The Routledge Handbook of Elections, Voting Behavior and Public Opinion

Justin Fisher, Edward Fieldhouse, Mark N. Franklin, Rachel Gibson, Marta Cantijoch, Christopher Wlezien, Justin Fisher, Edward Fieldhouse, Mark N. Franklin, Rachel Gibson, Marta Cantijoch, Christopher Wlezien

• Publisher: Routledge
• Year: 2017
• ISBN:
  9781317494805
• Edition: 1
• Topic: Politics & International Relations
• Subtopic: Politics
• Index: Politics & International Relations

The study of elections, voting behavior and public opinion are arguably among the most prominent and intensively researched sub-fields within Political Science. It is an evolving sub-field, both in terms of theoretical focus and in particular, technical developments and has made a considerable impact on popular understanding of the core components of liberal democracies in terms of electoral systems and outcomes, changes in public opinion and the aggregation of interests.

This handbook details the key developments and state of the art research across elections, voting behavior and the public opinion by providing both an advanced overview of each core area and engaging in debate about the relative merits of differing approaches in a comprehensive and accessible way. Bringing geographical scope and depth, with comparative chapters that draw on material from across the globe, it will be a key reference point both for advanced level students and researchers developing knowledge and producing new material in these sub-fields and beyond.

The Routledge Handbook of Elections, Voting Behavior and Public Opinion is an authoritative and key reference text for students, academics and researchers engaged in the study of electoral research, public opinion and voting behavior.

PART I

Theoretical approaches to the study of voter behavior

1
DEMOCRATIC THEORY AND ELECTORAL BEHAVIOR¹

Ian McAllister

Normatively, democracy is a system of institutions, procedures and conventions. In practice, however, democracy is nothing without the citizens who provide it with form and substance. At democracy’s core are the beliefs, behavior and actions of citizens who make up the actual or potential democratic polity. As Angus Campbell and his colleagues put it in the seminal The American Voter study, “our approach is in the main dependent on the point of view of the actor” (Campbell et al. 1960: 27). Placing the citizen at the core of our understanding of democracy therefore provides a different perspective on how democracy operates, how it is changing, and perhaps most important of all, how it may change in the future. The purpose of this chapter is to outline how recent advances in the study of electoral behavior are altering democratic theory.

The backdrop to understanding the role that electoral behavior plays in shaping democratic theory is the rapid post-war advances in social science methodology. These advances have made it possible to explore, among other things: what citizens understand about democracy and what their expectations of it are; how citizens behave within a democracy and the political consequences of their behaviors; and what political actions citizens might take under certain hypothetical conditions. The rise of public opinion surveys in the 1950s and 1960s provided an early and strong empirical base for analyzing these changes. Based on more than half a century of opinion research, we know an immense amount about how citizens think and behave politically.

In contrast to much recent research, many of the early scholars of political behavior were as interested in the normative implications of their work as they were in its substantive findings. Bernard Berelson’s famous 1952 Public Opinion Quarterly article first gave voice to this approach by arguing that “opinion research can help a democracy to know itself, evaluate its achievements, and bring its practices more nearly in accord with its own fundamental ideals” (Berelson 1952: 313).² Another earlier pioneer of this approach was V. O. Key, who was particularly concerned about the role of public opinion in constraining political elites, and about elite responsiveness to citizen demands (Key 1966).

Despite this early and vigorous attempt to integrate theory with empirical research, it is fair to say that democratic theory, with some notable exceptions, has signally failed to keep pace with advances in electoral behavior.³ Rather than post-war advances in methodology heralding a new collaboration between theorists and behaviorists, both sides have largely followed independent intellectual paths. The first section of this chapter examines how we define democracy as it relates to electoral behavior, while the second section focuses on the consequences of electoral system design for behavior and theory. The third section examines four fields of electoral behavior which have specific implications for democratic theory, while the conclusion highlights the benefits of a closer relationship between behavioral and normative approaches to democracy.

Defining democracy

Before we can evaluate how advances in electoral behavior affect democratic theory, we need to arrive at an understanding of what is meant by democracy, and what constitutes its main components. Such an exercise has obviously occupied many academic studies and a review of the relevant literature is beyond the scope of this chapter. Briefly, we can identify two general approaches.

