How protest voters choose

King’s Research Portal DOI: 10.1177/1354068817698857 Document Version Peer reviewed version Link to publication record in King's Research Portal Citation for published version (APA): Birch, S., & Dennison, J. (2017). How Protest Voters Choose. PARTY POLITICS. https://doi.org/10.1177/1354068817698857 Citing this paper Please note that where the full-text provided on King's Research Portal is the Author Accepted Manuscript or Post-Print version this may differ from the final Published version. If citing, it is advised that you check and use the publisher's definitive version for pagination, volume/issue, and date of publication details. And where the final published version is provided on the Research Portal, if citing you are again advised to check the publisher's website for any subsequent corrections. General rights Copyright and moral rights for the publications made accessible in the Research Portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognize and abide by the legal requirements associated with these rights. •Users may download and print one copy of any publication from the Research Portal for the purpose of private study or research. •You may not further distribute the material or use it for any profit-making activity or commercial gain •You may freely distribute the URL identifying the publication in the Research Portal Take down policy If you believe that this document breaches copyright please contact librarypure@kcl.ac.uk providing details, and we will remove access to the work immediately and investigate your claim. Download date: 10. Jun. 2020 How Protest Voters Choose Much ink has been spilt in recent years on political disaffection. Research has grappled with the impact of popular disenchantment with politics on electoral turnout1 and on unconventional forms of political participation.2 Political scientists have also pointed to protest voting – voting for an anti-establishment party as a protest against mainstream politics – as a consequence of dissatisfaction with traditional political options,3 but we know far less about the phenomenon of protest voting than we do about the effects of disaffection on participation. Although there is some existing evidence that disillusionment with mainstream parties can cause people to vote for minor and/or extreme parties in protest, there is a paucity of both theory and empirical evidence on the factors that condition the decision to vote for an anti-establishment party as a form of protest. We also know little about why some voters choose one protest option over the other (a decision we label the ‘protest choice’). The party spectra of most parliamentary democracies include minor parties on both the left and the right that are potential recipients of the votes of those whose decision is motivated primarily by disillusionment with a particular party, with certain policies or, more generally, with the existing political establishment or with political institutions. There is also evidence from studies of radical right-wing and anti-immigrant politics that parties in this portion of the political spectrum tend to draw support from those who are 1 Examples include e.g. Dalton, 2004; Finkel, 1985; Karp and Banducci, 2008; Norris, 2002; 2004; Powell, 1986; Schaefer and Streeck, 2013. 2 E.g. Dalton, 1996; 2004; Norris, 1999; 2011; Tarrow, 2000. 3 See, for example, Bélanger, 2004; Bergh, 2004; Brusis, 2016; Hetherington, 1999; Kang, 2004; Kselman and Niou, 2011; Pattyn et al., 2012; Pop-Eleches, 2010; van der Burg, Fennema and Tillie, 2000; Webb, 2005. 1 broadly discontented with politics.4 The same has been found of a number of radical parties of the left.5 Yet we know relatively little about why disaffected voters select one anti-establishment party over another as a means of voicing their discontent with politics as usual. In times of political turbulence, protest voting can surge from being a minor political novelty to a phenomenon of considerable import, as when the 2016 presidential election in Austria yielded a run-off between two non-establishment candidates, when Syriza formed a government following the 2015 Greek election, or when the Pirate party claimed one in seven the votes an the Icelandic election of 2016. In this context, it is all the more urgent to have a clearer understanding of the logic of protest voting in general, and why electors opt for one protest choice rather than another; this in turn will help unravel the puzzle of why in some countries protest voting has been in greater evidence than in others. Taking as its empirical referent the 2015 General Election in the UK, this paper assesses both the protest vote and protest choice. Though the first-past-the-post electoral system in the UK differs from that of many European democracies, the 2015 British election is in many other respects an ideal context in which to study protest voting and protest choice. The vote of the traditional third party – the centrist Liberal Democrats – collapsed following the party’s stint in government as coalition partner to the Conservatives. The Liberal Democrats had long been a haven for protest votes (Bélanger, 2004; Kang, 2004; Pattie and Johnston, 2001), but with a vote 4 E.g. Bélanger and Aarts, 2006; Bergh, 2004; Betz, 1994; Fennema, 1997; Lubbers, Gijsberts and Scheepers, 2002; Norris, 2005; Oesch, 2008. 5 See, for example, e.g. Bélanger and Nadeau, 2005; Bellucci, 2014; Bergh, 2004; Bomberg, 1998; Karyotis and Rudig, 2015; O’Neill, 1997; Poguntke, 1993; Ramiro and Gomez, 2016; Schumacher and Rooduijn, 2013. 2 for this party no longer an act of protest, disaffected voters fled elsewhere, principally, in England, at least – to anti-establishment parties of the right and left – the United Kingdom Independence Party (UKIP) and the Greens. This situation thus provides an opportunity to examine the predictors both of protest voting and also of the protest choice. In modeling protest voting we find support for the role of issue positions, trust and campaign effects. We then consider the factors that determine how protest voters choose between competing protest choices – in this case UKIP and the Greens. Though previous research on protest voting emphasizes the role of trust in politics and other measures of political disaffection, we find that the most powerful predictors in our models are policy positions on issues ‘owned’ by these parties, as well as campaign effects. From these findings we argue that protest voting is fundamentally a party political act, not simply an expression of discontent. The paper unfolds as follows: the next section defines more precisely what we mean by protest voting and sets out the theoretical justification for our argument; the following section reviews salient features of the UK context; the third section describes our choice of data and method, including how we identify protest voters, and gives a brief overview of the descriptive evidence. Multivariate analysis follows in the fourth section. A final section concludes. 1. Protest Voting and Protest Choice The period since 2008 has been a time of political turbulence across the democratic world. As in the 1970s, economic shock has led to party system instability, and previously marginal political forces have gained strength by questioning the ability of established political parties to deal effectively with the fall-out of economic strain (Tillman, 2016). The appeal of such minor parties has been especially pronounced among those who have borne the brunt of the economic 3 downturn (Bartels, 2014; March, 2012; Mayer, 2014), and the most successful such political organizations have frequently taken populist stances in the hopes of mobilizing the hard-pressed. Modern anti-establishment parties fall into two distinct categories: right-wing and left-wing, both of which have historically benefitted from poor economic conditions. This phenomenon has been in evidence across Europe in recent years. On the right, the French Front National and the True Finn party in Finland have enjoyed notable rises in their vote shares. On the left, the Left-Green Movement in Iceland, the Italian Five-Star Movement, Podemos in Spain and Greece’s Syriza have all experienced significant increases in political support. We conceptualize protest voting in terms of the vote, not the party. Van der Brug, Fennema and Tillie caution against the circularity is identifying certain parties as ‘protest parties’ on the grounds that they are believed to attract ‘protest votes’ (van der Brug, Fennema and Tillie, 2000: 82). We therefore abstain from characterizing parties in terms of protest and focus instead on the act of protest voting. Several scholars have offered definitions of protest voters or protest voting which it is useful to review. Bowler and Lanoue define protest voters as ‘those citizens who express general dissatisfaction with the performance of the incumbent government, or are more specifically convinced that the government's policies have had an adverse impact on their own lives’ (Bowler and Lanoue, 1992: 491). Van der Brug, Fennema and Tillie identify a protest voter as ‘a rational voter whose objective is to demonstrate rejection of all other parties’ (van der Brug, Fennema and Tillie, 2000: 82). Bergh describes protest voting as ‘the act of voting for a political party or candidate in order to express political distrust’ (Bergh, 2004: 376). Pop-Eleches defines protest voting as ‘the practice of voting for a party not because of the actual content of its electoral message but in order to “punish” other parties’ (Pop-Eleches, 2010: 223). Kselman and Niou 4 frame protest voting as ‘choosing a party other than one’s most-preferred to send that most preferred party a signal of dissatisfaction’ (Kselman and Niou, 2011: 400; italics removed). Common to all these definitions is an understanding of protest voting as an expression of dissatisfaction with mainstream parties and/or politicians.6 There are three points to note about this understanding: firstly, it conceives of protest voting as expressive voting (Brennan and Lomasky, 1993), voting whose aim is to send a signal by voting for a minor party (Kang, 2004; Kselman and Niou, 2011). The second point to note is that the object of dissatisfaction is mainstream parties, typically including parties in government and those that have been in government in the past. It is therefore a vote against a political ‘elite’ and potentially (though this remains to be demonstrated) an expression of dissatisfaction with the political elite’s consensus on certain issues and/or with the perceived manners, lifestyles and attitudes of incumbent politicians as a group. The third point is that the political ‘elite’ must be understood in relation to the electoral system in a given country (Kang, 2004). Studies which have maintained that voting for far right parties is not a protest vote have largely been carried out in European states with proportional representation electoral systems (van der Brug et al, 2000; van der Brug, 2003). Under such circumstances, a vote for a minor party very often helps to win that party seats in parliament, and in proportional representation systems even small parties can play a role as coalition partners, thereby having an opportunity to implement their policies. In electoral terms, 6 It is in theory possible that protest voting could be understood as a protest against the regime, but there appear to be few empirical examples where this has clearly been found to be the case. In a comparative study of protest voting in Austria, Denmark and Norway, Bergh found that protest voting mainly reflected a reaction against political elites, rather than against the political system itself (Bergh, 2004: 386). 