Party leader survival. Identifying the factors influencing the longevity of party leaders' tenures in the Netherlands

(1)

Party leader survival

Identifying the factors

influencing the longevity

of party leaders’ tenures

in the Netherlands

(2)

Master’s Thesis Comparative and European Politics

Party leader survival

Identifying the factors influencing the longevity of

party leaders’ tenures in the Netherlands

Author: Cees van Dijk (3029743)

Supervisor: Prof. dr. M. H. Leyenaar Second reader: Dr. A. S. Zaslove

Institute: Radboud University Nijmegen School of Management

Political Science

Word count: 34.042

(3)

i

Abstract

This purpose of this study is to identify the factors influencing the longevity of party leaders’ tenures, a subject of growing importance due to the increasing importance of political personalities. Although parties remain to play an important role in present-day liberal democracies, the importance of party leaders is increasing both in the relationship between politics and society and within political parties. Party leadership selection, survival and replacement are topics that receive a lot of attention in the media and it is surprising that little systematic research has been performed to identify which factors influence the longevity of a party leader’s tenure. This study investigates the influence of performance, timing, party organization, predecessors, personal characteristics and party characteristics on party leadership longevity. To determine the impact of these factors, a database of Dutch party leaders since the Second World War is build. Some of the major results are that longevities decrease over time, performance is important for party leader survival, there is a clear difference between male and female party leaders and some party characteristics have a significant effect as well. The results are mixed or inconclusive with regard to democratization of party leader selection methods and to the role of predecessors.

(4)

ii

Preface

The master thesis that is lying in front of you concludes six months of research on the subject of party leadership survival, but above all my Master of Science education in Comparative and European Politics.

It took some time to decide on the subject of this thesis, however, eventually, I found the right subject to immerse myself into. The time taken to decide on the right subject was well worth it. Working on a topic for a longer amount of time requires motivation and perseverance. Therefore, the subject had to align with my personal interests. The process of writing has known its ups and downs, however, in the end it all fell into place.

I would like to thank several people who have contributed to this thesis. My first word of thanks goes to my supervisor at the university, prof. dr. Monique Leyenaar. I would like to thank her for repeatedly reading parts of my thesis and giving valuable comments. Moreover, she gave input on the subject of party leadership and advised me on how to study this subject. Second, I would like to say a word of thanks to dr. André Krouwel, dr. Andrej Zaslove and prof. dr. Bertjan Verbeek. The latter two for giving me advice on interesting and useful literature, the former for several short brainstorm sessions on how to perform this research and where to get my data. Third, I would like to thank my fellow students Jessica Sciarone and Jeroen Romeijn for their advice and comments, especially with regard to the use of the English language. Moreover, they had to put up with my complaining when I got stuck with my thesis. Last but not least, I would like to thank my family for their never-ending support and for motivating me during the period of writing this thesis.

(5)

iii

List of tables

page

2.1 Party functions 9

4.1 Party selection 40

4.2 Overview of case selection 42

4.3 Party leaders and their functions 43

4.4 Number of preference votes (in percentages) 46

4.5 Democratization of party leader selection methods in the Netherlands 50

4.6 Selectorates of current parties and party leaders 51

5.1 Causes of party leader removal (in percentages) 59

5.2 Causes of party leader removal over time 59

5.3 Seat share development in the election before party leader removal 62 5.4 Correlation between seat share development and party leader survival by party 63 5.5 Correlation between government participation and longevity 66

5.6 Democratized party leadership selections 69

5.7 Longevity of party leaders with democratized leadership selection methods 70

5.8 Longevity of party leaders with long-serving predecessors 71

5.9 Longevity of party leaders with predecessors who have been Prime Minister 72

5.10 Longevity of male and female party leaders 73

5.11 Longevity of party leadership tenures by party 75

5.12 Longevity of party leaders of minor and major parties 76

5.13 Longevity of party leaders of left-wing and right-wing parties 77 5.14 Longevity of party leaders of populist and non-populist parties 78

6.1 Summary of the results 80

A1 Party names and abbreviations 94

B1 Dutch party leaders and their tenures 95

D1 Party leaders and their political functions 101

E1 Reasons for party leader removal 105

(8)

vi

List of figures

page

2.1 Different methods of party leadership selection 19

4.1 Different methods of party leadership selection 48

5.1 Longevity of party leadership tenures in months 53

5.2 Party leader survival over months 54

5.3 Number of party leader replacements by year 55

5.4 Overview of categories and factors investigated 60

5.5 Correlation between seat share development and party leader survival over time 63 5.6 Correlation between government participation and longevity over time 66

5.7 Longevity of party leadership tenures over the years 67

5.8 Correlation between grace period and longevity 68

(9)

1

1. Introduction

1.1 Introduction

Party leaders come and go. In the last five years alone there have been eight party leader replacements the Netherlands. The removal and replacement of party leaders are important events for a political party, since a new period under a new party leader commences. However, these periods are not the same for all party leaders. The eight recent party leader replacements occurred after party leadership tenures varying from only 21 months for SP party leader Kant, to an impressive 286 months for SGP party leader Van der Vlies.1_{Furthermore, these eight party} leaders do not only vary in the longevity of their tenure; there are differences in electoral success, government participation, age, gender, selection methods and multiple other factors. Party leadership changes have a large impact on both politics and in society. Therefore, it is striking that almost no research has been done to investigate which factors determine the durability and survival of party leaders. This study sets out to identify these factors using a self-build dataset on Dutch party leadership since the Second World War.

1.2 Societal relevance

Party leader exits and replacements receive a large amount of attention in the media2_{, which is} caused by the (growing) importance of party leaders. Due to processes of personalization, presidentialization and mediatization, party leaders have become more prominent political actors (Garzia, 2011; Karvonen, 2010; McAllister, 2007; Poguntke & Webb, 2005: pp. 10-11, 348-349). In the relationship between politics and citizens party leaders are important actors. They seem to partly take over the role of parties in educating citizens and mobilizing them to participate. Bittner (2011: pp. 113-114) shows that the evaluation of party leaders by citizens is of major importance when deciding who to vote for. Due to loosening ties between citizens and political parties, people ascribe greater importance to political personalities. This means that the resignation and succession of a party leader will have a great influence on the political choices citizens make. Although this provides the party leader with more personal power, it can have a downside as well. Party leaders face a harder task in convincing citizens to vote for them and their party in elections. Although an electoral win can strengthen the position of the party leader, an electoral loss might be attributed to the performance of the party leader to a greater extent than in the past. This might lead to shorter party leadership tenures. Because of the increasing importance of party leaders in elections, their role in election campaigns becomes more

1 A list of party names and abbreviations can be found in table A1 in appendix A.

2 For example, the day after the exit of Sap as party leader of GL, nine national newspapers devoted 35 articles to this event.

