Variations in Party Affiliation: Does Form Shape Content? Karina Kosiara-Pedersen Københavns Universitet

Transcription

1 Variations in Party Affiliation: Does Form Shape Content? Karina Kosiara-Pedersen Københavns Universitet Susan Scarrow University of Houston Emilie van Haute Université libre de Bruxelles Paper prepared for presentation at the ECPR Joint Session of Workshops, Workshop 7 Contemporary Meanings of Party Membership, Salamanca, April

2 Variations in Party Affiliation: Does Form Shape Content? Karina Kosiara-Pedersen, Susan Scarrow, Emilie van Haute Introduction The topic of this workshop is the changing nature of party membership; this topic was proposed as a response to changes that are happening in many countries and political systems. Our paper probes the political implications of some of these changes, trying to establish whether and how parties and citizens different approaches to membership affect partisan participation and partisan attitudes. Yet our focus is not only on change, but also on individual and organizational difference, because the term party affiliation has always covered a wide variety of relationships between individuals and organizations. Political parties have historically maintained multiple types of formal and informal links with their closest supporters, including direct membership and membership through collateral organizations. Direct membership has itself varied greatly in both its formality, and its accessibility. At one end of the accessibility spectrum some parties have considered all who attend a party event or donate to be party members; at the other end of the spectrum, parties have very formal procedures for joining that may include regular dues payments, probationary periods, and even endorsements from existing members. Indirect membership varies in the extent to which individual members of the collateral organization have participation rights in their local parties. Finally, contemporary parties have been introducing new affiliation options, including registered party sympathizers, cyber-members and registered primary voters. These new options potentially make it even easier for party supporters to acquire some sort of affiliation status, one that keeps them in regular contact with the party even though they are not formal members. This paper explores the implications of these variations in party affiliation, both at the party and the individual levels. Specifically, at the aggregate level, it asks whether party affiliation rules affect patterns of partisan participation and civic and political attitudes of citizens. At the individual level, the paper investigates how individual affiliation choices affect participatory behaviors. In doing so, this paper also tackles a broader question: Are new forms of individual linkage helping parties to offset the effects of declining participation and activism? After a brief development on declining levels of partisan participation and activism, we first discuss varieties of party affiliation. We outline our expectations about how party affiliation might matter, either by affecting participatory levels and behaviors. We then do a preliminary test of our expectations by combining party-level data on affiliation rules with individual-level survey data, drawing on recent surveys of party members. 2

3 1. Declining levels of partisan participation and activism Today, we have several tools and methods that allow for an estimation of the levels of partisan participation and activism. The data collected with these tools and methods all point toward a decline of participatory behaviors in connection with parties. As regard party membership figures, Katz and Mair (1992a,b) provided the first systematic comparative effort. They concluded that despite sometimes gaining in terms of raw numbers of members, the parties in Western Europe have generally failed to maintain their initial share of the available membership pool (Katz and Mair 1992b). Katz and Mair s data were updated in 2001 and expanded to included Central and Eastern European countries (Mair and van Biezen 2001). The update confirmed the (increasing) decline in the 1990s in all established democracies. Other scholars confirmed the diagnosis (Widfeldt 1995; Scarrow 2000; Webb et al. 2002). Ten years later, research continues to stress that membership decline has deepened over time (Scarrow and Gezgor 2010; Whiteley 2011; van Biezen et al. 2012) and now concerns almost all democracies. These analyses tend to focus on the national level. However, within-country variations are big, and the trends at the party level are far from uniform (Delwit 2011). Similarly, the literature emphasized an overall decline of party activism over time (Whiteley and Seyd 1998; Whiteley 2011). Yet again, this overall decline hides large within-country differences (Table 1). Table 1. % Worked in political party or action group last 12 months Country Party % N Belgium PS Belgium OpenVLD Belgium SP.A Belgium CD&V Czech Republic CSSD Czech Republic ODS Czech Republic KSCM Denmark Socialdemokraterne - the Danish social democrats Denmark SF- Socialistisk Folkeparti - the Socialist People's Party Denmark Venstre, Danmarks Liberale Parti Venstre Germany SPD Germany CDU\CSU Ireland Fine Gael Ireland Fianna Fáil Ireland Labour Israel Balad Israel Kadima Israel Hadash Israel Ha' Likud Netherlands Labour Party