5 many minor parties are parties like any other. Protest voting must therefore be understood as a reaction against current power-holders; a party that is already strongly represented in parliament cannot, by definition, be considered a political outsider, and is therefore not an attractive object for protest voting. In a study of Canadian elections, Bowler and Lanoue (1992) found protest voting to be less common in seats where the party for which the vote is cast has a realistic chance of winning. With these considerations in mind, protest voting will be defined for the purpose of this analysis as a change of vote, motivated by political dissatisfaction, away from a mainstream party and for a party that consciously adopts the political stance of an ‘outsider’. Protest voting can thus be an expression of disaffection with particular parties and their policies, or it with politicians as a class and political institutions more generally.7 We recognize that this definition is relatively expansive, encompassing both ‘positive’ voting for the policies favoured by antiestablishment parties that are highly unlikely to win power, and ‘negative’ voting against parties of the establishment, but we feel that as a definition it captures the core elements of previous definitions of protest voting, and also that it accords with common-sense understandings of this term as anti-elite expressive voting designed to ‘send a message’ rather than contribute to government formation. We are motivated to understand both what drives protest voting in general, and what explains how voters decide which protest vehicle should be the recipient of their vote. Analytically, protest voting could be conceptualized as a two-step process. It could be that voters 7 In one of the few studies to disentangle the various objects of discontent, Norris finds support for far right parties to be associated with distrust of political institutions as well as of incumbent parties (Norris, 2005: 157). 6 first become detached from their habitual party of identification, or, for nonidentifiers, they move away from the last party for which they voted. In a second step, voters then make the decision to vote for a particular protest option.8 Several studies employing panel data confirm this temporal ordering of protest voting: voters first become disillusioned with mainstream parties and politicians, and then opt to vote for a party among those that, they feel, will express their dissatisfaction, typically a small and/or new party (Bélanger and Aarts, 2006; Karyotis and Rudig, 2015). It is also possible, however, that in some contexts the order of these two steps is reversed, in the sense that voters are attracted to an ‘outsider’ party for which they have not previously voted, and then at some point they make the decision actually to vote for that party, leaving their habitual choice behind. We remain agnostic as to the order in which these steps occur, and our research design does not equip us to assess this question directly. We are interested rather in modeling the decision to cast a protest vote and the protest choice as two phenomena of interest, which we consider in tandem. We posit three principal explanations for protest voting and protest choice: (1) ideological affinity; (2) a lack of trust in mainstream parties and politicians in general; and (3) political communication factors specific to the campaign context, such as the ability of the party leader (and other party spokespeople) to communicate effectively in the national media, and campaigning on the ground at local level.9 8 For some voters, there may be an intermediary step where they decide whether or not to vote at all. 9 There is also a residual category of other explanations, such as random choice, etc. These motivations will not be considered here. 7 (1) Ideological protest voting: Single-member district electoral systems such as that used in the UK have the well-known effect of constraining the ideological space and driving political parties to the centre of the policy spectrum. This grouping effect might well vary over time, which will result in variations in the number of voters who find themselves located at points on this spectrum that are not occupied by mainstream parties. Many such voters can be expected to opt for mainstream parties nevertheless, but variations in the intensity with which voters hold ideological positions might result in changes in their willingness to opt for mainstream parties that do not fully reflect their views; under these circumstances they will no longer be prepared to remain loyal to their habitual party, and they will instead protest against that party’s position on policies that are important to them and vote for an alternative. It is thus possible that voting for anti-establishment parties is the result of ideological placement (Schumacher and Rooduijn, 2013). If a party’s policy shifts in one direction and/or the electorate shifts in the opposite direction, a portion of the party’s electorate may abandon their traditional allegiance in favor of the next most proximate option. The party they select depends on the ideology they find most appealing (even if that ideology does not appear to accord with their personal economic interests, as is often the case among the supporters of right-wing populist parties (Napier and Jost, 2008)). If there is another party whose views are acceptable to the voter, such a move may satisfy standard criteria of ‘rationality’ in the short term (Kang, 2004). Reaction against mainstream parties that is a consequence of the post-2008 economic downturn and the policy solutions proposed by all such parties (in the 2015 UK context these included austerity and continued European Union membership) might explain protest voting. Van der Brug, Fennema and Tillie reject the idea that protest voting might be ideological, arguing instead that ‘compared to ideological and pragmatic voters, protest voters cast their vote 8 not to affect public policies, but rather to express disenchantment with the political system or with the political elite’ (van er Brug, Fennema and Tillie, 2000: 83; cf van der Brug and Fennema 2005; van der Brug, Fennema and Tillie, 2003; Fennema, 1997). But it is possible that disenchantment with the political system derives precisely from the fact that the political parties that dominate a system have collectively failed to consider seriously certain policy options. The recent surge in support for anti-austerity parties such as Syriza in Greece and Podemos in Spain may be seen as a reaction against established party systems where the mainstream parties all accept the necessity of austerity policies. Likewise, support for far-right parties that propose the expulsion of immigrants and other radical anti-immigrant policies can be seen as a rejection of mainstream parties that have been unwilling to take such policy options seriously. It is in this sense that we argue, contra van der Brug, Fennema and Tillie, that protest voting can have its origins in ideological and policy preferences. Further support for this position comes from psychological research which finds a link between political cynicism and authoritarian attitudes (including intolerance and racism) (Pattyn et al. 2012; cf Grabb, 1979; Napier and Jost, 2008). This research suggests that personality traits such as authoritarianism are part of a nexus that includes a propensity toward disaffection and protest. This might explain why many protest voters choose small parties of the right rather than those of the left. An additional piece of evidence for the protest-ideology link comes from the role of emotion in shaping reasoning. Anger is an emotion that tends to lead people to make sweeping generalizations and to engage in stereotyping (Lerner, Goldberg and Tetlock, 1998; Goldberg, Lerner and Tetlock, 1999). It is not unexpected, therefore, that so many of those who see themselves losing out in the current economic climate might lash out both at politicians and at 9 the same time blame ethnic ‘others’. This is a further reason to expect there might be a connection between protest voting and voting for anti-immigrant parties. Moreover, small parties are often perceived to be single-issue parties, and they may well be symbolically identified in the minds of voters with that issue alone, even if they have a comprehensive range of policies (Meguid, 2005). Van der Brug, Fennema and Tillie (2000) find that the supporters of right-wing anti-immigrant parties tend focus on the issue of immigration to the (relative) exclusion of socio-economic and other issues. The same may be true of other protest voters also – they may respond to especially salient issues with which they identify antiestablishment parties. Another interpretation of ideological protest voting is that voters protest against specific shifts in their habitual party’s ideological position. For example, when a party lurches to the right, a portion of its habitual voters may vote for a party further to the left out of dissatisfaction with this development. Changes in the ideological positions of individual parties might thus explain trends in protest voting at a given election. In the medium term, mass defection from a party at one election many put pressure on that party to change course prior to the next election so as to retrieve errant supporters; protest voting of this sort may thus be seen as a ‘rational’ signaling device that can shape the direction of one’s habitual party (Kselman and Niou, 2011). (2) Trust-based protest voting: A second