Abstract
Parliamentary entry on the national level is a crucial achievement for any new party. But its repercussions are not necessarily beneficial. This article assesses the electoral consequences of parliamentary breakthrough by theorizing factors that shape (a) a new party organization’s capacity to cope with pressures generated by parliamentary entry and (b) the relative intensity of the new functional pressures a new party is exposed to after breakthrough. To test our hypotheses derived from these two rationales, we apply multilevel analyses to a new data set covering 135 organizationally new parties that entered national parliament across 17 advanced democracies over nearly five decades (1968–2015). Our findings stress the importance of party organizational characteristics (party origin, time for party building and leadership continuity) for parties’ capacity to sustain electoral support after breakthrough. In contrast, the intensity of functional pressures as generated by government participation immediately after breakthrough does not have significant effects on parties’ performance at the follow-up election.
Introduction: New party performance after parliamentary breakthrough
Only very few new parties that compete at elections ever reach sufficient levels of support to enter national parliament, even in systems with low parliamentary thresholds such as the Netherlands (Krouwel and Lucardie, 2008). While doubtlessly being an important achievement in a party’s development (Lucardie, 2000; Pedersen, 1982), the repercussions of national parliamentary breakthrough are not necessarily positive. This event confronts a new party with additional challenges that go beyond nominating candidates and running campaigns, which can create intra-organizational conflict and can have destabilizing effects (Deschouwer, 2008). Confronted with the new demands linked to parliamentary representation and sometimes even government participation, new entries might fade away quietly after the decline in initial electoral support or spectacularly disintegrate, a less frequent but much more noticed fate (e.g. Aylott, 1995; de Lange and Art, 2011; Lucardie and Ghillebaert, 2008). 1 Why some new parties can exploit parliamentary entry in their favour, while others become the ‘victim of their own success’ is an important question (Van Haute and Pilet, 2006) since only new parties that stay around are likely to broaden the representative offer to citizens or to push mainstream parties to adapt their profile. Nonetheless, cross-national studies rarely address this question across different party families and beyond a restricted number of democracies. 2
This article addresses this gap by examining how new parties that have entered national parliaments in 17 advanced democracies over nearly five decades performed electorally after their big breakthrough and why. As earlier research suggests that structural explanations such as institutional and sociological conditions seem not very helpful to account for short-term electoral effects (Akkerman and de Lange, 2012: 577–578), we propose – mainly building on Harmel and Svåsand (1993) – a party-centred perspective on new party performance instead (see also Bolleyer, 2008; de Lange and Art, 2011). We theorize factors that shape (a) a new party organization’s capacity to cope with new demands related to parliamentary entry (e.g. leadership continuity vs. leadership change) and (b) the relative intensity of these demands (e.g. parliamentary vs. government participation).
We test the hypotheses derived from these two contrasting rationales using a new data set encompassing 135 organizationally new parties that won a seat in their national parliament at least once (irrespective of vote share or programmatic profile) across 17 advanced democracies (1968–2015). Applying multilevel analyses, our findings stress the importance of new parties’ own characteristics rather than the intensity of institutionally generated pressures. We find significant and robust effects of the time the party had to build up its organization prior to its breakthrough and the nature of its origin on the electoral costs of parliamentary breakthrough as well as significant effects of the continuity of its leadership – with the latter not being robust in all democracies studied though. In part, this complements recent findings indicating the importance of party origin to other dimensions of new party success (e.g. party organizational survival) (Beyens et al., 2016; Bolleyer, 2013: 84; Bolleyer and Bytzek, 2013) than the one we examine here. More importantly, the relevance of the timing of breakthrough and of leadership continuity for the electoral costs of parliamentary breakthrough has – to our knowledge – not been examined in any large-N study. We conclude by discussing avenues for future research.
