Show Summary Details

Page of

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). (c) Oxford University Press, 2015. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Handbooks Online for personal use (for details see Privacy Policy).

date: 22 July 2017

Tactical Voting

Abstract and Keywords

This article reviews comparative research on electoral systems and voting behavior, identifying analytical gaps in the tactical voting literature. It starts by examining the core assumptions underpinning the classic approach to tactical voting. It then displays some empirical data about the estimates of this behavior reported in the literature. Contrasting experiences of theoretical and methodological issues in majoritarian and non-majoritarian systems are discussed. Focusing on some difficulties in the application of the “avoiding wasted vote” motivations under PR and mixed-member rules, the article furthers the consolidation of research on tactical voting beyond FPTP systems. The last section suggests that future scholarship in this field should move beyond the “wasted vote” approach to tactical voting and distinguish between instrumental and expressive motivations, consider the long-term consequences of tactical voting, and stress the endogenous nature of tactical voting and the role of parties and media in spreading it.

Keywords: tactical voting, strategic voting, wasted vote, first-past-the-post, proportional representation

A considerable literature stresses that tactical voting, broadly defined as a vote for a party that is not the preferred one, motivated by the intention of affecting the outcome of the election (Blais et al. 2001), matters to party system fragmentation.1 Building on the influential ideas of Maurice Duverger (1954), several empirical studies have found a negative relationship between the levels of tactical voting and the number of political parties. In this context, a broad consensus among political scientists views moderate levels of party system fragmentation as a prerequisite for accountability mechanisms to operate correctly (e.g., Carey and Hix 2011). Although this literature has made important advances in uncovering the individual and contextual causes of tactical voting, traditional accounts of this behavior are not without problems. Empirical works showing that tactical voting is the key determinant of electoral outcomes in first-past-the-post (FPTP) systems have been challenged on conceptual and methodological grounds. Most decisively, the previous literature that addresses the sources of this phenomenon still lacks a proper grasp of (1) the channels through which elites foster tactical voting and (2) the types of tactical voting that exist under other electoral rules (i.e., proportional representation [PR] and mixed-member systems).

Is tactical voting really confined to FPTP systems? Should scholars employ the same analytical tools to examine this phenomenon in other institutional settings? Unveiling the determinants of party system fragmentation is one of the oldest and most important research questions in political science. It is far beyond the scope of this article to exhaust the literature that has focused on this issue since the days of Duverger. In contrast, it provides a critical survey of the empirical literature on the sources of tactical voting from a comparative perspective. In particular, I seek to identify the conceptual and methodological problems associated with the accepted wisdom that attributes variation of electoral outcomes in FPTP systems to tactical voting. Moreover, I also question some of the assumptions underpinning the dominant approach to the study of tactical voting in this particular institutional context and their applicability to other electoral rules, using the experiences of several PR and mixed-member systems. By doing so, I clearly illustrate some analytical gaps present in the quantitative studies of the determinants of tactical voting in FPTP systems.

The interest in the causes and consequences of tactical voting is almost as old as the discipline. The earliest contemporaneous recognition of the phenomenon dates back to 1869, when Henry Droop stated:

Each elector has practically only a choice between two candidates or sets of candidates. As success depends upon obtaining a majority of the aggregate votes of all the electors, an election is usually reduced to a contest between the two most popular candidates or sets of candidates. Even if other candidates go to the poll, the electors usually find out that their votes will be thrown away, unless given in favour of one or other of the parties between whom the election really lies.

(Droop 1869, cited in Riker 1982, 756)

The relevance of tactical voting is not exclusively academic. In fact, political elites, mainly party leaders, have remarkably contributed to its spread, either directly by reminding the members of their parties about its importance, or indirectly by addressing their voters through mass media. For example, despite obtaining half a million more popular votes than his opponent, the Democrat sitting vice president, Al Gore, was narrowly defeated by the Republican challenger, George W. Bush, in the 2000 US presidential election. In that year the third candidate, Ralph Nader, collected almost three million votes nationwide. Tactical voting by Nader supporters for Gore could have easily made Bush lose the election (Burden 2005). The possibility that voting sincerely for Nader could give the election to Bush provoked some progressive celebrities and interest groups to mount a public relations campaign to persuade Naderites in battleground states to vote Democratic,2 claiming that “a vote for Nader is a vote for Bush” (Aldrich 1993, 160). Likewise, the Tactical Voting 1987 (TV 87) platform was created before the 1987 UK general elections to inform voters about what non-conservative candidates had a better chance of winning the seat at stake in each district facilitating as a result the coordination of centre-left voters (Johnston and Pattie 1991).3 Finally, the small Free Democratic Party (FDP) in Germany has repeatedly campaigned for “second” (party list) votes by trying to persuade supporters of the main center-right party (i.e., the Christian Democratic Union/Christian Social Union [CDU/CSU]) to cast a split (tactical) vote and support their first preference only in the nominal tier of the electoral system.4 In this respect, a general campaign message of the FDP in the 1987 election was, “A FIRST-class decision; your SECOND vote for a THIRD force!” (Roberts 1988).5 This pattern has been confirmed for more recent elections and in the case of the Greens (Saalfeld 2005).

Despite this long and explicit interest, and because the voting behavior literature on FPTP systems largely focused on the analysis of the electoral competition between the two main candidates, scholars did not incorporate tactical voting into their research agenda until the 1980s. There are several political and methodological reasons for this renewed interest in the topic. First of all, party systems during the first decades after World War II in Western Europe were described as “frozen” (Lipset and Rokkan 1967; Rose and Urwin 1970).6 In such a world, the number of parties was relatively small. In contrast, this pattern does not hold nowadays, and a considerable number of countries have registered rising levels of party system fragmentation very recently (Best 2010). The larger pool of available parties creates more opportunities to cast a tactical vote. Second, some strong third parties have emerged in the United States and the United Kingdom in the last couple of decades. The extraordinary performance of Ross Perot in the 1992 and 1996 US presidential elections and the increasing importance of the Liberal Democrats in the UK general elections, which generated the only two hung parliaments7 after general elections since 1929 (i.e., February 1974 and 2010), along with the so-called second-vote campaigns deliberately run by the minor parties (FDP and Greens) in Germany, mean that tactical voting can no longer be seen as inconsequential. Third, there has been an increase in multicandidate presidential primary races in the nomination processes in both of the US major political parties since the post-1968 reforms (Abramson et al. 1992). The fact that there were multiple viable candidates in these elections opened the door for tactical voting by primary voters. Finally, the development of maximum likelihood estimators (i.e., multinomial logistic regression specifications), which make it possible to estimate models of multiparty elections, and the increasing availability of cross-national survey data that capture voters’ preferences in different countries, have contributed to renewed academic interest in the study of tactical voting.

The rest of this article is organized as follows. First I review the “state of the art” in the literature on the definition and measurement of tactical voting. Then I use the contrasting experiences of FPTP and PR systems to underline the major insights of the works on the determinants of tactical voting and point out some of their most significant gaps. I conclude by outlining the substantive, methodological, and policy implications, focusing on ideas for further research.

Tactical Voting: Definition and Measurement

Research on the empirical causes of tactical voting has exploded in recent decades. Indeed, some of the hottest debates of the last part of the twentieth century in the field revolved around the ever-important but elusive question of what a tactical vote is and how to measure it. Political scientists have traditionally focused on the analysis of tactical voting in FPTP systems. However, it has been increasingly recognized that this behavior is also central to elections conducted under other types of electoral rules. This section focuses on instrumental tactical voters and discusses one of the most substantially contested topics in the field: the definition and measurement of tactical voting in FPTP systems. A brief terminological note is in order before addressing more substantive issues.

The terms “tactical” (e.g., Heath et al. 1991) and “strategic” (e.g., Cox 1997) voting have been used in the UK and the US literatures, respectively. I believe that “tactical” is the most appropriate term to refer to the behavior in which I am interested here, for etymological reasons. Whereas strategy suggests a sequence of actions, tactic usually refers to a single maneuver. For this reason, the term “tactical” is probably more accurate because, as we will see below, tactical voting is utility maximizing for a single election only.8

What is a tactical voter in single-member, simple-plurality electoral systems? According to Fisher (2004), these instances of tactical voting are defined by three criteria. First, a tactical voter has short-term instrumental motivations for his or her vote choice; that is, the voter wants to influence who wins in that constituency in that election. According to this logic, the tactical voter is only considering short-term factors. In fact, a tactical vote may harm the long-term (strategic) interests of the voter’s preferred party.9 In addition, a tactical voter only cares about which candidate receives more votes than the others at the district level and hence is elected. By contrast, voters are not tactical, at least not instrumentally tactical, if they do not vote for their first option for reasons that have nothing to do with their willingness to affect the electoral outcome; as result, a vote for a party that is not the preferred one is not necessarily a tactical vote (Franklin et al. 1994). Along these lines, expressive tactical voters should be identified as being motivated not by the desire to “avoid a wasted vote,” but rather by the wish to send some message or signal to their own party or to some other to alter their political course.10 This type of protest behavior also receives the label non-Duvergerian tactical vote. Likewise, standard tactical voters are not depicted as “portfolio maximizers,” who have an incentive to vote for parties that are likely to win the election at the national level (Cox 1997), or as “policy balancers,” who split their vote across different branches of government to obtain moderate policy outcomes (Alesina and Rosenthal 1995; Fiorina [1992] 1996).