possible explanation for protest voting, and that favored by the existing literature, is that voters cast voted in protest as an expression of lack of trust in mainstream political elites, and in particular, those in power. Trust has consistently been identified in comparative studies of protest voting as one of the main drivers of this phenomenon (Bélanger and Nadeau, 2005; Bergh, 2004; Hetherington, 1999). The 2015 election in the UK 10 was the first for a generation where two mainstream parties were in power; it was also a time of declining trust levels in the wake of the 2009 MP expenses scandal, which had further exacerbated already low levels of trust among the UK electorate (Allen and Birch, 2015). In this sense one might expect there to be relatively high levels of protest voting at this election, motivated by frustration with the ‘political class’ and behaviors that were widely perceived to be self-serving. It must be noted that a priori the trust explanation provides little help in accounting for which political option protest voters choose. (3) Campaign-based protest voting: A final possibility is that the decision to use one’s vote to voice protest is driven by factors specific to the electoral campaign in question, and that it is therefore a function of political communication. Political communication during an election campaign takes two main forms, mass media communication by the leader and other party representatives, and grassroots campaign by party activists. Both types of communication can be expected to play some role in enticing voters to defect from their habitual party of choice and to vote for an outsider, especially if leaders and campaigners of ‘outsider’ parties are especially attractive and/or effective in getting across their messages. Under the single-member plurality electoral system employed in the UK, voters may opt for the anti-establishment party which has greatest profile nationally or in their local constituency, as the familiarity of a minor party name has been shown to be linked to the persuasiveness of that party (Coan et al, 2008). Similarly, the popularity of the party leader may play a role in shaping anti-establishment party choice. Many successful anti-establishment parties have been led by charismatic figures such as Pim Fortuyn in the Netherlands and Jean-Marie Le Pen in France, and this has been one of the factors that has 11 been found to have affected their success (Lubbers, Gijsberts and Scheepers, 2002; Schumacher and Rooduijn, 2013). Also relevant during an electoral campaign are factors specific to the constituency in question, including marginality (as measured at the previous general election) and the extent to which a voter would regret ‘wasting’ their vote on a small party that had scant chance of success in a single-member district. To some extent, perceptions of marginality and wasted votes are shaped by political communication, but in as much as the partisan composition of most constituencies is relatively constant over extended periods of time, voters can also be expected to rely on historical knowledge in assessing the advisability of casting a protest vote in any given local context. According to the campaign-based argument, the decision to protest is thus the result of short-term conjunctural factors rather than long-term pre-dispositions, as it is the dynamics of the campaign itself that shapes the voter’s calculus. We might expect such voters to be less likely to cast protest votes at successive elections, though that is a topic for another paper. These differing explanations will be assessed in the empirical analysis below, but first it is necessary to review key features of the British case. 2. The UK Context Britain has traditionally had a relatively stable ‘two-and-a-half party system’, composed of the Conservatives on the centre-right, Labour on the centre-left, and the smaller Liberal Democrats in a centrist position. Between 2010 and 2015 this changed, as two previously small and ideologically more extreme parties witnessed dramatic increases in popular support: the United Kingdom Independence Party (UKIP) to the right of the Conservatives, and the Greens to the left of Labour. These shifts in vote choice, together with a substantial increase in support for the 12 regionally-confined Scottish National Party, meant that parties other than the three traditional anchors of the British political system increased their collective vote share from 3.54 per cent in 2010 to 24.8 per cent in 2015 (see Figure 1). As the data in Figure 1 demonstrate, this was a development unprecedented in post-war UK electoral history. Figure 1 about here This dramatic reconfiguration of the party system can be explained with reference to several developments during the 2010-2015 Parliament. Firstly, many members of the electorate had suffered economically between 2010 and 2015, as the UK experienced its longest period of economic stagnation since the 1930s; by 2015 real wages were nearly 10 per cent below their pre-crisis levels (Machin, 2015). Secondly, trust in politicians was at record low levels, following the 2009 MPs expenses scandal and the continuing repercussions of allegations of impropriety by elected representatives (Ipsos Mori, 2015), and those most critical of politicians’ behavior were already in 2010 more likely to vote for minor parties (Allen and Birch, 2015). Finally, the traditional ‘third’ party in British politics, the Liberal Democrats, saw their popular support plummet, as shown in Figure 2, following an undistinguished term as side-kick to the Conservatives in the 2010-2015 coalition government. Figure 2 about here The combined vote share of Labour and the Conservatives hardly changed between 2010 and 2015 (rising from 65.0 per cent in 2010 to 67.2 per cent in 2015)10 whereas UKIP and the Greens each saw their vote share quadruple in tandem with a decline of 15 percentage points in the Liberal Democrat vote. The UKIP vote share rose from 3.1 per cent in 2010 to 12.6 per cent 10 The Conservative vote share increased by .8 per cent from 36.0 to 36.8 per cent; the Labour vote share increased by 1.4 per cent from 29.0 to 30.4 per cent. 13 in 2015 (an increase of 9.5 percentage points), while the Green share of the vote went up from 1.0 per cent to 3.8 per cent (an increase of 2.8 percentage points). The combined vote share of these parties thus shot up from 4.1 per cent in 2010 to 12.3 per cent in 2015 as one in twelve of the voting public shifted their vote in these parties’ direction.11 The specific questions we address in this paper are why so many voters deserted mainstream parties, and why the right-wing populist offer of UKIP was so much more appealing than the Greens’ left-wing alternative. Citizens disenchanted with mainstream parties have two principal options: they can stay at home or they can vote for a minor party – ‘exit’ and ‘voice’ in Hirschman’s (1970) terms. At a close election, such as the 2015 British General Election, there is considerable incentive for voters to go to the polls and, indeed, turnout increased by about one percentage point over its 2010 level suggesting that even many disaffected electors still felt the desire to make their voice heard. Such voters must then decide which of the various minor parties to opt for. We focus for the purposes of this paper on England for three reasons. First, doing so brackets the rather different dynamics associated with the nationalist alternatives available in Scotland and Wales or the very different party system of Northern Ireland. Second, and related, the English party system in 2015 presented a clear distinction between, on the one hand, two parties of government (the Conservatives and the Liberal Democrats) and a clear opposition party hoping to take over as the government (Labour) and, on the other hand, two parties that had no history of government and, outside of a handful of constituencies (around one per cent) had no hope of winning seats. By contrast, both the Welsh and Scottish nationalists were capable of winning multiple seats (with the latter ultimately winning all but five of Scotland’s 59 seats). Third, both 11 The election result figures quoted here are taken from the House of Commons Library reports published following each election (Cracknell, 2010; Hawkins, Keen and Nakatudde, 2015). 14 the Welsh and Scottish nationalists had been in their respective devolved governments since the previous General Election in 2010 – creating the possibility of protest voting against these parties. The generalizability of the findings of this paper are therefore enhanced by testing only English seats in order to provide the clearest distinction between parties that were likely and highly unlikely to receiving votes that were motivated by the desire to express protest. Preelection polls had charted the steady rise of UKIP and the Greens, and on polling day, these were in most constituencies the most viable recipients of protest votes. Both parties had for the first time been allowed to take part in some of the televised pre-election debates and they both received considerably more media coverage during the 2015 election campaign than during previous campaigns. In this sense both UKIP and the Greens were potential options at the ballot box for many voters.12 Before moving on to empirical analysis, let us pause to review key features of each of these political organizations. UKIP: The United Kingdom Independence Party was founded in 1993 and has received an increasing vote share in European Parliament elections on the basis of its opposition to British membership of the European Union. For years the party’s support in Westminster elections was negligible, though it had inched up gradually from .03 per cent in 1997 to 3.1 per cent in 2010. Following the 2010 General Election, UKIP began to enjoy a precipitous rise in the polls, and it won the largest share of the vote – 27.5 per cent – at the European Parliament elections of 2014, two-thirds more than the 16.6 per cent it had gained in 2009. 12 It may be that what we observed at the 2015 British General Election is analogous to the shift that took place in Canadian politics in 1993 when two new parties – Reform and the Bloc Québécois – took over from the New Democratic Party as the preferred choice for protest voters (Bélanger and Nadeau, 2005). 