The party (un)faithful: Explaining party members’ defecting voting behaviour in different contexts (Belgium and Britain)

Abstract

Party members are often depicted as loyal voters that supply the party with a small but steady electoral base. Research, however, suggests that party members occasionally vote for other parties in general elections. This might either be a strategic choice or an indication of their discontent, possibly preceding a choice to exit the party. Using original survey data on more than 9000 party members in Flanders (Belgium) and Britain, this article aims to extend our understanding of why party members defect. We test hypotheses that take into account both individual and party-level considerations (ideological concerns, leadership evaluations and strategic choices) and the conditioning effects of party and electoral systems. Our analysis shows that institutional settings are crucial for strategic voting, whereas ideological concerns and leadership evaluations matter equally in both countries. As such, our study provides important insights in the attitudes and behaviour of party members, contributing to current debates on the contemporary meaning of party membership.

Keywords

comparative research disloyalty elections party members political parties survey vote defection voting behaviour

Introduction

Party members remain valuable assets for political parties, despite processes of membership decline (Van Biezen et al., 2012) and organizational change (Katz and Mair, 1995). Although they also bring costs (e.g. Scarrow, 1994; Seyd and Whiteley, 2004), members provide voluntary labour, financial support and policy input to the party. They constitute a pool of potential candidates, act as ambassadors to the public and symbolize the party’s support in society (Scarrow, 2014; Van Haute and Gauja, 2015; Webb et al., 2017b). Moreover, members are often said to supply the party with a small but steady electoral base, as they are seen as ‘vote-multipliers’ who gather electoral support through their everyday social contacts (Scarrow, 1994; Scarrow, 2014) and because they are assumed to be loyal voters themselves, supporting the party in ‘both good times and bad’ (Seyd and Whiteley, 2004: 360)

Recent empirical evidence nonetheless suggests that the electoral loyalty of party members should not be taken for granted as small but relevant segments of membership bases occasionally vote for another party (Polk and Kölln, 2017). Identifying whether and why party members defect is important, first, because it is reasonable to assume that a ‘vote-multiplying effect’ is provided only to the extent that members actually vote for the party themselves. Second, if even the most committed party supporters cast a defecting vote, it is not surprising that the less committed ones end up doing the same. Understanding members’ motivations then indirectly leads also to insights in the rationale of other voters. Third, as party members’ defection might indicate their discontent with the current state of affairs within the party, it may possibly precede the more drastic decision to exit the party, eventually leading to further membership decline (Bale et al., 2017; Wagner, 2017).

Using original survey data on party members in Flanders (Belgium) and Britain, this article aims to further explore the factors that drive party members towards a defecting vote. We build on the work of Polk and Kölln (2017) and, in addition to using a more generalizable operationalization of ‘pull factors’ (i.e. general left–right orientation instead of attitude towards migration) and testing this on original data, we supplement the explanatory framework with two major additional contributions. First, we add strategic voting considerations to ideological concerns and party leadership evaluations as explanations for defecting voting behaviour, and we hypothesize how these could have an impact. Second, and most importantly, in addition to individual-level explanatory factors, we also posit that these factors are conditional upon the institutional setting in which they take place. We choose our two cases, Flanders and Britain, as examples of two highly different party systems and electoral rules, and we develop arguments on why and when individual-level variables have (or not) an effect.

After reviewing the literature, constructing the explanatory framework and introducing the data, we test six hypotheses that take into account personal motives and institutional influences. Our data show that in Flanders and Britain between 4% and 16% of party members defect. Unlike ideological concerns and discontent with the party leadership, which seem to equally drive defecting voting behaviour in both systems, strategic considerations only matter in Britain, where its majoritarian electoral system decreases the likelihood for parties to obtain seats. This suggests that defecting members are not necessarily a sign of internal disagreement and that the impact of institutional incentives must not be forgotten when investigating party members’ voting behaviour.

Party members at the ballot box

Party membership can be considered as a formal and far-reaching form of party identification, which has traditionally been regarded as an important explanation for voting behaviour (Bartels, 2000). It is formal because people formally register to become a member, pay a fee and receive a membership card, which goes beyond just identifying oneself with a particular party. It is far-reaching, as members are not only expected to vote for the party but also to encourage others to follow their lead, performing a ‘vote multiplying’ function (Scarrow, 1994). They can do so by canvassing people in their community in the run-up to the election, by distributing campaigning material, by expressing their support through poster display and social media or by becoming a candidate themselves (Bale and Webb, 2015). Even members who are not very active within the party can still act as ‘party ambassadors’, spreading the party message and convincing others (Ponce and Scarrow, 2016). Despite trends of professionalized electoral campaigns (Lisi, 2013) and although non-members can also be involved in campaign activities, party members remain pivotal because they are much more likely to engage in ‘high intensity’ type of activities such as face-to-face and phone canvassing, as opposed to ‘medium’ or ‘low’ intensity ones such as leafletting or social media support (Webb et al., 2017b).