(10)

important as well. On the one hand parties are increasingly focusing on their leaders instead of on complex policy programs. They adapt themselves to the logic of the mass media and aim at simplifying complicated issues. On the other hand, the media is inclined to focus on personalities instead of on policy ideas. Therefore, leaders increasingly stand in the spotlights during elections and election campaigns.

Within political parties power resources of party leaders have increased at the expense of other bodies within the party. Important in this regard are the party leader selection methods. Currently, there is a trend towards democratization of these methods (Carty & Blake, 1999; Cross & Blais, 2012; Denham, 2012; LeDuc, 2001; Kenig, 2009b; Punnett, 1993; Wauters, 2013). The influence of grassroots members increases, while the influence of party elites and other party officials decreases. This might be a consequence of the increasing importance of party leaders, since parties want to have a leader who is supported by grassroots members. At the same time it is a cause of the increasing importance of party leaders, since the leader is the one who gets a personal mandate from the party’s supporters. This puts the leader in a central position in which he or she bears more responsibility for the performance of the party than what would otherwise be the case. Other ways in which party leaders can exert influence within the party is by influencing the careers of other politicians, the cooperation between parties and the way a political party is led (Ennser-Jedenastik & Müller, 2011: p. 2). Moreover, the party leader influences the policy positions of the party. An often heard complaint about modern politics is that it lacks stability. Political parties are less likely to change their policy positions drastically under the same leader. Often, such changes occur when new party leaders are in office. The replacement of a party leader offers opportunities to choose a new direction. Such an ideological shift is a form of political instability. Moreover, a high rate of party leader replacement harms political stability as well, since having a party leader for a longer period is in itself a form of political stability.

1.3 Scientific relevance

Despite the importance of party leaders and the societal relevance of party leader survival, little research has been done on this subject. To the best of my knowledge, there are only four systematic studies identifying determinants of party leader longevity. A first article is from 2007 (Bynander & Hart, 2007), showing that this research topic is still in its infancy. The three articles that followed are Andrews and Jackman (2008) (focusing on performance), Ennser-Jedenastik and Müller (2011) (focusing on intra-party factors) and Horiuchi et al. (2013) (focusing on predecessors).

However, on subjects related to party leadership longevity more research is done. Three different topics can be distinguished. The first topic is survival of other political actors than party

(11)

leaders. Studies have focused on the survival of national leaders (Bienen & Walle, 1991; Bueno de Mesquita & Siverson, 1995; Bueno de Mesquita et al., 2002; 2003; Chiozza & Goemans, 2004; Flores & Smith, 2011), autocratic leaders (Albertus & Menaldo, 2012; Gandhi & Przeworski, 2007; Magaloni, 2008), leaders in traditional communities (Gillett, 1973; Miller, 1968), cabinets (Grofman & Roozendaal, 1997; King et al., 1990; Laver, 2003; Warwick, 1992) and cabinet ministers (Berlinski et al., 2007; Fischer et al., 2012; Flores, 2009; Huber & Martinez-Gallardo, 2008). Some explanations for the survival of these political actors might be useful for this thesis. For example, Flores and Smith (2011) look at the internal party rivalry to explain the national leader’s longevity. Fischer et al. (2012) examine, among others, the influence of performance and of personal characteristics such as age and gender on the duration of ministers’ tenures. Such factors might be of influence on party leadership longevity as well.

A second topic concerns the institutional mechanisms regarding the selection, the survival and the replacement of leaders (Courtney, 1995; Cross & Blais, 2012; LeDuc, 2001; Quinn, 2004, 2005; Weller, 1983, 1994). Some of these articles focus on the democratization of party leader and candidate selection methods (Cross & Blais, 2012; LeDuc, 2001; Lundell, 2004; Pennings & Hazan, 2001), which is highly relevant for this study. As mentioned before, party leader selection methods are democratizing, which can have an effect on party leader longevity.

A third large group of studies aims at giving in-depth accounts of leadership selection, survival and replacement. Some of these studies focus at specific events (Alderman, 1998; DenHam & Dorey, 2006; Drucker, 1984; Griffen-Foley, 2001; Heppell & Crines, 2011; Naughtie, 2002), such as the election of Foot as party leader of the UK Labour Party in 1980. Other studies focus at party leadership in one country (Courtney, 1995; Manion, 1985; Punnet, 1993; Quinn, 2005; Wong & Zheng, 2002) or one political party (Bille, 1997; Ceruti, 2008; Jackson, 1975).

The articles focusing specifically on systematic analyses of party leadership longevity will provide input for hypotheses concerning this topic. However, these three related research areas will be helpful as well in doing this study. Moreover, since research on party leadership longevity is still in its infancy, my own ideas on possible important factors can be tested as well.

1.4 Research question, sub questions and methods

This study focuses on the longevity of party leaders’ tenures. The research question is:

“What factors determine the longevity of party leaders’ tenures in a time where individual political leadership is becoming more important?”

To answer this research question several sub questions are formulated. It is important to first describe the theoretical and societal context of party leadership longevity. Therefore, the following three sub questions are posed:

(12)

- What is the role of political parties in modern representative democracies? - What is political leadership and what is party leadership?

- How did the role of political leaders, and in particular that of party leaders, in modern representative democracies develop under the influence of the current developments of personalization, presidentialization and mediatization?

As was mentioned earlier, this study will investigate the longevities of Dutch party leaders. Two sub questions are posed to specifically describe the Dutch context of party leadership:

- Who are the Dutch party leaders?

