutionalization,9 and internal discipline. 10 Less attention was paid to internal organization than to other aspects.11 Until now, studies examining how parties are structured have been few and far between and have given little emphasis to comparative analyses. The multidimensional aspects of the topic and the need for empirical evidence have been obstacles in researching party organization and its impact on the components of democratic regimes. At the present time, there is much work to be done in generating systematic empirical knowledge on this subject. There are at least three possible paths to follow. One path is to analyze further organizational aspects in Latin America, through exploring concepts drawn from other geographical areas or through generating new concepts.12 In Europe and the United

A-5

10/2/07

3:36 PM

Page 115

THE ORGANIZATION OF POLITICAL PARTIES

115

States as well, despite the fact that the largest number of studies on organization have been carried out in these areas, relatively few attempts have been made to find indicators adapted to organization concepts.13 Thus, when researchers use such terms as “complexity,” “centralization,” “involvement,” or “factionalization”—to cite some dimensions of the organizational reality of political parties proposed by Kenneth Janda (1993)— there is still no consensus with respect to their operationalization. A second path is to rethink the suitability of existing classifications relating to organizational aspects for Latin American parties. In other regions, fundamentally in the case of Europe, a multitude of classifications exist that relate to organizational aspects; none has universal value, but rather is chosen depending on the purpose and needs of the study in question. 14 A significant number of these classifications are variations of Maurice Duverger’s outline (1951), which groups parties as cadre organizations (as opposed to mass parties). The organizational evolution of political parties has been different in Latin America, in part because of the different challenges faced in this region. In Europe, organizations have reacted to two basic phenomena: the increasing ability of social movements to attract representation and identification (and the decreasing ability of political parties to do the same), and the falling levels of party membership.15 In Latin America, however, parties have made organizational changes in response to political decentralization, reform of electoral systems, the change from populist to neoliberal states, and the increasing presence of mass media.16 Given that Latin America’s challenges are different, its resulting organizational structures have often been dissimilar to the ones in present-day Europe. Organizational evolution has also been different because the starting point was not the same: Latin American parties were, from the beginning, not like European parties. Nevertheless, the opposition underlined by Duverger is central to attempts to classify parties, in spite of these differences between the two regions and although this classification has some limitations.17 The third path is to expand the empirical evidence used to study organizational aspects. Studies that look at party organizations from the members’ points of view are practically nonexistent. To date, most data are based on official documents, such as those giving overall membership rates or rules for selecting candidates, or on the authorities that approve party programs.18 The opinions of party members on these documents and other aspects need to be collected. The first studies using such evidence in scholarly work on Latin American parties were written by Manuel Alcántara (2004a) and edited by Alcántara and Flavia Freidenberg (2001). The infrequent use of this kind of data is due to the high cost of surveying, or carrying out interviews with representative samples. Besides the scarcity of studies, another aspect puts limitations on the

A-5

10/2/07

3:36 PM

116

Page 116

PARTY POLITICS AND IDEOLOGY

study of types of party organizations. In the research on party organizations, there are underlying normative considerations about their effects on political systems as a whole. Although these considerations stimulate the study of the organizational side of political parties, they can occasionally take attention away from a deeper analysis of the subject. For example, it has been stated that understanding parties’ organizational strengths and institutionalizations is central to the understanding of the historical successes and failures of Latin American democracies.19 However, more research is needed to ascertain what factors lead to party situations that are more desirable from a normative point of view. Party organization impacts party behavior in at least three important ways. First of all, it affects the relationship parties establish with voters and members, for it has an effect on such important issues as the level of involvement possible in party structures or the degree of responsibility in decisionmaking processes. It is this issue that is most extensively studied in the present chapter. Second, a party’s organization affects its behavior and performance in the legislative and executive branches, among other venues. In this way, party organization determines key aspects of political life, such as discipline in a parliamentary group and the party’s ability to control the government once the party is in office, as Jean Blondel (1978) suggests with the concept of “partocracy.” Lastly, party organization affects the party’s survival and durability. Although the kind of management most likely to ensure party durability has not yet been extensively studied, it is recognized that this variable enables one to understand the origin and evolution of parties.20 This chapter addresses the question of party organization from the standpoint of the perceptions of its parliamentary elite, one of the most clearly identifiable and influential groups within a political party. The analysis concentrates on aspects related to party life more than on formal aspects of the organizational structure. To this end, what is fundamentally analyzed are levels of participation within political parties and the distribution of responsibilities in decisionmaking processes. Limiting the study to the intraparty aspect does not mean that other facets—such as the electoral or parliamentary side, or the consideration of the party in government—do not have relevant organizational traits. Similarly, although the focus of this chapter is the analysis of the opinions of the parliamentary elite, this does not mean that the perceptions of other groups within political parties are irrelevant.21 However, despite the need for caution in making generalizations, it is undeniable that the perceptions of legislators, a group with a special impact in party organizations, give an approximate idea of the global vision of the party. This is especially true in the context of Latin America, where elitism is traditionally very strong. The analysis of perceptions is useful, since perceptions condition behavior. For example, the normative opinion of a legislator on the best way of organizing

A-5

10/2/07

3:36 PM

Page 117

THE ORGANIZATION OF POLITICAL PARTIES

117

the party is likely to be channeled as pressure for the party to evolve in that direction.

■ Intraparty Life According to the Parliamentary Elite In order to study the internal management of parties, Duverger22 proposed an analysis from the viewpoints both of organizational structure and of intensity and type of internal party life. It is to be expected that this model, which divides the organizational aspect into two dimensions, will be better able to differentiate parties based on levels of participation and internal life than on organizational structure. With respect to organizational structure, existing literature seems to indicate that tendencies are more homogeneous within Latin American parties. Alcántara (2004a) states that, in general, Latin American political parties have continuous structures and, measuring by levels of infrastructure and bureaucratic presence in cities of a certain size, can be said to be fairly spread out throughout national territory. In addition to looking at the variation between parties in terms of party dynamism, studies have tended to concentrate on formal organizational structures more often than on party operation.23 Thus, this study concentrates not so much on the description of formal party structures, which have been addressed extensively in other studies,24 but rather on an analysis of internal party life from the perspective of its members. The objective is to capture the differing intensity of citizen participation in parties as well as the degree to which citizens have a say in party decisionmaking.25 Table 5.1 includes the questions used to shed light on both of these aspects. Consideration is given to how parliamentary elites perceive the levels of party militancy and of participation (dimension of participation in intraparty life). Legislators also give their opinions on the level of internal democracy, the amount of power that should ideally be assigned to party leadership, and on which groups are relevant in the decisionmaking process (dimension of the distribution of responsibility in intraparty life). This second aspect captures the ability of party members to influence decisions, in this case from the point of view of the parliamentary elite. Thus, we analyze the level of participation in parties and also, by evaluating the type of distribution of responsibilities, appraise what this degree of participation actually accomplishes. Attention to both of these aspects is necessary, for although a party may be very participatory, it may be so in activities that have very little impact in the most important decisionmaking processes. An example of this is a party that organizes highly mobilized electoral campaigns but at the same time operates in an elitist fashion when selecting its candidates. Conversely, the number of party militants may be low, but these militants may be highly involved in the elaboration of their party’s electoral program or in the selection of its leaders.

A-5

10/2/07

3:36 PM

118

Page 118

PARTY POLITICS AND IDEOLOGY

Table 5.1

Perceptions for the Analysis of Intraparty Life of Political Parties

Intraparty Life

Subject of Questions

Participation

Level of militancy in the party Level of participation in the party

Distribution of responsibility

Preferred level of power of party leaders Level of internal democracy within the party Important groups in the decisionmaking process

Source: Based on the PELA questionnaire.

Although a comprehensive explanation of the way in which political parties organize goes beyond the scope of this chapter, this study does explore the relationship of participation and responsibility to the ideological orientation of parties and their levels of ideological structuring. Although the relationship of many variables could be studied, the hypotheses put forward most frequently in the literature on party organization are those that link these issues with party ideology.26 In addition, as stated earlier, organization and ideology are variables that are constitutive of a party and that have a fundamental importance in party operation, since they are requisite to other party processes. Degrees of Participation in Political Parties

Among the answers of legislators from eleven countries, there are very few negative evaluations of internal militancy. “Militancy” here refers to the actions of those who officially participate in the party as recognized and accepted members, and who may also pay membership or similar fees. In parties such as the Panamanian PRD; the two Nicaraguan parties, FSLN and PLC; as well as the Paraguayan Movimiento de Colorados Éticos and the Salvadoran FMLN, legislators rate militancy as high (see Table 5.2). In contrast, the lowest ratings for party militancy are found in the three Chilean parties, the RN, PS, and PPD, followed by the Guatemalan Partido Unionista. Looking at party systems as a whole (see Table 5.2, first column), the countries where party militancy levels are rated the highest are Nicaragua, Panama, and Honduras, in contrast to Chile and Argentina, where legislators perceive militancy as being low. In general, legislators rate participation levels as high (see Table 5.2). Based on Susan Scarrow’s (2000) work on parties, “participation” can be defined as the integration of individuals who are not necessarily formal members of a party into party-organized activities that, on many occasions, are linked to the electoral field. These activities constitute a way of socializing and of forming networks of people who will eventually benefit from

A-5

10/2/07

3:36 PM

Page 119

THE ORGANIZATION OF POLITICAL PARTIES Table 5.2

119

Perceived Militancy by Party Legislators (scale of 1 to 5) Party Militancy

Party System

High (5.00–3.6)

Argentina Country average 2.99 Bolivia Country average 3.20

Medium (3.59–2.6)

Low (2.59–1.00)

PJ (3.48)

UCR (2.5)

MNR (3.26) MIR (3.37) NFR (3.07) MAS (3.13)

Chile Country average 2.19

UDI (2.56) PDC (2.4) RN (2.13) PS (2.0) PPD (1.9)

Costa Rica Country average 3.41 Dominican Republic Country average 3.66 Ecuador Country average 3.40

PAC (4.00)

El Salvador Country average 3.72 Guatemala Country average 3.03 Honduras Country average 3.89 Nicaragua Country average 4.24 Panama Country average 3.94 Paraguay Country average 3.71

ARENA (3.62) FMLN (4.07)

PRD (3.73) PRSC (3.72) PRE (3.62) PSC (3.61)

PLN (3.43) PUSC (2.82) PLD (3.53) PRIAN (3.44) ID (3.27) MUPP-NP (3.09) PCN (3.47) PAN (3.46) FRG (3.48)

PL (3.90) PN (3.88) PLC (4.15) FSLN (4.34) PRD (4.63)

PA (3.25)

UNACE (4.14) MPQ (3.71)

ANR (3.57) PLRA (3.44)

PU (2.17)

Source: Data based on PELA 1994–2005. Note: The question read: “How would you rate the level of internal militancy in your party?”

employment opportunities or derive other types of benefits that are not strictly political. In this sense, parties such as the FMLN or the FSLN are almost unanimous in considering that participation in their party is intense and constant. In contrast to these parties, legislators from the PLRA complain that their party’s organization is oriented exclusively toward elections. The case of Chile is a surprising one. In this party system, traditionally considered one of the most institutionalized,27 a high percentage of legislators from the PDC and the PPD believe that participation occurs in their parties only in elections. These answers contradict the stereotype of the Chilean parties as dynamic and as forming an integral part of society.28 In this case,

A-5

10/2/07

3:36 PM

120

Page 120

PARTY POLITICS AND IDEOLOGY

aspects such as the existence of a highly self-critical political class could, in part, possibly explain these results. On the other hand, it is difficult to interpret the legislators’ perceptions of differences between militancy and participation. For the purposes of this chapter, both of these issues are considered to be indicators that allow operationalization of the subject of party life based on individual perceptions. In fact, there is an empirical relationship between both aspects: the higher the perception of militancy, the higher the participation rate, and vice versa (see Table 5.7). In terms of level of participation in political parties, the traditional idea is that members of left parties are more committed and therefore more participatory, whereas right parties are more professionalized, in the sense of being more distant from their support base, which is less participatory. The data show that party ideology is, indeed, related to the level of perceived participation (see Table 5.7), and that in the cases studied here, members of right parties less frequently rate participation levels in their parties as intense and constant (see Table 5.3). Yet, the level of militancy perceived by party members does not have a statistically relevant relationship to ideological orientation. For the time being, our data do not support statements frequently made about the relationship between the variable of ideology and organizational aspects, such as that militancy in left parties tends to be more intense than in right parties, or that leaders in left parties tend to exercise a stricter control than is exercised in right parties. The testing of these hypotheses, which, according to our data, are true for a limited number of cases in Latin America, should be complemented with information on numbers of militants. Such statistics would allow us to see whether it is true that, in Latin America, “right parties are more likely to accentuate strategies of increasing the electoral bases whereas left parties tend to emphasize increasing the number of militants.”29 Then again, the data do show an interesting pattern in the relationship between militancy and the ideological structuring of political parties; more specifically, between militancy and party coherence.30 In parties for which there are data on coherence,31 there tend to be higher levels of militancy in those that are better structured around ideas, programs, and organizational strategies (see relationship in Table 5.7 between militancy and total coherence). Turning this argument around, this finding could lead us to state that members of parties with low militancy tend to have low levels of consensus around substantive issues. A correlation also exists between militancy and coherence in parties on more specific subjects, such as organizational aspects and ideological issues. The tendency of this relationship is the same: the more agreement between party members on how to organize the party, the more militancy in the party; and the more agreement around ideological aspects, the more militancy in the party. Thus, disagreement on substantive issues seems to be a factor correlated to lower party militancy.

A-5

10/2/07

3:36 PM

Page 121

THE ORGANIZATION OF POLITICAL PARTIES Table 5.3

121

Perceived Participation by Party Legislators (percentage) Party Participation

Party System Argentina Bolivia

Chile

Costa Rica Dominican Republic Ecuador

El Salvador Guatemala Honduras Nicaragua Panama Paraguay

Intense and Constant PJ (20.4) UCR (10.5) MNR (31.6) MIR (46.7) NFR (14.3) MAS (73.3) RN (7.1) UDI (73.3)

PUSC (23.5) PLN (35.7) PAC (85.7) PRD (38.9) PLD (59.4) PRSC (22.2) PSC (47.8) PRE (71.4) ID (46.7) MUPP-NP (72.7) PRIAN (60.0) ARENA (44.0) FMLN (96.6) PCN (73.3) PAN (53.8) FRG (50.0) PU (66.7) PL (45.5) PN (41.7) PLC (83.9) FSLN (91.7) PRD (93.1) PA (35.0) ANR (53.8) PLRA (20.0) UNACE (71.4) MPQ (80.0)

Only in Elections PJ (67.3) UCR (66.2) MNR (52.6) MIR (20.0) NFR (57.1) MAS (20.0) PDC (77.0) RN (50.0) UDI (45.8) PPD (73.3) PS (66.7) PUSC (64.7) PLN (42.9) PAC (14.3) PRD (50.0) PLD (31.3) PRSC (66.7) PSC (43.5) PRE (21.4) ID (26.7) MUPP-NP (18.2) PRIAN (20.0) ARENA (48.0) PCN (20.0) PAN (30.8) FRG (36.4) PU (16.7) PL (38.6) PN (43.8) PLC (9.7) FSLN (8.3) PRD (6.9) PA (40.0) ANR (42.3) PLRA (80.0) UNACE (28.6) MPQ (20.0)

Scarce and Marginal PJ (10.2) UCR (26.3) MNR (15.8) MIR (33.3) NFR (28.6) MAS (6.7) PDC (22.0) RN (42.9) UDI (20.8) PPD (26.7) PS (33.3) PUSC (11.8) PLN (21.0) PRD (11.1) PLD (9.4) PRSC (11.1) PSC (8.7) PRE (7.1) ID (26.7) MUPP-NP (9.1) PRIAN (20.0) ARENA (8.0) FMLN (3.4) PCN (3.7) PAN (7.7) FRG (13.6) PU (16.7) PL (15.9) PN (14.6) PLC (6.5) PA (25.0) ANR (3.8)

Source: Data based on PELA 1994–2005. Note: The question read: “How would you rate the level of militant participation in your own party: intense and constant, scarce and marginal, only in elections?”

Degrees of Responsibility in Political Parties

Knowing the distribution of responsibilities in a party’s decisionmaking processes allows one to draw conclusions about how inclusive that party is. “Responsibility” means having the possibility of affecting party decisions,

A-5

10/2/07

3:36 PM

122

Page 122

PARTY POLITICS AND IDEOLOGY

whether those decisions relate to the selection of leaders and candidates, to the elaboration of public policy, or to the definition of the party platform. To get an idea of members’ ability to affect party decisions, we first asked legislators to rate their own parties’ levels of internal democracy. Internal democracy is a per se value that underlies most reforms in political parties. However, despite the fact that the idea of using democratic procedures has become more popular, the concept of “internal democracy” presently has an ambiguous meaning. Even so, we can say that, in general terms, internal democracy is understood to afford the possibility of creating a direct democracy within parties, more than a representative democracy. More specifically, internal democracy is equated to the use of primaries to elect candidates, without taking into account the undesirable consequences that can stem from these kinds of practices, among which the most egregious are factionalization and the necessity of acquiring funding in order to be a successful candidate (see Linz, 2002). In contrast, other relevant aspects of internal democracy, such as involving a large part of the party in the definition of the party platform, tend to be overlooked. If we take internal democracy to mean the involvement of a large number of party members in crucial decisionmaking processes, the fact that parties differ in the levels of democracy they attain is demonstrated by the answers given to this question (see Table 5.4). As was true in evaluations of party militancy, here, too, there is a wide range of viewpoints on this issue within individual party systems. The highest evaluation of internal democracy is expressed by legislators of the Panamanian PRD, followed by those of the FMLN and the FSLN. In contrast, the most pessimistic perceptions of internal party democracy come from legislators of the Chilean PPD and the Panamanian PA. One can arrive at a deeper understanding of legislators’ perceptions of internal democracy by contrasting these answers to those concerning preferred distribution of power within the party, specifically the power exercised by party leaders (see Table 5.5).32 Some legislators clearly desire a change in their parties’ status quo, such as those from the Bolivian MIR, the Panamanian PA, and the PLH. These legislators favor decreasing the power of party leadership. Among all parties studied, legislators from parties commonly seen as caudillistas—that is, highly personalistic and displaying strong personal leaderships—prefer their leadership to have less power. Although no party displays the same degree of consensus in requesting more power for their leadership, in two parties a significant number of respondents favor increasing it; this is the case with the Paraguayan PLRA and the PRE. In the same vein, the question of which groups legislators consider to be “very important in making decisions” gives us an idea of the relevance legislators assign, respectively, to party leaders and party membership. This

A-5

10/2/07

3:36 PM

Page 123

THE ORGANIZATION OF POLITICAL PARTIES Table 5.4

123

Degree of Perceived Internal Democracy by Party Legislators (scale of 1 to 5) Party Degree of Internal Democracy

Party System Argentina Country average 3.01 Bolivia Country average 3.44

High (5.00–3.6)

MAS (4.12)

Chile Country average 3.34

UDI (3.7)

Costa Rica Country average 3.47 Dominican Republic Country average 3.11

PAC (3.93) PLN (3.64)

Ecuador Country average 3.52

ID (3.80) MUPP-NP (3.90)

El Salvador Country average 3.94

PCN (4.13) ARENA (3.62) FMLN (4.07) PAN (3.92) PU (3.86)

Guatemala Country average 3.71 Honduras Country average 3.37 Nicaragua Country average 3.87 Panama Country average 3.5 Paraguay Country average 3.47

Medium (3.59–2.6)

Low (2.59–1.00)

UCR (3.21) PJ (2.82) MIR (3.44) NFR (3.31) MNR (2.89) PDC (3.5) RN (3.4) PS (3.3) PPD (2.8) PUSC (2.53) PLD (3.38) PRSC (3.07) PRD (2.89) PRE (3.57) PRIAN (3.30) PSC (3.05)

FRG (3.36) PN (3.42) PL (3.33)

PLC (3.67) FSLN (4.07) PRD (4.20)

PA (2.80)

MPQ (3.75) UNACE (3.57)

ANR (3.57) PLRA (3.00)

Source: Data based on PELA 1994–2005. Note: The question read: “Party militants frequently complain of not being able to participate in party decisionmaking processes. How would you rate the degree of internal democracy in your own party: very high, high, medium, low or very low?”

question is not framed in mutually exclusive terms, but rather asks legislators to assign a value to different groups in terms of their degree of importance (see Table 5.6). Taking into account the answers of all parties, the order of perceived importance (from most to least important) is the following: district voters, party leaders, members, party voters, and party legislators. These answers show that parties and their internal opinions are not central criteria in the behaviors of legislators in parliament. Despite the fact that this chapter draws a distinction between participa-

A-5

10/2/07

3:36 PM

124

Page 124

PARTY POLITICS AND IDEOLOGY

Table 5.5

Preferred Degree of Power for Party Leadership (percentage) Party Degree of Leadership More Power over Legislators

Present Situation

PJ (12.2) UCR (44.4) MNR (11.8) NFR (8.3) MAS (15.4)

PJ (40.8) UCR (33.3) MNR (5.9) NFR (20.0) MAS (15.4)

Chile

PDC (23.5) PPD (6.7) PS (22.2)

Costa Rica

PUSC (11.8) PLN (7.1)

PDC (58.8) RN (81.3) UDI (80.8) PPD (33.3) PS (44.4) PLN (21.4)

Ecuador

PSC (25.0) PRE (42.9) ID (14.3) MUPP-NP (27.3) PRIAN (50.0) ARENA (19.2) FMLN (13.8) PCN (33.3) PAN (15.4) FRG (9.1) PU (14.3) PL (16.3) PN (16.7) PLC (15.6) FSLN (28.0) PRD (7.3) PA (5.0) ANR (12.0) PLRA (46.2) MPQ (28.6)

Party System Argentina Bolivia

El Salvador Guatemala Honduras Nicaragua Panama Paraguay

Dominican Republic

PRD (21.6) PLD (10.0) PRSC (3.8)

PSC (50.0) PRE (28.6) ID (35.7) MUPP-NP (9.1) PRIAN (20.0) ARENA (3.8) FMLN (34.5) PCN (6.7) PAN (7.7) FRG (22.7) PU (28.6) PL (2.3) PN (6.3) PLC (18.8) FSLN (28.0) PRD (16.7) PA (5.0) ANR (84.0) PLRA (46.2) UNACE (100.0) MPQ (71.4) PRD (13.7) PLD (15.0) PRSC (34.6)

Less Power PJ (40.8) UCR (11.1) MNR (82.4) MIR (100.0) NFR (83.3) MAS (69.2) PDC (17.6) RN (18.8) UDI (19.2) PPD (60.0) PS (33.3) PUSC (82.4) PLN (64.3) PAC (100.0) PSC (25.0) PRE (28.6) ID (50.0) MUPP-NP (63.6) PRIAN (30.0) ARENA (76.9) FMLN (51.7) PCN (60.0) PAN (76.9) FRG (63.6) PU (42.9) PL (79.1) PN (77.1) PLC (65.6) FSLN (44.0) PRD (70.0) PA (90.0) ANR (4.0) PLRA (7.7) PRD (64.7) PLD (75.0) PRSC (61.5)

Source: Data based on PELA 1994–2005. Note: The question read: “I would like to get your opinion on party leadership. In your opinion, should party leaders in your country have more power over legislators or should they, on the contrary, have less power over you?”

tion and distribution of responsibilities in the party, there is a relationship between the two: the more perceived participation there is in a political party, the higher the levels of internal democracy. If perceived participation is taken as a fairly accurate indicator of actual participation, it could be con-

A-5

10/2/07

3:36 PM

Page 125

THE ORGANIZATION OF POLITICAL PARTIES Table 5.6

Groups Perceived as Very Important in Making Decisions (percentage) Attention to Party Members

Party PJ UCR MNR MIR NFR MAS PDC RN UDI PPD PS PSC PRE ID MUPP-NP PRIAN PUSC PLN PAC PANguat PU PL PN FSLN PRDpa PA ARENA FMLN ANR PLRA UNACE MPQ PLC PRD PLD PRSC

125

Party Leaders Party Legislators 11.8 11.8 28.6 18.8 12.5 17.6 27.8 13.3 26.9 6.7 0.0 77.3 83.3 71.4 33.3 62.5 17.6 14.3 35.7 38.5 28.6 37.8 35.4 48.0 46.7 20.0 57.7 41.4 38.5 42.9 85.7 57.1 38.7 21.1 31.3 20.7

3.9 5.6 0.0 12.5 6.3 17.6 0.0 20.0 7.7 0.0 11.1 45.5 58.3 21.4 44.4 25.0 23.5 28.6 28.6 23.1 28.6 11.6 27.1 25.0 20.0 5.0 30.8 24.1 23.1 31.3 14.3 57.1 16.7 16.1 6.5 7.1

Attention to Voters

Members

District Voters

Party Voters

15.7 22.2 9.5 6.3 6.3 23.5 0.0 6.7 15.4 0.0 0.0 59.1 58.3 57.1 44.4 50.0 35.3 21.4 28.6 38.5 28.6 36.4 37.5 44.0 40.0 10.0 50.0 48.3 61.5 50.0 71.4 57.1 34.5 28.1 12.9 17.2

78.4 61.1 89.5 87.5 50.0 93.8 77.8 62.5 69.2 66.7 77.8 90.9 84.6 64.3 55.6 62.5 70.6 57.1 57.1 53.8 50.0 65.1 72.9 56.0 80.0 75.0 61.5 82.8 92.3 46.7 85.7 57.1 58.6 71.9 48.4 96.4

17.6 16.7 23.8 18.8 20.0 35.3 0.0 6.7 19.2 6.7 0.0 72.7 58.3 42.9 44.4 50.0 41.2 21.4 38.5 61.5 14.3 38.6 41.7 52.0 40.0 15.0 38.5 37.9 48.0 50.0 71.4 37.5 46.7 45.6 42.9 51.7

Source: Data based on PELA 1994–2005.

cluded that high levels of active participation and mobilization in a party are usually accompanied by a high degree of internal democracy. Regarding the lack of consensus on the meaning of “internal democracy,” we cannot overlook the fact that of all the questions on intraparty life analyzed in this chapter, this one is the hardest to apply. Ideology is one variable that can interfere in different interpretations of the concept of democracy; Table 5.7 shows that these two variables are correlated. Parties to the right tend to describe their organizations as less democratic than do

A-5 10/2/07

Table 5.7

Pearson Correlation Indexes

Participation Perceived as Intense

Perceived Militancy

Average of ideological self-placement and ideological placement of party n

1 38

Placement of the political party n

.993(**) 38

1 38

Participation perceived as intense n

–.280 38

–.279

1 38

Perceived militancy n

.130 38

.152 38

.585(**) 38

1 38

Perceived internal democracy n

–.318 38

–.345(*) 38

.587(**) 38

.317 38

Perceived Internal Democracy

Ideological Programmatic Organizational Total Coherence Coherence Coherence Coherence

Index of Party Life

1 38 (continues)

Page 126

Ideological Placement of Political Party

In Favor of Less Power for Party Leadership

3:36 PM

Average of Ideological SelfPlacement and Ideological Placement of Party

A-5 10/2/07

Table 5.7

continued

3:36 PM

Ideological Placement of Political Party

Participation Perceived as Intense

Perceived Militancy

Perceived Internal Democracy

In Favor of Less Power for Party Leadership

.030 38

.065 38

.169 38

.202 38

–.026 38

1 38

.515(*) 19

.508(*) 19

–.0.46 19

.471(*) 19

–.201 19

.194 19

1 19

Programmatic coherence n

.407 19

.395 19

–.229 19

–.188 19

–.005 19

.168 19

.062 19

1 19

Organizational coherence n

.250 19

.222 19

–.067 19

.579(**) 19

.137 19

.012 19

.173 19

–.045 19

1 19

Total coherence n

.662(**) 19

.639(**) 19

–.187 19

.458(*) 19

–.087 19

.217 19

.770(**) 19

.525(*) 19

.521(*) 19

1 19

–.207 38

–.209 38

.653(**) 38

.734(**) 38

.799(**) 38

.014 38

.064 19

–.209 19

.386 19

.093 19

In favor of less power for party leadership n Ideological coherence n

Index of party life

Notes: * Significant at the .05 level (two-tailed). ** Significant at the .01 level (two-tailed).

Ideological Programmatic Organizational Total Coherence Coherence Coherence Coherence

Index of Party Life

1 38

Page 127

Average of Ideological SelfPlacement and Ideological Placement of Party

A-5

10/2/07

3:36 PM

128

Page 128

PARTY POLITICS AND IDEOLOGY

parties to the left. For example, of all the parties analyzed in this study, the farthest right are the PSC of Ecuador and the Salvadoran ARENA (8.09 and 8.05, respectively). And among their party systems, these two parties paint the most pessimistic pictures of levels of internal democracy. The opposite occurs with the two parties that lean most heavily to the left, the MAS and the FSLN, with 2.14 and 1.65, respectively, in ideological placement. These two parties have very high levels of internal democracy, as does the FMLN. This finding endorses the research hypothesis that a party’s having revolutionary origins has an influence on the kind of party it later becomes. In this context, this influence can be seen in the level of internal debate on party ideology, program, and organization.33 Nevertheless, there is a possibility that left and right parties have different interpretations of the concept of internal democracy. This could lead to potential distortions in the answers that, for some scholars, would invalidate the comparison among parties. It could be hypothesized that left parties conceive of internal democracy as the involvement in the deliberation and decisionmaking processes of parties; whereas in right parties this concept is associated not so much with participation in debates on ideology and party programs, but rather with officially being able to vote for candidates or for people to occupy positions in the party structure. Even so, in order to prove this argument, it would be necessary to contrast perception with objective information on practices within political parties. Keeping in mind that parliamentary elites are the primary beneficiaries of the lack of internal democracy, it may be speculated that citizen assessment of democracy in political parties would be even more pessimistic than that of legislators. Based on legislators’ evaluations of their own parties in terms of levels of militancy, participation, and degree of internal democracy, we have calculated an index of what could be called “perceived intraparty life” (see Table 5.8). This is one possible way of looking at such organizational aspects as levels of participation and types of involvement in decisionmaking. Other ways of operationalizing this concept would be to analyze party statutes to determine the composition of decisionmaking bodies and to assess continuity and change in party leaders and electoral candidates. The advantage of this index is that it allows the examination of information on perceptions in an aggregate manner. The scale ranges from 0 to 1, with values toward the higher end of the scale indicating that legislators perceive the party as more participatory and democratic in its internal functioning.34 The highest ranked in perceived party life are two parties denominated “fronts” (FMLN and FSLN). Their legislators consider these parties democratic in their decisionmaking and participatory in terms of allowing involvement by different groups. The Chilean PPD, on the other hand, is perceived as having low organizational pluralism. This party is followed by the Costa Rican PUSC and the Panamanian PA, whose legislators consider

A-5

10/2/07

3:36 PM

Page 129

THE ORGANIZATION OF POLITICAL PARTIES Table 5.8

129

Perceived Party Life by Legislators Perceived Pluralism by Party

Low PPD PUSC PA PU PS UCR RN MNR PJ

Medium 0.457 0.488 0.515 0.553 0.563 0.592 0.593 0.604 0.612

FRG PLRA PDC PRDrd NFR PSC PRSC PANguat UDI PRIAN MIR PLD PLH PLN ANR PN PRE

High 0.620 0.644 0.646 0.651 0.669 0.671 0.686 0.688 0.702 0.704 0.725 0.729 0.756 0.771 0.771 0.772 0.776

ARENA ID MUPP-NP MPQ UNACE MAS PLC PCN PRDpa PAC FMLN FSLN

0.786 0.787 0.789 0.821 0.828 0.837 0.849 0.873 0.876 0.886 0.921 0.948

Source: Data based on PELA 1994–2005.

the democratic and participatory aspects of their parties to be limited. Similarly, information suggests a priori that parties ranked as caudillistas, like the PA, PJ, ANR, FRG, and PRE in Alcántara’s consideration (2004), do not have high levels of party life. This seems to confirm some arguments on political parties that have not yet been proven: it can be hypothesized that personalist leadership negatively affects levels of participation and involvement in the shaping of a party’s future.

■ The Relationship of Party Life to the Party’s Context The organization of parties is related to, and has an impact on, the context in which it is developed. This section explores two spheres of interaction between party life and the environment in which it is carried out: the relationship between levels of party life and types of party systems, and the relationship between party life and evaluation of political parties. Party Life and Party Systems

One of the most widely accepted classifications for the understanding of Latin American party systems is that suggested by Scott Mainwaring and Timothy Scully (1995). In this study, the authors propose the concept of “institutionalization of party systems” (understood as the degree of rou-

A-5

10/2/07

3:36 PM

130

Page 130

PARTY POLITICS AND IDEOLOGY

tinization of practices and interactions between parties) as the best conceptual tool for capturing variations between party systems. Organizational strength was one of the four dimensions employed to operationalize the concept of party institutionalization. The inclusion of this dimension was one of the most highly criticized aspects of this study. Indeed, in the revised version done by J. Mark Payne, Daniel Zovatto, Fernando Carillo, and Andrés Allamand (2002), this component is eliminated from the calculation of the index of institutionalization of Latin American party systems, mainly because of the lack of empirical evidence to capture the degree to which organizations are stable and strong within a specific party system. This is but one more proof of the difficulty of operationalizing the organizational element. For Mainwaring and Scully, “organizational strength” refers to situations in which political elites are loyal to their parties and where party discipline in legislatures is reasonably solid. In addition, in spite of their centralization, parties must be present not only at the national but also at the local level. The index of party life, or of perceived pluralism, is a possible improvement in the measurement of organizational aspects. It is meant to measure the degree of participation and democracy within a party, but captures neither the solidity nor the stability of an organization. Therefore, this index only maps some of the elements to consider when examining the organizational aspects of party institutionalization. The data on political parties used here can be approached in an aggregate form to gain an empirical approximation of party systems in terms of their levels of party life. Table 5.9 shows aggregate results by country. The Nicaraguan and Salvadoran party systems have the highest levels of participation and the most horizontal distribution of responsibilities between party members; Chile and Argentina, on the other hand, display the lowest levels of party life.35 The countries where levels of intraparty life are the most similar among all the parties of one system (see standard deviations in Table 5.9) are Honduras and the Dominican Republic; the party systems with the widest levels of variation in this regard are those of Panama and Costa Rica. We have not found statistically significant relationships between the institutionalization indexes and the index of perceived party life, something that may be partially explained by the low number of cases.36 However, there are also conceptual limitations in the index of perceived organizational pluralism. The dimensions of participation and distribution of responsibilities cannot be the only measurements of the organizational aspect of a party. It is, however, one of the issues that should be taken into account in indexes such as that of party institutionalization. Improving the institutionalization index by considering party life will allow us to test the hypothesis

A-5

10/2/07

3:36 PM

Page 131

THE ORGANIZATION OF POLITICAL PARTIES Table 5.9

131

Intraparty Life by Party System

Party System Argentina Bolivia Chile Costa Rica Ecuador El Salvador Guatemala Honduras Nicaragua Panama Paraguay Dominican Rep.

Average of Intraparty Life

Number of Parties Studied

Similarity Between Parties from the Same System (standard deviation)

.602 .708 .592 .701 .745 .860 .620 .764 .898 .695 .766 .688

2 4 5 3 5 3 3 2 2 2 4 3

.014 .098 .092 .227 .054 .068 .067 .011 .070 .255 .085 .039

Strength Institutionalization of Party of Organization Party System 2.00 2.03 1.00 1.74 3.00 2.46 3.00 2.38 1.00 1.43 Not calculated 2.11 Not calculated 1.32 Not calculated 2.65 Not calculated 2.02 Not calculated 1.94 3.00 2.21 Not calculated Not calculated

Source: Data based on PELA 1994–2005. The strength of party organization index stems from Mainwaring and Scully 1995. The institutionalization of the party system index stems from Payne et al. 2002:143.

that the level of institutionalization of party systems is the factor that most crucially affects differences in types of party organizations. We have tried to suggest that ideology interacts with the way in which parties are organized. Mainwaring (1999) points out that although there are organizational differences between center and conservative parties on one hand and left parties on the other, these differences are minimal compared to the differences between parties of more institutionalized systems and those of less institutionalized ones. This means that parties from the same system display fewer differences among themselves (on issues related to party life) than exist between parties from more institutionalized systems and parties from less institutionalized systems. For the moment, given the small variation in perception of party life between parties from the same system (see column 3, Table 5.9), there is no evidence to the contrary. Parties within the same political system are similar to each other in levels of party life, although to arrive at more exhaustive conclusions, parties should be studied at an individual level. The inclusion of organizational aspects that improve the party institutionalization index could be important in proving this hypothesis and comparing the relationship that party ideologies and the institutionalization of party systems have to the way in which parties are organized in Latin America.

A-5

10/2/07

3:36 PM

132

Page 132

PARTY POLITICS AND IDEOLOGY

Party Life and Evaluation of Political Parties

At the beginning of this chapter, we suggested that the organizational element is one of the most central issues for political parties. One idea that runs through the chapter is that issues related to party life have an important effect on citizens’ perceptions of parties. It may then be the case that party organization may be partially responsible for the present disrepute into which political parties and their members have fallen. One can measure this disrepute in many ways. Three questions solicit citizens’ answers on the necessity of political parties, their confidence in political parties, and their feelings of party identification. Table 5.10 shows the relationship of these questions—asked of a representative sample of citizens from each of the countries studied here37—to the index of party life we proposed previously. Although there is apparently no relationship between intensity of party life and the confidence in political parties or the impression that parties are necessary, it can be seen that the levels of participation and responsibility permitted do indeed affect feelings of identification with political parties. The more participation, militancy, and latitude for citizens to become involved in party decisionmaking there is, the more citizens tend to identify with that party. In spite of the fact that the measurement of the organizational element needs to be improved, the data show that there is a relationship between being in favor of party leadership having less power and the perception that political parties are necessary for the development of society. This means that those parties in countries whose citizens consider parties to be a key piece in the country’s development tend to be in favor of the party leadership’s having less power.

Table 5.10

Correlations Between Party Life and Opinion on Political Parties Parties Identification are with Trust Necessary Parties in Parties

Parties are necessary Pearson correlation n Identification with parties

Index of Party Life

1

Pearson correlation n

–.020 35

1

Trust in parties

Pearson correlation n

.450(**) 35

.277 35

1

Index of party life

Pearson correlation n

.026 35

.533(**) 35

.059 35

Source: Data from Payne et al. 2002 and PELA 1994–2005. Note: ** Significant at the .01 level (two-tailed).

1

A-5

10/2/07

3:36 PM

Page 133

THE ORGANIZATION OF POLITICAL PARTIES

133

■ Conclusion Party organization has a great impact on the nature of a party and its actions. In this chapter, we have defended the usefulness of studying party organization through the views of party members on this issue.38 The study of perceptions of the organizational aspects of political parties has focused on the intensity of participation and involvement in party structures. Up until now, the importance of party organization was, to a large extent, measured in terms of elections: a party strong in organization was a party successful in elections. However, despite the fact that the number of votes a party gets is indicative of strength, there are other spheres of action that do not have the same ability to mobilize as do elections, but which are also indicative of a party’s strength. For example, receiving a large number of votes is not automatically associated with a party’s ability to integrate different social groups within itself; this latter trait is more in line with the understanding of party life that informs this chapter. In this sense, the conceptualization of this chapter is similar to what is sometimes called the “vitality of party organizations” 39 and sometimes called “party strength.” This conceptual focus accords more with the present-day reality of Latin America, where parties are increasingly falling into disfavor and where citizens’ perception that parties do not defend their interests is one of the central elements of this feeling— thus, the importance of paying attention to the involvement of society in parties. Data suggest that organizational realities vary. Nevertheless, despite differences in members’ views on the functioning of parties, one may generalize that present-day party structures are far from being inclusive and democratic in their operation. These data on participation and involvement in Latin American political parties allow us to measure an element that has rarely been studied systematically. Perceptions on levels of participation, militancy, internal democracy, and the involvement of groups in party decisionmaking must be complemented by objective facts on levels of membership and the analysis of statutes and rules for party member participation in decisionmaking. Using the distinction proposed by Wolfang Muller and Kaare Strom (1999) to discuss the behavior of political parties, it could be said that Latin American political parties have tried to contribute to democratic functioning through interparty competition rather than through intraparty participation. The latter requires not that parties limit themselves to competing with other parties, but that they seek to organize themselves democratically and not simply as a team of leaders. Parties organized according to this model give citizens opportunities for political participation, both in the election of party leaders and in the drawing up of party programs. Representative democracy consists of both types of democratic behavior— interparty and intraparty.

A-5

10/2/07

3:36 PM

134

Page 134

PARTY POLITICS AND IDEOLOGY

On another subject, the characteristics of political parties allow us to make some predictions on the types of party systems in place. Thus, systemic considerations can be incorporated into conclusions on party life. In particular, the data presented in this chapter allow us to operationalize the element of a party’s organizational strength, an aspect of political systems for which there are presently no valid indicators. Indeed, in Mainwaring and Scully’s study (1995), this aspect constitutes the fourth dimension of the institutionalization of a party system and alludes to the idea that “organization matters.” In that book, the organizational dimension was studied using only qualitative and general considerations, which point up the lack of studies and systematic information. The later, revised measurement of this aspect excludes this fourth dimension, since there was not enough systematic information to develop a comparable measure. At the present moment, the impact of party organization on the legitimacy of a party does not refer to the existence (or not) of a formal organization established by statutes, which meets regularly, has leaders who make decisions, and is present throughout the national territory. This is not the case in Latin America, where formally organized parties exist. On the contrary, what our data suggest is that, rather than paying attention to formal structures, we should examine the vitality of these structures. The evidence shows that participation and militancy are not excessively low, but that internal democracy has not been achieved to an equal degree in all parties and that, at the same time, the political class itself favors decreasing the power of party leadership. The source of our data is the legislative elites, but the hypothesis can be made that these conclusions would be even more emphatic were the citizens to be polled. Given the low levels of legitimacy suffered by political parties, given the fact that they inspire little confidence, and given as well the generalized perception that parties are out of touch with their voters, it seems of vital importance to rethink the issues of participation and involvement in parties. If the debate centers around the need to reorganize parties before their disrepute affects democratic performance, rather than discussing redesigning formal party structures, attention must be paid to the life that takes place within these structures. Parties must become more inclusive and participatory, through incentives both for the elite to include the citizenry in decisionmaking and for citizens to involve themselves in parties. This noninclusiveness of the crises in party representation is what limits the effectiveness of political parties.

■ Notes 1. In the case of Europe, Bartolini refers to this phenomenon as the myth of the “golden age.” According to the author, this idea is defective not only because of the strong normative emphasis in the form of nostalgia for what a party should be,

A-5

10/2/07

3:36 PM

Page 135

THE ORGANIZATION OF POLITICAL PARTIES

135

but also because it is probably the result of a mythical vision and an idealization of historical reality (Bartolini, 1988:253). 2. Some of the factors that lead to this diagnosis are the mistrust of party leaders, the shared perception of the incompetence of parties, the growing mass of citizens that do not identify with a specific political party, and levels of volatility of party systems. On these issues, see the report of the Programa de Naciones Unidas para el Desarrollo (PNUD) (2004:71 and following). 3. For this analysis we have used data from the Project of Latin American Parliamentary Elites (Proyecto de Elites Parlamentarias de América Latina, or PELA), which brings together information gathered on thirty-eight Latin American political parties from twelve party systems, covering all Central American countries, the Dominican Republic, and the Southern Cone (Argentina, Bolivia, Chile, Ecuador, and Paraguay). This chapter uses a maximum of five parties per system and analyzes those who were the strongest in parliament at the time of the interviews. In the PELA, the number of interviews done per party is proportional to each party’s parliamentary presence at the time the interviews were done. In this chapter, however, only those parties with a minimum of six interviews were taken into account. 4. The PELA questionnaire does not include a great number of organizational questions since it is more oriented toward the program, ideology, and biography of legislators. 5. Cavarozzi (1991); Di Tella (1993). 6. Norden (1998); Alcántara (1994). 7. McDonald and Ruhl (1989); Mainwaring and Scully (1995). 8. Alcántara (2002); Freidenberg and Sánchez López (2001). 9. Levitsky (2003). 10. Carey (2002). 11. In all geographic areas, questions of party organization have received very little attention, as pointed out by Janda (1993), who calls attention to the imbalance between the number of studies on organizations in general and the number of studies on the organization of political parties. 12. Examples of the kind of conceptual work necessary for a reflection on organizational issues in Latin America are those of Levitsky (1998, 2003). This author studies the development and application that the concept of party institutionalization has had in recent times. 13. With respect to this question, Montero and Gunther (2002) point out that there is still a long way to go in order to get beyond simplistic classifications of party organization. They indicate the need to develop empirical indicators on internal party life, on the relationship between organizational change and electoral volatility, and of coming up with hypotheses that explain the diversity of change in party organizations. 14. Wolinetz (2002). 15. Ibid. 16. Alcántara (2004a). 17. Trying out other classifications causes problems that are familiar in the European case, such as the existence of different criteria of classification that are not coherent among themselves. 18. In this sense, there is an excellent and systematic study done by Marenghi (2001), which covers a large number of Lain American parties. 19. Diamond, Hartlyn, Linz, and Lipset (1999:25). 20. Alcántara (2004a).

A-5

10/2/07

3:36 PM

136

Page 136

PARTY POLITICS AND IDEOLOGY

21. Other groups whose perceptions are relevant include the party leadership without a legislative role, the middle-rank leaders, and groups with a lesser level of involvement in party activities (among others, members or sympathizers). 22. In Bartolini (1998). 23. The gathering of data on this aspect requires extensive fieldwork or an intense investment of resources that are not always available. 24. Such as that of Alcántara and Freidenberg (2001). 25. Janda (1993) states that the dimensions of analysis relative to the organization of parties are complexity, centralization, involvement of party members, factionalism, type of activities, and degree of cohesion. 26. For example, it is usual to affirm that traditional left parties tend to be more hierarchical in their organization than those of the right. 27. Mainwaring and Scully (1995); Payne, Zovatto, Carillo, and Allamand (2002). 28. There are, however, some dissenting voices from this stereotype. An example would be the work of Angell (2003), whose examination of the present party system in Chile includes a mention of the decrease in active involvement in political parties. 29. Alcántara (2004b:26). 30. Ruiz Rodríguez (2006). 31. Party coherence has been operationalized attending to parliamentary positions on a set of questions (ideological, programmatic, organizational). It is measured through standard deviations in their responses. The more dissension found among members of the same party, the less party coherence exists, and vice versa. For a detailed account of the party coherence index, see ibid. 32. In the questionnaire, “party leadership” is understood to mean those members who occupy positions of responsibility within the party and on occasion, although not always, within the political system as well (legislators are one example; others would be ministers, governors, mayors, and senators). 33. Alcántara (2004b). 34. Steps followed in the elaboration of this index: (1) creation of a new variable through the addition of the averages of each party in the answers to the following three variables: level of militancy, level of participation, degree of internal democracy; and (2) standardization of a new variable, “perceived organizational pluralism,” in order to achieve the scale from 0 to 1 using the formula for the transformation of variables to a 0 to 1 scale. Formula for the transformation of variables to a 0 to 1 scale: Value of new variable =

Value of old variable – Minimum value of old variable Maximum value of old variable – Minimum value of old variable

35. It is paradoxical that although the countries of Central America have only recently adopted a democratic system, they rank the highest in this classification. This, and the fact that Chile comes in last, fly in the face of commonly held beliefs. It must be remembered, however, that this classification is based on the opinions of the legislators from the parties analyzed. 36. Note that in some party systems the number of parties studied was too small to be representative of the system as a whole (the case of Argentina is perhaps the most problematic in this sense). 37. Data excerpted from Payne et al. (2002). 38. In the Latin American case, in general, little attention has been paid to examining the position of party members on organizational issues. An exception to

A-5

10/2/07

3:36 PM

Page 137

THE ORGANIZATION OF POLITICAL PARTIES

137

this trend is the data included in Alcántara and Freidenberg (2001). One of the most interesting aspects of this study is that it gathers the perceptions of members of different ranks on who the principal sources of power in their parties are. 39. Alcántara (2004b).

A-5

10/2/07

3:36 PM

Page 138

A-6

10/2/07

3:36 PM

Page 139

6 The Dimensions of Polarization in Parliaments Cristina Rivas Pérez

OVER THE LAST FOUR DECADES, A LARGE NUMBER OF STUDIES HAVE

been conducted on the dimensions of party competition in European democracies. This type of analysis has made it possible to assess the persistence and change of social and political cleavages.1 Democratization and democratic consolidation in Latin America have given rise to a growing number of studies on party systems in the region. Most often these studies have focused on three basic characteristics of the party system: fragmentation, electoral volatility, and the effective number of parties. By contrast, there have been few studies on other topics critical to party competition, such as ideology, orientations and values, and cleavages, or political divides.2 There have even been debates on the appropriateness of the concept of cleavages in the analyses of the region’s politics. However, if cleavages are understood as relatively stable patterns of polarization, by virtue of which certain groups endorse specific policies or parties,3 then their applicability to Latin American politics is entirely justified. “Cleavage” in this chapter refers to the dimensions of programmatic polarization that separate—or bring together—Latin American political parties labeled as either rightist or leftist, whereas “polarization” is understood as the distance between left and right regarding ideological preferences on issues of political interest. This chapter examines which cleavages condition the divisions and polarizations of political parties in Latin America. It also assesses whether the topics that traditionally have distinguished “left” and “right” are still important in Latin American politics. This chapter aims also at establishing what social, economic, religious, and political divides constitute the leftright dimension in Latin America.4 This empirical analysis is based on the assumption that the left-right dimension simplifies the political world and

139

A-6

10/2/07

3:36 PM

140

Page 140

PARTY POLITICS AND IDEOLOGY

reflects a number of underlying dimensions.5 With this purpose in mind, this chapter analyzes parties located in extreme positions in the left-right dimension, according to the PELA data (see Table 6.12 in the chapter appendix). 6 The selection of these cases facilitates the identification of those elements that affect ideological and programmatic polarization. The first part of this chapter evaluates the usefulness of the left-right dimension in classifying and distinguishing political parties in Latin America. The second part explores the main dimensions of variability among parties, as well as the aspects that allow us to distinguish between left and right parties. Finally, based on the analysis of the main dimensions of programmatic polarization, this chapter examines party families in Latin America. Methodologically, the chapter uses factor analysis and a biplot representation of the resulting average factors and of the selected political parties. Constructing factors for the whole region allows us to compare the dimensions of ideological and programmatic polarization across nations. This analysis is structured around five factors: state intervention in public policies, religion and moral values, social background, democracy, and the armed forces. This study hopes to make a substantive contribution to the study of left and right parties in Latin American parliaments, which in the future should be complemented by studies that include center parties. Such works will allow us to test whether the findings of this chapter hold when the whole spectrum of ideological positions is taken into consideration.

■ Left and Right Parties in Latin American Parliaments The vocabulary of “right” and “left” is standard in many countries. The use of the left-right dimension allows citizens to place parties, programs, and decisions in a coherent and clearly understandable framework.7 It has been frequently pointed out that the left-right dimension lacks analytical value in the study of Latin American party systems.8 However, it must be taken into account that definitions of “left” and “right” can vary regionally, cross-nationally, and across time.9 For these reasons, a comparative analysis of the left-right dimension must not focus on fixed ideological criteria, but on spatial considerations. Differences in regional and national historical trajectories, rooted in specific social and political experiences,10 limit the general applicability of understandings of the left-right dimension. Some studies refer to two basic components in the definition of the leftright dimension: a symbolic component and a substantive one; the latter is connected to specific ideological preferences regarding political, religious, economic, and social issues.11 This chapter focuses on the presence of these programmatic dimensions in Latin American parliaments and, more specifi-

A-6

10/2/07

3:36 PM

Page 141

THE DIMENSIONS OF POLARIZATION IN PARLIAMENTS

141

cally, aims at identifying the programmatic positions of political parties. This analysis will allow us to label political parties and to establish the main party families in Latin America.12 It is based on the legislators’ individual responses regarding their ideological self-placements and the ideological placements of their respective parties.13 Almost all Latin American legislators (98 percent) were able to place themselves in the left-right dimension. This fact endorses the use of this dimension as a criterion for the classification of parties. Table 6.1 shows the classification of thirty Latin American parties in the left-right dimension, with fourteen parties leaning to the right and sixteen parties to the left.14 There are few doubts about the placement of parties in the left-right dimension. Perhaps the most debatable case is that of the Panamanian PA, which, in spite of being defined by its members as rightist, could also be considered as a center party. Overall, the classification in Table 6.1 provides a picture which is very similar to those of other studies.15

■ The Main Dimensions of Polarization in Latin American Parliaments The left-right dimension has been traditionally linked to class-based economic conflicts, but other issues also contribute to party competition (religious and moral values, authoritarianism-democracy, neoliberalism-statism, nationalism-regionalism, etc.).

Table 6.1

Classification of Latin American Parties by Ideology Parties on the Right

Chile Colombia Dominican Republic Ecuador El Salvador Guatemala Honduras

UDI PC PRSC PSC, PRIAN ARENA, PCN GANA, PSN PNH

Nicaragua Panama

PLC PA, PS

Venezuela

COPEI

Parties on the Left Argentina Bolivia Chile

ARI MAS PPD, PS

Ecuador El Salvador

PRE, ID, MUPP-NP, PSP FMLN

Honduras Mexico Nicaragua

PUD PRD FSLN

Peru Uruguay Venezuela

PAP EP/FA MAS, MVR

Source: Based on PELA 1999–2005. Note: Only parties with at least five legislators were included in the analysis.