These are non-trivial contributions, particularly considering that previous empirical studies have demonstrated that a small but nonetheless significant association exists between parties’ electoral activity and their electoral performance (André and Depauw, 2016; Fisher and Denver, 2009; Johnston and Pattie, 2003; Karp et al., 2008). Research on the electoral importance of party memberships has not, however, led to unequivocal results. In Belgium, the number of party members before elections appears to be a good predictor for the election result of that party in the subsequent year (Hooghe and Dassonneville, 2012). A similar phenomenon was observed in the United Kingdom, although a straightforward relationship could not be found for all parties (Fisher et al., 2006). Mixed evidence is also found on Eastern Europe. Whereas Tavits (2012) finds that membership size is significantly associated with electoral success, Gherghina (2014) concludes that large memberships do not bring parties more electoral stability.

Whatever the case, it cannot be taken for granted that party members always vote for their own party. Looking at European Social Survey data, Polk and Kölln (2017) calculated that on average about 8% of party members cast a defecting vote in elections (see also Rahat et al. (2014)). The fact that a relevant share of their members appears to defect is a threat for parties. Not only average voters, but even party members might use cues other than their partisan affiliation when deciding on which party to vote for. As Katz suggested a few decades ago (1990: 151), although members tend to be more loyal than non-members, the cause of party loyalty is not membership per se: Members are rather a ‘self-selected sample of the most loyal supporters of a party’. This makes determinants of voting behaviour among party members a topic worth of investigation.

The determinants of members’ voting behaviour

Why do party members cast a vote for another party at general elections? Although it seems irrational to join a party but then not to vote for it, we argue that vote defection could be a well-considered choice based on (1) ideological concerns, (2) leadership evaluations and (3) strategic considerations.

A choice for a particular party is, first of all, influenced by ideological concerns. People generally vote for a party because the policies it proposes correspond with their own views and priorities (Bélanger and Meguid, 2008; Rabinowitz and Macdonald, 1989). Although one could expect that the opinions of party members and their parties are highly congruent, this is not necessarily always the case. May’s ‘law’ of curvilinear disparity (1973), for instance, posits that party members tend to be more extreme than the party elite.¹ Also, the fact that many parties are informally subdivided into factions may lead to members fostering policy preferences that differ from those of the party (elite). Perhaps even more interestingly, in their empirical analysis of Canadian and Belgian party members, van Haute and Carty (2012) estimate that between 10% and 25% of members within each party are ‘ideological misfits’.

Dissatisfied party members have a choice to either remain loyal to the party despite divergence (‘loyalty’), to raise their voice in order to try and change things (‘voice’) or to leave the party altogether (‘exit’) (Hirschman, 1970). However, if one does not leave the party (which does happen because of ideological disagreement (Wagner, 2017)), the choice between loyalty and voice might be more nuanced. Based on van Haute‘s (2015) four-fold typology of intra-party responses to dissatisfaction along a voice/silence, constructive/destructive classification, a vote for another party can be classified as sabotage, a destructive type of voice, and perhaps a first step towards considering to leave the party for good.²

As ideological incentives are just one reason to join parties (Whiteley and Seyd, 1998), members who have ideological concerns might choose not to leave the party – either because of other material or social incentives, or because they expect things will change for the better in the near future – but rather to cast a vote for another party in elections. When analysing the effect of ideological concerns on party members’ vote defection, the distinction made by Polk and Kölln (2017) between ‘push’ and ‘pull factors’ is particularly helpful. Using a Downsian understanding of spatial voting (Downs, 1957), when the ideological incongruence (or distance) between a party member and his or her party is large, that member might be ‘pushed’ to vote for another party (‘push factor’). A party member might, however, also be attracted by another party because its programme corresponds closely with his or her own policy preferences, leading the member to casting a defecting vote (‘pull factor’). Unlike Polk and Kölln (2017), who operationalize pull factors with anti-migrant attitudes, we use a more general measure because we believe party members tend to make a general evaluation (based on multiple policy dimensions instead of just one) of where they stand compared to their own party and other parties when they decide whether to vote loyally or disloyally. We then look at the absolute distance between a member’s perceived position on a general left–right scale and the perceived position on the same scale of his/her own party (push factor) as well as at the distance between a member’s self-positioning and the perceived position of what he or she considers the most ideologically close party (pull factor). Although pull and push factors are certainly interrelated, especially in crowded political spaces, we make a clear distinction between the two as they are analytically different concepts: A member might be ideologically close to one’s own party, while simultaneously experiencing the close ideological proximity of other parties. We therefore formulate two hypotheses:

H1a: The greater the ideological distance between a party member and his/her party, the more likely (s)he is to cast a defecting vote (‘push’ factor).

H1b: The closer a party member feels ideologically to another party, the more likely (s)he is to cast a defecting vote (‘pull’ factor).

This article also contributes in exploring how institutional settings related to the broader political system affect the individual considerations party members make when deciding how to cast a vote. Particularly relevant here are electoral rules and the party system, two aspects on which our cases differ vastly. While Belgium has a flexible-list proportional representation (PR)-system and a fragmentized multiparty system, the United Kingdom has a majoritarian single-member plurality system (also known as first-past-the-post (FPTP)) and a (still predominantly) two-party system.

When it comes to the influence of ideological concerns on members’ voting decision, especially the constellation of the party system seems to matter. Dejaeghere and Dassonneville (2015) have shown how the number of parties and the degree of ideological polarization in a party system have an impact on the probability of voters to switch parties. The former factor has a positive effect, the latter a negative effect on vote switching. The chance of finding another party that deserves one’s vote is higher when there are more parties to choose from. Moreover, the ideological distance between parties tends to be lower in systems with many parties, making vote switching more likely (Tavits, 2005).