- How did the role of Dutch party leaders develop under the influence of the current developments of personalization, presidentialization and mediatization?

The third and final set of sub questions are posed to identify the relations between a variety of factors and party leadership longevity:

- Does performance have an influence on party leadership longevity? - Does timing have an influence on party leadership longevity?

- Does party organization have an influence on party leadership longevity? - Do predecessors have an influence on party leadership longevity?

- Do personal characteristics have an influence on party leadership longevity? - Do party characteristics have an influence on party leadership longevity?

These three sets of sub questions are answered by a literature study and by empirical testing. For the empirical testing, Dutch party leaders since the Second World War are selected. All party leaders of significant parties that have won parliamentary seats are included in the analysis. By limiting the scope to one country, detailed and substantive analyses can be performed. At the same time, including leaders of almost all parties makes it possible to make statements about a whole range of political parties. Not only leaders of major parties are included, but also those of minor, new and populist parties. Limiting the scope to party leaders after the Second World War allows comparing party leaders over a substantial period of time, while at the same time keeping the national institutional circumstances largely similar. This research design results in the selection of 92 party leaders whose party leadership has ended. Performing a regression analysis is not possible since some variables vary, for example, most leaders have multiple election results.3_{Therefore, the influence of each factor is tested individually. This has as a disadvantage} that other factors cannot be controlled for. However, it does allow to go into specific circumstances and to give a more detailed analysis.

1.5 Structure

3 Event history analysis would be a possible research method. However, creating the dataset for 92 party leaders would be an endeavour beyond the limits of this thesis, since a case would have to be created for every party leader for every year.

(13)

The structure of this thesis is as follows. Following this introduction, the second chapter addresses the first set of sub questions by discussing the theoretical and societal context of party leadership. Issues such as political leadership, party leadership and societal developments are central concepts in this chapter. The third chapter will present possible explanations of the longevity of party leaders’ tenures. In order to do this, literature on topics related to party leadership longevity and literature focusing on party leadership longevity are reviewed. Based on the literature and on my own ideas, hypotheses are formulated. The fourth chapter addresses party leaders and party leadership in the Netherlands. In this chapter, Dutch party leaders are identified and developments related to the changing role of party leaders in Dutch politics and society are discussed. Thus, this chapter will address the second set of sub questions. The fifth chapter tests the hypotheses on the Dutch party leaders and discusses the results. By doing this, the third and final set of sub questions is answered. The factors influencing party leadership longevity are identified. The thesis will end with a conclusion in which the results are summarized and in which the research question is answered. Furthermore, in this sixth and last chapter the results are placed in a wider societal and scientific context.

2. Party leaders and party leadership

(14)

This theoretical chapter will focus on party leadership and the importance of party leaders in liberal democracies. After this chapter, the first set of sub questions have been answered:

- What is the role of political parties in modern representative democracies? - What is political leadership and what is party leadership?

- How did the role of political leaders, and in particular that of party leaders, in modern representative democracies develop under the influence of the current developments of personalization, presidentialization and mediatization?

There are two major reasons to be interested in and to study party leaders. First, parties play a central role in modern liberal democracies which are based on the responsible party model (Dalton, 1985: p. 270). Therefore, the first paragraph will discuss political parties and their functional importance for democratic political systems. Second, several societal developments point at an increasingly important role for political leaders, one of the most important developments being the personalization of politics. Politics is increasingly centered around individuals, party leaders amongst them. To look at these developments regarding political leadership the second paragraph will first look at what political leadership exactly is, followed by the third paragraph, which is focused on the societal developments and their relation with political leadership. The fourth and final paragraph will limit the scope by focusing on party leaders and their (changing) role in a representative democracy and in their political parties. Party leaders belong to a group of political actors who are of growing importance. Also, their roles in the organization of a political party have changed. It is interesting to see how these developments might relate to the longevity of their tenure. Research specifically aimed at the longevity of party leaders’ tenures will be discussed in the next chapter, this chapter will discuss the broader context of this topic. However, this discussion can often be linked with questions concerning the research question.

2.2 The importance of parties in liberal democracies

Political parties are one of the most important political actors in representative democracies. However, their role differs in each political system. This discussion on the central role of parties will focus on political parties in liberal democracies. In non-democracies, parties play a different role than in democracies. Sometimes only one party is allowed, other times different parties are allowed but do not have any real influence on policies. Although there is no consensus on how to define a liberal democracy, almost all definitions attribute one key feature to it: regular, free and fair elections in which politicians compete for the votes of the citizens to in the end be able to make political decisions (Diamond: 1999, pp. 8-10; Urbinati, 2011: p. 23). Often, there are other features that are attributed to a liberal democracy4_{: examples are pluralism in sources of}

(15)

information, legal equality, cultural freedom for minority groups, rule of law and an accountable military apparatus (Diamond, 1999: pp. 11-12; 2008: p. 22). However, even the thin, electoral, conception of democracy already shows the importance of representatives. As these representatives are almost always organized through political parties, this implies that the political party is a central actor in the liberal democratic political system. Appointing representatives by citizens to make political decisions on their behalf defines a representative democracy. This applies to any type of democracy; presidential, parliamentary or semi-presidential. It is hard to imagine national politics without these representatives organized through political parties. Indeed, there are only a few countries without actual political parties, and even fewer democracies without them (Ware, 1996: p. 1).

Political parties became prominent political actors in the nineteenth century. Two developments attributed to this, the first being the increasing power of political legislatures and the second being the enfranchisement of new groups in society (Scarrow, 2006: p. 17). In parliament the parties started competing more intensely with each other and organized themselves. Through the expansion of the suffrage more societal groups were represented in parliament, which led to more conflicting interests. Parties organized themselves more formally to increase their supporters’ base in society (ibid.). It is hard to define what exactly parties are. There is a wide variety of definitions that stress different key features of parties. The definition used in this study is the following (Ware, 1996: p. 5):

“A political party is an institution that (a) seeks influence in a state, often by attempting to occupy positions in government, and (b) usually consists of more than a single interest in the society and so to some degree attempts to ‘aggregate interests’.”