A-6

10/2/07

3:36 PM

142

Page 142

PARTY POLITICS AND IDEOLOGY

The five dimensions, or cleavages, considered in this study result from the application of an exploratory factor analysis with varimax rotation on twenty PELA variables.16 These variables are shown in Table 6.2.17 The goal of this principal components analysis is to identify the latent structures that will allow us to differentiate between right and left in Latin America on the basis of the opinions and preferences of their legislators. Table 6.3 shows the main results of this factor analysis (complete results are presented in Table 6.13 of the chapter appendix). Five factors explain 53.7 percent of the total variance. The first factor, connected to state intervention (the first eight variables in Table 6.2), accounts for 15.6 percent of the variance. The second factor, related to the evaluation of the military, accounts for 11.3 percent of the variance. The third factor is linked to religiosity and moral values; it accounts for 10.9 percent of variation. The fourth factor has to do with legislators’ social backgrounds, and accounts for 8.9 percent of the total variance. Finally, the fifth factor connects to attitudes toward democracy and democratic institutions, and accounts for 6.8 percent of variance. Once the five dimensions of polarization have been established, it is necessary to determine which dimensions give rise to higher levels of interparty polarization, as well as how party families are placed on these dimensions.

Table 6.2

Variables in the Analysis Variables

State intervention to subsidize housing State intervention to provide jobs State intervention to provide citizens with health care State intervention to guarantee a free and universal university education State intervention to provide unemployment benefits State intervention to provide unemployment insurance State intervention to protect the environment State intervention to cover basic citizen needs Strength of religious conviction and observance of religious practices Attitude toward divorce Attitude toward abortion Trust in the Catholic Church Evaluation of the role played by the military nowadays Trust in the armed forces Relations between the armed forces and the government threaten democratic stability Relations with the armed forces are a problem for democratic consolidation Education of the respondent’s father Education of the respondent’s mother Democracy is preferable to an authoritarian regime Role of political parties Elections as a mechanism of political representation Source: Based on PELA 1999–2005.

A-6

10/2/07

3:36 PM

Page 143

THE DIMENSIONS OF POLARIZATION IN PARLIAMENTS Table 6.3

143

Analysis of Rotated Factors

Variables State providing health care State providing unemployment benefits State covering basic social needs State guaranteeing university education State providing housing subsidies State providing jobs State protecting the environment Evaluation of the armed forces Trust in the armed forces Relationship between the armed forces and the government as a threat to democracy Armed forces as a threat to democratic consolidation Attitude toward abortion Strength of religious conviction Attitude toward divorce Trust in the Catholic Church Father’s educational level Mother’s educational level Role of political parties Role of elections Democracy vs. authoritarianism

Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 .721 .679 .669 .647 .638 .634 .627 .798 .728 –.723 –.672 –.755 .742 –.704 .640 .905 .901 .718 .639 –.603

Source: Data of factor analysis from PELA 1999–2005.

This factor analysis has been complemented by an HJ-Biplot analysis, an extension of Gabriel’s biplot techniques. 18 It is sometimes useful to graphically represent data stemming from multivariate analysis. This HJBiplot presents the average scores for each party based on factor analysis.19 These graphical representations are useful because they not only show the configuration of individuals and parties, but also show which variables account for such configurations. Although HJ-Biplot does not reproduce the elements of the original matrix, it has the advantage of presenting markers for political parties and for programmatic dimensions within one graphic representation.20 There are many possible biplots for a specific group of data: this chapter provides the representation that gives us the most interesting insights for the interpretation of these data. Parties are placed perpendicularly on the vectors representing the dimensions extracted from factor analysis, which allows us to identify those variables that help differentiate groups of parties. In order to assess how well this biplot representation fits, it is necessary to first consider the inertia (variability) absorption rate. A high variability absorption rate is a necessary condition for considering reliable our inter-

A-6

10/2/07

3:36 PM

144

Page 144

PARTY POLITICS AND IDEOLOGY

pretation of the position of political parties and of programmatic dimensions. However, it is not a sufficient condition.21 Table 6.4 presents the proportion of variability absorbed by the first three-factor axes.22 As this table shows, the first axis captures most of the information and absorbs 35.57 percent of the total variability. That is, the dimension or dimensions best represented in axis 1 are the cleavages that most strongly polarize left and right parties in the first factor. The second axis also provides us with important information, but less so than the first one (23.54 percent). The variability rate in the first factor plane, formed by axes 1 and 2, reaches 59.12 percent. The space formed by axes 1, 2, and 3 accounts for 78.58 percent of the variability. In the following pages, attention is focused on the first factor plane. Secondary attention is paid to the third axis, which, with an inertia absorption rate of 19.48 percent, clarifies the positions of some parties and the character of some of the programmatic dimensions considered here. Factor axes identify those programmatic dimensions that determine differences among political parties. The measure for the relationship between the biplot axes of representation and each of the observed variables is labeled “the relative contribution of the factor to the element.”23 This contribution allows us to establish what dimensions of programmatic division are most closely connected to each axis and, consequently, which variables account for the placement of parties in each of the factor axes. This scale ranges from 0 to 1,000. The closer the value is to 1,000, the higher the variability of this dimension in the study.24 Not all the variables included in the analysis are well represented in the first factor plane (axes 1 and 2). Table 6.5 shows that there are two dimensions that are characteristic of axis 1; namely, those that get a strong contribution from axis 1 and a low contribution from the other axes: state intervention, and religion and values. The democracy dimension attains the highest contribution from axis 2, whereas the third axis is determined by the armed forces dimension (with a value of 889 over 1,000). The social background factor shows less power to discriminate among political parties.25 A more detailed analysis would require consideration of why the contri-

Table 6.4

Axis 1 Axis 2 Axis 3

Proportion of Cumulative Variability Absorbed Variability Absorbed

Accumulated Variability Absorbed

35.57 23.54 19.48

35.57 59.12 78.58

Source: HJ-Biplot analysis with data from PELA 1999–2005.

A-6

10/2/07

3:36 PM

Page 145

THE DIMENSIONS OF POLARIZATION IN PARLIAMENTS Table 6.5

145

Relative Contributions of the Factors to the Programmatic Dimensions

Programmatic Dimensions State intervention Armed forces Religion and values Social background Democracy

Axis 1

Axis 2

Axis 3

588 52 540 396 203

125 52 205 276 519

56 889 11 0 17

Source: HJ-Biplot analysis with data from PELA 1999–2005. Note: Data in bold numbers reveal a high or very high contribution of some of the axes.

bution of the social background dimension is so low. A possible interpretation lies in the similarity of legislators’ positions on this dimension. Another possible interpretation is that this dimension is better represented in factor planes different from those we consider here. In that case, it would be necessary to examine the planes formed by the axes 1 and 4, 2 and 4, or 3 and 4. As for the political parties, the quality of their representation can be interpreted as the amount of information on a party captured by the axes. The addition of relative contributions of each factor to the element provides us with information on the quality of representation in the reduced dimensional space. The closer to 1,000 the value of a political party in the factor axes, the higher the quality of representation in the factor plane. Based on these considerations, it must be pointed out that the PNH, PRD, PRSC, and MAS in Venezuela are not well represented in the reduced dimensional space and, therefore, we cannot draw reliable conclusions about them. It is also necessary to be cautious regarding the Guatemalan GANA, since the contributions they get from these axes are low. The representation of the remaining political parties is good or acceptable in some of the factor planes analyzed (see Table 6.6). It is also important to take into account the relative contribution of each factor axis to each party in order to interpret the discriminating power of factor axes. The parties that get a strong contribution from axis 1 and a low contribution from the rest of the axes are mainly the PSN (867), PLC (779), UDI (641), and MUPPNP (682). This indicates that the positions of these parties in the state intervention, and religion and values dimensions are different from those of other parties. The PPD, the Chilean PS, the PSP, PRE, and PAP are the political forces with the highest contributions from axis 2 (986, 911, 900, 874, and 808, respectively). That is, they can be differentiated from the remaining parties principally in their democratic orientations. The highest contributions of axis 3 are for the PCN (663) and ARENA (501). In these cases, parties are mainly differentiated by their positions in the armed forces dimension.

A-6

10/2/07

3:36 PM

146

Page 146

PARTY POLITICS AND IDEOLOGY

Table 6.6

Relative Contributions of the Factors to Political Parties

Political Parties ARI MAS-B UDI PPD PS-C PC PSC PRE ID MUPP-NP PRIAN PSP GANA PSN PNH PUD PRD PLC FSLN PA PS-P PAP PRSC ARENA FMLN PCN EP/FA MVR MAS-V COPEI

Axis 1

Axis 2

Axis 3

330 472 641 16 3 269 113 87 410 682 219 1 49 867 83 417 183 779 237 466 311 7 9 358 536 218 486 255 37 404

605 134 187 986 911 178 1 874 233 65 196 900 0 1 12 374 106 0 12 9 4 808 2 132 219 99 502 168 90 3

0 374 6 4 48 357 322 0 232 6 16 16 209 21 141 110 16 42 439 71 7 79 89 501 40 663 6 498 64 488

Source: HJ-Biplot with data from PELA 1999–2005. Note: Data in bold numbers reveal a high or very high contribution of some of the axes.

Inspection of Figure 6.1 shows that, with a few exceptions, the Latin American left and right are clearly different in the dimensions of programmatic polarization extracted from factor analysis. Right parties are located to the left of the first factor plane and left parties are placed to the right of that plane. There are two types of left parties: a more revolutionary left, placed in the upper right quadrant, and a more moderate left, placed in the lower right quadrant. Since the ideological dimension is well represented in Figure 6.1, the following pages focus on establishing the main programmatic dimensions of variability among Latin American parties. For this purpose, it is necessary to take into consideration the length and angle of the vectors that represent the cleavages, as well as the perpendicular projection of the points, the

10/2/07

3:36 PM

Page 147

THE DIMENSIONS OF POLARIZATION IN PARLIAMENTS

147

Figure 6.1 Representation of the Polarization Dimensions, HJ-Biplot Axes 1 and 2 4 PSP

3

PUD 2

UDI

Religion and Values ARENA

1 Axis 2

A-6

PC Armed Forces PCN

PRIAN FMLN MVR MUPP-NP FSLN

COPEI

PA PS(P) PSN PLC PSC GANA PRSC PNH

0

MAS(V)

PAP

–1

Social Background

–2

MAS(B)

PRD ID

State Intervention

ARI EP/FA

Democracy –3 PPD –4 –5

–4

–3

–2

–1

PS(C)

0

1

PRE 2

3

4

5

Axis 1

Source: Created by the author based on PELA 1999–2005.

political parties, over the vectors that represent programmatic dimensions or cleavages. Thus, the longer the vector, the higher the variability the vector of this cleavage has in the analysis; and the smaller the angle formed by the vector and factor axis is, the higher its importance in the study will be. The perpendicular projection of parties onto the vectors allows us to establish what dimensions are the most useful for distinguishing among subsets of political parties. Figures 6.1 and 6.2 show that four programmatic dimensions polarize the thirty cases under study: religion and values, and state intervention, which characterize axis 1; democracy, which characterizes axis 2; and armed forces, which characterizes axis 3.26 The first factor plane (see Figure 6.1) distinguishes parties with a strong religious component; these parties are placed on the left side of the graph of parties espousing secularist views. It also distinguishes parties favorable to more state intervention (placed on the right side) from those adopting neoliberal economic views (placed in the upper left of the quadrant). The dimension related to democratic attitudes also allows us to distin-

10/2/07

3:36 PM

148

Page 148

PARTY POLITICS AND IDEOLOGY Figure 6.2 Representation of the Polarization Dimensions, HJ-Biplot Axes 1 and 3 4 Armed Forces 3

FSLN

ARENA

MVR

PCN

2

PC 1 Axis 3

A-6

0

PRSC PNH State Intervention PS(C) PA FMLN Democracy MAS(V) PLC UDI PRIAN PS(P) PRD Social Background PRE ARI EP/FA MUPP-NP PPD PSP Religion PSN and Values

–1

GANA PSC

ID

PUD

2

3

–2 PAP –3 COPEI –4 –5

–4

–3

MAS(B) –2

–1

0

1

4

5

Axis 1

Source: Created by the author based on PELA 1999–2005.

guish among political parties. This cleavage separates parties fully endorsing democratic institutions (in the lower half of Figure 6.1) from those harboring doubts about democratic institutions (in the upper half).27 In spite of its lower quality, the social background dimension allows us to distinguish those political parties in which a majority of its members are from families with a high cultural and educational level (in the lower half) from those whose families have a lower educational level (in the upper right-hand quadrant). The second factor plane, formed by axes 1 and 3, is the one that best captures party polarization in the armed forces dimension. Figure 6.2 shows that the vector representing the armed forces cleavage distinguishes between parties that positively evaluate the current role of the military (upper side of the figure) and those parties holding negative views of the military (lower side of the figure).28 Based on the previous analysis, particularly on the results of the HJBiplot representation, a classification of party families is proposed in the next section. This classification will take into consideration the distance between political forces: the higher the distances, the stronger the differ-

A-6

10/2/07

3:36 PM

Page 149

THE DIMENSIONS OF POLARIZATION IN PARLIAMENTS

149

ences between political parties will be. This is also true the other way around: the shorter the distances, the greater the similarities among parties will be. Classification of Parties in the Religion and Values Dimension

The importance of the religious cleavage in the political attitudes and behavior of citizens in both Europe and Latin America can hardly be disputed, particularly in those cases in which Christian-democratic parties have been present. The Catholic Church has played a very important role in Latin American history and society.29 In the political arena, the importance of the Church is felt, for instance, in the formation of conservative political forces or in the processes of democratic transition. Even if religion has lost some of its political importance, it continues to affect differences between left and right parties, as was previously shown. Its importance does not strictly relate to religious faith, but is connected to moral issues that have been particularly divisive in Latin America, such as abortion and divorce. Four variables are connected to this cleavage: strength of religious conviction (which takes into account religious feelings, beliefs, and practice); attitudes toward abortion; attitudes toward divorce; and trust in the Catholic Church. The conservative pole would be characterized by its strong religious conviction, its opposition to abortion and divorce, and its trust in the Catholic Church. By contrast, the progressive and liberal pole would be characterized by its weaker religious conviction, its endorsement of the right to abortion and divorce, and its mistrust of the Catholic Church. UDI, ARENA, COPEI, PSN, and PC are the most conservative and religious parties, followed by PLC, PCN, and PA. The Ecuadorian left parties PSP and PAP also hold predominantly religious values. At the opposite pole are the progressive liberal parties, such as PPD; the Chilean PS, PRE, EP/FA, ARI; the Bolivian MAS, PUD, and MUPP-NP; as well as ID, MVR, FMLN, FSLN, and the right party PRIAN (see Table 6.7).

Table 6.7

Classification of Political Parties in the Religion and Values Dimension Political Parties

Religious/Conservative Secularist/Progressive

UDI, ARENA, COPEI, PAP, PSP, PSN, PLC, PA, PC, PS-P PPD, PS-C, PRE, ARI, EP/FA, ID, MUPP-NP, MVR, PUD, FMLN, PRIAN, FSLN, MAS-B

Source: Based on the results of HJ-Biplot. Note: Parties for which the quality of representation is low are not included.

A-6

10/2/07

3:36 PM

150

Page 150

PARTY POLITICS AND IDEOLOGY

There is a small group of parties with an intermediate position in this cleavage. They adopt moderate positions either in the conservative area (like the Panamanian PS, PSC, and PCN) or on the progressive side, like the Venezuelan MAS and the Mexican PRD. We must, however, be careful regarding the positions of these parties, since the quality of their representation in Figure 6.1 is low and, therefore, not reliable. Overall, these data reveal that, in spite of the process of secularization, the religious cleavage continues to structure Latin American party systems, although in an attenuated manner. Classification of Parties in the State Intervention Dimension

The structural economic transformations consisting of neoliberal and promarket policies have become another object of competition for Latin American parties. The different positions held by political parties regarding the state’s role in the economy have made this cleavage central to the region’s politics since the 1980s. This is so despite widespread consensus on the implementation of promarket and orthodox fiscal policies. 30 Neoliberal and promarket policies have been traditionally linked to right parties, whereas statist positions have been espoused by left parties. The variables constituting the state-market dimension concern such issues as health care, housing, university education, jobs, unemployment benefits, the environment, and the basic needs of the population. As previously shown, social variables (e.g., health care, unemployment benefits, and providing for basic needs) make the strongest contribution to this dimension. The clearly statist parties are the left-oriented EP/FA, PPD, the Chilean PS, PRE, ARI, PUD, the Bolivian MAS, MUPP-NP, FMLN, MVR, and ID; and also the right party PRIAN, which has shown positions different from those of other right parties in all dimensions of programmatic polarization. 31 Right parties are neoliberal. This is the case of UDI, COPEI, ARENA, and the Guatemalan, Colombian, Panamanian, and Nicaraguan right. Again, PAP and PSP are closer to right, neoliberal than to left, statist positions (see Table 6.8).

Table 6.8

Classification of Political Parties in the State Intervention Dimension Political Parties

In favor of state intervention Against state intervention

EP/FA, PPD, PS-C, PRE, ARI, MAS-B, PUD, MUPP-NP, FMLN, PRIAN, MVR, ID, FSLN UDI, ARENA, COPEI, PC, PSP, PSN, PLC, PA, PAP

Source: Based on HJ-Biplot analysis. Note: Parties for which the quality of representation is low are not included.

A-6

10/2/07

3:36 PM

Page 151

THE DIMENSIONS OF POLARIZATION IN PARLIAMENTS

151

A group of right parties adopt moderate neoliberal positions; they are the PCN, the Panamanian PS, PNH, and PSC. A group of moderate statist parties also exists—the Venezuelan MAS and the Mexican PRD. However, the low quality of representation for this group of parties prevents us from drawing definitive conclusions on their positions in this dimension. Classification of Parties in the Democracy Dimension

The authoritarian experiences of many Latin American countries in the 1970s and 1980s left their imprint on party systems, as well, and gave rise to an authoritarianism-democracy cleavage. Democratic transition processes, however, gave rise to widespread positive evaluations of democracy.32 The variables included in the democracy cleavage relate to the political culture of the parliamentary elite. More specifically, they consist of support for democracy and for two key institutions of the democratic regime: political parties and elections. Legislators are most likely to support the democratic regime, but there may be significant differences on views of elections and political parties. When interpreting these data, it must be kept in mind that in some Latin American countries electoral processes were not competitive, and that some political parties had authoritarian connections and inclinations.33 Figure 6.1 allows us to classify political parties by two main poles, one “democratic,” which includes parties unconditionally endorsing democracy, political parties, and elections, and one “authoritarian,” which includes parties expressing doubts about democracy as the best possible regime, as well as doubts about political parties and elections. The group of democratic parties is mainly formed by the left parties PPD and PS in Chile, the PAP in Peru, PRE and ID in Ecuador, EP/FA in Uruguay, and ARI in Argentina; and by the right parties COPEI, PSN, and PLC (see Table 6.9). The “authoritarian” pole includes the Colombian PC, PSP, PUD, FMLN, the Bolivian MAS and PRIAN, as well as MUPP-NP, MVR, FSLN, and, to a lesser extent, ARENA. A third group of parties is at

Table 6.9

Classification of Political Parties in the Democracy Dimension Political Parties

Democracy Authoritarianism

PPD, PS-C, PAP, PRE, EP/FA, ARI, ID, PSN, PLC, COPEI, PA PC, PSP, PUD, MAS-B, FMLN, PRIAN, MVR, MUPP-NP

Source: Based on HJ-Biplot analysis. Note: Parties for which the quality of representation is low are not included.

A-6

10/2/07

3:36 PM

152

Page 152

PARTY POLITICS AND IDEOLOGY

the center of this dimension. It includes UDI, PA, PSC, and PNH. Overall, this dimension divides the Latin American left into two main groups: a moderate, democratic left and a revolutionary-authoritarian left. It also reveals that the specific position of the Peruvian PAP—close to the right parties both in the religion and moral values and in the state intervention dimensions—is different from that of other conservative parties, due to its evaluations of democracy. Classification of Parties in the Social Background Dimension

The process of political socialization and its influence on the behavior of politicians have been addressed often in the study of political elites. Most of the literature underlines the importance of two elements in political socialization processes: the time and the agents of socialization. Different authors pay special attention to family as one of the main agents in this process.34 Closely connected with time and agents is the social context of political socialization (the family’s socioeconomic status and educational level). Family and social contexts exert specific influences on political views and party identifications.35 The goal here is to establish whether the social context of legislators and, in particular, the educational levels of their fathers allow us to distinguish among political parties. Even if this is not one of the dimensions showing strong variability (see Figure 6.1), it also shows significant differences between left and right parties. Table 6.10 demonstrates that, whereas right parties are characterized by high educational levels, left political groups (with the exception of the Chilean and Uruguayan ones) are characterized by lower educational levels. This classification overlaps with our previous classification in the democracy dimension. Parties whose members come from more highly educated families are more supportive of democracy, whereas those with members from lower educational levels are more critical of democratic institutions. Therefore, it can be pointed out that there is an overlap between these two dimensions of polarization.

Table 6.10

Classification of Political Parties in the Social Background Dimension Political Parties

High educational level Low educational level

PPD, PS-C, PAP, PRE, EP/FA, ARI, ID, PSN, PLC, COPEI, UDI, PA, ARENA, PC PSP, PUD, MAS-B, FMLN, PRIAN, MVR, MUPP-NP, FSLN

Source: Based on HJ-Biplot analysis. Note: Parties with a low quality of representation are not included.

A-6

10/2/07

3:36 PM

Page 153

THE DIMENSIONS OF POLARIZATION IN PARLIAMENTS

153

Classification of Parties in the Armed Forces Dimension

This dimension is directly connected to the role played by the military in recent authoritarian regimes; it is also related to the quite different experiences of political parties during episodes of authoritarianism. Three variables constitute this dimension: evaluation of the role played nowadays by the military, trust in the armed forces, and view of whether the relationships between the military and the national government pose any risk to democratic consolidation. Table 6.11 shows that the promilitary pole includes ARENA, PCN, FSLN, MVR, and the Colombian PC. The PA, UDI, and PLC also display promilitary orientations. The FMLN also holds positive views of the military. Within the antimilitary pole, there is one subgroup of parties holding strongly antimilitary views, such as the Bolivian MAS, COPEI, PAP, and PUD, and perhaps also PSC and ID in Ecuador; and a second subgroup that displays a more moderately negative evaluation of the military, which is formed by EP/FA and MUPP-NP. The classification of left and right parties in this dimension is very similar to the one observed in the democracy dimension. Therefore, these two cleavages overlap: the promilitary parties are the same as those positioned in the authoritarian pole, and vice versa. This shows that issues related to the authoritarian past and the role of the military are connected to democratic evaluations, and that these two dimensions continue to divide Latin American political parties.

■ Conclusion This chapter has tried to show the analytical usefulness of the cleavage approach to the study of left-right politics in Latin American legislatures. It also has sought to demonstrate that left-right divisions have a multidimensional character. Instead of adopting a conception of cleavages centered around social conflicts, along the lines of Seymour Martin Lipset and Stein Rokkan (1967), this chapter has followed a more flexible approach, focusing on the

Table 6.11

Classification of Political Parties in the Armed Forces Dimension Political Parties

Promilitary Antimilitary

ARENA, PCN, FSLN, MVR, PC, PA, UDI, PLC, FMLN MAS-B, COPEI, PAP, PUD, PSC, ID, PSN, EP/FA, MUPP-NP

Source: Based on HJ-Biplot analysis. Note: Parties for which the quality of representation is low are not included.

A-6

10/2/07

3:36 PM

154

Page 154

PARTY POLITICS AND IDEOLOGY

different dimensions of programmatic polarization.36 This approach allows us to assess which cleavages give rise to stronger divisions among parties. In methodological terms, this analysis has attempted to establish dimensions of polarization that may be useful to the comparative study of Latin American party systems and to represent, simultaneously, the position of parties in all the dimensions of polarization. Our analysis has shown that five dimensions structure party competition in Latin America and that polarization is larger in four dimensions: religion and moral values, state intervention, democracy, and armed forces. By contrast, divisions connected to social background play a much lesser role. This analysis has pointed out an overlap between the armed forces and democracy dimensions; these divisions are often more connected to national experiences than to left-right ideological differences. There is also significant overlap between the social background and democracy dimensions. The five dimensions analyzed in this chapter are congruent with divisions in the left-right ideological dimension; they clearly separate left and right parties. However, this analysis also shows two subgroups within the Latin American left, one more moderate and another more radical. By contrast, the positions of the Latin American right are much more homogeneous in these five dimensions. Overall, the position of parties in the left-right dimension provides us with a useful map of the orientations of parties on economic, political, and socioeconomic issues. Therefore, the left-right dimension simplifies and synthesizes crucial information about policy preferences in Latin American legislatures. Finally, it is necessary to conduct more in-depth analyses of political parties (including centrist parties) and party systems to complement the results of this research. Inclusion of centrist parties will provide us with a more encompassing and accurate picture of the dimensions of polarization in Latin American party systems.

A-6

10/2/07

3:36 PM

Page 155

155 ■ Appendix Table 6.12

Party Placement and Individual Self-Placement in the Left-Right Dimension (scale from 1 to 10)

Country Argentina Bolivia Chile Colombia Dominican Republic Ecuador

El Salvador Guatemala Honduras Mexico Nicaragua Panama Peru Uruguay Venezuela

Political Party ARI MAS UDI PPD PS PC PRSC PSC PRE ID MUPP-NP PRIAN PSP ARENA FMLN PCN GANA PSN PN PUD PRD PLC FSLN PA PS PAP EP/FA COPEI MVR MAS

Average Ideological Average SelfParty Placement Placement 3.50 1.57 7.00 3.77 2.56 6.81 6.72 6.83 3.92 3.60 3.00 5.60 3.67 8.04 1.76 6.93 6.58 7.17 7.15 2.60 2.82 7.00 1.83 7.29 7.17 3.91 2.60 5.83 3.68 3.83

4.17 2.14 7.08 3.85 2.56 7.63 7.97 8.08 4.21 3.47 3.22 7.10 4.00 8.42 1.72 7.27 7.35 7.17 7.70 3.00 3.14 7.11 1.69 7.05 9.00 3.81 3.03 7.00 4.13 3.83

N N (Population) (Sample) 12 27 36 21 12 48 36 24 14 15 11 10 6 27 31 16 34 7 61 5 95 48 38 16 8 28 40 7 77 21

Source: Data based on PELA 1999–2005. Note: Only parties with at least five legislators are included in the analysis.

6 14 24 13 9 27 29 23 14 15 9 10 6 26 29 15 25 6 47 5 22 27 29 14 6 22 30 6 47 12

A-6

10/2/07

3:36 PM

Page 156

156 Table 6.13

Results of Factor Analysis (total variance explained) Initial Eigenvalues

Extraction Sums of Squared Loadings

Rotation Sums of Squared Loadings

Component

% of Cumulative % of Cumulative % of Cumulative Total Variance % Total Variance % Total Variance %

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

3.472 17.359 17.359 2.506 12.530 29.890 1.799 8.996 38.886 1.659 8.293 47.179 1.311 6.557 53.736 .907 4.533 58.269 .893 4.466 62.734 .844 4.220 66.955 .787 3.935 70.889 .739 3.693 74.583 .700 3.499 78.082 .668 3.340 81.422 .609 3.046 84.468 .564 2.820 87.288 .539 2.695 89.984 .492 2.461 92.444 .460 2.298 94.743 .428 2.139 96.882 .350 1.752 98.633 .273 1.367 100.000

3.472 17.359 2.506 12.530 1.799 8.996 1.659 8.293 1.311 6.557

17.359 29.890 38.886 47.179 53.736

Source: Based on PELA 1999–2005. Note: Extraction method used is principal component analysis.

3.123 15.613 2.265 11.325 2.199 10.997 1.792 8.962 1.368 6.840

15.613 26.938 37.934 46.897 53.736

A-6

10/2/07

3:36 PM

Page 157

157 Table 6.14

Questions Used

Cleavage Religion and values

Armed forces

Democracy

State intervention

Social background

Question • With reference to your religious feelings and beliefs, where would you place yourself on a scale from 1 to 10, with “1” corresponding to the lowest level of religious conviction and the observance of religious practices and “10” corresponding to the highest level? • Degree of support for divorce and abortion: “1,” totally opposed; “2,” only in some circumstances; and “3,” completely in favor. • To what degree do you trust in the actions of the Catholic Church in public life in your country? Very much, a lot, not much, not at all. • On this card, there are a series of squares to be filled in that go from “1,” very negative, to “10,” very positive. On this scale, where would you place the role of the armed forces at the present moment? • To what degree do you believe the relationship between the armed forces and the government to be a threat to democracy: very high, high, low, not at all? • To what degree do you trust in the actions of the armed forces in public life in your country: very much, a lot, not much, not at all? • Could you tell me how important the relationship with the armed forces is to the consolidation of democracy in your country: very important, important, not very important, or not at all important? • With which of the following statements do you most agree? “Democracy is preferable to any other form of government”; or, “In the context of economic crisis or political instability, an authoritarian government may be preferable to a democratic one.” • There are people who say that without parties there cannot be democracy. How would you rate your agreement with this statement: very much in agreement, in agreement, somewhat in agreement, not at all in agreement? • I would like to get your opinion on a series of functions traditionally assigned to the state. In general, tell me to what degree—very high, high, low or not at all—you believe the state should intervene in the following areas: subsidizing housing for citizens; creating employment; providing public health coverage; guaranteeing universal and free university education; providing unemployment insurance; protecting the environment; covering basic needs of the population. • What level of schooling did your father officially complete: none, primary school, high school or similar, junior college degree or similar, BA, or graduate degree? • What level of schooling did your mother officially complete: none, primary school, high school or similar, junior college degree or similar, BA, or graduate degree?

Source: Based on PELA 1999–2005.

A-6

10/2/07

3:36 PM

158

Page 158

PARTY POLITICS AND IDEOLOGY

■ Notes 1. Lipset and Rokkan (1967), and Inglehart (1984). 2. Coppedge (1997). 3. Inglehart (1984). 4. The relative importance of cleavages varies over time, and there is always a critical cleavage that predominates over others and that structures party competition. 5. See, for instance, Sartori (1999). 6. These data are based on the individual perceptions and evaluations of legislators. 7. Sani and Montero (1986:155). 8. On the utility of the left-right dimension in Latin America, see Alcántara (1991:88). This author argues that four factors make the use of this distinction less common in Latin America. The first one relates to the legacies of populist regimes; the second one lies in the quite recent character of many Latin American parties. The third factor pertains to the legacies of authoritarian regimes and to the divide between democracy and authoritarianism. Finally, the fourth factor has to do with the emergence of new issues, which might also weaken the importance of left-right differences in Latin America. 9. Coppedge (1997:6). 10. Crespo (1997:116). 11. Luna and Zechmeister (2003:6). 12. Alcántara (2004:134). Some authors point out that in order to speak of parties in which ideology plays a role, it is only necessary that they adopt clear positions on a number of issues (Coppedge, 1997). 13. Legislators placed themselves and their respective parties on a scale ranging from 1 to 10, in which “1” means “left” and “10” means “right.” Parties are considered to be left oriented when their average value is less than or equal to 4 in the party placement and self-placement variables, and to be right oriented when their average value in both variables is equal to or higher than 7. Table 6.12 in the appendix shows averages in the self-placement and party placement dimensions. Party positions are very similar in both variables, except for the cases of the Colombian PC, the Ecuadorian PSC and PRIAN, the Guatemalan GANA, the Dominican PRSC, and the Venezuelan COPEI. 14. In the Uruguayan case, the data for this analysis correspond to the 2000–2005 legislative period. Parties with less than five members in parliament were excluded from the analysis. 15. Thus, this classification coincides to a large extent with the typology elaborated by Alcántara (2004c:157). However, whereas Alcántara considers five ideological locations (left, center-left, center, center-right, and right), this chapter only distinguishes three basic placements (left, center, and right). 16. The results of this analysis meet the requirements of factor analysis. The value for the Kaiser Meyer Olkin measure of sample adequacy equals 0.736. The five factors reported here explain 53.74 percent of the variance. The number of cases considered in the analysis was 502. 17. Questions for these variables are reported in Table 6.14 in this chapter’s appendix. 18. Galindo (1986) and Galindo and Cuadras (1986). 19. These evaluations might not coincide exactly with the strategic positions adopted in parliament by political parties. This analysis assumes, however, that there is a strong association between the average positions of parliamentarians and those

A-6

10/2/07

3:36 PM

Page 159

THE DIMENSIONS OF POLARIZATION IN PARLIAMENTS

159

of their respective political parties. Due to the proportionality and representativeness of the samples, differences in the number of legislators are assumed not to affect the results of this analysis. 20. It must be kept in mind that in the HJ-Biplot representation, programmatic dimensions are presented through vectors, and political parties as points with their corresponding labels. Those dots close to the point of the arrow above the gravity center have a higher magnitude and correspond to positive values, whereas those projected below the gravity center have negative values. The more distant the points are from the center of gravity, the higher the variability of these parties in the study. Closeness between dots reveals similarity between parties. Finally, the longer the vector, the higher the contribution of that element (dimension) to total variability. 21. Galindo, Fernández, Ávila, del Río, and Fernández (1999). 22. Factor axes refer to the axes of biplot representation. These axes are new variables, consisting of linear combinations of the original variables, that allow us to represent more than two variables in one dispersion figure, often in two-dimensional space. Their meaning is given by the variables they consist of. Factor axes are not related to the factors extracted by factor analysis. 23. The relative contribution of a factor to the element represents the amount of variability of each of the variables explained by that factor, and must be interpreted in the same way as the determination coefficient in a linear regression. That is, it indicates the contribution of each of the factor axes to the biplot representation of the dimensions of ideological polarization. 24. Since the axes are independent, the contribution of each of them to each variable is also independent. Therefore, it is possible to calculate the contribution of a factor plane by adding the contributions of the axes that form it. For instance, the contribution of the plane to the state intervention dimension will be the result of adding the contributions of axes 1, 2, and 3 to this dimension; that is, 588, 125, and 56, respectively. Consequently, the contribution of the first plane to the state intervention dimension will be 769 (out of 1,000). 25. When one variable, in this case the social background dimension, gets a similar contribution from more than one axis (396 in axis 1 and 276 in axis 2), this indicates its low variability in the study and, therefore, its low discriminating power. 26. A dimension characterizes an axis when it must be interpreted with regard to that axis. 27. Differences in this dimension do not necessarily reveal the presence of antidemocratic values but the existence of different views regarding the role of key democratic institutions, such as elections and political parties. 28. Due to its low quality, the graphic representation of the factor planes formed by axes 2 and 3 is not shown in this chapter, nor would it add anything of interest to the information presented in Figures 6.1 and 6.2. 29. Correa and Viera-Gallo (1986:58). 30. Luna and Zechmeister (2003). 31. The PRIAN is considered to be a right party. However, the average ideological self-placement of its legislators is 5.60. The fact that this biplot analysis is based on legislators’ perceptions may account for the distinctive position of this party. 32. Martínez Rodriguez (1997). 33. Crespo (1997:131). 34. See, among others, Lipset, Lazarsfeld, Barton, and Linz (1954) and Almond and Verba (1965). Almond and Verba indicate that early socialization exerts an important influence on the political attitudes and participation patterns of individuals.

A-6

10/2/07

3:36 PM

160

Page 160

PARTY POLITICS AND IDEOLOGY

35. See, for instance, Soule (1969:446) for a study on the state legislators of Michigan, and Kornberg and Thomas (1965:768–771) for a study on legislators in the United States and Canada. 36. See Inglehart (1984).

A-7

10/2/07

3:37 PM

Page 161

7 The Determinants of Intraparty Ideological Differences Flavia Freidenberg, Fátima García Díez, and Iván Llamazares Valduvieco

THIS CHAPTER ASSESSES THE INFLUENCE OF DIFFERENT TYPES OF

variables (contextual as well as individual) on the levels of ideological heterogeneity of Latin American political parties. More specifically, our chapter examines what factors condition differences between the ideological self-placements of legislators (in the left-right dimension) and the ideological positions of parties (according to the views of each party’s legislators). Our chapter looks at the influence that national, partisan, and individual characteristics exert on intraparty ideological divisions of Latin American parliaments. More specifically, this chapter explores the role played by four national explanatory variables: electoral volatility, levels of electoral personalism, duration of democratic experiences, and economic development. In addition, the question of whether party characteristics also affect intraparty ideological differences is examined. We are interested in testing the influences of conjunctural circumstances (whether parties are in power or in opposition) and of the ideological positions adopted by parties (whether they are radical or moderate, and whether they are left or right) on the ideological differences between legislators and their parties. Finally, this chapter considers whether the legislators’ individual attitudes regarding party leaders and interest groups have also a bearing on their ideological closeness to their respective parties. Since the data we are analyzing have a hierarchical, nested structure, we have conducted a three-level analysis. This analysis will allow us to test how variation at three different levels (national, partisan, and individual) affects ideological differences between parties and legislators. 1

161

A-7

10/2/07

3:37 PM

162

Page 162

PARTY POLITICS AND IDEOLOGY

■ Political Parties, Ideology, and Intraparty Ideological Differences Quite a few studies have been conducted on the role of ideology in democratic politics. Many of these studies advocate a spatial theory of ideology.2 Under this perspective, ideology offers voters and politicians a simplified map of policy alternatives. This spatial map, often summarized as the leftright dimension, allows citizens to select those politicians or parties closer to their preferences as well as to minimize their information costs. That is, the left-right dimension synthesizes to a maximum the complexity of policy alternatives, and allows citizens to vote on criteria of spatial proximity.3 In this way, the left-right ideological space makes political behavior more transparent and predictable. It also makes it easier to hold politicians accountable for their decisions. For these reasons, the left-right dimension reduces the transaction costs of the political field. But the importance of the left-right dimension in democratic politics cannot be taken for granted. In particular, many authors underline the fact that other types of mechanisms linking voters and politicians, such as clientelist exchanges and personalist practices, are widely present in many Latin American societies.4 Recent studies also underscore the crucial influence that levels of institutionalization exert on democratic politics and party systems in Latin America.5 In spite of these caveats, different empirical analyses show the interest and importance of analyzing the role played by ideology in Latin American politics.6 This chapter follows the lines of this type of literature by examining how different types of contextual and individual factors affect ideological differences within Latin American parties. The study of intraparty ideological differences is particularly relevant to research on the role of ideology. Ideologically cohesive parties will be able to send clearer messages to voters and provide them with transparent clues regarding their future behavior in parliament and in office. Ideological cohesion will also facilitate coordination within parties and between institutions ruled by the same party. Intraparty ideological differences have been linked to different types of contextual explanatory factors. In the first place, several authors have underscored the importance of electoral institutions (such as vote structure and size of electoral district) for the ideological characteristics of parties.7 In the second place, levels of electoral volatility might also affect intraparty ideological cohesion. The more unstable a party system, the higher the difficulties of making parties ideologically cohesive and of establishing partyvoter linkages based on abstract information. In the third place, experience with democratic politics could also positively affect intraparty ideological cohesion. As Herbert Kitschelt and Elizabeth Zechmeister (2003) point out, programmatic and ideological structuration demands sufficiently extended democratic experiences so that politicians and voters can coordinate around

A-7

10/2/07

3:37 PM

Page 163

THE DETERMINANTS OF INTRAPARTY IDEOLOGICAL DIFFERENCES

163

party and ideological labels. Ideally, democracy consists of a network of complex communications in informal and formal institutional arenas,8 and political parties do necessarily play a role in communication flows among citizens and politicians. But for that scenario to develop, societies must enjoy sustained periods of full democracy. Finally, the socioeconomic characteristics of countries can also have a bearing on the ideological cohesion of parties. The effective functioning of the left-right dimension in the interactions between politicians and voters may demand a sufficiently developed economy. Replacing personalist and clientelist exchanges by general and abstract linkages between voters and politicians will be more likely in urban and economically developed societies, in which economic transactions are impersonal and in which basic education is widespread. In those contexts, politicians will have incentives to coordinate with each other and appeal to voters on ideological platforms. The characteristics and situation of parties are also relevant to the levels of intraparty ideological cohesion. Parties in government may include more heterogeneous social and political groups, and will tend to position themselves in centrist positions. In addition, access to government will allow them to stay united without emphasizing ideological cohesion. For those reasons, parties in government may be more prone to ideological heterogeneity than opposition parties.9 However, in order to assess the influence of access to national power on intraparty ideological differences, we need to control for the ideological characteristics of political parties. The placement of parties in the left-right dimension might be correlated with ideological differences in two ways: conservative parties might tend to be more heterogeneous than left-wing political forces, and the radicalism of political parties might also affect ideological heterogeneity. On the one hand, radical parties might more actively encourage ideological homogeneity; on the other, legislators of radical parties might have incentives, under certain electoral configurations, to adopt comparatively moderate ideological positions and thus detach themselves from their parties’ positions. Finally, a third group of variables relates to the individual charcteristics and perceptions of legislators. In particular, their perceptions regarding their representative role and their ties to voters, interest groups, and party leaders can affect their propensity to hold ideological positions different from those of their respective parties. Certainly, these perceptions are not independent of the characteristics of both political parties and the national contexts politicians move in. Our analysis takes into consideration both the contextual and individual factors conditioning the ideological distance between legislators and their respective parties. For this purpose, we have included in our analysis several national characteristics plausibly affecting intraparty ideological differ-

A-7

10/2/07

3:37 PM

164

Page 164

PARTY POLITICS AND IDEOLOGY

ences: levels of electoral volatility,10 duration of democratic experience from 1945 to 1999,11 index of electoral personalism,12 and per capita GDP in 1999. Our model includes three variables at the party level. The first one relates to membership in a party in government or in the opposition.13 The second one captures the degree of ideological radicalism of political parties;14 the third maps the left-right position of political parties, according to their legislators. Our individual variables map the importance that legislators assign to party leaders and interest groups when making political decisions. These variables take into consideration the legislators’ perceptions of party structures and of their links to external organizations.15

■ Analysis In order to analyze intraparty ideological differences, we have constructed an indicator that maps the differences, in absolute values, between the average party position in the left-right dimension and the individual self-placement of legislators.16 The higher this value is, the higher the ideological distance between a legislator and his or her party will be. Table 7.1 shows average ideological differences by Latin American country. Table 7.2 shows ideological differences by political party. The countries displaying the highest ideological differences are Bolivia

Table 7.1

Differences Between Ideological Self-Placement and Party Positions, by Country (absolute values)

Country

Mean

n

Standard Deviation

Mexico Chile Argentina El Salvador Bolivia Honduras Colombia Costa Rica Nicaragua Guatemala Paraguay Ecuador Dominican Republic Peru Uruguay Venezuela Total

1.07 .71 .73 1 1.45 1.42 1.11 .84 1.39 1.43 1.24 1.24 1.06 1.00 .76 1.29 1.10

121 82 111 46 90 67 77 45 59 53 65 100 97 67 62 96 1238

1.02 .52 .74 .72 .94 1.05 .85 .71 .99 1.08 1.00 1.06 .98 .89 .55 .98 .93

Source: Data based on PELA 1994–2005.

A-7

10/2/07

3:37 PM

Page 165

165 Table 7.2

Differences Between Ideological Self-Placement and Party Positions, by Political Party (absolute values)

Political Party Argentina PJ UCR FREPASO Bolivia ADN MNR MIR UCS CONDEPA Chile PDC RN UDI PPD PS Colombia PL PC Costa Rica PUSC PLN RD Dominican Republic PRD PLD PRSC BIS Ecuador PSC DP PRE ID MUPP-NP El Salvador ARENA FMLN PCN PDC Guatemala PAN FRG FDNG Honduras PL PN Mexico PAN PRD PRI Nicaragua PLC FSLN

Mean

n

Standard Deviation

.83 .61 .61

58 34 19

.83 .47 .81

1.75 1.08 1.39 1.63 1.33

24 20 17 16 13

1.19 .90 .56 1.00 .58

.63 .98 .67 .50 .75

29 17 17 11 8

.36 .56 .67 .47 .46

1.15 1.03

48 29

.98 .58

1.05 .63 .67

22 20 3

.79 .60 .58

.82 1.10 1.45 1.84

46 33 11 7

.83 .94 1.13 1.40

1.37 1.15 1.57 .66 1.56

25 30 22 17 6

.99 1.17 1.08 .58 1.29

1.33 .74 .97 .96

15 18 8 5

.79 .65 .79 .22

1.55 1.10 1.56

33 14 6

1.11 1.04 .98

1.24 1.64

37 30

1.07 .99

.90 1.32 1.03

31 31 59

1.06 .93 1.03

1.56 1.20

32 27

1.18 .66 (continues)

A-7

10/2/07

3:37 PM

166

PARTY POLITICS AND IDEOLOGY

Table 7.2

continued

Political Party Paraguay ANR PLRA PEN Peru UN FIM PP PAP Venezuela MVR AD MAS COPEI Uruguay EP/FA PC PN Total

Page 166

Mean

n

Standard Deviation

1.40 .82 1.60

36 21 8

1.02 .82 1.10

1.00 1.00 1.13 .85

12 6 27 22

.74 .82 1.15 .61

1.36 1.34 1.17 1.17

47 16 12 6

.97 1.24 .82 .73

.79 .74 .72

30 19 13

.55 .57 .59

1.10

1238

.93

Source: Data based on PELA 1994–2005.

(1.45), Guatemala (1.43), Honduras (1.42), and Nicaragua (1.39); whereas the lowest ideological differences are present in Chile (.71), Argentina (.73), Uruguay (.76), and Costa Rica (.83). Inspection of standard deviations shows that Mexico, Honduras, Guatemala, Paraguay, and Ecuador present the most heterogeneous levels of ideological distance, whereas Chile, Peru, and, to a lesser degree, Argentina, El Salvador, and Costa Rica are quite homogeneous countries in this respect. Comparison of means by parties shows that the highest differences in the individual and party positions are those of the Dominican BIS, the Bolivian ADN and UCS, and, to a lesser degree, of the Ecuadorian PRE and MUPP-NP, the Guatemalan PAN and FDNG, and the Paraguayan PNH and PEN. The lowest differences are those of the Chilean PPD, the Argentine FREPASO and UCR, the Costa Rican PLN, the Chilean DC, and the Ecuadorian ID. Before conducting the multilevel analysis, it is also necessary to examine the relationships between the national contextual variables we are considering. Table 7.3 reports the Pearson correlation indexes between all our contextual variables and the national average for the party legislators’ ideological differences. The Pearson correlation index is highest for the per capita GDP variable (–.68, significant at the .01 level). The Pearson correlation index with the years of democracy variable is moderately strong

A-7

10/2/07

3:37 PM

Page 167

THE DETERMINANTS OF INTRAPARTY IDEOLOGICAL DIFFERENCES Table 7.3

167

Pearson Correlation Coefficients Between the Aggregate National Variables and the Average National Differences of the Ideological Positions of Parties and Legislators (16 cases) Index of Electoral Personalism

Years of democracy Volatility Per capita GDP (1999) Mean ideological differences

.21 .21 .18 –.24

Years of Democracy

Volatility

Per Capita GDP (1999)

.12 .50* –.49

.04 .35

–.68**

Source: World Development Indicators, World Bank. Notes: * Significant at the .05 level. ** Significant at the .01 level.

(though not significant at the .05 level). Other variables are much more weakly related, at the aggregate level, to our dependent variable. For our multilevel analysis, all variables but membership in opposition parties have been grand-mean centered. As previously stated, our dependent variable is the absolute ideological difference between a legislator’s ideological self-placement and the ideological placement of his or her party, according to its legislators. As a prior step, we ran an unconditional model with random intercepts in order to calculate intraclass correlations. According to our analysis, most of the variation (92.6 percent) takes place at level 2; that is, at the party level. Variation at the individual and national levels is much lower (2.8 and 4.6 percent, respectively).17 As a second and definitive step, we conducted a variance components multilevel model with no cross-level interactions, in which only intercepts are random. We therefore avoided elaborating a model that is complex and difficult to interpret. The following equations describe our model: Level-1 Model: Y = P0 + P1*(Party Leaders) + P2*(Interest Groups) + E Level-2 Model: P0 = B00 + B01*(Party Ideology) + B02*(Opposition) + B03*(Party Radicalism) + R0 P1 = B10 P2 = B20 Level-3 Model: B00 = G00 + G0001(Volatility) + G002(Years Democracy) + G003(Per Capita GDP) + G004(Elect. Personalism) + U00 B01 = G010 B02 = G020 B03 = G030 B10 = G100 B20 = G200

A-7

10/2/07

3:37 PM

168

Page 168

PARTY POLITICS AND IDEOLOGY

The main results of our analysis are reported in Tables 7.4 and 7.5. Results indicate that two national contextual variables—per capita GDP and electoral volatility—have significant effects on our dependent variable. The higher the per capita GDP is, the lower the ideological differences between legislators and their parties are as well. And the higher the volatility levels, the higher the ideological differences between legislators and their parties. The coefficient for years of democracy points in the right direction (the higher the democratic experience, the lower the ideological differences), but does not reach statistical significance. Finally, electoral personalism does not seem to exert any influence on our dependent variable. Table 7.4

Fixed Effects Estimates (without robust standard errors)

Fixed Effect For INTRCPT1, P0 For INTRCPT2, B00 INTRCPT3, G000 VOLATIL, G001 Years Democracy, G002 Per capita GDP, G003 ELECPERS, G004 For Party ID, B01 INTRCPT3, G010 For OPPOSIT., B02 INTRCPT3, G020 For PARTYRAD, B03 INTRCPT3, G030 For Party LEAD. slope, P1 INTRCPT3, G100 For INT. Groups slope, P2 For INTRCPT2, B20 INTRCPT3, G200

Table 7.5

Coefficient

Standard Error

t Ratio

Approximate d.f. P Value

1.105147 0.009997 –0.003295 –0.126147 –0.210100

0.031752 0.002441 0.002238 0.034601 0.182577

34.805 4.095 –1.472 –3.646 –1.151

11 11 11 11 11

0.000 0.002 0.169 0.004 0.275

0.038705

0.024714

1.566

43

0.124

–0.132449

0.066397

–1.995

43

0.052

0.105589

0.053686

1.967

43

0.055

–0.081965

0.036835

–2.225

1143

0.026

0.096365

0.028514

3.380

1143

0.001

d.f.

Chi-square

P Value

0.10908 0.77889

28

60.33908

0.001

0.00009

11

15.63341

0.155

Estimation of Random Components

Random Effects

Standard Variance Deviation Components

Level 1 and level 2 INTRCPT1, R0 0.10908 Level 1, E 0.88254 Level 3 INTRCPT1/INTRCPT2, U00 0.00925 Deviance = 2998.197386 Number of estimated parameters = 13

A-7

10/2/07

3:37 PM

Page 169

THE DETERMINANTS OF INTRAPARTY IDEOLOGICAL DIFFERENCES

169

Among our level 2 variables, membership in the opposition and party ideological radicalism approach the .05 significance level. Legislators in opposition parties tend to be ideologically closer to their parties than are those in government. Other things being equal, members of ideologically radical parties and (to a lesser degree) members of conservative parties tend to be farther apart ideologically from their respective parties. Finally, our individual variables exert a significant influence on our dependent variable in the expected direction. Legislators more prone to taking into consideration party leaders when making decisions tend to be ideologically closer to their respective parties, whereas legislators who privilege their relationships with interest groups tend to be ideologically farther from their respective parties. The coefficients for these two variables reveal the role played by individual considerations regarding party leaders and links to social groups, and, indirectly, by party structures and party links to social groups in and of themselves. As for the random components of the model, the statistically significant R0 (.011) reveals that unmodeled variation across parties is quite important. The variance component for level 3, U00 is low and statistically insignificant. In combination with the importance of individual perceptions regarding the role of party structures in individual political decisions, these results underscore the need to explore and include in our model other party characteristics affecting intraparty ideological differences.

■ Conclusion Most analyses of intraparty ideological differences focus on a single level of analysis, be it individual, partisan, or national. By contrast, this chapter has examined in a combined manner the effects of contextual, partisan, and individual factors on intraparty ideological differences. The results of this analysis underscore the importance of two contextual, national explanatory factors: per capita economic wealth and electoral volatility. This finding reveals the importance of both structural processes and political-institutional characteristics for the development of ideologically homogeneous political parties. And although our two main level-three explanatory variables refer to contemporary characteristics of Latin American polities, they cannot be detached from long-term historical processes. Our direct measures of party characteristics seem to exert a quite moderate influence on our dependent variable. Members of opposition parties tend to be closer to their respective parties’ ideological positions than members of ruling parties. And legislators in nonradical parties and in left-oriented parties are closer to their respective parties than are members of radical and conservative parties. Considering the overwhelming importance of

A-7

10/2/07

3:37 PM

170

Page 170

PARTY POLITICS AND IDEOLOGY

variation at level 2, these results underscore the need to consider other party characteristics plausibly affecting intraparty ideological differences. Finally, our results show that individual perceptions regarding the role of party leaders and interest groups exert a significant influence on ideological differences. Legislators paying close attention to party leaders when making political decisions are more likely to be ideologically close to their respective parties. And legislators assigning importance to interest groups are more prone to have ideological differences with their respective parties. However, these individual perceptions are most likely conditioned by the characteristics of political parties. This possibility suggests the need to consider other party characteristics in the analysis. Our analysis helps clarify the role played by different types of factors affecting intraparty ideological differences. The influence exerted by national characteristics and, above all, party characteristics reveals the importance of contextual factors in accounting for intraparty ideological differences. Some of these factors (such as party characteristics, economic modernization, and political-institutional stability) are shaped in the course of history. It is for this reason that historical processes must be taken into consideration when analyzing the development of ideologically cohesive parties in Latin America. From a methodological perspective, in the future this multilevel analysis should be made more systematic and refined by considering other explanatory factors (particularly at level 2) and by examining, theoretically and empirically, how the effects of our explanatory variables are conditioned by the characteristics of the contexts in which they are nested. That is, a further development of this study must consist in the assessment of models in which level-three variables condition the effects of level-two and level-one variables, and in which level-two variables impinge on the effects of level-one variables. Conducting this type of analysis will demand a more thorough, prior theoretical exploration of the factors conditioning intraparty ideological differences.

■ Notes 1. On multilevel analysis, see Kreft and de Leeuw (1998), Snijders and Bosker (1999), Steenbergen and Jones (2002), and Luke (2004). 2. See, for instance, Downs (1957) and Hinich and Munger (1996). 3. Downs (1957). 4. Conaghan (1995) and Levitsky (2003). 5. Mainwaring and Torcal (2005). 6. See Alcántara (1991 and 2004a), Luna and Zechmeister (2005), Kitschelt and Zechmeister (2003), and Ruiz Rodríguez (2003a). 7. See Carey (2000); Bowler, Farrel, and Katz (1999); Ames (1992); and Hallerberg and Marier (2001). 8. Habermas (1998).