A party-centred perspective on new parties after breakthrough
New parties’ capacity to cope with public office
Although parliamentary entry is likely to create some strains in any new party, the capacity of new parties to cope with this challenge is likely to vary with the newcomers’ capacity to operate effectively in parliament. While much of the literature has been preoccupied with new entries’ ‘ideological newness’ and how their profiles affect their chances of survival or their level of success (e.g. Abedi, 2004; Lucardie, 2000; Megiud, 2007), comparative work on party persistence and decline as well as case study research urges us to consider how parties as organizational actors operate when confronted with new challenges (e.g. Akkerman and de Lange, 2012; Art, 2011; Deschouwer, 2008; Harmel and Svåsand, 1993; Mudde 2007; Rose and Mackie, 1988). We theorize four factors reflecting this rationale.
Building an organization able to sustain itself in face of (often conflicting) internal and external demands takes time. Simultaneously, Harmel and Svåsand have stressed (1993) that new parties often face the challenge that different stages of their development (each of which generating distinct demands) are squeezed in a short period of time. Once a party enters parliament, they argue, the attention of leading actors tends to shift from building and managing their party organization and consolidating a support base to their new responsibilities as parliamentarians (if they join government even as ministers). This shifting of priorities is naturally more problematic early in a party’s life cycle. If a party enters parliament right after its formation, party founders or leaders might simply not have had the time to build a viable extra-parliamentary infrastructure able to keep followers’ loyal, not to speak of the formation of a fully institutionalized party organization in which party followers and representatives consider the survival of the party as an end in itself (de Lange and Art, 2011; Panebianco, 1988). We therefore formulate the following hypothesis:
Classical studies on party development in Western democracies have long argued that any party’s organizational development is shaped by its origin (Duverger, 1981; Panebianco, 1988). Recent work by Bolleyer has distinguished parties formed by individual entrepreneurs from party formations that are promoted by existing societal organizations and stressed this distinction’s importance for new parties’ organizational survival (2013: 39–41; 84). The basic argument goes that thanks to followers’ already established group affiliation, they are inclined to identify with the party and are less likely to defect, even when at times disappointed by their party. If this rationale holds, it has implications for the relative costs of parliamentary breakthrough: rooted parties should be more able to cope with their first parliamentary entry and limit potential electoral losses in case of weak performance than entrepreneurial parties. At the same time, ties to promoter organizations can serve as a recruitment pool for possible party personnel already used to operate in organizational contexts and willing to prioritize the interests of the party rather than their own (Art, 2011). If elected to parliament, those might find it easier to adapt to the demands of operating as part of a parliamentary group, improving the likelihood of a decent performance of rooted formations during their first term.
3
While the loyal support from promoter organizations and time for party building are expected to facilitate institutionalization (i.e. a party’s organizational consolidation), the ability to cope with leadership turnover is generally considered a critical test whether a party as organization has truly institutionalized (Janda, 1980: 19). If organizationally new parties have not yet fully consolidated when entering parliament for the first time, leadership skills but also leadership continuity can be considered crucial to keep the organization together (Mudde, 2007; Pauwels, 2014: 66). Whether parties are new or well established, leadership changes are generally considered destabilizing events (Harmel et al., 1995: 5; see also Bynander and ‘t Hart, 2008). Parties have little incentive to replace a leader who successfully pursues its goals (Ennser-Jedenastik and Müller, 2015: 932), which holds all the more for a new party as ‘few visible actions are taken for it by other than its leader’ (Harmel and Svåsand, 1993: 72). This leader (often taking over one of the party’s first seats or its only seat) shapes the party’s communication with the public as its most visible spokesperson. If the leader embodying a party that just experienced its probably biggest ever success (i.e. winning its first national seats) is replaced or leaves during the party’s first parliamentary term – deliberately (i.e. as result of or regulated by party procedure
4
) or involuntary (i.e. an expression of crisis) – we expect negative effects on the party’s electoral performance at the next election. When electoral loyalties are still fluid, many voters are likely to connect to a new party solely or predominantly through the particular personality of the leader (which differs from active members who also know the organization and its programme). As Litton argued recently, the change of a party’s leader can be assumed to be more crucial for its public recognition than change, for instance, of its programme since ‘for the most part, voters are not knowledgeable, interested or equipped to understand political information in general’ (2015: 714). Consequently, particularly in regard to not very well-known new parties, discontinuity in this central identification figure opens up the possibility that those followers reorient themselves away from the newcomer towards one of its competitors. Taking all these together, we expect the repercussions of leadership discontinuity during a party’s first ever term in parliament to be negative.