The first part of this definition stems from the definition of Janda, who focused in his definition of political parties on their endeavor for government power (Beyme, 1985: p. 12). The influence can be a goal in itself, but it can also be a means to be able to influence certain things in society. Pursuing influence can manifest itself in trying to get as many votes as possible and to be substantially present in parliament or in focusing on the realization of policy proposals, but for a lot of parties being part of government is an important means as well (Strøm, 1990: pp. 566-569; Ware, 1996: p. 6). The second part of the definition points at the relation between political parties and interest groups. Both institutions defend interests, but there is a distinction between them. Political parties act as organizations that aggregate the interests of certain groups of citizens, while the main task of interest groups is to articulate specific interests (Beyme, 1985: p. 12). Parties bring together a wide variety of interests in their policy programs. In contrast, interest groups focus only on some particular interests. Another difference between parties and interest groups is that parties operate in a party system where they face competing parties with

(16)

different opinions (Beyme, 1985: p. 13). Most of the time, interest groups do not operate in a system where they have clear countervailing groups.

As mentioned before, nineteenth century parties started to serve as a connection between representatives in parliament and citizens (Dalton, 1985: p. 269). A new form of representative democracy developed: the responsible party model. This model has four important conditions (ibid.: p. 270). First, multiple parties compete for political power through elections. Second, these parties have distinct policy programs between which voters can choose. Third, voters have to have enough information to judge the incumbent parties on their actions. Fourth, parties control the government and the parliament and citizens recognize this. In sum: “The choice of parties provides the electorate with indirect control over the actions of individual legislators and the affairs of government. If the public is satisfied or dissatisfied with government performance, the next election offers the opportunity to implement these evaluations” (ibid.). Present-day Western democracies are based on this model, which highlights the importance of political parties. This description of the model focuses on choices and (electoral) opportunities citizens have with regard to politics. The definition of a political party given in this paper focuses on seeking influence and aggregating interests. To gain influence political parties aim at satisfying the public by representing the interests of certain groups of citizens.

Dalton and Wattenberg (2000) show what this central role of political parties in representative democracies means in practice. They divide the functions of parties in three categories: (1) parties in the electorate, (2) parties as organizations and (3) parties in government. Within these categories multiple party functions are mentioned, several of these functions directly relate to the responsible party model. These functions and their category are shown in table 2.1. This versatile and extensive list of functions of parties in liberal democracies shows how pivotal political parties are. It is important to highlight this, as every party leader derives his or her influence from his or her function as leader of such a political party. However, it is also important to notice that the extent to which parties are the only or dominant actor in fulfilling the abovementioned functions differs per function. One of the current developments is that political leaders have taken on a larger role at the expense of the party organization and other bodies within the party. This makes an even stronger case to look at party leaders as they belong to the most important national political leaders. To examine these processes, it is necessary to first discuss the concept of political leadership. This will be done in the following paragraph. The subsequent paragraph looks at the current developments regarding the increasing importance of political leaders.

Table 2.1: Party functions

(17)

- Simplifying choices for voters: based on the party people know to some extent what to expect with respect to positions on specific issues of individual politicians.

- Educating citizens: parties provide citizens with information about politics.

- Generating symbols of identification and loyalty: parties serve as an “political anchor” in the political system.

- Mobilizing people to participate: parties are important in getting citizens to participate in politics and especially in voting.

Parties as organizations:

- Recruiting political leadership and seeking government offices: parties have to select their party leaders and people who would be able to take on a role in the government.

- Training political elites: Being selected for office or as leader most of the time is the result of a longer period of training and activism within the party.

- Articulating political interests: parties have to bring forward their and their and their supporters’ stands.

- Aggregating political interests: parties have to bring together the interests of different groups to come to a coherent policy program.

Parties in government:

- Creating majorities in government: parties are responsible for creating a government after an election.

- Organizing the government: parties have to ensure cooperation between the individual legislators.

- Implementing policy objectives: parties have to make government policy.

- Organizing dissent and opposition: the parties that form the opposition should propose alternative policy objectives.

- Ensuring responsibility for government actions: with parties in control people know who is responsible.

- Controlling government administration: the parties are present within the government bureaucracy.

- Fostering stability in government: parties do not vary in the extent to which issues and leaders do.

Note: Derived from Dalton & Wattenberg, 2000: pp. 6-10. 2.3 Political leadership

This study looks at the determinants of the longevity of party leaders’ tenures. However, the overarching theme is (political) leadership. Burns (1977: p. 273) states that “leadership over human beings is exercised when persons with certain motives and purposes mobilize, in competition or conflict with others, their own institutional, political, psychological and other resources in such a manner as to arouse, engage, and satisfy the motives of followers.” According to

(18)

him, the power of leaders lies in the relationship between the leader and his followers; the leader and the followers have a common purpose. When applied to party leaders it means that party leaders share motives of party supporters and especially of active party members and use the resources they have at hand to “arouse, engage, and satisfy” these motives.

Here, an important distinction must be made. On the one hand there is the concept of leadership that is associated with occupying the formal position of a leader in a group (positional leadership), on the other hand there is the concept of leadership that is associated with someone who is not officially the leader of a group but still has a large influence on the group (behavioral leadership) (Blondel, 1987: p. 13). With regard to political leadership, positional leadership can be based on different positions that are held by a politician, for example being a monarch, president, prime minister, party leader, chairman or secretary general. This distinction between positional and behavioral leadership shows that someone’s positional power is not the same as that person’s actual power. However, the two concepts are closely interconnected, since leaders often become behavioral leaders because of the formal position they hold (ibid.: p. 14). Having a formal leadership position increases the resources that are at hand to someone. As Burns calls them, such a person probably has more institutional and political (and maybe even psychological) resources at hand to demonstrate leadership.

The subject of this study clearly falls within the concept of positional leadership, the population from which the cases are selected consists of people who formally were the leaders of their party. There might have been other members of the party who demonstrate more behavioral leadership, but the focus is on the formal position of party leader. Besides, it is likely that party leaders are often the ones who exhibit the largest amount of behavioral leadership. They are generally the ones within the party with the highest amount of institutional and political resources. Although the selection of the universe of cases is based on someone’s positional leadership, the behavioural leadership a party leader has does influence his or her chances on staying in power. Someone with a large amount of influence has more resources at hand to ensure that he or she is not removed from his or her position as party leader. This might lead to a longer leadership tenure.