A-7

10/2/07

3:37 PM

Page 171

THE DETERMINANTS OF INTRAPARTY IDEOLOGICAL DIFFERENCES

171

9. See Amorim Neto and Santos (1997) and Hawkins and Morgenstern (2003). 10. Aggregate electoral volatility was calculated taking into account the elections giving rise to the legislatures included in the analysis. Electoral data were excerpted from Payne, Zovatto, Carillo, and Allamand (2002) and Alcántara (2003). 11. Excerpted from Mainwaring, Brinks, and Pérez-Liñán (2001). 12. This variable consists in the index of electoral personalism developed by Hallerberg and Marier (2001). 13. Altman and Pérez-Liñán (1999). 14. This variable was created by standardizing and then squaring the ideological positions of each party according to its legislators. 15. The text of this question read as follows: “When you make political decisions, to what degree—very high, high, low or not at all—do you take into account (i) the leaders of your party, (ii) interest groups?” 16. All data come from the PELA. Our multilevel analysis included 16 countries, 47 parties, and 1,153 individuals. With the exceptions of Peru (2001–2006) and Guatemala (1995–1999), all legislatures covered the year 2000. Those legislators for which we lacked a specific party label were excluded from the analysis, as well as those parties with fewer than 8 members. 17. Deviance for this model equals 3142.036099 with four estimated parameters.

A-7

10/2/07

3:37 PM

Page 172

A-8

10/2/07

3:37 PM

Page 173

8 Programmatic Coherence Leticia M. Ruiz Rodríguez

■ The Program in Political Parties Two of the functions traditionally attributed to political parties are the aggregation of interests and the enunciation of projects for improving society. A party’s diagnosis of society’s problems and its proposal for solutions are generally referred to as the “party program.” The enunciation of a program is deemed crucial to the definition of a political party; for scholars who study electoral behavior, party program (in conjunction with the party’s promises and past behavior) is considered a decisive factor in voting behavior. In this rational approach to voting behavior, party program constitutes a way of orienting the electorate or of providing them with a cognitive shortcut.1 While this party model, in which the programmatic element is of central importance, is somewhat useful in looking at the dynamics of party competition in Europe, the phenomenon of “catch-all” parties has put into question the identification of parties with a specific program and ideology.2 Nevertheless, the binomial of “political party and program” continues to be valid to a great extent in the imaginary of both European scholars and citizens. This can be seen by the fact that some discussions on the present crisis of political parties make reference to the lack of well-defined, reliable programs as a causal factor in the ineffectiveness of parties as mobilizing agents and representatives of people’s interests.3 However, where this idea of parties as defined by a specific political program is especially problematic is in the case of Latin America. On numerous occasions, both explicitly and implicitly, it has been stated that the concept of parties as actors who hold a particular political program cannot be applied to the political parties of this region.4 Those who support this point of view recognize that elections motivate the creation of platforms, manifestos, and plans of action. But they also recognize that behind these elements, which have come to form part of electoral marketing, is another element: a human collective group that most probably is not completely in

173

A-8

10/2/07

3:37 PM

174

Page 174

PARTY POLITICS AND IDEOLOGY

agreement with the party program and that, in fact, may perceive it as something imposed externally by the party organization. This would help explain why views on party programs held by party members and party leaders can contradict each other.5 Other explanations for these differences have to do with elements underlying affiliation to political parties, such as the influence of personalist elements in the process of selecting party leaders. Studies on clientelist and personalist patterns of party competition have argued for the central importance of these variables in Latin America, where involvement and promotion within a party are results of personal loyalties and benefits. Other scholars have argued that some degree of programmatic structuring and party competition is incompatible with the aforementioned model.6 This chapter analyzes the nature of political parties in Latin America, with special attention to their programmatic identity. Taking party members as a unit of measurement, the chapter undertakes an exploration of the degree to which parties are united by similar diagnoses and proposals of actions. Party coherence is the instrument used to measure dissent/consent in the party system. Coherence captures the degree to which members of a political party are in agreement on issues relevant to the political and party arenas. When the members of a political party have similar stances, this party is seen to have high coherence, and vice versa. Previous studies defend a tridimensional conception of coherence, 7 with the three dimensions being ideological coherence, programmatic coherence, and organizational coherence. This chapter concentrates on programmatic coherence, analyzing political parties in terms of the degree of similarity or dissimilarity of their members’ political projects. To examine this similarity, we look both at aspects related to parties’ plans of action and at their views on problems and their solutions. The focus, therefore, will be on issues of a substantive nature. Ideological issues constitute the other group of substantive concerns that have an impact on the nature of a party and give a symbolic interpretation to programmatic questions, which we classify in this chapter as “specific proposals for action.” Some studies defend the image of parties as united around a common program, but this image is not supported by reality. 8 Nevertheless, our analysis confirms the existence of a minimal level of programmatic consensus between party members. This level of consensus varies greatly, allowing us to characterize Latin American party organizations. The chapter aims to provide empirical evidence to support the argument that, contrary to what is commonly thought, political parties do not always design their programs with electoralist and opportunist criteria in mind. On the contrary, members of a political party share, to a greater or lesser degree, a similar view of society, although their views can be bound together somewhat loosely.

A-8

10/2/07

3:37 PM

Page 175

PROGRAMMATIC COHERENCE

175

Among party members whose positions on programmatic issues can be examined, this chapter concentrates on the parliamentary elite.9 An analysis of programmatic coherence through the parliamentary elite is important for various reasons. For one, it provides evidence on the nature of political parties from the viewpoint of party members. Despite the fact that studies on parties have increased, for decades party members have been overlooked in analyses of political parties.10 Through its examination of the parliamentary elite, this chapter explores the existence of structuration around ideas, program, attitudes, and issues; in any case, this structuration exists side by side with clientelist and personalist practices. Another reason for concentrating on the parliamentary elite is that this group greatly affects party politics and the stability of political systems. The existence of coherent parties increases the predictability of their actions. For example, knowing the degree of agreement among party members helps predict how they will vote. Programmatic coherence is, in a way, a source of natural unity; thus, for issues around which there is a great deal of coherence, it is easier to predict the content of the vote and the possibility that a majority of the party will vote the same way. In addition, a coherent party is more effective in achieving its objectives in the electoral arena, the government, the legislative branch, and the intraparty sphere.11 Moreover, coherent parties are more easily held to account: promises and party identification are clearer, so voters more easily know whether promises are kept and/or positions shifted. All of these reasons are especially significant in the Latin American case, given that it is only since the transition to democracy that more attention has been paid consistently to these issues. In the panoramic spirit informing this book, an examination of coherence from the point of view of parliamentary elites gives us systematic information on the preferences and orientations of Latin American legislators. As stated throughout this book, rigorous and systematic data of this kind have previously been hard to come by. The next section of this chapter contains a brief discussion on the conceptualization and operationalization of the concept of “programmatic coherence.” The third section describes the levels of programmatic coherence of the parties studied. Finally, the fourth section examines the relationship between party positions on the role of the state in the economy and party ideology.

■ Programatic Coherence: Concept and Operationalization All studies associate party coherence with the existence of unity within political parties.12 Nevertheless, there are persistent debates about how “unity” should be understood and measured. With respect to the question of what the content of “coherence” should

A-8

10/2/07

3:37 PM

176

Page 176

PARTY POLITICS AND IDEOLOGY

be, the measurement of agreement may have to do with the attitudes or behavior of party members. Some scholars study agreement in the behavior of party members; most specifically, through the way they vote.13 For the purposes of their study, they use such terms as “party discipline” and “party cohesiveness.” On the other hand, attention has not often been paid to agreement among party members in attitudes and opinions, mainly because of the lack of systematic empirical evidence on the opinions of party members. There is, however, another reason for the lack of attention paid to coherence revealed through attitudes. Explanations of political dynamics tend to underestimate the importance of attitudes as opposed to behavior, a trend seen in the study of all facets of party behavior.14 This is especially true when legislators constitute the object of study; the tendency has been to consider the way legislators vote as being the relevant outcome of political parties and, thus, the aspect of party members’ behavior that should be analyzed. Voting has been considered a priority for study because it determines the passing of laws, as well as a large part of other parliamentary decisions. This chapter, on the other hand, accords importance to attitudes, since attitudes condition behavior, and since knowledge of opinions and perceptions helps to predict behavior. Attitudes, however, also constitute an independent variable with importance in its own right. Thus, the existence of attitudinal unity allows political parties to act more effectively. Whenever data are available, the study of similarities and differences in the thinking of party members, such as that carried out in this chapter, is a task that is eminently useful. The first studies that apply the concept of “coherence” to the study of parties through attitudinal data concentrate on organizational aspects.15 Lee Anderson’s study (1968) measures the degree to which members of the same political party are in agreement on ways of achieving certain goals through issues having to do with internal organization. Later, the application of this concept was widened to include other spheres of party life, and coherence as an organizational issue was left behind as the focus shifted to party ideology and program. The study of party coherence in Latin America has been comparatively less frequent than in other areas and has only recently begun to develop and be applied.16 In Latin American cases, the dimension of coherence that has received by far the most attention is unity of ideological principles. Although studies have not always paid attention to such aspects as convergence in the party’s ideological placement by its members or convergence in placements of the electorate, the concept of party coherence has been used, especially in the debate on the usefulness of the categories “right” and “left” (see, for example, Chapter 6). Within the framework of this debate, measurements of unity in ideological placement and its relationship to positions on program have been sketched out.17 Thus, we can say that there has long been interest in the concept of ideological unity in political parties.

A-8

10/2/07

3:37 PM

Page 177

PROGRAMMATIC COHERENCE

177

In this chapter, “programmatic coherence” is used to mean the degree of agreement that gives rise to the articulation of concrete proposals within parties.18 This characteristic of parties has received less attention than has ideological coherence, even taking into account the fact that programmatic principles have been the focus of recent and excellent studies.19 The main reasons for this lack of attention to programmatic as compared to ideological coherence have to do with the difficulty of capturing programmatic coherence. On one hand, in order to capture programmatic coherence, one must define the issues around which political parties have developed concrete plans of action. On the other, it is not easy to achieve comparability between data obtained from different party systems. The data gathered in the PELA allow us to measure the level of agreement or dissent on programmatic issues between the members of different parties. The parties chosen for this study are those on which there is comparable data for approximately the same time period. Cases were chosen based on Alcántara’s study (2004a),20 which, within thirteen political systems, selected the two parties that had had the best electoral results in the decade of the 1990s. The sample constitutes a representative selection of regional party reality, since it contains variations on aspects related to the parties (e.g., their ideologies and ages), as well as on aspects of the political system to which the parties belong (e.g., number of parties or polarization of the party system). The PELA questionnaire was not originally designed to measure the coherence of the parliamentary elites who were interviewed. Some questions that would have been relevant to this subject did not form part of the questionnaire. On the other hand, the list of issues on which legislators were surveyed was the same in all countries. The use of the same questionnaire for all countries has both advantages and disadvantages. Having asked the same questions in all countries and of all parties allows for an exact comparison of the level of coherence on the same issues in different contexts. The disadvantage is that some issues are not relevant in a given country at a given moment. The reasons are varied: national consensus on the issue may have erased it from the public consciousness, the issue may not yet have become relevant, or it simply never will be relevant in a given country. In the above-mentioned situations, the inclusion of an issue that is not relevant may, depending on the case, cause one to over- or underestimate the measure of programmatic coherence. It is difficult to get around this obstacle when the basis of the study is the comparison of identical questions for every case, but the limitation must be recognized. For this reason, it was necessary to carefully select the questions to be included in the operationalization of programmatic coherence, trying to make sure that all questions formed part of the debate in all national contexts. A factor analysis was done on legislators’ answers on various subjects,

A-8

10/2/07

3:37 PM

178

Page 178

PARTY POLITICS AND IDEOLOGY

chosen with an eye not only to the programmatic dimension of coherence, but also to ideological and organizational aspects. The analysis confirms the existence of seven components, four of which relate to questions that can be classified as “programmatic,” since they have to do with strategies and evaluations of specific issues. The answers had to do with the diagnosis of political and socioeconomic problems as well as with attitudes toward strategies for increasing public spending and toward the ideal role of the state. Since the components related to the diagnosis of problems and to strategies to be followed were so closely related in subject matter, these four components were reduced to two (evaluative and strategic). Thus, the factor analysis was only partially carried out. The first component of programmatic coherence, “denominated evaluation component,” is structured around questions on economic, social, and political issues of major importance in most Latin American countries. This component refers to the degree of coherence in legislators’ assessments of how important these issues are within their own countries. The second component, labeled as “strategy component,” refers to coherence in proposals for concrete solutions to problems. Figure 8.1 contains the questions used to measure party coherence in these two components. After the questions were chosen, three simple mathematical procedures were followed. First, all answers were coded on a single scale, which ranged from 0 to 100. Then, in order to calculate coherence in each component, the answers of each member of a party to that component were added up and divided by the number of questions. With the values obtained for each member, the standard deviation in this component was calculated for the party under study. Last of all, programmatic coherence, which takes both of these components into account, was obtained through adding the components and dividing the result by two, which is the number of components.21 The standard deviation of the average of party members’ answers has been considered the statistic that best captures the idea of proximity between stances within the same political party. Calculations of the spread of answers based on the mode would have captured convergence in one answer, but would not have captured the idea of similarity of opinions, which is what this study understands coherence to be. Given the fact that the calculation is carried out in standard deviations, for the purposes of the interpretation of results, higher values in the measurement of coherence in one component or in total programmatic coherence indicate higher incoherence. Thus, “0” would mean maximum coherence, or complete agreement between legislators from one party; and “100,” minimum coherence, or complete disagreement between legislators from one party. Together with the number for coherence, we put the averages of the programmatic positions of parties for each one of the components on a scale of 1 to 100 (Figure 8.1 shows the meaning of the scale in each component).

A-8

10/2/07

3:37 PM

Page 179

PROGRAMMATIC COHERENCE

179

Figure 8.1 Degree of Programmatic Coherence Among Party Legislators Evaluation Component None at all

Very high

The degree of importance you assign in your country to the problem of: Unemployment Public health/social security Education Democratization of public life Human and minority rights Conflicts between state powers Corruption Political violence Strategy Component None at all

Very high

The level of state intervention you feel there should be in: Primary education Housing Secondary education Social security Environment Action you feel should be taken regarding levels of public spending on: Public health Unemployment insurance Housing Pensions

■ Levels of Programmatic Coherence in Political Parties Coherence in the Evaluation of Problems and Coherence in the Definition of Strategies

The evaluation component measures each party’s coherence when assigning importance to eight political issues that are usually difficult to address in Latin America (unemployment, public health, education, democratization of

A-8

10/2/07

3:37 PM

180

Page 180

PARTY POLITICS AND IDEOLOGY

public life, human rights or minority rights, conflicts between state powers, corruption, and political violence). Since these are currently key issues, the programs of Latin American political parties tend to include rather extensive considerations related either to decreasing or to solving these problems. Our data allow us to determine to what degree there is contention or agreement within the party itself regarding these evaluations, which are presented to the public as accepted by the party as a whole. The axis of measurement for programmatic coherence in the evaluation component consists of the agreement or disagreement around how important various issues are. Table 8.1 shows the average importance assigned by each party to the eight issues listed above (second column). A higher score in the total average of all the issues means that legislators attribute more importance to the problems included in this component. Particularly striking in this respect is the low importance assigned to these issues by legislators from the FRG, as

Table 8.1

Coherence in the Evaluation Component

Party/Country FMLN/El Salvador PC/Uruguay PRSC/Dominican Republic ADN/Bolivia UCR/Argentina PRD/Dominican Republic EP/FA/Uruguay PLH/Honduras FSLN/Nicaragua DP/Ecuador ANR/Paraguay PLRA/Paraguay PAN/Mexico PRI/Mexico MNR/Bolivia PNH/Honduras PLN/Costa Rica PLC/Nicaragua PUSC/Costa Rica PAN/Guatemala PJ/Argentina PDC/Chile PPD/Chile ARENA/El Salvador FRG/Guatemala MUPP-NP/Ecuador

n

Importance of Problems (average)

Coherence in the Evaluation

20 21 11 24 34 47 30 37 27 31 36 21 31 60 20 30 20 32 22 35 59 29 12 16 14 7

76.94 42.59 71.72 70.95 69.93 74.41 64.17 75.90 79.63 67.83 73.84 73.15 76.43 75.88 71.67 70.83 61.25 70.05 60.48 64.13 46.89 39.08 49.77 58.33 18.25 68.65

9.12 9.47 10.22 10.35 10.80 10.94 11.21 11.22 11.72 12.44 12.46 12.99 13.55 13.89 13.98 14.59 14.80 15.02 15.30 16.52 17.28 17.48 17.48 20.57 21.13 21.88

Source: Data based on PELA 1994–2005.

A-8

10/2/07

3:37 PM

Page 181

PROGRAMMATIC COHERENCE

181

opposed to the evaluation of those from the Nicaraguan FSLN, which, among all parties, gives the most alarmist assessment of their national situation with respect to these problems. Coherence in assigning importance to these problems varies widely among parties. It must be remembered that when dealing with standard deviations, higher scores indicate a lower degree of coherence. Thus, legislators from the FMLN and the PC are those who most agree among themselves in the assessment of problems. On the other hand, the MUPP-NP has the lowest level of evaluation coherence, followed by the FRG and ARENA. In the case of the Ecuadorian party, the low level of coherence could be related to the fact that since this party was created in response to issues of identity and the rights of indigenous peoples, it needed to incorporate the demands of different sectors of society (unions, religious groups, entrepreneurs, etc.).22 However, the questions used for this study deal with a wide gamut of problems; in other words, it is possible that the MUPP-NP is very coherent in its diagnosis of some issues, but not with respect to these problems as a whole. It is probable that this is often the case in most of the parties with a large indigenous component that have proliferated in Latin America. The analysis of political parties that belong to the same legislature shows that these parties do not generally present similar levels of coherence in their evaluation of problems. The exceptions are the Uruguayan parties (EP/FA and PC) and the Chilean ones (PDC and PPD). However, the mean shows a general tendency for the two parties studied from each country to be relatively close together in their assessment of problems. In this respect, the parties from Bolivia, Paraguay, Costa Rica, and Ecuador are the pairs that show the most similarity between parties from the same system. The strategy component measures the internal coherence of parties when defining their actions. The answers to the PELA questionnaire solicit fundamentally individual positions on the ideal level of state intervention in different spheres of the economy. For this purpose, there is a battery of questions on the role of the state in various spheres and another on the most desirable level of public financing. Table 8.2 shows that the Dominican PRD is the most coherent in its programmatic strategy, which is based on a prominent role for the state. Since it was founded, this party has fortified its anti-Trujillo role with constant demands for social justice and for a commitment to increasing economic prosperity.23 The PRD is followed by the EP/FA and the FMLN, which have a high degree of coherence in strategic issues, and which support high levels of state intervention. In contrast, of all parties, it is the Guatemalan PAN, followed by the Salvadoran ARENA, that has the lowest level of coherence in terms of strategy. Among pairs of parties that belong to the same legislative body, the

A-8

10/2/07

3:37 PM

182

Page 182

PARTY POLITICS AND IDEOLOGY

Table 8.2

Coherence in the Strategy Component

Party/Country PRD/Dominican Republic EP/FA/Uruguay PPD/Chile FMLN/El Salvador UCR/Argentina PRSC/Dominican Republic PDC/Chile PLN/Costa Rica MUPP-NP/Ecuador DP/Ecuador FSLN/Nicaragua FRG/Guatemala ADN/Bolivia PLC/Nicaragua ANR/Paraguay PUSC/Costa Rica MNR/Bolivia PLH/Honduras PRI/Mexico PC/Uruguay PLRA/Paraguay PJ/Argentina PAN/Mexico PNH/Honduras ARENA/El Salvador PAN/Guatemala

n

Role of the State (average)

Increase in Public Spending (average)

Coherence in Strategy

47 30 12 20 34 11 29 20 7 31 27 14 24 32 36 22 20 37 60 21 21 59 31 30 16 35

93.62 96.22 96.67 89.33 92.75 91.52 84.60 93.33 86.67 80.43 88.40 93.33 87.50 91.25 92.04 84.55 86.00 86.67 87.78 85.71 86.67 80.56 80.22 83.78 88.75 80.95

94.68 92.92 89.58 94.38 90.81 93.18 90.09 87.50 91.07 92.74 93.98 81.25 81.77 89.06 78.82 77.84 85.00 81.08 78.13 66.07 89.88 84.11 79.84 82.08 78.91 71.07

5.91 6.40 6.54 6.94 6.96 8.02 8.62 8.77 8.78 9.17 9.46 10.39 10.93 11.79 12.91 12.99 13.35 13.68 14.07 14.26 14.93 15.49 15.58 15.80 16.27 17.48

Source: Data based on PELA 1994–2005.

Chilean parties are the ones with the highest levels of coherence; the Mexican and Honduran parties have the lowest levels. Nevertheless, in the systemic approach, evidence seems to suggest that in this component, levels of coherence within pairs of parties from the same system vary widely. We do not here present data on how the members of each party are distributed with respect to the answers of each component. Nevertheless, it is relevant to note that data analysis indicates that among legislators with typical stances, the highest level of consensus on desirable state intervention is achieved in primary education, public health, and social security. In most parties, legislators’ answers show a certain contradiction between, on one hand, the most desirable level of state intervention in the economy and the optimal level of public spending and, on the other hand, self-placements in the statism-market scale. This may be due to legislators’ opting for intermediate positions on the scale. This means that the differ-

A-8

10/2/07

3:37 PM

Page 183

PROGRAMMATIC COHERENCE

183

ence between parties in these programmatic issues would be more clearly shown in the questions separated by subject in relation to the degree of state intervention and the degree of increase in public spending. Even so, although it is not as strong as would be expected, a significant relationship between both these issues is shown in the results (see Table 8.3), with those who attribute more importance to the state being more favorable to increasing public spending. However, both averages (most desirable degree of state intervention and degree of increase in public spending) have a relationship to coherence in these issues, in which high levels of coherence are associated with positions more favorable to state intervention and to increases in public spending. The coherence of parties on programmatic issues tends to be slightly higher in the strategy component than in the evaluation component. This slight difference in degrees of coherence can be seen in the similarity of the ranges of the standard deviations that measure the coherence of each component (first row, Table 8.4). In turn, where there are the greatest differences between parties is in the evaluation of problems, as can be seen in the aggregation of standard deviations (second row, Table 8.4). Members of the political parties analyzed seem to be more in agreement on what kind of intervention should be assigned to the state and to the market than on the degree of importance of problems, which is where higher levels of disagreement and differences between parties can be seen. Figure 8.2 represents the distribution of agreements in the two components analyzed. All political parties are ordered from greatest to least coherence, and the shadowed box, or area, represents 50 percent of cases with central values. In other words, the box plot shows the space occupied by

Table 8.3

Pearson Correlations in the Strategy Component

Role of the state Significance n Public spending Significance n Coherence strategy component

Role of the State

Public Spending

1.000

.390* .049 26 –.764**

1.000

Source: Data based on PELA 1994–2005. Notes: * The correlation is significant at the 0.05 level (two-tailed). **The correlation is significant at the 0.01 level (two-tailed).

Coherence Strategy Component –.658** .000 26 .000 26 1.000

10/2/07

3:37 PM

184

Page 184

PARTY POLITICS AND IDEOLOGY

Table 8.4

Variation in Programmatic Coherence, by Component Programmatic Dimension Evaluation Component

Strategy Component

12.76 366.41

11.57 295.49

Range of deviations Aggregate standard deviations Source: Data based on PELA 1994–2005.

cases with central values in terms of coherence (the boxes of the figure). Thus, a smaller-sized box (which in this case corresponds to the strategy component) indicates that there is more concentration in parties with medium coherence. In turn, the arms situated to the right and left of the boxes represent parties in the top 25 percent and in the bottom 25 percent for coherence. Thus, the area in Figure 8.2 located more to the left in the strategy component indicates that coherence is greater in this component. Finally, the larger size of the evaluation component indicates that there are more differences among parties in levels of coherence in the evaluation component.

Figure 8.2 Distribution of Coherence of Party Legislators in the Evaluation and Strategy Components

Components of the Program Dimension

A-8

Evaluation

Strategy

0

10 Maximum Coherence

20

30 Minimum Coherence

A-8

10/2/07

3:37 PM

Page 185

PROGRAMMATIC COHERENCE

185

Total Programmatic Coherence

Table 8.5 shows the results of the calculations for total programmatic coherence, which bring together data for coherence in the components of evaluation and strategy. The FMLN, PRD, EP/FA, and UCR are the most coherent parties in the definition of problems and of strategies to solve these problems. It is worth pointing out that the two fronts are among the most programmatically coherent parties. A third front, the FSLN, shows high coherence in programmatic objectives. In spite of the fact that this party was formed from a collection of many diverse groups, and is thus very heterogeneous, in interviews its members display a great degree of coordination in their points of view on evaluation and strategy. The results for total programmatic coherence support conventional wisdom with respect to parties like the PRI, which have demonstrated a high degree of flexibility in positioning their policies in the left-right spectrum.24 Since this oscillation has not led to a lack of support among the

Table 8.5

Total Programmatic Coherence

FMLN/El Salvador PRD/Dominican Republic EP/FA/Uruguay UCR/Argentina PRSC/Dominican Republic FSLN/Nicaragua ADN/Bolivia DP/Ecuador PLN/Costa Rica PC/Uruguay PPD/Chile PLH/Honduras ANR/Paraguay PDC/Chile PLC/Nicaragua MNR/Bolivia PLRA/Paraguay PRI/Mexico PUSC/Costa Rica PAN/Mexico PNH/Honduras MUPP-NP/Ecuador FRG/Guatemala PJ/Argentina PAN/Guatemala ARENA/El Salvador

n

Evaluation Component

Strategy Component

Programmatic Coherence

20 47 30 34 11 27 24 31 20 21 12 37 36 29 32 20 21 60 22 31 30 7 14 59 35 16

9.12 10.94 11.21 10.80 10.22 11.72 10.35 12.44 14.80 9.47 17.48 11.22 12.46 17.48 15.02 13.98 12.99 13.89 15.30 13.55 14.59 21.88 21.13 17.28 16.52 20.57

6.94 5.91 6.40 6.96 8.02 9.46 10.93 9.17 8.77 14.26 6.54 13.68 12.91 8.62 11.79 13.35 14.93 14.07 12.99 15.58 15.80 8.78 10.39 15.49 17.48 16.27

8.03 8.43 8.81 8.88 9.12 10.59 10.64 10.81 11.79 11.87 12.01 12.45 12.69 13.05 13.41 13.67 13.96 13.98 14.15 14.57 15.20 15.33 15.76 16.39 17.00 18.42

Source: Data based on PELA 1994–2005.

A-8

10/2/07

3:37 PM

186

Page 186

PARTY POLITICS AND IDEOLOGY

electorate, this party’s case is one of the most frequently cited as proof for the thesis that when a party is able to distribute benefits in a clientelistic manner, programmatic ideological consistency is not necessary in order to stay in power. In the same way, the measurement of coherence reinforces the general notion about the PJ, which is the fourth most incoherent party. For Steven Levitsky (2003), this situation has a positive connotation, since the PJ has shown itself to be, in his words, a party with programmatic flexibility, a trait that has allowed it to adapt to whatever new situations arise.25 The discussion of what level of coherence is desirable goes beyond the scope of this chapter. However, it can be observed that a high degree of incoherence can be just as damaging as an excessive amount of coherence. In addition, levels of coherence vary depending on which aspect of the party is under discussion. For example, a great degree of coherence will help a party be more efficient in its parliamentary function. In the electoral sphere, however, there is evidence to support the idea that the most incoherent, “catch-all” parties tend to be the most successful ones in the Latin American region.26 The information on the average coherence in the Chilean PDC is also not surprising. In this party, there is a long history of an increasingly polarized division between one group, which tends toward the right, and a center-left group. 27 This tendency is so marked in the Chilean Christian Democratic party that PPD has less internal dissent on economic issues than does the PDC, despite the fact that, because of the context in which it arose, the PPD is perceived as an instrumental party. In this sense, one argument used to explain the consensus in the positions of parties formed by left-wing militants in Chile is that it is a result of lessons learned in the democratic transition. The PPD was formed by these militants, who conceived of it at the time as a temporary substitute for the banned PS. The Salvadoran party ARENA and the Guatemalan PAN are among the parties that display the most disagreement in programmatic terms. The fact that these parties were sponsored by the entrepreneurial class suggests a hypothesis, which would need to be tested, about ties between programmatic incoherence and financial dependence on a source external to the party. The PAN was supported by both the sugar industry and wealthy Guatemalan families. ARENA, made up of traditional economic elites, elements from the urban middle class, and owners of small businesses, was the party through which the Salvadoran right broke with their tradition of functioning outside of democratic channels, and began voting.28 The existence of different worldviews within these parties could be related to the fact that party groups have access to independent resources. On the contrary, when a party depends on only one economic group, this situation is conducive to that party’s maintaining a stable, well-defined ideological profile over time.29 This analysis, which takes party members as the unit of measurement,

A-8

10/2/07

3:37 PM

Page 187

PROGRAMMATIC COHERENCE

187

indicates that variability exists in the programmatic and ideological structuring of parties, and that it occurs more frequently than is often supposed. The analysis supports recent arguments that issues are important in Latin American politics and that they are not dependent on clientelistic and personalistic networks.30 In fact, these data contradict the argument that clientelistic practices and the structuration of party competition around issues are incompatible.31 For example, the Dominican PRSC is one example of a party that maintains itself in power through corruption and clientelism, and at the same time displays high levels of programmatic coherence.32 Finally, the more coherent a party appears to be in programmatic issues, the higher the predictive power of its average positions are, since these positions are subscribed to by a high percentage of party members. Conversely, in an organization with a high degree of incoherence, there are no shared criteria; the average answers given by this party’s legislators will not be indicative of the attitudes of the other members of this party.

■ Political Parties and Legislators and the Debate on Statism-Market We have operationalized programmatic coherence as the similarity of positions between legislators of the same party in the evaluation of problems and in preferences for strategies to solve these problems. When measuring the degree of agreement around these issues, more attention has been paid to the structuration of parties around programmatic aspects than to the average positions. Nevertheless, together with a description of the degree of programmatic structuration, the preceding section of this chapter has given information on the substantive orientation of parties on programmatic issues. In this context, perhaps the most striking element is that of the averages in the strategy component of coherence, which indicate that some parties (the Uruguayan EP/FA and the Dominican PRD) are more interventionist than others (the Guatemalan PAN and the Mexican PAN). The binomial of interventionism/statism constitutes an important dividing line between parties in Latin America. We will now explore the relationship of this binomial to ideology and the ideological coherence of parties.33 In addition, we will address the degree to which the programmatic positions of parties on the statism-market debate are related to each other. The debate on the desirable level of state participation in the economy is so important that it constitutes a central element in the definition of the categories of “right” and “left.” Despite the fact that positions have become closer on this subject, it is one of the issues that has great potential to create polarization. Table 8.6 reveals a high relationship between legislators’ ideological self-placements and their placements on the scale of statism-market, which also figures as a question in the PELA survey. The positive character

A-8

Pearson Correlation Coefficients Between Several Key Variables in the Study

1.000

.149 .468 26 1.000

Primary Education

Secondary Education

Public Health

.568** .002 26 .501** .009 26 1.000

–.202 .323 26 .085 .679 26 .007 .973 26 1.000

–.145 .481 26 .199 .330 26 .072 .726 26 .528** .006 26 1.000

–.092 .655 26 –.052 .800 26 –.129 .530 26 .176 .391 26 .052 .800 26 1.000

Source: Data based on PELA 1994–2005. Notes: **The correlation is significant at the 0.01 level (two-tailed). * The correlation is significant at the 0.05 level (two-tailed).

Unemployment

Ideology

–.232 .254 26 –.003 .990 26 .113 .584 26 .014 .946 26 .103 .615 26 –.101 .624 26 1.000

–.032 .876 26 .015 .943 26 .087 .674 26 .047 .818 26 .108 .598 26 .209 .306 26 .519** .007 26 1.000

Statism / Market –.306 .146 24 .051 .811 24 –.001 .995 24 .047 .828 24 .105 .624 24 .114 .59 24 .525** .008 24 .713** .000 24 –.421* .040 24 1.000

Page 188

Unemployment Significance n Education Significance n Public health Significance n Primary education Significance n Secondary education Significance n Public health Significance n Unemployment Significance n Ideology Significance n GDP Significance n Statism/market

Public Health

3:37 PM

Unemployment Education

Correlates to:

188

Evaluation of:

10/2/07

Table 8.6

10/2/07

3:37 PM

Page 189

PROGRAMMATIC COHERENCE

189

of this correlation index indicates that the further to the right a legislator places him- or herself, the more that legislator tends to prefer that the economy be guided only by market criteria. Table 8.6 also allows us to see if the positions of the legislators on both scales (statism-market and left-right) are predictors of their programmatic positions. Specifically, correlations have been calculated on three issues— unemployment, education, and public health—for which there are data on legislators’ evaluation of importance as well as their proposals for solutions. These correlations show a relationship only between the statism-market and left-right scales and strategies for unemployment. There is no such relationship for the other two issues (primary and secondary education and public health). Nevertheless, if, instead of checking the individual relationship among these three issues on the statism-market scale, we check the aggregate average of all proposals for increasing public spending, a significant relationship is indeed found (see Figure 8.3). There is also a significant relationship between the average of answers on the degree of state intervention (see Figure 8.4) and the statism-market scale. Thus it can be said that the

Figure 8.3 Degree of Public Spending on Statism vs. Market, by Party

Increase

100 FMLN PRD FSL PRSC DP MUPP UCR PPD PDC

EPF

90

PLRA

PLC

Public Spending

PLN PJ

80 ANR

Decrease

A-8

MNR ADN FRG PLH

PNH ARENA

PUSC

PANGU

70 PC

60 30

40 Statism

50

60

70

80 Market

Source: Created by author based on PELA 1994–2005.

90

10/2/07

3:37 PM

190

Page 190

PARTY POLITICS AND IDEOLOGY Figure 8.4 Degree of State Role Preferred in Statism vs. Market, by Party 100 PPD

Strong

EPF PLN

PRD

UCR PRSC

90

PLRA

ANR

PLC

State Role

FMLN FSL

ADN

ARENA

PLRA PLH MNR PC PDC PUSC

MUPP

80

DP

PNH PANGU

PJ

None

A-8

70 30

40 Statism

50

60

70

80

90

Market

statism-market scale sums up the programmatic positions used to measure programmatic coherence. Table 8.7 shows individually the relationship of party ideology to programmatic coherence. Both programmatic coherence and the strategy component have a relationship to ideology,34 and thus coherence in solutions appears to be related to party ideology, with parties more to the right being more incoherent about what measures to adopt. The evaluation component of coherence is not associated with ideology. That is to say, the assigning of importance to problems, which is what was used for content in the measurement of coherence in this component, does not have ideology as one of its explicative variables. As explained at the beginning of this section, with the information available we can analyze whether or not the average positions of parties in the strategy component are influenced by their evaluation of problems. The analysis of correlations between the evaluation of these issues and the strategies that should be followed shows that a relationship cannot be established between the positions adopted with respect to both groups of issues

A-8

10/2/07

3:37 PM

Page 191

191

PROGRAMMATIC COHERENCE Table 8.7

Correlation of Ideological Coherence and Ideology with Programmatic Coherence and Its Components Ideological Coherence

Ideological coherence 1 Significance n 26 Ideology .555(**) Significance .003 n 26 Coherence in evaluation –.138 Significance .500 n 26 Coherence in strategy .311 Significance .121 n 26 Coherence in program .110 Significance .591 n 26

Ideological Coherence Placement in Evaluation .555(**) .003 26 1 26 .154 .453 26 .608(**) .001 26 .487(*) .012 26

–.138 .500 26 .154 .453 26 1 26 .223 .273 26 .783(**) .000 26

Coherence in Strategy

Coherence in Program

.311 .121 26 .608(**) .001 26 .223 .273 26 1

.110 .591 26 .487(*) .012 26 .783(**) .000 26 .781(**) .000 26 1

26 .781(**) .000 26

26

Source: Data based on PELA 1994–2005. Notes: ** The correlation is significant to the 0.01 level (two-tailed). * The correlation is significant to the 0.05 level (two-tailed).

(see Table 8.7). For example, a legislator’s assigning a high degree of importance to public health is not related to the way in which s/he defines the role of the state in public health or the ideal level of public spending on health issues. If there were a relationship between averages in the evaluation and strategy components, by knowing the importance a legislator attributed to an issue, we could predict the strategy that s/he would recommend.

■ Conclusion In this chapter, I have studied the programmatic structuring of the most successful parties in Latin America in the 1990s. The parties chosen for this study showed variable levels of programmatic coherence that, although levels are minimal in some cases, suggest certain shared interests on the programmatic level of party competition. This does not mean that I do not recognize the importance of other dynamics in the region, such as personalism and clientelism, which are compatible with programmatic competition and occasionally exert more influence on party competition. This study shows that party coherence allows one to distinguish among Latin American political parties. Once party coherence has been operationalized and measured, it is possible to analyze the influence of party and

A-8

10/2/07

3:37 PM

192

Page 192

PARTY POLITICS AND IDEOLOGY

nonparty factors on party coherence on programmatic issues. This chapter offers new evidence to explore the relationship between party coherence and party behavior. At a moment in which there is significant public dissatisfaction with the role of parties in the aggregation of interests and the articulation of political programs, this study shows that programmatic consensus is not absent from Latin American parties. However, it is still necessary to measure differences between this consensus and electoral programs. Reforms advocating internal party democracy should also address the political processes through which party programs are created. These reforms will bring party programs closer to the average party positions this chapter has focused upon.

■ Notes 1. Downs (1957). 2. Kirchheimer (1966). 3. Kitschelt (2001). 4. Goodman (1990); McDonald and Ruhl (1989); and Dix (1989). 5. May (1973) proposes his law of curvilinear disparity to explain these differences. In Latin America there are various studies that have addressed this question (PNUD, 2004). 6. Rosas (2001). 7. Ruiz Rodríguez (2007); Ruiz Rodríguez (2006); and Ruiz Rodríguez and García Montero (2003). 8. Schattschneider (1942) and De Swaan (1973). 9. Research on programmatic coherence was carried out in twenty-six Latin American parties in thirteen countries, by applying questionnaires that were generated from the database of the PELA (1994–2005). From every country, the two parties that had had the most electoral success in the decade of the 1990s were selected. The degree to which the parliamentary side of the political parties is articulated around similar visions of problems and their solutions will be evaluated. Given the fact that in the PELA the number of legislators interviewed from each party was based on the party’s presence in congress, choosing the most successful parties guaranteed a large enough N to be able to draw conclusions about the party as a whole. This is a synchronic strategy, since the calculation and interpretation of coherence are based on data gathered more or less simultaneously within parties from different countries. 10. Schedler (1995) and Mainwaring (1998). An outstanding study on legislators and their parliamentary groups was edited by Bowler, Farrel, and Katz (1999). In Latin America specifically, the following contributions stand out: Jones (2002); the study edited by Morgenstern and Nacif (2002); the studies based on data gathered on party members in the PPAL project, included in Alcántara and Freidenberg (2001); and various articles published using the data collected in the PELA. 11. Foweraker (1998) and Stokes (2001) both support this argument regarding the impact of programmatic parties on the governments in Latin America. 12. Downs (1957), Anderson (1968), and Huntington (1965) were the first authors to apply the concept of coherence to the study of political processes. Later works that incorporated this concept into the specific study of political parties are those of Panebianco (1988), Janda (1993), and Maor (1997).

A-8

10/2/07

3:37 PM

Page 193

PROGRAMMATIC COHERENCE

193

13. Carey (2002 and 1998); Tsebelis (1997); Figuereido and Limongi (1995); and Bowler, Farrel, and Katz (1999). 14. The studies of political culture are among the few that have recognized the importance of the attitudinal dimension in political dynamics. 15. Huntington (1965) and Anderson (1968). Given the theoretical links between attitudes and behavior, it would be relevant to examine the empiric relationship between coherence in attitudes and unity of behavior, such as cohesion in voting. 16. Previous works by Hawkins and Morgenstern (2003) have systematically addressed these issues based on the PELA data. Other works on this topic are Ruiz Rodríguez (2006), Ruiz Rodríguez (2003a), and Ruiz Rodríguez and García Montero (2003), where political parties are taken as units of analysis and a dimensional approach to party coherence is proposed. 17. Colomer and Escatel (2005); Mainwaring and Torcal (2005); Alcántara and Luna (2004); Llamazares and Sandell (2000); Coppedge (1998); and Alcántara (1995a). 18. The distinction between ideological coherence and programmatic coherence follows the classic opposition between ideology and program, popularized by Sani and Sartori (1983). “Ideology” refers to orientations related to emotions and feelings, and thus is not based on rational considerations. “Program,” on the other hand, refers to cognitive and value-based orientations. There are two main reasons for using this distinction. In the first place, ideology and program work differently in the structuration of competition. Ideology works on the symbolic level, and therefore is not usually directly addressed in the political arenas, although it does function as a referent for identity. The symbolic character of ideology can be seen in the possibility of using it in order to sum up programmatic stances. Program, on the other hand, structures competition on a concrete level. In the second place, each of these dimensions of the political has a different effect on the structuration of the party and the competition between parties; there can be parties displaying a high degree of ideological coherence but a low degree of programmatic coherence, and vice versa. This distinction does not overlook the fact that both dimensions can be related empirically and theoretically, above all in their positions rather than in the degree of coherence they give rise to. 19. Middlebrook (2000); Alcántara and Freidenberg (2001); Mainwaring and Scully (2003); and Alcántara (2004a). 20. This is an index that sums up the behavior of parties in the electoral arena based on four dimensions: the mean percentage of votes achieved in elections held during the decade of the 1990s; the number of times the party has won the office of president of the republic over the total number of presidential elections held; the number of times the party has won elections for provincial, state, or departmental governors over the total; and the number of times the party has won elections for mayor in cities with a population above one million, again over the total number possible. 21. If this procedure is followed, the value of the final coherence of the component or dimension will be comparable to the value of other dimensions or components that may have a different number of questions. 22. Alcántara and Freidenberg (2001). 23. Agosto and Cueto (2001). 24. Moreno (1999:143). 25. While in Levitsky’s study this programmatic zigzagging is seen as having positive connotations, Stokes (2001) describes this same trait as a case of program-

A-8

10/2/07

3:37 PM

194

Page 194

PARTY POLITICS AND IDEOLOGY

matic switch. Together with other cases, Stokes analyzes the case of the PJ and the consequences of its trajectory for representation and democracy. 26. Ruiz Rodríguez (2004). 27. Hunneus (2003); Hinzpeter and Lehman (1999). 28. Middlebrook (2000b). 29. On these issues, see Panebianco (1988) and Goodman (1990). 30. Alcántara (2004a) and Coppedge (1998), among others. 31. Rosas (2001) and Kitschelt (2000). 32. Agosto and Cueto (2001). 33. The ideological coherence was taken from Ruiz Rodríguez (2003a). 34. Ideology for each party was calculated based on the average of the ideological self-placement of its legislators, the placement legislators assign to their party, and the position that other legislators assign to that party.

A-9

10/2/07

3:38 PM

Page 195

9 The Parliamentary Left Salvador Martí i Puig and Salvador Santiuste Cué

■ New Scenarios and Their Impact on Political Forces of the Left In the past three decades, the Latin American left faced two historical processes of social change that redefined its political and ideological evolution. In fact, the profound nature of these events occasioned a restructuring of the left’s strategies and political practices. The first of these processes, which took place outside of the region, was the fall of what is known as “real socialism”: the rapid collapse of the political and economic system that arose in the October Revolution and that had dominated the whole of Western Europe since the end of World War II. The collapse of the socialist Soviet bloc, exemplified in the fall of the Berlin Wall in 1989 and the subsequent disappearance of the USSR in 1991, had a profound effect on the political left throughout the world, including, of course, the Latin American left. In fact, the Latin American left was especially affected since, with very few exceptions, it had extrapolated from or imitated the Cuban and Eastern European political and ideological-political models.1 Because of that, “this left” lost its main political and ideological “rearguard” and was deprived not only of its symbolic references, but also of external allies in its own particular struggles for the radical transformation of the societies in which it was located.2 But in addition to this debacle, the Latin American left, taken as a whole, had to face another process, one which also began in the 1980s, but which, unlike the one discussed above, was an internal process: the peaceful transition from authoritarian regimes to democratic ones, which took place in practically all of the countries of the continent.3 Thus, the armed forces’ retreat to their barracks in the Southern Cone and Andean countries and the collapse of reactionary, despotic regimes in Central America opened the door to democratization and ultimately to a process of political inclusion in which, for the first time in the history of many countries, the left was

195

A-9

10/2/07

3:38 PM

196

Page 196

PARTY POLITICS AND IDEOLOGY

accepted as a legitimate political player. This phenomenon made it necessary for the left to conceive of itself, and to organize, as an attractive and electorally viable political option. The condition for this inclusion, however, should not be underestimated; the left had to renounce revolution and to adapt itself to the demands of the electoral battlefield.4 As if this were not enough of a challenge, the adoption (to a greater or lesser degree) of a neoliberal model of economic development in all of the region’s countries has imposed limits on the left’s economic programs. The cutbacks implemented by governments in the 1980s and ’90s—and prescribed by international organizations—undercut the classic socioeconomic commitment of the left to strong, direct state intervention in the economy. Faced with the prevailing liberal economic model, the left has also found itself obliged to look for new political alternatives capable of checking or transforming the direction of the economies of the region’s countries. This has increased the existing difficulties of the left both in redefining its role as a relevant political actor within the new social and economic panorama and in searching for new ways to influence the relationship between the state and the market. In general terms, these are the basic coordinates of the “context of change” that the Latin American left has had to face from the end of the twentieth to the beginning of the twenty-first centuries. The objective of this chapter is to analyze the impact of this context of change on the most important left parties of Latin America, paying special attention to the specific attitudes and perceptions of their political elites with respect to two main points of discussion: (1) representative democracy as a political form of government, after having now had a number of years of experience in the political and economic competition that democratization has brought; and (2) the role that the state and the market should play in the drawing up of public policy, with the understanding that for the left, the capacity to create a framework of equal opportunity and greater social unity forms part of its thinking. Thus, this chapter analyzes the changes in the Latin American left with respect to two questions that are, without doubt, crucial to the definition of its most recent political evolution. These questions also reveal the extent to which the left has adapted to its new context; for this reason, in the next section we empirically examine both of these matters. This examination will be undertaken in light of the positions adopted by the main political forces of the left at the beginning of the twenty-first century. Although our analysis has an essentially ideographic nature, we maintain that it contributes to a general comparative understanding of the changes and continuities in the left parties active in Latin American democracies today. In a more specific sense, the assessment of the left’s characteristics, with respect to the aspects mentioned above, can also help us understand the problems, opportunities, and challenges that this left is now facing.

A-9

10/2/07

3:38 PM

Page 197

THE PARLIAMENTARY LEFT

197

■ The Sample to Be Studied: Twelve Left Parties in Nine Latin American Countries In order to more precisely specify the scope of this chapter, we must also clearly define the object of our study. In the first place, we analyze opinions expressed by parliamentarians of different political parties who locate themselves and their parties in the left wing of the classic “left-right” dimension (on a scale of 1 to 10), as can be observed in Figures 9.1 and 9.2. The sample is composed of the representatives of the lower houses of twelve left parties from nine Latin American countries. Specifically, the parties we examine are the Partido Revolucionario Democrático (PRD) of México; the Frente Farabundo Martí para la Liberación Nacional (FMLN) of El Salvador; the Frente Sandinista de Liberación Nacional (FSLN) of Nicaragua; the Movimiento V República (MRV) and the Movimiento al Socialismo (MAS) of Venezuela; Izquierda Democrática (ID) and the Movimiento Unidad Plurinacional Patchakutick–Nuevo País (MUPP-NP) of Ecuador; the Movimiento al Socialismo (MAS) of Bolivia; the Partido Aprista Peruano (PAP), the Partido Socialista (PS) and the Partido Por la Democracia (PPD) of Chile; and the Encuentro Progresista/Frente Amplio (EP/FA) of Uruguay.5 Obviously, this sample of parties leaves out a certain number of left parties present in their respective national parliaments. However, the ones included are those that presently have greater legislative representation,

Figure 9.1 Ideological Self-Placement of Parliamentarians (means)

Left

Source: Created by the authors based on PELA 1994–2005.

Right

A-9

10/2/07

3:38 PM

198

Page 198

PARTY POLITICS AND IDEOLOGY

Figure 9.2 Ideological Self-Placement of Political Parties (means)

Left

Right

Source: Created by the authors based on PELA 1994–2005.

with the exception of the Brazilian Partido dos Trabalhadores (PT), for which we lack data. In fact, absences from our sample are a result of certain parties’ more or less nominal nature, from a parliamentary point of view, or of their provisional nature, in electoral terms. It could be argued that the parties chosen do not sufficiently represent the immense field of the Latin American left as a whole, above all, those left forces situated outside of party frameworks. Nevertheless, we must emphasize that the sample is especially significant, at least in relation to the spectrum of Latin American political parties that openly declare themselves to be of the left. An examination of Figures 9.1 and 9.2 reveals that, although all parliamentarians place themselves on the left of the ideological scale, there are significant differences between parties.6 Thus, between the ideological position of the PRD (4.18) and that of the FSLN (1.69), there is a difference of more than two points (see Figure 9.2), similar, for example, to the difference between the MVT of Venezuela (4.13) and the FMLN of El Salvador (1.72). There are, therefore, different ideological positions, which range from “moderate” to more “extreme.” This is only to be expected, given the historical and programmatic specificities of each political actor. What needs to be examined, however, is whether or not these different ideological positions also show up as significant differences between parties with respect to the topics of analysis previously suggested. This will be discussed in the following pages.

A-9

10/2/07

3:38 PM

Page 199

THE PARLIAMENTARY LEFT

199

■ Perceptions of Democracy and Its Institutions Do Latin American left parties really believe in democracy? That is to say, do they believe that it constitutes the best form of government possible? According to the data in Table 9.1 the answer is yes; an overwhelming majority of parliamentarians from all left parties considers democracy to be the most preferable form of government. In only five out of the twelve parties analyzed are any respondents against this preference, and their numbers are relatively small; the highest of these figures are found in Chávez’s MVR and in the Bolivian party MAS, led by Evo Morales. Nevertheless, since these parties’ positions in favor of an authoritarian regime constitute only 10 percent of the opinions expressed, there are no significant statistical differences between these parties and all other parties analyzed. What do these data indicate? First of all, a qualitative leap forward with respect to the past, for it must be remembered that liberal democracy was not, historically, one of the objectives most eagerly sought after by the Latin American left. Quite the contrary; for large sectors of the left, democracy was always a secondary concern that, in the best of cases, constituted a necessary but temporary stage on the road to socialism. It was thus not surprising that many members of left parties supported the Cuban regime more wholeheartedly than any other in the Latin American subcontinent. As it has pointed out: “[D]uring many years, the majority of the left thoroughly and mistakenly scorned democracy, for they regarded it as an impostor.”7 The

Table 9.1

Preferences for Democracy in the Principal Parties of the Latin American Left (percentage) Parties

PRD FMLN FSLN MVR MAS(V) ID Democracy is preferable to any other form of government 92.3 In the context of a crisis, an authoritarian government may be preferable 7.7 DK/DA 0

MUPPNP MAS(B) PAP

PPD

PS

EP/FA

100

86.2

78.7

100

100

90.9

82.4

100

100

100

100

0 0

6.9 6.9

14.9 6.4

0 0

0 0

9.1 0

11.8 5.9

0 0

0 0

0 0

0 0

Source: Data based on PELA 1994–2005.

