DOI: 10.5553/PLC/258999292020002002002

Politics of the Low CountriesAccess_open

Article

Like Mother, Like Daughter?

Linkage Between Local Branches and Their National Party Headquarters in Belgium

Trefwoorden local branches, national party headquarters, linkage, integration, multilevel parties
Auteurs
DOI
Toon PDF Toon volledige grootte
Samenvatting Auteursinformatie Statistiek Citeerwijze
Dit artikel is 31 keer geraadpleegd.
Dit artikel is 20 keer gedownload.
Aanbevolen citeerwijze bij dit artikel
Kristof Steyvers, "Like Mother, Like Daughter?", Politics of the Low Countries, 2, (2020):125-147

    This article scrutinises local-national linkage in Belgium to better understand territorial power relations in multilevel parties. Drawing on a survey of local chairs of national parties, it adopts an innovative, informal and bottom-up approach. The descriptive analysis reveals two central axes in the morphology of linkage: scope (downward support and upward influence) and surplus (benefits versus costs). However, (the valuation of) this interdependence appears as a matter of degree. The explanatory analysis therefore probes into the effect of macro- (between environments), meso- (between parties) and micro- (within parties) level factors. It demonstrates that variance is explained by different parameters. For scope, differences between parties trump those within them. For surplus, specific differences between parties as well as within them matter. The answer to our guiding question is therefore variegated: it depends on for what and for whom.

Dit artikel wordt geciteerd in

    • 1 Introduction

      Local branches of national political parties are often denoted as Janus-faced, combining potentially contradictory roles as autonomous actors in the municipal arena and as agents of their mother party on the ground (Geser, 1999; Lehmbruch, 1979; Müller & Narud, 2013). In practice, their faces often turn to one another as branch politics regularly intersects with its national stem. This article considers that mother-daughter linkage, aiming to describe and explain the relations between local branches and their national party headquarters through the eyes of the chairs of these branches in Belgium.
      Thereby, it taps into two broader streams in the party literature. First, the concession of parties as multilevel organisations instead of monoliths unified around the centre. This focuses on the territorial power relations in parties and the consequences for their organisational functioning (Deschouwer, 2003; Thorlakson, 2006, 2009). The layered approach supplements the default national party prism with a supra- and subnational view emphasising the diversity and dynamism of their mutual relations. When it comes to the subnational, most research has been oriented at the regional level (Detterbeck, 2012; Swenden & Maddens, 2009), while the local tier entered the scholarly limelight to a much lesser extent (Reiser & Vetter, 2011). This is the case despite its 1) potential to bring place-bound interests into decision-making processes within parties; 2) representative role of the national party in the municipal political arena; 3) persistent functionality for wider activities of the party such as providing the legitimacy of a membership base, recruitment and selection of candidates for elections, campaign efforts and electoral socialisation or mobilisation (Copus & Erlingsson, 2012; Egner, Gendzwill, Swianiewicz & Pleschberger, 2018; Pedahzur & Britcha, 2000). For a more comprehensive assessment, the local tier should thus be integrated in the study of multilevel relations within parties.
      Second, better understanding of the linkage between the local and the national levels is key to underpinning the assumption that parties have evolved from hierarchy to stratarchy. The former refers to a centripetal organisation in which subnational units radiate around their national headquarters exercising unilateral steering and control. The latter conceives of parties as agglomerations composed of strata functioning relatively autonomously. Whereas cooperation under a common label remains primeval, the various strata dispose of a considerable discretion. Consequently, power is spread and shared within the party (Mair, 1994). Some see stratarchy as an adaptive strategy towards the cartel party. Therein, national party leaders aim to maintain the benefits of local branches while circumventing the constraints these might impose on them as an organised intra-party opposition. Local elites equally try to avoid intervening in the (often directly elected) national leadership (Katz & Mair, 1995). This entails increasing mutual autonomy and an overall decentralised organisation. Others adopt the logic of the franchise model with the party as a brand having an internal division of labour (Carty, 2004). Such assertions consider autonomy as interdependence instead of independence and integration as a matter of organisational flexibility. Subsequently, different modes of stratarchy exist (Bolleyer, 2012; Poguntke et al., 2016). However, many empirical questions on the relationship between the organisational subunits remain.
      By including the hitherto largely neglected local stratum this article thus seeks to contribute to the study of territorial power relations in contemporary multilevel parties. To describe linkage between local branches and their national headquarters, it will draw on three dimensions often identified in the literature: vertical integration, influence and autonomy. These will be specified in various aspects of organisational life differentiating between the faces and/or functions of parties. To capture the morphology of linkage, the article will also scrutinise associations between these aspects and dimensions. Explicitly building thereon, our contribution further seeks to account for similarities and differences in linkage by taking three levels of explanation into account. Thereto, the results of the RepResent Local Chair Survey 2018 are presented. This survey among the local chairs of national parties in Belgium provides a unique opportunity to study multilevel relations bottom-up (adopting a local perspective) and informally (as perceived through the eyes of chairs as key informants). It consummates the top-down (national) and formal (statutory) perspective predominant in the literature (Poguntke et al., 2016; Scarrow, Webb & Poguntke, 2017). The survey also adds an assessment of the perceived (dis)advantages of linkage.
      Apart from the availability of complementary evidence, Belgium is also a context suited to scrutinising this type of linkage. It has a two-tier local government system (with 581 municipalities and 10 provinces) often situated in the Southern European state tradition of intergovernmental relations. Here, a limited range of municipal functions and a low level of discretion are mediated by direct access of local decision makers to the national centre through dual mandate-holding or party political networks (De Rynck & Wayenberg, 2010). Intra-governmental relations at the municipal level are commonly termed collective (with an indirectly elected collegiate executive). They are part of a wider consensual mode of governance with semi-open list variants of proportional representation and a rather fragmented multiparty system. Although local government is now largely within the constitutional orbit of the Regions, most of these features sustained (Wayenberg, De Rynck, Steyvers & Pilet, 2012). Belgium is also noteworthy for its partocracy and split party system (Deschouwer, Pilet & Van Haute, 2017). Apart from the regionalists and the radical right, all party families have Dutch- and French-speaking siblings. Strictly speaking, there are thus only regional parties (but this contribution will use the more common denominator of the ‘national’ to refer to the party headquarters). National political parties dominate in local elections. With their longer cycle (i.e. 6 years vis-à-vis 5 in regional or federal elections), the latter often display barometric trends. State-wide parties are thus prominent in the local-national nexus although stratarchy tendencies can be discerned (Buelens, Rihoux & Deschouwer, 2008).
      The remainder of this article proceeds as follows. First, the literature on dimensions and aspects of multilevel linkage in political parties is briefly reviewed and applied to the local-national nexus. Here, we will also further consider how this object of research so far manifests in the context of Belgium. Second, pertinent issues in the design, method and data of the RepResent Local Chair Survey 2018 are outlined. Third, the results of the descriptive and explanatory analysis are presented. Fourth, the final section summarises our findings and discusses them against the existing literature.