As is stated in Burns’ definition, leadership can be executed using different types of resources. Institutional, political and psychological resources are available to the leader to perform his or her task. These different types of resources are also reflected in theories of leadership. A rough division can be made in theories on leadership based on the position these theories take on emphasizing the role of structure or emphasizing the role of agency (Hartley & Benington, 2011: pp. 208-209). Structural theories focus on how leadership is shaped by the institutional framework. The formal rules of the law determine what a leader is allowed to do and also influence what a leader can do. The practices of an institution also matter for the degree

(19)

of freedom a political leader has, since it is not only based on formal rules. An example of how the institutional framework shapes leadership is the American constitution which defines what a president is allowed to do. Party leaders also have to deal with formal rules and institutional practices. However, (in the Netherlands) party leaders do not occupy a formal organizational function. This does not mean that their power is not limited in regulations, since other party organs do have certain rights a party leader officially cannot interfere with. Parties have practices that influence the position and the possibilities of the party leader as well. Structural theories look at the socially constructed role the leader has in relation to the state and the citizens as well. What lies within the leader’s possibilities is also influenced by what is socially acceptable (ibid.).

Theories that focus on the role of agency specifically look at the characteristics of the leader: his or her qualities, capabilities and leadership style (ibid.). These individual characteristics shape someone’s leadership and are of great importance to party leaders. A party leader can be, among other things, charismatic, domineering, intelligent, short-sighted and communicative. All of these features affect someone’s leadership style and can be of influence on the longevity of party leaders’ tenures. For example, trustworthiness and charisma might have a positive influence on longevity, while other characteristics may have a negative influence. It is important to note that the distinction between theories focusing on structure and theories focusing on agency is not that clear-cut. Most of the theories hold a position somewhere on a continuum between them and they certainly do not position themselves at one or the other end. The longevity of a party leader’s tenure is not influenced solely by individual characteristics, structural factors are important as well. The role of political leaders and party leaders is shaped through structure and individual characteristics. The following paragraph will deal with some structural societal changes which have made the individual political leader more important. 2.4 The increasing importance of political leaders

There are several developments that have influenced and still influence the roles different political actors play. Although political parties continue to play a central role in democratic politics, political leaders seem to have taken on a more prominent role than they have done previously. The role that political leaders play is constantly influenced by societal developments, developments that sometimes lead to institutional changes. In recent history these developments have led to a situation in which political and party leaders have become more important than they were before. Political leadership can be conducted in different political groups, for example in political parties, social movements, and the government. However, political parties have been the main political actors since the nineteenth century. Especially in the 1950s and 1960s, parties had a dominant position in politics. Citizens and parties were tied

(20)

together through ‘social cleavages’, such as religion, class and region (Blondel & Thiébault, 2010: p. 1). These social cleavages ‘determined’ to which social class one belonged and also, which is of greater importance for this study, which political party one adhered. However, these linkages have partly lost their strength in the last decades of the twentieth century. Models of social cleavages are nowadays less able to explain political behavior of citizens. Other theories focusing more on the individual and personalized relationship between citizens and politics have to be taken into account.

Those working on the ‘personalization’ of politics argue that political leaders have become of greater importance. This process can be defined as follows (partly derived from Karvonen (2010: p. 4)):

“Personalization of politics is the process that individual political actors become more prominent at the expense of parties and collective identities in the political arena.”

The literature suggests there are two major causes for this process (McAllister, 2007: p. 572, pp. 579-585; Karvonen, 2010: p. 4). The first is the more structural change of dealignment mentioned above. Citizens are not as loyal to parties and ideologies as they once were. Volatility rates have increased and party membership has decreased. The organizational linkages between parties and citizens have weakened. Nowadays, people do not base their vote on their social group identities to the same extent as once was the case. When group identities and ideological conflict do not determine or strongly influence the political choices people make, there is more room for political leaders to be influential (Karvonen, 2010: p. 4). The individual qualities of (party) leaders become more important for citizens to make political decisions. The second cause is the rise of electronic mass media, especially the television. In the 1950s and 1960s television became the dominant medium through which politics reaches the citizens. The role of television in the political process is especially important during elections and campaigning, but television also obtained a crucial role in transmitting information. However, I would argue that also internet could have an influence on the process of personalization. As Strömbäck (2008: p. 243) argues, the internet is not guided by just one logic. It encompasses a wide variety of methods that are focused on providing users with information. One method that is used is the more direct interaction between political leaders and citizens. This makes the distance between them smaller and reduces the necessity of a party organization through which politicians reach the public. Until today, the influence of television has been larger than that of internet. Television has changed the way politicians communicate with citizens. Television focuses more on personality and less on political programs, this way, the complexity of political issues is reduced (Poguntke & Webb, 2005: p. 15). Furthermore, in television programs party leaders or other high placed politicians are often the main representatives for their party, at the cost of other

(21)

politicians. The other way around, politicians and political parties adapt to the media logic. This is part of the process of mediatization, which is defined as follows (Mazzoleni & Schulz, 1999: p. 250):

“Mediatized politics is politics that has lost its autonomy, has become dependent in its central functions on mass media, and is continuously shaped by interactions with mass media.”

One of the key features of mediatization is that political actors adapt to the requirements of the media (ibid.: pp. 251-252). They act in certain ways to get attention in the media and to use this attention in a way that is the best for the party. One of the consequences is that parties often let party leaders be the person who represents the party in the media. When a party ‘uses’ the media for political goals, it is beneficial if there constantly is one person who represents the party. This way, citizens better recognize which party (and person) stands for which policy. If different people keep talking on behalf of a party, less politically interested citizens might not keep track of who is who. This way, politics is shaped by the interactions with mass media. Another consequence is that politicians will concentrate to a larger extent on symbolism and less on substance. Political issues are simplified in order to fit in with the media logic (Poguntke & Webb, 2005: p. 15). Thus, the mediatization of politics leads to an increasing role of political personalities, party leaders amongst them.