While the observation that new parties run the risk of becoming ‘victims of their own success’ is valid for inexperienced newcomers on the left and on the right (e.g. Deschouwer, 2008; Van Haute and Pilet, 2006), the argument has been most prominent in the literature on parties of the new right (e.g. Art, 2011; Carter, 2005; Mudde, 2007). Members of this new party family are portrayed as organizationally weak, electoral vehicles that are dominated by a single (more often than not founding) leader. This leader might be not interested in long-term investments such as party building or even undermine the development of a viable infrastructure to protect his or her position in the party (Bolleyer, 2013; Carter, 2005). Although recent studies rightfully stress the diversity of these parties’ societal and organizational underpinnings (Art, 2011) and their varying capacity to cope with public office – parliament or government (Akkermans and de Lange, 2012; Bolleyer et al., 2012; Pauwels, 2014), the most extreme examples of party disintegration and decline still fall in that group such as New Democracy in Sweden or One Nation Australia. These cases might point to a particular vulnerability of hierarchical and highly centralized parties as linked to an excessive organizational dependency on one single (often founding) leader on whom organizational power is concentrated. A leader’s concern to protect his or her position of (widely unrestricted) power in an organization might, in turn, be difficult to reconcile with managerial demands that confront a leader once represented in parliament for the first time.
5
As far as new right parties tend towards such leadership structures, we can formulate the following hypothesis:
The demands of parliament versus the demands of government
While the demands that parliamentary politics impose on parties (e.g. the intensified need to recruit competent representatives, to assure effective intra-party coordination, to cope with media scrutiny) are similar across the range of advanced democracies, performance pressure is widely considered to be more intense when parties take over government responsibility visible in the takeover of own ministries. While government participation can mean access to valuable resources (e.g. through patronage), this move is mostly considered as detrimental for a new party. Government participation creates the risk of strong negative incumbency effects. Parties holding ministries have policy influence which raises voter expectations towards what the party is likely to deliver (Rose and Mackie, 1983). With very few exceptions, new parties in advanced democracies play the role of the junior partner and minor parties usually find it difficult to shape government policy and implement core preferences against the resistance of their bigger partners and are consequently often punished by the electorate (e.g. O’Malley, 2010). Furthermore, new parties often find it difficult to cope with their new responsibilities professionally, for example, to assure internal communication and, more specifically, regulate conflicts between its representatives in public office and its extra-parliamentary base (Bolleyer, 2008). Finally, new parties often enter parliament with strong anti-establishment credentials with the help of a protest vote (Abedi, 2004). While they can maintain this image after breakthrough by playing the role of a principled opposition party in parliament, they face a considerable risk to alienate followers and thus to deteriorate their support base when being seen to actively collaborate with established elites in government (McDonnell and Newell, 2011). While this is not to say that government responsibility does not offer benefits (such as prestige, influential parliamentary posts, ministerial resources or policy influence), new parties often find it difficult to exploit these benefits, an observation leading to our final hypothesis:
Data and measurement
Defining and specifying new parties
We define ‘newness’ in terms of parties’ organizational development rather than the newness of the issues that they represent. Organizationally new parties still need to build a viable, self-sufficient infrastructure consolidated by a (relatively) stable support base, which makes these parties more vulnerable than and thus distinct from the group of established or ‘organizationally mature’ parties. Parties are classified as new if they are built from scratch (‘newly born’), and if they originate from minor splits of established parties. We further included mergers involving ‘newly borns’ or splits. Successor parties were excluded as well the mergers of old parties that start out from several established party organizations. The same goes for mergers between old and organizationally new parties, whenever old parties in effect ‘swallowed’ the remainders of organizationally new parties, that is, ‘new’ formations in effect are an organizational continuation of established parties (for more details, see Bolleyer and Bytzek, 2013).