15 The available survey data suggests that UKIP was particularly adroit at capturing the electoral favor of those who had suffered most from globalization and the economic downturn – ‘left behind’ voters who were low on skills and high in years (Ford and Goodwin, 2014). This can be explained by the fact that the party’s appeal is based largely on its opposition to immigration, which many poorly-educated Britons view as the cause of the economic problems they experience. UKIP was not always an anti-immigrant party, having started with an almost exclusive focus on the economic impact of European Union membership on small businesspeople, but in the run-up to the 2010 General Election it undertook a classic heresthetic move, reorienting its appeal to capture illiberal anti-immigrant sentiment which the main parties were not able or willing to pander to directly (Ford and Goodwin, 2014). In 2014 Ford and Goodwin summed up UKIP’s appeal by noting that ‘the two highest profile issues on the political agenda in the past five years have been immigration and the financial crisis, and on both issues UKIP have done very well among Eurosceptic voters who think the mainstream parties have failed to deliver’ (Ford and Goodwin, 2014: 202). UKIP has also positioned itself clearly as an anti-establishment party, making overt appeals to protest voters with slogans such as ‘Sod the Lot’ (a phrase that appeared on one of its 2010 campaign posters against images of mainstream politicians). The party has consciously positioned itself as the voice of the ‘average’ person speaking out against the corrupt political elite (Abedi and Lundberg, 2009). The successful mobilization of discontent with the metropolitan political elite has been one of the main keys to its recent success (Ford and Goodwin, 2014). Addressing his party in 2010, leader Nigel Farage expressed the sentiments of UKIP supporters he had met: 16 One thing many have in common, they are fed up to the back teeth with the cardboard cut-out careerists in Westminster. The spot-the-difference politicians. Desperate to fight the middle ground, but can’t even find it. Focus groupies. The triangulators. The dog whistlers. The politicians who daren’t say what they really mean. (cit. Ford and Goodwin, 2014: 4). This was a view that resonated strongly with large portions of the British electorate who were disillusioned with the political mainstream. But UKIP did not have a monopoly on antiestablishment views, for the Greens articulated similar sentiments, if in a different lingo. The Greens: Formed in 1985 as a successor to the short-lived Ecology Party, the Green Party of England and Wales had enjoyed periodic support in European Parliament elections but polled under one per cent in General Elections prior to 2005. In 2010, the party won its first seat in Parliament when then leader and MEP Caroline Lucas was elected to represent Brighton Pavilion, though the party’s overall vote share still hovered around the one per cent mark. By late 2014 the party was in a different place entirely, with opinion polls showing the party levelpegging with the Liberal Democrats at between 5 and 8 per cent, with a handful of polls giving the party double-digit support. By 2015 the Green Party of England and Wales had also seen its membership rise steadily over the previous five years, but membership growth accelerated considerably in the autumn of 2014. Increase in support for the Greens was given a further boost when the BBC announced that it would not be including the party in the televized debates to be held in advance of the election and outraged citizens joined the party in sympathy (Dennison, 2015). By April 2015, the party’s membership had shot up to 60,000, having more than trebled in the past year alone (Green Party of England and Wales, 2015). 17 The support base of the Greens has traditionally been the mirror image of that of UKIP. Whereas the latter is most popular among poorly-educated men over the age of 55 (Ford and Goodwin, 2014; 153), the Greens gain their greatest support among educated young people (Birch, 2009; Dennison, 2015). The party does best in university towns and in large cities, drawing on a reservoir of left-liberal sentiment that has traditionally also made many people vote for the Liberal Democrats (Birch, 2009; Carter and Rootes, 2006). Not surprisingly, the Greens explicitly targeted the Liberal Democrat vote in 2015 (Dennison, 2015). There are a number of factors that can potentially explain the ‘Green surge’ of 2015. First and most obviously, the decline in popularity of the Liberal Democrats released a sizeable chunk of the electorate who were favorably disposed to Green stances on many issues. Secondly, climate change was back on the global agenda with feverish preparations afoot for the Paris summit of December 2015 to agree a follow-on to the Kyoto Protocol. Thirdly, the Greens had higher political profile than at any point in the past, having placed fourth in the 2014 European Parliament elections, winning an additional seat and thereafter being embraced by the media as a ‘UKIP of the Left’ (Dennison, 2015). Caroline Lucas was also proving a vocal and popular MP who had garnered considerable media attention during her five years in Parliament. Fourthly, the Greens sought in 2015 to move out of the single-issue box into which many people placed them by proposing a range of popular policies, such as renationalization of the railways, increasing the minimum wage, removing the profit motive from the National Health Service as well as opposition to fracking, HS2 (High Speed Rail 2) and university tuition fees. Indeed, whereas in February 2014, a plurality of voters placed the Greens in a centrist position, at 5 on a 0-10 leftright spectrum, by May 2015 the most common placement was 0, indicating the electorate had become far more aware of the Greens’ leftist credentials (Dennison, 2016). They had also begun 18 to carve themselves a niche as the only party with a clearly non-conciliatory approach to UKIP’s anti-immigration rhetoric and an outright rejection of austerity as an economic policy. Finally, the Greens have long appealed to protest voters by lumping mainstream parties together as ‘grey parties’, and in 2015 this message appears to have been more resonant than at previous times. Writing in early 2015, Jennings and Stoker noted the common feelings of dissatisfaction with mainstream politics among the supporters of UKIP and the Greens: ‘The odds of someone intending to vote Green or Ukip are up to two and a half times higher […] if they express distrust in politicians. People who intend to vote for UKIP and the Greens are also more dissatisfied with British democracy, dislike both David Cameron and Ed Miliband, and more likely to agree that “politicians don’t care what people like me think”’ (Jennings and Stoker, 2014). Between them, UKIP and the Greens won over four million more votes than they had in 2010. Some of these voters had previously abstained, and a handful had voted for other minor parties at previous elections, but as we shall see, most of these voters were former adherents of one of Britain’s traditional main parties: Labour, the Conservatives or the Liberal Democrats. Approximately three times as many of these voters opted for UKIP than opted for the Greens. Our aim in the sections to follow is to try to understand this choice and to assess the implications of the British evidence for protest voting in parliamentary democracies more generally. 3. Data and methods In this section we introduce the data employed to test the hypotheses articulated above, and we discuss the operationalization of protest voting we employ. One of the major challenges in studying protest voting is that identifying such vote(r)s is not easy as it might seem. However, 19 there are some clues with which we can begin to identify these voters. For this purpose, we draw on the British Election Study Internet Panel, an internet survey conducted at multiple points over the 2014-2017 period. The total sample size is approximately 30,000 in each wave, enabling the thorough analysis of relatively unusual voter choices. For this study we draw primarily on Wave 6 of the survey, fielded in May 2015, following the General Election of 7 May. In this survey, voters are asked why they voted as they did in the 2015 UK General Election. The results, as indicated in Table 1, show that protest voting is not provided to respondents as a set answer, with habitual, best-party, best-candidate and tactical reasons instead given as potential responses. Table 1 about here The 8.0 per cent who responded with ‘other’ as their primary motivation for electoral choice – totaling 2,662 replies in total across the UK – were then asked to enter their own reason for their choice. Of these, 118 (4.4 per cent of ‘other’ respondents and just 0.42 per cent of total number of respondents) used the word ‘protest’ in their given reason. In the distinct, and generally simpler, party systems in English constituencies, these ‘self-described’ protest voters overwhelmingly voted for UKIP, with sizeable minorities voting for the Greens or ‘other’ parties and a handful also voted for Labour and one voted for the Liberal Democrats, as shown in Table 2. Table 2 about here Of those who said they were primarily motivated by a desire to express protest, 83.5 per cent opted for either UKIP or the Greens and 95.2 per cent voted for one of these parties or an ‘other’ 20 party.13 Of those giving an alternative reason for voting, after responding ‘other’, a number meet our definition of protest voting but did not use the word ‘protest’. We include a list of 140 of these in the Appendix, Table A3. The first ten alphabetically, by way of example, are: 1. “All other Parties lie. break the mould! [sic]” 2. “angry with lib dems for coalition with the tories [sic]” 3. “Basically I'm a Conservative who wants a stronger policy against immigration into the UK” 4. “because I didn't want eityer of the 2 main parties [sic]” 5. “because no party can do right by this country no matter their promises” 6. “Because the main parties are not listening to ordinary working class people on Immigration and it's effects on our lives and this country [sic]” 7. “because they are all as bad as on and other [sic]” 8. “Because they were not the others” 9. “Cameron forced through gay marriage not even on his manefestotout. a [sic]” 10. “Cameron refused to debate our future in the EU” In these examples alone, we can see evidence of voters exercising ‘voice’ by protesting against the mainstream parties in toto, against particular parties and their policies and against the entire political class. All also include as motivations a desire to express negative feelings 13 The ‘other’ choices of these self-described protest voters are listed in Table A1 in the Appendix. Overall, these ‘other’ responses can be divided between (1) spoilt ballots, (2) independent candidates, (3) minor parties and (4) refusal to say. This distribution is reflected in the overall distribution of ‘other’ voters at the election, as shown in A2. 