A-9

10/2/07

3:38 PM

200

Page 200

PARTY POLITICS AND IDEOLOGY

argument, in reality, was simple: formal democracy as it had developed in Latin America was a “bourgeois” democracy and thus a political instrument at the service not of the masses, but of local and foreign capital. Going by the above data, however, the left’s former mistrust of democracy seems to have disappeared. This is now true not only of more socialdemocratic or reformist parties, such as the PAP, the PPD, the ID, and, to a lesser degree, the PRD of Mexico, but also of parties of a more “classically socialist” type, such as the PS of Chile, the EP/FA of Uruguay, and the MAS of Venezuela.8 Table 9.1 reveals that democracy is supported as the most preferable form of government even by former armed groups—the case of the FMLN and the FSLN—that now act as political electoral platforms. This is also true, by the way, of the new party options of a “movimientista” nature; namely, the MAS of Bolivia and the MUPP-NP of Ecuador, whose orientations are social and indigenista. The agreement between parliamentarians from different parties about the desirability of democracy as a political regime demonstrates how, within the Latin American left, there is increasingly less room for orientations of an authoritarian or “conditionally authoritarian”9 type; that is to say, groups who would support an authoritarian regime, given certain political, economic, or social conditions. It is thus possible to state that democracy has acquired a legitimacy that it previously lacked within the left of Latin America. The fact, however, that democracy has been legitimized as a form of government with the party elites of the left does not mean that no differences exist as to how it should be carried out. Respondents’ agreement with democracy represents a change from the past, but one must go beyond this level to ascertain how respondents evaluate democracy’s institutions and fundamental mechanisms. It is one thing to support democracy as a form of government or as a general idea, as opposed to authoritarianism, but it is quite another to be necessarily in agreement with the basic ingredients essential to democracy’s workings. Because of this, parliamentarians were questioned on a key aspect of liberal democracy: electoral methods and the role of elections. Specifically, they were asked the following question: “In a context of pluralism and ample party competition, how would you rate your agreement with the following statement: ‘Elections are always the best way to express certain political preferences’? Are you very much in agreement, in agreement, somewhat in agreement, or not at all in agreement?” Various things can be highlighted about the answers obtained (see Table 9.2). In the first place, the majority of the parliamentarians agree that elections are the best way to express political preferences, with the only exception being the MAS of Bolivia. In second place, in six parties (PRD, MAS [V], PAP, PPD, PS, and EP/FA) this favorable opinion of elections is practically unanimous. The third fact to be noted, nonetheless, is that in six other parties (FMLN, FSLN, MVR, ID, MAS [B], and MUPP-NP) the percentage

A-9

10/2/07

3:38 PM

Page 201

THE PARLIAMENTARY LEFT

201

of those who oppose elections, while not a majority, is significant. What are the implications of all of this? That, although in principle elections are granted a fundamental role in democracy, a significant group of parliamentarians does not believe that elections are the only way in which to address citizens’ preferences and demands. That is to say, a reticence still exists toward the electoral component of liberal democracy. Could this reticence be due to a lack of faith in the electoral process? The data contained in Table 9.3, on the means of confidence that party legislators have in their country’s most recent elections, make it clear that this is not the case. All parties, except the PAP of Peru and the Sandinista party (the latter, to a lesser degree), display values that are relatively close to a high degree of confidence. What, then, is the explanation for the percentage of disagreement in some parties with respect to elections? One possible answer to this question would be to assume that “orthodox” groups still survive within the left and continue to oppose elections as a democratic façade that, in reality, functions as an instrument of traditional powerful elites. Electoral results that were poor or that prevented them from coming to power would reinforce this argument. However, another answer can be ventured. A large part of the left, particularly since the decade of the 1990s, has favored the development of an alternative to liberal democracy, an alternative structured precisely around the necessity of building a participatory democracy.10 Since the Porto Alegre Forum, the Latin American left, and especially the parties that claim to represent it, have enthusiastically embraced a model of democracy in which social movements would further a democratic political dynamic that would go beyond such formal mechanisms of representation as parties and elections. In this model, citizen par-

Table 9.2

Degree of Agreement with the Statement That Elections Are Always the Best Way to Express Political Preferences (percentage) Parties

PRD FMLN FSLN MVR MAS(V) ID Very much in agreement In agreement Somewhat in agreement Not at all in agreement DK/DA

MUPPNP MAS(B) PAP

PPD

EP/FA

55.6 44.4

70 23.3

46.2 46.2

31 51.7

37.9 41.4

40.4 40.4

58.3 41.7

40 33.3

54.5 18.2

5.9 41.2

7.7

17.2

17.2

14.9

0

26.7

18.2

41.2

0

0

0

3.3

0 0

3.4 0

4.3 0

0 0

0 0

9.1 0

11.8 0

0 0

0 0

0 0

3.3 0

0 0

Source: Data based on PELA 1994–2005.

72.7 100 27.3 0

PS

A-9

10/2/07

3:38 PM

202

Page 202

PARTY POLITICS AND IDEOLOGY

Table 9.3

Means of Confidence in Elections

Political Party PRD FMLN FSLN MVR MAS (V) ID MUPP-NP MAS (B) PAP PPD PS EP/FA

Mean

n

Standard Deviation

3.08 3.28 2.28 2.62 3.08 3.93 3.64 2.88 1.38 4.67 5.00 4.30

13 29 29 45 12 15 11 17 21 15 9 30

0.64 0.84 0.92 1.19 1.08 0.70 1.12 0.99 0.97 0.62 0.00 0.75

Source: Data based on PELA 1994–2005. Note: The scale of the variable on which the means are based goes from 1 (minimum confidence) to 5 (maximum confidence). In El Salvador, respondents were questioned about the elections from 1994 on and in Mexico, from 1989 on.

ticipation would increase through means of semidirect democracy and the participatory management of public affairs. Because of this, a significant portion of the left may not consider elections to be the only way for citizens to express their preferences. Along the same lines, the parliamentarians were posed the followed question about another basic and undoubtedly fundamental aspect of liberal democracy: “There are people who say that without parties there cannot be democracy. How would you rate your agreement with this statement: very much in agreement, in agreement, somewhat in agreement, not at all in agreement?” Once again, although to a much greater degree than with elections, members of various parties disagree with the statement that there cannot be democracy without political parties (see Table 9.4). Thus, in the PRD, the FMLN, the MVR, and the Bolivian MAS, the percentages are more against parties than for them. In fact, only the Chilean Partido Socialista is wholly in agreement with the idea of completely tying democracy to political parties. The data show, in this sense, a certain “antiparty” feeling, which is in line with the orientation expressed by the Latin American population in public opinion studies. These results may seem paradoxical, given the fact that the majority of the respondents are representatives of political parties. However, within the vision referred to previously of a more participatory and radical democracy, there is a drive to complement or even go beyond traditional party functions through the actions of nongovernmental organizations, such as social movements. It should not, therefore, be surprising that out of the twelve parties analyzed, at least

A-9

10/2/07

3:38 PM

Page 203

THE PARLIAMENTARY LEFT Table 9.4

203

Degree of Agreement with the Statement That Without Parties There Cannot Be Democracy (percentage) Parties

PRD FMLN FSLN MVR MAS(V) ID Very much in agreement In agreement Somewhat in agreement Not at all in agreement DK/DA

MUPPNP MAS(B) PAP

PPD

PS

EP/FA

30.8 15.4

20.7 17.2

58.6 20.7

29.8 17

75 8.3

64.3 7.1

45.5 27.3

17.6 17.6

72.7 18.2

46.7 40

77.8 22.2

53.3 20

53.8

48.3

3.4

25.5

8.3

14.3

18.2

35.3

0

13.3

0

13.3

0 0

13.8 0

17.2 0

25.5 2.1

8.3 0

14.3 0

9.1 0

29.4 0

9.1 0

0 0

0 0

13.3 0

Source: Data based on PELA 1994–2005.

seven of them do not include the word “party” in their official names. This does not necessarily mean that the left considers political parties to be obsolete institutions, but it does imply a different vision of representative democracy than the one that traditionally exists in the most consolidated democracies of the Western world. Given this state of affairs, it is also relevant to examine what party elites consider to be the “advantages of democracy,” for although all these elites describe themselves as favorable to this form of government, it is also likely that each holds a different belief of what its benefits and advantages are. In addition, examining this question requires studying a certain conception of what democracy is or should be. This can help us to understand the previous evaluation of elections and political parties by the legislators of each left party analyzed. Presumably, there will be variations in the subjective bases of democracy, since they contain at least two dimensions that have always been in dispute: freedom and equality.11 In Table 9.5, the main advantage alluded to by the majority of left parliamentarians is related to the foundations of liberal democracy by definition: the protection of individual freedom and rights. This is, in fact, the opinion expressed by the majority of the legislators of the FMLN, MVR, MAS (V), PAP, PPD, PS, and EP/FA. There is less unanimity about other possible advantages of democracy, with a great variety of answers among parliamentarians of different parties. Having said this, we stress that in some parties (this is the case of the FMLN, the FSLN, the MAS [V], and the MUPP-NP), a significant number of parliamentarians support advantages relative to economic growth and the redistribution of income, a fact that would highlight a more socialist conception of democracy. Similarly,

A-9

10/2/07

3:38 PM

204 Table 9.5

Page 204

PARTY POLITICS AND IDEOLOGY Main Advantage of a Democratic Regime (percentage) Parties

PRD FMLN FSLN MVR MAS(V) ID Economic growth 7.7 Protection of individual rights and liberties 15.4 Electing government authorities 7.7 Greater equality of opportunities 7.7 Respect for human rights and for minorities 0 Participating in decisions 30.8 A better distribution of income 0 Peaceful resolution of conflicts 7.7 Participating in politics through parties 23.1 DK/DA 0

MUPPNP MAS(B) PAP

PPD

PS

EP/FA

3.3

3.4

27.6

14.9

25

13.3

27.3

5.9

4.5

0

0

41.4

17.2

40.4

41.7

20

18.2

17.6

40.9

60

90

3.4

3.4

6.4

0

20

27.3

5.9

18.2

20

0

16.7

24.1

6.9

10.6

8.3

20

0

23.5

13.6

0

6.7

6.9

3.4

8.5

16.7

20

0

5.9

9.1

0

10

13.3

3.4

10.3

14.9

0

0

18.2

17.6

4.5

6.7

0

13.8

24.1

2.1

8.3

0

0

5.9

4.5

0

0

6.7

0

3.4

0

0

0

9.1

0

0

6.7

0

0

3.4 0

0 0

2.1 0

0 0

6.7 0

0 0

4.5 0

0 0

0 0

3.3 0

5.9 11.8

6.7

Source: Data based on PELA 1994–2005.

although in another direction, in various parties a high percentage of members give great value to participation in decisions (PRD, MVR, MUPP-NP). In any case, it can be seen that in almost all parties, the answers of their parliamentarians are quite unequally distributed. At this point, some conclusions can be ventured. First of all, we can say that a majority of representatives from the Latin American left accept democracy as the form of government to be followed and, thus, that authoritarian political options are openly rejected. Second, however, and without implying any contradiction of the first point, there is a certain anti-institutional emphasis on a part of the left that is still mistrustful of the mechanisms that structure liberal democracy, namely, elections and political par-

30

20

A-9

10/2/07

3:38 PM

Page 205

THE PARLIAMENTARY LEFT

205

ties themselves. Third, the degree of this anti-institutional emphasis varies significantly from party to party. It is relevant to point out that, in general, the left parties that adopt a more critical stance are those that have been most recently constituted as party organizations. This is in contrast to parties that date from further back, such as the Partido Socialist Chileno, the PPD (split off of the PS), the PAP, and the Frente Amplio from Uruguay. This aspect may be related to these more recently created parties being less attached to institutions and having less experience of them. In addition, having originated in movements and social groups that until recently were excluded from the political system, they may be still skeptical of the conventional methods of liberal democracy.

■ Perceptions of the Market What perception does the Latin American left have of the market? How does it conceive of the role of the state in economic regulation, policy design, and social welfare benefits? These questions are not unjustified, since economic and political patterns in the region have shifted radically during the past three decades. Until the mid-eighties, most Latin American governments adopted a strategy for economic development based on the substitution of importations (Bulmer-Thomas, 1994).12 This strategy, which required a considerable intervention of public powers in the industrialization process, was adopted without taking into account either the diverse nature of the regimes in question (whether they were representative democracies or corporatist or authoritarian regimes) or the specific policies each country followed to stimulate development. Thus, in this undertaking, in some countries the state maintained an important role in the economy through social-reformist policies (as in Costa Rica) or corporatist policies (as in Mexico, Argentina, and Brazil), while in others, monetarist and free exchange policies were established, as in El Salvador and Colombia.13 This model of development implied the articulation of a national-popular state, where the public was omnipresent in political life and the economy, despite the considerable weakness and lack of autonomy of institutions, which were continually at the mercy of corporative interests and actors of various kinds.14 In this context, where the relationship—and the borders—between the state and the market were not clearly defined, the intervention of politics in economic decisions (and vice versa) was easy to perceive. However, from the late eighties on, and especially in the decade of the nineties, due to the phenomenon of globalization, both the institutional and the socioeconomic coordinates of the region ended up coming together. In the 1990s, in economic terms, statist policies were abandoned in favor of a dynamic characterized by the adoption of neoliberal, fondomonetarista poli-

A-9

10/2/07

3:38 PM

206

Page 206

PARTY POLITICS AND IDEOLOGY

cies similar to those advocated by the IMF: opening of economies to foreign markets, decentralization of the administration of the state, and paring down social benefits to a minimum.15 Thus, this new context changes the way in which the institutions of Latin American states work together. This change is due not only to the wave of transitions to democratic regimes, but also to the increasing insertion of Latin American countries into an open and competitive international context. This new framework is characterized by (1) the intensification of commercial competition due to the appearance of new centers of production; (2) the proliferation and rearticulation of transnational companies; (3) rapid technological change—particularly with respect to communication and transportation technologies; (4) the appearance of a global financial system; (5) the existence of a monopolar international system, dominated by the United States; (6) the appearance of regional commercial alliances; and (7) the increase of inequalities within countries and between them. The relationship between the state and the market in Latin American countries has been redefined by the new context described above, to the point of eroding the very concept of sovereignty and, finally, that of politics itself.16 Politics, as it had been developed within the parameters of the “classic state society,” was characterized by (1) the establishing of processes of government and the drawing up of policies based on the division between, on one hand, a public-representative sphere anchored in the vote and, on the other, a private sphere based on consumption; (2) the rigidity of agendas defined almost exclusively in relation to the socioeconomic axis, through the logic of growth and (re)distribution; (3) party life conceived of as the expression of the homogenous interests of large social classes; and (4) the complete dominance of the nation-state in the regulation of social policies and benefits as well as in the bureaucratic production of these policies.17 According to various theoreticians, this new relationship between the state and the market meant the appearance of a new sociopolitical matrix, which has been called (in contrast to the state-centered model) the “neoliberal market-driven and multicentered matrix.”18 In this model, the ability to satisfy social demands is attributable not only to the action of political institutions, but also to the combined effect of the intervention of a wider group of actors. Thus, the way of doing politics is attributed neither to the “coordination by the market” (based on the supposed spontaneous harmony of economic exchanges) nor to the imperative capacity of the state (founded on hierarchical enforcement from a position of power). Rather, it is seen as the control of social processes through the constant interaction of all types of agents, where the state is just one of many actors, albeit a decisive and strategic one.19 Taking the above into account, the new way of doing politics not only depends on institutions and formal rules, but also springs from a continuous

A-9

10/2/07

3:38 PM

Page 207

THE PARLIAMENTARY LEFT

207

process of exchange and adjustment between collective and individual subjects, both public and private. This process, rather than depending on the existence of a directing center from which political power or other types of leadership are exercised, is configured as a network for the exchange of resources. Thus, when these parameters lose their validity, the definition and the final result of any policy are now no longer perceived as exclusively due to the government that proposes it and applies it with a specific end in mind, but rather as the consequence of the constant interaction among all kinds of social agents.20 However, this does not mean that the state and its politicians have been reduced to mere spectators since the beginning of the nineties. During these years, a double dynamic of privatizations of former public monopolies and the creation of regulating agencies—in principle of an independent and technical nature—was carried out throughout the whole region. This phenomenon, which obeyed the ideological leanings of the “Washington consensus,” was also spread by a “domino effect.”21 With these agencies installed, and with a sizeable part of the strategic economic sectors in the hands of private businesses, the discourse maintained by the left political class (now in power in some countries) at the beginning of the twenty-first century has begun to pay more attention to social inequalities. The left’s attitude has shown more disagreement with privatization policies and the work of the above-mentioned agencies, which, together with the conditions imposed by the IMF, appear to leave future elected officials with their hands tied when it comes to economic and social politics. Within this framework, the “problem of the left” becomes one of how to stop and reverse the process of social polarization present throughout the region, without falling into the temptation of using populist criteria of short-term value. Obviously, some analysts22 request maturity of this new left, arguing the need to achieve institutional balances that will make it possible to stimulate markets, achieve long-term commitments with economic investors, and give credibility to government policies that increase social welfare. The objective of this epigraph is to reveal the perceptions of the left parliamentary elites. To this end, interviews have emphasized: preferences between statism and the market (see Table 9.6); the type of taxes preferred (see Table 9.7); views on the privatization of public services (see Table 9.8); and the degree of public intervention necessary to control prices (see Table 9.9), create employment (see Table 9.10), cover basic necessities (see Table 9.11) and health care (see Table 9.12); and provide universal access to free university education (see Table 9.13). Table 9.6 gives us an idea of the moderate nature of the positions presently defended by left parliamentary elites of the region. The representatives of the Bolivian MAS, the PS, and the FA come down most heavily in

A-9

10/2/07

3:38 PM

208

Page 208

PARTY POLITICS AND IDEOLOGY

Table 9.6

Preferences Between Statism and the Market

Political Party PRD FMLN FSLN MVR MAS(V) ID MUPP-NP MAS (B) PAP PPD PS EP/FA

Mean

n

Standard Deviation

2.69 2.62 2.86 3.09 3.08 2.73 2.91 2.21 2.95 3.07 2.44 2.47

13 29 29 46 12 15 11 14 21 15 9 30

0.63 0.56 0.88 0.69 0.67 1.03 1.58 1.48 0.50 0.59 0.53 0.63

Source: Data based on PELA 1994–2005. Note: Scale of 1 to 5 where “1” represents statist and “5” economic policies.

favor of state intervention. They are also the only representatives who fall below the 2.5 mark—that is to say, more toward the state than the market— although the numbers indicate substantial moderation (no party is below 2). Conversely, the MVR, the Venezuelan MAS, and the PPD state that they are more oriented toward the market; and, indeed, their numbers are above 3. The rest of the groups (including the ex–guerrilla members of the FMLN and the FSLN and the indigenistas from the MUPP-NP) are situated between 2.5 and 3. The information contained in Tables 9.7 and 9.8 is closely related to the subject of the previous table: the types of taxes preferred and attitudes toward privatizations. The data indicate that when it comes to raising taxes,

Table 9.7

Preferences with Type of Taxes (percentage) Parties

PRD FMLN FSLN MVR MAS(V) ID Capital gains and income taxes 58.3 Consumption taxes 25 DK/DA 16.7

MUPPNP MAS(B) PAP

PPD

PS

EP/FA

86.2

89.7

44.7

50

42.9

63.6

17.6

54.5

80

55.6

80

6.9 6.8

10.3 0

46.8 8.5

50 0

42.9 14.3

18.2 18.2

17.6 64.7

31.8 13.6

13.3 6.7

33.3 11.1

13.3 6.7

Source: Data based on PELA 1994–2005.

A-9

10/2/07

3:38 PM

Page 209

THE PARLIAMENTARY LEFT Table 9.8

209

Preferences in the Privatization of Public Services (percentage) Parties

PRD FMLN FSLN MVR MAS(V) ID I would privatize all public services I would privatize only those that report a minimal profit I would privatize all except those that have the most impact I would leave things as they are I would not privatize any public service DK/DA

0

MUPPNP MAS(B) PAP

PPD

PS

EP/FA

0

0

0

8.3

0

0

0

0

0

0

0

20.7

51.7

29.8

8.3

46.7

36.4

11.8

36.4

26.7

44.4

26.7

75

20.7

24.1

48.9

75

46.7

54.5

5.9

54.5

60

44.4

36.7

0

0

0

0

0

0

0

16.7 0

58.6 0

24.1 0

10.6 10.6

8.3 0

6.7 0

9.1 0

8.3

0

11.8 70.6

0

6.7

11.1

0

9.1 0

6.7 0

0 0

16.7 20

Source: Data based on PELA 1994–2005.

almost all the representatives of the political groups (with the exception of the MVR and the tie between differing opinions within the Venezuelan MAS, the Bolivian MAS, and the ID) choose to tax capital gains and income over consumption (taxing consumption, by definition, is regressive taxation). In order of strength of preference for imposing capital gains and income taxes rather than consumer taxes: FSLN, FMLN, PPD, FA, MUPPNP, PRD, PAP, PS, MVR, MAS (V), ID, and MAS from Bolivia. In terms of the privatization of public services (see Table 9.8), the only party in which the majority supports the statement “I would not privatize any public service” is the FMLN (58.6 percent of the FMLN’s answers support this option). It is followed at a distance by the FSLN (24 percent) and the EP/FA and the PRD (16.7 percent). None of the parliamentary class of the rest of the left groups show any significant support for this position. The majority of the preferences, thus, are divided between those who would leave in public hands only those services that have a real importance for the

A-9

10/2/07

3:38 PM

210

Page 210

PARTY POLITICS AND IDEOLOGY

population (advocating for a normative criterion of coverage) and those who would privatize services that have minimal profits (privileging an economic criterion over principles of fairness, since public services that serve to redistribute resources or to maintain social cohesion are generally not profitable).23 The parliamentarians who opt for the first of the above-mentioned criteria (that of leaving services of real importance in the public domain) are those of the Venezuelan MAS (75 percent), the PRD (75 percent), the PPD (60 percent), the PAP (54 percent), the MUPP-NP (54 percent), and the MVR (48.9 percent). On the other hand, those who would be in favor of policies based on economic efficiency are the FSLN (51.7 percent), the ID (46.7 percent), the PS (44.4 percent), and the PAP and the MUPP-NP (both with 36.4 percent). The relatively high percentage of the Bolivian MAS—as well as that of the FA of Uruguay—who responded “DK/DA” should also be noted. Regarding the level of public intervention and the spheres in which this intervention should be effected, the data present parliamentarians’ opinions on the degree to which basic necessities should be covered, control of prices by the administration, the creation of employment, the public health care system, and access to free university education. In these tables, then, we have the opinions of parliamentarians of left groups on the classic spheres of public services—health care, aid to cover basic necessities, and higher education. We have, as well, their opinions on state intervention in the job market and in the market of goods and services. With respect to their views on the control of the market of goods and services (that is, consumer prices), Table 9.9 shows a dividing line between groups that support higher levels of intervention in order to control prices and those who believe that little or no control is necessary. The first group is made up of the parliamentarians of MUPP-NP (90.9 percent), FSLN (86.2 percent), FMLN (82.7 percent), ID (80 percent), PRD (77 percent), MVR (65 percent), EP/FA (62.4 percent), and MAS of Bolivia (58.8 percent). In Table 9.9

Level of State Intervention in Price Controls (percentage) Parties

PRD FMLN FSLN MVR MAS(V) ID High Fairly high Not much None DK/DA

46.2 30.8 15.4 7.7 0

24.1 58.6 17.2 0 0

34.5 51.7 10.3 3.4 0

29.8 36.2 29.8 2.1 2.1

Source: Data based on PELA 1994–2005.

0 33.3 33.3 33.3 0

60 20 20 0 0

MUPPNP MAS(B) PAP 54.5 36.4 0 9.1 0

29.4 29.4 23.5 0 17.6

4.5 31.8 54.5 9.1 0

PPD

PS

EP/FA

6.7 20 46.7 20 6.7

11.1 0 66.7 22.2 0

36.7 26.7 36.7 0 0

A-9

10/2/07

3:38 PM

Page 211

THE PARLIAMENTARY LEFT

211

the second group are the PS (88.9 percent), PPD (66.7 percent), the Venezuelan MAS (66.6 percent), and PAP (63.6 percent). Regarding regulation of the job market (see Table 9.10), in all groups there is a minimum of more than half in favor of higher levels of intervention (high and fairly high), even if the strength of this preference varies from one group to another. In groups such as the Bolivian MAS (88.2 percent), ID (86.7 percent), MUPP-NP (81.8 percent), FMLN (75.9 percent), and FSLN (72.4 percent), the great majority of their parliamentarians believe in a high level of intervention; however, this choice (that of “high”) is the majority opinion in all parties except the PAP and the PPD, where the majority is in favor of little intervention (45.5 percent and 46.7 percent, respectively). Finally, Tables 9.11, 9.12, and 9.13 lay out these parliamentarians’ opinions on the three above-mentioned spheres of public services: public services to cover basic necessities, health care, and higher education. In terms of levels of state intervention for basic necessities (see Table 9.11), more than two-thirds of the legislators of all groups agree that it should be high or fairly high, the only exception being the PAP, where 45.4 percent prefer not much or none. The same tendency can be observed to an even greater degree when state intervention in public health care is under discussion (see Table 9.12); in this case more than 90 percent agree that the level should be high or fairly high. Similarly, a minimum of two-thirds of those interviewed are in favor of a high or fairly high level of state intervention to guarantee free access to university education (see Table 9.13). With respect to this last point, however, in three parties the majority rate their support as fairly high rather than high: the PS (77.8 percent), the PAP (50 percent), and the FSLN (41.4 percent). In view of the data presented in the tables on socioeconomic questions (see Tables 9.6 to 9.13), we believe it is of value to apply a tree-analysis classification in order to group the parliamentarians studied, according to

Table 9.10

Level of State Intervention in Creating Employment (percentage) Parties

PRD FMLN FSLN MVR MAS(V) ID High Fairly high Not much None DK/DA

46.2 30.8 7.7 0 15.4

75.9 20.7 3.4 0 0

72.4 24.1 3.4 0 0

48.9 27.7 21.3 2.1 0

Source: Data based on PELA 1994–2005.

41.7 16.7 25 16.7 0

86.7 13.3 0 0 0

MUPPNP MAS(B) PAP 81.8 18.2 0 0 0

88.2 11.8 0 0 0

18.2 36.4 45.5 0 0

PPD

PS

EP/FA

26.7 26.7 46.7 0 0

44.4 33.3 22.2 0 0

46.7 40 13.3 0 0

A-9

10/2/07

3:38 PM

212 Table 9.11

Page 212

PARTY POLITICS AND IDEOLOGY Level of State Intervention to Cover the Basic Necessities of Citizens (percentage) Parties

PRD FMLN FSLN MVR MAS(V) ID High Fairly high Not much None DK/DA

46.2 38.5 7.7 7.7 0

62.1 34.5 34 0 0

58.6 34.5 6.9 0 0

66 19.1 10.6 4.3 0

66.7 8.3 8.3 16.7 0

73.3 20 0 6.7 0

MUPPNP MAS(B) PAP 45.5 45.5 9.1 0 0

64.7 29.4 0 0 5.9

22.7 31.8 40.9 4.5 0

PPD 60 40 0 0 0

PS

EP/FA

66.7 11.1 22.2 0 0

66.7 26.7 3.3 0 3.3

PS

EP/FA

Source: Data based on PELA 1994–2005.

Table 9.12

Degree of State Intervention in Covering Health Care (percentage) Parties

PRD FMLN FSLN MVR MAS(V) ID High Fairly high Not much None DK/DA

92.3 7.7 0 0 0

89.7 10.3 0 0 0

65.5 34.5 0 0 0

83 12.8 4.3 0 0

91.7 0 8.3 0 0

86.7 13.3 0 0 0

MUPPNP MAS(B) PAP 81.8 18.2 0 0 0

94.1 5.9 0 0 0

50 45.5 4.5 0 0

PPD 80 20 0 0 0

100 0 0 0 0

Source: Data based on PELA 1994–2005.

their perception of the state, the market, and their political party.24 In order to do so, we have constructed the variable “level of statism,” which is a synthesis of six variables that measure the level of state intervention in different spheres (prices, housing, employment, public health, education, and environment). This variable, considered as a dependent variable, incorporates two values, or “groups”: one that includes the least statist parliamentarians and one that includes the most statist. The results can be seen in Figure 9.3. The results in this last figure distinguish between three groups of parliamentarians, according to their level of statism and their political party: • Group 1: Almost three-quarters of the parliamentarians—all from the PRD, FSLN, and PAP—consider themselves to be “less statist.” • Group 2: In this group, there is a nearly equal division between “less statist” and “more statist” parliamentarians. This may explain why

93.3 6.7 0 0 0

A-9

10/2/07

3:38 PM

Page 213

THE PARLIAMENTARY LEFT Table 9.13

213

Level of State Intervention in Guaranteeing Free Access to University Education (percentage) Parties

PRD FMLN FSLN MVR MAS(V) ID High Fairly high Not much None DK/DA

76.9 23.1 0 0 0

51.7 41.4 6.9 0 0

37.9 41.4 20.7 0 0

66 10.6 21.3 2.1 0

58.3 8.3 16.7 16.7 0

60 20 20 0 0

MUPPNP MAS(B) PAP 54.5 27.3 18.2 0 0

70.6 17.6 11.8 0 0

27.3 50 18.2 4.5 0

PPD

PS

EP/FA

46.7 33.3 20 0 0

22.2 77.8 0 0 0

66.7 33.3 0 0 0

Source: Data based on PELA 1994–2005.

we find the parliamentarians of the majority of the parties in this group, which is composed of the FMLN, MRV, MAS (V), ID, MUPP-NP, MAS (B), PPD, and PS. • Group 3: This group of parliamentarians is composed exclusively of members of the EP/FA. The most significant aspect of this group is that 90 percent of its members are “statist politicians.” Before concluding, we must address the question of what the possible logic behind this classification (of “statist”) is. To be sure, at first glance one could respond that such a classification is due, on one hand, to the degree of radicalism of the groups in question or, on the other, to the internal logic of the individual countries to which these groups belong. Nevertheless, it is not irrelevant to include the variable (V1) of experience in power and (V2), the degree of success achieved in the task of governing, when applying the classic policies from the repertoire of the “old” left. After applying the two variables suggested, it can be affirmed that the presence of the FSLN and the PAP in group 1 is due to their having been in power (V1) and to their disastrous governmental performance as a result of applying statist policies (V2). It can also be stated that the presence of the EP/FA in group 3 can be explained by their lack of governmental experience at the time the interviews were carried out (absence of V1 and V2). Finally, it is difficult to interpret the reasons for group 2 containing such a large number and such a great variety of parties, for the parties present in this group vary from those that have never had governmental responsibilities— the cases of the FMLN, the MUPP-NP, the MAS (B), and the MAS (V)— and others that have, such as the MRV, the ID, the PPD, and the PS. Perhaps the only coherent answer is that the internal dynamics present in these parties’ countries, together with their ideological location and (for those that have been in power) their governmental experience, have led these groups

A-9

10/2/07

3:38 PM

214

Page 214

PARTY POLITICS AND IDEOLOGY

Figure 9.3 Classification Tree Showing Level of Statism by Political Party Statism Level Less statist: 54.10% More statist: 45.90% n = 244

Political Parties

PRD, FSLN, PAP Less statist: 74.60% More statist: 25.40% n = 63 1

FMLM, MRV, MAS (V), ID, MUPP-NP, MAS (B), PPD, PS Less statist: 54.30% More statist: 45.70% n = 151 2

EP/FA Less statist: 10.00% More statist: 90.00% n = 30 3

to a perception of the relationship between the public and private spheres that distances them from the radicalism expressed by group 3, but without the necessity of renouncing previous unsuccessful experiences. In summary, it can be stated that the opinions of the surveyed parliamentarians demonstrate that the left of the region is, on one hand, moderately in favor of state intervention, very favorably inclined toward the market, and relatively indulgent of the dynamic of privatization. On the other, it is an enthusiastic supporter of the role of the state as a regulator of the market (in the area of prices as well as employment) and as a provider of goods and services. From the above, we can conclude that the left’s leadership has evolved from populist and state-centered positions toward others more in agreement with a market-centered reality. Even so, there is present a proactive concept of direct state intervention in the elaboration of public policies that satisfy the demands of its social constituencies in their struggle to attain their social rights. This transformation could be said to demonstrate the left’s evolution toward a certain modernity and pragmatism regarding the state-market relationship. At the same time, however, the left maintains a marked utopianism regarding the state’s ability to intervene in the design of public policies and to function as a regulating force in the market; this is so precisely in a moment when such functions often lie in the hands of regulating agencies or, due to the recent waves of privatization, in the hands of private enterprise.

A-9

10/2/07

3:38 PM

Page 215

THE PARLIAMENTARY LEFT

215

■ Conclusion After analyzing the data presented here, we conclude by sketching out a tentative synthesis of the “new profile” of the political groups that represent the left in Latin America. Along these lines, it can be said that the presentday Latin American left is completely conscious and informed of the transformations that have taken place in today’s world. It is thus absurd to conceive of the left of this region as a political tendency anchored in an unmoving, statist past or nostalgic for “times gone by” that never existed in reality. Nevertheless, the position of the left is weak, due to tensions caused by: 1. Responding to the demands of its social constituencies (to be provided with goods and services below market prices and to the intervention of institutions in defense of their interests as salaried workers, small business owners, or unemployed workers). 2. Acting within the framework of pluralist and competitive institutions, which take for granted an unceasing dynamic of negotiations and compromises (be it in parliament or with regional governments). This process invariably diminishes the radical nature and the coherence of the left’s proposals. 3. Accepting the demands of economic groups and international agencies that limit the left’s ability to act and whose attitude toward the left affects the left’s chances to become a credible player in the electoral and political arenas. All in all, the left parties of the region have chosen their direction with their “eyes wide open,” and, within their ranks, a visionary, authoritarian approach that aims to radically change reality is almost completely absent. The left parties are completely democratic in form and in procedure (whether because of conviction or strategic necessity); they are respectful of legality and, in economic terms, are pragmatic and modern. They have left behind their former revolutionary aspirations as well as the desire to make a radical break with the past, and no longer press for economic solutions of a populist, corporatist, or interventionist type. At the same time, it may well be true that these left parties are not completely comfortable in a context that they often perceive as constantly demanding concessions and imposing limitations, for the new setting in which they must operate demands that they carry out an extremely complex balancing act. This is so not only in terms of ideology, but also with regard to the left’s approach to their hypothetical natural allies,25 with whom they must establish relationships without threatening a status quo that imposes considerable constrictions on them. It is perhaps because of this situation that present-day left political parties in Latin America have complex rela-

A-9

10/2/07

3:38 PM

216

Page 216

PARTY POLITICS AND IDEOLOGY

tionships with their traditional allies (unions, squatters, and neighborhood movements, or campesino movements). Seeing that the benefits from “their” party’s coming to power are no longer as direct or tangible as they were, these traditional allies have become progressively more independent of party organizations. Although they often support left political options when the latter are in the opposition, once left parties gain some power, the relationship can become conflicted and energy draining.

■ Notes 1. With the exception of a few social democratic parties (for example, the PLN of Costa Rica), the majority of the Latin American left organizations, whether they were parties, guerrillas, or social movements, took their inspiration from Marxism; their objective was the construction of an anticapitalist utopia (Gilberg, 1989). 2. It must not be forgotten that the pro-Soviet world (especially Cuba) lent the left not only ideological support but also logistical and material help. Above all, it provided a refuge for those fleeing for their lives from the bloody repression of the authoritarian regimes (or, in the cases of Colombia and Venezuela, from authoritarian policies) that populated the subcontinent. 3. This process of political change belongs to the phenomenon that Huntington (1991) denominated “the third wave of democratization.” Nevertheless, it must be pointed out that each country followed its own path. In some, the process of democratization was slow and progressive (Chile, Brazil); in others, abrupt (Argentina); in some it occurred within the framework of a war (El Salvador, Guatemala). There is one case of democratization through revolution (Nicaragua) and even through invasion (Panama), and one case is unclear (Panama). The exception was the Cuban regime, which did not crumble after the the wall fell, nor was it transformed by the wave of democratizations. Quite a bit has been written about this phenomenon. One seminal work, essential to the study of this case, is that of O’Donnell, Schmitter, and Whitehead (1986). A critical analysis of the literature written on this case can be found in Martí (2001). 4. This premise—that of the renunciation of the revolutionary dream—and its impact on the Latin American left during the first half of the 1990s is the central subject of the work of Castañeda (1995:15), who advocates for a “disarmed Utopia.” It is also true that in some sectors of the left, revolutionary symbols are still in use, but these are exceptions. However, it must also be pointed out that some nonparty organizations (who have some influence within parties) have emerged around the defense of anticapitalist, socially transforming utopian projects. The World Social Forum is one of the main contexts of these new demands. 5. The number of parliamentarians interviewed for each party is based on the party’s relative weight within the national congress. Those interviewed were thirteen PRD parliamentarians of the 2000–2003 legislature; twenty-nine FMLN parliamentarians of the 2003–2006 legislature, twenty-nine FSLN parliamentarians of the 2002–2006 legislature, forty-seven MVR and twelve MAS parliamentarians of the 2000–2005 legislature, fifteen ID and eleven MUPP-NP parliamentarians of the 2002–2006 legislature, seventeen PAP parliamentarians of the 2001–2006 legislature, nine PS and fifteen PPD parliamentarians of the 2002–2006 legislature; and thirty EP/FA parliamentarians of the 2000–2005 legislature. 6. This is confirmed by a variance analysis of the left-right placements

A-9

10/2/07

3:38 PM

Page 217

THE PARLIAMENTARY LEFT

217

assigned by parliamentarians to their respective parties. In this analysis, the squared eta (ε2) equals 0.451. This figure is higher than the one we obtain by comparing the means of ideological self-placement by political party (0.330). Both coefficients are statistically significant. 7. Castañeda (1995:388). 8. In reality, the EP/FA represents a coalition of parties that include, among others, the Partido Socialista from Uruguay. 9. Maravall (1995:102). 10. Nilsson (2003). 11. Held (2001). 12. There were some exceptions, with respect both to being the first to adopt neoliberal policies (as in the case of Chile) and to the late maintenance of statist policies, until the end of the eighties and the beginning of the nineties (as in the cases of Nicaragua and Brazil). 13. Gomà (1998). 14. Cavarozzi (2002). 15. Gomà and Jordana (2004). 16. Despite the fact the debate on globalization is relatively recent, the bibliography generated by this subject is extensive. Among the most influential studies are Held et al. (2001); Beck (1994 and 1998); Gray (2000); and Castells (1998). As far as the impact of this phenomenon on the Latin American left, see Eckstein and Wickham-Crowley, 2003. 17. A more extensive debate on this subject, linked to the appearance of new forms of collective action and their impact on public policies, can be found in the collective work edited by Ibarra, Martí, and Gomà (2002). 18. Garretón (2003:93–100). 19. Moureau (2003). 20. Rittberger (1993:1). 21. Mesaguer (2004). 22. Jordana (2005). 23. For a debate on the role of public services, redistribution, and social cohesion, see Adelantado (2000). 24. The CHAID algorithm is used to classify the cases. 25. The question of who currently supposedly constitutes the social base of the left in developing countries in general and in Latin America in particular has given rise to an interesting and polemical debate. Among the factors that complicate this question: (1) the economic transformation of the last few decades, which has led to an increase of the informal economy and of the employed who live in situations of precariousness; (2) the heavy migration to the United States and Europe; and (3) social disarticulation, as well as the increase of criminality within the most disadvantaged social sectors. These factors have been a heavy blow for the left, with its static vision of “el pueblo” and anchored, as it has traditionally been, in a discourse oriented toward salaried workers organized into unions.

A-9

10/2/07

3:38 PM

Page 218

A-10

10/2/07

3:38 PM

Page 219

10 The Parliamentary Right Manuel Alcántara Sáez and Iván Llamazares Valduvieco

STUDIES OF LATIN AMERICAN POLITICS HAVE OFTEN UNDERLINED THE

difficulty of interpreting such politics using concepts and analytical approaches developed in the United States and Western Europe. As is pointed out in other pieces in this volume, these caveats particularly concern the relevance of the left-right categories to the analyses of Latin American politics. On the other hand, different empirical and comparative analyses have shown that, despite important differences in this respect,1 the left-right vocabulary is often employed by Latin American politicians; left-right selfplacements help to predict party membership; and left-right orientations are connected to other political attitudes among both citizens and politicians.2 Using statistical techniques, but in a comparative analytical framework, this chapter explores the main characteristics of the legislators belonging to right parties. In particular, our work tries to determine which are the attitudinal, programmatic, and ideological factors that help to predict membership in right parties. We attempt to establish the degree to which legislators from the right are different from those belonging to other parties—a substantively important goal. The Latin American right has played crucial roles throughout the region’s history: in nineteenth-century conflicts concerning the Catholic Church; in twentieth-century social and economic struggles around distribution of resources; in clashes over political regimes (be they inclusive or excluding, democratic, hybrid, or authoritarian); and in current divisions over the economic and social politics of the democratic governments. From a comparative perspective, both synchronic and diachronic, it becomes crucial to establish where the main ideological and attitudinal differences between members of right and nonright political parties lie, and whether members of right parties display orientations clearly different from those of their main political competitors. These are the main goals of this analysis.

219

A-10

10/2/07

3:38 PM

220

Page 220

PARTY POLITICS AND IDEOLOGY

The development and evolution of conservative parties are linked to the emergence of religious, political, and socioeconomic conflicts.3 The articulation of these cleavages shaped the political articulation of political groups in Latin America. In the nineteenth century, sharp conflicts arose in several countries around the role of the Catholic Church.4 These conflicts resulted from a combination of international, domestic, and ecclesiastic developments.5 This was the context in which conservative, strongly pro-Catholic forces arose in Chile, Colombia, Paraguay, and Uruguay. At the beginning of the twentieth century, the growing importance of the United States in the region affected the position and policies of the Latin American right. In some cases, US political and economic interests contributed to the shaping of conservative politics in the region.6 But in some others, the emerging US influence would give rise to nationalist responses.7 From the 1930s to the 1960s, antidemocratic values flourished among the Latin American right, legitimating authoritarian regimes in Peru (under Odria), Argentina (under Uriburu and Onganía), Paraguay (under Stroessner), Guatemala (under Castillo Armas), and Panama.8 Also during the twentieth century, new right parties connected to the Catholic Church and to Catholic social and political activism arose. Christian-democratic sectors were strongly influenced by the French thinker Jacques Maritain. These were the cases of the Chilean PDC, initially called the Falange Nacional,9 the Mexican PAN,10 the Costa Rican PUSC,11 and the Venezuelan COPEI.12 In other Latin American countries, weak conflicts over religious issues combined with circumstances hostile to the development of stable and strong conservative parties. In Argentina, Brazil, and Bolivia, conservative sectors had to face very complex, ideologically ambivalent populist movements, such as Peronism, Varguism, and the MNR respectively. The conservative, authoritarian regimes that flourished between the 1960s and the 1980s were often based on the cooperation between conservative sectors and the military. Although conservative parties did not play an official role in these regimes, conservative politicians occupied key positions in them.13 After democratic transitions, the Latin American right adapted to the new polyarchic and competitive context. In some cases, it maintained old identities and organizations; in others, it coalesced around new labels, sometimes following new and temporary personalist leaderships. Overall, Latin American conservative parties have followed different historical paths and also operated in often-idiosyncratic party systems. Our study does not analyze the process by which the current Latin American conservative parties arose and developed, but explores the ways they articulate different positions in several ideological and programmatic arenas.

A-10

10/2/07

3:38 PM

Page 221

THE PARLIAMENTARY RIGHT

221

In the following section, we identify the right parties on which this chapter focuses. For that purpose, we used surveys covering the 2000 legislatures. Our initial analysis considered the following countries: Argentina, Bolivia, Chile, Colombia, Costa Rica, Ecuador, Guatemala, El Salvador, Honduras, Mexico, Nicaragua, Paraguay, the Dominican Republic, Peru, Uruguay, and Venezuela. However, the cases of Venezuela and Argentina had to be dropped because no political force in the survey met the requirements we define below. The third section examines the profiles of the Latin American right by conducting country-by-country statistical analyses. These analyses examine the degree to which four general dimensions (left-right self-placements, socioeconomic preferences, attitudes toward political issues, and cultural and moral attitudes) allow us to predict membership in right parties. The statistical technique used in these analyses is logistic regression. In this section we also compare the results from national analyses, and examine some possible factors affecting the cross-national differences revealed by our data.

■ The Identification of Right Parties We selected right parties based on the left-right self-placement of party members. More specifically, we considered “right parties” those parties that met three characteristics. First, the left-right self-placement of their legislators had to be above the mean point in the scale (5.5). Second, the party self-placement had to be above the national mean. Finally, all selected parties had to have at least ten legislators in our sample. The use of these criteria led to the selection of the political parties included in Table 10.1. No Argentine and Venezuelan party met all these criteria. In the case of Argentina, the reasons are more structural than conjunctural: conservative parties were rarely able to overcome territorial and ideological differences, and were often exposed to the strategic shifts of the main national parties, in particular, of the Peronist party. In Venezuela, the collapse and re-creation of the party system did not give rise to a stable and electorally successful articulation of conservative political groups. There are fifteen parties in our classification. Seven of them present the most rightist positions (above 7 on the left-right scale): the Chilean UDI, the Colombian PC, the Dominican PRSC, the Ecuadorian PSC, the Honduran PN, the Nicaraguan PLC, and the Salvadoran ARENA. Some of theses parties are more than a century old (the PC and the PN), while others are just a few decades old (like UDI, PLC, and ARENA). By contrast, the most centrist parties (below 6 on the scale) are the Uruguayan PN and the Paraguayan ANR, both of which are more than a century old.14

A-10

10/2/07

3:38 PM

222

Page 222

PARTY POLITICS AND IDEOLOGY

Table 10.1

Left-Right Self-Placement Among Legislators of Right Parties

Political Party ADN (Bolivia) n=24 RN (Chile) n=17 UDI (Chile) n=17 PC (Colombia) n=29 PUSC (Costa Rica) n=22 PRSC (Dominican Republic) n=11 ARENA (El Salvador) n=15 PSC (Ecuador) n=25 PAN (Guatemala) n=32 PN (Honduras) n=30 PAN (Mexico) n=31 PLC (Nicaragua) n=31 ANR (Paraguay) n=36 UN (Peru) n=12 PN (Uruguay) n=13

Left-Right Self-Placement Mean Std. Dev. Mean Std. Dev. Mean Std. Dev. Mean Std. Dev. Mean Std. Dev. Mean Std.Dev. Mean Std. Dev. Mean Std. Dev. Mean Std. Dev. Mean Std. Dev. Mean Std. Dev. Mean Std. Dev. Mean Std. Dev. Mean Std. Dev. Mean Std. Dev.

6.5 (2) 6.8 (1.03) 8.18 (1.24) 7.2 (1.26) 6.6 (1.33) 8 (1.41) 8.6 (1.24) 7.5 (1.48) 6.7 (1.68) 7.9 (1.48) 6.00 (0.86) 7.13 (1.91) 5.9 (1.76) 6.1 (1.31) 5.77 (0.83)

Source: Data based on PELA 1994–2005.

■ The Nature and Intensity of the Differences Between Right Parties and Their Party Competitors When examining the position of the right in the Latin American party systems, one of the main questions relates to the ideological and programmatic specificity of right parties. To what extent does the left-right dimension allow us to distinguish legislators from right parties from those of nonright parties? And, even more important, which are the programmatic and ideological issues that allow us to distinguish legislators of right parties from those of other political forces? In order to address these questions, we have conducted logistical regressions in each of these countries. In each of these regressions, we try to predict membership in right parties by using different types of independent variables.15 More specifically, we have taken as predictors the left-right self-placements of legislators; a group of variables that

A-10

10/2/07

3:38 PM

Page 223

THE PARLIAMENTARY RIGHT

223

measures the preferences regarding the economic role of the state; a group of variables that measures the cultural and moral orientations of legislators (for instance, regarding abortion and divorce); and also a group of variables regarding the importance of political issues (such as the relationships between governments and the military, corruption, and human rights). Table 10.2 shows the results of these logistical regressions. It includes information about the independent variables that are significant at least at the .10 level, and also the Cox and Snell R2, an indicator of the strength of the statistical association between our predictors and the dependent variable. The dependent variable in each model is membership in right parties.16 This statistical analysis has an inductive character. Instead of assessing the degree to which the same variables predict membership in the right in each country, we have included in our analysis only those variables that are linked to our dependent variable in a statistically significant manner. The reason for this is that our main interest lies in capturing the degree of differentiation between members and nonmembers of the right in each dimension (economic, cultural, political), irrespective of the specific variables we are including in each analysis. In the ideological dimension, right parties are most distinctive in Chile, Nicaragua, and El Salvador, and least different in Paraguay, Bolivia, Peru, and Mexico.17 In the socioeconomic dimension, the most different right parties are those of Chile, Mexico, Guatemala, Nicaragua, and Ecuador, whereas those of Colombia, El Salvador, Bolivia, the Dominican Republic, Honduras, Peru, and Paraguay are hardly distinguishable. Nicaragua, Mexico, Chile, and Colombia are the countries in which the right shows the most differentiated positions in the cultural dimension. With the exception of El Salvador, the remaining cases show very little degrees of differentiation in the cultural field. Finally, in the political field, the most differentiated parties are those of Chile, El Salvador, Nicaragua, and Guatemala. By contrast, the least differentiated parties are those of Uruguay, Costa Rica, Colombia, Peru, and Paraguay. Table 10.3 offers a summary of the importance of the levels of differentiation of the parties we selected. This summary is based on the Cox and Snell coefficients reported in Table 10.2. Differences between members of the right and of other parties were considered important when the Cox and Snell coefficient were above the Latin American mean in that dimension. This summary allows us to explore the main differences among the cases included in our analysis. Table 10.3 allows us to distinguish groups of countries that present extreme degrees of differentiation and, within each of these groups, to examine the degree to which countries with similar levels of differentiation share common traits. The cases with the highest numbers of differences are Chile, Nicaragua, and, to a lesser degree, El Salvador. Interestingly, these countries present quite different characteristics in areas

A-10 10/2/07

Logistic Regressions: Differentiation of Right Parties in the Left-Right Dimension, and on Socioeconomic, Cultural, and Political Issues Left-Right Differences

Issues helping to predict membership in right parties (significant at the 0.10 level) State reduction Price controls

Cultural Issues Cox and Snell R2

.36

Issues helping to predict membership in right parties (significant at the 0.10 level) Abortion Divorce

Political Issues Cox and Snell R2

.33

Mexico (PAN)

.16

Chile (RN/UDI)

.66

State reduction State providing jobs

.47

Divorce

.26

El Salvador (ARENA/PCN) Bolivia (ADN)

.53 .12

.11 .11

Abortion Abortion

.13 .09

Honduras (PN)

.34

State providing jobs Provide unemployment benefits State reduction (.01)

.14



.03

Colombia (PC)

.36

.04

Abortion

Costa Rica (PUSC)

.27

Provide secondary education State reduction Price controls

.23



Issues helping to predict membership in right parties (significant at the 0.10 level) Crime Democratization Relations with the military Conflicts among state institutions Political violence Crime Relations with the military Corruption

Cox and Snell R2

.15

.51

Human Rights Democratization Relations with the military Conflicts among state institutions Corruption

.34 .13

.19



.07

.03



.05

.23

(continues)

Page 224

Cox and Snell R2

Socioeconomic Issues

3:38 PM

Table 10.2

A-10 10/2/07

Table 10.2

continued Socioeconomic Issues

Cultural Issues

.64

State reduction

.29

Guatemala (PAN)

.25

Ecuador (PSC)

.31

Dominican Republic (PRSC)

.36

Paraguay (ANR)

.02

Peru (UN)

.13

Uruguay (PN)

.25

State reduction Provide secondary education State reduction Price controls Provide secondary education Provide unemployment benefits Price controls Provide housing Provide social security Meet basic needs State reduction Provide primary education

Mean Std. Dev.

.31 .19

.32

Abortion Divorce Abortion

.27

Political Issues .38

.32

.07

Corruption Political violence Corruption

.30

Abortion

.05

Relations with the military

.09

.13



.00

Relations with the military

.21

.16



.03



.09

.14



.03



.11

.24



.06



.03

.22 .22

Source: Data based on PELA 1994–2005. Note: Cox and Snell R2 higher than the Latin American mean in bold numbers.