The process of personalization is closely related with another process: the process of presidentialization. This can be defined as follows (ibid.: p. 1):

“Presidentialization [of politics] denominates a process by which regimes are becoming more presidential in their actual practice without, in most cases, changing their formal structure, that is, their regime-type.”

Regimes becoming more presidential entails two processes. The first concerns the increasing power resources for leaders within their political party and within the executive body of the national political system. The second is an increasing focus on leaders in electoral processes (ibid.: p. 5). Presidentialization is a concept that is very closely related to the personalization of politics; presidentialization implies there is to some extent a process of personalization. However, presidentialization puts more emphasis on the institutional and organizational context of politics. Causes of presidentialization are partly the same as those of personalization: the increasing importance of mass communication and the erosion of traditional social cleavage politics foster presidentialization (ibid.: pp. 13-17). Two other causes are the internationalization of politics and the growth of the state (ibid.). The internationalization of politics leads to more international problems that have to be dealt with. Often, it is the national leader who is the main political actor in the international world. The growth of the state means

(22)

governments take on more responsibilities than they once had, which has led to an increased organizational framework. This can result in the centralization of power to coordinate this framework. It can also lead to a situation in which the leader becomes more important with regard to contacts within the cabinet, fewer decisions are made collectively by the cabinet and more by the relevant minister and the leader.

Personalization, mediatization and presidentialization are all developments that lead to a growing importance of individual politicians and especially political leaders. The party leader is one of those political leaders and this is what the next paragraph will focus on. Part of this paragraph will assess to what extent the abovementioned developments influence the role of party leaders in liberal democracies.

2.5 The party leader

The second paragraph focused on the importance of political parties in liberal democracies. The fourth paragraph discussed the growing importance of political leaders, sometimes at the expense of the central role of political parties. The conclusion drawn from these discussions is that it is important to look at party leadership. On the one hand party leaders form one type of political leadership and are becoming more prominent in daily politics. On the other hand they lead a political party, an institution that continues to be a central political actor in liberal democracies. Because party leaders play this central role in contemporary politics, this paragraph will focus on party leaders. The first section discusses who the party leader is and that it is sometimes problematic to identify him or her. Due to large variations between party leaders in different countries, no definition of party leader is given. However, this is done in chapter four, where party leadership in the Netherlands is discussed. The second section focuses on how functions of party leaders are changing due to the developments discussed in the third paragraph. Politics tends to center more and more on political leaders and party leaders are no exception to this. The final section discusses the recent trend of the democratization of party leadership selection. These three sections focus on the role that party leaders play in liberal democracies. Explanations of different longevities of party leaders’ tenures are not given yet, this will be discussed in the following chapter.

2.5.1 Identifying the party leader

The party leader, sometimes referred to as the political leader of a party, is the most important person of a political party. He or she represents the party towards the general public and has to keep an eye on the political unity of the party. However, due to three issues it often is problematic to identify the party leader. The first issue is that being a party leader is not always a legal political function. For example, what party leadership entails in the Netherlands is not

(23)

written down in any laws or regulations. The opposite is the case in some other countries, for example the UK. In those countries party leadership is a legal political function. A second issue is that parties use varying terminology to identify the party leader (So, 2011: p. 5). Party leaders go by different names, for example president, secretary general, leader of X party and parliamentary leader. Sometimes, it is hard to determine what meaning these titles have and who the actual party leader is. For example, in Belgium the president is the party leader, where in the Netherlands he or she is not. A third issue is that the person viewed as the party leader can shift political positions. Many parties do not have strict rules about who the party leader is. This means that in those parties the party leader can for example sometimes be the parliamentary leader while at other times he or she will be Prime Minister. This leads to a situation in which the party leader sometimes is someone in the parliamentary group, sometimes in the cabinet and sometimes in another part of the party organization.

Commonly, the party leader is the person leading a party in times of elections. It is the party leader who represents the party towards the public and who appears most often in the media. Party leaders often compete for high political offices, such as presidency, prime ministership or chancellorship (Kenig, 2009a: p. 434). This means that one of the party leaders of the major parties will often become the new national leader. Once the party leader obtains this function, it is possible he or she will remain to be the party leader. For example, this is the case in the UK and in other Westminster countries such as Ireland, Canada, Australia and New Zealand (Cross & Blais, 2012: p. 131). When a party leader does not succeed in bringing the party to victory, he or she can remain on his or her original position (as long as they are not forced to quit). This can, for example, be the position of a parliamentary group leader or party president. When a party does not win, but is able to participate in a coalition government, the party leader faces different options. He or she can participate in the cabinet, but can also choose to stay on his or her old position and refrain from participating in the cabinet. This is of course only the case in countries with coalition governments. Often, when a party leader participates in the cabinet without being the national leader, he or she will have an important position of a different kind, such as being Deputy Prime Minister. Sometimes, the party leader does become the national leader or does take place in the cabinet, but that the party leadership is transferred to someone else. For example, this is the case when parties have the custom to locate the party leadership in the parliament, as was the case with the Dutch KVP.

2.5.2 Developing party leadership

Political leaders are becoming more important due to the processes of personalization, presidentialization and mediatization. This discussion will focus on the effects this has on the role party leaders play within political parties and in election time. Within political parties there

(24)

is a shift in power towards the party leader at the expense of other bodies within the party (Poguntke & Webb, 2005: pp. 9-11). The democratization of party leadership selection procedures is a development which shows this shift in power. Instead of party activists and party elites, it are often party members or even the whole electorate who are nowadays deciding on leadership issues. Since it is the party leader who gets a personalized mandate from a larger group of people, it is likely that this will also lead to more opportunities for the leader to influence policy (ibid.). After all, he or she is the one who is judged by the public. There is a downside of this personalized mandate for the party leader as well. Although the performance of a party is dependent on more factors than only the performance of the party leader, the party leader is increasingly held accountable due to his increasing importance. As long as the party performs well, he or she often has little to fear. However, when the party is performing poorly, the party leader is held accountable to a larger extent than once was the case. Moreover, when party leaders are selected and removed by a larger group of people, he or she has less chance to influence them. When decisions with regard to party leadership are made by a small party elite, the party leader can personally try to persuade the decision makers.