Second, a tactical vote is a vote for a party other than the first preference. However, a vote for the most preferred party could conceivably be a tactical vote, because it could comprise tactical considerations as well (Abramson et al. 2010). For example, a voter could have come to like a party different from the one that she or he originally preferred, because the voter appreciates the fact of being on the winning side (Niemi et al. 1993). In this kind of situation, voters would still be tactical as long as they prefer to see another, nonvoted party win (Fisher 2004).

Third, the vote choice should be consistent with the expectations of the constituency result, the utility generated by the victory of each party, and the principle of utility maximization. Hence, those who weigh their preferences and the parties’ expected performance cast a tactical vote (Abramson et al. 1992). If the expected utility is higher for another party than for the preferred party, the voter is expected to vote tactically in favor of the lower-ranked party with the highest expected utility (Fisher 2004). In contrast, those who vote for their most preferred party without considering its electoral chances cast a sincere vote (Abramson et al. 1992).

This combination of motivations and preferences of the tactical voter closely mimics Downs’s theoretical characterization of the behavior of the rational voter. According to Downs (1957, 48), “a rational voter decides what party he believes will benefit him most: then he tries to estimate whether his party has any chance of winning.” Building on Downs’s original description of the rational voter, Blais and Nadeau (1996, 40) define a tactical vote as “a vote for a second-preferred party (candidate) rather than for the first-rated one, motivated by the perception that the former has a better chance of winning the election.” Hence, in principle we need information about a person’s vote motivations in order to study tactical voting (Fisher 2004). Accepting this last point has relevant implications for the measurement of tactical voting. According to Alvarez and Nagler (2000), the literature employs three different ways to measure the extent of tactical voting registered in a given election. The first involves the use of aggregate electoral returns and suffers from the classic ecological fallacy problem (Robinson 1950). Moreover, it lacks information regarding voters’ motivations. This is why scholars have become increasingly reluctant to employ this method, and others suggest that it was only justifiable in the past in the absence of high-quality individual data (Fisher 2004).

The second way to measure the extent of tactical voting is using reports of postelection survey respondents about the motivations for their voting behavior (Heath et al. 1991; Niemi et al. 1992). This approach was made possible in the early 1980s when the British Election Study asked the appropriate questions for the first time, and it has become the most widely used to study tactical voting, in spite of the intense methodological scrutiny it has been subjected to since then (Alvarez and Nagler 2000; Blais et al. 2005; Evans 2002; Evans and Heath 1993; Fisher 2004; Niemi et al. 1992). In their report on the 1987 UK elections, Heath and his colleagues defined tactical voters as those who reply to the reasons-for-voting question thus: “I really preferred another party, but it had no chance of winning in this constituency.” The other three possible responses were “I always vote that way,” “I thought it was the best party” and “other.” The respondents who chose “other” were subsequently prompted to describe their motives. Heath et al. (1991) also coded as tactical voters those who answered the latter question by saying that they had voted tactically and those who said that they had voted against a party or candidate.

The first obvious disadvantage of this method is that including tactical motivations among the list of preformulated alternatives risks the possibility that the question will elicit the desired answer, leading to an overestimate of tactical voting. In this sense, respondents who do not fit into the other two possible answers but find it difficult to articulate their motives may jump to the “smart” alternative of tactical voting. However, it is also true that voters may not readily accept that they have engaged in a behavior that sounds suspiciously unethical, leading to an underestimate of its actual frequency (Niemi et al. 1992). Hence, this approach does not necessarily involve an overreporting problem.

Within this second way of measuring tactical voting, Niemi et al. (1992) combine the response to the aforementioned closed-ended question with the answer to three other items to identify tactical voters. These additional items are responses to open-ended questions about why respondents voted for a particular party, why they no longer supported another party, and why they have never supported a particular party. Niemi et al. (1992) code as tactical voters all respondents who gave an answer to any of these questions that they judge to be indicative of tactical voting, including responses in which the tactical is at best implicit (e.g., “Wanted Conservative party to have more seats in Parliament”). This method, however, creates problems because of the inappropriate aggregation of items that do not measure the same construct (Evans and Heath 1993). The lack of relationship between tactical voters coded as such according to the open-ended questions and the so-called distance from contention11 leads Evans and Heath to advocate the sole use of the “reasons-for-voting” item. Niemi et al. (1993) and Franklin et al. (1994) react to this critique by distinguishing between instrumental and expressive tactical voters and identifying the former as people whose preferred party came third or lower in their constituency and who voted for a party that came first or second. However, the new procedure advocated by Franklin and his colleagues to measure instrumental tactical voting also fails to persuade Heath and Evans (1994), because motivational information from neither the closed-ended nor the open-ended questions is used.

In effect, a tactical vote is a vote mainly driven by a tactical consideration, as opposed to a vote motivated by habit or party identification/ideological closeness. The tactical voter may vote for a positively assessed party (as opposed to “the lesser evil”), but the purpose of “avoiding a wasted vote” offsets the importance of other types of motivations. For this reason, not measuring tactical voting on motivational terms is essentially wrong. In fact, Niemi et al. (1992, 239) claim in their first work on the topic that “there is much to be gained by asking respondents directly about tactical voting.” Finally, Franklin et al.’s (1994) method is also criticized because it does not allow for the possibility of tactical voters being mistaken in their perceptions of the likely chances of the various parties of winning the constituency.

Although the Heath et al.’s (1991) measure is relatively robust, some inconsistencies can arise. First of all, as these scholars themselves acknowledge, tactical voters will be considered inconsistent (i.e., sincere) if they end up voting for the party they really prefer. Second, Heath et al.’s measure can be easily revised and improved with the strength-of-feeling scores (Fisher 2004). In this sense, tactical voters will also be considered inconsistent in the following three circumstances: first, if they have a clear preference for a party on the basis of the strength-of-feeling scores, but this is the same as the party voted for; second, if they have a clear preference for a party on the basis of the strength-of-feeling scores, but it is not the “really preferred party”; and third, if they have a tie for first preference on the strength-of-feeling scores that does not include the “really preferred party.” These considerations lead to a substantial reduction in the level of tactical voting.12 Even more important, the decision to exclude all the inconsistent respondents from the pool of tactical voters appears to be a good one, because there is no consistent monotonic association between the excluded voters and distance from contention.

Finally, Alvarez and Nagler (2000) show that there is a strong postelection bias toward reporting a vote for winning parties the further away in time the interview is conducted from election day, hence increasing the likelihood of finding a tactical vote.13 For this reason, they advocate the use of a third approach to measure tactical voting that infers it from models of party preference. However, Alvarez and Nagler´s measure is not completely satisfactory, for two main reasons. First, it is of limited use because we cannot identify tactical voters or test the explanatory power of this phenomenon. Second, its implementation can induce a wrong coding of tactical voters due to an incorrect model specification (Fisher 2004). Moreover, if we only take the close-ended method of eliciting tactical voting, as Evans and Heath (1993) argue for, the length of time elapsed after the election does not bias the estimates of tactical voting obtained using self-reported motivations. In contrast, the bias is observed with the open-ended or the combined measures, affecting only the calculations of one set of authors, that is, Niemi and colleagues. Hence, using reports of postelection surveys is still the best way to measure tactical voting.

In short, there are two common measures of tactical voting with individual data in the current state of the art: direct and indirect (Blais et al. 2005). The direct method consists of specifying the conditions that need to be satisfied in order to code a vote as tactical. There are basically two such conditions: having short-term instrumental motivations and voting for a party other than the first preference. The indirect method involves constructing a model of vote choice and estimating, based on simulations, how many individuals would have voted differently if perceptions of the likely outcome of the election had had no effect on their decision. After hard and protracted discussions, I conclude, following Fisher (2004, 166), that tactical voting should ideally be measured directly rather than indirectly because of the formal concepts behind its definition.