.12 .12

.19 .14

Page 225

Nicaragua (PLC)

3:38 PM

Left-Right Differences

A-10

10/2/07

3:38 PM

226

Page 226

PARTY POLITICS AND IDEOLOGY

Table 10.3

Summary of the Levels of Differentiation Between Members of Right and Nonright Parties

Chile, Nicaragua El Salvador Mexico Guatemala, Ecuador Honduras Colombia Dominican Republic Uruguay, Costa Rica Bolivia, Paraguay, Peru

Left-Right

Economy

Culture

Politics

Number

+ + – – + + + – –

+ – + + – – – + –

+ + + – – + – – –

+ + – + + – – – –

4 3 2 2 2 2 1 1 0

Source: Data based on PELA 1994–2005.

as relevant to this analysis as their levels of economic development, socioeconomic structures, party systems, quality of democratic institutions, and experiences with democracy during the twentieth century. Technically speaking, a comparison centered around these three cases would follow a most-different comparative design, separating Chile from the two Central American cases. What these countries share are experiences of sharp political conflicts in which economic, cultural, and political divides reinforced each other. All these countries also experienced polarizing political regimes (conservative or revolutionary). Finally, these polarizing political regimes presented strong links with broad social coalitions. It is quite likely that these experiences reinforced the distinctive features of the right, as well as of other political groups. The fact that the Salvadoran right presents lower levels of differentiation could be accounted for by several factors. In the first place, due to the selection criteria we used, we left out of our selection parties that, in ideological and attitudinal terms, could share many important features with ARENA; that is particularly the case of the PCN. This could have lowered the degree of differentiation of the ARENA members. Besides, it is possible that socioeconomic, political, and cultural divisions did not reinforce each other in such a stark manner in El Salvador. In particular, the existence of a Christian-democratic political party both close to and different from the more conservative political groups could help explain the fact that the Salvadoran right is not so different in the economic dimension. Those cases situated in the lower end of the table do not show any kind of differentiation in the dimensions we examined here. With the partial exception of Bolivia, these countries had very limited experiences with democracy. Besides, although these regimes suffered authoritarian regimes, nondemocratic experiences were not linked to broad social coalitions based

A-10

10/2/07

3:38 PM

Page 227

THE PARLIAMENTARY RIGHT

227

on a combination of socioeconomic, cultural, and political issues. Economically, these countries are characterized by their low (Bolivia and Paraguay) or middle-low (Peru) levels of per capita income. Some other characteristics of the parties’ party systems may also be important for this analysis. In the first place, their party systems are highly volatile (above the Latin American average in all cases). And, perhaps more importantly, the existence of centrist parties, very close to those we selected in this analysis (like the MNR in Bolivia, Peru Posible in Peru, and the PLRA and the PEN in Paraguay), may have also helped limit the degree of differentiation of these right parties. Finally, a comparison between these countries and Nicaragua and El Salvador, similar in several economic and social respects, reveals the importance of the regime trajectories of these countries, as well as of the intensity of the social and political confrontations these countries experienced. The low number of countries makes it impossible to develop a multivariate analysis on the social and political conditionants of the different patterns of differentiation revealed by our analysis. It is possible, however, to examine the bivariate associations between the indicators we considered above and several potentially relevant political (electoral volatility, years of democracy, effective number of parties) and economic (per capita GDP) independent variables. Interestingly, ideological differences present stronger levels of association with political and, to a lesser degree, cultural variables. By contrast, the association with ideological and economic differences is not significant at the .05 level. The strength of ideological differences is also connected to the levels of electoral volatility. This low level of statistical association between economic and ideological differences does not indicate that economic differences are not important or that they are not connected to other types of differences. Economic differences are significantly associated with political differences, and one reason for this might lie in the connection between strong socioeconomic conflicts and polarizing political regimes with strong social linkages. In turn, political conflicts and divisions seem to be connected with ideological differences. The fact that economic differences are not directly related to ideological divisions might result from the role of other, possibly crucial economic and political factors. As shown in Table 10.4, there is a statistically significant association between the level of economic differentiation and per capita income. That, in spite of the fact that two countries with comparatively high per capita income (Costa Rica and Uruguay) present characteristics that are adverse to the existence of strong differences between members of the right and other parties. These adverse circumstances lie in the centripetal character of the Costa Rican party system and in the existence of two traditional parties that

A-10

10/2/07

3:38 PM

228 Table 10.4

Ideological Economic Cultural Political

Page 228

PARTY POLITICS AND IDEOLOGY Pearson Correlation Coefficients Between the Cox and Snell R2 and Other Political and Economic National Characteristics (14 cases)

Economic

Cultural

Political

.31 (.28) —

.53* (.05) .46 (.10) —

.73** (.003) .57* (.03) .45 (.11) —

.46 (.10) .57* (.03)

.45 (.11)

Per Capita GDP

Electoral Volatility

–.07 (.82) .59* (.03) .34 (.24) .01 (.98)

–.62* (.02) –.07 (.81) –.16 (.59) –.35 (.22)

Effective Number of Years of Parliamentary Democracy Parties .11 (.70) .21 (.48) –.28 (.32) –.15 (.61)

–.03 (.93) .21 (.48) –.18 (.55) .29 (.32)

Source: Data based on PELA 1994–2005. Note: Coefficients significant at the .05 level in bold numbers.

are close to each other (the PC and PN) in Uruguay. The association between economic development and the economic differentiation of the right is consistent with the expectation that economic development will facilitate the articulation and coordination of political groups around different alternatives of economic policy. Now, as previously stated, nondemocratic regimes based on broad social coalitions would also be linked to strong social and political conflicts. For that reason, levels of economic differences between the right and other forces would be connected to both levels of economic development and to political divisions. By contrast, political divisions and trajectories would not be connected to economic development. Finally, another dimension potentially connected to the preferences of Latin American conservative legislators relates to the relationships with the United States. We mentioned previously in this chapter how US influence contributed to shaping the position of the right in party systems in Latin America. One of the questions in this survey asks what importance legislators assign to the relationships between their countries and the United States. Table 10.5 shows the Pearson correlation coefficients between membership in right parties and the importance assigned to the relationships with the United States. Correlation coefficients are statistically significant in just three countries: Nicaragua, Colombia, and Mexico. The weakness of this association in the remaining countries could be connected to the fact that this question has both descriptive and prescriptive components. Legislators might be taking into account not only their ideological preferences, but also the hard facts of US influence in the region.18 The special position of Mexico, Colombia, and Nicaragua could be explained by the specific relationships between the United States and these three

A-10

10/2/07

3:38 PM

Page 229

THE PARLIAMENTARY RIGHT Table 10.5

229

Pearson Correlation Index Between Membership in Right Parties and the Importance Governments Should Assign to the US in Foreign Policy

Country Mexico (n=116) Chile (n=82) El Salvador (n=44) Bolivia (n=90) Honduras (n=67) Colombia (n=77) Costa Rica (n=45) Nicaragua (n=59) Guatemala (n=53) Paraguay (n=64) Ecuador (n=97) Dominican Republic (n=96) Peru (n=66) Uruguay (n=61)

Importance of the US Pearson correlation Sig. (two-tailed) Pearson correlation Sig. (two-tailed) Pearson correlation Sig. (two-tailed) Pearson correlation Sig. (two-tailed) Pearson correlation Sig. (two-tailed) Pearson correlation Sig. (two-tailed) Pearson correlation Sig. (two-tailed) Pearson correlation Sig. (two-tailed) Pearson correlation Sig. (two-tailed) Pearson correlation Sig. (two-tailed) Pearson correlation Sig. (two-tailed) Pearson correlation Sig. (two-tailed) Pearson correlation Sig. (two-tailed) Pearson correlation Sig. (two-tailed)

.198(*) .033 .044 .695 .078 .617 .154 .148 .220 .073 .237(*) .038 .126 .409 .321(*) .013 .261 .059 –.027 .832 .120 .240 .151 .141 .029 .817 .051 .697

Source: Data based on PELA 1994–2005. Note: * Countries where correlation coefficients are statistically significant.

countries: in Mexico, proximity to the United States and the importance of the North American Free Trade Agreement; in Colombia, the strong alliance between this country’s administration and that of the US administrations in the recent years; and in Nicaragua, the legacies of US intervention and civil war.

■ Conclusion This exploration has allowed us to identify the main right parties in fourteen Latin American countries, as well as to analyze the main ideological and attitudinal characteristics of their legislators. The results of our analyses show, in the first place, the diversity of the Latin American right. Conservative parties present quite different levels of differentiation with

A-10

10/2/07

3:38 PM

230

Page 230

PARTY POLITICS AND IDEOLOGY

their competitors, and articulate themselves around issues that vary from country to country. Whereas in Chile, Nicaragua, and El Salvador members of right parties are clearly distinguishable from those of other political forces in most attitudinal dimensions, in Paraguay, Peru, and Bolivia degrees of differentiation are lower in all the dimensions we analyzed in this chapter. The absence of a clearly distinguishable right coincides with the low institutionalization of the party systems of these countries.19 Comparisons among these groups of cases (both among those cases in which our dependent variables present similar values and among those countries that, although similar in some of their plausibly conditioning factors, are different in the values of our dependent variables) allow us to point out the importance of several explanatory factors. Critical among them is whether countries experienced polarizing, often openly authoritarian political regimes, which were also connected to conflicts in which social, political, and economic divisions reinforced each other. Our comparative and statistical analyses also underscore the influence of economic development on levels of economic differentiation and the role that the major centrist parties can play in the articulation and differentiation of the political right. In all these respects, the modest findings of this exploratory analysis can pave the way for both in-depth, comparative studies and more systematic statistical analyses on the factors affecting the quite diverse situations of the right in Latin American party systems.

■ Notes 1. Coppedge (1998). 2. Alcántara (2004a); Colomer and Escatel (2005). 3. On the emergence and historical trajectories of Latin American conservative parties, see Middlebrook (2000b), Di Tella (1993), and Gibson (1996). 4. These conflicts followed previous intra-elite divisions regarding the Spanish colonial past. Traces of them can be found, for instance, in the denomination of Colombian conservatives as “godos” (“goths”), a term used in the past to refer to peninsular Spaniards. 5. Such as the 1848 European revolutions and the Catholic Church’s reaction against the threats posed by liberalism and social revolution, and in defense of its autonomy and social influence. 6. That was particularly the case of Central America and the Caribbean. Following US military interventions (Gil, 1975), sultanistic regimes arose in Nicaragua and the Dominican Republic. These regimes also gave rise to political parties, such as the Nicaraguan Liberal Party and the Dominican Trujillista Party. In Honduras, the rise of the two main political parties was also influenced by the US presence in the region; in this case, by the large US banana companies. In El Salvador, the development of the conservative PCN followed a different, more autonomous path, resulting from the alliances between economic and political oligarchies and the military. On these two countries, see Alcántara and Freidenberg (2001). 7. An instance of this type can be found in Leopoldo Lugones’s speech “La hora de la espada.” See Iturrieta (1994:39).

A-10

10/2/07

3:38 PM

Page 231

THE PARLIAMENTARY RIGHT

231

8. In some of these cases, Spanish Francoism exerted an important ideological influence. See Delgado Gómez-Escalonilla (1988). 9. The National Falange was created out of the National Movement of the Conservative Youth by young people influenced by Catholic ideas on social issues. Among the founders of this group were Bernardo Leighton, Eduardo Frei Montalba, Manuel Garretón, and Radomiro Tomic. In 1957, the Falange and conservative Christian democrats created the PDC (Grayson, 1968). 10. The PAN was created in 1939 under the leadership of Manuel Gómez Morín, former president of the Catholic Association of the Mexican Youth (Reveles, 2003:24). See also Loaeza (1999:106 and the following) on the relationships between the Catholic Church and the PAN. 11. On the influence exerted by Catholic social thought, see Alfaro Salas (2001:72). 12. Although COPEI follows Christian-democratic ideas, it also presents itself as a party independent from the Catholic Church (Combellas, 1985:45). The centrist positions adopted by COPEI were followed by other Latin American Christian-democratic parties in the 1950s (Mainwaring and Scully, 2003). 13. These were the cases of the authoritarian regimes in Argentina, Chile, El Salvador, Brazil, and Uruguay. See Beltrán (1970) and Loveman and Davies (1989). 14. On the formation of these parties, see Alcántara (2004a). 15. The dependent variable in these analyses takes the value of 1 when legislators belong to right parties, and of 0 when they do not belong to right parties. 16. The socioeconomic variables we include in our analysis are the following ones: “Nowadays the role of the state is under discussion. In that respect, could you tell me if you are more in agreement or more in disagreement with each of the following sentences? ‘The state should intervene as little as possible in society, and let private initiative meet the needs of the citizens.’ I would also like to know your opinion on several traditional functions of the state. Thinking in general terms, could you tell me for each of the following items to what degree the state should intervene: a lot, a fair amount, little or none at all? Price controls; Guarantee universal and free primary education; Provide housing to citizens; Provide jobs to those who want to work; Provide universal social security; Provide universal and free secondary education; Provide unemployment benefits; Protect the environment; Meet the basic needs of all citizens.” Cultural and religious variables were the following: “What is your personal opinion on divorce?”; “What is your personal opinion on abortion?”; “Could you tell me if you are more in favor of the presence of Christian values in politics or of secular ones?” The political variables we included in our analysis were the following ones: “I am going to quote several problems which are common to many countries. Could you tell me which degree of importance you assign to each one of them in your country today: a lot, a fair amount, little or none? Insecurity and crime; Education; Democratization of public life and institutions; Human rights or rights of ethnic and cultural minorities; Relationships with the military; Conflicts among state institutions; Corruption; Drug traffic; Political violence.” 17. On the levels of ideological polarization in these countries, see Alcántara (2004b). 18. The association between left-right self-placements and the same variable renders coefficients significant at the 0.05 level also in Mexico and Nicaragua, but not in Colombia. The coefficient for this correlation is also significant in Uruguay. 19. Thus, Peru, Bolivia, and Paraguay are the countries with the lowest levels of party system institutionalization in Mainwaring and Scully’s study on this topic (1995:17).

A-10

10/2/07

3:38 PM

Page 232

A-11

10/2/07

3:39 PM

Page 233

PART 3 Conclusion

A-11

10/2/07

3:39 PM

Page 234

A-11

10/2/07

3:39 PM

Page 235

11 Legislative Careers: Does Quality Matter? María del Mar Martínez Rosón

THERE ARE MANY REASONS FOR UNDERTAKING A STUDY OF PARLIA-

mentary careers,1 despite the fact this kind of research is difficult; it is most productive when legislators remain in congress for decades, but does not tend to be very useful when they stay for only a few months.2 Nevertheless, there are convincing reasons to study parliamentary careers, even in the latter case. First of all, although the average duration of Latin American legislators’ stay in congress is far from decades, there is a core group—whose size varies from country to country—with longer careers. This group should be studied, since these legislators tend to take on important positions in congress. In the second place, ways must be found to overcome the difficulties generated by this situation. Very little is presently known about the expertise of Latin American legislators. For instance, how much experience do they have? Where does it comes from? How is it obtained? Although studies on political careers have taken on a certain importance in other countries and regions, such research concerning Latin America is very scarce; only recently has this line of study begun to open up. The centrality that the relationship between the executive and the legislative branches has had in Latin America means that we have very little information about what type of representatives make them up and in what proportion.3 Only a few studies have been done, and the majority of them have to do with the case of Brazil.4 In Latin America, legislators with previous legislative experience range from 61.4 percent in Chile to 7.8 percent in Costa Rica, but in no country is the average stay in congress greater than two legislatures. In these situations, where the average representative has not made a career of being a legislator, how does the house take advantage of individuals’ strengths and experiences? What are these strengths, and where are these experiences gained? As stated by Nancy Burns, Laura Evans, Gerald Gamm, and Corrine McConnaughy (2004) in their research on state houses in the

235

A-11

10/2/07

3:39 PM

236

Page 236

CONCLUSION

United States, the importance of legislative expertise is generally underestimated since new legislators draw on experiences from other contexts, which has a positive effect on their work in the legislature. Although answering all these questions is beyond the scope of this chapter, these issues need to be addressed in the Latin American context. As we will show, information on careers can be used as an explanatory variable in different contexts; for example, with respect to voting. Although other information, such as ideology or party affiliation, is important in elections, voters also use information about legislators’ careers.5 In certain circumstances, such as with open-list electoral systems or in small districts, voters use this information to choose between mediocre candidates and quality candidates. To analyze this issue, the chapter first reviews the literature on parliamentary careers and their quality. The sociopolitical and sociodemographic characteristics of Latin American legislators from the past ten years are described and classified based on a typology created to reflect quality. In this context, “quality” is understood to be a group of personal skills that give the legislator the ability and competence needed in order to carry out his or her duties. The second part of the chapter specifies relationships of dependence or independence between the quality of legislators and several political and institutional factors. Finally, the relationship between the quality of representatives and voting is analyzed.

■ Parliamentary Careers: What Does “Quality” Mean? One angle from which it is possible to study parliaments is legislators’ careers. There are many different ways of approaching this subject. John Hibbing (1999) identifies four objectives that justify studying parliamentary careers: (1) the literature that studies legislators, their reasons for running for office, the recruitment process, and the legislators’ reasons for leaving parliament or not running for office again; (2) improving knowledge on parliaments, above all on their level of institutionalization; (3) improving understanding of the sociopolitical system of which congress forms a part. Knowledge is sought on the sociodemographic characteristics of legislators, as well as on activities they were involved in before or after government service. Their level of representation of the population is studied, or their position in parliament in the structure of political opportunities; that is to say, in relation to other institutions. (4) Finally, a small part of the literature on parliamentary careers consists of comparative studies that seek to encounter generalities that describe different sociopolitical systems. Despite the precision of these four objectives, it is not clear what exactly is meant when “parliamentary careers” are spoken of. Most scholars use this term to refer to previous occupations, the number of terms that a representative remains in parliament, the drive for reelection, or the holding of

A-11

10/2/07

3:39 PM

Page 237

LEGISLATIVE CAREERS

237

other political offices. The concept of parliamentary career farthest from this vision is that used by Hibbing himself (1991), who approaches the study of career within the North American congress by focusing on all those activities that could be considered to form part of a parliamentary career: the race for reelection, the search for formal positions within the institution, roll-call votes, legislative activities, and activities carried out in the electoral district. On the other hand, there are two basic ways of understanding the professional level, or quality, of parliamentary careers. The first, and most frequently encountered, identifies a group of normative characteristics that are desirable for individuals or parliaments. A characteristic from this group is assigned a certain number of points (which may be the same for all characteristics or may not), and the sum of these points defines the level of competence, or quality.6 The second method, which attempts to come to a flexible definition, is used less than the first, more rigid method. It consists of defining all those elements that lead to a candidate’s electoral success as elements of quality. In this approach, the definition of “quality” varies among countries, or even between regions and parties, since the definition depends on the weight given to different elements that led to a candidate’s becoming elected.7 The differences between these two ways of constructing and defining “quality”—with one approach normative and the other practical—have an important effect. For example, if quality is defined by those characteristics that lead to more votes, then it could be said that corruption and clientelism are characteristics of quality. On the other hand, a formal definition of quality allows us to see to what degree quality influences voting behavior, and to compare this variable to other classical variables used to explain voting, such as ideology or party. Thus, for the purposes of this chapter, “quality” will be understood as a constellation of characteristics present (or absent) in each legislator, and not those characteristics that differentiate candidates who have achieved election from those who have not. What researchers consider the characteristics of quality vary from one author to another. Even when scholars use the same sociodemographic and personal characteristics to define a legislator, these characteristics can be grouped under different concepts. Some scholars use the expression “quality legislator” to mean an honest, competent person,8 or just a competent one.9 Others speak of “efficient legislators,10 or “legislators with expertise.”11 Others speak of “careerism,”12 “professionalism,”13 or “professionalization.”14 The quality of a legislator has also been identified as a two-pronged set of characteristics, consisting of a set of both personal abilities and strategic resources (e.g., the individual’s ability to organize a successful electoral campaign).15 Nevertheless, a certain order can be established among this variety of aspects and designations. The most broad-based and complex

A-11

10/2/07

3:39 PM

238

Page 238

CONCLUSION

understanding of quality is offered by Jeffrey Mondak (1995), who sees it as a combination of competence and integrity; the rest of the concepts can be included under one of these two characteristics.16 Besides the variety of understandings of the term “professional” or “professionalization,” the use of this concept is quite different when applied to legislatures in the United States than when applied to other cases in general. The historical development of legislative power in the United States means that a parliament is considered “professionalized” with respect to three variables: length of the sessions, salary, and having personnel and other resources.17 We are thus dealing with a formal question, in which neither the characteristics of the legislators nor the results of their work are important. At the same time, another kind of literature is much more diverse in terms of the cases it uses and the definitions it offers. These studies share the view that professionalization as a process and professionalism (or quality) as a characteristic are linked to the individual, and that the design or characteristics of the parliament are not that important.18 Returning to a consideration of the general concept, Mondak (1995), and also Carl McCurley and Mondak (1995), formulate the best possible definition of quality. This is so for two reasons. In the first place, this definition is the most inclusive; all others may be grouped under it. In the second place, and most important of all, it is closest to the voters’ definition. As shown in these two 1995 articles, when an individual decides how to vote, s/he is influenced by two pieces of information: the level of the candidate’s personal abilities and his or her integrity.19 In spite of this, in this chapter only information relating to the competence profile will be used, since data have been gathered on the individual level. To get indicators of honesty for each Latin American legislator is a very difficult task, which goes beyond the scope of this study. Thus, for the purposes of this chapter, a “high-quality legislator” will mean one who has the characteristics of competence. Having defined the theoretical elements, it is now possible to address this issue in the region. Table 11.1 contains some of the sociodemographic and sociopolitical characteristics of Latin American legislators most frequently used in scholarly studies. Table 11.2 presents the general characteristics of the legislators polled in the third round of the PELA,20 broken down by country. A glance at the data for Latin America in Table 11.1 shows that the majority of the percentages remain constant throughout time. The only three exceptions are the percentage of legislators with previous parliamentary experience, the percentage of legislators with political representative experience in a post won through popular elections, and legislators’ evaluations of their salaries. In these cases, there is at least a ten-point difference between the first and the third rounds. On one hand, there was a decrease in the percentage both of legislators who felt that their salary was insufficient and of those who had had parliamentary experience. On the

A-11

10/2/07

3:39 PM

Page 239

LEGISLATIVE CAREERS Table 11.1

239

Sociodemographic and Sociopolitical Characteristics of Latin American Legislators Third Round (n)

Gender (%) Men Women Average age Education (%) None Primary school Secondary school 2–3 years of university Bachelor’s degree Graduate studies Religious believer (%) Legislative experience (%) Political experience in representation (%) Family members in politics (%) Ideological selfplacement (average) Exclusive dedication (%) Salary insufficient (%) Economic improvement (%) Number of legislatures (%) One Two Three Four or more Average number of legislatures

Second Round (n)

First Round (n)

84.1 15.9 47.4

(1,239) (234) (1,456)

87.2 12.8 47.5

(1,192) (175) (1,355)

85.5 14.5 47.2

(634) (107) (772)

0.1 1.1 8.5 11.9 46.1 32.3 89.5 32.7

(2) (17) (126) (177) (689) (481) (1,326) (487)

0.1 1.4 9.5 15.1 47.5 26.5 88.3 33.6

(1) (19) (131) (206) (652) (359) (1,198) (458)

0.8 3.0 11.9 10.3 48.4 25.7 92.6 42.3

(6) (23) (92) (80) (376) (199) (692) (327)

41.1

(609)

39.8

(446)

25.0

(30)

54.0

(803)

53.3

(724)

49.1

(381)

5.05 (1,461) 57.7 (859) 36.1 (535)

5.04 (1345) 59.9 (820) 38.6 (526)

5.15 54.7 46.2

(750) (425) (358)

39.1

(577)

40.9

(463)

37.6

(45)

68.0 19.1 8.5 4.4

(1,002) (282) (125) (65)

67.1 17.1 11.1 4.8

(917) (234) (152) (65)





1.5

(1,474)

1.5

(1,368)





Source: Data based on PELA 1994–2005.

other hand, there was a considerable increase in legislators who had previously had other representative posts: from 25 percent to 41.1 percent. These data could indicate that in the past ten years—and especially at the end of the 1990s—important changes have taken place in Latin American parliaments. The first change has to do with the decrease in experienced legislators and with what could be considered the greater mobility of the elites. (Unfortunately, these data cannot be contrasted with the average number of terms the representatives remained in the legislature for the first wave of surveys.) The second change gives us a new understanding of this mobility, since these new legislators come to parliament having held other representative posts. Thus, the loss of legislative experience is compensated for with another type of experience, of a varied nature (local and regional institu-

A-11 10/2/07

Table 11.2

Sociodemographic and Political Characteristics of Latin American Legislators

Colombia

Costa Rica

El Ecuador Salvador Guatemala Honduras Mexico Nicaragua Panama Paraguay

Peru

DR

URU

VEN

67.6 32.4 51.8

83.8 16.3 44.3

89.8 10.2 48.5

87.4 12.6 45.1

62.7 37.3 46.7

83.7 16.3 47.0

91.3 8.8 47.3

91.7 8.3 48.7

89.2 3.9 49.3

75.0 25.0 43.6

73.3 26.7 46.5

89.1 10.9 50.6

88.9 11.1 44.0

81.5 18.5 48.4

89.6 10.4 46.7

88.2 11.8 50.5

88.9 11.1 47.1

0.0 0.0 18.3 8.7 52.9 20.2 83.7

0.0 5.0 8.8 7.5 42.5 36.3 79.7

0.0 0.0 8.0 12.5 42.0 37.5 86.4

0.0 1.1 3.2 6.3 23.2 66.3 94.7

0.0 3.9 5.9 9.8 41.2 39.2 94.1

0.0 0.0 11.2 3.1 56.1 29.6 91.8

1.3 1.3 9.0 15.4 52.6 20.5 93.8

0.0 2.5 11.6 17.4 43.8 24.8 91.7

0.0 2.9 17.6 5.9 58.8 14.7 96.1

0.0 0.0 3.2 7.3 59.7 29.8 90.2

1.7 0.0 5.0 8.3 36.7 48.3 83.3

0.0 0.0 10.9 7.8 50.0 31.3 98.4

0.0 0.0 12.5 28.6 30.4 28.6 94.6

0.0 0.0 2.4 4.8 45.8 47.0 98.8

0.0 0.9 0.9 23.1 50.4 24.8 99.1

0.0 1.5 11.8 19.1 42.6 25.0 51.6

0.0 1.0 5.0 19.0 37.0 38.0 83.7

20.2

17.5

61.4

36.8

7.8

36.7

55.0

32.2

50.0

16.9

28.8

46.9

14.3

29.3

25.6

32.4

37.0

58.7

31.3

37.9

72.6

37.3

37.8

33.8

33.9

25.5

53.7

35.0

29.7

47.3

37.3

26.5

44.1

50.5

43.3

58.8

55.7

62.1

56.0

54.1

52.5

53.7

71.6

50.0

61.7

53.1

66.1

37.3

48.7

50.7

51.5

4.5 71.2 21.8

4.1 75.0 48.8

5.4 64.8 21.6

5.6 86.3 11.7

5.1 58.8 42.0

4.6 61.2 76.5

5.1 45.0 23.8

5.4 43.8 62.8

5.8 15.7 67.6

5.0 62.1 16.3

4.4 50.0 18.3

6.1 40.6 17.2

5.5 50.0 46.4

4.8 84.1 23.5

5.8 42.7 47.9

4.0 69.1 13.2

4.4 64.0 33.0

31.7

43.8

24.4

36.2

34.0

26.5

41.3

25.6

12.8

60.2

50.0

57.1

29.6

45.1

58.1

45.6

43.9

80.6 12.6 2.9 3.9

83.5 10.1 5.1 1.3

38.6 25.0 35.2 1.1

63.2 25.2 7.4 4.2

92.2 5.9 2.0 0.0

64.6 21.9 5.2 8.3

45.0 21.3 22.5 11.3

70.7 13.8 7.8 7.8

50.0 24.5 16.7 8.8

83.7 11.4 2.4 2.4

71.2 23.7 3.4 1.7

53.1 32.8 10.9 3.1

85.7 12.5 1.8 0.0

70.7 17.1 6.1 6.1

75.7 20.0 4.3 0.0

68.7 22.4 6.0 3.0

64.3 25.5 3.1 7.1

1.3

1.2

2.0

1.5

1.1

1.6

2.0

1.5

1.8

1.2

1.4

1.6

1.2

1.5

1.3

1.4

1.5

Source: Data based on PELA 1994–2005.

Page 240

Gender (%) Men 84.1 Women 15.9 Average age 47.4 Education (%) None 0.1 Primary school 1.1 Secondary school 8.5 2–3 years university 11.9 Bachelor’s degree 46.1 Graduate studies 32.3 Religious believers (%) 89.5 Legislative experience (%) 32.7 Political experience of representation (%) 41.1 Family members in politics (%) 54.0 Ideological self-placement (average) 5.1 Exclusive dedication (%) 57.7 Salary insufficient (%) 36.1 Economic improvement (%) 39.1 Number of legislatures (%) One 68.0 Two 19.1 Three 8.5 Four or more 4.4 Average number of legislatures 1.5

Chile

3:39 PM

Latin America Argentina Bolivia

A-11

10/2/07

3:39 PM

Page 241

LEGISLATIVE CAREERS

241

tions or subnational and supranational representative bodies), but equally valuable. In general, it can be said that the majority of Latin American legislators are male, are religious believers, and have an average age close to fifty. The educational level of present-day Latin American legislators interviewed is high: more than 90 percent have had some degree of university study; moreover, the percentage of legislators with only a high school education or lower is on the wane. Ideologically, legislators can be situated on the line that separates the center from the center-left. As has been mentioned, it is increasingly common for legislators to come to parliament after having held other representative positions; and it is increasingly less common for them to have had prior experience in the legislative body that they are in. Even so, the average length of stay, measured in legislative terms, has remained stable throughout the last few years. The proportion of legislators who carry out another professional activity while they serve as legislators is high: more than 40 percent in the last round of interviews. Likewise, around 40 percent feel that their legislative salary is insufficient, although this same percentage of legislators also states that their income level has improved. Referring to the information for each country contained in Table 11.2, there are some percentages that depart significantly from the Latin American mean. The percentage of legislators with prior parliamentary experience is high in Chile, El Salvador, Honduras, and Panama, but in the cases of Paraguay and Costa Rica, this percentage is very low. The average stay in parliament, measured in terms (of office), is not very high in Latin America as a whole, averaging around 1.5 terms. Chile and El Salvador, on the other hand, have an average of 2 terms, which makes them, together with Honduras (1.8), the countries with the longest stays. In contrast, in Costa Rica, where immediate reelection is not permitted, the average stay is only very slightly above one term. The percentage of legislators previously elected to other representative posts is a majority in Colombia (72.6 percent) and very high in Argentina (58.7 percent) and Mexico (53.8 percent), both of which are countries with a federal structure.21 On the other hand, in the Dominican Republic (26.5 percent) and Honduras (25.5 percent), this situation is less common. Most legislators state that they dedicate themselves exclusively to their jobs as legislators in Colombia (86.3 percent), Peru (84.1 percent), and Bolivia (75 percent), while in Honduras, situated on the other extreme, this is true only in 15.7 percent of all cases. The majority of Latin American legislators are satisfied with their salaries as representatives, above all in the cases of Colombia and Uruguay. On the other hand, the majority are dissatisfied in Ecuador (76.5 percent), Honduras (67.6 percent), and Guatemala (62.8 percent). Although the majority of Latin American legislators are satisfied with their salaries, for 60.9 percent of them, this salary does not mean that they are now better off

A-11

10/2/07

3:39 PM

242

Page 242

CONCLUSION

economically than when they were carrying out other professional activities. The proportion of satisfied legislators drops to 39.8 percent in the case of Mexico, but rises to 87.2 percent in Honduras. These are some of the variables most frequently used to discover the characteristics of the individuals who make up the government. However, not all of them are valid sources of information on the level of competence that exists in congress. The variables that make it possible to analyze the relationship between who is in the government and their quality are all those aspects that reveal a certain level of expertise or knowledge that puts the legislator in a better position to carry out his or her tasks. Thus, the variables that we use to analyze the degree of quality are the following: educational level, years of party militancy and political trajectory, number of times elected as legislator, and having other offices of popular representation.22 The individuals included are all the legislators interviewed in the third round. The results of the cluster analysis23 allow us to speak of six different types of legislators in Latin America (see Table 11.3), labeled in the following way: highly experienced, experienced legislator, experienced politician (A and B), and inexperienced (A and B). The first category (highly experienced) is made up of legislators whose political and parliamentary experience is of the highest order in the Latin American context. The second category (experienced legislator) includes those with parliamentary experience, but no experience in any other elected office. In both of these categories, as well as having a high educational level, legislators have been party militants for many years and have spent years in politics in general. The third category (experienced politician) is made up of those who have held elected offices and whose familiarity with political and party circles is of many years’ standing. The qualities of the politicians (A) are different from the next category of politicians (B), in that their educational level is

Table 11.3

Classification of Quality of Experience of Latin American Legislators

Category

%

(n)

Party Experience

Legislative Experience

Elected Offices Held

Educational Level

H-Exper L-Exper P-Exp-A P-Exp-B Inexp-A Inexp-B

8.3 11.6 24.1 8.6 36.3 11.1

116 161 336 120 506 155

Very high Very high High High Medium High

Yes Yes No No No No

Yes No Yes Yes No No

High High High Medium–Low High Medium–Low

Source: Data based on PELA 1994–2005. Note: L-Exper: Experienced legislator; H-Exper: Highly experienced; Inexp-A: Inexperienced A; Inexp-B: Inexperienced B; P-Exp-A: Experienced politician A; P-Exp-B: Experienced politician B.

A-11

10/2/07

3:39 PM

Page 243

LEGISLATIVE CAREERS

243

higher. Finally, legislators in the two categories labeled “inexperienced” are characterized by their lack of legislative and representative experience. It is obvious that the most numerous group is the inexperienced group (A), which represents 36 percent of the total. The second largest group is that of experienced politicians (A), with approximately 24 percent. The sum total of the categories of highly experienced and experienced legislator is only 20 percent. The distribution by country (see Table 11.4) reveals important differences in the distribution of categories. In Chile and Colombia, “highly experienced” legislators make up around 18 percent of those interviewed. The countries above the Latin American average in the second category are Chile, El Salvador, and Honduras. By contrast, there is a high percentage of inexperienced legislators (B) in the Dominican Republic and in Paraguay. Inexperienced legislators are in the majority in Bolivia, Costa Rica, Ecuador, Peru, and the Dominican Republic. In Figure 11.1, these six categories have been reclassified into three, which differentiate between representatives with legislative experience, those with political experience, and those with neither experience at all. The first category, experienced legislators (H and L)—where, as a minimum, all members have parliamentary experience—is concentrated mainly in Chile,

Table 11.4

Distribution of Categories by Country

Country Argentina Bolivia Chile Colombia Costa Rica Ecuador El Salvador Guatemala Honduras Mexico Nicaragua Panama Paraguay Peru Dominican Republic Uruguay Venezuela Latin America

H-Exper 5.9 2.9 18.8 18.0 2.0 11.6 13.2 6.1 7.8 4.3 1.7 9.7 3.8 10.0 4.5 6.2 11.8 8.3

L-Exper P-Exp-A P-Exp-B 9.9 7.2 23.5 3.4 4.1 8.1 23.7 13.9 23.5 5.1 11.9 17.7 0.0 8.6 7.2 16.9 8.2 11.6

35.6 27.5 12.9 47.2 28.6 17.4 11.8 12.2 8.8 45.3 28.8 16.1 18.9 28.6 18.0 24.6 24.7 24.1

16.8 2.9 7.1 7.9 8.2 7.0 7.9 13.9 8.8 4.3 5.1 4.8 24.5 2.9 2.7 12.3 11.8 8.6

Inexp-A

Inexp-B

(n)

22.8 44.9 28.2 21.3 46.9 48.8 31.6 39.1 37.3 35.0 42.4 40.3 35.8 47.1 45.0 24.6 32.9 36.3

8.9 14.5 9.4 2.2 10.2 7.0 11.8 14.8 13.7 6.0 10.2 11.3 17.0 2.9 22.5 15.4 10.6 11.1

101 69 85 89 49 86 76 115 102 117 59 62 53 70 111 65 85 1,394

Source: Data based on PELA 1994–2005. Note: L-Exper: Experienced legislator; H-Exper: Highly experienced; Inexp-A: Inexperienced A; Inexp-B: Inexperienced B; P-Exp-A: Experienced politician A; P-Exp-B: Experienced politician B.

10/2/07

3:39 PM

244

Page 244

CONCLUSION

tin

A

m

er

ic

a

Figure 11.1 Level of Experience Among Latin American Legislators

La

A-11

Source: Created by the authors based on PELA 1994–2005.

El Salvador, Honduras, Panama, and Uruguay. The second category, experienced politicians, has high percentages in Colombia, Argentina, Mexico, and Paraguay. And in the last category, inexperienced legislators, the countries with the highest percentages are the Dominican Republic, Bolivia, Costa Rica, and Ecuador. If we use a dichotomous classification, experienced versus inexperienced (where the first four categories figure as “experienced” and the last two as “inexperienced”), the countries with the best percentages are Colombia, Argentina, Chile, Uruguay, and Mexico. On the other end of the scale are the Dominican Republic, Bolivia, Costa Rica, Ecuador, and Guatemala, with the highest percentage of legislators without experience. Only two of the three democracies considered stronger and with higher levels of quality in Latin America—Chile and Uruguay—have a majority of high-quality legislators. In Costa Rica, on the other hand, the group of inexperienced legislators is higher than 50 percent.24 Weaker democracies in the Andes region and Central America—and/or with low scores on the democratic quality scale,25 such as Guatemala, Ecuador, or Bolivia—also receive lower scores in quality of legislators. However, other cases such as Colombia rank among the highest in quality of parliaments. While the distance between the quality of a democratic regime and the quality of its

A-11

10/2/07

3:39 PM

Page 245

LEGISLATIVE CAREERS

245

politicians is very great, we cannot overlook the fact that even when legislatures have the same levels of expertise, their results can be completely different. Although Chile and Mexico have similar levels of experience, the way in which these human resources are used is very different in the two countries. Although there is no index of legislative quality that corroborates this statement, it is very probable that the effectiveness of the Chilean legislators is greater than of those from Mexico. Other, perhaps less controversial examples from the lowest end of the scale would be those of Costa Rica and Guatemala. Although these two countries both have a similar starting point, we can ask, What are the elements that allow Costa Rica to take advantage of its human resources? Why is it that other countries, such as Guatemala, are incapable of using these resources efficiently? In order to answer these questions, an analysis should be done taking into account, among other things, the legislatures’ levels of quantitative and qualitative efficiency, the imbalance in the roles played by the legislators of the party in power in comparison to those in the opposition, the organization of the congress, and work done in committees. Even though the state of research on Latin American legislatures does not yet allow these questions to be answered, we can sketch out certain relationships between each of the categories in which we have included parliamentary careers and other sociopolitical and sociodemographic variables, based on a biplot analysis. In this analysis we use, on one hand, the six types of legislators and, on the other, a group of variables with which these types can be related. The variable of family members is taken into account if the legislator has or has had relatives involved in politics. The gender variable indicates if the legislator interviewed is a man or a woman; the legislator’s age is included as well. The variable of religious belief tells us if the legislator is a believer or not. The variable of socialization captures the frequency with which politics was discussed during the legislator’s childhood and youth. The father’s educational level is also taken into account. Finally, ideology is reflected in the classic scale of “1” (left) to “10” (right). Figure 11.2, obtained from the HJbiplot representational technique,26 allows us to represent at the same time all of the variables and the relationship between each variable and the classification of parliamentary careers. In order to interpret the relationship of each category (bold) with the rest of the variables (not bold), that category must be projected perpendicularly onto each of the vectors. Figure 11.2 shows a discontinuous line such that for the family members variable, the “inexperienced (B)” category takes high values, which indicates absence of relatives dedicated to politics, whereas the L-Experienced and HExperienced categories take low values (they cut the variable in points A and B), which, according to our codification, indicates that these legislators have relatives in politics. In general, if the projection is close to the extreme

10/2/07

3:39 PM

246

Page 246

CONCLUSION

Figure 11.2 Relationship Between Different Parliamentary Careers and Sociodemographic Variables, Biplot Axes 1 and 2 1.50 L-EXPER

Exclusive dedication

L-EXPER Experienced legislator

Religious conviction

H-EXPER Highly experienced

INEXP-B

0.75 Age

Family members

Socialization Axis 2

A-11

Gender

0.00

B A

–0.75

INEXP-B Inexperienced B P-EXP-A Experienced politician A P-EXP-B Experienced politician B

P-EXP-B

H-EXPER

INEXP-A Inexperienced A

P-EXP-A

Father’s education level

INEXP-A Ideology

–1.50 –1.50

–0.75

0.00 Axis 1

0.75

1.50

Source: Created by the authors based on PELA 1994–2005.

of the variable (not bold), that category takes high values. If the projection is in the opposite extreme (in the discontinuous part of the line, imaginary for the rest of variables), this category takes low values for this variable.27 Given the fact that the total variability of the data cannot be included in just one figure, since we are talking about a bidimensional representation, in this analysis three dimensions, or axes, are maintained. Figure 11.2 contains the first two dimensions and Figures 11.3 and 11.4 contain the rest of the possible combinations. The three axes explain 88.85 percent of the total variability. Similarly, the quality of representation of the variables and the categories must be considered before interpreting the figure. In Figure 11.2, the position of all categories of parliamentary careers can be interpreted, except for the category of experienced politician (B), which is poorly represented on the plane, and its location is not correct. Ideological placement and father’s educational level are the other variables that are not of high enough quality in this figure (axes 1 and 2). The remaining variables and categories are well represented 28 and can be interpreted. In Figure 11.3, which corresponds to plane (1, 3), the categories of highly experienced and

10/2/07

3:39 PM

Page 247

LEGISLATIVE CAREERS

247

Figure 11.3 Relationship Between Different Parliamentary Careers and Sociodemographic Variables, Biplot Axes 1 and 3 1.50 L-EXPER Experienced legislator

Father’s education level

H-EXPER Highly experienced

INEXP-A Gender

0.75 Socialization L-EXPER Axis 3

A-11

Exclusive dedication

0.00 H-EXPER Age

Religious conviction P-EXP-A Ideology Family members INEXP-B

INEXP-A Inexperienced A INEXP-B Inexperienced B P-EXP-A Experienced politician A P-EXP-B Experienced politician B

–0.75

P-EXP-B –1.50 –1.50

–0.75

0.00 Axis 1

0.75

1.50

Source: Created by the authors based on PELA 1994–2005.

experienced politicians (B), as well as the variables of socialization, family members, father’s educational level, gender, and age, can be interpreted.29 Finally, in plane (2, 3)—Figure 11.4—the categories of inexperienced (A), experienced politician (B), and experienced legislators—as well as the variables of father’s educational level, religious convictions, and exclusive dedication—are of sufficiently high quality.30 The variable of ideology is not of a high enough quality to be interpreted in any of the figures. The category of highly experienced displays high values for the variables of socialization and age, and low values for the variables of exclusive dedication, religious conviction, family members in politics, and gender. This means that legislators from this category are older men with religious beliefs and that they dedicate themselves exclusively to politics. On the whole, politics was often discussed at home during their childhood and youth, they had family members involved or who had been involved in politics, and the educational level of their fathers was mid-range. Representatives in the category of experienced legislator show high values in the variables of exclusive dedication, religion, age, and socialization and low values for the variables of family members and gender. Thus,

10/2/07

3:39 PM

248

Page 248

CONCLUSION

Figure 11.4 Relationship Between Different Parliamentary Careers and Sociodemographic Variables, Biplot Axes 2 and 3 1.50

L-EXPER Experienced legislator

Father’s education level

H-EXPER Highly experienced

INEXP-A

INEXP-A Inexperienced A

Gender

0.75

Socialization

Axis 3

A-11

P-EXP-A Ideology H-EXPER

0.00

L-EXPER Religious conviction Exclusive dedication

Family members INEXP-B Age

INEXP-B Inexperienced B P-EXP-A Experienced politician A P-EXP-B Experienced politician B

–0.75

P-EXP-B –1.50 –1.50

–0.75

0.00 Axis 2

0.75

1.50

Source: Created by the authors based on PELA 1994–2005.

they share various traits with the previous category, the main difference being that as well as working in the parliament, they carry out other professional activities. Experienced politicians (A) show medium values for the variables of gender and family members, and low values for exclusive dedication, religious conviction, age, and socialization. This means that this group of legislators is composed of both men and women, that it is a younger group, and that these legislators dedicate themselves exclusively to their work in parliament. Their socialization is not due to family conversations on politics, and only a few in this group have family members in politics. The category of expert politician (B) is very similar with respect to the variables, although the members of this group tend to be a bit older, and there are more men than women. The variables of religious conviction, exclusive dedication, socialization, and age show low values for the category of inexperienced (A), which means that this is a young group where neither men nor women form an overwhelming majority of the members. These legislators dedicate themselves exclusively to their parliamentary job. Finally, the variables of reli-

A-11

10/2/07

3:39 PM

Page 249

LEGISLATIVE CAREERS

249

gious conviction, family members, gender, and exclusive dedication show high values for inexperienced legislators (B) and low values in the variables of age and socialization. This group is quite different from the previous one; although the members of this group are young as well, they are more likely to be women who carry out other activities in addition to their work in parliament and whose political socialization did not take place in the home. It is quite clear that in the two categories of Latin American legislators with the highest quality, this high quality is related not only to age, but also to the context of political socialization in which these legislators move. No other group moves in circles that are so favorable to political socialization. A majority of female legislators can be found in the category of inexperienced politicians (B). They have a significant, although lower, presence in the category of expert politicians (A) and inexperienced (A). In the rest of the categories, women are in a minority. Neither an exclusive dedication to parliament nor the father’s educational level seems to have a clear relationship to level of quality. The interpretation of the figures is summed up in Table 11.5. Having examined the characteristics of Latin American legislators, their levels of quality, the distribution of these categories by country, and other important variables, the question can be asked, Does quality matter? A number of strategies can be used to answer this question. The following section of this study attempts to ascertain in what fields quality has been used as a fundamental explanatory variable. The aspects on which quality is dependent are also detailed briefly, although this issue will not be discussed here.

Table 11.5

H-Exper L-Exper P-Exp-A P-Exp-B Inexp-A Inexp-B

Biplot Outcomes

Age

Gender

High High Low Med. Low Low

M M M-W M M-W W

Family Educational Members Level Religious Exclusive in Political of Conviction Dedication Politics Socialization Father Yes No Yes Yes Yes No

Yes No Yes Yes Yes No

Yes Yes Both Both Both No

Yes Yes No No No No

Med. Low — Low High —

Source: Data based on PELA 1994–2005. Notes: L-Exper: Experienced legislator; H-Exper: Highly experienced; Inexp-A: Inexperienced A; Inexp-B: Inexperienced B; P-Exp-A: Experienced politician A; P-Exp-B: Experienced politician B. Italics indicates information from Axes 2 and 3 in Figure 11.4. Bold indicates information from Axes 1 and 3 in Figure 11.3.

A-11

10/2/07

3:39 PM

250

Page 250

CONCLUSION

■ Quality as a Dependent and Independent Variable The different patterns of parliamentary careers and their level of quality depend on both individual characteristics and elements originating in the political system and the party organization. The quality of those elected depends on the quality of all the candidates and the ability of voters to discriminate among candidates and choose wisely.31 In Latin American cases, the quality of candidates does not depend only on their individual merits; it is also influenced by parties.32 Thus, as a dependent variable, quality is tied to constitutional and electoral norms, especially to rules referring to district size, electoral lists, and term limits. Quality is also dependent on whether parties’ recruitment processes are centralized or decentralized, hierarchical or democratic.33 In this study, quality is measured with information related to accumulated expertise. Thus, when a country’s electoral and constitutional norms limit the number of terms, the possibility of gaining this experience decreases; in these cases, the aggregate level of competence diminishes.34 When the district is larger and the electoral lists are closed, the quality of the candidates will be lower, since more restrictions are put on the voters’ ability to choose legislators of higher quality. Similarly, when districts are small and electoral lists are open, the probability is greater that a highquality candidate will be elected.35 On the individual level, the costs and benefits of the post to which the individual aspires must be considered. According to Francesco Caselli and Massimo Morelli (2004), although voters prefer to elect better candidates, multiple equilibria are possible, and these give rise to path-dependent trajectories. The presence of bad politicians in a parliament lowers the incentives for good candidates to come forward and, at the same time, these lowquality politicians perpetuate this dynamic by keeping (low) rewards. Particularly interesting in this respect is a study done by Ernesto Dal Bó, Pedro Dal Bó, and Rafael Di Tella (2002) on how threats, bribes, and punishments alter the costs and benefits of legislative posts. According to this model, punishments (be they physical violence, legal harassment, or defamation campaigns) lead to a lower quality (understood here not as honesty but as competence) of political representatives, since punishments make political life less attractive to highly capable individuals. Information on parliamentary careers has been used as an independent variable to explain various issues within the political system. Different research studies have tried to ascertain the impact of a legislator’s or politician’s quality on primary elections and candidates, campaign spending, voting, electoral margin, organization of the legislature, level of institutionalization of parliament, representation, public policy,36 executive/legislative relations, and political behavior in general. For example, research on US legislatures found that, given the need for a period of socialization37 and for time to learn how to do legislative tasks,

A-11

10/2/07

3:39 PM

Page 251

LEGISLATIVE CAREERS

251

the chances that organizational difficulties will arise are higher in a parliament with a large number of inexperienced members. Nevertheless, although the quality of a parliament may be affected, its efficiency, measured in terms of laws passed, will not be affected. Gerard Padró i Miguel and James Snyder (2004), on the other hand, maintain that the average performance of a legislature does, indeed, increase when legislative experience is higher. Given the fact that one of the variables included here as a component of competence takes this knowledge into account, high quality should lead to higher performance in parliament. In comparison to those who only seek reelection or who are not interested in being elected again, US politicians who aspire to higher office spend more time listening to public opinion.38 It has also been found that in more formally professionalized parliaments, legislators have more contact with their voters,39 are more attentive to their voters’ needs, and public policy has a closer relationship to public opinion. 40 In the case of Latin America, one would expect that, since the level of turnover is higher, those legislators who seek reelection will act differently and be more attentive to the needs of their districts, above all in small and single-member districts and where electoral norms favor personal vote. This, however, is an issue which remains to be analyzed. One field in which information on careers has been most employed is in the study of the institutionalization of parliament. The question of whether information on careers is indicative of level of institutionalization is still under debate; such information is used by Nelson Polsby (1968) to indicate just that. He notes that changes in parliament are due to the individual interests of representatives who have the possibility of remaining in parliament for a long period of time. However, the length of careers cannot be taken as an indicator of the level of institutionalization of a legislature. In addition, the potential for institutionalization is not the same in all parliaments.41 Renovation or length of careers is not sufficient to measure the capacity of a legislature to overcome the effects of renovation or lack of experience.42 One recurrent subject in the field of legislative organization is organization of committees. When legislators see parliamentary office as a longterm career, they develop standardized systems of career advancement.43 The choice of one method over another has to do with the characteristics of the parties and of the electoral and constitutional systems. Whether the system for career advancement is based on committees or on the party depends on who decides if a legislator has the possibility to be reelected. When parties are weaker, a legislator will seek reelection by increasing his or her participation in committees that shape public policy. This decentralized system arises where parties are heterogeneous and where there is personalized electoral voting and systems based on separation of powers. The opposite case,

A-11

10/2/07

3:39 PM

252

Page 252

CONCLUSION

where career advancement is centered on the party, arises in situations where the average length of stay in parliament is short. In this situation, legislators do not concentrate on influencing public policy, and committees are made up of legislators who share party lines. However, situations vary widely throughout Latin America; in some cases the drive for reelection is more probable than in others. In addition, the existence of mixed electoral systems should be taken into account. In cases of single-member districts, or with open electoral lists, in what way does the organization of committees take into account the presence of experienced legislators? Even though a parliament may not have a majority of career legislators, even though it may not have perfectly established and routinized systems of organization, it may still take advantage of the accumulated experience that does exist within it. Even when electoral incentives do not stimulate work in committees and the drive for reelection is not representatives’ main objective, there are ways in which legislators’ needs can be adjusted and the level of parliamentary work can be maintained. A study of state legislatures in the United States found that in nonprofessionalized legislatures, the most important committees tend to be made up of the most experienced legislators.44 With respect to Latin America, however, the way in which the organization of committees adapts to different types of parliamentary careers,45 and to different circumstances due to electoral norms, still remains to be studied.46 Despite the many relationships that exist among information on career, level of quality, and other variables of the political system, in Latin America there are still no analyses of whether results from previous studies hold true across different contexts. In order to contribute to this area of study, the following section analyzes the relationship between quality and voting.

■ The Quality of Legislators and Voting As mentioned previously, quality influences the voting choices of the electorate. This influence, however, is different depending on the political context, especially in terms of district size and the type of electoral lists.47 In a line of argument similar to that used by Matthew Shugart, Melody Ellis Valdini, and Kati Suominen (2005), an analysis cannot limit itself to national differences on the basis of different national electoral systems. Electoral norms do not operate identically in all circumstances. Differences within an electoral system due to district size, for instance, can give rise to different electoral systems within a single country.48 In their study on the quality of US representatives, Mondak (1995) and McCurley and Mondak (1995) analyze the way in which integrity and competence influence voting. These qualities influence voting in different manners. Integrity directly influences voters: the voter evaluates the candidates

A-11

10/2/07

3:39 PM

Page 253

LEGISLATIVE CAREERS

253

and their levels of integrity, then votes according to this evaluation. On the other hand, according to these authors, legislators’ levels of competence have an indirect influence, in that it dissuades less competent candidates from running for office. Casting a vote for a high-quality candidate may be strongly connected to the legislator’s seeking a personal vote. A study done by Brian Crisp, Maria Escobar-Lemmon, Bradford Jones, Mark Jones, and Michelle TaylorRobinson (2004) clearly details situations in which a legislator seeks a personal vote, and the relationship between this objective and different institutional designs. (The study tries to adapt the relationships that David Mayhew [1974] finds for the United States between seeking reelection and working in districts.) The case studies used are the upper and lower houses of Argentina, Chile, Colombia, and Venezuela, and the Costa Rican and Honduran assemblies. In these countries, unlike in the United States, elections are not always held in single-member districts, the selection of candidates is not always decentralized, and parties are not weak. In Latin American countries, legislators who run for reelection can follow a strategy of seeking votes or of improving the reputation of the party49 with respect to the characteristics of the electoral system, the methods of candidate selection, and the size of the district. Personal vote-seeking takes place when party lists are open and the selection of candidates is not under total party control. Party vote-seeking occurs when lists are closed and candidates are selected in a centralized manner. District size exerts influence in two ways. Where lists and candidate selection promote party voting, the larger the district, the more likely it is that legislators will work for the benefit of the whole party and the less likely it is that they will work for the district. By contrast, where personal votes are encouraged, the probability that legislators will work in an individual manner increases in large districts. However, the objective here is not the response of candidates to institutional designs, but the response of voters given these institutional designs. Our objective here consists of showing that quality is higher for legislators elected in small districts, where the knowledge and control of voters are more similar to those of single-member districts than they are in large districts, where, in addition to other factors, electoral results are more difficult to predict.50 In the latter cases, parties are not under pressure to place their best candidates on their lists. Personal characteristics are important; candidates can change their behaviors to be elected, but they cannot change their personal traits.51 When electoral districts are small, voters have a better knowledge of candidates and, in addition, can predict final results more accurately. As the size of districts increases, processing all the information on candidates becomes more difficult. In these circumstances, voters can exert some control on the quality of representatives if there are open party lists or party sub-

10/2/07

3:39 PM

254

Page 254

CONCLUSION

lists. In the case of closed lists, the possibility of voting on the basis of the candidates’ quality is very small. When districts are small and lists are open, the possibility of voting with quality in mind is very high. For this reason, we expect the percentage of legislators of quality to be higher in small districts than in large districts. When the district size is small and there are open lists, the difference between the percentage of legislators with quality and of those without quality will be higher. By contrast, legislators elected in large districts, and with closed lists, will display lower quality. In order to assess these expectations, the six categories of legislators previously mentioned have been reduced to two. The group of quality includes the first four categories, and the low-quality group includes the two categories of inexperienced legislators. The district size has been divided into three categories.52 Small districts elect one to four legislators, middle districts elect five to seven, and large districts more than eight. Figures 11.5 to 11.8 show the distribution of legislators by quality levels considering district sizes. Figure 11.5, which includes all Latin American legislators,53 shows that 40.7 percent of legislators are elected in small dis-

Figure 11.5 Distribution of Elected Legislatures by Level of Experience and District Size, Latin America 50

40

Percentage

A-11

30

20

10

0 Small

Medium

❒ Experienced

Large ■ Inexperienced

A-11

10/2/07

3:39 PM

Page 255

LEGISLATIVE CAREERS

255

tricts, 23.2 percent in medium-sized districts, and 36 percent in large districts. The percentage of inexperienced legislators is highest in large districts, while the percentage of experienced legislators is highest in small districts. These data support the idea that the quality of legislators matters and that it affects electoral results. The process through which quality influences electoral results can be twofold. In a direct manner, voters can elect the candidates with the highest quality among all candidates proposed by a party. Indirectly, parties can anticipate which qualities citizens will take into consideration when voting, and select candidates with higher quality when districts are small and lists are open. We cannot distinguish here between both types of effects, and will focus our analysis on the general implications of open lists and small districts. To examine these effects, we have included the figures of all countries. The first group of countries has closed and blocked lists.54 This group includes Argentina, Colombia, Costa Rica, El Salvador, Guatemala, Honduras, Nicaragua, Paraguay, and Uruguay. The second group includes countries with mixed electoral systems, using both single- and multimember districts. These are the cases of Bolivia, Mexico, and Panama. Finally, Ecuador is the only country with an open-list system. In the first group (see Figure 11.6), we expect to find more experienced legislators in small districts and more inexperienced legislators in large districts. In eight out of nine cases, in large districts the percentage of inexperienced legislators is higher than that of experienced representatives; the only exception is the case of El Salvador. In small districts, in five out of eight cases a higher percentage of experienced (than inexperienced) legislators gets elected, the exceptions being Colombia, El Salvador, and Nicaragua. In the last two cases, in small districts, inexperienced legislators are chosen over experienced legislators by a significant majority of votes. By contrast, in large districts there is no overwhelming election of inexperienced legislators. Therefore, our expectations are not confirmed in the cases of El Salvador and Nicaragua. Mixed systems are displayed in Figure 11.7. In the cases of Bolivia and Mexico, there are closed and blocked lists, and closed but not blocked in Panama in multimember districts. Figures show that, in Bolivia and Mexico, in small districts, which in most cases are single-member districts, there are more experienced than inexperienced legislators; whereas in large districts there are more inexperienced than experienced legislators. However, in Panama experienced legislators attain better results in multimember districts. This result is consistent with the thesis of Shugart et al. (2005) that if the characteristics of the electoral system favor seeking the personal vote, the probabilities of seeking the personal vote are higher in large districts. That is, when lists are unblocked, the influence of district size is less than the influence of list type. As a result, the possibility of

3:39 PM

Page 256

Figure 11.6 Distribution of Elected Legislatures by Level of Experience and District Size, Closed-List Systems 60

Percentage

Argentina

50 40 30 20

❒ Experienced ■ Inexperienced

10 0 Medium

Large

50

Percentage

40 Colombia

30 20 10 0 Small

Medium

Large

Small

Medium

Large

60 50

Percentage

10/2/07

Costa Rica

A-11

40 30 20 10 0

3:39 PM

Page 257

Figure 11.6 Continued 50

Percentage

El Salvador

40 30 20 ❒ Experienced ■ Inexperienced

10 0 Small

Medium

Large

Small

Medium

Large

Small

Medium

Large

70 60 50 Percentage

Guatemala

40 30 20 10 0 50 40

Percentage

10/2/07

Honduras

A-11

30 20 10 0

3:39 PM

Page 258

Figure 11.6 Continued 50

Percentage

Nicaragua

40 30 20 ❒ Experienced ■ Inexperienced

10 0 Small

Medium

Large

Small

Medium

Large

50

Percentage

40 Paraguay

30 20 10 0 80

60 Percentage

10/2/07

Uruguay

A-11

40

20

0 Small

Large

3:39 PM

Page 259

Figure 11.7 Distribution of Elected Legislatures by Level of Experience and District Size, Mixed Systems 80

Percentage

Bolivia

60

40 ❒ Experienced ■ Inexperienced

20

0 Small

Medium

Large

Small

Medium

Large

100

Percentage

80 Mexico

60 40 20 0

100 80 Percentage

10/2/07

Panama

A-11

60 40 20 0 Small

Medium

10/2/07

3:39 PM

260

Page 260

CONCLUSION

electing candidates of higher quality is greater in cases of unblocked lists than in cases in which districts are small or lists are single-member. In the Ecuadorian system (see Figure 11.8), due to the fact that lists are open, we would expect results similar to those of Panama, which has unblocked lists. However, results are similar to those of systems with blocked and closed lists, where experienced legislators are more frequent in small districts. One reason for this may lie in the effective number of parties: whereas in Panama the effective number of parties is 3.26, in Ecuador it reaches 7.54.55 The complexity of information that must be assessed in a system like the Ecuadorian one is therefore much higher. Thus, evaluations based on quality will be easier in small districts. Overall, the literature on personal vote pays attention to district size and the types of electoral lists. However, if the election of good candidates partially depends on the ability of voters to process information on the quality of candidates, the number of parties must also be taken into account. The ability of voters to distinguish between good and bad candidates diminishes as the number of parties increases. As the case of Ecuador shows, the influence of the effective number of parties can even counterbalance the expected effects of open lists.