In Belgium, many parties compete for voters’ support and hence it is reasonable to expect that substantive and ideological party differences are smaller than in a majoritarian country such as the United Kingdom (Bouteca, 2011). When we look at the ‘effective number of parties’ (Laakso and Taagepera, 1979), an indicator that takes into account the number and strength of parties in a party system, we see that at the UK General Election of 2015 the effective number of parties that obtained votes was 3.9, and 2.5 if we only take into account parties that obtained seats. In Belgium (as a whole), for the 2014 elections, these figures are much higher (9.6 and 7.8, respectively) (Döring and Manow, 2016). Even when we calculate the effective number of parties only for the regional Flemish Parliament, these are clearly higher than the United Kingdom with 5.1 and 4.5, respectively. In sum, as the effective number of parties is larger in Flanders than in the United Kingdom, we expect that ideological differences with neighbouring parties are smaller and that members more easily find political parties whose ideological positions are close to their own. This is indeed confirmed by our data: whereas the mean distance to the closest party in Flanders is 1.04 (confidence interval: 1.00–1.07) on a 11-point left–right-scale, this is significantly different in Britain with 1.60 (CI: 1.56–1.64). As such, we expect that both ideological considerations (push and pull factors) will play a more limited role in Britain than in Flanders.

H2: Ideological concerns are less important for party member defection in Britain than in Flanders.

A second factor that could lead party members to casting a defecting vote is leadership evaluations, which seem particularly important in light of the growing literature on the personalization of politics (Karvonen, 2010; Rahat and Sheafer, 2007). Evaluations of individual candidates, and party leaders in particular, are expected to be an increasingly important determinant of voting behaviour. Empirical evidence of leadership effects on individual vote choices has, however, been mixed (Balmas et al., 2014; King, 2002; Wauters et al., 2018). It appears that leaders tend to matter more when partisanship is low (Holmberg and Oscarsson, 2011: 50).

Although party members are by definition those who strongly identify with their party, we do expect that leadership evaluations are an important explanatory factor for their defecting voting behaviour. Party members are more directly affected by the performances of party leaders than regular voters. This is the case, for instance, when members perceive that a new leader is dramatically shifting the party’s direction, as in the recent case of Jeremy Corbyn’s Labour party. Polk and Kölln (2017) indeed find that members’ negative leader evaluations increase the probability of electoral disloyalty in Sweden and Britain.

Casting a defecting vote might be a tempting option not only of dissatisfaction with (the policy preferences of) a party leader (e.g. Hirschman, 1970; van Haute, 2015), but also in case of psychological attachments to another aspirant leader (especially if an intra-party contest takes place) (Cross and Pruysers, 2017; Southwell, 2010). In any case, it is a less radical option than leaving the party: Party members remain to some extent loyal to the ‘party brand’ by staying on board as members (perhaps in the hope that one day the party leader will resign) but might once (or as long as the party leader is in function) cast a defecting vote (e.g. Webb et al., 2017a). Thus, we posit that the more negative party members evaluate their party leader, the more likely it will be that they cast a defecting vote.

H3: The more negatively a party member evaluates her/his party leader, the more likely (s)he is to cast a defecting vote.

The degree to which leadership evaluations impact party members’ voting decisions might also be influenced by the electoral system. Research suggests that electoral leadership effects exist in all systems, but that they are more outspoken in majoritarian than in proportional systems (Holmberg and Oscarsson, 2011) because majoritarian systems tend to put more emphasis on individual representatives, whereas PR-systems promote parties rather than candidates. Moreover, ensuring that the leader of the party with most votes will become prime minister, as generally is the case in majoritarian systems, might work as an incentive to focus more on the leader in the voting calculus. We expect this may also be the case for party members’ voting behaviour.

H4: Evaluations of party leaders are more important for party member defection in Britain compared to Flanders.

Finally, we argue that party members’ decision to cast a defecting vote is not necessarily an indication of their dissatisfaction with party leaders or the ideological course of the party, but may well be a strategic consideration. Although members will usually cast a vote for the party they prefer, sincere voting is not the only option. Like voters, members can choose to cast their vote strategically or tactically (Fisher, 2004) for a party with a better chance of being pivotal and influential. They might do so because they believe that their preferred party will not obtain any seats or because they aim to bring a particular government majority to power (Gschwend, 2007). Although their party attachment might lead them to overestimate their parties’ chances of winning (Blais, 2002), higher levels of political sophistication, corresponding with the average profile of party members (Scarrow and Gezgor, 2010), have been found to positively impact the likelihood of a strategic vote (Gschwend, 2007).

Strategic voting does not only occur in majoritarian electoral systems in order not to ‘waste’ votes, but also in systems of PR (Irwin and Van Holsteyn, 2012). Whereas in majoritarian system like Britain voters tend to strategically defect from their preferred small party because winning seats is unlikely, strategic voters in PR systems like Flanders mainly aim to influence the coalition formation by voting for large parties who have a better chance of ending up in government (and weighing on decision-making while in power). Although voters could also strategically vote for small parties that they expect to be pivotal to coalition formation, smaller parties mostly tend to be victims of strategic behaviour. Thus, when it comes to voting at elections, we expect party size to affect party members in both countries in similar directions: Smaller parties are more vulnerable for strategic voting than larger parties.