The basic vulnerability of new parties as relatively immature organizations is particularly important to a party’s evolution when the latter is confronted with new challenges and pressures as the case when a party enters national parliament. While parliamentary entry constitutes a significant short-term success and opens access to new resources, organizationally new parties are not only less consolidated but also less experienced in holding public office than established parties. As argued earlier, the exposure to new functional pressures (e.g. involvement in law-making) combined with intense media scrutiny can easily have destabilizing effects which is why this event is so important in a party’s life cycle. Reflecting our analytical focus on organizationally new party performance in the decisive phase after their national breakthrough, we look at the subset of organizationally new parties that won seats in their national parliament at least once. 6 We identified new formations meeting these two criteria from 1968 onwards, a period when citizens’ party affiliations underpinning formerly ‘frozen party systems’ in advanced democracies started to de-align (e.g. Bartolini and Mair, 2007). 7 The increasing flexibility of voters’ choices created a ‘window of opportunity’ for new party entry and thus their electoral sustainability, showing in higher numbers of newly formed parties participating in national elections, entering national parliaments and defending a niche on the national level (e.g. Hug, 2001; Mair, 1999). Thus, to pin down the factors that shape new party performance gained particular importance in the period studied.
Data set and dependent variable
Our data collection focused on organizationally new parties operating in established party systems to assure a basic comparability of the challenges new parties face when entering the parliamentary stage (a situation that is substantially different if the majority of rivalling parties are new and inexperienced as well). More specifically, our data set covers new parties that entered their national parliament in 14 European democracies plus Canada, Australia and New Zealand 8 (i.e. won one seat at least once) between 1968 and 2015 irrespective of their vote share (i.e. the first breakthrough election took place in 1968, the last follow-up election in 2015). 9 Our sample covers 135 parties, which is highly inclusive, not only in electoral terms (parties are included irrespective of their national vote share, which in some cases is as low as 0.1% of the national vote) but also when compared to studies that restrict themselves to particular, ideologically defined, new party families.
To capture parties’ performance patterns, we chose the difference between the percentage of the national vote a party won at the first election after breakthrough and the share it won at its breakthrough election divided by the party’s vote share at the breakthrough election as our dependent variable ((VoteElect2-VoteElect1)/VoteElect1). This captures the relative vote loss of parties between two elections and thus measures a party’s vulnerability best since it effectively distinguishes between a party that has won 10% of the vote in its breakthrough election and loses 2% in the upcoming election (in this case our dependent variable is 0.2, a mean vote loss of 20% compared to the vote share at the breakthrough election) and a party that won 3% at its breakthrough election and loses 2% (hence facing a relative vote loss of 0.67, or a 67% vote loss compared to the vote shares at the breakthrough election). Using relative vote loss as dependent variable thus assures comparability between parties. In our sample, 56.3% of the parties lose votes at the election after their parliamentary breakthrough; 8.1% maintain their vote share, whereas only 35.6% gain votes. The average absolute loss among those 76 new parties that are punished electorally is 2.7%, with the relative vote loss being 0.5 (or 50%).