■ Conclusion Many Latin American legislatures demonstrate significant percentages of quality legislators. These are the cases of the Colombian, Argentine,

Figure 11.8 Distribution of Elected Legislatures by Level of Experience and District Size, Open-List Systems Ecuador 60 50

Percentage

A-11

40 30 20 10 0 Small

Medium

❒ Experienced

Large

■ Inexperienced

A-11

10/2/07

3:39 PM

Page 261

LEGISLATIVE CAREERS

261

Chilean, Uruguayan, and Mexican lower chambers. In the Dominican Republic, Bolivia, Costa Rica, Ecuador, and Guatemala, more than half the legislators are totally inexperienced. The quality of those elected depends on both the quality of candidates and the capacity of voters to distinguish between good and bad candidates. The votes cast by citizens do not exclusively depend on party preferences or ideological leanings, but are conditioned also by the quality of candidates. Single-member districts and open lists favor the election of experienced legislators; large districts and closed lists, conversely, favor the election of inexperienced legislators. There is a third variable affecting electoral results: the effective number of parties. When the effective number of parties is high, the percentage of inexperienced legislators is higher in large districts, as the Ecuadorian data show. The ability of voters to elect good legislators decreases as the number of parties increases. Future research will have to take into consideration not only the effective number of parties, but also other party-related variables, such as processes of candidate selection. Candidate selection is usually considered in studies on the personal vote, but it is not clear what effects this variable has on voters’ perceptions and, above all, on electoral results. However, the results presented here are based on a concept of quality that takes into consideration the knowledge and experience, but not the honesty and moral integrity, of candidates. Future studies will have to address this limitation in order to make the information on the quality of Latin American legislators more complete. It is also necessary to consider what factors condition the quality of legislators, in both the expertise and integrity dimensions. These factors could include constitutional rules, electoral systems, party regulations, individual costs, and expected benefits. Another pending task is an in-depth assessment of the relationship between quality and other variables. There are three main issues relevant to the analysis of quality in Latin American legislators. The first one is the relationship between legislative efficiency and the quality of legislators; it would require quantitative and qualitative information on the efficiency of Latin American legislatures. It is also important to assess whether there is a learning curve in the legislative job, and how and when legislative knowledge and expertise are acquired. This would allow us to calculate the costs of experienced incumbents leaving legislatures. The second issue is the connection between the quality of legislators and the organization of the legislature. If seniority is not used as a criterion to organize legislative work, we must ask what, then, is the role played by senior legislators in the legislative chamber. Finally, the third issue involves the patterns of representation of legislators: Are experienced legislators more likely to devote their time to common-interest issues, or do they pay more attention to their districts’ interests? Many topics relevant to our understanding of legislative careers in Latin America remain still unexplored. However, the continuity and consol-

A-11

10/2/07

3:39 PM

262

Page 262

CONCLUSION

idation of democratic regimes now make it possible to define and carry out a long-term research agenda on the characteristics and functioning of Latin American legislatures.

■ Notes 1. I would like to thank the Education Ministry of Spain for its support. I also want to thank several colleges at the Research Seminar in Salamanca and two anonymous referees for their helpful comments. 2. Hibbing (1991). 3. Calvert (2004). 4. In addition to the recent study done by Alcántara, Gárcia Montero, and Sánchez (2005) for the Brazilian case, the following works can be consulted: Leoni, Pereira, and Renno (2004); Madeira (2003); Miguel (2003); Samuels (2003); Santos (2001); Santos and Renno (2004). In the case of Argentina, see Jones (2001) and Jones, Saiegh, Spiller, and Tommasi (2002). For the case of Mexico, see Beer (2004). Llanos and Sánchez López (2004) have done a comparative study of upper houses in Latin America, and Crisp, Escobar-Lemmon, B. Jones, M. Jones, and Taylor-Robinson (2004) have written a comparative study that includes both legislative branches in various countries of the region. See a theoretical typology in the introduction to Morgenstern and Siavelis (2007), based on Latin American legislators’ loyalties. 5. Shugart, Ellis Valdini, and Suominen (2005). 6. Squire and Hamm 2005); Stone, Maisel, and Maestas (2004); and Moncrief, Squire, and Jewell (2001). 7. Arnold (2003). 8. Padró i Miguel and Snyder (2004). 9. E. Dal Bó, P. Dal Bó, and Di Tella (2002). 10. Padró i Miguel and Snyder (2004). 11. Burns, Evans, Gamm, and McConnaughy (2004). 12. Epstein, Brady, Kawato, and O’Halloran (1997); Katz and Sala (1996); and Brady, Buckley, and Rivers (1999). 13. Moncrief et al. (2001); Squire and Hamm (2005); and Kousser (2005). 14. Norris (1997); Uriarte (2000); and Reiser (2003). 15. Stone et al. (2004). 16. Kinder (1983) points out four elements: competence, leadership, integrity, and charisma; although further studies have underlined the major importance of competence and integrity. 17. Fiorina (1994); Moncrief et al. (2001); Maddox (2002); Sanbonmatsu (2002); Reiser (2003); and Squire and Hamm (2005). 18. Botella (1997); Caselli and Morelli (2004); Cotta and Best (2000); Delgado (2000); Jerez (1997); Linz (2002); Norris (1997); Padró i Miguel and Snyder (2004); Stone et al. (2004); Uriarte (1997 and 2000); and Von Beyme (1995). 19. In the context of government action and the stabilization of democracy in Latin America, this issue is also considered by Hagopian (2005:321). This author states that citizens should perceive that their leaders are governing fairly, honestly, and effectively and, above all, that their voices can be heard. 20. Included here are the cases of Argentina (2003–2007), Bolivia (2002–2007), Chile (2001–2005), Colombia (2002–2006), Costa Rica (2002–2006), Ecuador (2002–2006), El Salvador (2003–2006), Guatemala (2004–2008), Honduras (2001–2005), Mexico (2000–2003), Nicaragua (2001–2006), Panama

A-11

10/2/07

3:39 PM

Page 263

LEGISLATIVE CAREERS

263

(1999–2004), Paraguay (2003–2008), Peru (2001–2006), the Dominican Republic (2002–2006), Venezuela (2000–2005), and Uruguay (2000–2005). The second round corresponds to the following periods of time: Argentina (1997–2001), Bolivia (1997–2002), Chile (1997–2001), Colombia (1998–2002), Costa Rica (1998–2002), Ecuador (1998–2002), El Salvador (2000–2003), Guatemala (2000–2004), Honduras (1997–2001), Mexico (1997–2000), Nicaragua (1996–2001), Paraguay (1998–2003), Peru (1995–2000), the Dominican Republic (1998–2002), Venezuela (1993–1998), and Uruguay (1995–2000). The first round includes Argentina (1995–1997), Bolivia (1993–1997), Chile (1993–1997), Costa Rica (1994–1998), Ecuador (1996–1998), El Salvador (1997–2000), Guatemala (1995–1999), Honduras (1994–1997), Mexico (1994–1997), Paraguay (1993–1998), and the Dominican Republic (1994–1998). 21. On the importance of federal structures, see Jones (2007) regarding Argentina; Langston (2007), on Mexico; and Samuels (2007), on Brazil. 22. The educational level takes values between “1” (no education) and “6” (graduate education). The accumulated years of party militancy and political participation are included in the analysis as a mean of both values. The number of times that a legislator has been elected as a representative varies between “1” and “4 or more legislatures,” and, finally, holding other offices of popular representation is a dichotomous variable. 23. The goal of this cluster analysis (K-means) is to classify individuals according to certain variables in such a way that the resulting groups are as differentiated as possible among themselves and, at the same time, the individuals within each group are as similar to each other as possible. 24. Although it is true that in Costa Rica there is a constitutional limitation that does not allow legislators to serve consecutive terms, this is not a sufficient explanation of why this country obtains such a low score on the quality scale, given the fact that the index includes more variables. This is pointed up by the Mexican case, where although the same limitation exists, the legislative branch ranks as one of the highest in terms of quality. 25. On different measurements of democratic quality in the Latin American region, see Altman and Pérez-Liñán (2002). On the state of democracy in Latin America, see Hagopian and Mainwaring (2005). 26. On classic biplot techniques, see Gabriel (1971) and Galindo (1986). 27. High values indicate older (age), women (gender), politics frequently spoken about during childhood and youth (socialization), high educational levels (father’s educational level), nonbelievers (religion), right ideological position (ideology), and no relatives in politics (relatives). 28. The quality of representation is on a scale of 0 to 1; the closer to “1” a variable or individual is, the higher that variable’s or individual’s quality of representation. The category of highly experienced has a quality of representation of .869; experienced legislator, .925; experienced politician (A), .667; inexperienced (A), .635; and inexperienced politician (B), .831. The variables have the following levels of quality of representation: socialization, .699; age, .901; works exclusively in parliament, .936; religious conviction, .897; family members in politics, .978; and gender, .733. 29. The levels of quality are, respectively, .665, .850, .732, .867, .978, and .843. 30. With a quality of representation of .751, .773, .630, .820, .856, and .980, respectively. 31. Caselli and Morelli (2004). 32. For an in-depth study of the importance of parties and also of the federal or centralized structure of the state, see Jones, Saiegh, Spiller, and Tommasi (2002).

A-11

10/2/07

3:39 PM

264

Page 264

CONCLUSION

33. On recruitment and candidate selection, see Morgenstern and Siavelis (2007). Argentina is studied by Jones (2007); Brazil by Samuels (2007); Chile by Navia (2007); México by Langston (2007); and Uruguay by Moraes (2007); see also Siavelis (2005) for Chile. 34. Mondak (1995). 35. These same variables, in relation to personal vote and local needs, are used by Shugart et al. (2005). 36. On the relationship between quality of parliamentary career and public policy, see Moncrief (1999); Fowler (1993); and Maestas (2000). On constituency building in Colombia, see Crisp and Desposato (2004). 37. The study done by Hibbing (1991:6) for the US legislative branch in the decade of the 1980s states that this learning curve still exists. Of all the activities carried out by a legislator, the legislative ones have the most pronounced learning curve. The activity in districts, however, decreases over time. On this same issue, see Squire (1998). 38. Maestas (2003). Moraes (2007) applies this idea for Uruguay. 39. Squire (1993). 40. Maestas (2000 and 2003). 41. Hibbing (1991:166). 42. On this debate, see Squire and Hamm (2005). 43. Epstein et al. (1997:992). 44. Burns et al. (2004). 45. On the selection of committee presidents in Brazil, see Santos and Renno (2004). 46. In a comparison between theoretical models applied to the US Congress and Latin American cases, Morgenstern and Nacif (2002) mention, as one of the basic differences, the fact that not all Latin American legislators seek reelection. In addition, legislators from this region follow different strategies once they are in office, and act under different constitutional arrangements and party alignments. 47. An experimental study (Chin and Taylor-Robinson, 2005) shows how electoral rules affect information acquisition. Voters under proportional representation (PR) and closed lists access less candidate information compared to individuals under single-member rules (SMD). Also, subjects in the PR system were more likely to vote for their party than subjects in the SMD system. 48. This idea is applied to Central American countries in García Díez (2005:57). 49. Both strategies are not excluded, as Moraes (2007) shows for Uruguay, where both are developed simultaneously. 50. A similar argument is presented in Shugart et al. (2005) regarding personal vote. 51. Ibid. 52. Information on district sizes comes from Alcántara, García Montero, and Sánchez (2005). 53. The cases of Chile, Peru, the Dominican Republic, and Venezuela have not been included in this analysis, either because of our lack of data on district sizes or because there is no variation in district size. 54. On the characteristics of electoral systems, see Payne, Zovatto, Carillo, and Allamand (2002). This classification considers electoral norms that ruled the election of those representatives, leaving aside further (posterior) changes. 55. On the effective number of parties in Latin America, see Alcántara (2004b).

A-12

10/2/07

3:40 PM

Page 265

12 Politicians and Politics in Latin America After Twenty-Five Years of Democracy Manuel Alcántara Sáez

■ The Return of an Old Paradigm? Almost three decades of democratic transition have witnessed numerous political reforms. No country in the region, with the obvious exception of Cuba, has been left out of this wave of political transformations. Profound constitutional reforms, decentralization processes, and changes in the relationships between the branches of government, in electoral laws, and in rules regarding political participation are just some of the transformations experienced by Latin American polities.1 As of 2006, the agenda of political reforms is patently open in Bolivia and Ecuador, and is also significant in Chile, El Salvador, Paraguay, and Peru. Stimulated by a sort of constitutional obsession, institutional and political reforms have derived from a combination of several factors. In the first place, from the 1980s on, the abundant and fertile academic literature on constitutional engineering,2 informed by the development of neoinstitutional theory in political science, has supplied the intellectual energy necessary to address this issue.3 In the second place, this trend toward political reform was facilitated by the malleable character of formal institutions, easily manipulated by political reformers. In the third place, skillful and ambitious political leaders used reforms to advance their political goals; for instance, making constitutional changes to enable the reelection of presidents. Finally, the sometimes-improvised policy of international cooperation with organizations and donor countries, in combination with the ambition of international consultants, has inspired all kinds of institutional reform projects, without much consideration of whether or not they are feasible. In the 1980s, discussions started to center on the advantages and disadvantages of presidentialism. Later on, debates focused on electoral repre-

265

A-12

10/2/07

3:40 PM

266

Page 266

CONCLUSION

sentation and led to the introduction of voting by ballot in presidential elections and of mixed electoral systems. Institutional debates addressed other aspects and types of elections and electoral systems (such as reelection, electoral thresholds, and preferential vote) and the implications of uni- and bicameralism.4 Although it is impossible to give, in this chapter, an accurate and all-encompassing picture of all these reforms, it is possible to say that most of them were designed in an improvised manner and without careful study of their implications and alternatives. Reforms were effected without a systemic consideration of political problems, and were driven by unrealistic expectations regarding their short-term effects. These results must not lead one to the conclusion that neoinstitutionalist theories were intrinsically flawed; it was, rather, their narrow-minded application to Latin American politics that failed to meet the quite often unrealistic expectations they gave rise to. In an intellectual context dominated by structural class-based and dependence theories, neoinstitutionalist theories had the advantage of drawing attention to political institutions. Institutions matter, but the characteristics and orientations of politicians matter as well. Political elites push changes in institutions, and are critical to political democratization and modernization. It is also important to take politicians into account in analyzing the shifts in economic policies of the 1980s; here, too, their perceptions and attitudes on economic issues and problems matter. However, with a few exceptions,5 the study of promarket economic policy reforms has so far ignored this facet to focus on economic conditions and the role of international multilateral organizations. Years ago, Robert Scott (1967:117) pointed out that political elites play a key role in the transition from traditional to modern politics. Even more important is their role in the construction of democratic regimes and the profound economic reforms Latin American countries have experienced in the past few decades. The generalized renewal of the political elite makes it even more urgent to focus our analysis on its characteristics, attitudes, and evaluations. However, current analyses of elite opinions are not common among political scientists because academics no longer rely much on the opinions of strategic politicians; they prefer to base their research on more objective criteria, like behavior. This book attempts to reformulate that trend. A key change in Latin American politics during the past twenty-five years is the increasingly important role played by a political elite formed by professionals with access to power through elections, most of the time under the umbrella of political parties. Legislators are an important part of this elite. They play a crucial role in democratic political systems, both as legislators and as prominent members of their parties. They play their legislative roles for fixed terms, which range from three to seven years.6 Their knowledge of institutional, political, and substantive issues contributes to the professionalization and specialization of the political class. Where

A-12

10/2/07

3:40 PM

Page 267

POLITICIANS AND POLITICS AFTER TWENTY-FIVE YEARS OF DEMOCRACY

267

reelection is possible, legislators also develop a feeling of belonging to a political elite. After more than two decades of representative democracy, the routinization of democratic politics in Latin America makes the study of opinions, values, and socialization processes especially important. Political parties have become key players in democratic politics as channels for the formation, aggregation, and articulation of interests. In turn, politicians over time have developed new political identities, personal goals, professional careers, and strategies for action. One of the most critical issues of political life concerns the values and identities of political actors. Focus on strategic and instrumental considerations has often led to the denial of the role played by values and ideology. However, the analysis of politics reveals the important role values play in shaping political actions and events. Studying legislators’ perceptions and values helps us to understand both intraparty politics and interparty competition. Latin American elites share many political values, particularly their support of democracy. The values and perceptions of political elites are also linked to those of Latin American citizens, particularly when it comes to the definition and evaluation of the most urgent problems faced by Latin American countries.

■ Politicians and Representative Democracy One reason many Latin American citizens feel alienated politically relates to public perceptions of the democratic channels for representation. Public opinion surveys indicate that legislatures and political parties are the two institutions receiving the worst evaluations.7 Doubts about or criticisms of the representative mechanisms of democracy account for the interest in establishing formal mechanisms for citizen participation, and were behind the constitutional reforms in Colombia (1991) and Venezuela (1999). Debates opposing representative and direct democracy have become so important due in part to public dissatisfaction with the performance of politicians. Inclusion of participatory mechanisms in institutional designs may serve the interests of populist leaders, but does not eliminate the importance of public grievances about the functioning of representative mechanisms and the performance of political elites. As Araceli Mateos Díaz shows in this book, support for democracy has been constant and widespread among Latin American politicians. This support is one of the fundamental reasons for the success of democracy in the region.8 Latin American legislators see democracy as a regime protecting individual rights and liberties, and consider consensus around the constitutions and their basic institutions to be the crucial factor in democratic con-

A-12

10/2/07

3:40 PM

268

Page 268

CONCLUSION

solidation. Politicians hold thus a procedural rather than substantive or economic view of democracy. In spite of this, Latin American legislators assign the highest importance to economic problems. However, as Mateos Díaz shows, many politicians put into question the role of parties as a fundamental institution of democracy. This is particularly true in the cases of three Andean countries: Bolivia, Venezuela, and Ecuador. In these three countries, party systems have experienced severe crises, leading, in the cases of Bolivia and Venezuela, to the disappearance of the old party system. In these three countries, profound political turmoil and institutional crises took place. In Venezuela, pronounced political polarization paved the way for failed coup attempts; in Bolivia and Ecuador, five presidents were replaced in the past decade. The idea that parties are not always necessary for democracy, in combination with the relativizing of the role of elections in channeling the population’s demands, puts some limits on the development of fully democratic orientations among the Latin American legislators. These limitations are parallel to the existence of a significant political divide around democracy, which, in turn, overlaps with left-right divisions and differences on the role of the military, as Cristina Rivas Pérez shows. The issue of legislators indicating some doubts about the necessity of political parties to democracy is complemented by Mercedes García Montero and Patricia Marenghi’s analysis of representation. Despite the fact that constitutions emphasize the national character of representation, particularist interests prevail among Latin American legislators, leading to public dissatisfaction with politicians. Ironically, clientelism does not lead to public approval, but to the alienation of citizens. According to legislators, they tend to be guided by mandates, particularly from voters in their constituencies, rather than their own personal views and assessments. This, however, does not make citizens’ evaluations of their representatives more positive. The low legitimacy of political parties can be connected to the absence of inclusive and participatory mechanisms. As Leticia Ruiz Rodríguez shows, both political elites and citizens must have incentives in order to create and participate in democratic decisionmaking processes. As was stated years ago, the absence of communication between Latin American political leaders and their followers gives rise to independent and frequently irresponsible political elites.9

■ The Difficult Path Toward Professionalization The professionalization of politicians concerns both legislative and party life. Leticia Ruiz Rodríguez’s study of party life shows that interparty competition prevails over intraparty participation. This fact has negative consequences for the vitality of party life. When parties focus strictly on electoral

A-12

10/2/07

3:40 PM

Page 269

POLITICIANS AND POLITICS AFTER TWENTY-FIVE YEARS OF DEMOCRACY

269

contests and disregard internal organization, it is very difficult for legislators to have a professional life within a party. By contrast, political and personal opportunities flourish in the often-well-funded legislatures; high salaries and logistical and physical resources facilitate and encourage legislative careers. In spite of this, more than one-third of Latin American legislators consider their remuneration too low; although, as María Martínez Rosón shows, there are wide cross-national differences in this respect. In this regard, there is a pronounced contrast between Ecuador, Honduras, and Guatemala, where salaries are considered by legislators very low, and Colombia and Uruguay, where this perception is not present at all. Two elements that can help identify the level of professionalization of Latin American legislators are previous political experience and degree of dedication to legislative activities. In both cases, the situation of Latin American legislators is far from that of complete political professionalization. As far as dedication to legislative activities is concerned, the most extreme cases are those of Honduras, where only one out of six legislators are exclusively dedicated to their legislative careers, and Colombia and Peru, with more than 80 percent of legislators dedicated exclusively to parliament. Overall, most Latin American legislators dedicate themselves parttime to their legislative activities. Martínez Rosón’s data show the low legislative experience of most legislators. The most important exceptions are those of El Salvador, where terms last three years, and Chile. In these two countries, more than one-half of legislators have been in parliament for more than three terms. With the exception of Colombia, less than one-half of Latin American legislators have had previous experience as representatives in other institutions. Formal education is essential to both the professionalization and the quality of legislators’ activities. On average, the educational level achieved by Latin American legislators is high. Nine out of ten legislators have some type of university degree, while more than one-third of them have done at least some graduate study. Martínez Rosón’s study shows an interesting association between the quality of legislators and the quality of politics. Whereas the cases of Chile and Uruguay are marked by the high quality of their legislatures, those of Guatemala, Ecuador, and Bolivia are characterized by their low quality. These defining characteristics of legislatures and legislators are connected to individual features as well as to party system and party organization characteristics.

■ The Persistence of the Left-Right Dimension Contributions in this book clearly reveal the persistence and importance of the left-right dimension in Latin American party systems.10 Individual posi-

A-12

10/2/07

3:40 PM

270

Page 270

CONCLUSION

tions in the left-right dimension provide us with consistent information about Latin American politics and also clarify the characteristics of political parties. Abstract ideological positions account for issue preferences and perceptions of the problems of everyday political life. Since the 1980s, the persistence of these connections has contributed to organizing and anchoring Latin American political life. Ideological differences clearly structure axes of competition, as Cristina Rivas Pérez’s study on the dimensions of programmatic polarization shows. Her study demonstrates the importance of five basic dimensions: state intervention, armed forces, democracy, religion and values, and, to a lesser degree, social background. Nevertheless, political parties have quite significant differences. There are important cross-national differences in levels of ideological coherence (understood as agreement among party members on several issues). These differences also connect to party organizational models. As Leticia Ruiz Rodríguez shows in this volume, high levels of militancy go hand in hand with high levels of consensus on ideas, programs, and organizational strategies. By contrast, parties with low militancy levels display low consensus on substantive issues. Flavia Freidenberg, Fátima García Díez, and Iván Llamazares Valduvieco show the importance of links between the ideological characteristics of parties and legislators’ perceptions regarding their party structures. Their study shows, in addition, that opposition parties are more coherent than governmental parties, a result of the fact that governmental decisions on political issues are not always compatible with ideological consistency and unity. The ideological placement of Latin American parties in the left-right dimension helps us to predict the attitudes and perceptions of legislators regarding issues ranging from the economy to religious and moral values. In addition, as Cristina Rivas Pérez writes in this volume, legislators’ positions allow us to identify higher levels of homogeneity among the Latin American right in those dimensions she analyzes. By contrast, the Latin American left displays significant differences in its attitudes toward the military and democratic institutions. Exploring similar problems, Leticia Ruiz Rodríguez analyzes differences across parties in the evaluation of problems and the definition of political strategies. This analysis allows her to identify two key elements, which she labels “evaluation” and “strategy” components. Her analysis of programmatic coherence, which focuses on parties with high levels of electoral success in the past decade, shows that left-wing parties display higher levels of coherence. Her analysis also shows a relationship between programmatic incoherence and party dependence on external funding. Salvador Martí i Puig and Salvador Santiuste Cué show the crucial role played by state economic intervention among left parties, whereas Manuel Alcántara Sáez and Iván Llamazares Valduvieco analyze the role of socio-

A-12

10/2/07

3:40 PM

Page 271

POLITICIANS AND POLITICS AFTER TWENTY-FIVE YEARS OF DEMOCRACY

271

economic, cultural, and political attitudes among legislators of right parties. Both types of party families (left and right) display very important internal differences. However, a more comprehensive understanding of Latin American party systems also demands an analysis of center parties, a topic that this volume does not address.

■ Politicians and Politics Democratic institutions are now a solid reality in Latin America, but their efficient functioning requires the presence of certain types of politicians in general and of legislators in particular. In contrast to the past, in this day and age politicians do not gain access to positions of political responsibility as a result of their membership in the agrarian elites, the armed forces, or oligarchic groups, nor is it enough for them to enjoy the support of the Catholic Church, business circles, or the US embassy. The political game consists primarily of regularly held elections. It is through these elections, which are, on most occasions, fair and competitive, that citizens participate and candidates get elected. Politicians know the rules and they know that their political future will depend on their performance and on how voters evaluate them. Under a democratic system, severe efficiency problems in addressing social issues will lead to public alienation and political crisis. The new political situation experienced by Latin American countries directs our attention to the quality of democratic politics, understood as the ability of politicians to respond to public demands and solve social problems in a framework of equal rights and liberties. Addressing these tasks is impossible without the existence of a highly qualified political elite. Politicians act now in a context characterized by freedom, autonomy, and, to some degree, meritocracy. Political learning and the assimilation of the basic institutions of democracy by politicians are now critical to the quality and stability of political systems. Once a fair and impartial institutional framework has been achieved, democratic quality requires that political leaders have certain characteristics. Particularly important are the socialization and professionalization of politicians and the development of the skills necessary to carry out political life in the legislative arena. “Professionalization” here refers to the presence of appropriate institutional mechanisms and incentives for the development of a stable, consistent, and predictable legislative career. Politicians belong to party organizations, which typically exist before most politicians enter politics. They socialize politicians and shape their behavior according to historical legacies and organizational dynamics. In some cases, however, political leaders enjoy the freedom necessary to shape the character of political parties. The relationship between politicians and parties is one of the most important aspects of political life. Differences

A-12

10/2/07

3:40 PM

272

Page 272

CONCLUSION

between individual preferences and party positions are part of this interactive game in which party organizations and politicians act. Also crucial in this regard are the tradeoffs between hierarchical and democratic organizations. One useful way of approaching the interactive dynamic between politicians and parties is the direct analysis of the political elite. This type of analysis will also benefit from the development of contextual studies considering the importance of both parties and nations, and will probably entail the development of multilevel analyses of Latin American politics.11 Reforms advocating for an increase in internal party democracy often focus on internal electoral processes and candidate selection, but do not address the elaboration of party platforms and programs. Democratic reforms in the elaboration of party platforms would bring party programs and legislators’ positions closer together. Theoretical models on representation, mandates, and representative autonomy are also relevant to the analysis of legislators’ conceptions of representation. Still, differences between revealed preferences and actual behavior are always possible, and Latin American legislators may have different conceptions of representation. The Latin American legislative elite is an important, though numerically small, part of the total set of Latin American politicians. Findings regarding Latin American legislators must be complemented by the study of other groups of politicians, such as the local and regional legislative elites, with which national elites might share some important traits. These findings could also be contrasted with those of analyses focusing on governmental elites. A comprehensive knowledge of Latin American politicians requires the study of these two different types of politicians.

■ Notes 1. Payne, Zovatto, Carrillo, and Allamand (2002). 2. An influential book by Sartori (1994) included these two words in its title. 3. March and Olsen (1984). 4. There is an abundant literature on these issues. See, for instance, García Díez (2005); Jones (1995); Linz and Valenzuela (1994); Martínez (2004); Molina (2000); and Sabsay (1991). 5. One exception is Stokes (2001), who centers her study on the perceptions of those politicians who switched from statist to market-oriented policies. Her study focuses, however, on changes by incumbents in national executives. 6. It must be kept in mind that, in contrast to the executive powers, legislative powers have maintained total stability. In no situation have legislators witnessed a shortening of their mandates. 7. PNUD (2004). 8. This coincides with the argument of Mainwaring and Pérez-Liñán (2005:46) on the virtuous circles created by changing attitudes toward democracy among both the right and left in Latin America. 9. Scott (1967:127).

A-12

10/2/07

3:40 PM

Page 273

POLITICIANS AND POLITICS AFTER TWENTY-FIVE YEARS OF DEMOCRACY

273

10. This argument is also made in Alcántara (2004a). However, these two studies stem from different data. Data for this book stem from the PELA (1994–2005), an annual study focusing on incumbent legislators. The data for Alcántara (2004a) stem from the PPAL (1999–2001). Alcántara’s study was carried out at a single point in time and included party militants with different types of political roles (high-ranking party leaders and militants in different areas of government and in representative office). 11. Thus, if we consider party levels of ideological heterogeneity, they are not only shaped by individual characteristics, but also by national political-institutional characteristics (such as political trajectories and electoral volatility) and party features.

A-12

10/2/07

3:40 PM

Page 274

A-BM

10/2/07

3:40 PM

Page 275

Political Parties, by Acronym and by Country

■ Party Acronyms AD Acción Democrática/Democratic Action (Venezuela) ADN Acción Democrática Nacionalista/Nationalist Democratic Action (Bolivia) ANR Asociación Nacional Republicana/National Republican Association (Paraguay) ARENA Alianza Republicana Nacionalista/Nationalist Republican Alliance (El Salvador) ARI Alternativa por una República de Iguales/Alternative for a Republic of Equals (Argentina) BIP Bloque Independiente Peñagomista/Independent Peñagomista Block (Dominican Republic) BIS Bloque Institucional Socialista Demócrata/Socialist Democratic Institutional Block (Dominican Republic) CONDEPA Conciencia de Patria/Nation’s Conscience (Bolivia) COPEI Comité de Organización Política Electoral Independiente/Committee for the Independent Political Electoral Organization (Venezuela) DP Democracia Popular/Popular Democracy (Ecuador) EP/FA Encuentro Progresista/Frente Amplio/Progressive Alliance/Broad Front (Uruguay) FDNG Frente Democrática Nueva Guatemala/Democratic Front of New Guatemala FIM Frente Independiente Moralizador/Independent Moralizing Front (Peru) FMLN Frente Farabundo Martí para la Liberación Nacional/Farabundo Martí Front for National Liberation (El Salvador) FREPASO Frente País Solidario/Front for the Country of Solidarity (Argentina)

275

A-BM

10/2/07

3:40 PM

276 FRG

Page 276

POLITICAL PARTIES

Frente Republicano Guatemalteco/Republican Front of Guatemala FSLN Frente Sandinista de Liberación Nacional/Sandinista Front of National Liberation (Nicaragua) GANA Gran Alianza Nacional/Grand National Alliance (Guatemala) ID Izquierda Democrática/Democratic Left (Ecuador) MAS Movimiento al Socialismo/Movement for Socialism (Bolivia, Venezuela) MIR Movimiento de Izquierda Revolucionario/Leftist Revolutionary Movement (Bolivia) MNR Movimiento Nacionalista Revolucionario/Nationalist Revolutionary Movement (Bolivia) MPQ Movimiento Patria Querida/Movement of the Beloved Nation (Paraguay) MUPP-NP Movimiento Unidad Plurinacional, Pachakutik–Nuevo País/New Nation–Pachakutik Plurinational Unity Movement (Ecuador) MVR Movimiento V República/V Republic Movement (Venezuela) NFR Nueva Fuerza Republicana/New Republican Force (Bolivia) PA Partido Arnulfista/Arnulfista Party (Panama) PAC Partido de Acción Ciudadana/Party of Citizen Action (Costa Rica) PAN Partido Acción Nacional/National Action Party (Mexico) PAN or PANguat Partido de Avanzada Nacional/Party of National Vanguard (Guatemala) PAP Partido Aprista Peruano/Aprista Party of Peru PC Partido Colorado/Colorado Party (Uruguay) PC Partido Conservador de Colombia/Conservative Party of Colombia PCN Partido de Conciliación Nacional/Party of National Conciliation (El Salvador) PDC Partido Demócrata Cristiano/Christian Democratic Party (Chile, El Salvador) PEN Partido Encuentro Nacional/National Reunion Party (Paraguay) PJ Partido Justicialista/Justicialist Party (Argentina) PL Partido Liberal de Colombia/Liberal Party of Colombia PL or PLH Partido Liberal/Liberal Party (Honduras) PLC Partido Liberal Constitucionalista/Constitutionalist Liberal Party (Nicaragua)

A-BM

10/2/07

3:40 PM

Page 277

POLITICAL PARTIES

PLD PLN PLRA PN PN or PNH PP PPD PRD PRD or PRDpa PRD or PRDrd PRE PRI PRIAN

PRSC PS PS or PSp PSC PSN PSP PU PUD PUSC RD RN UCeDé UCR

277

Partido de la Liberación Dominicana/Dominican Liberation Party Partido de Liberación Nacional/National Liberation Party (Costa Rica) Partido Liberal Radical Auténtico/Authentic Radical Liberal Party (Paraguay) Partido Nacional/National Party (Uruguay) Partido Nacional/National Party (Honduras) Perú Posible/Peru Possible Partido por la Democracia/Party for Democracy (Chile) Partido de la Revolución Democrática/Democratic Revolutionary Party (Mexico) Partido Revolucionario Democrático/Democratic Revolutionary Party (Panama) Partido Revolucionario Dominicano/Dominican Revolutionary Party Partido Roldosista Ecuatoriano/Roldosista Party of Ecuador Partido Revolucionario Institucional/Institutional Revolutionary Party (Mexico) Partido Renovador Institucional Acción Nacional/Party of National Action and Institutional Renovation (Ecuador) Partido Reformista Social Cristiano/Social Christian Reformist Party (Dominican Republic) Partido Socialista/Socialist Party (Chile) Partido Solidaridad/Solidarity Party (Panama) Partido Social Cristiano/Social Christian Party (Ecuador) Partido Solidaridad Nacional/National Solidarity Party (Guatemala) Partido Sociedad Patriótica/Patriotic Society Party (Ecuador) Partido Unionista/Unionist Party (Guatemala) Partido de Unificación Democrática/Democratic Unification Party (Honduras) Partido de Unidad Social Cristiana/Party of Social Christian Unity (Costa Rica) Renovación Democrática/Democratic Renovation (Costa Rica) Renovación Nacional/National Renovation (Chile) Unión de Centro Democrático/Union of the Democratic Center (Argentina) Unión Cívica Radical/Radical Civic Union (Argentina)

A-BM

10/2/07

3:40 PM

278

Page 278

POLITICAL PARTIES

UCS

Unidad Cívica Solidaridad/Civic Union of Solidarity (Bolivia) Unión Democrática Independiente/Independent Democratic Union (Chile) Unidad Nacional/National Unity (Peru) Unión Nacional de Ciudadanos Éticos/National Union of Ethical Citizens (Paraguay)

UDI UN UNACE

■ Parties, by Country Argentina ARI

FREPASO PJ UCeDé

Bolivia

UCR ADN CONDEPA MAS MIR MNR NFR UCS

Chile

PDC PPD PS RN UDI

Colombia

PC PL

Alternativa por una República de Iguales/Alternative for a Republic of Equals Frente País Solidario/Front for the Country of Solidarity Partido Justicialista/Justicialist Party Unión de Centro Democrático/Union of the Democratic Center Unión Cívica Radical/Radical Civic Union Acción Democrática Nacionalista/ Nationalist Democratic Action Conciencia de Patria/Nation’s Conscience Movimiento al Socialismo/Movement for Socialism Movimiento de Izquierda Revolucionario/ Leftist Revolutionary Movement Movimiento Nacionalista Revolucionario/ Nationalist Revolutionary Movement Nueva Fuerza Republicana/New Republican Force Unidad Cívica Solidaridad/Civic Union of Solidarity Partido Demócrata Cristiano/Christian Democratic Party Partido por la Democracia/Party for Democracy Partido Socialista/Socialist Party Renovación Nacional/National Renovation Unión Democrática Independiente/ Independent Democratic Union Partido Conservador de Colombia/ Conservative Party of Colombia Partido Liberal de Colombia/Liberal Party of Colombia

A-BM

10/2/07

3:40 PM

Page 279

POLITICAL PARTIES

Costa Rica PAC PLN PUSC RD Dominican BIP Republic BIS PLD PRD or PRDrd PRSC Ecuador

DP ID MUPP-NP

PRE PRIAN

PSC PSP El Salvador ARENA FMLN

PCN PDC Guatemala FDNG

279

Partido de Acción Ciudadana/Citizen Action Party Partido de Liberación Nacional/National Liberation Party Partido de Unidad Social Cristiana/Party of Social Christian Unity Renovación Democrática/Democratic Renovation Bloque Independiente Peñagomista/ Independent Peñagomista Block Bloque Institucional Socialista Demócrata/ Socialist Democratic Institutional Block Partido de la Liberación Dominicana/ Dominican Liberation Party Partido Revolucionario Dominicano/ Dominican Revolutionary Party Partido Reformista Social Cristiano/Social Christian Reformist Party Democracia Popular/Popular Democracy Izquierda Democrática/Democratic Left Movimiento Unidad Plurinacional, Pachakutik–Nuevo País/New Nation–Pachakutik Plurinational Unity Movement Partido Roldosista Ecuatoriano/Roldosista Party of Ecuador Partido Renovador Institucional Acción Nacional/Party of National Action and Institutional Renovation Partido Social Cristiano/Social Christian Party Partido Sociedad Patriótica/Patriotic Society Party Alianza Republicana Nacionalista/ Nationalist Republican Alliance Frente Farabundo Martí para la Liberación Nacional/Farabundo Martí Front for National Liberation Partido de Conciliación Nacional/Party of National Conciliation Partido Demócrata Cristiano/Christian Democratic Party Frente Democrática Nueva Guatemala/ Democratic Front of New Guatemala

A-BM

10/2/07

3:40 PM

280

Page 280

POLITICAL PARTIES

FRG

Honduras

Mexico

Nicaragua

Panama

Paraguay

Peru

Frente Republicano Guatemalteco/ Republican Front of Guatemala GANA Gran Alianza Nacional/Grand National Alliance PAN or PANguat Partido de Avanzada Nacional/National Vanguard Party PSN Partido Solidaridad Nacional/National Solidarity Party PU Partido Unionista/Unionist Party PL or PLH Partido Liberal/Liberal Party PN or PNH Partido Nacional/National Party PUD Partido de Unificación Democrática/ Democratic Unification Party PAN Partido Acción Nacional/National Action Party PRD Partido de la Revolución Democrática/ Democratic Revolutionary Party PRI Partido Revolucionario Institucional/ Institutional Revolutionary Party FSLN Frente Sandinista de Liberación Nacional/ Sandinista Front of National Liberation PLC Partido Liberal Constitucionalista/ Constitutionalist Liberal Party PA Partido Arnulfista/Arnulfista Party PRD or PRDpa Partido Revolucionario Democrático/ Democratic Revolutionary Party PS or PSp Partido Solidaridad/Solidarity Party ANR Asociación Nacional Republicana/National Republican Association MPQ Movimiento Patria Querida/Movement of the Beloved Nation PEN Partido Encuentro Nacional/National Reunion Party PLRA Partido Liberal Radical Auténtico/ Authentic Radical Liberal Party UNACE Unión Nacional de Ciudadanos Éticos/ National Union of Ethical Citizens FIM Frente Independiente Moralizador/ Independent Moralizing Front PAP Partido Aprista Peruano/Aprista Party of Peru PP Perú Posible/Peru Possible UN Unidad Nacional/National Unity

A-BM

10/2/07

3:40 PM

Page 281

POLITICAL PARTIES

Uruguay

Venzuela

EP/FA PC PN AD COPEI

MAS MVR

281

Encuentro Progresista/Frente Amplio/ Progressive Alliance/Broad Front Partido Colorado/Colorado Party Partido Nacional/National Party Acción Democrática/Democratic Action Comité de Organización Política Electoral Independiente/Committee for Independent Political Electoral Organization Movimiento al Socialismo/Movement for Socialism Movimiento V República/V Republic Movement

A-BM

10/2/07

3:40 PM

Page 282

A-BM

10/2/07

3:40 PM

Page 283

Bibliography

Adelantado, José, ed. (2000). Transformaciones del estado de Bienestar. Barcelona: Icaria. Agosto, Gabriela, and Francisco Cueto (2001). “República Dominicana.” In M. Alcántara and F. Freidenberg, eds., Partidos políticos de América Latina: Centroamérica, México y República Dominicana, pp. 615–698. Salamanca: Ediciones Universidad de Salamanca. Ajenjo, Natalia (1999). “La cultura política de las elites como condición para la democracia: Los casos de Chile y México.” Paper delivered at the VI encuentro de Latinoamericanistas en España, Cáceres, Spain. Alcántara, Manuel (1991). “La relación izquierda-derecha en la política latinoamericana.” Leviatán 43–44, pp. 73–93. ——— (1994). “Análisis comparado del papel de los partidos políticos en los procesos de transición política.” In Silvia Dutrenit and Leonardo Valdés, eds., El fin de siglo y los partidos políticos en América Latina. Mexico City: Instituto Mora. ——— (1995a). “La elite parlamentaria latinoamericana y el continuo izquierdaderecha.” In W. Hofmeister and J. Thesing, eds., La transformación de los sistemas políticos en América Latina, pp. 385–410. Buenos Aires: CIEDLA. ——— (1995b). “El estudio de las elites parlamentarias en América Latina.” Inguruak: Revista Vasca de Sociología y Ciencia Política, no. 13, pp. 25–37. ——— (1997). “Democracia y valores democráticos en la clase política latinoamericana.” In Helena González and Heidulf Schmidt, eds., Democracia para una nueva sociedad (Modelo para armar). Caracas: Nueva Sociedad. ——— (1999). Sistemas políticos de América Latina. Madrid: Tecnos. ——— (2000a). “De la perception par les elites parlamentaires latino-américaines du rôle des États-Unis dans la région.” In Isabelle Vagnoux, ed., Les États-Unis et les élites latino-américaines, pp. 141–154. Aix-en-Provence: Publications de l’Université de Provence. ——— (2000b). South American Legislatures: Thinking About Economic Integration and Defense Policy. Washington, D.C.: CSIS. ——— (2002).“Experimentos de democracia interna: Las primarias de partidos en América Latina.” Working Paper no. 293. Hellen Kellogg Institute for International Studies, University of Notre Dame, Notre Dame, Indiana. ——— (2003). Sistemas políticos de América Latina. 3rd ed. Madrid: Tecnos. ——— (2004). “Quality of Democracy or Quality of Politics?” In Guillermo O’Donnell, Jorge Vargas Cullell, and Osvaldo M. Iazzetta, eds., The Quality of Democracy: Theory and Applications, pp. 234–238. Notre Dame, Ind.: University of Notre Dame Press.

283

A-BM

10/2/07

3:40 PM

284

Page 284

BIBLIOGRAPHY

——— (2004a). ¿Instituciones o máquinas ideológicas? Origen, programa y organización de los partidos latinoamericanos. Barcelona: ICPS. ——— (2004b). “Partidos en América Latina: Precisiones conceptuales, estado actual y retos futuros.” Documentos CIDOB, no. 3. Barcelona: Fundación CIDOB. Alcántara, Manuel, and Flavia Freidenberg (1997). “Elites parlamentarias e integración subregional en América Latina.” America Latina Hoy, no. 17, pp. 11–17. ———, eds. (2001). Partidos políticos de América Latina. Vols. 1–3. Salamanca: Ediciones Universidad de Salamanca. Alcántara, Manuel, and Iván Llamazares (1997). “El análisis de los diputados latinoamericanos en el contexto de los estudios sobre la clase política: Características, objetivos y estrategias de investigación.” América Latina Hoy, no. 16 (August): pp. 15–28. Alcántara, Manuel, and Juan Pablo Luna (2004). “Ideología y competencia partidista en dos post-transiciones: Chile y Uruguay en perspectiva comparada.” Revista de Ciencia Política 24, no. 1, pp. 128–168. Alcántara, Manuel, Mercedes García Montero, and Francisco Sánchez (2005). Funciones, procedimientos y escenarios: Un análisis del poder legislativo en América Latina. Salamanca: Ediciones de la Universidad de Salamanca. Alcántara, Manuel, and Marisa Ramos (2000). “Los diputados latinoamericanos y las opciones de la inserción internacional: El caso de la Comunidad Iberoamericana.” In Francisco Rojas Arevena, ed., Las Cumbres Iberoamericanas: Una mirada global. Santiago and Caracas: FLACSO and Nueva Sociedad. Alfaro Salas, Sergio (2001). “Costa Rica.” In Manuel Alcántara and Flavia Freidenberg, eds., Partidos políticos de América Latina: Centroamérica, México y República Dominicana, pp. 33–134. Salamanca: Ediciones de la Universidad de Salamanca. Almond, Gabriel, and Sydney Verba (1965). The Civic Culture, Political Attitudes and Democracy in Five Nations: An Analytic Study. Boston: Little, Brown. Altman, David, and Aníbal Pérez-Liñán (1999). “Más allá de la poliarquía: Una aproximación a la calidad de las democracias.” Revista Uruguaya de Ciencia Política, no. 11, pp. 83–105. ——— (2002). “Assessing the Quality of Democracy: Freedom, Competitiveness, and Participation in Eighteen Latin American Countries.” Democratization 9, no. 2, pp. 85–100. Ames, Barry (1992). “Electoral Strategy Under Open-List Proportional Representation.” American Journal of Political Science 2, no. 39, pp. 406–433. ——— (2000). “Disciplina partidaria en la legislatura brasileña.” Política y Gobierno 7, no. 1, pp. 15–61. Amorin Neto, Octavio, and Fabiano G. M. Santos (1997). “The Executive Connection: Explaining the Puzzles of Party Cohesion in Brazil.” Paper delivered at the LASA Congress, Guadalajara, Mexico. Anderson, Lee (1968). “Organizational Theory and the Study of State and Local Parties.” In William J. Crotty, ed., Approaches to the Study of Party Organization. Boston: Allyn and Bacon. Angell, Alan (2003). “Party Change in Chile in Comparative Perspective.” Revista de Ciencia Política 28, no. 2, pp. 88–108. APSA. Committee on Political Parties (1950). “Towards a More Responsible TwoParty System.” American Political Science Review (September). Arnold, Charles (2003). “Candidate Quality and Candidate Emergence: An

A-BM

10/2/07

3:40 PM

Page 285

BIBLIOGRAPHY

285

Exploratory Study Utilizing a New Method for Measuring Quality.” Paper delivered at the meeting of the American Political Science Association, Philadelphia, Pennsylvania. Aron, Raymond (1989). Estudios sociológicos. Madrid: Espasa Calpe. Barnes, Samuel (1977). Representation in Italy: Institutionalized Tradition and Electoral Choice. Chicago: University of Chicago Press. Bartolini, Stefano (1998). “Partidos y sistemas de partidos.” In Gianfranco Pasquino, Stefano Bartolini, Maurizio Cotta, Leonardo Morlino, Angelo Panebianco, eds., Manual de Ciencia Política, pp. 217–264. Madrid: Alianza Universidad. Beck, Ulrich (1994). La sociedad del riesgo. Barcelona: Paidos. ——— (1998). ¿Qué es la globalización? Falacias del globalismo, respuestas a la globalización. Barcelona: Paidos. Beer, Carolina (2004). Electoral Competition and Institutional Change in Mexico. Notre Dame, Ind.: University of Notre Dame Press. Beltrán, Virgilio Rafael (1970). El papel político y social de las fuerzas armadas en América Latina. Caracas: Monte Ávila Editores. Blondel, Jean (1978). Political Parties: A Genuine Case for Discontent? London: Oxford University Press. Bobbio, Norberto (1995). Derecha e Izquierda: Razones y significados para una distinción política. Madrid: Taurus. Botella, Joan (1997). “Parlamento y carreras políticas.” In M. Ramírez, ed., El parlamento a debate. Madrid: Trotta. Bowler, S., David M. Farrel, and Richard Katz (1999). Party Discipline and Parliamentary Government. Columbus: Ohio State University Press. Brady, David, Kara Buckley, and Douglas Rivers (1999). “The Roots of Careerism in the U.S. House of Representatives.” Legislative Studies Quarterly 24, no. 4, pp. 489–510. Brutonel, Michael, Richard Gunther, and John Higley (1992). “Introduction: Elite Transformations and Democratic Regimes.” In Elites and Democratic Consolidation in Latin America and Southern Europe. Cambridge: Cambridge University Press. Burns, Nancy, Laura Evans, Gerald Gamm, and Corrine McConnaughy (2004). “Pockets of Expertise: Careers and Professionalism in Twentieth-Century State Legislatures.” Paper delivered at the meeting of the American Political Science Association, Chicago, Illinois. Calvert, Peter (2004). “Executive Leadership and Legislative Assemblies: Latin America.” The Journal of Legislative Studies 10, nos. 2–3, pp. 218–229. Carey, John. (1998). Term Limits and Legislative Representation. New York: Cambridge University Press. ——— (1999). “Partidos, coaliciones y el Congreso chileno en los años noventa.” Política y Gobierno, no. 2, pp. 365–405. ——— (2000). “Party Unity in Legislative Voting.” Paper delivered at the meeting of the American Political Science Association, Washington, D.C. ——— (2001). “Party and Coalition Unity in Legislative Voting.” Paper delivered at the meeting of the American Polítical Science Association, Washington, D.C. ——— (2002). “Getting Their Way, or Getting in the Way? Presidents and Party Unity in Legislative Voting.” Paper delivered at the Comparative Politics Workshop, University of Chicago, Chicago, Illinois. Caselli, Francesco, and Massimo Morelli (2004). “Bad Politicians.” Journal of

A-BM

10/2/07

3:40 PM

286

Page 286

BIBLIOGRAPHY

Public Economics 88, nos. 3–4, pp. 759–782. Available at http://ideas.repec. org/p/cpr/ceprdp/2402.html#author. Castañeda, Jorge G. (1995). La utopía desarmada: Intrigas, dilemas y promesas de la izquierda en América Latina. México: Editorial Joaquín Mortiz, Grupo Planeta. Castells, Manuel (1998). La era de la información. Vol. 2. Madrid: Alianza. Cavarozzi, Marcelo (1991). “Más allá de las transiciones a la democracia en América Latina.” Revista de Estudios Políticos, no. 74, pp. 85–112. ——— (2002). Autoritarismo y democracia. Buenos Aires: Eudeba. Cavarozzi, Marcelo, and Juan Abal Medina, eds. (2002). El asedio a la política: Los partidos latinoamericanos en la era neoliberal. Buenos Aires: Politeia. CEPAL (2003). Panorama Social de América Latina 2002–2003 (August). Cheibub, José Antonio, and Adam Przeworski (1999). “Democracy, Elections and Accountability for Economic Outcomes.” In Adam Przeworski, Susan Stokes, and Bernard Manin, eds., Democracy, Accountability, and Representation. New York: Cambridge University Press. Chin, Michelle L., and Michelle M. Taylor-Robinson (2005). “The Rules Matter: An Experimental Study of the Effects of Electoral Systems on Shifts in Voters’ Attention.” Electoral Studies 24, no. 3, pp. 465–483. Close, David, ed. (1995). Legislatures and the New Democracies in Latin America. Boulder, Colo.: Lynne Rienner Publishers. Cohen, Joshua (1989). “Deliberation and Democratic Legitimacy.” In A. Hamlin and B. Pettit, eds., The Good Polity. Oxford: Oxford University Press. Collier, David, and Steven Levistky (1997). “Democracy with Adjectives: Conceptual Innovation in Comparative Research.” World Politics 49, no. 3, pp. 430–451. Colomer, Josep M. (2001). Instituciones políticas. Barcelona: Ariel. Colomer, Josep M., and Luis E. Escatel (2005). “La dimensión izquierda-derecha en América Latina.” Desarrollo Económico 44, no. 177, pp. 123–136. Combellas, Ricardo (1985). COPEI: Ideología y liderazgo. Caracas: Editorial Ariel. Conaghan, Catherine (1995). “Políticos versus partidos: Discordia y desunión en el sistema de partidos ecuatoriano.” In Scott Mainwaring and Timothy Scully, eds., La construcción de las instituciones democráticas, pp. 355–447. Santiago: CIEPLAN. Converse, Philip, and Roy Pierce (1986). Political Representation in France. Cambridge: Harvard University Press. Coppedge, Michael (1997). “A Classification of Latin American Political Parties.” Working Paper no. 24. Kellogg Institute, Notre Dame University, Notre Dame, Indiana. ——— (1998). “The Dynamic Diversity of Latin American Party Systems.” Party Politics 4, no. 4, pp. 547–568. Corral, Margarita (2004). “El focus y el estilo de representación parlamentaria en América Central: Un análisis de las percepciones y actitudes de los diputados.” Paper delivered at the X encuentro de Latinoamericanistas Españoles, Salamanca, Spain. Correa, Enrique, and José A. Viera-Gallo (1986). Iglesia y dictadura. Chile: Chile y América. Cotta, Maurizio, and Heinrich Best (2000). Parliamentary Representatives in Europe 1848–2000: Legislative Recruitment and Careers in Eleven European Countries. Oxford: Oxford University Press. Crespo, Ismael (1997). “La cultura política de la clase parlamentaria centroamericana.” Revista Mexicana de Sociología 59, no. 1, pp. 115–139.