4 Netherlands Christian Democratic Party Netherlands Christian Union Netherlands Social Reformed Party Norway Conservative Party (H) Norway Labour Party (A) Norway Centre Party (SP) Norway Christian Democratic Party (KRF) Portugal Partido Social Democrata (PSD) Portugal Partido Socialista (PS) Spain Partido Popular Spain Partido Socialista Obrero Español (PSOE) Sweden Moderata samlingspartiet Sweden Socialdemokraterna Sweden Centern United Kingdom Labour United Kingdom Conservative Source: ESS-5, Design weight applied; Parties with 10 or more members who responded to survey. Most explanations on declining membership and activism focus on the supply-side and are rooted in the literature on political participation: modernization theories (Norris 2002), institutional explanations (Bartolini 1983; Tan 2000; Norris 2002; Weldon 2006), or short-term effects such as electoral cycles (Bartolini 1983; Widfeldt 1999; Fisher 2000; Fisher et al 2006). Yet, these views are often challenged at the individual level. Furthermore, whilst these supplyside explanations can partly account for the general decline of party membership and activism, they cannot explain why this trend affects some political parties more than others. Interparty differences within countries cannot be the result of national political culture, institutional design or proximity of elections. Therefore, this paper intends to investigate variations in partisan participation by mobilizing both demand-side explanations rooted in the literature on party organizations and supply-side explanations rooted in the literature on participation. More specifically, this paper explores the implications of variations in party affiliation on partisan participation. From a demand-side perspective, at the aggregate level, it asks whether party affiliation rules affect patterns of partisan participation. At the individual level, the paper investigates how individual affiliation choices affect participatory behaviors. 4

5 2. Varieties of Party Affiliation, old and new From a demand-side perspective, party organizations make decisions as to which type of link they want to establish with citizens. They set rules regarding party affiliation, and these rules differ across parties, and within single parties over time (Heidar 1994; Heidar 2006; van Haute 2011). Indeed, one observer concluded, It is less easy to define a member of a political party than of any major organization (von Beyme 1985: 168). Thus, it is no surprise that such relations continue to evolve. In this perspective, the literature on party organizations has provided numerous party models or party types, all referring more or less directly to the way they organize party affiliation. At the basic level, these models and classifications revolve around two questions that can help us to distinguish some of the basic modes of party affiliation: (1) What are the specific rules regarding formal individual membership?, and (2) How diverse are the affiliation options on offer? (1) What are the specific rules regarding formal individual membership? Duverger (1951) developed one of the first distinctions between party organizational types based on the way they organize party affiliation. Duverger systematically questioned homogeneity of party affiliation types and was one of the first to postulate that depending on the type of party organization and party culture, the modes of affiliation differ (Duverger 1951). Following his typology of party organizations, Duverger schematized modes of partisan affiliation as concentric circles from voters to supporters, members, and activists; he hypothesized that only mass parties would be characterized by formal membership. Since then, the way parties organize formal membership has always been crucial in the party organization literature (Katz and Mair 1995; Krouwel 2006), in accordance with the dominance of the mass party model. One of the literal distinctions between formal membership and other types of partisan affiliation is that members have to fill out a form and pay some dues (Duverger 1951). Formal membership is therefore often but not always distinguished by formal enrollment procedures and regular dues payments. Next to dues payment, some parties set additional procedures or barriers to entry, while others make access to formal membership much easier (Scarrow 1996). In that sense, there is variation across parties in the financial and procedural barriers to entry they set for formal membership. But parties are also expected to have lowered their barriers to entry over time (Katz and Mair 1995; Detterbeck 2005; Sandri and Pauwels 2011). (2) How diverse are the affiliation options on offer? Parties can provide multiple types of affiliation to citizens. Beyond traditional direct membership, these types of affiliations may be grouped into three additional basic categories. First, there is traditional indirect affiliation, where an individual is enrolled due to their membership of an organization affiliated with the party. Whereas direct members join the party itself, indirect members join another organization that is affiliated with the party (Panebianco 1988). This could be a non-party organization, such as a trade union, a co-op, or a farmer s organization, or it could be a party-sponsored organization, such as a women s group or youth group. Parties have varied greatly in their treatment of indirect members. In some cases, particularly when the indirect members have joined party organizations, the parties have automatically granted them full direct membership in the party, giving them rights to vote at local party meetings and to become candidates. In other cases, indirect membership has very 5