    • 2 Multilevel Linkage: Unpacking the Local-National Nexus

      Multilevel linkage is likely to be varying between and within parties and/or territorial levels. Such a complex phenomenon thus requires careful unpacking. For analytical grip, we will discuss three dimensions often discerned in the literature and adapt them to the particularities of the local-national nexus (Geser, 1999; Reiser & Vetter, 2011). We argue that the aspects further considered in the empirical sections of the article have the greatest potential to contribute to theory building on multilevel linkage and to complement existing empirical evidence in Belgium.

      2.1 Vertical Integration

      The first dimension is vertical integration. In general, this refers to the extent and mode of organisational linkage between local branches and their national headquarters. Integration can appear in terms of structures, means and goals and comes in formal as well as informal manifestations (Thorlakson, 2006). The territorial reach and depth of a political party is evidently an important precondition for effective linkage (Deschouwer, 2006). Therefore, their presence (i.e. the number of levels the party runs for in elections) and pervasiveness (i.e. the number of subnational units the party runs for in elections and/or holds seats in) should also be considered. Locally, this pertains to the density and participation of national party branches in the municipal arena (Reiser & Vetter, 2011).
      In Belgium, local branches of national parties are nowadays present almost anywhere and dominate the offer in municipal elections. While the traditional pillar parties (Christian-democrats, socialists and liberals) have the strongest local anchorage, their newer counterparts (greens, regionalists and radical left or right) are catching up. Consequently, in many places the partisan offer increasingly mirrors that of the national level. National lists also acquire the bulk of local votes, although differences remain in the relative position of specific parties between regions and vis-à-vis national elections. In addition, non-national lists (i.e. lacking a national name and/or recognition) often complement their national counterparts. Under the surface of this denominator, diverse phenomena gather, of which many actually have implicit organisational links with national parties. Genuinely independent local lists operate in an electoral niche. Non-national lists are less present and successful in the Brussels and in the Flemish than in the Walloon Region. Finally, pre-electoral alliances combine different (types of) lists (Dandoy, Dodeigne, Matagne & Reuchamps, 2013; Steyvers & Heyerick, 2017).
      Once this precondition is met, the organisation of vertical integration enters the fray. Thorlakson (2011, p. 717) ultimately specifies three levels therein. First, the extent and form of legal integration as set out in party statutes. This is exemplified by a common membership (and/or finances), the recognition of subunits as component entities of the party and their representation in the governance of higher levels. Second, the degree and nature of cooperation and coordination between levels in the party, including meeting arrangements or resource and service provision. The latter differs in institutionalisation (distinguishing a hard variant from a softer counterpart) and flow (opposing a downward from an upward direction). Third, the existence of shared identity, values and intentions as an attitudinal component.
      The first level is outside the scope of this article. In general, most parties in Belgium formally adhere to the model of a subscriber democracy with overlapping membership organised at the chapter level and its indirect representation in national governance structures (Scarrow et al., 2017). Our efforts are complementary to the Political Party Database Project comparing the formal aspects of party organisation.1xSee, https://www.politicalpartydb.org. We will thus elaborate empirically on the second level. Here, we scrutinise downward vertical integration as the provision of various resources and services by the national to the local level. This can be considered as a form of support to local branches. It merits closer inspection, as: “we lack comparative data on … aspects of integration, especially ‘soft’ practices of parties, such as provision of … services and cooperative relations” (Thorlakson, 2011, p. 715). It is also in line with the call to: “adding the non-formal ‘tissue’ to the formal ‘skeleton’ in … analyses of multi-level parties” (Moon & Bradberg, 2010, p. 53). Regarding the third level, previous research has shown a strong left-right congruence between the local chapter and its national headquarters in Belgium (Devos, Verlet & Reynaert, 2007).