A-BM

10/2/07

3:40 PM

Page 287

BIBLIOGRAPHY

287

——— (1998). “Clase política, diseños institucionales y gobernabilidad en Centroamérica.” In J. Botella and J. M. Sanahuja, eds., Centroamérica después de la crisis. Barcelona: ICPS. Crewe, Ivor (1985). Electoral Change in Western Democracies: Patterns and Sources of Electoral Volatility. New York: St. Martin’s Press. Crisp, Brian F., and Scott W. Desposato (2004). “Constituency Building in Multimember Districts: Collusion or Conflict?” Journal of Politics 66, no. 1, pp. 136–156. Crisp, Brian F., Maria C. Escobar-Lemmon, Bradford S. Jones, Mark P. Jones, and Michelle M. Taylor-Robinson (2004). “Vote-Seeking Incentives and Legislative Representation in Six Presidential Democracies.” Journal of Politics 66, no. 3, pp. 823–846. Cutright, Philips (1963). “National Political Development: Its Measurement and Social Correlates.” In Nelson W. Polsby, Robert A. Dentler, and Paul A. Smith, eds., Politics and Social Life, pp. 569–582. Boston: Houghton Mifflin. Czudnoswski, Mosheon, ed. (1983). Political Elites and Social Change: Studies of Elite Roles and Attitudes. DeKalb: Northern Illinois University Press. Dahl, Robert A. (1971). Polyarchy: Participation and Opposition. New Haven, Conn.: Yale University Press. Dal Bó, Ernesto, Pedro Dal Bó, and Rafael Di Tella (2002). “Plata o Plomo”: Bribe and Punishment in a Theory of Political Influence.” Working Paper no. 28. Department of Economics, Brown University, Providence, Rhode Island. Available at http://www.econ.brown.edu/2002/wp2002-28.pdf. Dalton, Russell (1985). “Political Parties and Political Representation: Party Supporters and Party Elites in Nine Nations.” Comparative Political Studies 18, no. 3, pp. 267–299. de Swaan, Abram (1973). Coalition Theories and Cabinet Formation. Amsterdam: Elsevier. Del Campo, Esther, and Elena Martínez Barahona (2000). “La derecha en Chile.” Paper delivered at the LASA Congress, Miami, Florida. Del Campo, Esther, and Marisa Ramos (1997). “La paradoja partidista: Institucionalización y representación en los partidos políticos latinoamericanos.” America Latina Hoy, no. 16, pp. 29–39. Delgado, Irene (2000). “Elites políticas y vida parlamentaria: Actividades y motivaciones de los diputados españoles.” In A. Martínez, ed., El Congreso de los diputados en España: Funciones y rendimientos, pp. 295–342. Madrid: Tecnos. Delgado Gómez-Escalonilla, Lorenzo (1988). Diplomacia franquista y política cultural hacia Iberoamérica, 1939–1953. Madrid: Centro de Estudios Históricos. Di Tella, Torcuato S. (1993). Historia de los partidos políticos en América Latina. Buenos Aires: Siglo XX, Fondo de Cultura Económico. Diamond, Larry (1992). “Economic Development and Democracy Reconsidered.” In Gary Marks and Larry Diamond, eds., Reexamining Democracy: Essays in Honor of Seymour Martin Lipset. Newbury Park, Calif.: Sage. ——— (1999). Developing Democracy Toward Consolidation. Baltimore and London: Johns Hopkins University Press. Diamond, Larry, Jonathan Hartlyn, Juan J. Linz, and Seymour Martin Lipset (1999). Democracy in Developing Countries: Latin America. 2nd ed. Boulder and London: Lynne Rienner Publishers. Diamond, Larry, and Leonardo Morlino (2004). “The Quality of Democracy: An Overview.” Journal of Democracy 15, no. 4, pp. 20–31. Diaz, Borja, and Sebastián Linares (2005). “Fortalecimiento de la independencia

A-BM

10/2/07

3:40 PM

288

Page 288

BIBLIOGRAPHY

judicial en centroamérica: Un balance tras veinte años de reformas.” América Latina Hoy, no. 39, pp. 47–96. Dix, Robert H. (1989). “Cleavage Structures and Party Systems in Latin America.” Comparative Politics 22, no. 1, pp. 23–37. Domínguez, Jorge (1998). “Free Politics and Free Markets in Latin America.” Journal of Democracy 9, no. 4, pp. 10–84. Downs, Anthony (1957). An Economic Theory of Democracy. New York: Harper. Duverger, Maurice (1951). Los partidos políticos. Mexico DF: Fondo de Cultura Económico. Eckstein, Susan, and Timothy Wickham-Crowley (2003). Struggles for Social Rights in Latin America: Race, Gender, Labor and Subsistence. New York and London: Routledge. Eldersveld, Samuel (1989). Political Elites in Modern Societies. Ann Arbor: University of Michigan Press. Epstein, David, David Brady, Sadafumi Kawato, and Sharyn O’Halloran (1997). “A Comparative Approach to Legislative Organization: Careerism and Seniority in the United States and Japan.” American Journal of Political Science 41, pp. 965–998. Esaiasson, Peter, and Knut Heidar (2000). Beyond Westminster and Congress: The Nordic Experience. Columbus: Ohio State University Press. Esaiasson, Peter, and Soren Holmberg (1996). Representation from Above: Members of Parliament and Representative Democracy in Sweden. Aldershot: Dartmouth Publishing. Escobar-Lemmon, María, and Erika Moreno (forthcoming 2007). “Mejor solo que mal acompañado: Political Entrepreneurs in Colombia.” In S. Morgenstern and P. M. Siavelis, eds., Pathway to Power: Political Recruitment and Candidate Selection in Latin America. University Park: Pennsylvania State University. Etzioni-Halevy, Eva (1993). The Elite Connection: Problems and Potential of Western Democracy. Cambridge, Eng.: Polity Press. Eulau, Heinz, John Wahlke, William Buchanan, and Leroy Ferguson (1959). “The Role of Representative: Some Empirical Observations on the Theory of Edmund Burke.” American Political Science Review, no. 53, pp. 742–756. Farah, Barbara (1980). “Political Representation in West Germany: The Institution and Maintenance of Mass-Elite Linkages.” Doctoral dissertation, Department of Political Science, University of Michigan, Ann Arbor, Michigan. Figuereido, Argelina, and Fernando Limongi (1995). “Partidos políticos na Câmara dos Deputados: 1989–1994.” Dados 38, no. 3. Fiorina, Morris P. (1994). “Divided Government in the American States: A Byproduct of Legislative Professionalism?” American Political Science Review 88, no. 2, pp. 304–316. ——— (1999). “Further Evidence of the Partisan Consequences of Legislative Professionalism.” American Journal of Political Science 43, no. 3, pp. 974–977. Foweraker, Joe (1998). “Institutional Designs, Party Systems, and Governability: Differentiating the Presidential Regimes of Latin America.” British Journal of Political Science, no. 28, pp. 651–676. Fowler, Linda L. (1993). Candidates, Congress, and the American Democracy. Ann Arbor: University of Michigan Press. Freidenberg, Flavia (1999). “Actitudes a la democracia de las elites parlamentarias en Argentina, Ecuador, Perú y Uruguay.” Master ’s thesis, Instituto Interuniversitario de Estudios de Iberoamérica y Portugal, Universidad de Salamanca, Salamanca, Spain.

A-BM

10/2/07

3:40 PM

Page 289

BIBLIOGRAPHY

289

——— (2000). “Posiciones ideológico-programáticas de los políticos y los partidos en Ecuador.” Paper delivered at the seminar “Party Politics in Latin America,” School of International Studies of the University of Miami, Miami, Florida. ——— (2003). Jama, caleta y camello: Las estrategias de Abdalá Bucaram y el PRE para ganar las elecciones. Quito: Universidad Andina Simón Bolívar. Freidenberg, Flavia, and Manuel Alcántara (2001). “Cuestión regional y política en Ecuador: Partidos de vocación nacional y apoyo regional.” América Latina Hoy, no. 27, pp. 123–152. Freidenberg, Flavia, and Francisco Sánchez López (2001). “Como se escolhe um candidato a presidente? Regras e práticas nos partidos políticos da América Latina.” Opiniao Pública 8, no. 2, pp. 158–188. Gabriel, K. Ruben (1971). “The Biplot-Graphic Display of Matrices with Application to Principal Component Analysis.” Biometrika, no. 58, pp. 453–467. Galindo, Purificación (1986). “Una alternativa de representación simultánea: HJbiplot.” Questíio 10, no. 1, pp. 13–23. Galindo, Purificación, and C. Cuadras (1986). “Una extensión del método biplot y su relación con otras técnicas.” Publicaciones de Bioestadística y Biomatemática, no. 17. Galindo, Purificación, M. J. Fernández, C. Ávila, A. Del Río, and M. Fernández (1999). “El análisis HJ-biplot como herramienta de un ecosistema acuático.” Statistic and Environment, pp. 85–94. Gangas, Pilar (2000). “Los diputados españoles (1977–1996).” In A. Martínez, ed., El Congreso de los Diputados en España: Funciones y rendimientos, pp. 271–294. Madrid: Tecnos. García Díez, Fátima (2000). “Partidos políticos y reforma electoral en América Central y Panamá.” Paper delivered at the LASA Congress, Miami, Florida. ——— (2005). Procesos contemporáneos de reforma electoral en América Central y Caribe. Salamanca: AECI–Fundación General de la Universidad de Salamanca. García Montero, Mercedes, and Patricia Marenghi (2003). “Entre lo que dicen y lo que hacen: La congruencia en la disciplina legislativa de los partidos latinoamericanos.” Paper delivered at the VII Congreso de Ciencia Política y de la Administración, Barcelona, Spain. Garretón, M. Antonio, ed. (2003). Latin America in the Twenty-First Century: Toward a New Sociopolitical Matrix. Miami: North/South Center Press. Gibson, Edward (1996). Class and Conservative Parties: Argentina in Comparative Perspective. Baltimore: Johns Hopkins University Press. Gil, Federico (1975). Latinoamérica y Estados Unidos: Dominio, cooperación y conflicto. Madrid: Tecnos. Gilberg, Trond, ed. (1989). Coalition Strategies of Marxist Parties. Durham, N.C., and London: Duke University Press. Gomà, Ricard (1998). “Rendimientos sustantivos: Políticas económicas y sociales en la CentroAmérica posrevolucionaria.” In Ana S. Cardenal and Salvador Marti, eds., América Central las democracias inciertas. Madrid: Tecnos. Gomà, Ricard, and Jacint Jordana, eds. (2004). Descentralización y políticas sociales en América Latina. Barcelona: Fundació CIDOB. Goodman, Louis (1990). “Political Parties and the Political Systems of Central America.” In L. Godman, W. M. Leogrande, and J. Mendelson Forman, eds., Political Parties and Democracy in Central America, pp. 3–16. Boulder, Colo.: Westview Press. Gray, John (2000). Falso amanecer: Los engaños del capitalismo global. Barcelona: Paidós.

A-BM

10/2/07

3:40 PM

290

Page 290

BIBLIOGRAPHY

Grayson, George (1968). El Partido Demócrata Cristiano chileno. Buenos Aires: Editorial Francisco de Aguirre. Gutiérrez Saxe, M., and J. Vargas Cullell (1998). “Auditoría Ciudadana sobre la calidad de la democracia: Propuesta para su ejecución en Costa Rica.” Costa Rica: Iniciativa del Proyecto “Estado de la región en desarrollo humano sostenible,” PNUD and European Union. Habermas, Jürgen (1998). Facticidad y validez. Madrid: Trotta. Hagopian, Frances (2005). “Conclusions: Government Performance, Political Representation, and Public Perceptions of Contemporary Democracy in Latin America.” In Frances Hagopian and Scott Mainwaring, eds., The Third Wave of Democratization in Latin America: Advances and Setbacks, pp. 319–361. Cambridge: Cambridge University Press. Hagopian, Frances, and Scott Mainwaring, eds. (2005). The Third Wave of Democratization in Latin America: Advances and Setbacks. Cambridge: Cambridge University Press. Hallerbeg, Mark, and Patrick Marier (2001). “Executive Authority, the Personal Vote, and Budget Discipline in Latin American and Caribbean Countries.” Working Paper no. 17. Bonn: Zentrum für Europäische Integrationsforschung, Rheinische Friedrich-Wilhelms Universität, Bonn, Germany. Hawkins, Kirk, and Scott Morgenstern (2003). “Cohesion of Legislators in Latin America: Patterns and Explanations.” Unpublished manuscript. Held, David (2001). Modelos de democracia. Madrid: Alianza Editorial. Held, David, Anthony McGrew, David Goldblatt, and Jonathan Perraton (1999). Global Transformations: Politics, Economics, and Culture. Cambridge, Eng.: Polity Press. Hibbing, John R. (1991). “Congressional Careers: Contours of Life in the U.S. House of Representatives.” Chapel Hill: University of North Carolina Press. ——— (1999). “Legislative Careers: Why and How We Should Study Them.” Legislative Studies Quarterly 24, no. 2, pp. 149–171. Higley, John, and Richard Gunther (1992). Elites and Democratic Consolidation in Latin America and Southern Europe. New York: Cambridge University Press. Higley, John, Jan Pakulski, and Wlodzimierz Wesolowski, eds. (1998). Postcommunist Elites and Democracy in Eastern Europe. Houndmills, Eng.: Macmillan. Hinich, M., and M. C. Munger (1996). Ideology and the Theory of Democratic Choice. Ann Arbor: University of Michigan Press. Hinzpeter, Ximena, and Carla Lehmann, S.B. (1999). “Las alas de la DC: La fuerte tensión interna del principal partido.” Puntos de Referencia, no. 211. Holmberg, Soren (1989). “Political Representation in Sweden.” Scandinavian Political Studies, no. 12, pp. 1–36. ——— (1999). “Collective Policy Congruence Compared.” In Warren Miller et al., Policy Representation in Western Democracies. Oxford: Oxford University Press. ——— (2000). “Issue Agreement.” In Peter Esaiasson and Knut Heidar, Beyond Westminster and Congress: The Nordic Experience. Columbus: Ohio State University Press. Hunneus, Carlos (2003). “A Highly Institutionalized Political Party: Christian Democracy in Chile.” In Scott Mainwaring and Timothy Scully, eds., Christian Democracy in Latin America: Electoral Competition and Regime Conflicts, pp. 121–161. Stanford, Calif.: Stanford University Press.

A-BM

10/2/07

3:40 PM

Page 291

BIBLIOGRAPHY

291

Huntington, Samuel (1965). “Political Development and Political Decay.” World Politics, no. 17, pp. 386–430. ——— (1991). La Tercera Ola: La democratización a finales del siglo XX. Mexico City: Paidós. Ibarra, Pedro, Salvador Martí, and Ricard Gomà, eds. (2002). Creadores de democracia radical: Movimientos sociales y redes de políticas públicas. Barcelona: Icaria. IDD-Lat (2004). Índice de desarrollo democrático. Buenos Aires: Konrad Adenauer Stiftung. Inglehart, Ronald (1984). “The Changing Structure of Political Cleavages in Western Society.” In Russell J. Dalton, Scott Flanagan, and Paul Allen Beck, eds., Electoral Change in Advanced Industrial Democracies: Realignment or Dealignment? pp. 25–69. Princeton, N.J.: Princeton University Press. Irwin, Galen, and Jacques Thomassen (1975). “Issue-Consensus in a Multi-Party System: Voters and Leaders in the Netherlands.” Acta Politica, no. 10, pp. 389–420. Iturrieta, Anibal (1994). “El primer nacionalismo argentino.” In Aníbal Iturrieta, ed., El pensamiento político argentino contemporáneo. Buenos Aires: Grupo Editor Latinoamericano. Janda, Kenneth (1993). “Comparative Political Parties: Research and Theory.” In A. W. Finifter Political Science: The State of the Discipline II, pp. 163–191. Washington, D.C: American Political Science Association. Jerez, Miguel (1997). “La élite parlamentaria.” In M. Ramírez, ed., El parlamento a debate. Madrid: Trotta. Jones, Mark P. (1995). Electoral Laws and the Survival of Presidential Democracies. Notre Dame, Ind.: University of Notre Dame Press. ——— (2001). “Carreras políticas y disciplinas partidarias en la Cámara de Diputados argentina.” POSTData, no. 7, pp. 189–230. ——— (2002). “Legislator Behaviour and Executive-Legislative Relations in Latin America.” Latin American Research Review 37, no. 3, pp. 176–188. ——— (2007). “The Recruitment and Selection of Legislative Candidates in Argentina.” In Morgenstern and Siavelis, eds., Pathway to Power. Jones, Mark P., Sebastian Saiegh, Pablo T. Spiller, and Mariano Tommasi (2002). “Amateur Legislators–Professional Politicians: The Consequences of PartyCentered Electoral Rules in a Federal System.” American Journal of Political Science 46, no. 3, pp. 656–669. Jordana, Jacint (2005). “Agencias reguladoras no tan inocentes.” La vanguardia (Barcelona). May 26. Katz, Jonathan N., and Brian R. Sala (1996). “Careerism, Committee Assignments, and the Electoral Connection.” American Political Science Review 90, no. 1, pp. 21–33. Katz, Richard (1997). “Representational Roles.” European Journal of Political Research 32. Kinder, Donald R. (1983). Presidential Traits: Report to the NES Board of Overseers. Ann Arbor: Center for Political Studies, University of Michigan. Kirchheimer, Otto (1966). “The Transformation of the Western European Party Systems.” In J. Lapalombara and M. Weiner, eds., Political Parties and Political Development, pp. 177–200. Princeton, N.J.: Princeton University Press. Kitschelt, Herbert (2000). “Party Cohesion, Accountability and Responsiveness: Democratic Institutions and Political-Economic Change.” Paper delivered at the meeting of the American Political Science Association, Washington, D.C.

A-BM

10/2/07

3:40 PM

292

Page 292

BIBLIOGRAPHY

——— (2001). “Panoramas de intermediación de intereses políticos: Movimientos sociales, grupos de interés y partidos a comienzos del siglo XXI.” In R. Maiz, ed., Construcción de Europa, Democracia y Globalización. Vol. 1. Santiago de Compostela, Spain: Universidade de Santiago de Compostela. Kitschelt, Herbert, Zdenka Mansfeldova, Radoslaw Markowski, and Gabor Toka (1999). Post-Communist Party Systems: Competition, Representation and Inter-Party Cooperation. New York: Cambridge University Press. Kitschelt, Herbert, and Elizabeth Zechmeister (2003). Patterns of Party Competition and Electoral Accountability in Latin America. Paper delivered at the annual meeting of the American Political Science Association, Philadelphia, Pennsylvania. Klingemann, Hans-Dieter, Richard Hofferbert, and Ian Budge (1993). Parties, Policies and Democracy. Boulder, Colo.: Westview Press. Kornberg, Allan, and Norman Thomas (1965). “The Political Socialization of National Legislative Elites in the United States and Canada.” Journal of Politics 27, no. 4, pp. 761–775. Kousser, Thad (2005). Term Limits and the Dismantling of State Legislative Professionalism. New York: Cambridge University Press. Kreft, Ita, and Jan De Leeuw (1998). Introducing Multilevel Modeling. London: Sage. Krehbiel, Keith (2004). “Legislative Organization.” Journal of Economic Perspectives 18, no. 1, pp. 113–128. Langston, Joy. (2007). “Legislative Recruitment in Mexico.” In Morgenstern and Siavelis, eds., Pathway to Power: Political Recruitment and Candidate Selection in Latin America. University Park: Penn State University Press. Leiras, Marcelo (2002). “Instituciones de gobierno, partidos y representación política en las democracias de América Latina: Una revisión de la literatura reciente.” Contribuciones 1, no. 73, pp. 7–57. Leoni, Eduardo, Carlos Pereira, and Lucio Renno (2004). “Political Survival Strategies: Political Career Decisions in the Brazilian Chamber of Deputies.” Journal of Latin American Studies, 36, no. 1, pp. 109–130. Levitsky, Steven (1998). “Institutionalization and Peronism: The Concept, the Case and the Case for Unpacking the Concept.” Party Politics 4, no. 1, pp. 77–92. ——— (2003). Transforming Labor-Based Parties in Latin America: Argentine Peronism in Comparative Perspective. New York: Cambridge University Press. Limongi, Fernando, and Argelina Figueiredo (1995). “Partidos políticos na Camara dos Deputados: 1989–1994.” Dados 38, no. 3, pp. 497–527. Linz, Juan J. (2002). “Parties in Contemporary Democracies: Problems and Paradoxes.” In R. Gunther, J. R. Montero, and J. J. Linz, eds., Political Parties: Old Concepts and New Challenges, pp. 292–317. Oxford: Oxford University Press. Linz, Juan J., and Alfred Stepan (1996). Problems of Democratic Transition and Consolidation: Southern Europe, South America and Post-Communist Europe. Baltimore, Md., and London: Johns Hopkins University Press. Linz, Juan J., and Arturo Valenzuela (1994). The Failure of Presidential Democracy. 2 vols. Baltimore, Md.: Johns Hopkins University Press. Lipset, Seymour Martin (1959). “Some Social Requisites of Democracy: Economic Development and Political Legitimacy.” American Political Science Review 53, pp. 69–105. Lipset, Seymour A., Paul Lazarsfeld, Allen Barton, and Juan Linz (1954). “The Psychology of Voting: An Analysis of Political Behaviour.” In Lindzey Garden, ed., Handbook of Social Psychology II, pp. 1124–1177. Cambridge, Mass.: Addison Wesley.

A-BM

10/2/07

3:40 PM

Page 293

BIBLIOGRAPHY

293

Lipset, Seymour Martin, and Stein Rokkan (1967). Party Systems and Voter Alignments: Cross-National Perspectives. New York: Macmillan. Lipset, Seymour M., and Aldo Solari, eds. (1967). Elites in Latin America. New York: Oxford University Press. Llamazares, Iván, and Rickard Sandell (2000). “Partidos politicos y dimensiones ideológicas en Argentina, Chile, México y Uruguay: Esbozo de un análisis espacial.” Polis, pp. 43–69. Llanos, Mariana, and Francisco Sánchez López (2004). “Councils of Elders? The Senates and Its Members in the Southern Cone.” Working Paper no. 17. Institut für Iberoamerika-Kunde, Hamburg, Germany. Loaeza, Soledad (1999). El Partido Acción Nacional: La larga marcha, 1939–1994; Oposición leal y partido de protesta. Mexico City: Fondo de Cultura Económica. Londegran, John, and Keith Poole (1996). “Does Income Promote Democracy?” World Politics 49, pp. 1–30. Loveman, Brian, and Thomas M. Davies, Jr. (1989). The Politics of Antipolitics: The Military in Latin America. 2nd ed. Lincoln: University of Nebraska Press. Lujambio, Alonso (1993). “Presidentes y congresos: Estados Unidos, la experiencia latinoamericana y el futuro mexicano.” Foro Internacional, no. 33, pp. 87–121. Luke, Douglas A. (2004). Multilevel Modeling: Quantitative Applications in the Social Sciences. Thousand Oaks, Calif.: Sage. Luna, Juan Pablo, and Elizabeth Zechmeister (2003). “Estructuración ideológica e izquierda en los sistemas de partidos latinoamericanos circa 1996–1998.” In Jorge Lanzaro, ed., La izquierda latinoamericana. Buenos Aires: CLACSO. ——— (2005). “Political Representation in Latin America: A Study of Elite-Mass Congruence in Nine Countries.” Comparative Political Studies, no. 38, pp. 388–416. Maddox, Jerome (2002). “Working Outside of the House (and Senate): Outside Careers and Partisan Bias in Professionalized State Legislatures.” Paper delivered at the meeting of the Annual American Political Science Association, Boston, Massachusetts. Madeira, Rafael Machado (2003). “O padrao de carreira politica dos deputados federais da arena baiana: Trajetorias estaveis e lealdades pessoais.” Revista Mediacoes 8, no. 2, pp. 83–117. Maestas, Cherie (2000). “Professional Legislatures and Ambitious Politicians: Policy Responsiveness of State Institutions.” Legislative Studies Quarterly 25, pp. 663–690. ——— (2003). “The Incentive to Listen: Progressive Ambition, Resources and Opinion Monitoring Among State Legislators.” Journal of Politics 65, no. 2, pp. 439–456. Maiguashca, Juan, ed. (1994). Historia y región en el Ecuador: 1830–1930. Quito: FLACSO-CEN. Mainwaring, Scott (1998). “Party Systems in the Third Wave.” Journal of Democracy 9, no. 3, pp. 67–82. ——— (1999). Rethinking Party Systems in the Third Wave of Democratization: The Case of Brazil. Stanford, Calif.: Stanford University Press. Mainwaring, Scott, Daniel Brinks, and Aníbal Pérez-Liñán (2001). “Classifying Political Regimes in Latin America, 1945–1999.” Studies in Comparative International Development 36, no. 1, pp. 37–65. Mainwaring, Scott, and Aníbal Pérez-Liñán (1997). “Party Discipline in the Brazilian Constitutional Congress.” Legislative Studies Quarterly, no. 4, pp. 453–483.

A-BM

10/2/07

3:40 PM

294

Page 294

BIBLIOGRAPHY

——— (2004a). “Nivel de desarrollo y democracia: El excepcionalismo latinoamericano (1945–1996).” América Latina Hoy 36, pp. 189–248. ——— (2004b). “Regional Effects and Region-wide Diffusion of Democracy: Why Regions of the World Are Important in Comparative Politics.” Paper delivered at the meeting of the American Political Science Association, Chicago, Illinois. ——— (2005). “Latin American Democratization Since 1978: Democratic Transitions, Breakdowns, and Erosions.” In Frances Hagopian and Scott Mainwaring, ed., The Third Wave of Democratization in Latin America: Advances and Setbacks, pp. 14–61. Cambridge: Cambridge University Press. Mainwaring, Scott, and Timothy Scully, eds. (1995). Building Democratic Institutions: Party Systems in Latin America. Stanford, Calif.: Stanford University Press. ——— (2003). Christian Democracy in Latin America: Electoral Competition and Regime Conflicts. Stanford, Calif.: Stanford University Press. Mainwaring, Scott, and Matthew Shugart, eds. (1997). Presidentialism and Democracy in Latin America. New York: Cambridge University Press. Mainwaring, Scott, and Mariano Torcal (2005). “La institucionalización de los sistemas de partidos y la teoría del sistema partidista después de la tercera ola democratizadora.” América Latina Hoy, no. 41, pp. 141–173. Manin, Bernard (1998). Los principios del gobierno representativo. Madrid: Alianza Editorial. Maor, Moshe (1997). Political Parties and Party Systems: Comparative Approaches and the British Experience. London: Routledge. Maravall, José María (1995). Los resultados de la democracia. Madrid: Alianza Editorial. ——— (2003). El control de los políticos. Madrid: Taurus. March, James G., and Johan P. Olsen (1984). “The New Institutionalism: Organizational Factors in Political Life.” American Political Science Review 78, no. 3, pp. 734–749. Marenco dos Santos, Andre (2001). “Sedimentacao de lealdades partidarias no Brasil: Tendencias e descompasaos.” Revista Brasileira de Ciencias Sociais 16, no. 45, pp. 69–83. Marenghi, Patricia. (2001). “Anexo: Organización de los partidos políticos; Dimensiones comparadas.” In Manuel Alcántara and Flavia Freidenberg, eds., Partidos políticos de América Latina. Salamanca: Ediciones Universidad de Salamanca. Martí, Salvador (2001). “¿Y después de las transiciones qué? Un balance y análisis de las teorías del cambio político.” Revista de Estudios Políticos, no. 113, pp. 101–124. Martínez, Rafael, ed. (2004). La elección presidencial mediante doble vuelta en Latinoamérica. Barcelona: ICPS. Martínez Barahona, Elena (2000). “La democracia interna en los partidos centroamericanos.” Paper delivered at the seminar “Political Parties in Latin America,” Institute of Latin American Studies (ILAS), London, England. Martínez Rodríguez, Antonia (1997). “Cultura política, gobernabilidad y elites parlamentarias en América Latina.” Revista Mexicana de Sociología 59, no. 1, pp. 89–115. Martínez Rosón, Mar (2004). “El concepto de profesionalización parlamentaria y su aplicación en los parlamentos centroamericanos.” Paper delivered at the II Congreso de la Asociación Latinoamericana de Ciencia Política (ALACIP), Mexico City.

A-BM

10/2/07

3:40 PM

Page 295

BIBLIOGRAPHY

295

Mascott, María de los Ángeles (2004). “Representantes y representados en México: ¿Mandato o delegación?” Metapolítica 8, pp. 30–42. Matthews, Donald, and Henry Valen (1999). Parliamentary Representation: The Case of the Norwegian Storting. Columbus: Ohio State University Press. May, John D. (1973). “Opinion Structure of Political Parties: The Special Law of Curvilinear Disparity.” Political Studies, no. 21, pp. 135–151. Mayhew, David (1974). Congress: The Electoral Connection. New Haven, Conn.: Yale University Press. McAllister, Ian (1991). “Party Elites, Voters and Political Attitudes: Testing Three Explanations of Mass-Elite Differences.” Canadian Journal of Political Science, no. 24, pp. 237–268. McCoy, Jennifer L., ed. (2000). Political Learning and Redemocratization in Latin America: Do Politicians Learn from Political Crises? Miami: North/South Center Press. McCurley, Carl, and Jeffrey J. Mondak (1995). “The Influence of Incumbents, Competence, and Integrity in U.S. House Elections.” American Journal of Political Science 39, no. 4, pp. 864–885. McDonald, Ronald H., and Mark Ruhl (1989). Party Politics and Elections in Latin America. Boulder, Colo.: Westview Press. Mendez Lago, Mónica, and Antonia Martinez Rodríguez (2002). “Political Representation in Spain: An Empirical Analysis of the Perception of Citizens and MPs.” Journal of Legislative Studies 8, no. 1, pp. 63–90. Meseguer, Covadonga (2004). La difusión de las privatizaciones en la OCDE y en América Latina: Un proceso de aprendizaje? Working Paper no. 1. CIDOB, Barcelona, Spain. Michels, Robert (1991 [1st ed., 1915]). Los partidos políticos: Un estudio sociológico de las tendencias oligárquicas de la democracia moderna. Buenos Aires: Amorrortu Editores. Middlebrook, Kevin J., ed. (2000a). Conservative Parties, the Right, and Democracy in Latin America. Baltimore, Md.: Johns Hopkins University Press. ——— (2000b). Introduction. In Middlebrook, ed., Conservative Parties, the Right and Democracy. Miguel, Luis Felipe. (2003). “Capital politico e carreira eleitoral: Algumas varia veis na eleicao para o Congresso brasileiro.” Revista de Sociologia e Politica 20, pp. 115–134. Miller, Warren, ed. (1999). Policy Representation in Western Democracies. Oxford: Oxford University Press. Miller, Warren, and Donald Stokes (1963). “Constituency Influence in Congress.” American Political Science Review 57, no. 1, pp. 165–177. Molina, José Enrique (2000). “Los sistemas electorales de América Latina.” Cuadernos de Capel 45. Molinelli, Guillermo (1991). Presidentes y Congresos en Argentina: Mitos y realidades. Buenos Aires: Grupo Editor Latinoamericano. Moncrief, Gary F. (1999). “Recruitment and Retention in U.S. Legislatures.” Legislative Studies Quarterly 24, no. 2, pp. 173–208. Moncrief, Gary F., Peverille Squire, and Malcom E. Jewell (2001). Who Runs for the Legislature? Upper Saddle River, N.J.: Prentice Hall. Mondak, Jeffrey J. (1995). “Competence, Integrity, and the Electoral Success of Congressional Incumbents.” Journal of Politics 57, no. 4, pp. 1043–1069. Montero, José Ramón, and Richard Gunther (2002). “Reviewing and Reassessing Parties.” Introduction to Richard Gunther, José Ramón Montero, and Juan J.

A-BM

10/2/07

3:40 PM

296

Page 296

BIBLIOGRAPHY

Linz, eds., Political Parties: Old Concepts and New Challenges, pp. 1–38. Oxford: Oxford University Press. Moraes, Juan Andrés. (2007). “Why Factions? Candidate Selection and Legislative Politics in Uruguay.” In Morgenstern and Siavelis, eds., Pathway to Power: Political Recruitment and Candidate Selection in Latin America. University Park: Penn State University Press. Morán, Maria Luz (1997). “Elites y cultura política en la España democrática.” In Pilar del Castillo and Ismael Crespo, eds., Cultura política: Enfoques teóricos y análisis empíricos. Valencia, Spain: Tirant lo Blanch. Moreno, Alejandro (1999). Political Cleavages, Issues, Parties and the Consolidation of Democracy. Boulder, Colo.: Westview Press. ——— (2004). El público y sus representantes: Opiniones de las elites y la sociedad en México.” Metapolítica 8, pp. 30–42. Morgenstern, Scott (2000). “Explaining Voting Unity in the Legislatures of the United States and Latin America.” Paper delivered at the meeting of the American Political Science Association, Washington, D.C. ——— (2001). “Organized Factions and Disorganized Parties: Electoral Incentives in Uruguay.” Party Politics 7, no. 2, pp. 265–286. ——— (2003). “Explicando la unidad de los parlamentarios en el Cono Sur.” In M. Alcántara and Elena Martínez-Barahona, eds., Política, dinero e institucionalización partidista en América Latina, pp. 357–388. Mexico DF: Universidad Iberoamericana. Morgenstern, Scott, and Benito Nacif, eds. (2002). Legislative Politics in Latin America. Cambridge: Cambridge University Press. Morgenstern, Scott, and Peter M. Siavelis (forthcoming 2007). Pathway to Power: Political Recruitment and Candidate Selection in Latin America. University Park: Penn State University Press. Moureau, Defargues (2003). La Mondialisation. Paris: Presses Universitaires de France. Muller, Edward N., and Mitchell A. Seligson (1994). “Civil Culture and Democracy: The Question of Causal Relationships.” American Political Science Review 82, no. 3, pp. 635–652. Muller, Wolfang C., and Kaare Strom (1999). “Conclusions: Party Behavior and Representative Democracy.” In Wolfgang C. Muller and Kaare Strom, eds., Policy, Office or Votes? How Political Parties in Western Europe Make Hard Decisions, pp. 279–310. Cambridge: Cambridge University Press. Nacif, Benito (2002). “Para comprender la disciplina de partido en la Cámara de Diputados de México: El modelo de partido centralizado.” Foro Internacional 42, no. 1, pp. 5–38. Navia, Patricio (2007). “Legislative Candidate Selection in Chile.” In Morgenstern and Siavelis, eds., Pathway to Power: Political Recruitment and Candidate Selection in Latin America. University Park: Penn State University Press. Nilsson, Martin (2003). “Left Parties and Democratic Consolidation in Latin America.” Paper from the ECPR–Joint Sessions of Workshops, University of Edinburgh, Edinburgh, Scotland. Norden, Deborah L. (1998). “Party Relations and Democracy in Latin America.” Party Politics 4, no. 4, pp. 423–443. Norris, Pippa (1996). “Legislative Recruitment.” In Lawrence LeDuc, Richard Niemi, and Pippa Norris, eds., Comparing Democracies: Elections and Voting in Global Perspective. London: Sage Publications. ———, ed. (1997). Passages to Power: Legislative Recruitment in Advanced Democracies. Cambridge: Cambridge University Press.

A-BM

10/2/07

3:40 PM

Page 297

BIBLIOGRAPHY

297

O’Donnell, Guillermo (1994). “Delegative Democracy?” Journal of Democracy 5, no. 1, pp. 55–69. ——— (1999a). “Horizontal Accountability and New Polyarchies.” In Andreas Schedler, Larry Diamond, and Marc F. Plattner, eds., The Self-Restraining State: Power and Accountability in New Democracies, pp. 29–52. Boulder and London: Lynne Rienner Publishers. ——— (1999b). “Polyarchies and the (Un)Rule of Law in Latin America: A Partial Conclusion.” In Juan E. Méndez, G. O’Donnell, and P. S. Pinheiro, eds., The (Un)Rule of Law and the Underprivileged in Latin America, pp. 303–339. Notre Dame, Ind.: University of Notre Dame Press. ——— (2002). “Ilusiones sobre la consolidación.” Nueva Sociedad, no. 180–181, pp. 311–332. O’Donnell, Guillermo, Philippe Schmitter, and Laurence Whitehead (1986). Transiciones desde un gobierno autoritario. 4 vols. Buenos Aires: Paidós. O’Donnell, Guillermo, Jorge Vargas Cullell, and Osvaldo M. Iazzetta, eds. (2004). The Quality of Democracy: Theory and Applications. Notre Dame, Ind.: University of Notre Dame Press. Otero, Patricia (2004). “Partidos y sistemas de partidos en Panamá: Un estudio de la estructuración ideológica.” Paper delivered at the X encuentro de Latinoamericanistas Españoles, Salamanca, Spain. Pachano, Simón (1996). Democracia sin sociedad. Quito: ILDIS. Padró i Miquel, Gerard, and James M. Snyder (2004). “Legislative Effectiveness and Legislative Life.” Working Paper no. 04-28. Department of Economics, MIT, Cambridge, Massachusetts. Available at http://ssrn.com/abstract= 576406. Pan American Health Organization (PAHO) (2004). PAHO Regional Mortality Database. Estimates based on world population prospects, 2004 revision. Available at www.paho.org. Panebianco, Angelo (1988). Political Parties: Organization and Power. Cambridge: Cambridge University Press. Payne, J. Mark, Daniel Zovatto, Fernando Carrillo, and Andrés Allamand (2002). Democracies in Development: Politics and Reform in Latin America. Washington, D.C.: Inter-American Development Bank. PELA (Proyecto Elites Parlamentarias en América Latina) (1994–2005). Salamanca: Universidad de Salamanca. Pharr, Susan J., and Robert Putnam (2000). Disaffected Democracies: What’s Troubling the Trilateral Countries? Princeton, N.J.: Princeton University Press. PNUD (Programa de las Naciones Unidas para el Desarrollo) (2004). La Democracia en América Latina: Hacia una democracia de ciudadanas y ciudadanos. New York: United Nations. Polsby, Nelson W. (1968). “The Institutionalization of the US House of Representatives.” American Political Science Review 62, pp. 144–168. Powell, Bingham (2001). “Democratic Representation: Two Contributions from Comparative Politics.” Paper delivered at the annual meeting of the American Political Science Association, San Francisco, California. Powell, Bingham, and Guy Whitten (1993). “A Cross-National Analysis of Economic Voting: Taking Account of Political Context.” American Journal of Political Science, no. 37, pp. 391–414. PPAL (Proyecto Partidos Políticos en América Latina) (1999–2001). Salamanca: Universidad de Salamanca. Przeworski, Adam, and Fernando Limongi (1997). “Modernization: Theories and Facts.” World Politics 49, pp. 155–183.

A-BM

10/2/07

3:40 PM

298

Page 298

BIBLIOGRAPHY

Przeworski, Adam, Susan Carol Stokes, and Bernard Manin, eds. (1999). Democracy, Accountability, and Representation. Cambridge: Cambridge University Press. Putnam, Robert D. (1973). The Beliefs of Politicians: Ideology, Conflict and Democracy in Britain and Italy. New Haven, Conn.: Yale University Press. Quintero, Rafael, ed. (1991). La cuestión regional y el poder. Quito: Corporación Editora Nacional. Ranney, Austin (1962). The Doctrine of Responsible Party Government: Its Origins and Present State. Urbana: University of Illinois Press. Reiser, Marion (2003). “From Political Amateurs to Professional Politicians? An Analysis of Councilors in Four German Cities.” Paper delivered at the meeting of the American Political Science Association, Philadelphia, Pennsylvania. Reveles, Francisco (2003). El PAN en la oposición. Mexico City: Ediciones Gernika. Rittberger, Volker, ed. (1993). Regime Theory and International Relations. London: Clarendon Press. Rohrschneider, Robert (1994). “Report from the Laboratory: The Influence of Institutions on Political Elites’ Democratic Values in Germany.” American Political Science Review 88, no. 4, pp. 927–941. Rosas, Guillermo (2001). “Policy Preferences, Ideology, and Political Competition in Latin American Legislatures.” Paper delivered at the meeting of the American Political Science Association, Washington, D.C. Rosenbaum, Walter (1975). Political Culture. London: Thomas Nelson & Son. Ruiz Rodríguez, Leticia (2000). “Clivajes y competencia partidista en Chile (1990–1999).” In Nuevo gobierno: Desafíos de la reconciliación; Chile 1999–2000, pp. 159–190. Santiago: FLACSO-Chile. ——— (2003a). La coherencia de los partidos políticos: Estructuración interna de la élite parlamentaria latinoamericana. Doctoral dissertation, Departamento de Derecho Pulbico General, Universidad de Salamanca, Salamanca, Spain. ——— (2003b). “Polarización en el Chile post-autoritario: Elites partidistas.” In Manuel Alcántara and Elena Martínez Barahona, eds., Política, dinero e institucionalización partidista en América Latina. Mexico DF: Universidad Iberoamericana. ——— (2004). “Rendimiento electoral y coherencia partidista en América Latina: El desacuerdo exitoso.” In Consejo Español de Estudios Iberoamericanos, Actas X encuentro de Latinoamericanistas españoles, Salamanca, Spain. ——— (2006). “Coherencia partidista: La estructuración interna de los partidos políticos en América Latina.” Revista Española de Ciencia Política, no. 14, pp. 87–114. ——— (2007). Partidos y coherencia: Parlamentarios en América Latina. Madrid: Centro de Estudios Políticos y Constitucionales. Ruiz Rodríguez, Leticia, and Mercedes García Montero (2003). “Coherencia partidista en las élites parlamentarias latinoamericanas.” Revista Española de Ciencia Política, no. 8, pp. 71–102. Sabsay, Daniel Alberto (1991). “El ballotage: Su aplicación en América Latina y la gobernabilidad.” Cuadernos de Capel 34. Samuels, David (2003). Ambition, Federalism, and Legislative Politics in Brazil. Cambridge: Cambridge University Press. ——— (2007). “Political Ambition, Candidate Recruitment, and Legislative Politics in Brazil.” In Morgenstern and Siavelis, eds. Pathway to Power: Political Recruitment and Candidate Selection in Latin America. University Park: Penn State University Press.

A-BM

10/2/07

3:40 PM

Page 299

BIBLIOGRAPHY

299

Sanbonmatsu, Kira (2002). “Political Parties and the Recruitment of Women to State Legislatures.” Journal of Politics 64, no. 3, pp. 791–809. Sani, Giacomo, and José Ramón Montero (1986). “El espectro político: Izquierdaderecha y centro.” In Juan J. Linz and José Ramón Montero, eds., Crisis y cambio: Electores y partidos en la España de los años ochenta, pp. 155–200. Madrid: Centro de Estudios Constitucionales. Sani, Giacomo, and Giovanni Sartori (1983). “Polarization, Fragmentation and Competition in Western Democracies.” In P. Daadler and P. Mair, eds., Western European Party Systems, pp. 307–340. Beverly Hills, Calif.: Sage. Santos, Fabiano, and Lucio Renno (2004). “The Selection of Committee Leadership in the Brazilian Chamber of Deputies.” Journal of Legislative Studies 10, no. 1, pp. 50–70. Sartori, Giovanni (1994). Comparative Constitutional Engineering: An Inquiry into Structures, Incentives and Outcomes. New York: New York University Press. ——— (1999 [1st ed., 1976]). Partidos y sistemas de partidos. Madrid: Alianza Editorial. Scarrow, Susan (2000). “Parties Without Members? Party Organization in a Changing Electoral Environment.” In R. J. Dalton and M. P. Wattenberg, eds., Parties Without Partisans: Political Change in Advanced Industrial Democracies, pp. 79–101. Oxford: Oxford University Press. Schattschneider, Elmer (1942). Party Government. New York: Rinehart and Winston. Schedler, Andreas (1995). “Under- and Overinstitutionalization: Some Ideal Typical Propositions Concerning Old and New Party Systems.” Working Paper no. 213. Kellogg Institute for International Studies, University of Notre Dame, Notre Dame, Indiana. Schmitt, Hermann, and Jacques Thomassen (1999). Political Representation and Legitimacy in the European Union. Oxford: Oxford University Press. Schmitter, Philippe (1995). “The Consolidation of Political Democracies.” In Geoffrey Pridham, ed., Transitions to Democracy: Comparative Perspectives from Southern Europe, Latin America, and Eastern Europe. Aldershot: Ashgate Publishing. Scott, Robert E. (1967). “Political Elites and Political Modernization: The Crisis of Transition.” In Seymour M. Lipset and Aldo Solari, eds., Elites in Latin America, pp. 117–145. New York: Oxford University Press. Searing, Donald (1985). “The Role of the Good Constituency Member and the Practice of Representation in Great Britain.” Journal of Politics 47, no. 2, pp. 348–381. ——— (1987). “New Roles for Postwar British Politics: Ideologues, Generalists, Specialists, and the Progress of Professionalization in Parliament.” Comparative Politics 19, no. 4, pp. 431–442. ——— (1994). Westminster´s World: Understanding Political Roles. Cambridge: Cambridge University Press. Shepsle, Kenneth A., and Mark S. Bonchek (2005). Las fórmulas de la política: Instituciones, racionalidad y comportamiento. Mexico DF: TaurusCIDE. Shin, Doh Chill (1994). “On the Third Wave of Democratization: A Synthesis and Evaluation of Recent Theory and Research.” World Politics 47, no. 1, pp. 135–170. Shugart, Matthew Søberg, Melody Ellis Valdini, and Kati Suominen (2005). “Looking for Locals: Voter Information Demands and Personal Vote-Earning

A-BM

10/2/07

3:40 PM

300

Page 300

BIBLIOGRAPHY

Attributes of Legislators Under Proportional Representation.” American Journal of Political Science 49, no. 2, pp. 437–449. Siavelis, Peter (2005). “La lógica oculta de la selección de candidatos en las elecciones parlamentarias chilenas.” Estudios Públicos 98, pp. 189–225. Snijders, Tom, and Roel Bosker (1999). Multilevel Analysis. London: Sage. Soule, John W. (1969) “Future Political Ambitions and the Behavior of Incumbent State Legislators.” Midwest Journal of Political Science 13, no. 3, pp. 439–454. Squire, Peverill (1993). “Professionalism and Public Opinion in State Legislatures.” Journal of Politics 55, no. 2, pp. 479–491. ——— (1998). “Membership Turnover and the Efficient Processing of Legislation.” Legislative Studies Quarterly 23, no. 1, pp. 23–32. Squire, Peverill, and Keith E. Hamm (2005). 101 Chambers: Congress, State Legislatures, and the Future of Legislative Studies. Columbus: Ohio State University Press. Steenbergen, Marco, and Bradford S. Jones (2002). “Modeling Multilevel Data Structures.” American Journal of Political Science 46, no. 1, pp. 218–237. Stokes, Susan (1999). “What Do Policy Switches Tell Us About Democracy?” In Adam Przeworski, Susan Carol Stokes, and Bernard Manin, eds., Democracy, Accountability, and Representation. Cambridge: Cambridge University Press. ——— (2001). Mandates and Democracy: Neoliberalism by Surprise in Latin America. Cambridge: Cambridge University Press. Stone, Walter J., L. Sandy Maisel, and Cherie D. Maestas (2004). “Quality Counts: Extending the Strategic Politician Model of Incumbent Deterrence.” American Journal of Political Science 48, no. 3, pp. 479–495. Studlar, Donley, and Ian Mcallister (1996). “Constituency Activity and Representational Roles Among Australian Legislators.” Journal of Politics 58, no. 1, pp. 69–90. Thelen, Kathleen, and Sven Steinmo (1992). “Institutionalism in Comparative Politics.” In Sven Steinmo, Kathleen Thelen, and Frank Longstreth, eds., Structuring Politics: Historical Institutionalism in Comparative Analysis. New York: Cambridge University Press. Thomassen, Jacques (1994). “Empirical Research into Political Representation: Failing Democracy or Failing Models?” In Kent Jennings and Thomas Mann, eds., Elections at Home and Abroad: Essays in Honor of Warren E. Miller. Ann Arbor: University of Michigan Press. Torcal, Mariano (2001). Political Disaffection and Democratization History in New Democracies. Working Paper no. 308. Kellogg Institute, University of Notre Dame, Notre Dame, Indiana. Tsebelis, George (1997). “Processo decisorio en sistemas políticos: Veto players no presidencialismo, parlamentarios, multicameralismo e pluripartidismo.” Revista Brasileira de Ciencias Sociais 12, no. 34. ——— (2000). “Veto Players and Institutional Analysis.” Governance, no. 13, pp. 441–474. Ugalde, Luis Carlos (2002). “La disciplina partidista en México.” Paper delivered at the El Gobierno Dividido en México, Riesgos y Oportunidades, Mexico DF, Mexico. Uriarte, Edurne (1997). “El análisis de las elites políticas en las democracias.” Revista de Estudios Políticos, no. 9, pp. 249–275. ——— (2000). “La política como profesión y vocación: Análisis de las motivaciones y de la carrera política de los diputados españoles.” Revista Española de Ciencia Política, no. 3, pp. 97–124.

A-BM

10/2/07

3:40 PM

Page 301

BIBLIOGRAPHY

301

Von Beyme, Klaus (1995). La clase política en el estado de partidos. Madrid: Alianza Editorial. Vowles, Jack, Peter Aimer, Susan Banducci, and Jeffrey Karp (1998). Voters’ Victory? New Zealand’s First Election Under Proportional Representation. Auckland: University of Auckland Press. Weissberg, Robert (1978). “Collective Versus Dyadic Representation in Congress.” American Political Science Review 72, no. 2, pp. 535–547. Wolinetz, Steven B. (2002). “Beyond the Catch-All Party: Approaches to the Study of Parties and Party Organization in Contemporary Democracies.” In R. Gunther, J. R. Montero, and J. J. Linz, eds., Political Parties: Old Concepts and New Challenges, pp. 136–165. Oxford: Oxford University Press. World Bank (2005). World Development Indicators [CD-ROM]. Washington, D.C.: Development Data Group. Wright Mills, Charles (1981). Poder, política, pueblo. Mexico DF: Fondo de Cultura Económica.

A-BM

10/2/07

3:40 PM

Page 302

A-BM

10/2/07

3:40 PM

Page 303

The Contributors Manuel Alcántara Sáez is professor of political science at the University of Salamanca. Flavia Freidenberg is lecturer of political science at the University of Salamanca. Fátima García Díez is lecturer of political science at the University of Salamanca. Mercedes García Montero is a researcher at the Instituto Interuniversitario de Iberoamérica of the University of Salamanca. Iván Llamazares Valduvieco is senior lecturer of political science at the University of Salamanca. Patricia Marenghi is a researcher at the Instituto Interuniversitario de Iberoamérica of the University of Salamanca. Salvador Martí i Puig is senior lecturer of political science at the University of Salamanca. María del Mar Martínez Rosón is a researcher at the Instituto Interuniversitario de Iberoamérica of the University of Salamanca. Araceli Mateos Díaz is lecturer of political science at the University of Salamanca. Cristina Rivas Pérez is a researcher at the Instituto Interuniversitario de Iberoamérica of the University of Salamanca. Leticia M. Ruiz Rodríguez is lecturer of political science at the University of Salamanca, and has a PhD in Political Science and Public Administration. Salvador Santiuste Cué is lecturer of sociology at the University of Salamanca.