6 little implication for the individual member of the collateral organization, granting the individual no rights to participate in the activities of the party to which her organization is affiliated. Second, parties may encourage more informal links with citizens, especially parties that eschew formalities such as individual membership rosters and suggested or required dues rates (Duverger 1951). Parties are expected to have increasingly encouraged these more informal modes of affiliation such as guest member, party friend, supporter, or registered sympathizer that enable individuals to sign up with a party without becoming a party member (Katz and Mair 1995; Krouwel 2006). Parties sometimes even formalize informal affiliation by stating in their statutes that they allow for these friends/supporters/etc. and what their role is (and isn t). However, it is also enabled more informally when parties allow non-members to sign up to do, e.g. a specific campaign task. Third, parties may connect with citizens via social media affiliations, creating a category of virtual members, who are the friends that parties or party leaders have on social media such as Facebook and Twitter. Parties may use these virtual members to identify and mobilize supporters, for instance by offering opportunities to participate in party debates and even party votes. However, besides accepting affiliation and the possible option of excluding individuals, parties are not as such in control of this mode of affiliation in the same way as they are with the other categories. These three categories, along with traditional direct membership, describe the various modes of party affiliation, or its breadth. Some parties offer a wide range of modes of affiliation, while others only allow more selective modes of affiliation. And here too, parties are expected to have introduced changes over time in the modes of affiliation they establish with citizens. This emphasizes that parties vary greatly in the way they organize their links with citizens, and this linkage has evolve over time. Our purpose in this paper is to examine how parties use of affiliation options can affect participatory behaviors. Because of this, we deliberately avoid conceptions of affiliation or membership that include behavioral or attitudinal components, such as the strength of individual partisan attachment. Instead, we conceptualize varieties of party affiliation as running along the two above-mentioned dimensions: the barriers to entry for formal membership and the breadth of affiliation options. From a supply-side perspective, at the individual level, it means that citizens have to assess these two dimensions when entering a relationship with a party organization: (1) which costs do they have to pay for formal membership?, and (2) which affiliation option(s) to chose from when offered the choice by the party organization? This paper explores the implications of these variations in party affiliation for partisan participation, both at the party and the individual levels. Specifically, at the aggregate level, it asks whether party affiliation rules affect patterns of partisan participation. At the individual level, the paper investigates how individual affiliation affects participatory behaviors. When looking at party affiliation, we look at our two dimensions of party affiliation: the barriers to entry or costs of affiliation, and the breadth or modes of affiliation. If party membership is a crucial form of linkage between citizens and those who govern, does the form of these linkages matter, or are all forms of linkage equal in their impact? 6

7 3. The Impact of Party Affiliation: Some Expectations We start with two main assumptions about how and why we may find a link between party affiliation and participatory behaviors. These assumptions lead us to expectations about how party affiliation may affect its content. Our aim is to lay out a series of predictions, some of which we will explore using newly gathered data, some of which may be explored at a later date with other data. First, parties set the rules for formal direct party membership. They set up the system of dues (financial rules) and/or additional criteria or conditions to formally join the organization (procedural rules). These rules correspond to barriers to entry set up by parties for individuals who want to formally join the party. They represent the financial and procedural costs of joining for individuals. We assume that affiliation rules act like a sieve, with narrower openings admitting only those who already are likely to be politically active. This argument is often developed in models of party organizations (Katz and Mair 1995; Krouwel 2006) but it has never systematically been empirically tested. Expectation 1a: At the aggregate level, this implies that parties with higher barriers to entry are expected to have higher aggregate levels of party activity and lower levels of party membership. Furthermore, we assume that those (few) willing to pay higher costs should be more disposed to also pay the opportunity costs of engaging in other sorts of partisan activities. This party-level factor is often ignored in the explanatory models of party membership and activism that rely heavily on individual-level factors inspired by models of political participation (Whiteley et al 1994; Whiteley and Seyd 1996; Gallagher et al 2002; Whiteley and Seyd 2002). Expectation 1b: At the individual level, this implies that there is a linear relation between the costs of affiliation and the extent of an individual s level of partisan participation. In other words, in parties with high barriers to entry, those who do join are more likely to get involved in party activities. Aside from any party efforts to mobilize their members, those who pay higher costs to join should be more likely to participate in partisan activities. Second, parties can potentially provide various types of affiliation to citizens: direct and indirect, member, supporter, friend, social media followers, instant members, etc. Individual may therefore face the choice of how to affiliate to the part of their choice. We assume that when parties offer multiple types of affiliation, it increases their ability to mobilize self-identified supporters. Once parties have contact information for supporters, they are able to send them targeted communications. Parties are likely to make special efforts to mobilize their affiliates to engage in partisan activities of all kinds, whether making a donation, attending a political rally, or helping with a local campaign. These efforts may increase the positive or negative psychic rewards of activity, rousing affiliates to help the party cause or to try to defeat unpopular competitors. Expectation 2a: At the aggregate level, this implies that parties offering more types of affiliation should be able to stimulate extra partisan activity. In other words, there should 7