Tactical Voting in First-Past-the-Post Systems

It seems fair to credit Maurice Duverger with being the most distinguished French political scientist of the last century (Benoit 2006). His chief contribution deals with party politics and electoral systems and can be summarized in what have come to be called Duverger’s law and Duverger’s hypothesis (1954). Since their original formulation, a vast majority of works on electoral systems have revolved around two main questions: How do the electoral rules shape the party system? To what extent are voters influenced by electoral systems? According to Duverger, the negative consequences of restrictive electoral rules on party system fragmentation are understood as the result of two mechanisms. First, minor parties are typically awarded a much smaller share of seats than the share of votes they receive. Second, the existence of this mechanical effect creates incentives for electoral coordination. As defined by Gary Cox (1999, 49), electoral coordination “refers to a variety of processes by which groups of voters and politicians coordinate their electoral actions in order to win more legislative seats or executive portfolios” (see also Riker 1982). Therefore, we expect electoral restrictiveness to decrease the number of parties by generating incentives for strategic entry or withdrawal on the part of political entrepreneurs and tactical voting on the part of voters (Cox 1997). Duverger uses the term psychological for these behavioral consequences of nonpermissive electoral laws for the party system’s size (Blais and Carty 1991).

The most frequent example of a nonpermissive electoral system is the one that comprises single-member districts and allocates its seats using plurality rule. In fact, Duverger (1954, 217) devotes his famous law to this type of electoral rules when he states that “the simple-majority single-ballot system favors the two-party system.” Since tactical voting is at the heart of Duverger’s law, most studies of tactical voting focus on FPTP systems.14 The modern empirical study of tactical voting was inaugurated by Shively in 1970. In the following decade, this subfield flourished, with several studies focused on Canada (Black 1978; Blais et al. 1974), the United Kingdom (Cain 1978; Spafford 1972), the United States (Bensel and Sanders 1979; Brody and Page 1973), and even the nominal tier of the German electoral system (Fisher 1973).

One of the main goals of this literature was to estimate the amount of tactical voting in a number of different countries. Without being exhaustive, Table 1 summarizes some of the estimates found in previous studies. As can be seen in this table, there is considerable variation in the amount of tactical voting estimated in FPTP systems, ranging from 3% (Canada 1997) to 17% (UK 1987).15 Most of this variation takes place across countries and over time. However, the amount of tactical voting estimated also varies within the same election. For example, there is a more than 10% gap in the amount of tactical voting estimated by Evans and Heath (1993) on the one hand and Niemi et al. (1993) on the other in the case of the UK 1987 general elections. These discrepancies are mainly due to the differences in the approaches used to measure tactical voting; as previously noted, studies that use responses of open-ended questions on why respondents voted for a particular party to code tactical voters (e.g., Niemi et al. 1993) tend to find higher levels of tactical voting.

Table 1. Published Estimates of Tactical Voting

Study

Election

Estimate of Tactical Voting (%)

FPTP Systems

Blais and Nadeau (1996)

Canada 1988

6

Blais et al. (2001)

Canada 1997

3

Cain (1978)

UK 1970

14.6

Fisher (1999)

UK 1983

5

Lanoue and Bowler (1992)

UK 1983

5.1

Johnston and Pattie (1991)

UK 1987

5

Evans and Heath (1993)

UK 1987

6.3

Heath et al. (1991)

UK 1987

6.5

Lanoue and Bowler (1992)

UK 1987

6.6

Evans (1994)

UK 1987

6.8

Franklin et al. (1994)

UK 1987

6.8–7

Alvarez and Nagler (2000)

UK 1987

7.2

Johnston and Pattie (1991)

UK 1987

7.7

Galbraith and Rae (1989)

UK 1987

10–12

Niemi et al. (1993)

UK 1987

17

Evans (1994)

UK 1992

9

Heath and Evans (1994)

UK 1992

9

Fisher (1999)

UK 1992

7.7

Fisher (1999)

UK 1997

8.5

Evans et al. (1998)

UK 1997

10

Fisher and Curtice (2006)

UK 2001

9

Fisher and Curtice (2006)

UK 2005

9

Abramson et al. (1992)

US 1988 (Dem. Primaries)

13

Abramson et al. (1992)

US 1988 (Rep. Primaries)

12.7–13.9

PR and Mixed-Member Systems

Pappi and Thurner (2002)

Germany 1994c

15

Gschwend (2001)

Germany 1998c

11.7

Duch and Palmer (2002)

Hungary 1997c

13.6

Blais et al. (2006)

Israel 2003

9

Irwin and Van Holsteyn (2012)

Netherlands 2002

29

Lago Peñas (2005)

Spain 1979-2000

0.4–2.6

García Viñuela et al. (forthcoming)

Spain 2000

7

García Viñuela and Artés (2012)

Spain 2000-2008

12–33a

García Viñuela et al. (forthcoming)

Spain 2008

3

Fredén (2014)

Sweden 2010

18

Kriesi (1998)

Switzerland 1995

16.2

Hobolt and Karp (2010)

Several countries

3.7–42.1b

(a) This percentage refers to the electoral base of United Left, the major victim of tactical voting in Spain.

(b) The highest value in the examined elections by these authors corresponds to Chile in 2005, while the lowest one refers to Italy in 2006.

(c) Mixed-member systems.

In recent years a game-theoretic literature has increasingly joined the aforementioned empirical works on tactical voting. These formal works also focus on party competition in single-member districts with three or more candidates and usually make the following four assumptions:

  1. 1. Voters only have one vote.

  2. 2. They are short-term instrumentally rational.

  3. 3. They have a party preference order.

  4. 4. They also have some expectations about the constituency result.

The Determinants of Tactical Voting in First-Past-the-Post Systems

Five factors increase the probability of casting a tactical vote in FPTP systems. First, a voter is more likely to be tactical, either instrumental or expressive, as the intensity of her or his first preference decreases; that is, voters are less reluctant to abandon their favorite options when they do not particularly like their first preferences. For example, Niemi et al. (1992) discovered that tactical voting rises among weak partisans in the United Kingdom. In contrast, Blais (2002) attributes the small amount of tactical voting registered in Canada to the strong preference for their party among third-party supporters. Second, a voter is more likely to be tactical, either instrumental or expressive, as the intensity of his or her second preference increases; that is, voters are keener on the idea of voting for a less preferred candidate when they do not particularly dislike their second options. Third, instrumental tactical voting is higher among citizens who have strong negative feelings about the expected winning party (Niemi et al. 1992). In contrast, expressive tactical voting is more frequent among those who are indifferent to the top parties in the constituency (Franklin et al. 1994). Fourth, a voter is more likely to be instrumentally tactical as the expected district marginality increases, that is, as the expected gap in the vote share of the two top parties goes down.16 This relationship is exactly the opposite when considering expressive tactical voters. Finally, a voter is more likely to be instrumentally tactical as the expected distance from contention increases, that is, as the expected gap in the vote share between the two preferred options goes up (e.g., Fieldhouse et al. 1996).17

District marginality and distance from contention imply that the voter should be able to form expectations about the electoral performance of parties. These are probably the most consistent features of works on tactical voting. However, they also create problems, because voters are not necessarily well informed about the electoral prospects of the parties, for three main reasons (Blais and Nadeau 1996). First of all, not all citizens are equally knowledgeable about politics. In overall terms, the better educated, the more politically interested, and the consumers of news media have better information about the electoral prospects of the different parties (Bowler et al. 2010), increasing the likelihood of casting a tactical vote (Niemi et al. 1992). In contrast, in a subsequent article, Franklin et al. (1994) show that tactical voting is unrelated to the level of education.

Second, consistent with voters’ need for dissonance reduction (e.g., Beasley and Joslyn 2001), voters are overly optimistic about the chances of their preferred options (Blais 2002). Likewise, political parties tend to systematically foster this wishful thinking among their supporters by presenting excessively positive calculations of their electoral prospects during campaigns. Hence, the assumed independence in most definitions of expectations from preferences is problematic. Moreover, voters come to truly like the top parties because either there is a documented strong tendency toward preferring winning options (the bandwagon effect; Rothschild and Malhotra 2014), or they tactically vote for them and try to avoid the subsequent cognitive dissonance generated by this behavior (Bølstad et al. 2013). In the latter cases, voters adjust preferences rather than expectations, and these adjustments take place after and not before elections.

Third, it is important to understand which sources of information voters rely on when forming their expectations about parties’ chances of winning. Earlier literature tended to assume voters use lagged information from the actual results of the previous election for these purposes. Similarly, other scholars claim that voters learn about the viability of the different competitors by using the electoral history heuristics (e.g., Lago 2008, 2012; Lago and Martinez i Coma 2012; Tavits and Annus 2006). In contrast, recent research has shown that contemporary information from opinion polls on the upcoming election is more likely to affect voters’ expectations (Guinjoan et al. 2014; Irwin and Van Holsteyn 2012). Bearing this in mind, Cox (1997, 122) argues that voters’ information about who will win and who will lose should decline as the number of polls available decreases.

In short, traditional studies of tactical voting used to focus on single-member, simple-plurality systems with three or more competing candidates/parties because of the important incentives that this type of electoral rules generates for this particular behavior. Although the list of determinants of tactical voting in this specific institutional context is long, tactical voters in FPTP systems are described as especially sensitive to two parameters when deciding to vote for a party other than the first preference for short-term instrumental motivations: the preferred party’s (lack of) electoral viability and the closeness between the two top candidates/parties. In the next section I examine whether these ideas apply to other institutional settings as well.