Measuring the explanatory variables
Our analytical framework identified five explanatory variables that are operationalized in the following manner. First, the time for party building is measured through the number of years between the foundation of the party and its parliamentary breakthrough. The average time between foundation and breakthrough is 3.9 years and ranges from 0 (i.e. breakthrough in the year of foundation) to breakthrough 40 years after foundation. Second, to capture the distinction between entrepreneurial versus rooted new parties operationally, we relied on Bolleyer’s study that classified 140 new parties regarding whether their foundation was supported by one or several identifiable promoter organizations or groups or not. The types of promoter organizations were leftist groups, environmental groups, women’s organizations, religious groups, unions/employer organizations, conservative/nationalist/far right groups, regionalist/separatist groups and ethnic groups (2013: 43–53, table 2.2). In essence, new formations qualified as rooted when their foundation was supported by a societal group that predated the actual formation and at that time had in place at least rudimentary organizational infrastructure including voluntary members or affiliates contributing to the organization’s maintenance (for more details see Bolleyer, 2013: 40–41). Of the 135 cases in our data set, 66 (48.9%) parties were entrepreneurs. Third, drawing on available primary sources and earlier case study research, we coded a new variable capturing leadership discontinuity, coding each party in our sample either as 0 if the same leadership stayed in office from its breakthrough until after the follow-up election and as 1 if it experienced leadership change during its first term. We defined the party leader as the party official (or the party leadership as the officials) in charge of managing the party’s relations and communications with the general public (distinct from officials who run a party’s day-to-day operations). 10 While the party leader usually takes over one of the party’s first seats, this does not have to be the case. In parties with collective leadership structures we considered as sufficient for discontinuity that one core figure withdraws or is replaced. 11 This avoided a bias in favour of coding discontinuity in centralized, leader-centred parties had we required in parties with several leadership figures several changes to establish discontinuity rather than just one. Importantly, in line with our theoretical argument we measured discontinuity as such. We did not distinguish changes that resulted from internal crises from a (supposedly) ‘regular’ or ‘neutral’ leadership (re)selection or from the simple implementation of a rotation system change incompatibility rules that force party leaders to step back when winning a seat to avoid a concentration of power. This had considerable methodological advantages since the mere occurrence of change can be more reliably coded (especially when covering minor parties that ceased to exist decades ago) than requiring the coders to distinguish ‘problematic’ leadership discontinuity resulting from some sort of crisis from ‘unproblematic’ or ‘rule-based’ discontinuity. Although extreme cases such as leaders leaving their party in protest (possibly to form a new party of their own) can be easily identified, the scheduling of a new leadership selection process can be the result of intra-organizational difficulties or can trigger them unintentionally. In cases that did not involve severe, publically displayed infighting, this distinction is not very clear-cut and making it requires evaluations easily coloured by the party’s performance after the new leader took over, leading to circularity when trying to assess the electoral costs of leadership discontinuity. All parties were double-coded. Of our 135 parties, 46 (34.1%) experienced leadership discontinuity during their first parliamentary term. Fourth, we coded all organizationally new parties in our sample as new right that were qualified in the literature as anti-immigrant party (Art, 2011), far or extreme right party (Ignazi, 2003) or populist radical right (Mudde, 2007). As van Spanje (2011) points out, despite considerable terminological disagreements, these groups overlap widely, and which parties form members of this party family is empirically little contested among experts. The coding of new right parties was mainly based on the classifications provided in the appendices of Mudde’s (2007) and van Spanje’s (2011). New right parties were coded 1, other parties 0. Thirty parties (22.2%) in our sample qualified as new right. Finally, a party takes over government responsibility when it formally participates in government coalitions visible in the taking over of ministerial posts. If parties held one or more ministerial posts during their first term in parliament they were coded 1, otherwise they were coded 0. The relevant information was provided by the ParlGov Database (Döring and Manow, 2012). Fourteen parties (10.4%) in our sample took over government responsibility (i.e. held ministries) right after their breakthrough election.