303

A-BM

10/2/07

3:40 PM

Page 304

A-BM

10/2/07

3:40 PM

Page 305

Index

Abortion, 149, 224–225 Accountability, 12, 175 AD (Venezuela). See Venezuela, political parties ADN (Bolivia). See Bolivia, political parties; Parliamentary right Age of legislators, 239–241, 245–249 Alcántara, Manuel, 1, 22, 26 (n1), 87, 219, 265, 270; on relationship between democracy and economic risks, 80; and studies of ideology, 158 (nn8,15); and studies of party organization, 115, 117, 137 (n38); and studies of programmatic coherence, 177 Allamand, Andrés, 130 Almond, Gabriel, 159 (n34) Anderson, Lee, 176 Angell, Alan, 136 (n28) ANR (Paraguay). See Paraguay, political parties; Parliamentary right ARENA (El Salvador). See El Salvador, political parties; Parliamentary right Argentina: attitudes toward authoritarian government, 66, 71; attitudes toward democracy, 66, 71, 75; attitudes toward role of parties in democracy, 69; authoritarian regimes, 220; closed and blocked lists, 255; constitutional regulations on representation, 31; democratic stability, 83; democratization process, 78, 216 (n3); district size and level of experience of legislators, 256; focus and style of representation, 34–37, 41–44, 51–52, 54; high number of

quality legislators, 260; influence of different actors on political decisions, 45–46; institutional trust, 82, 83; lack of conservative parties, 221; level of experience of legislators, 241–245, 256; party vs. territorial interests, 43–44; perceptions of problems compared to national realities, 106, 108; perceptions of problems faced by government, 88–90, 99, 105; perceptions on advantages of democracy, 75; Peronism, 220, 221; regional interests, 42–44; state intervention in economy, 205; threats to democratic consolidation, 78; type of democracy, 71; and United States, 229; voting and quality of legislators, 253 Argentina, political parties (ARI, FREPASO, PJ, UCeDé, UCR): and democracy vs. authoritarianism, 151; focus and style of representation, 36–39; ideology, 141; importance of different groups in decisionmaking, 125; internal democracy, 123–125; legislators’ ideological closeness to party positions, 155, 164–166; militancy, 119; and military, 153; perceived intraparty life, 129; perceptions on levels of participation, 121; preferred degree of power for party leaders, 124; programmatic coherence, 180–182, 185–187; public spending on statism vs. market, 189; relation between party life and types of party systems, 130, 131; and reli-

305

A-BM

10/2/07

3:40 PM

306

Page 306

INDEX

gion and values, 149; representation of polarization dimensions, 147–148; and social background, 152; and state intervention, 150, 190 ARI (Argentina). See Argentina, political parties Authoritarian governments, 65–66, 71–72, 83; and classification of political parties, 151–152; and the military, 220; parliamentary left and transition to democracy, 195–196; and parliamentary right, 220, 230, 230 (n6); preferences for democracy in left parties, 199; and United States, 220 Banana companies, 230 (n6) Behavior of political elites. See Style of representation BIP (Dominican Republic). See Dominican Republic, political parties BIS (Dominican Republic). See Dominican Republic, political parties Blondel, Jean, 116 Bolivia: attitudes toward authoritarian government, 66, 71–72; attitudes toward democracy, 66, 71, 72, 73, 75; attitudes toward elections, 67; attitudes toward role of parties in democracy, 69; constitutional regulations on representation, 31, 40; democratic stability, 83; district size and level of experience of legislators, 257; focus and style of representation, 34–38, 41–44, 51–52, 54, 55; high number of inexperienced legislators, 261, 269; influence of different actors on political decisions, 45–46; influence of interest groups, 56–57; institutional trust, 81, 82, 83; level of experience of legislators, 241–245, 257, 261, 269; mixed lists, 255; party vs. territorial interests, 43–44; perceptions of problems compared to national realities, 106, 108; perceptions of problems faced by government, 88–90, 93, 105; perceptions on advantages of democracy, 75; political and institutional disaffection, 84; political crises, 72, 73, 266; populist movements, 220;

regional interests, 42–44; threats to democratic consolidation, 78; type of democracy, 71; and United States, 229 Bolivia, political parties (ADN, CONDEPA, MAS, MIR, MNR, NFR, UCS): ADN as right party, 222; and democracy vs. authoritarianism, 151; focus and style of representation, 36–39; ideology, 141; importance of different groups in decisionmaking, 125; internal democracy, 123–125; legislators’ ideological closeness to party positions, 155, 164–166; militancy, 119; and military, 153; MNR as populist movement, 220; party leaders, 122; perceived intraparty life, 129; perceptions on levels of participation, 121; preferred degree of power for party leaders, 124; programmatic coherence, 180–182, 185–187; public spending on statism vs. market, 189; relation between party life and types of party systems, 131; and religion and values, 149; representation of polarization dimensions, 147–148; and social background, 152; and state intervention, 150, 190 Brazil: constitutional regulations on representation, 31, 40; democratization process, 216 (n3); focus and style of representation, 34–37; state intervention in economy, 205 Burns, Nancy, 235 Careers of political elites, 48, 50–51, 55–57, 235–264; and committee organization, 251–252; and competence, 238, 252–253; costs and benefits of legislative posts, 250; directions for further research, 261; and district size, 252–261; high number of inexperienced legislators, 261; level of experience, 241–249, 269; level of experience correlated with sociodemographic variables, 246– 249; and level of institutionalization of parliament, 251; and open vs. closed/blocked lists, 252–261; and personal voting, 253, 260; profes-

A-BM

10/2/07

3:40 PM

Page 307

INDEX sionalization, 238, 251, 268–269; quality as a dependent and independent variable, 250–252; “quality” definitions, 236–238, 242; quality of legislators and voting, 252–261; reelection strategies, 253; research methodology, 263 (nn22,23,27,28); sociodemographic and sociopolitical characteristics of legislators, 239– 241; and style of representation, 48, 50–51, 54–57; systems of career advancement, 251–252 Carillo, Fernando, 130 Caselli, Francesco, 250 Castañeda, Jorge G., 216 (n4) Catholic Church, 149, 220, 231 (n12) Caudillistas, 122, 129 Centrist parties, 221, 227, 231 (n12) Chile: attitudes toward authoritarian government, 66, 71; attitudes toward democracy, 66, 71; attitudes toward role of parties in democracy, 69; Catholic Church and conservative parties, 220; constitutional regulations on representation, 31; democratic stability, 83; democratization process, 216 (n3); focus and style of representation, 34–38, 41–44, 51–52, 54–56; high number of quality legislators, 260, 269; influence of different actors on political decisions, 45–46; influence of interest groups, 56–57; institutional trust, 81, 82, 83; level of experience of legislators, 241–245, 257, 260; party vs. territorial interests, 43–44; perceptions of problems compared to national realities, 106, 108; perceptions of problems faced by government, 88–90, 93, 99, 105, 106; perceptions on democratic consolidation, 77; regional interests, 42–44; threats to democratic consolidation, 78; type of democracy, 71; and United States, 229; voting and quality of legislators, 253 Chile, political parties (PDC, PPD, PS, RN, UDI): and analysis of cleavages/polarization, 145; and Catholic Church, 220; and democracy vs. authoritarianism, 151, 152; focus and style of representation, 36–39; ideol-

307

ogy, 141; importance of different groups in decisionmaking, 125; internal democracy, 122–125; legislators’ ideological closeness to party positions, 155, 164–166; militancy, 118, 119; and military, 153; perceived intraparty life, 128, 129; perceptions on levels of participation, 121; PPD, PS as left parties, 197– 198; preferred degree of power for party leaders, 124; programmatic coherence, 180–182, 185–187; public spending on statism vs. market, 189; relation between party life and types of party systems, 130, 131; and religion and values, 149; representation of polarization dimensions, 147–148; RN, UDI as right parties, 221–222; and social background, 152; and state intervention, 150, 190 Chin, Michelle L., 264 (n47) Christian-democratic groups. See Parliamentary right Citizens: civil rights/human rights, 74–76, 84, 203–204; and consolidation of democracy, 11, 12, 77, 84; degree of congruence between citizens and political parties, 69–70; degree of congruence between citizens and representatives, 30; equality of opportunity, 74–76; influence of elite opinions on, 68; influence on political decisions, 45, 125; perceptions of problems faced by government, 105; political disaffection, 78–81, 113, 132, 267–268; preferences regarding democracy and authoritarianism, 68; relation to political parties, 116, 117–121, 125. See also Education policy; Public health; Public welfare; Representation Civil rights, 74–76, 84, 203–204 Cleavages. See Political party cleavages and polarization Clientism, 174, 187, 237, 268 Closed/blocked list elections. See Open vs. closed/blocked list elections Coherence of political parties, 120; defined, 174; ideological, programmatic, and organizational coherence,

A-BM

10/2/07

3:40 PM

308

Page 308

INDEX

174, 193 (n18). See also Programmatic coherence of political parties Collier, David, 65 Colombia: attitudes toward authoritarian government, 66, 71; attitudes toward democracy, 66, 71, 267; attitudes toward elections, 67; attitudes toward role of parties in democracy, 69; Catholic Church and conservative parties, 220; closed and blocked lists, 255; constitutional regulations on representation, 31; democratic stability, 83; district size and level of experience of legislators, 256; focus and style of representation, 34–37, 41–44, 51–52, 54–55; high number of quality legislators, 260; influence of different actors on political decisions, 45–46; influence of interest groups, 56–57; institutional trust, 81, 82, 83; level of experience of legislators, 241–245, 256, 260; party vs. territorial interests, 43–44; perceptions of problems compared to national realities, 106, 108; perceptions of problems faced by government, 88–90, 94, 105, 108; regional interests, 42–44; salaries for legislators, 269; state intervention in economy, 205; threats to democratic consolidation, 78; type of democracy, 71; and United States, 228–229; voting and quality of legislators, 253 Colombia, political parties (PC, PL): and democracy vs. authoritarianism, 151; focus and style of representation, 36–39; ideology, 141; importance of different groups in decisionmaking, 125; internal democracy, 123–125; legislators’ ideological closeness to party positions, 155, 164–166; militancy, 119; and military, 153; PC as right party, 221– 222; perceived intraparty life, 129; perceptions on levels of participation, 121; preferred degree of power for party leaders, 124; programmatic coherence, 180–182, 185–187; public spending on statism vs. market, 189; relation between party life and types of party systems, 131; and reli-

gion and values, 149; representation of polarization dimensions, 147–148; and social background, 152; and state intervention, 150, 190 Committee organization, 251–252 Competence of legislators, 238. See also Careers of political elites; Legislative experience Competition, as dimension of democracy quality, 12 CONDEPA (Bolivia). See Bolivia, political parties Constitutions, regulations on representation, 31–32, 40, 62 (n7), 63 (n14) COPEI (Venezuela). See Venezuela, political parties Corruption, 109 (n2); and definitions of “quality,” 237; and national realities, 105, 107; and parliamentary right, 224–225; perceptions of problems by country, 88–90, 92; perceptions of problems by country and ideology, 93–105; and programmatic coherence, 187 Costa Rica: attitudes toward authoritarian government, 66, 71–72; attitudes toward democracy, 66, 71, 73, 75; attitudes toward role of parties in democracy, 69; closed and blocked lists, 255; constitutional regulations on representation, 31, 263 (n24); democratic stability, 83; district size and level of experience of legislators, 256; focus and style of representation, 34–38, 39, 41–44, 51–52, 54– 56; high number of inexperienced legislators, 261; influence of different actors on political decisions, 45–46; institutional trust, 81, 82, 83; level of experience of legislators, 241–245, 256, 261; national mandates, 56; party vs. territorial interests, 43–44; perceptions of problems compared to national realities, 106, 108; perceptions of problems faced by government, 88–90, 94, 100, 105; perceptions on advantages of democracy, 75; political party crises, 72; regional interests, 42–44; state intervention in economy, 205; threats to democratic consolidation, 78; type of

A-BM

10/2/07

3:40 PM

Page 309

INDEX democracy, 71; and United States, 229; voting and quality of legislators, 253, 263 (n24) Costa Rica, political parties (PAC, PLN, PUSC, RD): and Catholic Church, 220; and democracy vs. authoritarianism, 151; focus and style of representation, 36–39; ideology, 141; importance of different groups in decisionmaking, 125; internal democracy, 123–125; legislators’ ideological closeness to party positions, 155, 164–166; militancy, 119; and military, 153; perceived intraparty life, 128, 129; perceptions on levels of participation, 121; preferred degree of power for party leaders, 124; programmatic coherence, 180–182, 185–187; public spending on statism vs. market, 189; PUSC as right party, 222; relation between party life and types of party systems, 130, 131; and religion and values, 149; representation of polarization dimensions, 147–148; and social background, 152; and state intervention, 150, 190 Crime, 77–80, 224–225. See also Violence and public insecurity Crisp, Brian, 253 Cuba, 199, 216 (nn2,3) Cultural issues, and parliamentary right, 223–228. See also Abortion; Divorce Dahl, Robert, 11 Dal Bó, Ernesto, 250 Dal Bó, Pedro, 250 Data collection, 15–16. See also Research methodology Debt, external, 78–80 Decisionmaking in political parties, 113–114, 116–118, 121–129 Democracy: advantages perceived by political elites, 73–77, 84, 203–205; citizens’ perceptions of, 11–12, 267–268; and classification of political parties, 151–152; components of, 65; consolidation of, 11, 12, 75, 77–80, 82–84; and constitutional engineering, 265–266; decentralization and regional democratization,

309

77, 84; definitions, 65; democratic transition in Latin America, 195– 196, 265–267; duration of democratic experiences, 161–163, 167; and institutional trust, 81–82; noninstitutionalist theories, 266; and political crises, 72–73, 83–84, 268; and political disaffection, 268; political elites’ attitudes toward, 11–12, 65–85, 199–205, 267–268; and political parties, 66–71, 83–84, 133–134; quality of, 12–13, 67; stability of, 11–12, 55, 81–84; survey questions, 157; types of, 71–72. See also Elections; Political elites; Political parties; Representation Di Tella, Rafael, 250 Direct electoral democracy, 71–72 Direct party democracy, 71–72 Discrimination, 77–80 Divorce, 149, 224–225 Domínguez, Jorge, 30 Dominican Republic: attitudes toward authoritarian government, 66, 71; attitudes toward democracy, 66, 71–73; attitudes toward role of parties in democracy, 69; authoritarian regime, 230 (n6); constitutional regulations on representation, 32; democratic stability, 83; focus and style of representation, 34–38, 41–44, 51, 53–55, 57; high number of inexperienced legislators, 260–261; influence of different actors on political decisions, 45–46; institutional trust, 81, 82, 83; level of experience of legislators, 241–245, 260–261; national mandates, 56; party vs. territorial interests, 43–44; perceptions of problems compared to national realities, 106, 108; perceptions of problems faced by government, 88–90, 95, 105; regional interests, 42–44; threats to democratic consolidation, 78; type of democracy, 71; and United States, 229 Dominican Republic, political parties (BIP, BIS, PLD, PRD, PRSC): and analysis of cleavages/polarization, 145; corruption and clientism, 187; and democracy vs. authoritarianism,

A-BM

10/2/07

3:40 PM

310

Page 310

INDEX

151; focus and style of representation, 36–39; ideology, 141; importance of different groups in decisionmaking, 125; internal democracy, 123–125; legislators’ ideological closeness to party positions, 155, 164–166; militancy, 119; and military, 153; perceived intraparty life, 129; perceptions on levels of participation, 121; preferred degree of power for party leaders, 124; programmatic coherence, 180–182, 185–187; PRSC as right party, 221–222; public spending on statism vs. market, 189; relation between party life and types of party systems, 130, 131; and religion and values, 149; representation of polarization dimensions, 147–148; and social background, 152; and state intervention, 150, 187, 190 DP (Ecuador). See Ecuador, political parties Drug trafficking, 78–80; perceptions of problems by country, 88–90, 92, 108; perceptions of problems by country and ideology, 93–105 Duverger, Maurice, 115, 117 Easton, David, 7 Economic crisis, 77–80, 84 Economic development, 74–76, 84, 161, 162, 167; changing state-market relations, 205–207; and neoliberalism, 196, 205–206; and perceptions of democracy, 203–204; and perceptions of the market, 205; and state intervention vs. market, 187–191 Economic problems, 109 (n2); disjunction between perceptions and national realities, 105–106; as most prominent part of political agendas, 108; perceptions of problems by country, 88–90, 92; perceptions of problems by country and ideology, 93–105 Ecuador: attitudes toward authoritarian government, 66, 71–72; attitudes toward democracy, 66, 71, 73, 75; attitudes toward role of parties in democracy, 68, 69; democratic stability, 83; district size and level of

experience of legislators, 260; focus and style of representation, 34–38, 41–44, 51–52, 54–55, 57; high number of inexperienced legislators, 261, 269; influence of different actors on political decisions, 45–46; institutional trust, 82, 83; level of experience of legislators, 241–245, 260, 261, 269; open lists, 255; party vs. territorial interests, 43–44; perceptions of problems compared to national realities, 106, 108; perceptions of problems faced by government, 88–90, 100, 105, 106; perceptions on advantages of democracy, 75; political and institutional disaffection, 84; political crises, 72, 73, 268; regional fragmentation of party system, 63 (n11); regional interests, 42–44; regulations on representation, 31, 40; salaries for legislators, 269; threats to democratic consolidation, 78; type of democracy, 71; and United States, 229 Ecuador, political parties (DP, ID, MUPP-NP, PRE, PRIAN, PSC, PSP): and analysis of cleavages/ polarization, 145; and democracy vs. authoritarianism, 151, 152; focus and style of representation, 36–39; ID, MUPP-NP as left parties, 197–198; ideology, 128, 141; importance of different groups in decisionmaking, 125; internal democracy, 123–125, 128; legislators’ ideological closeness to party positions, 155, 159 (n31), 164–166; militancy, 119; and military, 153; perceived intraparty life, 129; perceptions on levels of participation, 121; preferred degree of power for party leaders, 124; programmatic coherence, 180–182, 185–187; PSC as right party, 221– 222; public spending on statism vs. market, 189; relation between party life and types of party systems, 131; and religion and values, 149, 150; representation of polarization dimensions, 147–148; and social background, 152; and state intervention, 150, 151, 190

A-BM

10/2/07

3:40 PM

Page 311

INDEX Education of political elites/fathers of elites, 152, 157, 239–241, 245–249, 269 Education policy: and parliamentary left, 211, 213; and parliamentary right, 224–225; and programmatic coherence, 188, 189 El Salvador: attitudes toward authoritarian government, 66, 71; attitudes toward democracy, 66, 71, 72; attitudes toward elections, 67; attitudes toward role of parties in democracy, 68, 69; closed and blocked lists, 255; constitutional regulations on representation, 31; democratic stability, 83; democratization process, 216 (n3); focus and style of representation, 34–37, 41–44, 51, 53–56; influence of different actors on political decisions, 45–46; institutional trust, 82, 83; legislator quality, 269; level of experience of legislators, 241– 245; party vs. territorial interests, 43–44; perceptions of problems compared to national realities, 106, 108; perceptions of problems faced by government, 88–90, 95, 101, 105, 106; perceptions on democratic consolidation, 77; political party crises, 72; regional interests, 42–44; state intervention in economy, 205; threats to democratic consolidation, 78; type of democracy, 71; and United States, 229 El Salvador, political parties (ARENA, FMLN, PCN, PDC): and analysis of cleavages/polarization, 145; ARENA as right party, 221–222; and democracy vs. authoritarianism, 151; development of PCN, 230 (n6); differences between right parties and nonright parties, 226; FMLN as left party, 197–198; focus and style of representation, 36–39; ideology, 128, 141; importance of different groups in decisionmaking, 125; internal democracy, 123–125, 128; legislators’ ideological closeness to party positions, 155, 164–166; militancy, 118, 119; and military, 153; perceived intraparty life, 128, 129; perceptions on

311

levels of participation, 121; preferred degree of power for party leaders, 124; programmatic coherence, 180–182, 185–187; public spending on statism vs. market, 189; relation between party life and types of party systems, 130, 131; and religion and values, 149, 150; representation of polarization dimensions, 147–148; and social background, 152; and state intervention, 150, 151, 190 Elections: attitudes toward, 67, 69, 71–72, 200–203; and careers of legislators, 236; criteria for democratic elections, 67; and democratic transition, 265–266; and district size, 252–261; and effective number of parties, 260, 261; electoral personalism, 161, 167, 253, 255, 260; electoral volatility, 161–163, 167, 227; fairness of, 67; and information acquisition, 264 (n47); and legislator quality, 252–261; legislators’ confidence in, 202; open vs. closed/ blocked lists, 252–261 EP/FA (Uruguay). See Parliamentary left; Uruguay, political parties Equality, as dimension of democracy quality, 12 Equality of opportunity, 204 Escobar-Lemmon, Maria, 253 Europe, 115, 135 (n1), 173 Evans, Laura, 235 Falange Nacional, 220, 231 (n9) Family background of political elites, 152, 239–240, 245–249 FDNG (Guatemala). See Guatemala, political parties FIM (Peru). See Peru, political parties FMLN (El Salvador). See El Salvador, political parties; Parliamentary left Focus of representation (general vs. specific interests), 32–41; defined/ described, 33; explanatory factors, 44–46, 49–52; and political parties, 35–39; research methodology, 59–60; and territories, 38–41 Freidenberg, Flavia, 115, 161, 270 FREPASO (Argentina). See Argentina, political parties

A-BM

10/2/07

3:40 PM

312

Page 312

INDEX

FRG (Guatemala). See Guatemala, political parties FSLN (Nicaragua). See Nicaragua, political parties; Parliamentary left Gamm, Gerald, 235 GANA (Guatemala). See Guatemala, political parties García Díez, Fátima, 7, 161, 270 García Montero, Mercedes, 29, 135 (n13), 268 Gender of political elites, 239–241, 245–249 Globalization, 205–206 Gómez, Manuel, 231 (n10) Guatemala: attitudes toward authoritarian government, 66, 71; attitudes toward democracy, 66, 71, 72, 75; attitudes toward role of parties in democracy, 69; authoritarian regime, 220; closed and blocked lists, 255; constitutional regulations on representation, 31; democratic stability, 83; democratization process, 216 (n3); focus and style of representation, 34–37, 41–44, 54; high number of inexperienced legislators, 261, 269; influence of different actors on political decisions, 45–46; institutional trust, 81, 82, 83; level of experience of legislators, 241–245, 261, 269; party vs. territorial interests, 43–44; perceptions of problems compared to national realities, 106, 108; perceptions of problems faced by government, 88–90, 96, 101, 105, 108; perceptions on advantages of democracy, 75; perceptions on democratic consolidation, 77; regional interests, 42–44; salaries for legislators, 269; threats to democratic consolidation, 78; type of democracy, 71; and United States, 229 Guatemala, political parties (FDNG, FRG, GANA, PAN, PSN, PU): and analysis of cleavages/polarization, 145; and democracy vs. authoritarianism, 151; ideology, 141; importance of different groups in decisionmaking, 125; internal democracy, 123–125; legislators’ ideological

closeness to party positions, 155, 164–166; militancy, 118, 119; and military, 153; PAN as right party, 222; perceived intraparty life, 129; perceptions on levels of participation, 121; preferred degree of power for party leaders, 124; programmatic coherence, 180–182, 185–187; public spending on statism vs. market, 189; relation between party life and types of party systems, 131; and religion and values, 149; representation of polarization dimensions, 147–148; and social background, 152; and state intervention, 150, 187, 190 Gunther, Richard, 135 (n13) Gutiérrez, Lucio, 72 Hagopian, Francis, 68, 262 (n19) Hibbing, John, 236–237, 264 (n37) Honduras: attitudes toward authoritarian government, 66, 71; attitudes toward democracy, 66, 71, 75; attitudes toward role of parties in democracy, 69; closed and blocked lists, 255; democratic stability, 83; focus and style of representation, 34–37, 41–44, 51–52, 54; influence of different actors on political decisions, 45–46; institutional trust, 81, 82, 83; level of experience of legislators/ quality of legislators, 241–245, 253, 269; party vs. territorial interests, 43–44; perceptions of problems compared to national realities, 106, 108; perceptions of problems faced by government, 88–90, 96, 105; perceptions on advantages of democracy, 75; perceptions on democratic consolidation, 77; regional interests, 42–44; regulations on representation, 31, 40; salaries for legislators, 269; threats to democratic consolidation, 78; type of democracy, 71; and United States, 229, 230 (n6) Honduras, political parties (PL/PLH, PN/PNH, PUD): and democracy vs. authoritarianism, 151, 152; focus and style of representation, 36–39; ideology, 141; importance of different groups in decisionmaking, 125;

A-BM

10/2/07

3:40 PM

Page 313

INDEX internal democracy, 123–125; legislators’ ideological closeness to party positions, 155, 164–166; militancy, 119; and military, 153; perceived intraparty life, 129; perceptions on levels of participation, 121; PN as right party, 221–222; preferred degree of power for party leaders, 124; programmatic coherence, 180–182, 185–187; public spending on statism vs. market, 189; relation between party life and types of party systems, 130, 131; and religion and values, 149; representation of polarization dimensions, 147–148; and social background, 152; and state intervention, 150, 151, 190 Human rights, 74–76, 84, 224–225 ID (Ecuador). See Ecuador, political parties; Parliamentary left Ideology: classification of parties, 141; determinants of intraparty ideological differences, 161–171; influence on political decisions, 46–47; and internal democracy in political parties, 125, 128; and levels of experience, 245–249; and levels of party participation, 120; moderation of extremes, 77; and party coherence, 120, 174, 176, 270; and perceptions of advantages of democracy, 75–76; and perceptions of problems faced by governments, 93–105; and programmatic coherence, 190–191, 193 (n18); and research methodology, 15; and style of representation, 54, 56, 64 (n19); utility of left-right dimension, 140–141, 154, 158 (n8), 162, 269–271. See also Parliamentary left; Parliamentary right; Political party cleavages and polarization Import substitution, 205 Incentives, and legislators’ perception of representation roles, 32 Income distribution, 203–204, 227 Informal practices, and legislators’ perceptions of representation roles, 32 Institutions, political, 11, 32, 81–82 Instituto Interuniversitario de Salamanca, 7–8

313

Interactionist frameworks for studying political elites, 10 Interest groups, 33–35, 41–59 Janda, Kenneth, 115, 135 (n11), 136 (n25) Jones, Bradford, 253 Jones, Mark, 253 Judiciary, 77–80 Kitschelt, Herbert, 162 Labor conflicts, 78–80 Left parties. See Parliamentary left Legislative experience, 239–241, 257–261; level of experience correlated with sociodemographic variables, 246–249. See also Careers of political elites Legislators. See Political elites Levitsky, Steven, 65, 135 (n12), 193 (n25) Linz, Juan, 12, 75, 113 Lipset, Seymour Martin, 1, 153 Llamazares Valduvieco, Iván, 161, 219, 270 Luna, Juan Pablo, 30 Mainwaring, Scott, 14, 65, 129–130, 131, 134 Manin, Bernard, 62 (n1) Marenghi, Patricia, 29, 268 Maritain, Jacques, 220 Market forces, 187–188, 205–214 Marti i Puig, Salvador, 195, 270 Martínez Rosón, María del Mar, 235, 269 MAS (Bolivia). See Bolivia, political parties MAS (Venezuela). See Parliamentary left; Venezuela, political parties Mass media, 45, 115 Mateos Díaz, Araceli, 7, 65, 267–268 Mayhew, David, 253 McConnaughy, Corrine, 235 McCurley, Carl, 238 Mexico: attitudes toward authoritarian government, 66, 71; attitudes toward democracy, 66, 71, 72, 75; attitudes toward elections, 67; attitudes toward role of parties in democracy,

A-BM

10/2/07

3:40 PM

314

Page 314

INDEX

69; constitutional regulations on representation, 31; democratic stability, 83; district size and level of experience of legislators, 257; focus and style of representation, 34–37, 41–44, 51, 53–56, 64 (n19); high number of quality legislators, 260; influence of different actors on political decisions, 45–46; institutional trust, 82, 83; level of experience of legislators, 241–245, 257, 260; mixed lists, 255; national mandates, 56; party vs. territorial interests, 43–44; perceptions of problems compared to national realities, 106, 108; perceptions of problems faced by government, 88–90, 97, 102, 105, 106, 108; political party crises, 72; regional interests, 42–44; state intervention in economy, 205; threats to democratic consolidation, 78; type of democracy, 71; and United States, 228–229; voting and quality of legislators, 263 (n24) Mexico, political parties (PAN, PRD, PRI): and Catholic Church, 220; creation of PAN, 231 (n10); and democracy vs. authoritarianism, 151; focus and style of representation, 36–39; ideology, 141; importance of different groups in decisionmaking, 125; internal democracy, 123–125; legislators’ ideological closeness to party positions, 155, 164–166; militancy, 119; and military, 153; PAN as right party, 222; perceived intraparty life, 129; perceptions on levels of participation, 121; PRD as left party, 197–198; preferred degree of power for party leaders, 124; programmatic coherence, 180–182, 185–187; public spending on statism vs. market, 189; relation between party life and types of party systems, 131; and religion and values, 149; representation of polarization dimensions, 147–148; and social background, 152; and state intervention, 150, 151, 187, 189, 190 Militancy, party, 117–118, 134; defined, 118; and party coherence, 120, 270

Military: and authoritarian governments, 220; civil control of, 77; and classification of political parties, 153; and parliamentary right, 224–225; relation to government, 78–80; survey questions, 157 MIR (Bolivia). See Bolivia, political parties MNR (Bolivia). See Bolivia, political parties Mondak, Jeffrey, 238 Morales, Evo, 199 Morelli, Massimo, 250 Moreno, Alejandro, 64 (n19) Morgenstern, Scott, 264 (n46) Motivational frameworks for studying political elites, 10 MPQ (Paraguay). See Paraguay, political parties Muller, Wolfang, 133 MUPP-NP (Ecuador). See Ecuador, political parties; Parliamentary left MVR (Venezuela). See Venezuela, political parties Nacif, Benito, 264 (n46) National mandates, 30–32, 56 Neoliberalism, 205–206 NFR (Bolivia). See Bolivia, political parties Nicaragua: attitudes toward authoritarian government, 66, 71; attitudes toward democracy, 66, 71, 75; attitudes toward elections, 67; attitudes toward role of parties in democracy, 68, 69; authoritarian regime, 230 (n6); closed and blocked lists, 255; constitutional regulations on representation, 31; democratic stability, 83; democratization process, 216 (n3); focus and style of representation, 34–38, 41–44, 51–52, 54–56; influence of different actors on political decisions, 45–46; institutional trust, 81, 82, 83; level of experience of legislators, 241–245; party vs. territorial interests, 43–44; perceptions of problems compared to national realities, 106, 108; perceptions of problems faced by government, 88–90, 102, 105, 106; perceptions on

A-BM

10/2/07

3:40 PM

Page 315

INDEX advantages of democracy, 75; perceptions on democratic consolidation, 77; political party crises, 72; regional interests, 42–44; threats to democratic consolidation, 78; type of democracy, 71; and United States, 228–229 Nicaragua, political parties (FSLN, PLC): and analysis of cleavages/ polarization, 145; and democracy vs. authoritarianism, 151; focus and style of representation, 36–39; FSLN as left party, 197–198; ideology, 141; importance of different groups in decisionmaking, 125; internal democracy, 123–125; legislators’ ideological closeness to party positions, 155, 164–166; militancy, 118, 119; and military, 153; perceived intraparty life, 128, 129; perceptions on levels of participation, 121; PLC as right party, 221–222; preferred degree of power for party leaders, 124; programmatic coherence, 180–182, 185–187; public spending on statism vs. market, 189; relation between party life and types of party systems, 130, 131; and religion and values, 149; representation of polarization dimensions, 147–148; and social background, 152; and state intervention, 150, 190 O’Donnell, Guillermo, 12, 30 Open vs. closed/blocked list elections, 252–261 PA (Panama). See Panama, political parties PAC (Costa Rica). See Costa Rica, political parties Padró i Miquel, Gerard, 251 PAN (Guatemala). See Guatemala, political parties; Parliamentary right PAN (Mexico). See Mexico, political parties; Parliamentary right Panama: attitudes toward role of parties in democracy, 69; authoritarian regime, 220; constitutional regulations on representation, 32, 40, 62 (n7), 63 (n14); democratic stability,

315

83; democratization process, 216 (n3); district size and level of experience of legislators, 257; focus and style of representation, 34–37, 41–44, 54; influence of different actors on political decisions, 45–46; institutional trust, 82, 83; level of experience of legislators, 241–245, 257; mixed lists, 255; party vs. territorial interests, 43–44; perceptions of problems compared to national realities, 106, 108; perceptions of problems faced by government, 88–90, 97, 103, 105; regional interests, 42–44; threats to democratic consolidation, 78; type of democracy, 71; and United States, 229 Panama, political parties (PA, PRD, PS): and democracy vs. authoritarianism, 151, 152; ideology, 141; importance of different groups in decisionmaking, 125; internal democracy, 122–125; legislators’ ideological closeness to party positions, 155, 164–166; militancy, 118, 119; and military, 153; perceived intraparty life, 128, 129; perceptions on levels of participation, 121; preferred degree of power for party leaders, 124; programmatic coherence, 180–182, 185–187; PS as left party, 197–198; public spending on statism vs. market, 189; relation between party life and types of party systems, 130, 131; and religion and values, 149, 150; representation of polarization dimensions, 147–148; and social background, 152; and state intervention, 150, 151, 190 PAP (Peru). See Parliamentary left; Peru, political parties Paraguay: attitudes toward authoritarian government, 66, 71; attitudes toward democracy, 66, 71–73, 75; attitudes toward role of parties in democracy, 69; authoritarian regime, 220; Catholic Church and conservative parties, 220; closed and blocked lists, 255; constitutional regulations on representation, 32, 62 (n7); democratic stability, 83; focus and style of repre-

A-BM

10/2/07

3:40 PM

316

Page 316

INDEX

sentation, 34–37, 41–44, 51, 54; influence of different actors on political decisions, 45–46; institutional trust, 82, 83; level of experience of legislators, 241–245; party vs. territorial interests, 43–44; perceptions of problems compared to national realities, 106, 108; perceptions of problems faced by government, 88–90, 103, 105; perceptions on advantages of democracy, 75; perceptions on democratic consolidation, 77; regional interests, 42–44; threats to democratic consolidation, 78; type of democracy, 71; and United States, 229 Paraguay, political parties (ANR, MPQ, PEN, PLRA, UNACE): ANR as right party, 221–222; and democracy vs. authoritarianism, 151; focus and style of representation, 37–39; ideology, 141; importance of different groups in decisionmaking, 125; internal democracy, 123–125; legislators’ ideological closeness to party positions, 155, 164–166; militancy, 118, 119; and military, 153; perceived intraparty life, 129; perceptions on levels of participation, 121; preferred degree of power for party leaders, 124; programmatic coherence, 180–182, 185–187; public spending on statism vs. market, 189; relation between party life and types of party systems, 131; and religion and values, 149; representation of polarization dimensions, 147–148; and social background, 152; and state intervention, 150, 190 Parliamentary elites. See Political elites Parliamentary left, 195–217; and advantages of democracy, 76; anti-institutional emphasis, 204–205; conclusions about, 215, 268; and education, 211, 213; and employment policy, 211; ideological self-placement of legislators, 197; ideological selfplacement of parties, 198; and neoliberalism, 196, 205–206; parties listed, 197; perceptions of democracy and its institutions (elections, etc.), 199–205; perceptions of the

market, 205–214; and public health, 211; and public services, 211; and public welfare, 211–212; research methodology, 197–198; social base of, 215–216, 217 (n25); and state intervention vs. market, 187, 207–214; and taxes, 208; and transition to democracy, 195–196 Parliamentary right, 219–231; and advantages of democracy, 76; and authoritarian governments, 220, 230, 230 (n6); and Catholic Church, 220; characteristics of, 221–222; and colonial past, 230 (n4); conclusions about, 229–230, 270; and cultural issues, 223–228; differences between right parties and nonright parties, 222–229; ideological self-placement of legislators, 222; and ideology, 223–225; parties listed, 221; and political issues, 223–228; research methodology, 221–223; and socioeconomic issues, 223–228; and state intervention vs. market, 187; and United States, 220, 228 “Partocracy,” 116 Payne, J. Mark, 130 PC (Colombia). See Colombia, political parties; Parliamentary right PC (Uruguay). See Uruguay, political parties PCN (El Salvador). See El Salvador, political parties PDC (Chile). See Chile, political parties PDC (El Salvador). See El Salvador, political parties PEN (Paraguay). See Paraguay, political parties Pérez-Liñan, Aníbal, 14, 65 Peronism, 220, 221 Peru: attitudes toward authoritarian government, 66, 71; attitudes toward democracy, 66, 71; attitudes toward role of parties in democracy, 69; authoritarian regime, 220; constitutional regulations on representation, 32; democratic stability, 83; focus and style of representation, 34–37, 41–44, 51, 53–54, 56; influence of different actors on political decisions, 45–46; institutional trust, 82,

A-BM

10/2/07

3:40 PM

Page 317

INDEX 83; level of experience of legislators/quality of legislators, 241–245, 269; party vs. territorial interests, 43–44; perceptions of problems compared to national realities, 106, 108; perceptions of problems faced by government, 88–90, 98, 104, 105; political party crises, 72; regional interests, 42–44; threats to democratic consolidation, 78; type of democracy, 71; and United States, 229 Peru, political parties (FIM, PAP, PP, UN): and analysis of cleavages/ polarization, 145; and democracy vs. authoritarianism, 151, 152; focus and style of representation, 37–39; ideology, 141; importance of different groups in decisionmaking, 125; internal democracy, 123–125; legislators’ ideological closeness to party positions, 155, 164–166; militancy, 119; and military, 153; PAP as left party, 197–198; perceived intraparty life, 129; perceptions on levels of participation, 121; preferred degree of power for party leaders, 124; programmatic coherence, 180–182, 185–187; public spending on statism vs. market, 189; relation between party life and types of party systems, 131; and religion and values, 149; representation of polarization dimensions, 147–148; and social background, 152; and state intervention, 150, 190; UN as right party, 222 PJ (Argentina). See Argentina, political parties PL (Colombia). See Colombia, political parties PL (Honduras). See Honduras, political parties PLC (Nicaragua). See Nicaragua, political parties; Parliamentary right PLD (Dominican Republic). See Dominican Republic, political parties PLN (Costa Rica). See Costa Rica, political parties PLRA (Paraguay). See Paraguay, political parties PN (Honduras). See Honduras, political parties; Parliamentary right

317

PN (Uruguay). See Parliamentary right; Uruguay, political parties PNH (Honduras). See Honduras, political parties Polarization. See Political party cleavages and polarization Political agendas, 87–109; conclusions about, 108–109; disjunction between perceptions and national realities, 105–109; and ideology, 91; perceptions of problems by country, 88–90; perceptions of problems by country and ideology, 93–105 Political crises, 88–90, 268 Political disaffection, 78–81, 113, 132, 267–268 Political elites: defined, 2; and democratic transition, 266–267; distinguished from political class, 10; level of experience, 241–249, 257–261; level of experience correlated with sociodemographic variables, 246– 249; relation to parties, 3, 271–272; salaries, 239–241, 269; socialization of, 245–249, 271; sociodemographic and sociopolitical characteristics, 239–241; traditional route to politics, 2–3. See also Careers of political elites; Democracy; Ideology; Parliamentary left; Parliamentary right; Political elites, attitudes and perceptions of; Programmatic coherence of political parties; Research methodology Political elites, attitudes and perceptions of: attitudes toward authoritarian government in the presence of political instability, 65–66, 71–72, 83; attitudes toward democracy, 11–12, 65–85, 199–205, 267–268; attitudes toward elections, 67, 69, 71–72, 200–203; attitudes toward party leaders and interest groups, 161, 163, 170; attitudes toward state intervention vs. market, 187–191; ideological closeness to parties, 161, 163–166, 169–170; ideological self-placement, 197, 222, 239–240; ideological selfplacement compared to party placement, 155, 164–166; importance of, 267; influence on citizens’ values,

A-BM

10/2/07

3:40 PM

318

Page 318

INDEX

68. See also Parliamentary left; Parliamentary right Political instability: and attitudes toward authoritarian government, 65–66, 71–72, 83 Political parties, 268–269; “catch-all” parties, 186; caudillistas, 122, 129; citizens’ distrust of, 113, 132, 268; and clientism, 174, 187, 237, 268; crises in, 72, 83, 268; decisionmaking, 113–114, 116–118, 121–129; deficiencies of, 68; and democratic process, 66–67, 133–134; distribution of interviews by round, country, and political party, 23–25; and elites’ attitudes toward democracy, 68–71, 83, 268; elites’ confidence in, 83; evolution of, 115; and focus and style of representation, 35–39, 45–47, 49–50, 54–55; influence of party leaders and party militants on political decisions, 45–46; internal democracy, 123–125; intraparty ideological cohesion, 163; intraparty ideological differences, 161–171; level of activism, 47, 50; levels of participation, 113, 116–121, 124– 125; low levels of legitimacy, 113, 268; militancy, 117–118, 120, 134, 270; open vs. closed/blocked lists, 252–261; party coherence, 120; party vs. territorial interests, 43–44; perceptions of problems compared to national realities, 106–107; politicians’ roles in, 3; and quality of legislators, 250; and regulations on representation, 40, 63 (n14); relation between party life and types of party systems, 131; relation between society and parties, 69–70; systems of career advancement, 251–252; and transfuguismo (party changing), 21. See also Ideology; Parliamentary left; Parliamentary right; Political elites, attitudes and perceptions of; Political party organization; specific countries Political party cleavages and polarization, 139–160, 269–271; analysis and variables, 142–148; cleavage defined, 139; conclusions about,

153–154; and ideology, 141; polarization defined, 139; polarization dimensions, 141–149; and religion and values, 149–150; research methodology, 157, 159 (nn20–28); and self-placement of legislators on left-right dimension, 155; and social background, 152; survey questions, 157; utility of left-right dimension, 140–141, 154, 158 (n8), 269–271; views on democracy vs. authoritarianism, 151–152; views on military, 153; views on state intervention, 150 Political party organization, 113–137; caudillistas, 122, 129; coherence, 126–127, 136 (n31); conclusions about, 133–134; distribution of decisionmaking responsibilities, 113– 114, 116–118, 121–129; and ideology, 120, 125–128; institutionalization of party systems, 129–131; internal democracy, 118, 122–123, 126–128, 134; interparty vs. intraparty competition, 268–269; levels of participation, 113, 116–121, 124–125; militancy, 117–118, 126–127; militancy and party coherence, 120; organizational coherence, 174; and party behavior, 116; party leaders, 118, 122–127; Pearson correlation indexes, 126–127; perceived intraparty life, 128–129; preferred degree of power for party leaders, 124, 126–127; relation between party life and citizens’ evaluations of parties, 113–114; relation between party life and types of party systems, 129–131; research methodology, 117–118, 135 (n3); study of, 114–117; vitality of party organizations/party strength concept, 133 Political problems, perceptions of, 92–105, 108 “Political Reform and Values of Parliamentary Representatives in Four National Cases in Latin America” project, 8 Political socialization, 12, 245–249, 271 Political systems: and focus of representation, 49, 50; and style of representation, 48, 50, 54

A-BM

10/2/07

3:40 PM

Page 319

INDEX Porto Alegre Forum, 201 Poverty, 77–80, 109 (n2); disjunction between perceptions and national realities, 107 PP (Peru). See Peru, political parties PPD (Chile). See Chile, political parties; Parliamentary left PRD (Dominican Republic). See Dominican Republic, political parties PRD (Mexico). See Mexico, political parties; Parliamentary left PRD (Panama). See Panama, political parties PRE (Ecuador). See Ecuador, political parties PRI (Mexico). See Mexico, political parties PRIAN (Ecuador). See Ecuador, political parties Price controls, 210–211, 224–225 Principal-agent theory, 33 Privatization, 207, 209 Procedural representation, studies on, 29, 62 (n2) Professionalization of legislative careers, 238, 251, 268–269 Programmatic coherence of political parties, 173–194, 270; and accountability, 175; coherence in evaluation of problems, 178–184, 270; coherence in problem-solving strategies, 178–184, 270; conclusions about, 191–192; defined, 177; distinguished from ideological and organizational coherence, 174, 193 (n18); and ideology, 190–191; operationalization of concept, 175–179; party program defined, 173; previous research, 176; research methodology, 177–179, 192 (n9); significance of legislators’ attitudes vs. voting behavior, 176; and state intervention vs. market, 187–191; total programmatic coherence, 185–187 PRSC (Dominican Republic). See Dominican Republic, political parties; Parliamentary right PS (Chile). See Chile, political parties; Parliamentary left PS (Panama). See Panama, political parties; Parliamentary left

319

PSC (Ecuador). See Ecuador, political parties; Parliamentary right PSN (Guatemala). See Guatemala, political parties PSP (Ecuador). See Ecuador, political parties PU (Guatemala). See Guatemala, political parties Public health, 188, 189, 211–212 Public insecurity. See Violence and public insecurity Public opinion, influence on political decisions, 45. See also Citizens Public services, 211 Public welfare, 211–212, 224–225 PUD (Honduras). See Honduras, political parties PUSC (Costa Rica). See Costa Rica, political parties; Parliamentary right “Quality” of legislators, 235–264, 269, 271. See also Careers of political elites RD (Costa Rica). See Costa Rica, political parties Regional/territorial interests, 38–45, 57 Religion, 149, 157, 239–241, 245–249 Representation, 29–64; constitutional regulations on, 31–32, 40; degree of congruence between preferences of citizens and representatives, 30; as dimension of democracy quality, 12; directions for further research, 58; discrepancy between words and actions, 58; factors that condition representatives’ perceptions of roles, 32, 44–59; focus and style in political parties, 35–39; focus of (general vs. specific interests), 32–41, 44–46, 49–52; influence of different actors on political decisions, 45–46; and national mandates, 30–32; party vs. territorial interests, 43–44; perceptions on ownership of seat, 47, 50– 54, 57, 64 (n17); research methodology, 59–62; style of (following instructions vs. following individual preferences), 32–35, 41–48, 50, 54–58; and territorial interests, 38–44; types of research on, 29–30, 62 (n2)

A-BM

10/2/07

3:40 PM

320

Page 320

INDEX

Research methodology, 8–25; analysis and variables, 19; careers of political elites, 263 (nn22,23,27,28); data collection/sampling methods, 15–16; description and theoretical justification for study, 8–13; distribution of interviews by country and year, 17–18; distribution of interviews by round, country, and political party, 23–25; example questions, 20; and focus and style of representation, 59–62; methodological decisions, 16–22; objectives and methodological design, 13–16; overview of previous research, 8–10; and parliamentary left, 197–198; and parliamentary right, 221–223, 231 (n16); and political party cleavages/polarization, 157; and political party organization, 117–118, 135 (n3); and programmatic coherence, 177–179, 192 (n9); sampling issues, 21; theoretical frameworks, 10–11 Responsiveness, as dimension of democracy quality, 12 Rivas Pérez, Cristina, 139, 268, 270 RN (Chile). See Chile, political parties; Parliamentary right Rokkan, Stein, 153 Ruiz Rodríguez, Leticia M., 113, 173, 268, 270 Rule of law, 12 Salaries of legislators, 239–241, 269 Santiuste Cué, Salvador, 195, 270 Scarrow, Susan, 118 Schmitter, Philippe, 75 Scott, Robert, 266 Scully, Timothy, 129–130, 134 Shin, Doh Chill, 9 Shugart, Matthew, 252, 255 Snyder, James, 251 Social cleavages, and legislators’ perceptions of representation roles, 32 Social problems, 109 (n2); perceptions of problems by country, 88–90, 92; perceptions of problems by country and ideology, 93–105; perceptions of problems compared to national realities, 106; secondary role in agendas, 108

Socioeconomic issues: and parliamentary left, 211–212; and parliamentary right, 223–228 Socioeconomic status, and classification of political parties, 152, 157 Solari, Aldo, 1 Soviet Union, 195, 216 (n2) State intervention, 187–188; and classification of political parties, 150–151, 157; and parliamentary left, 205–217; and parliamentary right, 224–225; and programmatic coherence, 181–183 Stepan, Alfred, 12, 75 Stokes, Susan, 30, 193 (n25) Strom, Kaare, 133 Studies of political elites. See Research methodology Style of representation (following instructions vs. following individual preferences), 32–35, 41–44; defined/described, 33; explanatory factors, 44–48, 50, 54–58, 64 (n19); and political parties, 35–39, 47, 50–51; research methodology, 60–61 Substantive representation, studies on, 29–30 Suominen, Kati, 252 Taxes, 208 Taylor-Robinson, Michelle, 253, 264 (n47) Territorial interests. See Regional/territorial interests Terrorism and political violence, 77–80 The Political System: An Inquiry into the State of Political Science (Easton), 7 Torcal, Mariano, 81 Transfuguismo (party changing), 21 UCeDé (Argentina). See Argentina, political parties UCR (Argentina). See Argentina, political parties UCS (Bolivia). See Bolivia, political parties UDI (Chile). See Chile, political parties; Parliamentary right UN (Peru). See Parliamentary right; Peru, political parties

A-BM

10/2/07

3:40 PM

Page 321

INDEX UNACE (Paraguay). See Paraguay, political parties Unemployment, 109 (n2), 188, 189, 211, 224–225 United States: and authoritarian governments, 230 (n6); and globalization, 206; influence of integrity and competence on voting, 252–253; and parliamentary right, 220, 228, 230 (n6); and public opinion, 251; quality of legislators, 250–251, 264 (n37) Uruguay: attitudes toward authoritarian government, 71; attitudes toward democracy, 66, 71; attitudes toward role of parties in democracy, 69; Catholic Church and conservative parties, 220; closed and blocked lists, 255; constitutional regulations on representation, 32; democratic stability, 83; focus and style of representation, 34–37, 39, 40–44, 51, 53–54, 56–57; high number of quality legislators, 260, 269; influence of different actors on political decisions, 45–46; influence of interest groups, 57; institutional trust, 81, 82, 83; level of experience of legislators, 241–245, 260, 269; party vs. territorial interests, 43–44; perceptions of problems compared to national realities, 106, 108; perceptions of problems faced by government, 88–90, 104, 105; preferences regarding democracy and authoritarianism, 66; regional interests, 42–44; salaries for legislators, 269; threats to democratic consolidation, 78; type of democracy, 71; and United States, 229 Uruguay, political parties (EP/FA, PC, PN), 228; closeness of PC and PN, 228; and democracy vs. authoritarianism, 151; EP/FA as left party, 197–198; focus and style of representation, 37–39; ideology, 141; importance of different groups in decisionmaking, 125; internal democracy, 123–125; legislators’ ideological closeness to party positions, 155, 164–166; militancy, 119; and military, 153; perceived intraparty life, 129; perceptions on levels

321

of participation, 121; PN as right party, 221–222; preferred degree of power for party leaders, 124; programmatic coherence, 180–182, 185–187; public spending on statism vs. market, 189; relation between party life and types of party systems, 131; and religion and values, 149; representation of polarization dimensions, 147–148; and social background, 152; and state intervention, 150, 187, 189, 190 Valdini, Melody Ellis, 252 Varguism, 220 Venezuela: attitudes toward authoritarian government, 66, 71; attitudes toward democracy, 66, 71, 73, 267; attitudes toward role of parties in democracy, 69; constitutional regulations on representation, 32; democratic stability, 83; focus and style of representation, 34–37, 41–44, 54; influence of different actors on political decisions, 45–46; institutional trust, 81, 82, 83; lack of conservative parties, 221; level of experience of legislators, 241–245; party vs. territorial interests, 43–44; perceptions of problems compared to national realities, 106, 108; perceptions of problems faced by government, 88–90, 98, 105; political crises, 72, 268; regional interests, 42–44; threats to democratic consolidation, 78; type of democracy, 71; and United States, 229; voting and quality of legislators, 253 Venezuela, political parties (AD, COPEI, MAS, MVR): and analysis of cleavages/polarization, 145; centrist position of COPEI, 231 (n12); COPEI and Catholic Church, 220; and democracy vs. authoritarianism, 151; focus and style of representation, 37–39; ideology, 141; importance of different groups in decisionmaking, 125; internal democracy, 123–125; legislators’ ideological closeness to party positions, 155, 164–166; MAS as left party, 197–

A-BM

10/2/07

3:40 PM

322

Page 322

INDEX

198; militancy, 119; and military, 153; perceived intraparty life, 129; perceptions on levels of participation, 121; preferred degree of power for party leaders, 124; programmatic coherence, 180–182, 185–187; public spending on statism vs. market, 189; relation between party life and types of party systems, 131; and religion and values, 149; representation of polarization dimensions, 147–148; and social background, 152; and state intervention, 150, 151, 190 Verba, Sydney, 159 (n34)

Violence and public insecurity, 88–90; disjunction between perceptions and national realities, 107; and national realities, 105; and parliamentary right, 224–225; perceptions of problems by country, 92, 108; perceptions of problems by country and ideology, 93–105 Voting. See Elections World Social Forum, 216 (n4) Zechmeister, Elizabeth, 30, 162 Zovatto, Daniel, 130

A-BM

10/2/07

3:40 PM

Page 323

About the Book

THE PREMISE OF THIS BOOK IS, SIMPLY, THAT POLITICIANS MATTER— that an understanding of the role played by politicians in the way that politics is carried out in their countries, far from constituting a resurrection of outdated elitist theories, is of vital importance in present-day Latin America. The authors consider politicians as both cause and effect. Drawing on pioneering field research on the opinions of Latin American legislators in seventeen countries, they focus on two broad topics: political systems, as reflected in politicians’ perspectives on the democracies and party organizations of which they are a part, and the role played by ideology, not only in shaping positions on issues, but also as a factor that gives coherence to parties. Their seminal work is central to the current political dynamics in the region, as well as to broader debates in political science.

Manuel Alcántara Sáez is professor of politics and director of the Iberoamerican Institute at the University of Salamanca. His most recent books include ¿Instituciones o máquinas ideológicas? Origen, programa, y organización de los partidos latinoamericanos and Sistemas políticos de América Latina.

323