H5: Members of smaller parties are more likely to cast a defecting vote than members of larger parties.

As briefly touched upon above, the electoral system, and particularly the proportionality between vote shares and seat shares, can be expected to condition the effect of strategic considerations on members’ vote shares. While in a PR system the explicit purpose is to maximize correspondence between both shares, a plurality system first and foremost aims to create a clear and stable government majority based on territorial representativeness (Duverger, 1954; Lijphart, 1990). The Gallagher (1991) index of disproportionality for the most recent elections in Belgium (2014) is 4.7, and this clearly differs from that of the United Kingdom which is 15.0 (Döring and Manow, 2016). Larger disproportionality creates a larger disadvantage for smaller parties and makes strategic voting more likely (Crisp et al., 2012; Gschwend, 2007). Tactical voting in PR systems (which is rather about coalition formation) is less likely since it presupposes more voters’ knowledge: Government and coalition formation are harder to understand and require more information (Irwin and Van Holsteyn, 2012).³ Or it is less visible as when voters expect small parties to be pivotal during government formation. Thus, we expect that members will vote strategically more often in Britain than in Flanders.

H6: Strategic considerations are more important for party member defection in Britain than in Flanders.

Data

We rely on broad-scale surveys conducted among grassroots members of five Flemish parties (Belgium) and six British parties. In contrast with public opinion or national election surveys that are conducted among the entire population and only contain a relatively small number of party members in the sample (when it is even possible to identify them at all), our surveys allow us to investigate party members’ voting behaviour and motivations in close detail.

In Flanders, a postal survey was conducted in the course of 2012 among members of the Flemish-regionalist party N-VA and the liberal-democratic Open VLD, following Dillman’s total design method. The same method was used 1 year later to survey party members of the Christian-democratic party CD&V and the ecologist party Groen, and again in 2015 for the social-democratic sp.a.⁴ Despite following the same method, response rates varied between parties (see Table 1). Due to funding reasons, surveys were conducted over different years. It is, however, unlikely that this has biased our reports of members’ vote defection to a large extent, as the Flemish parties’ electoral results have been very stable from 2010 onwards. Moreover, misremembering vote choices might occur, but this is more likely to occur among voters with low levels of political interest and political knowledge, while party members exhibit higher levels of both in comparison with the average voter. This makes misremembering less of an issue in our case.

Table 1.

Party members survey details, five Flemish parties.

Party	Party family	N	Response rate (%)	Period
Groen	Ecologist	931	62.0	April–June 2013
Sp.a	Social-democratic	583	38.9	June–October 2015
CD&V	Christian-democratic	666	44.3	March–June 2013
OpenVLD	Liberal-democratic	430	28.9	May–September 2012
N-VA	Regionalist (Flanders)	990	65.5	April–September 2012

N-VA members recorded the highest response rate with 65.5%, whereas for Open VLD we obtained a response rate of only 28.9%. In order to control for underrepresentation, the data were weighted according to sex and age category.

In Britain,⁵ 1 week after the May 2015 UK’s General Election, an online survey was conducted among party members of the Conservative party, the Labour party, the Liberal Democrats, the United Kingdom Independence Party (UKIP) and the Greens and the Scottish Nationalist Party (SNP; see Table 2). YouGov, an international Internet-based market research firm, recruited the survey respondents from a panel of around 300,000 volunteers who are paid a fee for completing a survey.⁶ At the beginning of the fieldwork period, 8840 YouGov panellists who were party members were invited to take part in the poll, and 5696 respondents subsequently took part in the survey (response rate of 64.4%). Although data are not weighted in any way since there are no known official population parameters for party memberships in the United Kingdom, data triangulation gives us confidence in the quality of the data.⁷

Table 2.

Party members survey details, six British parties.

Party	Party family	N	Period
Conservative	Conservative	1192	May 2015
Labour	Social-democratic	1180	May 2015
Lib Dem	Liberal-democratic	730	May 2015
UKIP	Eurosceptic, right-wing populist	784	May 2015
Greens	Ecologist	845	May 2015
SNP	Nationalist and social- democratic (Scotland)	963	May 2015

Note: UKIP: United Kingdom Independence Party; SNP: Scottish Nationalist Party. Total response rate: 64.4%

The dependent variable of our analysis is a dichotomous variable indicating whether a party member has voted for his/her own party in the last general elections (0 = vote for the own party; 1 = vote for another party). Only party members who casted a vote at the last general elections are included in the analysis.⁸

As for the independent variables, for the ideological ‘push factors’, we calculate the absolute difference between the score of party members on a self-placement left (0)–right (10) scale and their placement of the party on the same scale. For the ideological ‘pull factors’, we calculate the absolute difference between members’ self-placement on the left (0)–right (10) scale and their placement of the most adjacent other party on the same scale.⁹ For factors related to leadership evaluations, we asked Flemish party members to express the sympathy they feel for their current party leader on a scale from 0 to 10. Similarly, we asked to British party members to express how they feel their party leader performed in the electoral campaign on a scale from 0 to 10.¹⁰ Finally, for strategic considerations, we take the size of the party into account, that is, the total share¹¹ of votes they obtained in the previous general elections.¹² In order to control for the effect of other socio-demographic and political behaviour variables in our regression model, we include sex, three age categories (15–34; 35–65; 65+), educational level (graduates vs. non-graduates), party membership length (in number of years) and level of party activism. The latter is measured with a factor score resulting from country-specific factor analyses of campaign, selection and policy activities.¹³