8 be a positive relation between the breadth of affiliation options and aggregate levels of party activity. Furthermore, we assume that when being offered multiple options of affiliation, it makes it easier for affiliates to find ways to partisan participation without having to formally register to the party. Expectation 2b: At the individual level, this implies that all partisan affiliates will be more likely to engage in partisan activities than those who are non-affiliated. This relationship should hold when controlling for other resources normally associated with political participation. However, at the individual level, we expect that the individual s mode of affiliation may be associated with different types of partisan activity. How an individual joins is symptomatic of participation preferences, and is also symptomatic of how the party is likely to engage her. These patterns should be reflected in observed participation levels. Expectation 2c: At the individual level, we therefore expect that affiliates who hold only non-traditional affiliations (social media membership, registration for primary voting) are more likely to engage in low-cost and one-time activities, be that sharing partisan information on social media or attending a campaign rally (Copeland and Römmele 2014). Traditional dues paying members are more likely than non-traditional affiliates to participate in local party meetings and serve as campaign volunteers. We expect these relations to hold when comparing varieties of direct affiliation. We are less certain about the impact of indirect affiliation, i.e. party membership acquired through membership in a collateral organization. On the one hand, those who join a party indirectly may not have paid any direct costs to join, and may not even think of themselves as party members. In this sense, they should be no more disposed than other citizens to engage in partisan activity. On the other, if the party and/or the collateral organization use contact information to try to mobilize these indirect members to participate in election campaigns and other partisan activities, the indirect members may find it both easier and more rewarding to become active partisans. The extent to which organizational linkage affects individual behavior may depend on the type of links between collateral organizations and the parties. In addition to these two assumptions, we also expect interaction effects between the two dimensions of our two dimensions of party affiliation on activism. At the aggregate level, we expect that the relationship between the barriers to entry and party membership figures is mediated by the breadth of affiliation. In a party with a lot of barriers to entry for members, we expect party membership to be lower; especially if the party allows multiple modes of affiliation (we expect higher levels of other affiliates). Conversely, if the party does not allow for multiple modes of affiliation, barriers to entry should affect less the levels of party membership. At the individual level, we expect that the relationship between the costs of affiliation and the level of members activism is mediated by the existence of multiple modes of affiliation. When the costs of affiliation are high, we expect party activism to be higher; but this would be less the case if individual face multiple affiliation options (we expect higher levels of activism in the other types of partisan activities). In the rest of the paper we attempt some preliminary tests of these expectations about the connection between party affiliation and participatory behaviors. 8

9 4. Methods, data, and descriptive statistics To conduct these tests we employ several newly available data sources that allow us to connect information about party affiliation and individual party members and citizens. The first, the Political Party Database (PPDB), provides information about the structures, rules and resources of contemporary political parties. This multi-national collaborative effort is currently gathering data on parties in 23 parliamentary and presidential democracies 1. In this paper, we use PPDB data from the countries and parties covered by our other sources. The second type of new data used here comes from recent party members surveys from individual countries. These surveys are not the results of a coordinated cross-national effort, but were conducted separately in various countries by various national teams. Therefore, the methods of data collection and sampling, the questionnaires etc. differ 2. However, they all provide information about the activities and attitudes of contemporary party members and when keeping to simple measures, the data are fairly comparable. These surveys were collected at roughly the same time as the initial round of data collection for the PPDB, which allows us to connect the party-level and individual-level data. In order to measure barriers to entry or costs of formal affiliation, we use information from the PPDB. The financial costs are measured by the minimum dues level for formal membership (converted in Euros) 3. The procedural costs are measured by two indicators. The first one relates to the ease of joining online. The second indicator relates to the existence of specific restrictions to formal affiliation, such as a minimum age, a probationary period, the need for sponsorship from a current member, the requirement to agree with the party s principles, or the interdiction to belong to another party or political organization. The data shows that parties do vary a lot in terms of the requirements or barriers to entry they establish for formal membership (Table 2). The fees asked for formal membership varies from 1.3 euros per year for Jobbik (Hungary) up to above 150 euros a year for some of the Danish or Dutch parties that apply differentiated rates based on the applicant s level of income. Today, more than ¾ of the parties offer some form of option for potential members to join online. The same proportions apply an exclusivity criterion and forbid applicants to belong to another party organization. Minimum age requirements and the requirement to agree with the party s principles are applied in roughly ¼ of the parties included in our database. Lastly, probationary periods or the requirement to be sponsored by another member are much less frequent (respectively 13.4% and 8.3% of the cases). 1 For information about the countries included and the country team leaders, see the Appendix to this paper. 2 For information on these national party membership surveys, see MAPP web site: 3 Some parties apply different fees dependng on the level of income of the applicant. In that case, we have computed an average fee (highest-lowest fee possible/2). At this stage, we use this absolute measure. In a later stage it will be converted into a relative measure for the purposes of a good comparison (based on the proportion of minimum income). 9