21 about those for whom the respondents did not vote rather than those for whom the respondent did vote. All of these match our definition of protest voters; this suggests that many of those who did not use the term ‘protest’ were in fact protest voting. If many more meet our definition of protest voters than provided protest as a motive, we need a different strategy for estimating who the protest voters are. Our definition implies that protest voting is an expressive, rather than purposive, act that seeks to send a message rather than affect who wins the election. That message is to voice dissatisfaction with a party and its policies, politicians as a class or the political system as a whole. The above definition can be operationalized in the following way using the British Election Study data (all data are weighted14 and are drawn from England only): 1. Spoiling one’s ballot. 1.03 per cent of the population. 2. Wishing to send a message of dissatisfaction with one’s habitual party. Operationalized as both identifying most closely with a mainstream party at the time of the election and voting for non-mainstream party. A total of 6.6 per cent of the entire sample fitted this description. Of these voters who voted against their mainstream party identification, 68.2 per cent voted for UKIP, 25.2 per cent voted for the Greens, 1.0 per cent spoiled their ballots, 2.7 per cent voted for Independent candidates and 2.9 per cent voted for other minor parties. Forty per cent of UKIP voters identified with one of the three mainstream parties (24.0 per cent identified with the Conservatives, 13.0 per cent with Labour and 3.0 per cent with the Liberal Democrats). Thirty-six per cent of Green voters identified with 14 Weights used are the ‘core wave 6’ weights as recommended by the British Election Study team (http://www.britishelectionstudy.com/) 22 one of the three mainstream parties (4.0 per cent with the Conservatives, 18.0 per cent with Labour and 13.0 per cent with the Liberal Democrats). Of those voters who selfdescribed as protest voters, 70.1 per cent both identified with a mainstream party and voted for a non-mainstream party. 3. Sending a message of dissatisfaction about politicians as a class. We operationalize this element of ‘voice’ as both expressing ‘no trust’ (1 on a 7 point scale) in MPs in general and voting for an outsider party. A total of 29.6 per cent of UKIP voters expressed no trust in MPs in general while 19.9 per cent of Green voters did so. Twenty-four per cent of self-described protest voters expressed no trust in MPs. 4. Sending a message of dissatisfaction about political institutions. We operationalize this as both responding ‘very dissatisfied’ (1 on a 4 point scale) in British democracy and voting for an outsider party. 25.8 per cent of UKIP voters and 36.4 per cent of Green voters were ‘very dissatisfied’ with UK democracy. Twenty-six per cent of self-described protest voters were ‘very dissatisfied with UK democracy’. If a voter matches any of these definitions, we consider them a ‘protest voter’, as shown schematically in Table 3. Overall, 11.6 per cent of the England-only sample was classed as having cast a protest vote by this definition at the 2015 UK General Election. Of protest voters, 66.6 per cent voted for UKIP, 27.4 per cent voted for the Greens, 1.3 per cent spoiled their ballot, 1.8 per cent voted for independent candidates and 3.0 per cent voted for minor parties. Conversely, 68.3 per cent of UKIP voters were protest voters, 68.2 per cent of Green voters were protest voters, 100 per cent of spoiled ballots (by definition) were protest voters, 86.2 per cent of independent voters were protesting and 81.5 per cent of minor party voters were protesting 23 voting. Our measure of protest voting also captures the vast majority (87.1 per cent) of selfdescribed protest voters. Table 3 about here 4. Empirics We start by analyzing the determinants of protest voting in general. Following the theoretical considerations set out above, we offer four logit models to explain protest voting – models based on ideology, trust, and campaign effects, as well as a combined model including variables significant in the other three. Details of operationalization of the independent variables in these models can be found in Table A4 of the Appendix. The regression results are presented in Table 4. Table 4 about here As can be seen from these models, there is some support for all three hypotheses. Voter ideology, trust in politicians and campaign effects all predict propensity to cast a protest vote. At the same time, only certain variations of each of these hypotheses have significant impacts on vote choice. The most general indicator of ideological position – left-right self-placement – is not significant; rather it is voter views on specific issues ‘owned’ by anti-establishment parties that appear to drive protest voting: attitudes toward immigration and the European Union are especially strongly connected to protest voting. When it comes to trust, all three variables prove significant, though to varying degrees: trust in MPs and satisfaction with democracy are strongly (and negatively) linked to protest voting, while external efficacy is only weakly significant, and with an unexpected sign. Perhaps one of the most interesting findings is the importance of campaign effects on protest voting, as this is not a relationship that has been identified in previous literature on the topic. Party contact, constituency marginality and belief that a vote for 24 a small party will be wasted are all linked to protest voting, though party contact loses significance in the combined model. It seems that protest voting can be induced by context and political communication. It is also worth noting that in three of the four models there is a modest but statistically significant relationship between older age and protest voting. The impacts of the significant variables in the combined model are presented graphically in Figure 3. Figure 3 about here The next stage in our analysis is to assess how protest voters choose parties – what we term ‘protest choice’. The first step in this process is to ascertain descriptively which antiestablishment parties the adherents of different mainstream parties vote for. The data in Tables 5 and 6 demonstrate that Conservative-identifying protest voters generally opt for UKIP, whereas Labour- and Liberal Democrat-identifying protest voters divide between UKIP and the Greens. This finding accords with common perceptions of patterns of vote flows, though it is worth noting that protest votes do not necessarily travel to the ideologically most contiguous party. Supporters of the centrist Liberal Democrat party are more likely to plump for the Greens, in keeping with the pro-environmental, pro-EU values shared by these parties; whereas Labour identifiers are almost twice as likely to cast a protest vote for far-right UKIP as for the more ideologically proximate Green party (cf Ford and Goodwin, 2014). Tables 5 and 6 about here We next model the protest choice. Again, for the purposes of this analysis we only include voters in English constituencies that had both Green and UKIP candidates at the 2015 election, and here we only include protest voters who voted for UKIP or the Greens.15 Again, all 15 We include voters from the few constituencies in which the Greens and UKIP were electorally competitive. Though we recognize that our operationalization of protest voting may include 25 of the variables used in constructing our protest variable are taken from Wave 6 of the 20142017 British Election Study Internet Panel study, with the exception of leader affect variables, lagged one wave so as to prevent endogeneity, and the variable that asks respondents whether they would trust each party in government, which is only asked in, and taken from, Wave 5. The models, presented in Table 7, take the same general format as the models for protest voters, though leader affect (lagged) is also included as a political communication variable, on the grounds that the personality of the leader could well be a factor that adds or detracts from the attractiveness of a particular party, even if it is unlikely on its own to be sufficient to motivate the casting of a protest vote. Given that we are here modeling a binary choice between the two anti-establishment parties, our dependent variable needs to be a vote for one of the two, and we select UKIP for this purpose. Table 7 about here As can be seen from the data in Table 7, the choice to vote for the far-right anti-European party rather than the radical left environmental party is, not surprisingly, strongly conditioned by attitudes toward the European Union, as well as by (negative) attitudes toward environmental protection. What is perhaps most interesting about these findings is the strong impact of (lagged) leader affect and contact by the Green party, which appears to have gone some way toward swaying vote choice toward the Greens. Contact by UKIP is not significant. The relative effect of these factors comes out quite clearly in Figure 4, which presents the significant coefficients tactical voting in such cases, we keep these voters because (1) they represent a tiny proportion of the total votes for both parties (2) the distinction between electorally competitive and uncompetitive is not clear and (3) we do not make assumptions about the voters’ ability to forecast their constituency elections. 