Results

Descriptive analysis

Although most party members still vote for their own party, Table 3 represents that both in Flanders and Britain defecting voting behaviour is not a negligible phenomenon. On average, about 6% of the surveyed party members did not vote for the party they are member of in the previous general elections. At first glance, the share of defecting party members seems to be higher in Britain than in Flanders but only when we take into account all British parties, and not only those who won the highest number of seats (i.e. Tories, Labour and SNP).

Table 3.

Electoral results and members’ defecting voting behaviour.

	Flanders (Belgium)				United Kingdom
	Vote share 2010/2014¹⁴ (%)	Seats 2010/2014¹⁵	Defecting members (%)		Vote share 2015 (%)	Seats 2015	Defecting members (%)
Groen	7.1	5	4.0	Greens	3.8	1	15.8
Sp.a	14.2	13	7.8	Labour	30.4	232	4.9
Open VLD	14.0	13	6.1	Lib Dem	7.9	8	12.1
CD&V	17.6	17	4.9	Tories	36.9	331	3.8
N-VA	28.2	27	5.7	SNP	4.7	56	3.8
				UKIP	12.6	1	6.9

Note: SNP: Scottish Nationalist Party; UKIP: United Kingdom Independence Party.

However, parties that struggled to obtain many, if any, seats in Britain’s majoritarian electoral system (the Greens and Liberal Democrats in particular, but also UKIP) exhibit much higher shares of defecting votes (15.8, 12.1 and 6.9%, respectively) than mainstream parties (Conservative with 3.8% and Labour with 4.9%). In Flanders, we do not observe similar differences between small and large parties. Already from these descriptive data, strategic voting considerations seem to be a likely explanation for these differences across the two countries. In Belgium’s PR system, the risk of gaining no seats is much smaller for small parties than it is in Britain. Even when considering the possibility of strategic votes for coalition purposes, strategic voting seems to be less prevalent in Flanders, providing a first confirmation of our expectation as formulated in H6.

Explaining party members’ defecting voting behaviour

We now test the possible explanations for defecting voting behaviour in a more systematic way, by conducting a logistic regression analysis with a variable indicating whether a party member has voted for another party as the dependent variable.

The results in Table 4 show highly similar patterns for Flemish and British parties. First of all, ideological concerns appear to play a role in deciding to vote for another party. Unlike Polk and Kölln (2017), we find that not only ‘push factors’ (i.e. the perceived ideological distance to the own party) but also ‘pull factors’ (i.e. the ideological distance to the most adjacent party) have a significant effect in the expected direction on the probability to cast a defecting vote, leading to a confirmation of H1a and H1b. Ideological heterogeneity among members might seem logical or even desirable from the perspective of the catch all party (Kircheimer, 1966) as it gives parties more room for manoeuvre. However, these results do show that members who perceive a larger ideological distance between their opinions and those of their own party, and members who position themselves close to another party, are more likely to cast a vote for a different party than the party they are member of. This is because they either disagree with their own party or because they are attracted by the programme of an adjacent party.

Table 4.

Logistic regression for Flemish and British party members casting a defecting vote.

	Flemish parties			British parties
	B	SE	Exp(B)	B	SE	Exp(B)
Ideological concerns
Ideological distance own party (push factor)	0.172**	0.055	1.187	0.147**	0.046	1.159
Ideological distance adjacent party (pull factor)	−0.211*	0.100	0.810	−0.113**	0.039	0.893
Leader evaluations
Sympathy/performance score for own party leader	−0.236***	0.044	0.789	−0.246***	0.021	0.782
Strategic considerations
Vote share most recent elections (%)	0.009	0.012	1.009	−0.025***	0.005	0.975
Controls
Level of education (ref: graduate degree)	−0.099	0.184	0.906	−0.056	0.119	0.946
Age (ref: 65+)
15–34 years old	−0.383	0.312	0.682	−0.536**	0.181	0.585
35–65 years old	−0.273	0.231	0.761	−0.550***	0.149	0.577
Sex (ref: women)	0.059	0.183	1.060	0.044	0.125	1.045
Activity rate in the party	−0.682***	0.132	0.506	−0.467***	0.074	0.627
Years of party membership	−0.016*	0.008	0.984	−0.011*	0.005	0.989
Constant	−0.941	0.513	0.390	0.199	0.336	1.220
Nagelkerke pseudo-R²	0.12			0.10
N	2780			4891