Tactical Voting in Proportional Representation and Mixed Systems

It could be argued that a fully proportional electoral system makes tactical voting irrelevant for most voters (Duverger 1954).18 In such a scenario, almost all parties would be viable, so nobody would have an incentive to cast a tactical vote. However, this institutional context does not actually exist, and most PR systems have, in practice, some way of limiting proportionality (Gallagher and Mitchell 2005). The most common institutional devices to avoid the proliferation of parties associated with very permissive rules are low-magnitude districts and legal thresholds (Carey and Hix 2011). Because of these two features, which are frequently present in PR and mixed-member systems, these types of electoral rules share with majoritarian systems the same pattern of redistribution of seats from small to large parties (Rae [1967] 1971, ch. 9),19 giving rise to several “tactical situations” (Cox 1997).20 Moreover, government coalitions are more likely to be formed when permissive electoral rules are used and hence the number of legislative parties is large (Cheibub et al. 2004). The higher frequency of multiparty cabinets under non-majoritarian systems also contributes to increasing the opportunities for tactical voting.

Empirical evidence from the small but growing literature on tactical voting in non-majoritarian systems suggests that a significant number of voters do indeed cast tactical votes in elections conducted in these institutional contexts (Bargsted and Kedar 2009; Blais et al. 2006; Bowler et al. 2010; Cox 1997; Cox and Shugart 1996; Duch et al. 2010; Fredén 2014; García Viñuela and Artés 2012; Hobolt and Karp 2010; Irwin and Van Holsteyn 2012; Lago Peñas 2005; Meffert and Gschwend 2010). Looking at Table 1, we see that tactical voting is far from uncommon under PR and mixed-member rules (Abramson et al. 2010). Even more important for my purposes here, the typology of tactical votes registered in these non-majoritarian systems is richer. Following Meffert and Gschwend (2010), we can identify at least three possible forms of tactical voting under non-majoritarian rules: (1) the typical desertion of a small party supporter in favor of the second preference, because the former is not expected to obtain a seat; (2) a rental vote of a major party supporter in favor of a preferred junior coalition partner perceived as uncertain to obtain a seat; and (3) explicit tactical coalition voting to influence the composition and/or portfolio allocation of the next coalition government.

Similar to what happens in FPTP systems, the first type of tactical vote under non-majoritarian rules implies avoiding a wasted vote. In fact, initial attempts to study tactical voting in PR systems extrapolated theoretical arguments developed in FPTP contexts (e.g., Cox 1997, chs. 5 and 10; Cox and Shugart 1996; Tsebelis 1986). To the extent that this kind of tactical voter in non-majoritarian countries wants to best influence the allocation of seats at either the district or the national level, the theory could be easily adapted to these other electoral rules. In the science of electoral systems, district magnitude has been traditionally characterized as “the decisive factor” to explain party system fragmentation and electoral disproportionality (e.g., Taagepera and Shugart 1989, 112). It is obvious that district magnitude is positively associated with the number of viable parties winning seats—the so-called M+1 rule—(Cox 1997; Reed 1990) and negatively associated with the level of disproportionality (Lijphart 1994: 95–117). This is why tactical voting exists in low-magnitude districts, single-member districts being the most extreme case, while it tends to disappear in larger (i.e., of magnitude six or more), multimember districts (Cox 1997, 103–121; Lago 2008).21

As previously noted, significant vote thresholds that parties need to cross to get any representation are the second device traditionally used to limit the degree of proportionality of electoral results in PR systems (Gallagher and Mitchell 2005). The functional equivalence between low-magnitude districts and high legal thresholds leads Lijphart (1994, 12) to claim that “legal thresholds and district magnitudes can be seen as two sides of the same coin.” Thresholds may also encourage voters to defect from parties they fear will fail to pass the threshold and vote for larger parties. The higher the threshold, the larger the number of voters facing the “wasted vote” dilemma (Gallagher 1991, 49), and, hence the higher the likelihood of registering a tactical vote.

Regarding the second type of tactical vote, Hobolt and Karp (2010), Meffert and Gschwend (2010), and Fredén (2014) have linked a rental vote of a major party supporter in favor of a small party in non-majoritarian systems to Cox’s (1997, 197) concept of “threshold insurance.” According to this idea, supporters of large parties may vote for junior coalition parties to ensure that they qualify for compensatory seats. Likewise, Tsebelis (1986) coins the term “inverse tactical vote” to refer to the instances in which supporters of a big party might vote for a small party. This type of tactical voting has been typically studied for the German case (e.g., Gschwend 2007; Pappi and Thurner 2002) and is usually associated with the existence of compensatory or corrective tiers. These higher electoral tiers allocate seats (sometimes called “adjustment”) to compensate the parties that are underrepresented at the lower level(s) and to correct disproportionalities that arise there (Shugart 2000). Parties usually need to pass a threshold in order to qualify for any of these seats, and some parties are known to be in danger of falling below it. When one of these parties is the potential junior coalition partner of a major party, supporters of the latter may decide to rent their vote to the small party and tactically vote for it. Unlike in majoritarian systems, this is not necessarily a zero-sum game, because the major party may lose very few seats as a consequence of this behavior, but it makes a huge difference in terms of seats for the small party and the potential coalition between the two of them whether the small party passes the threshold or not.

In regard to the third type of tactical vote, most elections in non-majoritarian systems result in the formation of coalition governments (Gallagher 2005). Although Downs (1957) argues that estimating expected utilities in PR systems with coalition governments is too complicated for voters, several arguments have been presented in the literature in this respect: (1) voters take coalitions into consideration because they are more interested in government policy (utility) than in the distribution of seats in the legislature (Blais et al. 2006; Hobolt and Karp 2010); (2) they vote tactically because they see a vote for a party outside a coalition as “wasted” (Bargsted and Kedar 2009; Bowler et al. 2010); and (3) they vote for a party other than the first preference because they want to move the coalition’s political center of gravity (Duch et al. 2010). Hence, these tactical coalition voters in non-majoritarian systems resemble those of the policy-balancing model in presidential systems (Alesina and Rosenthal 1995; Fiorina [1992] 1996), in the sense that they take policy considerations into account when making up their minds.

Finally, the combination of the first and the second types of tactical voting described in this section gives rise to the existence of split-ticket voting in mixed-member systems (Bawn 1999; Benoit et al. 2006; Duch and Palmer 2002; Karp et al. 2002).22 On the one hand, nominal seats are allocated on a FPTP basis in most mixed-member systems, and we already know that voters have incentives to desert their first preference if it has little chance of winning the election in plurality rule elections. So minor party supporters typically vote for one of the two major parties at the majoritarian tier. On the other hand, the existence of minimum vote thresholds that parties need to pass in order to qualify for PR seats can create strong incentives for large party supporters to cast a tactical rental vote for a small and preferred coalition partner at the list tier. Hence, if minor party supporters cast a tactical vote in favor of a major party candidate at the majoritarian tier, and major party supporters cast a tactical vote in favor of a minor party list at the proportional tier, they end up behaving the same way. In other words, they become split-ticket voters because they choose candidates and lists from different political parties (a major party and a minor party at the nominal and the list tier, respectively) when an office is being decided by a single election.

Four main problems emerge regarding the extrapolation of the assumptions made for the study of tactical voting in FPTP systems to other institutional settings. First of all, as previously noted, a tactical vote is a vote for a party other than the first preference. Hence, we need information about voters’ first-party preferences to establish whether a vote is tactical or not. However, voters are often equally attracted to two (or more) parties under non-majoritarian rules. For example, about a fifth of all Dutch voters seem to experience this situation (Irwin and Van Holsteyn 2012).23 In order to break ties in the party preference ordering, Blais and Gschwend (2010) propose to use the party identification question, then the party leaders’ evaluations, and finally, to do it randomly.24

The second main problem concerning the application of the classic definition of tactical voting outside simple plurality electoral systems lies in the higher level of sophistication and/or political knowledge voters need to become tactical there. FPTP systems have probably the least complex electoral rules in the world (Blais and Massicotte 1996). In contrast, the operation of some PR and mixed-member systems is very complicated. Paraphrasing Eckstein (1963, 249), it is the easiest thing in the world to become inextricably tangled among the complexities of non-majoritarian electoral systems. For example, the allocation of seats among parties in Germany depends on the aggregate (national) number of votes cast for the parties’ land lists (“second votes”) (Saalfeld 2005). However, it is not clear that German voters generally understand the precise operation of their electoral system. In this sense, Jesse (1988) shows that only 45% of German voters identify the party vote as the most important.

A different, albeit related, problem has to do with the number of parties in elections conducted under PR (or mixed-member) rules. The simplicity of FPTP is associated with the presence of only a few viable parties in each district (Duverger 1954). In contrast, large party systems are the norm in countries that employ permissive electoral institutions (Gallagher 2005). The instrumental goal of maximizing expected utility by voting for a second or lower preferred party quickly becomes a very challenging task, and the actual operation of tactical voting depends heavily on the “electoral history heuristics” when the number of parties is large (Lago 2008).

