The two books I am reviewing both revolve around voting and parties in Latin America, but they differ in terms of the focus of their analyses (voters vs. parties and electoral outcomes), purpose (micro analysis of timeless voter motivations vs. macro explanation of partisan trends), and methods (statistics and methodological individualism vs. qualitative interpretations of partisan positions and shifts in voter support). While the differences are stark, together these books provide a comprehensive view of parties and elections in the region.
Kenneth Roberts’ book, Changing Course in Latin America, updates by a generation the foundational work on partisan politics in Latin America, Ruth Berins Collier and David Collier’s Shaping the Political Arena, in order to understand the leftward shift in the region’s politics that started at the end of the 1990s. Roberts accomplishes this by tracing the varied impacts of the 1980s economic crises and the subsequent neoliberal responses. The first half of the book is dedicated to developing the theory of critical junctures, while the second half provides an account of changes in the party systems based on antecedent conditions and other factors.
Roberts pushes and probes the theory of critical junctures in many ways. The overriding idea is that economic crises forced responses by the parties, and these responses have had long-term impacts on the party system itself. The “neoliberal critical junctures” he argues, “eroded party-society linkages and blurred the distinctions between the parties.” While the economic crisis affected all countries, its depth and duration varied, and the prior shapes of the party systems also varied. Roberts thus works at the structural and country level to explain the “divergent outcomes” he identifies. By the late 1990s, he finds a turn to the left, but the changes are conditioned by whether the systems in place were “elitist” or “labor mobilizing.”
The theoretical discussion is rich but challenging. To develop the complex theory in the first part of the book, Roberts uses names and countries as illustrative examples in ways that assume much knowledge. The second half of the book, which revolves around the difference between “elitist” and “labor-mobilizing” party systems, provides some of the necessary background and a rich, expansive discussion. He covers all of Central and South America plus Mexico, providing useful details about the dynamics of the party systems across the region. In emphasizing the impact of the economic crisis and the antecedent shape of the party system as he interprets the changes, Roberts leaves the reader with a theoretical basis for a comprehensive interpretation of change across time and differences across countries. The recent shift to the right in several countries may require a new analysis, but Roberts sets the stage by explaining the reasons behind the existing partisan alignments.
The Latin American Voter is an edited volume that aims to contribute to broad theories of voting using evidence from Latin America, tying chapters theoretically with the idea that voting is a function of structural and individual level factors. The more notable thread in the book is that all but one of the chapters are based on survey data from the Latin American Public Opinion Project. Latin Americanists and comparative behavioralists owe an important debt to LAPOP for the high-quality survey data that it has provided for over a decade. As evidenced by a multitude of studies, the project has enabled the empirical study of political attitudes and behavior. The current book only scratches the surface of potential projects, but the value of the survey data is clearly evident in the range of themes its authors analyze, including gender, voting turnout, polarization, crime and corruption, voters’ views of the economy and retrospective voting, and ethnicity.
Reviewing an edited volume is often unjust, because it is not possible to discuss the rich detail of the books’ 12 studies plus the theoretical introduction and conclusion. My goal, therefore, is to review the book’s theoretical outlines and highlight a few of the analytical insights.
As I noted, the central theoretical perspective emphasized by the book’s editors is that understanding Latin American (or other) voters requires consideration of both the voters themselves, as well as the context in which they live. This leads to the idea, tested in several chapters, that voters in some countries use policy positions of candidates to make voting decisions, while others are influenced by ethnicity, religion, partisanship, or other factors. But, these macro differences, which may be the result of political context and institutions, still interact with factors that differentiate the behavior of individual voters. In the conclusion, the editors state the finding in a negative way: “When it comes to who turns out and on what basis voters make decisions, factors outside the public’s direct control are of immense importance. Briefly put, political institutions and elites play critical roles in creating the conditions that facilitate or impede the public’s efforts to secure representation and accountability” (p. 347).
The book begins with a consideration of three models of voting: Voters can choose based on identity (membership in a group), they can pick candidates based on issues and policy positions, or they can choose based on the performance of the incumbent government. The authors set these traditional approaches to voting theory apart from a more pernicious view, that Latin American voters are captured by populists and/or by clientelism, thus breaking the party-voter bonds that support representation and accountability. In some cases more directly than others, the book’s authors explore these models and produce a complicated picture that varies across the region and by the manner in which subjects are addressed. For example, Elizabeth Zechmeister finds clearer left-right semantics in Chile, Uruguay, and Venezuela than in Brazil or Ecuador. Scott Mainwaring, Mariano Torcal, and Nicolás Somma find that class voting is strong in a different set of countries (Argentina, Bolivia, Brazil, Guatemala, Peru, and Venezuela). In another chapter, François Gélineau and Matthew Singer find that the economy plays a bigger role in elections in the Dominican Republic, Venezuela, El Salvador, Uruguay, and Peru than in other countries. Their study is based on voters’ responses to questions about the economy, but it emphasizes country-level variables (the role of the economy in the country, concentration of power, and degree of partisan polarization). To tie the findings to the book’s theory, it would be useful as a next step to also explore how voters’ individual characteristics make them more or less likely to consider the economy in their voting decisions. It would also be useful to synthesize the findings by considering the relationships among the dependent variables and explaining the different grouping of countries. Why, for example, is the set of countries where class drives voting different from those where left-right semantics are clear?
One variable that is left out of the studies is geography. The LAPOP surveys are nationally representative, and they have begun to do some oversampling for particular regions. This is a crucial variable for studies of voter behavior, because rural voters, for example, do not view the economy as do urban voters, and in some cases voters in one region face different partisan choices than in another. Daniel Moreno Morales’ chapter on ethnic voting is a case in point. The chapter is very valuable in exploring voters’ self-identity and how that identity conditions electoral choices. He finds, for example, that voters are more prone to use skin color than language for their identities. He then explores whether that identity is tied to voting and finds, as in other chapters in the book, variation across countries. Presumably there is also intra-country variation. Do voters in majority-indigenous areas, for example, react similarly to those in more heterogeneous zones? If so, perhaps living in a diverse area would move voters away from the use of identity and towards a retrospective view of policy in their electoral choices.
In looking at voting from micro and macro perspectives, these books could fit nicely as a core to a graduate course. The limitations of one are the strengths of the other. Roberts’s emphasis on partisan change as a reaction to neoliberal reforms, for example, is complemented by the authors in The Latin American Voter, who focus attention on the ways in which individual voters are influenced by policy and context. Those authors’ emphasis on other important factors that influence the vote (gender, class, ethnicity, religion, and partisanship) further helps contextualize the relative importance of the economic factors driving Roberts’ analysis. Finally, courses would benefit from debating the differing levels of analysis and methodological approaches; the micro questions from the Latin American Voter are open to sophisticated statistical testing given the resource of the LAPOP data, while Roberts’s interpretive and mostly qualitative approach seems appropriate for evaluating the broad theory and trends that motivate his book. Within this debate, it would also be useful to discuss the limitations of these approaches and alternatives, such as the use of panels to clarify changing voter moods. As political science continues to debate the merits of these approaches, both in terms of the value of the questions asked and the appropriate means for testing, these two books provide interesting counterpoints for discussion.