***p < 0.001; **p < 0.01; *p < 0.05.

Based on differences in the party system, we hypothesized that ideological concerns would play a larger role in Flanders than in the United Kingdom (H2), but the analysis gives only limited evidence for this. It is actually the case that both the odds ratios (ORs) for the ideological distance to the own party and to the most adjacent party are slightly higher in Flanders than in Britain, but differences remain rather small and might only be due to the fact that models are based on different samples (OR 1.187 vs. 1.159 the former; 0.810 vs. 0.893 the latter). H2 is therefore mostly rejected. Consequently, ideological differences with one’s own party seem to play a similar role in defecting voting behaviour irrespective of the party system in which parties operate. Ideological distance from one’s own party as well as ideological proximity to a different party than one’s own are only slightly more likely to trigger vote defection in a PR system such as Flanders than in a majoritarian system such as Britain. Even though ideological differences with neighbouring parties seem to be on average smaller in Flanders than in Britain, our findings can be explained by the fact that not only in Britain greater defection happens (mostly for strategic reasons) from small parties to larger ones, but, crucially, also because in 2015 smaller parties are not too ideologically distant from larger ones. In fact, not only the most defecting voters are among members of the Greens (15.8%), Lib Dem (12.1%) and UKIP (6.9%), but virtually every defecting member in the Green party and more than half of the Lib Dem voted for Labour, whereas almost every defecting member in UKIP voted for the Tories. The average position of each party on the ideological spectrum, calculated from average self-positioning of its members, shows that the distance between Greens and Labour and between UKIP and Tories is only 0.49 and 0.46, respectively. The distance between Lib Dem and Labour is slightly larger, but still relatively small on a 11-point scale (1.7). This calls for a contextualization of the findings. Majoritarian systems might discourage vote defection for ideological reasons compared to a PR system because the number of parties tends to be lower and ideological party distance higher. However, strategic thinking and the actual distribution of parties in the left–right space of a specific country might end up masking this: Party members are changing to another party but tend to stay within the same ‘ideological bloc’.

Second, our results show that evaluations of party leaders affect the chance of casting a defecting vote in a significant way. Although the formulation of the questions slightly differed in Flemish and British questionnaires, the effects are again to a great extent similar and go in the same direction (OR 0.789 vs. 0.782). This means that we find support for H3: Party members who evaluate their party leader more negatively are more likely to cast a defecting vote.

However, majoritarian electoral systems do not seem to result in a greater leadership effect, leading us to reject hypothesis H4. This could be related to the increasing role of party leaders even in elections in list-PR systems (Wauters et al., 2018). Although they position themselves not as potential prime ministers, they act as the electoral leader of the party and increasingly determine the electoral result of their party. Moreover, it is perhaps no coincidence that the Flemish social-democratic party sp.a has the highest share of defecting members (7.8%). Its party leader (Bruno Tobback) was not very popular and was voted away as party leader in a leadership contest 1 year after the general elections.

We do find, third, that strategic considerations have a larger impact in Britain than in Flanders on defecting voting behaviour, at least in terms of smaller parties being the victim of such considerations. Members of smaller British parties have a significantly greater chance of voting for another party than members of larger parties.¹⁶ This is not the case in Flanders, where the vote share of the party has a small positive, but non-significant effect. This leads to a confirmation of H6 and only a partial confirmation of H5. As we have seen, in Britain, it is the three smallest parties, the Greens, the Lib Dem and UKIP, that suffer the most at election time, even among their stronger supporters, because of FPTP. It might certainly be the case that, given access to power is less of an issue for smaller parties in Flanders, strategic considerations are masked by a focus on coalition power rather than access to power. Whatever the case, the implication is that members of smaller parties in Flanders are less likely to cast a defecting vote than members of smaller parties in Britain.

Finally, we discuss the effects of the control variables. Not surprisingly, the activity rate in the party has a significant effect both in Flemish and British parties: The more active party members are, the less likely they are to vote for another party. If party members are considering voting for another party, they might be less active in the party’s campaign activities from the start. Also years of membership have a significant effect: The longer a person is a member of a party, the less likely that (s)he will cast a defecting vote. Socio-demographic variables (such as sex and age) do not have a significant effect on defecting voting behaviour in Belgium, while it is the over 65s that, controlling for everything else, tend to be least loyal to British parties (and in the Conservative party in particular).

Conclusion

In this article, we examined intra-party cohesion at the level of the ‘party on the ground’ by describing and explaining the voting behaviour of party members. Our findings in Flanders and the United Kingdom show that the electoral loyalty of party members is not absolute nor unconditional. While the majority of members vote for their own party, a significant share (of around 4–16%) casts a defecting vote in the ballot box.

Similar to previous research (Polk and Kölln, 2017), we showed that ideological motives and negative leadership evaluations partially explain defecting voting behaviour. Contrary to our context-driven expectations, however, these motives appeared equally important in Flanders and in Britain. Still, our results and interpretation indicate that the impact of the institutional setting should not be neglected. Strategic considerations are clearly affected by the electoral system: British party members’ voting decisions are (partially) shaped by the incentives provided by the electoral system. Although our research design does not allow us to generalize across different electoral and party systems outside of our two cases, this study provides a first contribution stressing the importance of taking into account the influence of the political context both in the theoretical framework and at the empirical level.

What are the practical implications for political parties? The strategic considerations of British members should probably be the least worrisome. Many ‘defecting’ members in this country would probably cast a vote for their own party if they were living in an electoral district where the chance of it winning a seat is higher. Their voting decisions in some cases seem to be a pragmatic choice rather than a choice out of disappointment. This means that they can still possibly act as an ‘ambassador’ for the party, although they might find it more satisfactory to do so in a different electoral district than their own, either face-to-face or through phone canvassing, or maybe using a national platform such as the Internet.

