Variable 049–

• Privatize all state industries.

• Privatize only those industries that produce scarce profits.

• Privatize all the industries that are not strategically relevant to the development of the nation.

• Leave things in their current state.

• Other [specify]

• Does not know [cannot read]

• No response

Variable 050–

• Privatize all public services.

• Privatize only those services of scarce profits.

• Privatize all public services except services that are highly necessary to the majority of the populace.

• Leave things in their current state.

• Other [specify]

• Does not know [cannot read]

• No response

Variable 054–

• Complete/total

• Very strong

• Moderate

• Low

• None

• Does not know

• No response

Variable 057, 059, and 061–

This book is the result of a long, didactic, and collegial process. In the time between the start and finish of this project, we experienced significant changes in our academic careers and our personal lives. Four of us made the transition from graduate student to assistant professor during the course of writing this book. As a group tally, we began the project with one little girl in our midst and ended with eight young children. At the project's start, we all lived a distance of roughly ten miles from each other around Duke University; at one point toward the project's end, we were spread across three continents.

Our research cluster gathered for the first time in the spring of 1998 in the office of the only author of this study who then already held a faculty position, Herbert Kitschelt. We had recently been given an early peek at a significant new dataset, the first round of the Parliamentary Elites of Latin America survey, created by scholars at the University of Salamanca under the leadership of Manuel Alcántara Sáez. Most of us present at this meeting were second- and third-year graduate students whose research interests focused on Latin America. The project would be to undertake a thorough investigation into the nature of programmatic party competition in Latin America, an analytical question on which Kitschelt had just completed a book with data from postcommunist Eastern Europe.

What political, economic, and social issues shape the political arena of Latin American legislatures? During the past two decades, dramatic changes have altered the economic and political makeup of Latin American societies, perhaps shifting the ideological bases of party competition. Though comparative researchers have delved into the deep cleavages that might shape politics in the region (cf. Dix 1989; Collier and Collier 1991), we know little about how these divides are mapped into political preferences at the elite level. Party systems are shaped by both societal cleavages and political institutions such as electoral rules (Lipset and Rokkan 1967; Cox 1997). In the stage set by cleavages and institutions, strategic politicians devise political programs to capture votes, sometimes energizing dormant societal divides to garner electoral support, sometimes reactively aligning their political platforms to shifting preferences in the electorate. In any case, politicians “map” multiple issues on essentially “either-or” propositions, in order to convey their own political preferences to cognitive misers in the electorate. The spatial analysis of politics refers to these issue bundles as “ideologies.” This chapter explores Latin American legislatures by analyzing the structure and contents of issue bundles.

We begin our empirical analysis of programmatic structuration by focusing on the attitudes, values, and opinions of legislative elites. In doing so, we confirm that the contents and structures of what we call ideological and partisan spaces vary immensely throughout the region.

Economic governance/privatization

1. v49 Privatize industry

2. v50 Privatize services

Social protection/social policy

1. v64 Price controls bad

2. v65 Do not sponsor job creation

3. v69 Do not provide housing

4. v70 Do not provide social security

5. v71 Do not provide unemployment insurance

6. v72 Basic subsidies bad

7. v150 More guns, less butter

Financial openness/closure

1. v54 Let IMF in

2. v57 US investment good

3. v59 European investment good

4. v61 Latin American investment good

Law and order

1. v34 Delinquency/robbery are threats

2. v35 Labor unrest threatens democracy

3. v87 Violence an important problem

4. v168 Corruption always existed

Traditional values/cultural libertarianism

1. v210 I go to church

2. v235 No abortion

3. v236 No divorce

4. v240 I'm very religious

Postmodern sensibilities

1. v90 Human/minority rights are important

2. v92 Environmental issues are important

Authoritarian propensities

1. v11 Democracy is never the best system

2. v15 Elections are never the best way

3. v26 Parties not needed for democracy

4. v32 Justice doesn't work

Other variables

1. v234Left-right self-placement is a continuous variable in the range 1 (left) to 10 (right).

2. v132bParties are a series of dummy variables specific to each country.

Where do national profiles of programmatic party system structuration come from? What are their correlates and consequences and what are their potential implications for the future of democracy in the region, particularly given that we can see a decade beyond our point of observation in the late 1990s and consider intervening developments? Part II is devoted to these causal questions, whereas the conclusion takes up more speculative themes concerning the subsequent development of Latin American politics after the turn of the millennium. Chapters 6 and 7 deal with the long-term and short-term developments that might have influenced economic PPS, the most powerful issue dimension in Latin American party systems insofar as they have any programmatic structuration. In Chapter 8, we take up religion and regime as further programmatic dimensions that make a difference in some of our countries.

Long-term causal mechanisms turn out to be decisive in accounting for the rise and tenacity of economic and religious programmatic alignments, whereas regime divides are driven by recent episodes of authoritarianism or manifest performance failures of current democracy. Nevertheless, exceptionally sharp and prolonged economic decline since the 1980s also erodes established economic PPS, at least within the subgroup of countries with favorable long-term conditions for economic PPS. Our analysis, however, suggests that neither good nor bad performance builds up economic PPS in countries with inauspicious long-term dispositions to programmatic party competition.

By the end of the twentieth century, most Latin American countries could look back on their longest uninterrupted and, as of this writing, still open-ended period of political democracy. No country except Chile and Uruguay had enjoyed a longer run of full or even constrained democratic competition in any previous era. Before 1980, only five Latin American countries had achieved episodes of more than twelve continuous years of democratic or semidemocratic rule, but by 1999 that threshold had been crossed by no fewer than six additional countries. By the turn of the millennium, four of the five countries with comparatively durable democratic experiences before 1970 had once again accumulated a record of more than twelve years of continuous democracy since their most recent democratic transitions. Consequently, Latin America stands out as the most uniformly democratic region on earth behind the established Western democracies and Japan. This is evidenced in Table I.1, which is based on Larry Diamond's recoding of Freedom House regime scores (Diamond 2002: 29–30, table 2).

Because of the regional prevalence and durability of democracy in Latin America since the 1980s, research on the region has begun to move away since the mid-1990s from the theme of political regime change and has started to focus more on the empirical “quality” of democratic institutions and processes of authoritative decision making. Our study contributes to this field of investigations.

In Part I we present data that describe the degree of programmatic structuration of Latin American party systems. Each chapter focuses on one of four indicators: the dimensionality of the issue space (Chapter 2), left-right semantics (Chapter 3), elite-mass programmatic linkage on key issues or representation (Chapter 4), and the ideological cohesion of individual parties (Chapter 5). The data that we use to measure each of these indicators come from the 1997 Parliamentary Elites of Latin America (PELA) project and, in the case of representation, are complemented by data from the 1998 Latinobarometer survey. Full descriptions of the PELA survey, the variables we use, and a discussion of several data issues are presented in Appendixes A–C and in Web Appendixes D and E (at www.cambridge.org/9780521114950).

Although many of our key findings are summarized in the introduction to the next section of the book, Part II, analysts of party systems more generally and of Latin American politics more particularly will want to inspect these data carefully. Many of our indicators are new and are better understood when analyzed in their appropriate context within each chapter. Moreover, the scope of each analysis is broad and, in several cases, unprecedented; the data permit the comparison of Latin American party systems not only to each other but to party systems in the advanced industrial democracies and the former communist regimes of Eastern Europe. This introduction to Part I begins by justifying our operationalization of each indicator.