Control variables
We add five control variables that – drawing on the extensive literature on new parties – can be expected to shape a new party’s electoral performance. First, we control for the institutional entry barrier (parliamentary threshold) new parties need to cope with to access parliament. The national threshold of inclusion for each country is measured following the logic laid out in Taagepera (2002) capturing the minimum percentage of the national vote necessary to win one seat under the most favourable circumstances at a specific election. 12 In systems, in which this threshold is high, we consider vote losses to be more likely than vote gains at the elections following breakthrough. The relationship between parliamentary threshold and our dependent variable should thus be negative. The average threshold is 0.5% of the national vote with a minimum of 0.1% and a maximum of 1.3%. Second, we control for the provision of state funding access and broadcasting access (standard controls in the study of new party formation and entry, Bolin, 2014; Zons, 2013), as these resources should make it easier for new parties to assure re-election. Each of the two variables takes on the value 1 if direct (electoral and/or organizational) state funding or, respectively, broadcasting access is provided, and 0 otherwise. Each variable has been coded on the party-level and captures the nature of the regimes in place when a party entered parliament. Third, we control for a new party’s ideological distinctiveness. Irrespective of the content of its ideology, a new entry that presents a genuinely new profile makes it harder for mainstream parties to credibly take over its core issues and win voters back than when being confronted with a party that represents, in principle, old issues but criticizes the moderate position of the mainstream parties or criticizes mainstream parties for neglecting them as compared to other issues in their programme (Lucardie, 2000). Consequently, parties with a distinct offer are expected to be less vulnerable to vote losses. To capture newcomers’ relative ideological distinctiveness in their respective party systems, we use a dummy variable based on Abedi’s authoritative classification of ‘challenger’ parties, parties that challenge the status quo in terms of major policy and political system issues (2004: 11–14). We followed his classification to identify those new parties that qualified as distinct. Exceptions were made when parties had been classified as challengers that represented a version of an ideology already occupied by another (in some cases also new) party in the party system. This was appropriate since unlike Abedi, we attempted to capture distinctiveness relative to the offer provided by competitors at the time of breakthrough. More recent formations in our data set not covered by Abedi were classified along the same criteria. Fifty (37.0%) among our 135 cases were classified as ideologically distinct. Finally, we control for gross domestic product (GDP) since economic performance is often considered an important variable that affects the electoral fate of parties. We measured GDP development (in US$1 million per capita) from breakthrough to follow-up election by making use of the Organization for Economic Cooperation and Development data 2015 (doi: 10.1787/dc2f7aec-en, accessed on 17 July 2015). 13
For details on all variables see Table A2 in the online appendix.
Method and findings
Our data includes parties nested within elections nested within countries (76 of the 135 parties in our sample are competing against at least one new party in the same election). Accordingly, we need a modelling strategy that accounts for the fact that our units of analysis are not independent of each other and enables us to model effects on all three levels (Steenbergen and Jones, 2002). We, therefore, estimate a three-way nested random-effect multilevel regression model to explain new party electoral performance after breakthrough. The relative difference between the percentage of the national vote a party won at the first election after breakthrough and the share it won at its breakthrough election is our dependent variable ((VoteElect2-VoteElect1)/VoteElect1). Table 1 shows the results of our model:
Results of random-effect multilevel model.
GDP: gross domestic product.
Note: Coefficients of linear random-effect multilevel model. Standard errors are given within parentheses.
Levels of significance: ***p < 0.01; **p < 0.05; *p < 0.10.
The findings provide support for three of our five hypotheses, all of which stress the importance of party characteristics helping the party to cope with new challenges. The time for party building a party had before breakthrough has a significant positive effect on its electoral performance after breakthrough as expected by H1. Accordingly, the timing of breakthrough affects whether parliamentary entry is likely to have negative rather than positive repercussions for a party’s electoral performance, stressing the need to take into account when a party reaches a qualitatively new stage in its development (Pedersen, 1982). Similarly, our theoretical expectations were confirmed as far as the role of party origin is concerned (H2): Entrepreneurs perform significantly worse than rooted new formations, which underlines recent findings indicating that rooted formations persist longer as organizations and are more likely to re-enter into parliament (Beyens et al., 2016; Bolleyer, 2013; Bolleyer and Bytzek, 2013). 14 Similarly, leadership discontinuity affects a party’ electoral performance negatively (H3) (on a 10% significance level which we deem appropriate due to the low number of cases in the analysis). This finding, however, has to be interpreted carefully, since it is less robust than the other significant variables in our model. 15 The finding suggests that leadership discontinuity might be often but not always negative for a new party’s electoral performance after breakthrough. Finally, one of our control variables – GDP – has a positive effect on electoral performance, indicating the importance of the economic situation for parties’ electoral fortunes, which enhances confidence in our findings. Neither belonging to the new right family (H4), immediate government participation (H5) nor any other control variable has a significant effect on our dependent variable.