Strategic considerations aside, in order for party members not to cast a defecting vote (and not to convince others to do so), they need to have a degree of ideological congruence with the direction in which their party is headed, not to be ideologically too close to one or more of its competitors, and to be relatively positive about their party leader. In other words, they need not to be too dissatisfied with the current state of affairs within their party. If they are, then this raises the possibility that they will not only deny the party their own vote, but they may convince other party supporters to cast a defecting vote too. This could have effects in marginal constituencies of majoritarian system such as Britain, but also in the case of multiparty systems with small electoral margins like Flanders.

The long-term danger is that, if they remain dissatisfied for an extended period, defecting party members might consider the ‘exit option’ (Hirschman, 1970) by leaving the party through cancellation or not renewing one’s membership, further contributing to the widespread declining party membership trend.

Footnotes

Acknowledgements

The authors would like to thank Sergiu Gherghina, Caroline Close, as well as the other participants of the 2017 ECPR Joint Sessions Workshop ‘Rethinking Intra-Party Cohesion in Times of Party Transformation’ and the three anonymous referees for their valuable suggestions and comments. They would also like to express their gratitude to the respective party secretaries for their cooperation in the Flemish party member survey and Flemish Research Foundation (FWO) and Economic and Social Research Council for the grants.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Flemish Research Foundation (FWO; grant numbers 1521212N, 1504113N and 1518314N) and by the Economic and Social Research Council (Grant Number ES/M007537/1].

ORCID iD

Benjamin de Vet

Notes

References

Abramson

Aldrich

Blais

, et al. (2010) Comparing strategic voting under FPTP and PR. Comparative Political Studies 43(1): 61–90.

André

Depauw

(2016) The electoral impact of grassroots activity in the 2012 local elections in Flanders. Acta Politica 51(2): 131–152.

Bale

Webb

(2015) Grunts in the ground game: UK party members in the 2015 general election. In: Conference on ‘The 2015 British general election: parties, politics, and the future of the United Kingdom’, UC Berkeley: University of California. (unpublished paper).

Bale

Poletti

Webb

(2017) The secret of leaving: who quits their party and why? Grassroots members in the UK. In: EPOP conference, Nottingham, UK, 8–10 September 2017.

Balmas

Rahat

Sheafer

, et al. (2014) Two routes to personalized politics: centralized and decentralized personalization. Party Politics 20(1): 37–51.

Bartels

(2000) Partisanship and voting behavior, 1952-1996. American Journal of Political Science 44(1): 35–50.

Bélanger

Meguid

(2008) Issue salience, issue ownership, and issue-based vote choice. Electoral Studies 27(3): 477–491.

Blais

(2002) Why is there so little strategic voting in Canadian plurality rule elections? Political Studies 50(3): 445–454.

Bouteca

(2011) Van weermakers tot parapluverdelers: een onderzoek naar de mate van ideologische convergentie tussen Sp. a en Open VLD op de sociaaleconomische breuklijn (1961-2010). Ghent: Ghent University.

10.

Crisp

Olivella

Potter

(2012) Electoral contexts that impede voter coordination. Electoral Studies 31(1): 143–158.

11.

Cross

Pruysers

(2017) Sore losers? The costs of intra-party democracy. Party Politics. DOI: 10.1177/1354068817728216.

12.

Dejaeghere

Dassonneville

(2015) A comparative investigation into the effects of party-system variables on party switching using individual-level data. Party Politics 23(2): 110–123.

13.

Döring

Manow

(2016) Parliaments and governments database (ParlGov): information on parties, elections and cabinets in modern democracies. Available at: http://www.parlgov.org/

14.

Downs

(1957) An economic theory of political action in a democracy. Journal of Political Economy 65(2): 135–150.

15.

Duverger

(1954) Political parties their organization and activity in the modern state, Barbara and Robert North (Trans.). York: Methuen & Company Limited.

16.

Fisher

Denver

(2009) Evaluating the electoral effects of traditional and modern modes of constituency campaigning in Britain 1992–2005. Parliamentary Affairs 62(2): 196–210.

17.

Fisher

Denver

Hands

(2006) The relative electoral impact of central party co-ordination and size of party membership at constituency level. Electoral Studies 25(4): 664–676.

18.

Fisher

(2004) Definition and measurement of tactical voting: the role of rational choice. British Journal of Political Science 34(1): 152–166.

19.

Gallagher

(1991) Proportionality, disproportionality and electoral systems. Electoral Studies 10(1): 33–51.

20.

Gherghina

(2014) Party Organization and Electoral Volatility in Central and Eastern Europe: Enhancing Voter Loyalty. London: Routledge.

21.

Gschwend

(2007) Ticket-splitting and strategic voting under mixed electoral rules: evidence from Germany. European Journal of Political Research 46(1): 1–23.

22.

Hirschman

(1970) Exit, Voice, and Loyalty. Responses to Decline in Firms, Organizations, and States. Cambridge: Harvard University Press.

23.

Holmberg

Oscarsson

(2011) Party leader effects on the vote. In: Aarts

Blais

Schmitt

(eds) Political Leaders and Democratic Elections. Oxford: Oxford University Press, pp. 36–52.

24.