10 Table 2. Barriers to entry / costs of formal affiliation Barrier/Cost N Average 35.6 Dues Lowest Highest % Yes, complete 50.0 Online Membership % Yes, incomplete % No 24.0 Minimum age % Yes Probationary Period % Yes Sponsorship % Yes Agree principles % Yes Exclusivity % Yes These indicators are computed in one single index measuring the barriers to entry / costs of membership. The dues were recoded in two categories (0 = below the average dues level, 1 = above the average). The three categories for joining online were recoded 0 for a full option to join online, 0.5 for a partial option to join online, and 1 for no option to join online. All other procedural requirements were coded 1 when applied and 0 when non existent. The final index is a simple sum of the score of each party on each category. It varies from 0 to 7, and the higher the score on the index, the higher the barriers to entry set by the party, and the higher the costs of formal affiliation for the individual member. It confirms the idea that there is a lot of variation in terms of how parties regulate formal affiliation. Figure 1. Distribution of Parties on the Barriers to Entry Index (0-7)

11 The breadth or modes of affiliation is measured by various indicators assessing the extent to which citizens have other options than traditional and formal direct membership if they want to become affiliated with a party, using again information from the PPDB. In line with our theoretical discussion, we distinguish between four affiliation options, looking at whether parties allow (1) formal direct membership, (2) formal indirect membership, (3) other affiliates (friends, supporters, sympathizers, one-taskers, etc.), and (4) virtual members. These different types of affiliation allow us to distinguish between different groups of individuals (Table 3). All parties in our database allow for formal direct membership. Conversely, formal indirect membership is an exception (3.1%). There is more variation in how parties allow for other types of affiliation than formal membership, be it in the form of friend or supporter (30.7%) or in the possibility for supporters to sign up for specific tasks on the party website (48.7%). As regard the friendship or supportership option, parties tend to be clustered, opposing Belgium, Germany, Hungary, Italy, and Spain where the option is offered by (some) parties to Australia, Canada (as of 2011), Israel, Norway, or Sweden where no parties offer this option. Lastly, parties offer virtual channels of affiliation when they solicit citizens on their websites via surveys (16.2%), forums (41.9%), or the possibility to comment on certain issues (50.0%) Table 3. Breadth or modes of party affiliation 4 % N Formal PM Indirect membership Friend Sign up tasks Virtual survey Virtual forum Virtual comment Here too, these indicators are computed in one single index measuring the breadth or modes of affiliation. Since all parties provide the option of formal membership, it was not included in the index. Indirect membership was recoded 1 when applied; friendship and one-tasking were recoded 0.5 when applied; and the virtual options of affiliation were recoded 0.33 when applied; all these categories were coded 0 when non existent. The final index is a simple sum of the score of each party on each category. It varies from 0 to 3, and the higher the score on the index, the higher the breadth of affiliation to the party, and the more diverse the modes of affiliation for the individuals. It confirms the idea that there is a lot of variation in terms of breadth of affiliation offered by parties or affiliation options for individuals. 4 Formal membership = CR6MBRRUL in PPDB; Indirect membership is A42GROUPMB; Friend is CR7FRIEND; Sign up tasks is A106HELP; virtual survey, forum and comment are respectively A107SRVEY, A108FORUM, A109COMMNT. 11