26 from the combined model graphically. Also interesting is the fact that once the issue and campaign variables are controlled, the trust variables are not significant. And even in the ‘trust’ model, only satisfaction with democracy is significant, not trust in MPs or external efficacy. Given that trust features very prominent as a factor in the literature on protest voting, and given that it is significant in our own models of this behavior presented above, the fact that trust is less closely linked to protest choice confirms that there is insight to be gained from modeling these two phenomena separately, as we have done in this paper. Figure 4 here 5. Conclusion The analysis presented in this paper has yielded some interesting findings about protest voting, drawing on a case in which a spike in attrition from mainstream parties created an ideal laboratory in which to test this phenomenon. We have framed our analysis in terms of three possibilities: the decision to protest and the choice of party could both be based on ideological factors, on trust, or on campaign effects. We find support for all three hypotheses, but particularly for campaign effects and for certain forms of ideological commitment. As far as ideology is concerned, our findings suggest that protest voters are not voicing concern about changes in the positions of their erstwhile parties of choice in toto, but about specific shifts on certain issues. At first glance, the testing of single-issue protesting voting for UKIP – for much of its history a single-issue anti-EU party – and the Green Party – often perceived as a single-issue environmentalist party – should be obvious. However, in recent years both of these parties have moved beyond their core policy concerns to broader issues a trend that is common amongst populist parties that use a ‘host ideology’ from which to espouse other arguments. Whether or not UKIP and the Greens represent populist parties is not the concern of 27 this paper, but both parties have arguably moved on from their single-issues to anti-immigration in the case of UKIP (Dennison and Goodwin, 2015) and anti-austerity in the case of the Greens (Dennison, 2016). Our analyses show that protest voters are clearly sensitive to parties’ core issues, but less so issues such as austerity. This suggests that these parties’ recent hereshethic forays into new policy territory in search of votes are not likely to have won them votes from protest voters. We also find considerable support for campaign effects, which is not a factor widely considered in previous literature on protest voting; this finding can thus be viewed as one of the principal theoretical contributions of our paper. The political communication variables of party contact and leader affect played important roles in shaping protest voting. Feelings for UKIP leader Nigel Farage were the strongest such variable, but attitudes toward Green leader Natalie Bennett also played a strong role in conditioning support for her party, a result that many Greens averse to traditional leadership may find discomfiting. The leader and contact effects may be welcome news for both parties, as these are aspects of party appeal that they can address without altering their fundamental policy commitments. Whether these findings would hold in other political contexts remains to be determined in future research. The UK’s first-past-the post system is unusually effective in marginalizing small parties, but by the same token, relatively small changes in vote totals can lead to large changes in the share of seats won by a party, as was the case with the Scottish National Party, which increased its seats in Parliament from 6 to 56 in 2015. Protest voting under systems of proportional representation takes place under different institutional constraints, and this could well alter the calculus that subtends it. The next logical step in advancing this research will therefore be to extend it to a wider range of states. 28 Future research could also usefully drill down into sub-groups of protest voters to identify in greater detail the specific characteristics and behaviors typical of people motivated by different factors. It would also be interesting to take advantage of the panel structure of the British Election Study data to follow the voters analyzed here over time and discern changing patterns in party support before and after the 2015 General Election. Furthermore, comparing the findings presented in this study to analyses of the recent rise in support for nationalist parties in Scotland and Wales may offer insights into the nature of their supporters. For now, we can conclude that, when offered a choice, protest voters do not simply choose randomly. Instead, it seems that a mix of core party policies and party campaigning go some way to dictating the protest voter’s decision. Protest voting is therefore fundamentally a political act, not simply an expression of disenchantment with politics. 29 References Abedi, Amir and Lundberg, Thomas Carl (2009), ‘Doomed to Failure? UKIP and the Organisational Challenges Facing Right-Wing Populist Anti-Political Establishment Parties’, Parliamentary Affairs 62(1), 72-87. Allen, Nicholas and Sarah Birch (2015), Ethics and Integrity in British Politics: How Citizens Judge Their Politicians’ Conduct and Why it Matters, Cambridge: Cambridge University Press. Bakker, Ryan, Erica Edwards, Liesbet Hooghe, Seth Jolly, Jelle Koedam, Filip Kostelka, Gary Marks, Jonathan Polk, Jan Rovny, Gijs Schumacher, Marco Steenbergen, Milada Vachudova, and Marko Zilovic. 2015. "1999-2014 Chapel Hill Expert Survey Trend File." Version 2015.1. Available on chesdata.eu. Chapel Hill, NC: University of North Carolina, Chapel Hill. Bartels, Larry M. 2014. ‘Ideology and retrospection in electoral responses to the Great Recession’. In Nancy Bermeo and Larry M. Bartels (eds.). Mass Politics in Tough Times: Opinions, Votes and Protest in the Great Recession. Oxford: Oxford University Press, 185-223. Bélanger, Eric (2004), ‘Antipartyism and Third Party Vote Choice: A Comparison of Canada, Britain, and Australia’, Comparative Political Studies 37(9), 1054-78. Bélanger, Eric and Kees Aarts (2006), ‘Explaining the Rise of the LPF: Issues, Discontent, and the 2002 Dutch Election’, Acta Politica 41(1), 4-20. Bellucci, Paulo (2014), ‘The Political Consequences of Blame Attribution for the Economic Crisis in the 2013 Italian National Election’, Journal of Elections, Public Opinion and Parties 24(2), 243–63. 30 Bergh, Johannes (2004), ‘Protest Voting in Austria, Denmark, and Norway’, Scandinavian Political Studies 27(4), 367-89. Betz, Hans Georg (1994), Radical Right-Wing Populism in Western Europe, Basingstoke: Macmillan. Birch, Sarah (2009), ‘Real Progress: Prospects for Green Party Support in Britain’, Parliamentary Affairs 62:1, 53–71. Bomberg, Elizabeth (1998), Green Parties and Politics in the European Union, New York and London: Routledge. Bowler, Shaun and David J. Lanoue (1992), ’Strategic and Protest Voting for Third Parties: The Case of the Canadian NDP’, Western Political Quarterly 45(2), 485-99. Brennan, Geoffrey and Loren Lomasky (1993), Democracy and Decision: The Pure Theory of Electoral Preference, Cambridge: Cambridge University Press. van der Brug, Wouter (2003), ‘How the LPF Fuelled Discontent: Empirical tests of explanations of LPF support’, Acta Politica 38(1), 89-106. van der Brug, Wouter and Meindert Fennema (2003), ‘Protest or Mainstream? How the European Anti-immigrant Parties Have Developed into Two Separate Groups by 1999’, European Journal of Political Research 42(1), 55-76. van der Brug, Wouter, Meindert Fennema and Jean Tillie (2000), ‘Anti-immigrant Parties in Europe: Ideological or Protest Vote?’, European Journal of Political Research 37, 77102. --- (2005), ‘Why Some Anti-Immigrant Parties Fail and Others Succeed: A Two-Step Model of Aggregate Electoral Support’, Comparative Political Studies 38(5), 537-73. 31 Brusis, Martin. 2016. ‘Democracies Adrift: How the European Crises Affect East-Central Europe’. Problems of Postcommunism, DOI: 10.1080/10758216.2016.1201772 Carter, Neil and Christopher Rootes (2006), ‘The Environment and the Greens in the 2005 Elections in Britain’, Environmental Politics 15(3), 473-78. Coan, Travis G, Jennifer L. Merolla, Laura B. Stephenson and Elizabeth J. Zechmeister (2008), ‘It’s Not Easy Being Green: Minor Party Labels as Heuristic Aids’, Political Psychology 29(3), 389-405. Cracknell, Richard (2010), ‘General Election 2010 Preliminary analysis’ House of Commons Library Research Paper 10/36, London: House of Commons Library. Dalton, Russell J. (1996), Citizen Politics: Public Opinion and Political Parties in Advanced Western Democracies (2nd ed.), Chatham NJ: Chatham House. --- (2004), Democratic Challenges, Democratic Choices: The Erosion of Political Support in Advanced Industrial Democracies, Oxford: Oxford University Press. Dennison, James (2015), ‘The Other Insurgency? The Greens and the Election’, Parliamentary Affairs 68(1): 188-205 Fennema, Meindert (1997), ‘Some Conceptual Issues and Problems in the Comparison if AntiImmigrant Parties in Western Europe’, Party Politics 3(4), 473-92. Finkel, Steven E. (1985), ‘Reciprocal effects of participation and political efficacy: a panel analysis,’ American Journal of Political Science 29(4): 891–913. Ford, Robert and Matthew Goodwin (2014), Revolt on the Right: Explaining Support for the Radical Right in Britain, London and New York: Routledge. Goldberg, J. H., J. S. Lerner and P. E. Tetlock (1999), ‘Rage and Reason: The Psychology of the Intuitive Prosecutor’, European Journal of Social Psychology, 29, 781–795. 32 Grabb, Edward A. (1979), ‘Working-Class Authoritarianism and Tolerance of Outgroups; A Reassessment’, Public Opinion Quarterly 43, 36-47. Green Party of England and Wales (2015), ‘Membership of the Green Party of England and Wales Passes 60,000’, https://www.greenparty.org.uk/news/2015/04/15/membership-ofthe-green-party-of-england-and-wales-passes-60,000/ Hawkins, Oliver, Richard Keen and Nambassa Nakatudde (2015), ‘General Election 2015’, House of Commons Library Briefing Paper CBP7186, London: House of Commons Library. Hetherington, Marc (1999), ‘The Effect of Political Trust on the Presidential Vote, 1968-96’, American Political Science Review 93(2), 311-26. Hirschman, Albert O. (1970), Exit, Voice and Loyalty: Responses to Declines in Firms, Organizations and States, Cambridge, MA: Harvard University Press. Ipsos Mori (2015), ‘Politicians Trusted Less than Estate Agents, Bankers and Journalists’, https://www.ipsos-mori.com/researchpublications/researcharchive/3504/Politicianstrusted-less-than-estate-agents-bankers-and-journalists.aspx Jennings, Will and Gerry Stoker (2015), ‘Two Polarities of Anti-Politics: why trying to be friends with both Ukip and Green supporters won’t work for the mainstream parties’ http://sotonpolitics.org/tag/gerry-stoker/ Kang, Wong-Taek (2004), ‘Protest Voting and Abstention under Plurality Rule Elections: An Alternative Public Choice Approach’, Journal of Theoretical Politics 16(1), 79-102. Karp, Jeffrey A. and Susan A. Banducci (2008), ‘Political efficacy and participation in twentyseven democracies: how electoral systems shape political behaviour,’ British Journal of Political Science 38(2): 311–334. 