Hooghe

Dassonneville

(2012) Party members as an electoral linking mechanism: an election forecasting model for political parties in Belgium, 1981–2010. Party Politics 20(3): 368–380. DOI: 10.1177/1354068811436053.

25.

Irwin

Van Holsteyn

JJM

(2012) Strategic electoral considerations under proportional representation. Electoral Studies 31(1): 184–191.

26.

Johnston

Pattie

(2003) Do canvassing and campaigning work? Evidence from the 2001 general election in England. British Elections & Parties Review 13(1): 248–273.

27.

Karp

Banducci

Bowler

(2008) Getting out the vote: party mobilization in a comparative perspective. British Journal of Political Science 38(1): 91–112.

28.

Karvonen

(2010) The Personalisation of Politics: A Study of Parliamentary Democracies. Colchester: ECPR Press.

29.

Katz

(1990) Party as linkage: a vestigial function? European Journal of Political Research 18(1): 143–161.

30.

Katz

Mair

(1995) Changing models of party organization and party democracy the emergence of the cartel party. Party Politics 1(1): 5–28.

31.

King

(2002) Leaders’ Personalities and the Outcomes of Democratic Elections. Oxford: Oxford University Press.

32.

Kircheimer

(1966) The transformation of the western European party systems. In: LaPalombara

Weiner

(eds) Political Parties and Political Development. Princeton: Princeton University Press, pp. 177–200.

33.

Laakso

Taagepera

(1979) The “Effective” number of parties: a measure with application to West Europe. Comparative Political Studies 12: 3–27.

34.

Lijphart

(1990) The political consequences of electoral laws, 1945–85. American Political Science Review 84(2): 481–496.

35.

Lisi

(2013) The professionalization of campaigns in recent democracies: the Portuguese case. European Journal of Communication 28: 259–276.

36.

May

(1973) Opinion structure of political parties: The special law of curvilinear disparity. Political Studies 21(2): 135–151.

37.

Norris

(1995) May’s law of curvilinear disparity revisited: leaders, officers, members and voters in British political parties. Party Politics 1(1): 29–47.

38.

Polk

Kölln

(2017) Electoral infidelity: why party members cast defecting votes. European Journal of Political Research 57(2): 539–560.

39.

Ponce

Scarrow

(2016) Which members? Using cross-national surveys to study party membership. Party Politics 22(6): 679–690.

40.

Rabinowitz

Macdonald

(1989) A directional theory of issue voting. American Political Science Review 83(1): 93–121.

41.

Rahat

Sheafer

(2007) The personalization(s) of politics: Israel, 1949–2003. Political Communication 24(1): 65–80.

42.

Rahat

Kenig

Tuttnauer

(2014) Shifting political sands: when politicians, voters and (even) party members are on the move. In: ECPR joint sessions of workshops, University of Salamanca, Spain, 10–15 April 2014.

43.

Scarrow

Gezgor

(2010) Declining memberships, changing members? European political party members in a new era. Party Politics 16(6): 823–843.

44.

Scarrow

(1994) The ‘paradox of enrollment’: assessing the costs and benefits of party memberships. European Journal of Political Research 25(1): 41–60.

45.

Scarrow

(2014) Beyond Party Members: Changing Approaches to Partisan Mobilization. Oxford: Oxford University Press.

46.

Seyd

Whiteley

(2004) British party members: an overview. Party Politics 10(4): 355–366.

47.

Southwell

(2010) The effect of nomination divisiveness on the 2008 presidential election. PS: Political Science & Politics 43(2): 255–258.

48.

Tavits

(2005) The development of stable party support: electoral dynamics in post-communist Europe. American Journal of Political Science 49(2): 283–298.

49.

Tavits

(2012) Organizing for success: party organizational strength and electoral performance in postcommunist Europe. The Journal of Politics 74(1): 83–97.

50.

Van Biezen

Mair

Poguntke

(2012) Going, going,…gone? The decline of party membership in contemporary Europe. European Journal of Political Research 51(1): 24–56.

51.

van Haute

(2015) Joining isn’t everything: exit, voice, and loyalty in party organizations. In: Johnston

Sharman

(eds) Parties and Party Systems: Structure and Context. Vancouver: UBC Press, pp. 184–201.

52.

van Haute

Carty

(2012) Ideological misfits: a distinctive class of party members. Party Politics 18(6): 885–895.

53.

Van Haute

Gauja

(2015) Party Members and Activists. London: Routledge.

54.

Wagner

(2017) Why do party members leave? Parliamentary Affairs 70(2): 344–360.

55.

Wauters

Thijssen

Van Aelst

, et al. (2018) Centralized personalization at the expense of decentralized personalization. The decline of preferential voting in Belgium (2003–2014). Party Politics 24(5): 511–523.

56.

Webb

Bale

Poletti

(2017a) ‘All mouth and no trousers?’ How many Conservative Party members voted for UKIP in 2015–and why did they do so? Politics 37(4): 432–444. DOI: 10.1177/0263395717697344.

57.

Webb

Poletti

Bale

(2017b) So who really does the donkey work in ‘multi-speed membership parties’? Comparing the election campaign activity of party members and party supporters. Electoral Studies 46: 64–74.

58.

Whiteley

Seyd

(1998) The dynamics of party activism in Britain: a spiral of demobilization? British Journal of Political Science 28(1): 113–137.