This relates to the increasing role of party leaders in times of elections. Poguntke and Webb mention three ways in which this happens (2005: pp. 10-11). First, election campaigns are focused more on the party leader and less on the party and the policy program. Political parties present themselves to a greater extent through the leader in times of elections. Mediatization has already been discussed in the former paragraph, one of the key features of this process that political actors (in this case: political parties) adapt themselves to the media logic (Mazzoleni & Schulz, 1999: pp. 251-252). Parties try to simplify complicated issues and to put forward one person to transfer their message. Second, media focus more on party leaders in times of elections. The medium of television concentrates to a greater extent on personalities and less on policy programs (Poguntke & Webb, 2005: p. 15). Focusing on personalities simplifies politics, this way, television makers try to make political information more approachable and comprehensible for viewers. Third, citizens base their votes to a larger extent on party leaders competing in elections and less on other factors such as parties and policy programs (for example: Bean & Mughan, 1989).

Garzia (2011: pp. 705-706) mentions three conditions under which the perceptions of political leaders are especially important for the political behavior of citizens. The first is the absence (or presence) of a dominant climate of opinion. Such a climate is present when people perceive a state of opinion as dominant in society (Barisione, 2009: p. 480). Barisione (ibid.) argues that when this is the case, electoral communication will have little effect. However, Garzia (2011: p. 705) argues that it can be the other way around as well. He states that different climates of opinion ask for different types of leaders and takes the US presidential elections in

(25)

1976 and 2004 as examples. In the 1976 presidential elections citizens favored candidates who were perceived as honest and moral integer, due to the Watergate scandal in the early 1970s. In the 2004 presidential elections citizens favored candidates who were perceived as a strong leader, due to the attacks of 9/11. When such a dominant climate of opinion is absent, the effect of party leader characteristics will be smaller. The second condition is a systemic crisis. In such an exceptional situation citizens often prefer a highly charismatic leader. Examples mentioned by Garzia are the systemic transitions of Soviet Russia and Italy in the 1990s. The third and final condition is a close electoral contest. Party leaders will have a larger effect on the outcome when two or several parties are engaged in an electoral struggle of which the outcome has yet to be determined. When there is one party that is likely to win the election, leadership effects will be smaller. Next to these three conditions, Garzia mentions a larger trend as well that influences the impact of party leader during elections. As was mentioned before in this thesis, dealignment is a cause of personalization of politics. Garzia argues that dealignment leads to an increasing importance of party leader characteristics in determining the outcome of an election (ibid.). As King (2002, p. 41) argues: “… the impact of leaders' personalities and other personal characteristics will be greatest when voters' emotional ties to parties are at their weakest.”

This section has discussed the (increasing) importance of party leaders. One of the issues mentioned is the democratization of party leadership selection. The next section will focus on this issue by discussing what it exactly entails, what is causing it and what it’s consequences are. 2.5.3 Democratization of party leadership selection

In recent years there has been a trend towards less-centralized selection bodies (referred to as “selectorates”) deciding on who the candidates and the leader of the party will be. The focus of this section will be the shift in party leadership selection methods, a shift that is apparent in political parties in multiple countries. A lot has been written on the United Kingdom and Canada (Carty & Blake, 1999; Courtney, 1995; Cross & Blais, 2012; Denham, 2012; LeDuc, 2001; Kenig, 2009b; Punnett, 1993), but other countries are mentioned as cases of democratization as well, for example the US, Israel, Ireland, Japan, Portugal, Belgium and the Netherlands (Cross & Blais, 2012; Kenig, 2009b; LeDuc, 2001; Lisi, 2010; Voerman, 2005; Wauters, 2013). Democratization is not happening in all political parties in these countries, however, within some of the (major) political parties selectorates have become more inclusive. Three issues are addressed in this section. The first issue concerns reasons for parties and party elites to change their selection methods: when does democratization occur? The second issue concerns different practices on party leadership selection: which selection methods are used? The third issue concerns consequences of changing selectorates, especially for the longevity of party leadership tenures.

(26)

There are several explanations for the democratization of party leader selection methods. Cross and Blais (2012: pp. 129-130, p. 145) mention four major causes. The first is an electoral defeat. Parties are more willing to reform when they are not performing well electorally, since reforming the internal organization might seem unnecessary when the party is performing well. A second explanation is that parties are more willing to reform when in opposition. The party has failed in getting into government and is more likely to reform than when it would have been successful. Moreover, not getting into government causes party elites to lose out on influence. Grassroots party members want to interfere in order to turn the tide and to implement new selection rules. The third cause Cross and Blais mention is that new political parties often have parliamentary groups with less influence in the party organization. Therefore, there is less opposition to implement inclusive selection methods. Contributing to this is the aspiration of new parties to differentiate themselves from the older, often more hierarchal, parties. The fourth cause is that parties will adopt new methods when there is another party that has more inclusive methods. Democratization is generally seen as a positive change and parties do not want to fall behind on other parties. This is especially the case when they see the other parties having success with the new selection methods. Thus, contagion of democratizing leadership selection methods is an important factor as well. Scarrow (1999) and Wauters (2013: p. 4) focus on yet another cause: membership decline of political parties. By letting grassroots members in on the leadership decision parties try to make party membership more attractive. Furthermore, making the selectorate more inclusive can also be a means to keeping existing members on board.

It is important to specify what is meant by the democratization of party leader selection methods. Figure 2.1 shows different possibilities of selection bodies (Kenig, 2009a: pp. 435-437). The most inclusive option is to include the whole electorate. An example is the election of Veltroni, the leader of the Democratic Party in Italy. However, this method is hardly ever used in parliamentary democracies. The next option is that party members are allowed to decide on who the new party leader will be. The following possibility is the election by a selected party agency, which can be a congress, assembly, conference or convention. In the Netherlands, for example, it used to a common practice that a congress of delegates would vote for the new leader. Thus, the difference between this option and the former option is that here only party delegates are allowed to vote where the former option allowed participation of all party members. The next selectorate possibility is the parliamentary party group: the Members of Parliament for a party elect a new leader. For a small party in a parliament with relatively little seats this means that the selection of a new leader is in the hand of a few persons. For a larger party in a parliament with a large amount of seats this means that hundreds of members have to participate in the decision. The second to last method is the selection by the party elite. Most of the times, there are no formal rules for this selection method. A candidate will come forth out of informal deliberation.