12 Figure 2. Distribution of Parties on the Breadth of Affiliation Index (0-3) < < 1 1 < < 2 2 and more Figure 3 combines our two indexes and shows the relationship between the two dimensions of party affiliation: the barriers to entry or costs of affiliation, and the breadth or modes of affiliation. This relationship is linear and positive, meaning that parties with higher barriers to entry tend also to offer more affiliation options. This is interesting and perhaps counterintuitive. One could have expected that parties that highly regulate formal membership would regulate other forms of affiliation as well, and would not offer multiple alternatives. Yet, we find the opposite: parties that regulate formal membership more also tend to offer more affiliation options. Figure 3. Barriers to Entry * Breadth of Affiliation 2.5 Breadth / Costs of af.iliation Barriers to entry / Costs of af.iliation 12

13 The dependent variables are formulated at two levels: the individual-, and the party-levels. At the party level, our dependent variables are the aggregate levels of party membership and of party affiliates (expectations 1a and 2a), as well as the overall level of party activism. The level of party membership is measured by the M/E ratio, using MAPP membership data. The level of party affiliates is measured as the A/E ratio, meaning to total number of affiliates in the four categories (members, supporters, corporate members and FB likers) divided by the total number of voters. For the level of overall party activity, we have different data sources. For some parties, we have the share of party members having attended a party meeting within a year (or the like), for others, the share of party members devoting more than 0 hours to party work on average (using data from national party member surveys), and in certain cases we also have the share of members who declared that they have been active in their party in the last 12 months (ESS). In order not to lose too many cases in our analyses, we use one or the other measure depending on what s available for the party under study. In case we had multiple data points for one party, the level of party activism corresponds to the mean between these data points. At the individual level, our dependent variables are the individual level of party activism (expectations 1b and 2b) and the type of activities affiliates engage in (expectation 2c). 5. Analyses: Party Affiliation Differs: So What? Our first set of expectations relate to the effect of barriers to entry or costs of affiliation on participatory behaviors. At the aggregate level, we expect that parties with higher barriers to entry will have lower levels of party membership and higher aggregate levels of party activity. With regard to the impact on membership levels, our expectation does not seem to hold (Figure 4). There is no clear and significant relationship between barriers to entry and overall levels of party membership. If there is a relationship, it is even in the opposite direction: higher barriers to entry seem to have a positive impact on membership ratios, although this relationship is not significant. Figure 4. Barriers to Entry and Membership Figures 7 6 Barriers to Entry M/E Ratio 13

14 With regards to the impact on overall levels of party activism, our expectation seems to hold: higher barriers to entry seem to have a positive impact on party activism, although the relationship is weak. Our expectations at the individual level will be tested at a later stage when ESS data will be integrated in the database. Figure 5. Barriers to Entry and Party Activism 7 6 Barriers to Entry Overall Level of Party Activism Our second set of expectations relate to the effect of breadth or modes of affiliation on participatory behaviors. At the aggregate level, we expect that parties offering more types of affiliation should be able to stimulate extra partisan activity. In other words, there should be a positive relation between the breadth of affiliation options and aggregate levels of party affiliation and party activity. Our results (Figure 6) tend to go along our expectations: the more a party offers alternative modes of affiliation, the higher the number of party affiliates. Figure 6. Breadth of Affiliation and Party Affiliates 14

15 At this stage, we cannot test our expectations at the individual level (expectations 2b and 2c) Lastly, we assumed interaction effects between our two dimensions of party affiliation. At the aggregate level, we expect that the relationship between the barriers to entry and party membership figures is mediated by the breadth of affiliation. In a party with a lot of barriers to entry for members, we expect party membership to be lower; especially if the party allows multiple modes of affiliation (we expect higher levels of other affiliates). Conversely, if the party does not allow for multiple modes of affiliation, barriers to entry should affect less the levels of party membership. Figure 7 tends to confirm the idea that when parties offer alternative modes of affiliation, it negatively impacts party membership. Citizens are offered other types of relation to the party and would not pay the extra costs of formal membership. Figure 7. Breadth or Modes of Affiliation and Membership Figures 15