      2.2 Influence

      Whereas vertical integration highlights organisational and strategic linkage, “it does not describe how power is exercised in an integrated party” (Thorlakson, 2009, p. 160). Therefore, the second dimension traced out is influence. This can be considered as the extent to and way in which local branches have an impact and/or exercise control on the politics of their national headquarters (Fabre, 2010; Reiser & Vetter, 2011). It reflects whether and how the local branch is regarded as an important force in the national party structures (Thorlakson, 2009). It, equally, tends to have formal and informal attributes.
      The first can be found in the ex officio representation of local branches in national governance bodies or other forms of statutory involvement in central decision-making processes. The literature suggests three important intra-party arenas where such influence could occur: 1) the selection and/or functioning of coordinating extra-parliamentary bodies such as the national party executive or leader, 2) the recruitment and/or selection of candidates for national elections and 3) the elaboration and/or formulation of national party policy, including electoral manifestos (Allern & Saglie, 2012; Fabre, 2010; Pruysers, 2018). Statutory frameworks are fundamental guides to the nature of specific parties and likely to constrain partisan action. Still, informal practices “are established alongside official, prescribed, forms or organization and interaction” (Fabre, 2010, p. 351), especially in terms of a somewhat intangible feature as influence practice might deviate from principle. Also, the formal approach is unable to capture how autonomy is shaped by “the understandings of and subsequent applications by the party actors involved” (Moon & Bradberg, 2010, p. 56). Therefore, this article takes an informal view addressing the upward influence local branches perceive to have on their national headquarters in the aforementioned domains of intra-party decision-making.
      Previous research in Belgium suggests some but not extensive influence for local branches. Candidate lists for supralocal elections are compiled in a staged process involving negotiation between multiple levels. An informal list formation committee usually develops a draft that needs approval of all members in the constituency. This process is assorted as national party headquarters tend to be most concerned with the so-called eligible positions. There is more room for manoeuvre by other party actors (including local branches) for the remaining list positions (Vandeleene, De Winter & Baudewyns, 2018). In addition, most parties made the selectorate of their national leader more inclusive, giving all members a vote. Nevertheless, local branches play a role in the proposition of candidates in many parties. Moreover, their executives tend to be composed indirectly, also taking territorial interests into account. Finally, party central offices continue to exercise a disciplinary effect on their representatives in public office (Pilet & Wauters, 2014).

      2.3 Autonomy

      The third dimension of linkage is autonomy. This regards the extent to and the way in which local branches have discretion vis-à-vis their national headquarters. It is the ability to deviate from the centre without interference in domains similar to those of influence such as the selection of local leaders and candidates, the formulation of election manifestos, the choice of a coalition partner or control over the content of ongoing policy processes and voting behaviour in the local council or executive (Thorlakson, 2011). To these more specific party functions, the general articulation of a chosen political discourse and the discretionary performance of own competences by the local chapter can be added as well as overall organisational or financial degrees of freedom (Moon & Bradberg, 2010; Fabre, 2010). Ultimately, these are about the local chapter’s right to exist without much threat of unilateral intervening or disciplinary action by its national headquarters. This is affected by the strength and the form of available sanctions against local branches and the existence of independently operating organs to settle conflicts with the national level (Thorlakson, 2009).
      Previous research in Belgium suggests a substantial discretion for local branches to compose their candidate lists for municipal elections. Among a range of actors potentially involved, supralocal party actors are rather seldom mentioned. Allegedly, they also exercise a rather limited influence therein. Tickets are thus much more often filled and balanced by local party instances and ditto candidates (De Winter, Erzeel, Vandeleene & Wauters, 2013). In addition, the congruence between the local and the national majority or the perceived pressure by supralocal party instances were not very often indicated as factors affecting the formation of a local governing majority. Many local coalitions were interpreted as place-bound demonstrations of minimal connected winning. In addition, they formed because of positive experiences in past coalitions and good personal relations between local actors (Ackaert, De Winter, Dumont & Hennau, 2013).2xThis is the general assertion coming out of aggregated research. As pointed out in a review of a first version, the national party headquarters might keep a close watch on local coalition formation in large cities or be willing to secure the so-called cordon sanitaire against the radical right Vlaams Belang. National parties may also provide guidance in candidate selection or list composition. Autonomy will be less empirically central in this article (apart from some of its aspects in explaining other dimensions of linkage).