The first approach is to see democracy in terms of institutions and procedures. One of the earliest expressions was Schumpeter’s (1976 [1942]) definition of democracy as a competitive struggle for the popular vote. This view was further developed by Robert Dahl (1971), who saw democracy as involving contestation or competition on the one hand, and participation or inclusion on the other. Later studies have focused on outcomes, such as freedom and liberty, which can be defined in terms of the legal protections that citizens rely on (for reviews, see Diamond 1999; Ringen 2009).⁴ Whatever the focus, electoral behavior has relatively little role to play in this approach since it sees democracy as consisting of a framework of rules, rather than as a pattern of behavior.

A second approach has viewed democracy in terms of the social benefits that it can deliver, such as providing access to welfare and ensuring a basic standard of living. This view that democracy acts as a social safety net has been especially popular in low-income societies; there it is argued that political rights are meaningless in the absence of clearly demarcated social rights (see, for example, Przeworski 1985). Moreover, following Sen’s (1999) argument that democracy is more likely to increase living standards compared to other systems – since governments will have an electoral incentive to ensure fewer of its citizens remain in poverty – political and economic freedoms are often bracketed together. In this approach, too, behavior is seen to be of secondary importance to democracy when compared to social and economic performance.

One response to these varying definitions of democracy, and the implication that citizen opinions and behaviors are of secondary importance, has been to ask citizens themselves what they understand by the concept of democracy (Dalton, Shin and Jou 2007). This research has found that the public’s understanding of democracy is driven as much by political history as by a society’s economic development. Another response has been to use large-scale comparative surveys to measure various aspects of democracy (see Elkins 2000; Munck and Verkuilen 2002). Ferrín and Kriesi (2015), for example, show that Europeans mostly associate democracy with the rule of law and with free and fair elections. However, much of this research has been concerned with methodology rather than conceptualization. This has resulted in a poor fit between the normative and empirical aspects of democracy, and makes it difficult to arrive at agreed measures of democratic performance.

The approach taken by these studies is to define liberal democracy in terms of four key concepts which relate directly to electoral institutions; all four also overlap to a greater or lesser degree. First, accountability ensures that citizens can use regular elections to hold governments responsible for their performance while in office. Following O’Donnell (1998), vertical accountability is often distinguished from horizontal accountability. The former refers to the ability to hold elected representatives publicly accountable for their decisions while in office (Schedler 1999; Schmitter and Karl 1991), while the latter refers to the ability of government agencies to sanction one another for breaches of law or procedure. The ability of citizens in this way to steer governments toward policies they prefer has been a preoccupation of a large literature on dynamic representation (sometimes referred to as “thermostatic governance”).⁵

The second aspect of democracy is responsible government, which means that governments are appointed by a popularly elected legislature, remain answerable to them and can be removed from office by them. This system had its roots in the nineteenth-century Westminster system of government. In the mid-twentieth century, the concept of responsible government was revised to take into account a third aspect of democracy: political parties. In this approach, political parties are identified as the central agents of mass mobilization and the mechanism through which the system of responsible government operates (American Political Science Association 1950). In turn, responsible party government exists to ensure that the system performs to the satisfaction of its citizens; thus system performance is the fourth aspect of democracy focused on in this chapter. Before turning to how advances in electoral behavior have revised these four key concepts of democracy, the next section examines the key role of the electoral system and its impact on political behavior.

Electoral systems and democracy

What citizens think about democracy is agreed to be a key driver of the health of a democracy. If citizens are generally supportive of democracy, then it should thrive; if citizens are equivocal, then democracy may be under threat, particularly during periods of existential crisis. A large literature has examined public opinion during the transition from authoritarianism to democracy – what is called the process of democratic consolidation (see, for example, Bratton, Mattes and Gyimah-Boadi 2004; Diamond 1999). Once citizens come to hold a widespread view that democracy is what Linz and Stepan (1996: 5) characterize as “the only game in town,” then it can be properly said to have consolidated. But what constitutes widespread public support for democracy? And what are its mainsprings and how does public support vary according to differing institutional arrangements?