16 2.5 Breadth of Af.iliation M/E Ratio 6. Conclusion This paper is a first attempt to investigate the relation between how parties organize their affiliation and partisan participation. We know from the literature that patterns of partisan participation vary a lot over time, but also across parties. Party membership is in decline, but this decline does not affect all party organizations in the same proportions. There is also a large variation in terms of levels of activism across parties. Yet, most of the literature has so far focused on macro- or micro-level explanations of these trends. While these explanations shed some light on these trends, they can hardly account for within-country variations across parties. This paper argues that party organization matters, and that taking into account how parties organize may contribute to a better understanding of these variations in partisan participation. More specifically, it argues that how parties organize affiliation is crucial, as it affects the links between citizens and parties. In a sense, party affiliation rules may be considered as opportunity structures that frame individual participatory behaviors. In this paper, we looked at two dimensions of party affiliation rules: the barriers to entry for formal membership or the costs that individuals have to pay to formally become members, and the breadth of affiliation or the different modes of affiliation offered to citizens. Our analyses emphasized that party affiliation rules vary a lot across parties. Parties do set different requirements for formal membership. They also vary a lot as regard the affiliation options that they offer to citizens. We have also shown that the two dimensions of party affiliation rules are related: parties with higher barriers to entry tend also to offer more affiliation options. Lastly, we have attempted to see how variations in party affiliation rules may be related to the variation in partisan participation. Our analyses are still at a preliminary stage and we haven t fully completed our efforts to test whether those variations are related. Yet, our first steps 16

17 confirm that there is at least a possible relationship. These analyses have to be refined on various levels. First, we want to adjust our two main indexes to make sure that they capture the dimensions that we want to study. Second, we will develop our measures of partisan participation by adding information from FB surveys and ESS surveys in order to develop the individual-level analyses. Finally, the database will be expanded with more cases from PPDB project as these become available. These developments should allow for more robust tests of our expectations. These analyses aim in the end to allow us to assess whether party organization matters for partisan participation. If parties are changing their approaches to affiliation, so what? We are resisting the urge to just assume that this is a bad thing for either the parties, or for citizens. Instead, our intention is to establish the likely impacts of this type of change by empirically analyzing the links between party affiliation and participatory behaviors. 17

18 7. Appendix Table 4. Political Party Database Project - (April 2014) Country Coordinators Australia Anika Gauja University of Sydney (Australia) Austria Wolfgang Mueller University of Vienna (Austria) Belgium Kris Deschouwer Free University Brussels (Belgium) Brazil Manuel Alcántara University of Salamanca (Spain) Canada William Cross Carleton University (Canada) Czech Republic Petr Kopecký Leiden University (Netherlands) Denmark Karina Pedersen University of Copenhagen (Denmark) France Elodie Fabre Queen s University Belfast (United Kingdom) Germany Thomas Poguntke Heinrich Heine University (Germany) Hungary Zsolt Eneydi Central European University (Hungary) Ireland David Farrell University College Dublin (Ireland) Israel Gideon Rahat Hebrew University (Israel) Italy Luciano Bardi University of Pisa (Italy) Mexico Manuel Alcántara University of Salamanca (Spain) Netherlands Ruud Koole Leiden University (Netherlands) Norway Elin Allern University of Oslo (Norway) Poland Aleks Szczerbiak University of Sussex (United Kingdom) Portugal Marina Costa-Lobo Lisbon University (Portugal) S. Korea Hannes Mosler Free Unviersity, Berlin (Germany) Spain Tania Verge Maestre University Pompeu Fabra (Spain) Sweden Nicholas Aylott Södertörn University, Stockholm (Sweden) Taiwan Hannes Mosler Free University, Berlin (Germany) United Kingdom Paul Webb University of Sussex (United Kingdom) Database Editor Susan Scarrow Project Coordinators Thomas Poguntke Susan Scarrow Paul Webb Advisory Board Ingrid van Biezen Kenneth Janda Richard Katz Miki Caul Kittilson University of Houston (USA) Heinrich Heine University (Germany) University of Houston (USA) University of Sussex (United Kingdom) Leiden University (Netherlands) Northwestern University (USA) Johns Hopkins University (USA) Arizona State University (USA) 18