The lacking impact of the new right dummy makes sense in light of the significant effect of the party origin variable suggesting it is only particular new right parties that are vulnerable, that is, those lacking any social roots such as the Lijst Pim Fortuyn (Lucardie and Ghillebaert, 2008), not those which are rooted such as the French Front National or the Belgian Vlaams Belang which profit from linkage to nationalist movements (Art, 2011). This is echoed by recent studies on the relative electoral costs government imposes on new right parties which vary with these parties’ organizational strategies that assure the functioning of the organization inside and outside public office (Akkermans and de Lange, 2012). The absence of a significant effect of immediate government entry has similar implications, namely, that government responsibility can be detrimental or beneficial, depending on the party we look at. 16 Even though the functional pressures on new parties clearly increase when taking over such responsibilities and we know that some parties disintegrated as a consequence, some new parties successfully exploited government and benefited from the resources attached to it (Bolleyer et al., 2012). Overall our findings suggest that ‘objective’ differences in the intensity of functional pressures are less crucial to understand the electoral implications of new parties’ national entry than the organizational capacities of a particular party, which are shaped by its origin, the timing of parliamentary entry in a party’s life cycle, that is, the point in its development that a party is confronted with new challenges and pressures and the continuity of its leadership.
To assess the real-world meaning of our findings, Table 2 reports the predicted values of Relative Vote Loss for different values of the three significant independent variables (the values of the respective other independent variables are set to their mean): Time for Party Building, Entrepreneur and Leadership Discontinuity.
Predicted values of relative vote loss.
Table 2 shows that entrepreneurial new parties lose on average 16.1% of their initial vote share at the next election, while rooted parties gain on average 16.5%. Consequently, the societal origins of parties are crucial for their fate at elections. Furthermore, the time for party building a party had prior to its breakthrough affects its electoral support at the following election and thus whether breakthrough is electorally beneficial or rather damaging: While parties that are less than 10 years old tend to lose about 12% of their initial vote share at the next election, older parties tend to win votes. Finally, parties that experience leadership discontinuity lose on average 15% of their initial vote share, whereas parties with leadership continuity gain on average 8.6%. In comparison, time for party building is more important for the electoral performance of a new party after breakthrough than being rooted in societal organizations or leadership continuity in its first parliamentary term. Also, the simultaneous relevance of time for party building before breakthrough and party origin is insightful. While rooted parties might have insufficient time to build a strong organization if they enter parliament very early on and might (despite good starting conditions thanks to their links to promoter organizations) suffer electorally, entrepreneurial parties might be able to build up a viable organization prior to entry if breakthrough success does not come too early.
Conclusion and outlook
Parliamentary entry is a crucial achievement for any new party. After all, most new formations never gain representation in their national parliaments. Yet the electoral consequences of this event are not necessarily beneficial. In this article, we assessed the electoral consequences of entering national parliament for the first time. We presented a party-centred perspective to theorize core factors shaping (a) a party’s organizational capacity to cope with demands coming with entry and (b) the intensity of functional demands a new party is exposed to after its breakthrough. Applying multilevel analyses to a new data set covering organizationally new parties entering their parliament across 17 advanced democracies over nearly five decades revealed the following: New parties that enter parliament quickly after their foundation, entrepreneurial parties formed without societal roots and parties that experience a change in leadership during their first term in parliament tend to lose more votes at the follow-up election than other newcomers. Early government performance did not have the expected detrimental effect, neither were new right parties particularly vulnerable or ideologically distinct parties more resilient in the short term. These findings stress the importance to systematically consider organizational characteristics when studying new party performance, an issue already raised in the literature (e.g. Mudde, 2007), but so far little considered in large-N comparative analyses.