While scholars disagree about many aspects of public support for democracy, there is broad agreement that popular support for democracy is a highly complex phenomenon which is in a constant state of change (Klingemann 1999). This is especially the case in new democracies, which frequently retain elements of authoritarianism and where the norms, values and practices of democracy are often unfamiliar to many citizens. Measuring public support for democracy is therefore complex, and has to take into account different histories, cultures and institutional settings. Much of this complexity can be reduced to the distinction between whether the electoral system is majoritarian or proportional in its design. This is a distinction that was first highlighted by Lijphart (1999), who showed that it had particular consequences for the public’s views of democracy.

In the majoritarian model of government, the goal of the political system is to satisfy the demands of the majority of the electorate, and the selection of a government becomes the first priority of the electoral system. The majority controls the government through two mechanisms. The first is by providing the government with a mandate, so that they have the popularlyderived authority to implement the legislative program on which they won the election. The second mechanism is via accountability, so that political elites are held accountable by citizens for their actions while in office. These mechanisms have been criticized by electoral behavior scholars as unrealistic in terms of how party policies guide voters’ choices, and because winning parties rarely secure support from a majority of the electorate (see Thomassen 1994, 2014).⁶

In contrast to the majoritarian model, a proportional model of government is based on the assumption that elections produce elected representatives who reflect as closely as possible the characteristics of the electorate. Since it is rare under such a system for one party to achieve a governing majority, it is almost always the case that two or more parties must come together to form a governing coalition. An inevitable consequence of coalition government is that clarity of responsibility becomes blurred since it is never clear which party has been responsible for government policy (Aarts and Thomassen 2008). The ability of citizens to deliver an electoral reward or punishment for government performance is therefore weakened.

How do these two models of government shape the public’s view of democracy? This question has generated considerable research. In general, proportional or consensual models of government seem to work more efficiently than majoritarian models, at least from the electorate’s perspective. Powell (2000) argues that if elections are seen as a mechanism for linking citizen preferences with government outcomes, then the proportional model is more effective. Lijphart (1999) comes to similar conclusions, and he finds that the public’s satisfaction with their democracy is around 17 percentage points higher in proportional systems compared to majoritarian ones (see also Klingemann 1999, but cf. Aarts and Thomassen 2008).

Along with the public’s views of democracy, voting rules also have a wide range of consequences for electoral behavior. However, the interactions are complex and often intertwined with a country’s history and culture (for reviews, see Klingemann 1999; Norris 2004). In general, majoritarian systems have higher thresholds for election, which encourages political parties to aggregate their support and adopt a “catch-all” strategy in seeking votes. In proportional systems, the threshold for election is usually lower, which produces a larger number of parties that are then encouraged to pursue a narrower base of support around a more particularistic range of issues. This, in turn, has consequences for how parties mobilize opinion, and for partisanship, which is generally weaker in countries with multiple parties. This can lead to higher levels of electoral volatility (see Heath, this volume; Bowler, this volume).

Electoral behavior and democracy

While the framework of democratic theory is ever changing, this chapter argues that our understanding of citizens’ behavior and attitudes has advanced at a much greater pace. The period that has elapsed since the start of the quantitative revolution in the study of electoral behavior in the 1950s has resulted in numerous insights, and the last two decades have seen an unprecedented increase in the breadth and depth of such studies. Space prevents a detailed consideration of this research, but four areas are outlined here where developments have already fundamentally revised current theoretical approaches to democracy, or are likely to in the future. For each of these four areas, an outline of the findings of recent studies is provided, together with an evaluation of how and in what ways they influence democratic theory.

Political competency

One of the assumptions of liberal democratic theory is the existence of a competent electorate which is capable of making informed choices. Without a competent electorate, the public’s ability to use elections to hold governments accountable for their performance while in office will be limited. This, in turn, will undermine the core assumptions of a liberal democracy. As Berelson, Lazarsfeld and McPhee (1966: 308) put it half a century ago:

The democratic citizen is expected to be well informed about political affairs … to know what the issues are, what their history is, what the relevant facts are, what alternatives are proposed, what the party stands for, what the likely consequences are.

This optimistic view of the public’s politi...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Table of Contents
• Lists of figures
• List of tables
• Notes on contributors
• Acknowledgments
• Editors’ introduction – in defense of political science
• Part I Theoretical approaches to the study of voter behavior
• Part II Turnout: why people vote (or don’t)
• PART III Determinants of vote choice
• Part IV The role of context and campaigns
• Part V The nature of public opinion
• Part VI Methodological challenges and new developments
• Index