19 8. References Bartolini, Stefano (1983), The Membership of Mass Parties: The Social Democratic Experience In Hans Daalder and Peter Mair (eds) Western European Party System. Continuity and Change. Beverly Hills: Sage, pp Copeland, Lauren and Andrea Römmele (2014) Beyond the Base? Political Parties, Citizen Activists, and Digital Media Use in the 2009 German Federal Election Campaign, Journal of Information Technology & Politics forthcoming. Delwit, Pascal (2011), Still in Decline? Party Membership in Europe. In Emilie van Haute (ed.) Party membership in Europe. Exploration into the anthills of party politics. Brussels: Editions de l Université de Bruxelles. Detterbeck, Klaus (2005) Cartel parties in Western Europe?, Party Politics 11(2): Duverger, Maurice (1951) Les partis politiques. Paris: A. Colin. Fisher, Justin (2000), Small Kingdoms and Crumbling Organisations: Examining the Variation in Constituency Party Membership and Resources, British Elections and Parties Review 10: Fisher, Justin, David Denver and Gordon Hands (2006), Party Membership and Campaign Activity in Britain. The Impact of Electoral Performance, Party Politics 12(4): Gallagher, Michael and Michael Marsh (2002) Days of Blue Loyalty. The politics of membership of the Fine Gael party. Dublin: PSAI Press. Heider, Knut (1994) The Polymorphic Nature of Party Membership, European Journal of Political Research 25(1): Heidar, Knut (2006) Party Membership and Participation. In William Crotty and Richard S. Katz (eds) Handbook of Party Politics. London: Sage Publications, pp Katz, Richard S. and Peter Mair (1992a), Membership of political parties in European democracies, , European Journal of Political Research 22(3): Katz, Richard S. and Peter Mair (eds) (1992b) Party organizations: a data handbook on party organizations in western democracies, London: Sage Publications. Katz, Richard S. (1990) Party as linkage: A vestigal function?, European Journal of Political Research 18(1): Katz, Richard S. and Peter Mair (1995) Changing Models of Party Organizations and Party Democracy: The Emergence of the Cartel Party Party Politics 1(1): Krouwel, André (2006) Party Models. In Richard S. Katz and William Crotty (eds) Handbook of Party Politics. London: Sage Publications: Mair, Peter and Ingrid van Biezen (2001), Party Membership in Twenty European Democracies: , Party Politics 7(1): Norris, Pippa (2002) Democratic Phoenix: Reinventing Political Activism. New York: Cambridge University Press. 19

20 Panebianco, Angelo (1988) Political Parties: Organization and Power. Cambridge: Cambridge University Press. Pedersen, Karina (2003) Party membership linkage: The Danish case. København: Institut for Statskundskab, Københavns Universitet. Römmele, Andrea et al (2005) Political Parties and Political Systems: The Concept of Linkage Revisited. Westport: Praeger Publishers. Sandri, Giulia and Teun Pauwels (2011) The Role of Party Members in Belgian and Italian Parties: A Cross-National Analysis. In Emilie van Haute (ed) Party Membership in Europe: Explorations in the Anthills of Party Politics. Brussels: Ed. de l Université de Bruxelles: Scarrow, Susan (2000), Parties without members?. In Russel J. Dalton and Martin P. Wattenberg (eds) Parties Without Partisans. Political Change in Advanced Industrial Democracies. Oxford: Oxford University Press: Scarrow, Susan and Burcu Gezgor (2010), Declining memberships, changing members? European political party members in a new era, Party Politics 16(6): Van Biezen, Ingrid, Peter Mair and Thomas Poguntke (2011), Going, going, gone? The decline of party membership in contemporary Europe, European Journal of Political Research 51(1): Van Haute, Emilie (2009) Adhérer à un parti. Aux sources de la participation politique. Brussels: Editions de l Université de Bruxelles. Van Haute, Emilie (2011) Party Membership: An Understudied Mode of Political Participation. In Emilie van Haute (ed) Party Membership in Europe: Explorations in the Anthills of Party Politics. Brussels: Editions de l Université de Bruxelles: Von Beyme, Klaus (1985) Political Parties in Western Democracies, trans. Eileen Martin. Aldershot UK: Gower Publishing. Webb, Paul, David Farrell and I. Holliday (eds) (2002), Political parties in advanced industrial democracies. Oxford: Oxford University Press. Whiteley, Paul (2011), Is the party over? The decline of party activism and membership across the democratic world, Party Politics 17(1): Whiteley, Paul and Patrick Seyd (1998), The Dynamics of Party Activism in Britain: A Spiral of Demobilization?, British Journal of Political Science 28: Whiteley, Paul and Patrick Seyd (1996) Rationality and Party Activism: Encompassing tests of Alternative Models of Political Participation, European Journal of Political Research 29(2): Whiteley, Paul and Patrick Seyd (2002) High Intensity Participation. The Dynamics of Party Activism in Britain. Ann Arbor: University of Michigan Press. Whiteley, Paul et al (1994) Explaining Party Activism: The Case of the British Conservative Party, British Journal of Political Science 24(1): Widfeldt, Anders (1995), Party Membership and Party Representativeness. In Klingemann H.- D., Fuchs D. (ed), Citizens and the State. Oxford: Oxford University Press:

21 Widfeldt, Anders (1999) Linking Parties with People? Party Membership in Sweden London: Ashgate. 21