In our theoretical part, we presented reasons why time for party building or party origin should help parties to cope with parliamentary breakthrough better: Entrepreneurs are likely to perform less well after breakthrough than rooted formations because they are less able to recruit loyal and capable members and candidates, especially early on when they have not yet formed a viable infrastructure, while rooted formations can rely on their ties to organized groups serving as a recruitment pool right from the start (Art, 2011). Time for party building serves as a proxy for the relative fluidity of newcomers’ support base as well as of their infrastructure. A very young party that enters parliament right after formation has little time to institutionalize and consolidate support. This implies that the party easily falls into disarray when being confronted with new demands that trigger tensions in the organization at a time when voters and followers still desert the party easily being not (yet) emotionally affiliated to it (de Lange and Art, 2011).
Our findings lead to three broader issues that deserve closer examination in future cross-national comparative research. First, it is crucial to explore further how the timing of major events in a party’s life cycle transforms what constitutes success in the short term (such as electoral success, parliamentary entry or government participation) into a detrimental factor in the long term (Van Haute and Pilet, 2006: 307–310). Overall, our findings suggest that to understand performance patterns, the nature of events and the challenges they pose are less relevant than the relative vulnerability of a party organization exposed to them at the time a particular event occurs. While a range of cross-national studies assess parties’ strategic entry into electoral contests (most of which never gain representation) (Hug, 2001), more attention needs to be paid to parties’ organizational evolution before they participate successfully in elections and enter parliament, a theme dealt with predominantly by case study research. Second, and linked to this, we need to learn more about the repercussions of leadership discontinuity, another dimension difficult to study in large-N designs (but see Litton, 2015). Our findings imply that leadership discontinuity is often but not necessarily negative for a new party’s’ electoral performance after breakthrough. Indeed, the change of leader might improve a party’s image and its functioning if a successor is more popular or skilful than its predecessor. Thus, discontinuity can at times constitute a beneficial renewal (Harmel et al., 1995: 4). That said, major changes in political parties such as change of leadership are usually associated with a party suffering heavy defeat (Cross and Blais, 2012). This suggests that if a party changes its leader after a major success such as entering national parliament (if it is not a mere procedural requirement which is rare), such a step is often linked to some internal difficulty. At the very least our findings imply that change of leadership in crucial stages of party development is likely to be a risky enterprise. Thus, future cross-national work on new party performance needs to explore in greater depth when and how a party manages leadership turnover and how it impacts on a party’s evolution in different political settings. And finally, while the (in)stability of leadership can be an important indication of how a party copes during its first legislative term, we still lack appropriate tools to capture the actual parliamentary performance of newcomers across a wider range of parties and political systems as existing measures are tailored to the study of major parties. 17 To develop such measures will be another important challenge.
Footnotes
Acknowledgements
We are grateful to Tomas Cirhan for his research assistance. We also thank participants of the seminar ‘The Rise (and Fall) of New Parties’ held by André Kaiser at the University of Cologne and Gijs Schumacher who discussed an earlier version of this article at the European Consortium for Political Research General Conference in Glasgow in 2014 for helpful input as well as the three Party Politics referees for their constructive feedback. All remaining errors are ours.
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 research has received funding from the European Research Council (ERC) under the European Union’s Seventh Framework Programme (FP7/2007–13)/ERC grant agreement 335890 STATORG). This support is gratefully acknowledged.
Notes
Supplemental material
Supplementary material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