Finally, fewer votes are “wasted”, comparatively speaking, in non-majoritarian systems (Uggla 2008). Hence, fewer voters find themselves in the classic tactical situation described for FPTP rules in these other contexts. As Duverger (1954) and Cox (1997) have claimed, the “psychological” effect plunges when the number of seats in an electoral district reaches five, because the margin between success and failure becomes blurred for the voters. In other words, the expected utility of voting for the preferred party is rarely zero in PR and mixed-member systems as a result of their higher permissiveness. However, this does not mean that wasted votes do not exist at all in such institutional frameworks. Conceptually, all the “excess” votes cast for the preferred party are wasted in the sense that they do not matter for the final allocation of the seats (Cox and Shugart 1996), but this is something that only few voters know, and even more important, that almost none of them take into consideration when they cast their votes. Hence, tactical voting to avoid an excess vote is the exception rather than the norm in non-majoritarian systems.

Conclusion

The significant development of the tactical voting research from the quantitative perspective over the last thirty years has also involved a qualitatively important transformation in how scholars approach this issue. This change mainly involves the publication of several studies whose scope is increasingly broader; its approach is progressively comparative and its methodological sophistication is gradually increasing. In the early days the study of tactical voting largely focused on FPTP systems. Moreover, it lacked an adequately comparative approach and often fell short of quantitative rigor. Fortunately, this is no longer the case. The progress recently made in this regard has been decisively assisted by enormous improvements in the development of maximum likelihood estimators and the accessibility of cross-national survey data. Likewise, the study of the different types of tactical voting in non-majoritarian systems is an additional subfield that has attracted the attention of increasing numbers of political scientists in recent decades.

The end of cleavage politics (Franklin et al. 1992) and the declining importance of party identification (Dalton and Wattenberg 2000) are two recent political developments that have fueled the rising interest in tactical voting research. Regarding the former, suffice it to say that several previous studies found substantial evidence of the decreasing impact of social class on voting behavior throughout Europe (Oskarson 2005), though this phenomenon varies significantly across countries (Evans and de Graaf 2013). Likewise, there has been a large body of literature in recent decades referring to the growing number of people who have no party identification and the subsequent reduction in the power of this variable to explain voting behavior (Holmberg 2007), though this decline is far from linear (Berglund et al. 2005) and is largely confined to Europe (Bartels 2000). On the supply side of the political market, the growing depolarization of party systems in European democracies and the subsequent decline in ideological voting (Mair 2007) have also contributed to higher volatility rates (Mair 2013), and more important for our purposes here, to an increasing fragmentation of established party systems due to the emergence of new parties (Hino 2012). Within this context, new theories and refreshing evidence on tactical voting are particularly useful.

As I have reported in Table 1, there is a great deal of variation in the estimates of tactical voting reported in the literature, ranging between the 0.4% found by Lago Peñas (2005) for the 1996 Spanish election to the 42.1% reported by Hobolt and Karp (2010) for the 2005 Chilean election. These differences in previous works hinge on the type of electoral system employed in each case and the various approaches adopted by scholars to measure the behavior of interest. Surprisingly enough, no clear empirical pattern has emerged with regard to the variation in the amount of tactical voting registered over time. Despite the aforementioned recent political processes, which should foster the estimated amounts of tactical voting, there is no evidence of the general positive trend that other scholars have documented for analogous political behaviors (i.e., ticket splitting between 1957 and 1998 in Germany; Klingemann and Wessels 2001).

I have sketched the main types of tactical voting registered in non-majoritarian systems and argued that the study of instances of this behavior under such electoral rules is one of the most promising avenues for further research in the field. In PR countries with a national threshold of representation, voters concerned about the possibility of casting a wasted vote may tactically abandon small parties in danger of failing to pass it, whereas others may do the opposite precisely to avoid this outcome. Who are the former, and who are the latter? For instance, it could be argued that the distinction between these two behaviors is rooted in the individuals’ ideology. According to this idea, right-wing voters would be more afraid of wasting their vote and hence more likely to tactically defect from a party in danger of not entering the parliament.

A different possible strand of future research involves the refinement of the theory on the determinants of expressive (protest) tactical voting, as well as its potential long-term consequences. The proposition that the vote for small parties in second-order elections should have spillover effects that manifest at the national level remains largely untested, even though we should see such effects if voting is habit forming, as expected (Franklin 2004). Within this framework, it would be interesting to examine whether tactical voting can also have significant effects on citizens’ satisfaction with democracy. Although there is a vast empirical literature on the impact of electoral systems on overall satisfaction with democracy (Aarts and Thomassen 2008; Anderson and Guillory 1997; Dalton and Anderson 2011), there has been little systematic research on the microfoundations of this effect. Theoretically, tactical voters should be less satisfied than sincere voters, controlling for the winner/loser gap.

Methodologically speaking, the preference for the comparative approach and direct measurement of tactical voting should be reflected in the inclusion of the appropriate questions in the questionnaires of the major cross-national survey research projects. As explained above, Heath et al.’s (1991) approach is a priori the most attractive way of measuring tactical voting. According to these authors, identification of tactical voting should be based on information regarding vote choice, party preferences, and the motivations for vote choice. Unfortunately information on party preferences is sometimes missing, and we too often lack the proper survey questions to infer voters’ motivations in our determination of how much voting is tactical (e.g., in the Comparative Study of Electoral Systems project).

It is also always advisable to field as many panel surveys as possible to study the relationship between voters’ expectations of the constituency result and their subsequent behavior. In recent decades available comparative panel data on elections have expanded from a limited subset of nations (primarily the United States and the Netherlands) to include many other democracies (established and new), such as the United Kingdom (British Election Study), Spain (Center for Sociological Research Studies), New Zealand (New Zealand Election Study), Mexico (Mexican Election Panel Study), and Germany (German Longitudinal Election Study), among others. The increasing accessibility of this type of information also refers to some comparative data sets, such as the 2014 European Election Studies (on European Union countries) or the Making Electoral Democracy Work project (for twenty-six elections in total in Canada, France, Germany, Spain, and Switzerland).

Panel data have the additional virtue of allowing us to address potential endogeneity problems. Distinguishing causes from effects in real-world settings has become a first-order concern for voting behavior specialists in the last decade. If panel data are available, we can exploit the time dimension of the subjects to control for the unobserved. In addition, we can overcome research limitations posed by endogeneity problems by performing laboratory experiments. Experiments permit the researcher to re-create existing conditions in actual politics and to ascertain their effects on behavior while holding other factors constant; the results are therefore unambiguous with respect to causal direction. Finally, I believe that some institutional settings offer an excellent opportunity for theory building and testing by providing us with quasi-experimental designs. If in a particular place two different electoral systems are operating at the same time, we can take advantage of this controlled comparison to study how the rules of the game influence the occurrence of tactical voting.

Finally, how voters form their perceptions of the parties’ chances of winning and whether political actors (parties and media) can modify these views remain open empirical questions. Do voters basically take into consideration the previous electoral results when forming their perceptions of the parties’ chances of winning? Or do they, in contrast, update their beliefs about parties’ electoral prospects based on the information conveyed by preelection surveys? In my view, highly politically aware voters should be better able to form rational expectations about the outcome of an election. However, this should depend on parties’ (and media’s) attempts to prime or deprime tactical considerations during an electoral campaign.

References

Aardal, B. 2011. “The Norwegian Electoral System and Its Political Consequences.” World Political Science Review 7 (1). doi: 10.2202/1935-6226.1105Find this resource:

Aarts, K., and J. Thomassen. 2008. “Satisfaction with Democracy: Do Institutions Matter?,” Electoral Studies 27 (1): 5–18.Find this resource:

Abramson, P. R., J. H. Aldrich, A. Blais, M. Diamond, A. Diskin, I. H. Indridason, D. Lee, and R. Levine. 2010. “Comparing Strategic Voting under FPTP and PR Systems.” Comparative Political Studies 43 (1): 61–90.Find this resource:

Abramson P. R., J. H. Aldrich, P. Paolino, et al. 1992. “Sophisticated Voting in the 1988 Presidential Primaries.” American Political Science Review 86: 55–69.Find this resource:

Aldrich, J. H. 1993. “Rational Choice and Turnout.” American Journal of Political Science 37: 246–278.Find this resource:

Alesina, A., and H. Rosenthal. 1995. Partisan Politics, Divided Government, and the Economy. New York and Melbourne: Cambridge University Press.Find this resource:

Alvarez, R. M., F. J. Boehmke, and J. Nagler. 2006. “Strategic Voting in British Elections.” Electoral Studies 25: 1–19.Find this resource:

Alvarez, R. M., and J. Nagler. 2000. “A New Approach for Modelling Strategic Voting in Multiparty Elections.” British Journal of Political Science 30: 57–75.Find this resource:

Anderson, C. J., and C. A. Guillory. 1997. “Political Institutions and Satisfaction with Democracy: A Cross-national Analysis of Consensus and Majoritarian Systems.” American Political Science Review 91 (1): 66–81.Find this resource:

Bargsted, M. A., and O. Kedar. 2009. “Coalition-Targeted Duvergerian Voting: How Expectations Affect Voter Choice under Proportional Representation.” American Journal of Political Science 53: 307–323.Find this resource:

Bartels, L. B. 2000. “Partisanship and Voting Behavior, 1952–1996.” American Journal of Political Science 44: 35–50.Find this resource:

Bawn, K. 1999. “Voter Responses to Electoral Complexity: Ticket Splitting, Rational Voters and Representation in the Federal Republic of Germany.” British Journal of Political Science 29 (3): 487–505.Find this resource:

Beasley, R. K., and M. R. Joslyn. 2001. “Cognitive Dissonance and Post-Decision Attitude Change in Six Presidential Elections.” Political Psychology 22 (3): 521–540.Find this resource:

Benoit, K. 2006. “Duverger’s Law and the Study of Electoral Systems.” French Politics 4 (1): 69–83.Find this resource:

Benoit, K., D. Giannetti, and M. Laver. 2006. “Voter Strategies with Restricted Choice Menus.” British Journal of Political Science 36 (3): 459–485.Find this resource:

Bensel, R. F., and M. E. Sanders. 1979. “The Effect of Electoral Rules on Voting Behaviour: The Electoral College and Shift Voting.” Public Choice 34 (1): 69–85.Find this resource:

Berglund, F., S. Holmberg, H. Schmitt, et al. 2005. “Party Identification and Party Choice.” In The European Voter, edited by J. Thomassen, 105–123. Oxford: Oxford University Press, ECPR.Find this resource:

Best, R. E. 2010. “Increasing Irrationality? The Equilibrium Relationship between Electoral and Legislative Party System Size, 1950–2005.” Electoral Studies 29 (1): 105–116.Find this resource:

Black, J. H. 1978. “The Multicandidate Calculus of Voting: Application to Canadian Federal Elections.” American Journal of Political Science 22 (3): 609–638.Find this resource:

Blais, A. 2002. “Why Is There So Little Strategic Voting in Canadian Plurality Rule Elections?” Political Studies 50: 445–454.Find this resource:

Blais, A. 2004. “Strategic Voting in the 2002 French Presidential Election.” In The French Voter: Before and After the 2002 Election, edited by M. Lewis-Beck, 93–109. Basingstoke, UK: Palgrave Macmillan.Find this resource:

Blais, A., J. H., Aldrich, Indridason, I. H., et al. 2006. “Do Voters Vote for Government Coalitions? Testing Downs’ Pessimistic Conclusion.” Party Politics 12: 691–705.Find this resource:

Blais, A., and R. K. Carty. 1991. “The Psychological Impact of Electoral Laws: Measuring Duverger’s Elusive Factor.” British Journal of Political Science 21 (1): 79–93.Find this resource:

Blais, A., and T. Gschwend. 2010. “Strategic Defection across Elections, Parties, and Voters.” In Citizens, Context, and Choice: How the Political Context Shapes Electoral Behavior, edited by C. J. Anderson and R. J. Dalton, 176–195. Oxford: Oxford University Press.Find this resource:

Blais, A., and L. Massicotte. 1996. “Electoral Systems.” In Comparing Democracies. Elections and Voting in Global Perspective, edited by L. LeDuc, R. Niemi, and P. Norris, 49–81. Thousand Oaks, CA: Sage.Find this resource:

Blais, A., and R. Nadeau. 1996. “Measuring Strategic Voting: A Two-Step Procedure.” Electoral Studies 15 (1): 39–52.Find this resource:

Blais, A., R. Nadeau, E. Gidengil, and N. Nevitte. 2001. “Measuring Strategic Voting in Multiparty Plurality Elections.” Electoral Studies 20: 343–352.Find this resource:

Blais, A., F. Renaud, and R. DesRosiers. 1974. “L’Effet en Amont de la Carte Electorale: Le cas de la Region de Quebec a l’Election Federale de 1968.” Revue Canadienne de Science Politique 7: 648–672.Find this resource:

Blais, A., R. Young, and M. Turcotte. 2005. “Direct or Indirect? Assessing Two Approaches to the Measurement of Strategic Voting.” Electoral Studies 24: 163–176.Find this resource:

Bølstad, J., E. Dinas, and P. Riera. 2013. “Tactical Voting and Party Preferences: A Test of Cognitive Dissonance Theory.” Political Behavior 35 (3): 429–452.Find this resource:

Bowler, S., J. A. Karp, and T. Donovan. 2010. “Strategic Coalition Voting: Evidence from New Zealand.” Electoral Studies 29: 350–357.Find this resource:

Brody, R., and B. Page. 1973. “Indifference, Alienation and Rational Decisions.” Public Choice 15 (1): 1–17.Find this resource:

Burden, B. C. 2005. “Minor Parties and Strategic Voting in Recent U.S. Presidential Elections.” Electoral Studies 24: 602–618.Find this resource:

Cain, B. E. 1978. “Strategic Voting in Britain.” American Journal of Political Science 22 (3): 639–655.Find this resource:

Carey, J. M., and S. Hix. 2011. “The Electoral Sweet Spot: Low-Magnitude Proportional Electoral Systems.” American Journal of Political Science 55 (2): 383–397.Find this resource:

Cheibub, J. A., A. Przeworski, and S. M. Saiegh. 2004. “Government Coalitions and Legislative Success under Presidentialism and Parliamentarism.” British Journal of Political Science 34: 565–587.Find this resource:

Cox, G. 1999. “Electoral Rules and Electoral Coordination.” Annual Review of Political Science 2 (1): 145–161.Find this resource:

Cox, G. W. 1997. Making Votes Count: Strategic Coordination in the World’s Electoral Systems. Cambridge, UK: Cambridge University Press.Find this resource:

Cox, G. W., and M. S. Shugart. 1996. “Strategic Voting under Proportional Representation.” Journal of Law, Economics, and Organization 12: 299–324.Find this resource:

Dalton, R. J., and C. Anderson. 2011. Citizens, Context, and Choice: How Context Shapes Citizens’ Electoral Choices. Oxford: Oxford University Press.Find this resource:

Dalton, R. J., and M. P. Wattenberg, eds. 2000. Parties without Partisans: Political Change in Industrial Democracies. Oxford: Oxford University Press.Find this resource:

Downs, A. 1957. An Economic Theory of Democracy. New York: Harper & Row.Find this resource:

Duch, R. M., J. May, and D. A. Armstrong II. 2010. “Coalition-directed Voting in Multiparty Democracies.” American Political Science Review 104: 698–719.Find this resource:

Duch, R. M., and H. D. Palmer. 2002. “Strategic Voting in Post-Communist Democracy?” British Journal of Political Science 32: 63–91.Find this resource:

Duverger, M. 1954. Political Parties. London: Methuen.Find this resource:

Eckstein, H. 1963. “A Perspective on Comparative Politics: Past and Present.” In Comparative Politics, edited by H. Eckstein and D. Apter, 3–32. New York: Free Press.Find this resource:

Evans G. 1994 “Tactical Voting and Labour’s Prospects.” In Labour’s Last Chance?, edited by A. F. Heath, R. Jowell, J. Curtice and B. Taylor, 65–84. Aldershot: Dartmouth.Find this resource:

Evans, G. 2002. “On the Validation of Measures of Strategic Motivations: A Critical Comment on Alvarez and Nagler.” British Journal of Political Science 32: 185–188.Find this resource:

Evans, G., J. Curtice, and P. Norris. 1998. “New Labour, New Tactical Voting? The Causes and Consequences of Tactical Voting in the 1997 General Election.” British Elections and Parties Review 8: 65–79.Find this resource:

Evans, G., and N. D. de Graaf, eds. 2013. Political Choice Matters: Explaining the Strength of Class and Religious Cleavages in Cross-national Perspective. Oxford: Oxford University Press.Find this resource:

Evans, G., and A. Heath. 1993. “A Tactical Error in the Analysis of Tactical Voting: A Response to Niemi, Whitten and Franklin.” British Journal of Political Science 23: 131–137.Find this resource:

Fieldhouse, E. A., C. J. Pattie, and R. J. Johnston. 1996. “Tactical Voting and Party Constituency Campaigning at the 1992 General Election in England.” British Journal of Political Science 26 (3): 403–418.Find this resource:

Fiorina, M. (1992) 1996. Divided Government. Needham Heights, MA: Allyn & Bacon.Find this resource:

Fisher, S. D. 1999. “Tactical Voting in England 1987 to 1997.” Ph D diss., University of Oxford.Find this resource:

Fisher, S. D. 2004. “Definition and Measurement of Tactical Voting: The Role of Rational Choice.” British Journal of Political Science 34: 152–166.Find this resource:

Fisher, S. D., and J. Curtice. 2006. “Tactical Unwind? Changes in Party Preference Structure and Tactical Voting from 2001 to 2005.” Journal of Elections, Public Opinion and Parties 16 (1): 55–76.Find this resource:

Fisher, S. L. 1973. “The Wasted Vote Thesis: West German Evidence.” Comparative Politics 5 (2): 293–299.Find this resource:

Franklin, M., T. Mackie, and H. Valen. 1992. Electoral Change: Responses to Evolving Social and Attitudinal Structures in Western Countries. Cambridge, UK: Cambridge University Press.Find this resource:

Franklin, M., R. Niemi, and G. Whitten. 1994. “The Two Faces of Tactical Voting.” British Journal of Political Science 24: 549–557.Find this resource:

Franklin, M. N. 2004. Voter Turnout and the Dynamics of Electoral Competition in Established Democracies since 1945. Cambridge, UK: Cambridge University Press.Find this resource:

Fredén, A. 2014. “Threshold Insurance Voting in PR Systems: A Study of Voters’ Strategic Behavior in the 2010 Swedish General Election.” Journal of Elections, Public Opinion and Parties 24: 473–492.Find this resource:

Galbraith, J. W., and N. C. Rae. 1989. “A Test of the Importance of Tactical Voting: Great Britain, 1987.” British Journal of Political Science 19 (1): 126–136.Find this resource:

Gallagher, M. 1991. “Proportionality, Disproportionality and Electoral Systems.” Electoral Studies 10: 33–51.Find this resource:

Gallagher, M. 2005. Conclusion to The Politics of Electoral Systems, edited by M. Gallagher and P. Mitchell, 535–578. Oxford: Oxford University Press.Find this resource:

Gallagher, M. and P. Mitchell. 2005. “Introduction to Electoral Systems.” In The Politics of Electoral Systems, edited by M. Gallagher and P. Mitchell, 3–24. Oxford: Oxford University Press.Find this resource:

García Viñuela, E., J. Artés, and I. Jurado. Forthcoming. “Strategic Voting and Non-voting in Spanish Elections.” Party Politics. http://ppq.sagepub.com/content/early/2013/09/12/1354068813499869.abstractFind this resource:

García Viñuela, E. G., and J. Artés. 2012. “Strategic Voting and Effective Representation Thresholds: Evidence from Three Spanish General Elections.” European Journal of Political Research 51: 289–315.Find this resource:

Gschwend, T. 2001. “Strategic Voting in Mixed Electoral Systems.” PhD diss., State University of New York, Stony Brook.Find this resource:

Gschwend, T. 2007. “Ticket-Splitting and Strategic Voting under Mixed Electoral Rules: Evidence from Germany.” European Journal of Political Research 46: 1–23.Find this resource:

Guinjoan, M., P. Simón, S, Bermúdez, and I. Lago. 2014. “Expectations in Mass Elections: Back to the Future?” Social Science Quarterly 95 (5): 1346–1359.Find this resource:

Heath, A., and G. Evans. 1994. “Tactical Voting: Concepts, Measurement and Findings.” British Journal of Political Science 24 (4): 557–561.Find this resource:

Heath A., R. Jowell, J. Curtice, G. Evans, J. Field, and S. Witherspoon. 1991. Understanding Political Change: The British Voter 1964–1987. Oxford: Pergamon.Find this resource:

Hino, A. 2012. New Parties’ Success in Western Europe: A Comparative Analysis. New York: Routledge.Find this resource:

Hobolt, S. B., and J. A. Karp. 2010. “Voters and Coalition Governments.” Electoral Studies 29: 299–307.Find this resource:

Holmberg, S. 2007. “Partisanship Reconsidered.” In The Oxford Handbook of Political Behavior, edited by R. J. Dalton and H.-D. Klingemann, 557–570. Oxford: Oxford University Press.Find this resource:

Irwin, G. A., and J. M. Van Holsteyn. 2012. “Strategic Electoral Considerations under Proportional Representation.” Electoral Studies 31: 184–191.Find this resource:

Jesse, E. 1988. “Split-voting in the Federal Republic of Germany: An Analysis of the Federal Elections from 1953 to 1987.” Electoral Studies 7: 109–124.Find this resource:

Johnston, R. J., and C. J. Pattie. 1991. “Tactical Voting in Great Britain in 1983 and 1987: An Alternative Approach.” British Journal of Political Science 21: 95–108.Find this resource:

Karp, J. A., J. Vowles, S. A. Banducci, and T. Donovan. 2002. “Strategic Voting, Party Activity, and Candidate Effects: Testing Explanations for Split Voting in New Zealand’s New Mixed System.” Electoral Studies 21: 1–22.Find this resource:

Klingemann, H.-D., and B. Wessels. 2001. “Political Consequences of Germany’s Mixed-Member System: Personalization at the Grass-Roots?” In Mixed-Member Electoral Systems: The Best of Both Worlds?, edited by M. S. Shugart and M. P. Wattenberg, 279–296. Oxford: Oxford University Press.Find this resource:

Kriesi, H. 1998. “Straightforward and Strategic Voting in the Elections for the Swiss Council of States in 1995.” Electoral Studies 17: 45–59.Find this resource:

Laakso, M., and R. Taagepera. 1982. “Proportional Representation and Effective Number of Parties in Finland.” In Power, Voting, and Voting Power, edited by M. Holler, 107–120. Wiirzburg-Wien: Physica-Verlag.Find this resource:

Lago, I. 2008. “Rational Expectations or Heuristics? Strategic Voting in Proportional Representation Systems.” Party Politics 14: 31–49.Find this resource:

Lago, I. 2012. “Strategic Voting in Proportional Representation Systems: Evidence from a Natural Experiment.” Party Politics 18: 653–665.Find this resource:

Lago, I., and F. Martínez i Coma. 2012. “Forgetting to Make Votes Count: The Role of Previous Democratic Experience.” Electoral Studies 31: 413–421.Find this resource:

Lago Peñas, I. 2005. El Voto Estratégico en las Elecciones Generales en España (1977–2000). Madrid: Centro de Investigaciones Sociológicas.Find this resource:

Lanoue, D. J., and S. Bowler. 1992. “The Sources of Tactical Voting in British Parliamentary Elections, 1983–1987.” Political Behavior 14 (2): 141–157.Find this resource:

Lijphart, A. 1994. Electoral Systems and Party Systems: A Study of Twenty-seven Democracies, 1945–1990. Oxford: Oxford University Press.Find this resource:

Lipset, S. M., and S. Rokkan. 1967. Party Systems and Voter Alignments: Cross-National Perspectives. Toronto: The Free Press.Find this resource:

Mair, P. 2007. ‘Left–Right Orientations.” In The Oxford Handbook of Political Behavior, edited by R. J. Dalton and H.-D. Klingemann, 206–222. Oxford: Oxford University Press.Find this resource:

Mair, P. 2013. Ruling the Void: The Hollowing of Western Democracy. London and New York: Verso.Find this resource:

Meffert, M. F., and T. Gschwend. 2010. “Strategic Coalition Voting: Evidence from Austria.” Electoral Studies 29: 339–349.Find this resource:

Neal G. Jesse 2000. “A Sophisticated Voter Model of Preferential Electoral Systems.” In Elections in Australia, Ireland, and Malta under the Single Transferable Vote, edited by S. Bowler and B. Grofman, 59–78. Ann Arbor: University of Michigan Press.Find this resource:

Niemi R. G., G. Whitten, and M. N. Franklin. 1992. “Constituency Characteristics, Individual Characteristics and Tactical Voting in the 1987 British General Election.” British Journal of Political Science 22: 229–240.Find this resource:

Niemi, R., G. Whitten, and M. Franklin. 1993. “People Who Live in Glass Houses: A Response to Evans and Heath’s Critique of Our Note on Tactical Voting.” British Journal of Political Science 23 (4): 549–553.Find this resource:

Oskarson, M. 2005. “Social Structure and Party Choice.” In The European Voter. A Comparative Study of Modern Democracies, edited by J. Thomassen, 84–105. Oxford: Oxford University Press, ECPR.Find this resource:

Pappi, F. U., and P. W. Thurner. 2002. “Electoral Behaviour in a Two-vote System: Incentives for Ticket Splitting in German Bundestag Elections.” European Journal of Political Research 41: 207–232.Find this resource:

Rae, D. W. (1967) 1971. The Political Consequences of Electoral Laws. New Haven, CT, and London: Yale University Press.Find this resource:

Reed, S. R. 1990. “Structure and Behaviour: Extending Duverger’s Law to the Japanese Case.” British Journal of Political Science 20: 335–356.Find this resource:

Riera, P. 2009. “Non Bis in Idem: Voto Escindido en Sistemas Electorales Mixtos. Los Casos de Nueva Zelanda en 1999 y 2002.” Revista Española de Ciencia Política 20: 97–123.Find this resource:

Riera, P. 2015. “Electoral Systems and the Sheriff of Nottingham: Determinants of Disproportionality in New and Established Democracies.” Party Politics 21 (2): 222–233.Find this resource:

Riker, W. H. 1982. “The Two-Party System and Duverger’s Law: An Essay on the History of Political Science.” American Political Science Review 76 (4): 753–766.Find this resource:

Roberts, G. K. 1988. “The ‘Second-Vote’ Campaign Strategy of the West German Free Democratic Party.” European Journal of Political Research 16: 317–337.Find this resource:

Robinson, W. S. 1950. “Ecological Correlations and the Behavior of Individuals.” American Sociological Review 15: 351–357.Find this resource:

Rose, R., and D. W. Urwin. 1970. “Persistence and Change in Western Party Systems since 1945.” Political Studies 18 (3): 287–319.Find this resource:

Rothschild, D., and N. Malhotra. 2014. “Are Public Opinion Polls Self-Fulfilling Prophecies?” Research and Politics 3: 1–10.Find this resource:

Saalfeld, T. 2005. “Germany: Stability and Strategy in a Mixed-Member Proportional System.” In The Politics of Electoral Systems, edited by M. Gallagher and P. Mitchell, 209–230. Oxford: Oxford University Press.Find this resource:

Shepsle, K. A. 1991. Models of Multiparty Electoral Competition. New York: Harwood Academic Publishers.Find this resource:

Shively, W. P. 1970. “The Elusive ‘Psychological Factor’: A Test for the Impact of Electoral Systems on Voters’ Behavior.” Comparative Politics 3 (1): 115–125.Find this resource:

Shugart, M. S. 2000. “Compensatory Seats.” In International Encyclopaedia of Elections, edited by R. Rose, 42–43. Washington, DC: CQ Press.Find this resource:

Spafford, D. 1972. “Electoral Systems and Voters’ Behavior: Comment and a Further Test.” Comparative Politics 5 (1): 129–134.Find this resource:

Taagepera, R., and M. S. Shugart. 1989. Seats and Votes: The Effects and Determinants of Electoral Systems. New Haven, CT: Yale University Press.Find this resource:

Tavits, M., and T. Annus. 2006. “Learning to Make Votes Count: The Role of Democratic Experience.” Electoral Studies 25: 72–90.Find this resource:

Tsebelis, G. 1986. “A General Model of Tactical and Inverse Tactical Voting.” British Journal of Political Science 16: 395–404.Find this resource:

Uggla, F. 2008. “Incompetence, Alienation, or Calculation? Explaining Levels of Invalid Ballots and Extra-parliamentary Votes.” Comparative Political Studies 41 (8): 1141–1164.Find this resource:

Van der Brug, W., M. Franklin, and G. Tóka. 2008. “One Electorate or Many? Differences in Party Preference Formation between New and Established European Democracies.” Electoral Studies 27 (4): 589–600.Find this resource:

Notes:

(1) In this article, parties and candidates are used interchangeably to refer to actors that run in elections.

(2) See Ronald Borwstein, “Liberals Beat Drum for Gore, Hope Nader Backers Listen,” Los Angeles Times, November 1, 2000. http://articles.latimes.com/2000/nov/01/news/mn-45221

(3) In fact, this organized political campaign in favor of tactical voting triggered the first inclusion of questions regarding this behavior in the 1983 British Elections Study (Heath et al. 1991). A significant number of Labour and Liberal Democratic supporters still engaged in this type of behavior in the 1997 election (Evans et al. 1998). However, Fisher and Curtice (2006) document a small decline in Labour to Liberal Democrat tactical voting as a consequence of the collapse in hostility toward the Conservatives in the 2005 election.

(4) In recent decades, Germany has been experiencing a bipolar two‐bloc system, with the Social Democratic Party (SPD) and the Greens to the left of the political center pitted against the CDU/CSU and the FDP to the right of the center (Saalfeld 2005). Parties in Germany must obtain at least 5% of the national vote at the list tier to qualify for PR seats. According to the idea of “threshold insurance,” developed below, supporters of large parties may vote for potentially junior coalition parties at the list tier to ensure that they pass this threshold and are eligible for PR seats.

(5) A second example of promotion of tactical voting in non-majoritarian systems by political elites involves the continuous appeals of the Spanish Socialist Party to the voters of United Left so that they cast the so-called voto útil (useful vote) (Lago 2008).

(6) According to the first of these approaches, structural cleavages like national identity or religious denomination were considered a major determinant of political attitudes and voting behavior. However, voting would be above all an expression of citizens’ social position and the well-established values and interests associated with it. Because of these voting patterns, party systems in stable democracies would have remained immutable since the 1920s.

(7) In British politics, a hung parliament refers to the situation in which the elections fail to deliver a party with an overall majority in the House of Commons.

(8) The term “sophisticated voting” has also been used in this field (e.g., Abramson et al. 1992), though, as Shepsle (1991, 62–63) argues, it is more appropriate for voting on bills and amendments in committee or legislative proceedings.

(9) This is an open empirical question, though preliminary evidence seems to suggest that, for example, tactical voters improve the evaluations of the party they tactically vote for in order to avoid cognitive dissonance (Bølstad et al. 2013).

(10) For example, citizens can vote for a party that is not the preferred one for reasons other than affecting the electoral outcome in order to induce the latter to change its policies or its leader.

(11) Distance from contention is the difference between the percentage vote for the respondent’s most preferred candidate in his or her constituency and the percentage vote for the lower of the two top contenders. The distance from contention measure scores zero if the respondent’s most preferred candidate is in first or second place. The importance of a clear ordering between the second and the third candidates is documented by Galbraith and Rae (1989).

(12) Moreover, using the strength-of-feeling scores presents the obvious advantage of being able to confidently code as tactical those voters for whom we lack information about their “really preferred party” if they have a clear favorite on the strength-of-feeling scores and they do not vote for it (Fisher 2004).

(13) These results could be consistent with the idea of outcome-based dissonance reduction advanced by Beasley and Joslyn (2001): if information about the election result becomes more readily available over time, it would be no surprise if those who voted for a hopeless candidate increasingly change their reported behavior.

(14) Note that two-round majority, single nontransferable vote, and alternative vote systems are cases of majoritarian rules as well. Although these electoral systems create analogous tactical situations, I focus on FPTP contexts, based on the comparatively higher number of works that have done so in the past. For a study of tactical voting in a two-round majority system, see Blais (2004). For a study of tactical voting in a single nontransferable vote system, see Cox (1994). For a study of tactical voting in an alternative vote system, see Jesse (2000).

(15) In the case of PR systems, Lago (2005) finds a minimum of 0.4% for the 1996 Spanish election.

(16) According to Lanoue and Bowler (1992), the role of district marginality has been exaggerated.

(17) Heath et al. (1991, 52) use these two last criteria to define a “tactical situation” as a situation that encourages tactical considerations among some voters (see also Alvarez and Nagler 2000; Alvarez et al. 2006; Blais and Nadeau 1996; Niemi et al. 1992).

(18) In this section I deliberatively leave aside systems with ordinal voting, such as the single-transferable vote. Although it should still be true that voters whom we could define as tactical on the basis of their short-term instrumental motivations exist in those systems, the intricacies of this type of rules make the translation of the definition of tactical voting to them rather complicated. For a study of tactical voting in these preferential systems, see Jesse (2000).

(19) In fact, Rae ([1967] 1971, 86) graphically depicted electoral systems as the Sheriff of Nottingham, “apt to steal from the poor and give to the rich: strong parties usually obtain more than their proportionate share of legislative seats while weak parties receive less than their proportionate share of seats” (see Riera 2015 in this regard).

(20) Likewise, electoral formulas used in most PR and mixed-member systems work to the same effect and benefit, hence, the largest parties. For example, the frequently employed D’Hondt method of seat allocation favors the large parties (Laakso and Taagepera 1982). In Sweden and Norway the modification of the Sainte-Laguë algorithm punishes small parties by using 1.4 (not 1) as the first divisor. This modification makes the first district seat harder to win than the second and subsequent ones (Aardal 2011).

(21) However, Garcia Viñuela and Artés (2012) argue that it is not the district magnitude as such but the “effective threshold” and the distribution of votes between the parties running in the district that matters for tactical voters in a PR country like Spain.

(22) For a different account of slit-ticket voting in New Zealand based on nontactical considerations, see Riera (2009).

(23) According to van der Brug et al. (2008), when two or more parties are about equally attractive to the voters, they tend to vote for the largest of these parties because it has the best chance of achieving its policy goals. This could be considered an alternative form of tactical voting.

(24) More specifically, this method assumes that those who are identified with a party have at least a slight preference for that party. For the remaining ties, those who have a unique first preference for a party leader are considered to have at least a slight preference for her or his party. The final remaining ties are broken randomly.