33 Karyotis, Georgios and Wolfgang Rudig (2015), ‘Protest Participation, Electoral Choices and Public Attitudes’ in Georgios Karyotis and Roman Gerodimos (eds.), The Politics of Extreme Austerity: Greece in the Eurozone Crisis, Basingstoke: Palgrave-Macmillan, 123-41. Kselman, Daniel and Emerson Niou (2011), ‘Protest voting in plurality elections: a theory of voter signaling’, Public Choice 148, 395–418. Lerner, J.S., J.H. Goldberg and P.E. Tetlock (1998), ‘Sober Second Thought: The Effects of Accountability, Anger and Authoritarianism on Attributions of Responsibility’, Personality and Social Psychology Bulletin, 24(6), 563-74. Lubbers, Marcel, Mérove Gijsberts and Peer Scheepers (2002), ‘Extreme right-wing voting in Western Europe’, European Journal of Political Research 41: 345–378. McGuinness, Feargal (2012), ‘UK Election Statistics; 1918-2012’, House of Commons Library Research Paper 12/43, London: House of Commons Library. Machin, Stephen (2015), ‘Real Wages and Living Standards’, LSE Centre for Economic Performance Paper EA024, http://cep.lse.ac.uk/pubs/download/EA024.pdf March, Luke. 2012. Radical Left Parties in Europe. London and New York: Routledge. Mayer, Nonna. 2014. ‘The electoral impact of the crisis on the French working class: More to the right?’ In Nancy Bermeo and Larry M. Bartels (eds.). Mass Politics in Tough Times: Opinions, Votes and Protest in the Great Recession. Oxford: Oxford University Press, 266-96. Meguid, Bonnie M. (2005), ‘Competition Between Unequals: The Role of Mainstream Party Strategy in Niche Party Success’, American Political Science Review 99(3), 347-59. 34 Norris, Pippa (1999), ‘Conclusions: The Growth of Critical Citizens and its Consequences’, in Pippa Norris (ed.), Critical Citizens: Global Support for Democratic Governance, Oxford: Oxford University Press, 257-72. --- (2002), Democratic Phoenix: Reinventing Political Activism, Cambridge: Cambridge University Press. --- (2004), Electoral Engineering: Voting Rules and Political Behaviour, Cambridge: Cambridge University Press. --- (2005), Radical Right: Voters and Parties in the Electoral Market, Cambridge: Cambridge University Press. --- (2011), Democratic Deficit: Critical Citizens Revisited, Cambridge: Cambridge University Press. Oesch, Daniel (2008), ‘Explaining Workers’ Support for Right-Wing Populist Parties in Western Europe: Evidence from Austria, Belgium, France, Norway, and Switzerland’, International Political Science Review 29(3), 349-73. O’Neill, Michael (1997), Green Parties and Political Change in Contemporary Europe: New Politics, Old Predicaments, Aldershot: Ashgate. Pattie, Charles and Ron Johnston (2001), ‘Losing the Voters’ Trust: Evaluations of the Political System and Voting at the 1997 British General Election’, British Journal of Politics and International Relations 3(2), 191–222. Pattyn, Sven, Alain van Heil, Kristof Dhont and Emma Onraet (2012), ‘Stripping the Political Cynic: A Psychological Exploration of the Concept of Political Cynicism’, European Journal of Personality 26: 566–579. 35 Poguntke, Thomas (1993), Alternative Politics: The German Green Party, Edinburgh: Edinburgh University Press. Pop-Eleches, Grigore (2010), ‘Throwing out the Bums: Protest Voting and Unorthodox Parties after Communism’, World Politics 62(2), 221-60. Powell, Jr., G. Bingham (1986), ‘American Voter Turnout in Comparative Perspective,’ American Political Science Review 80(1): 17–43. Ramiro, Luis and Raul Gomez. 2016. ‘Radical-Left Populism during the Great Recession: Podemos and Its Competition with the Established Radical Left’. Political Studies, DOI: 10.1177/0032321716647400 Schumacher, Gijs and Matthijs Rooduijn (2013), ‘Sympathy for the ‘Devil’? Voting for Populists in the 2006 and 2010 Dutch General Elections’, Electoral Studies 32, 124-33. Schaefer, Armin and Wolfgang Streeck (2013), ‘Introduction’, in Armin Schaefer and Wolfgang Streeck (eds.), Politics in the Age of Austerity, Cambridge: Polity, 1-25. Tarrow, Sidney (2000), ‘Mad Cows and Social Activists: Contentious Politics in the Trilateral Democracies’, in Pharr and Putnam (eds.), Disaffected Democracies: What’s Troubling the Trilateral Countries?, Princeton, NJ: Princeton University Press, 270-90. Tillman, Erik. R (2016), ‘Has the Global Financial Crisis Changed Citizen Behaviour? A FourCountry Study’, in Jack Vowles and Georgios Xezonakis (eds.), Globalization and Domestic Politics, Oxford: Oxford University Press. Webb, Paul (2005), ‘The Continuing Advance of the Minor Parties’, Parliamentary Affairs 58(5), 757-75. 36 Table 1: Party choice motivation at the 2015 UK General Election Which one of the reasons on this card comes closest to the main reason you voted? None 2.0% I always vote that way 18.4% I thought it was the best party 53.6% I thought that party had the best candidate 10.1% I really preferred another party but it had 7.2% no chance of winning in this constituency Other 8.0% Don’t Know 0.8% N 27,771 Source: Wave 6 of the 2014-2017 British Election Study Internet Panel survey www.britishelectionstudy.com. 37 Table 2: Party choice of self-described protest voters (England only) Party Labour Liberal Democrats UKIP Green Other Total Frequency 4 1 71 15 12 103 % 3.9 1.0 68.9 14.6 11.7 100 Source: Wave 6 of the 2014-2017 British Election Study Internet Panel survey www.britishelectionstudy.com. 38 Table 3: Four-way Operationalization of Protest Voting (England only) (a) Unspecified protest Spoilt Ballot (b) Protest against habitual party Identify as a supporter of mainstream party (Con, Lab, LD) + Vote for nonmainstream party (UKIP or Greens) (c) Protest against the political class Minimum trust in MPs + Vote for nonmainstream party (UKIP or Greens) 39 (d) Protest against political institutions Minimum satisfaction with UK Democracy + Vote for nonmainstream party (UKIP or Greens) Table 4: Logistic Models of Protest Voting at the 2015 UK General Election (England only) Male Age Class (ref: professional) Intermediate Small employer/self Routine/semi-routine Degree Household Income Left-right self-placement Anti-Austerity Pro-immigration Leave EU Environmentalism (1) Ideological model 0.971 (0.116) 1.008 (0.00443) (2) Trust model 0.960 (0.144) 1.073 (0.203) 0.996 (0.208) 1.137 (0.140) 1.004 (0.0183) 0.980 (0.0294) 1.035 (0.0693) 0.907*** (0.0182) 2.720*** (0.385) 1.148* (0.0500) Trust MPs 1.053 (0.110) 1.016*** (0.00424) (3) Political context model 1.062 (0.109) 1.014*** (0.00413) (4) Combined model 0.946 (0.113) 1.014** (0.00453) 0.961 (0.128) 1.104 (0.188) 0.864 (0.162) 0.902 (0.102) 0.990 (0.0162) 0.936 (0.124) 1.191 (0.200) 0.850 (0.157) 0.871 (0.0949) 0.961** (0.0148) 0.873 (0.132) 0.958 (0.184) 0.838 (0.181) 1.141 (0.142) 1.001 (0.0182) 0.935*** (0.0184) 2.837*** (0.412) 1.080 (0.0490) 0.818*** (0.0373) 1.116* (0.0567) 0.773*** (0.0560) 0.827 (0.0972) 1.023*** (0.00485) 0.723*** (0.0375) 0.201*** (0.0939) 6,452 0.759*** (0.0291) 1.114* (0.0502) 0.769*** (0.0494) External inefficacy Satisfaction with UK democracy Contact by any party 0.0845*** (0.0488) 0.128*** (0.0497) 0.672*** (0.0695) 1.019*** (0.00462) 0.697*** (0.0348) 0.108*** (0.0353) 6,100 8,774 8,892 Constituency margin Small party wasted vote Constant Observations 40 Table 5: How do Conservative, Labour and LD identifiers vote? No vote Con Lab LD UKIP Green Spoilt ballot Independent Minor Con ID 5.3 82.3 1.8 2.7 7.1 0.5 0.1 0.17 0.11 Lab ID 6.0 4.0 79.7 3.3 4.0 2.4 0.1 0.16 0.32 LD ID 6.5 12.8 16.3 55.5 3.2 4.8 0.2 0.5 0.3 Source: Wave 6 of the 2014-2017 British Election Study Internet Panel survey www.britishelectionstudy.com. 41 Table 6: How do Conservative, Labour and LD identifiers protest vote? Protest only UKIP Green Spoilt ballot Independent Minor Con ID 89.4 6.6 0.6 2.1 1.4 Lab ID 57.0 34.8 1.3 2.3 4.6 LD ID 35.6 54.3 1.8 5.2 3.1 Source: Wave 6 of the 2014-2017 British Election Study Internet Panel survey www.britishelectionstudy.com. 42 Table 7: Logistic models of electoral support for UKIP, rather than the Greens, in the 2015 General Election (England only) Male Age Class (ref: professional) Intermediate Small employer/self Routine/semi-routine Degree Household Left-right self-placement Anti-Austerity Pro-immigration Leave EU Environmentalism (1) Ideological model 3.034* (1.326) 1.024 (0.0202) (2) Trust model 2.038** (0.528) 1.058*** (0.0116) (3) Political context model 1.222 (0.488) 1.040* (0.016) (4) Combined model 1.802 (1.367) 1.030 (0.025) 0.818 (0.491) 1.306 (1.247) 0.760 (0.516) 0.441* (0.178) 1.115 (0.0871) 1.514*** (0.174) 0.790 (0.210) 0.807*** (0.0513) 7.111*** (3.021) 0.326*** (0.081) 1.177 (0.363) 2.556 (1.233) 1.387 (0.584) 0.378*** (0.0976) 1.048 (0.0441) 1.206 (0.703) 1.516 (1.312) 0.804 (0.599) 0.734 (0.338) 1.018 (0.064) 0.354 (0.438) 0.154 (0.225) 0.155 (0.243) 0.650 (0.552) 1.008 (0.152) 1.363 (0.237) 0.931 (0.115) 5.493* (4.715) 0.399* (0.143) Trust MPs 0.917 (0.0826) 1.128 (0.130) 2.043*** (0.354) External inefficacy Satisfaction with UK democracy Like Farage 2.326*** (0.270) 0.480*** (0.067) 3.883** (2.073) 0.188** (0.104) 1.017 Like Bennett UKIP contact Green contact Constituency margin 43 1.185 (0.583) 2.398*** (0.459) 0.583** (0.104) 3.258 (2.162) 0.176* (0.124) 0.00996* (0.0188) 0.0185*** (0.0175) (0.016) 0.823 (0.155) 0.178 (0.286) 533 692 596 Small party wasted vote Constant Observations 44 0.856 (3.132) 472 Figure 1: Vote share of minor parties in the UK, 1979-2015 30 25 20 15 10 5 0 1940 1950 1960 1970 1980 1990 2000 2010 Sources: McGuinness, 2012; Hawkins, Keen and Nakatudde, 2015 45 2020 Figure 2: Change in polling figures of major parties 2010-2015 50 45 40 35 30 25 20 15 10 5 0 Conservatives Labour Liberal Democrats Source: ukpollingreport.co.uk 46 UKIP Greens Figure 3: Standardized estimated effects of determinants of protest voting 47 Figure 4: Standardized estimated effects of determinants of voting for UKIP rather than the Greens 48 Appendix Table A1: ‘Other’ electoral choices amongst protest voters (England only) Response I spoilt my paper Independent Independent candidate Monster Raving Looney NHS Action Party National Health Alliance None of the above North East Party SNP TUSC Don’t wish to say Frequency 1 2 1 1 1 1 1 1 1 1 1 Source: Wave 6 of the 2014-2017 British Election Study Internet Panel survey www.britishelectionstudy.com. 49 Table A2 Distribution of ‘other’ voters at the 2015 UK General Election (England only) Spoilt ballot Independent Minor parties Refused to answer Mainstream parties (Con, Lab, LD) Outsider parties (UKIP, Grn) Total Frequency 40 58 91 114 5 % (weighted) 11.7 19.4 31.3 34.2 1.3 5 313 2.2 100.1 Source: Wave 6 of the 2014-2017 British Election Study Internet Panel survey www.britishelectionstudy.com. 50 Table A3: ‘Other’ reasons for voting that constitute a protest motivation All other Parties lie. break the mould! angry with lib dems for coalition with the tories Basically I'm a Conservative who wants a stronger policy against immigration into the UK because I didn't want eityer of the 2 main parties because no party can do right by this country no matter their promises Because the main parties