(27)

The most exclusive selectorate is the single individual. This does not happen very often. It can be the case that the incumbent leader will decide who will be his or her successor or that a party is created by a single person and is completely dependent on that person. For example, this last option occurs in the case of Wilders of the Dutch PVV. He founded the party and decided he would be the leader (and even the sole member). Democratization of party leadership selection occurs when a party changes its selection method in a way that the selectorate becomes more inclusive. This can mean that instead of the party elite, the electorate may decide. But it can also mean that the party elite takes over this task from a single individual. Democratization does not necessarily mean there is a high level of democratic decision-making and that it is necessary for the electorate or all the party members to participate, but only that the level of democracy increases.

Figure 2.1: Different methods of party leadership selection

Note: Derived from Kenig (2009b: pp. 243-244)

The democratization of party leadership selection has several consequences. Overall, more candidates participate in the selection process. However, Kenig (2009b: p. 246) found that the contests are nonetheless less competitive with a high level of democratization than with a low level of democratization. Moreover, democratized party leadership selection methods increase the transparency of the selection, the participation of the members in party politics and the accessibility of the competition (ibid.). McSweeney argues that more inclusive selectorates lead to a different kind of party leaders. The party leaders will be less politically experienced, less socially elitist and less consensual (1999: p. 482). Another consequence of the democratization of leadership selection, but also that of candidates, is that the party will have less influence on who will be chosen. This might lead to less party cohesion, stability and partisan discipline; candidates are less accountable to their colleagues and more to the electorate (Rahat & Hazan,

Sin gle ind ivi du al Party elite Parliamentary party group

Selected party agency Party members

(28)

2001: pp. 312-313). In this sense, it weakens the political party. For this study, it is important to note that the democratization of party leadership may influence the chances of survival for a party leader. When the selectorate is exclusive the incumbent leader is able to influence the members of the selectorate directly (Ennser-Jedenastik & Müller, 2011: pp. 7-8). He or she is to a greater extent able to influence the decision that will be made through informal talks and personal contact. Moreover, a leader or leadership candidate can make promises about future political positions to politicians who are part of the selectorate. Another action that can be taken is placing loyal politicians in advance on positions in which they are able to exert influence about the future career of the incumbent party leader. However, when a leader has been chosen by an inclusive selectorate the claim can be made that he or she has a clear mandate from the electorate, ordinary party members or other party members. This effect will be stronger when a party leader has been chosen by a large majority. This way, he or she can counter criticism from the party elite, which may lengthen his or her tenure.

2.6 Conclusion

The primary goal of this chapter has been to discuss party leadership and the importance of party leaders. First of all, party leaders are in charge of a political party, a political actor that continues to play a pivotal role in liberal democracies (and especially parliamentary democracies). At the same time, processes of personalization, mediatization and presidentialization have increased the importance of political leaders in politics. Being one type of political leaders, party leaders are no exception to this. Therefore, the last paragraph limited the scope to party leaders and what role they play in politics. Moreover, it was discussed how one could identify party leaders, since that is essential to know when discussing their political role. This chapter has refrained from formulating hypotheses, since it was meant to discuss the context of party leadership. The next chapter will discuss the literature on party leadership longevity and develop hypotheses.

3. Party leadership longevity

The previous chapter discussed the concept of party leadership. The importance of party leaders has been examined and the influence of several societal developments on their roles in society and in politics has been discussed. Instead of focusing on the larger picture, this chapter will be

(29)

more specific in its subject: it focuses on possible determinants of party leader survival. First, I give an overview of research related to party leadership longevity, which can be categorized in three groups. Second, I give an overview of research specifically aimed at identifying the factors that are influencing party leadership longevity. Third, based both on the literature presented in these overviews and on my own ideas I discuss expectations concerning possible explanations for the longevity of party leaders’ tenures. Thirteen hypotheses are formulated to test these expectations. This way, the following sub questions are covered.

- Does performance have an influence on party leadership longevity? - Does timing have an influence on party leadership longevity?

- Does party organization have an influence on party leadership longevity? - Do predecessors have an influence on party leadership longevity?

- Do personal characteristics have an influence on party leadership longevity? - Do party characteristics have an influence on party leadership longevity?

After discussing party leadership in the Netherlands in the next chapter, the hypotheses will be tested in the fifth chapter. Thus, in that chapter the sub questions listed above are answered. This chapter will end with a short conclusion.

3.2 Research related to party leadership longevity

Although there are only a few systematic analyses on party leadership longevity, there are numerous studies on closely related topics. This paragraph will focus on these topics. Not all research on these topics will provide input for this thesis, however, there are several interesting arguments made by scholars working on these topics. As was mentioned in the introduction, a distinction can be made between three large research areas. The first category focuses on the survival of other political actors than party leaders, the second on institutional mechanisms concerning leadership selection, survival and replacement, and the third on in-depth analyses of leadership selection, survival and replacement. These different research areas will now be discussed. Even though there is an overlap between these research areas, this categorization creates some order in the literature.

3.2.1 Survival of other political actors

The first category includes studies on survival of other political actors than party leaders. These studies focus on a variety of political actors, however, the focus is often on survival of national leaders. Some scholars investigate leaders of countries with various political systems (Bienen & Walle, 1991; Bueno de Mesquita & Siverson, 1995; Bueno de Mesquita et al., 2002; 2003; Chiozza & Goemans, 2004; Flores & Smith, 2011), others focus on leaders in autocratic systems (Albertus & Menaldo, 2012; Gandhi & Przeworski, 2007; Magaloni, 2008). With regard to national leaders,

Party leader survival. Identifying the factors influencing the longevity of party leaders' tenures in the Netherlands

Party leader survival

Identifying the factors

influencing the longevity

of party leaders’ tenures

in the Netherlands

Master’s Thesis Comparative and European Politics

Party leader survival

Identifying the factors influencing the longevity of

party leaders’ tenures in the Netherlands

Abstract

Preface

Table of contents

List of tables

List of figures

1.

Introduction

2.

Party leaders and party leadership

3.

Party leadership longevity