are not listening to ordinary working class people on Immigration and it's effects on our lives and this country because they are all as bad as on and other Because they were not the others Cameron forced through gay marriage not even on his manefestotout. a Cameron refused to debate our future in the EU could not see MIlliband as PM. dislike Cameron. so as use my vote I voted the way I did could not vote for cameron did not want to vote for main parties did not want to vote for the party that allways wins here but did want them to loose didn't like the main parties manifestos Didn't trust the main parties didn't want to vote for others, gave them a chance Didn't want to vote for the big two Didn't want to vote for the main players didnt like the other parties disenchanted Disillusioned with main parties disilusioned Dissilussioned with 3 main parties Do not trust 3 main parties don't trust the main three dont like other parties Fancied a change fed up with the main parties Fed up of arrogance of main parties fed up with the main party;s Fed up with the main stream parties doing nothing fed up with the other parties fed up with the other parties ignoring the man on the street feed up with all the other party Feed up with the main oarties FPTP makes votes in my constituency worthless had no faith in other 3 main parties hate the SNP Hopefully it will make the other parties wake up to what people want hopefully to kick the main parties out of their complacency. I always vote for another party, but disagree with a number of their implementations over the last 5 years, feeling that their policies disagreed with the parties historic vision and objectives I am a Tory but cannot forgive Cameron for the bedroom tax I could not vote for major parties because they introduced gay marriage I couldn't bring myself to vote for any of the other parties i did not really like any of theothers.It was alast minute decision.The least worst choice I didn't believe in Labour or Conservative (who had the best chances) or the lesser publicised candidates. I didn't like the other parties I didn't trust the other parties I dislike other party's main policies more I don't trust the other parties I felt I couldn't vote for other parties I felt that the mainstream parties were all so similar and to be honest I just did not trust any of them with regards to the promises they made and I now wait to hear the rhetoric and the blame and excuses when they do not deliver on some of them I guessed The Conservatives could win so I wanted a party which would be a thorn in is side regarding Europe and immigration I have lost faith in All other parties as they don't care enough for Britain. I have no trust in the other parties, ukip have a good plan for immigration issues I live in a labour safe seat but wanted to send a message to labour that they were not far enough to the left on many issues i think it is time for a fresh party I thought it was time the Politicians got a wake-up call, they are so far removed from the electorate I voted for UKIP to show other parties that current high levels of immigration and issues with the EU are very important. Other parties do not appear to be addressing this issue. I voted on a single-issue (the EU) and primarily to have my vote counted as an anti-EU vote in the total because I live in a 51 Labour-Liberal Democrat marginal where UKIP cannot win. I would not vote UKIP otherwise. I wanted a change a shake up of establishment I wanted a Conservative government but, as I live in a safe constituency, I voted UKIP as I feel strongly about the EU I wanted change I wanted some competition to the big two parties i wanted someone differnt to have a chance i wanted to do something different to usual I wanted to make a point in a very safe seat I wanted to send a message that more and more people are concerned about environmental issues. I wanted to send the message to the Government that they really need to listen to people and do something to tackle immigration and stop branding concerned people as bigots and racists! There are too many people in the country, it's full, and letting more I wanted to upset the 2 party status quo I was looking for a change, other parties stated the same old policies which they never keep too. I wished the Tories to know the depth of support for UKIP policies. I would normally vote Conservative but I wanted UKIP to get a few seats. in a safe Conservative seat area and wanted to make point re EU and a few other issues without shurting chance of Conservatives not winning in the area in a safe Labour seat my vote wouldn't change anything but more votes for Greens might encourage people to see past the old situation of two or three main parties. It needs someone to shake the whole system up It's time for a real change in UK politics Its about time we had a change from the same old waffle from both the main parties just wanted to deliver a kick to the main parties Kicking of the main parties for all lying last time ley down by mainstream polical parties lib dem lost trust, the tories have increased the national debt, and miliband is a geek that should never have been leader let alone PM main party's didn't deserve my vote Make a point Making a statement mix things up a bit.and hoped others would do the same. my area is very strongly labour who I DIDN'T want to get in so I voted Ukip to shop support for an alternative party in politics when coun ting votes for each party overall, none3 are going to be able to sort out this country My estimation of UKIP as a party is very low, but voting for it sends a signal about issues which concern me - immigration, the EU, civil liberties. no clarity in the policies of the main parties No obvious choice no true party to vote for None of the major parties worth a vote not mainstream can't be tworse than the main parties Not part of Westminster bubble Not totally happy with the Conservatives, but they would win here anyway politics need freshening up Send a message to larger parties someone different the labour and conservative parties needed a shake up The main partys need to realize immigration and the EU are VERY important issues The other big 2 needed a shake up. the rest have let us down the three main parties all voted for this new marriage bill the two main parties need a shake up They were the only choice that wasn't any of the four main parties or UKIP things need a shake-up This area would vote in a monkey if it were supported by Labour, however, I hoped that enough disaffected supporters from everywhere would give UKIP a chance and the major parties deserve a kicking this country needs a shake up and a new and fresh party is a step forward thought it was time for a change Time for Britain to change time for change in country To focus the attention of the Conservative Party on important issues that they weren't fully addressing. to kick the tories up the backside to make a statement about my concerns about immigration to punish the conservatives To register disapproval of EU membership. To send a message to new government to do something about imigration to send a signal to the Labour Party 52 To send message to Labour Party to shake up the others to shake up the others to show the main parties they need to get real. to stir the pot To try and make point on immigration To try and shake the main parties up To try to shake up the status quo Totally disallusioned with Labour who are too far right UK politics needs an overhaul UKIP had no chance of winning but I think they got other parties thinking about what the ordinary person wanted. vote against apathy Wake up other parties want to encourage other parties actually likely to win to take environmental issues more seriously Wanted a change but knew Labour would win in my area anyways wanted change for the UK wanted to vote for a left wing party but send a message to the established order that they need to change warning to our MP we need more honesty i cant believe what the others are saying We need to break the two party system of boom and bust We needed a change from the usual corrupt parties we so need change,and we won't get it for another five years. Source: Wave 6 of the 2014-2017 British Election Study Internet Panel survey www.britishelectionstudy.com. 53 Table A4: Operationalization of Independent Variables and Descriptive Statistics Variable Question Male Age Class Are you male or female? What is your age? National Statistics Socio-economic Classification Degree What is the highest educational or work-related qualification you have? Household What is your gross household Income income? (1-15 categorical scale) Left-right In politics people sometimes talk of selfleft and right. Where would you place placement yourself on the following scale? (0 = left) AntiDo you think cuts to public spending Austerity in general have gone too far or not far enough? (5 = gone much too far) ProDo you think immigration is good or immigration bad for Britain's economy? And do you think that immigration undermines or enriches Britain's cultural life? Leave EU If there were a referendum on Britain's membership of the European Union, how do you think you would vote? (Leave = 1) Environment Do you think measures to protect the alism environment have gone too far or not far enough? (1 = gone much too far) Trust MPs How much trust do you have in Members of Parliament in general? (7 = a great deal) External How much do you agree politicians inefficacy don't care what people like me think? Satisfaction On the whole, how satisfied or with UK dissatisfied are you with the way that democracy democracy works in the UK as a whole? Like Farage How much do you like or dislike Nigel Farage? (10 = like) Like Bennett How much do you like or dislike Natalie Bennett? (10 = like) Contact by Have any of the political parties any party contacted you during the past four 54 Observ ations 29656 29656 Mean 0.469 48.102 Std. Min Max Dev. 0.499 0 1 17.147 16 95 18153 34.832 17.782 12 70 29472 0.398 0.490 0 1 21631 6.802 3.595 1 15 18202 5.103 2.486 0 10 20158 3.629 1.065 1 5 21359 7.371 3.667 2 14 17913 0.446 0.497 0 1 19336 3.396 1.041 1 5 21196 3.385 1.573 1 7 21055 2.160 1.108 1 5 20317 2.378 0.875 1 4 29539 3.090 3.260 0 10 24714 3.482 2.667 0 10 20923 0.578 0.494 0 1 weeks? UKIP contact Have UKIP contacted you during the past four weeks? Green Have the Green Party contacted you contact during the past four weeks? 2010 Constituency margin Small party How much do you agree people who wasted vote vote for small parties are throwing away their vote? 55 29656 0.185 0.388 0 1 29656 0.106 0.308 0 1 29650 17.831 11.704 20922 2.621 1.142 63.2 0.08 1 1 5