      2.4 (Dis)Advantages of Linkage

      Apart from scrutinising linkage as such, this article also adds an assessment of its potential advantages and disadvantages for “we also lack data on key attitudinal orientations” (Thorlakson, 2011, p. 715). This refers to the perceived benefits and costs of the aforementioned connections between local branches and their national headquarters. As it comes to the former, downward support might be crucial for organisational survival, whereas integration helps to defend local interests or mediate in case of conflict with the centre. As to the latter, the goals and the interests of the national headquarters might prevail in case of conflict, and integration could hinder the ability of the chapter to react to local circumstances or make its own strategic choices (Thorlakson, 2011). Here, no previous research on Belgium exists.

    • 3 RepResent Local Chairs Survey 2018

      To estimate local-national linkage, the article draws on the RepResent Local Chairs Survey 2018. Its design, methods and data were already outlined in the introduction to this special. Here, we discuss only the issues pertinent to our research.
      First, the value of the population for our aims. Chairs formally head the local extra-parliamentary party (i.e. central office). Owing to the partocratic nature of decision-making (equally present at the local level) and as a linchpin between the different faces of the local chapter, they also tend to wield considerable informal influence within it even when not holding a local elected mandate (i.e. public office). Moreover, local chairs act as the liaison (i.e. on the ground) between the chapter and its national headquarters in all parties. Therefore, they can be considered as key informants of the internal functioning and external integration of the local chapter into the national party structure (Deschouwer, Rihoux & Verthé, 2013). We thus take their perceptions as a reliable proxy for the local chapter as a whole (our unit of analysis), acknowledging these may deviate from those of others.
      Second, the cross-sectional nature of our analysis. The 2018 endeavour added a set questioning the views of local chairs on the support received from their national headquarters, the influence exercised on it, and the (dis)advantages of the integration of the chapter therein. This precludes a longitudinal comparison with previous versions of the survey on local-national linkage.
      The third issue is our limitation to the data for the Dutch-speaking parties in Belgium (active in the Flemish or Brussels Capital Region) with an acceptable response rate of 31%. As the introduction clarified, this rate is comparable for most of the parties in this linguistic group except for the Flemish liberals. It is much lower and varied among the French-speaking ones (particularly for those left of centre having a strong presence and success in the local arena). Hence, we start from the 487 responses in the Dutch-speaking group, including branches of the following families: radical left (PVDA; N = 13), greens (Groen; N = 78), socialists (sp.a; N = 104), Christian-democrats (CD&V; N = 120), liberals (Open VLD; N = 32), regionalists (N-VA; N = 91) and the radical right (Vlaams Belang; N = 34). In the remainder of this article, the parties will be discussed by their family denomination. The data are also weighted according to the response rate for each family.

    • 4 Local-National Linkage Described

      4.1 Downward Vertical Integration

      Table 1 National party headquarters’ support of local branches3xOriginal question: National party headquarters can support their local branches with different resources and services. Please indicate how much support your local chapter receives for each of the following aspects of its functioning.
      Aspect(Very) Much support (%)Mean (0-5)Standard deviationN
      Training for candidates or staff 39.9 3.1 1.2 348
      Staffing and organisation 38.7 2.9 1.4 360
      Mediation or rule enforcement (in case of conflict) 31.3 2.6 1.5 340
      Policy and programme 30.7 2.8 1.3 356
      Media training 27.0 2.6 1.4 340
      Membership recruitment and retention 19.0 2.3 1.3 351
      Finance 18.0 2.3 1.3 345
      Voter research 18.0 2.1 1.4 322

      First, we focus on downward vertical integration as the support local branches receive from their national headquarters in terms of resources and services. Table 1 reports the extent of that support (on a 0-5 scale) perceived by the local chairs for different aspects of their chapter’s functioning. The table displays the percentage of chairs indicating (very) much support (also the basis for ordering), the mean (for additional ordering) and the standard deviation for each aspect as well as the number of valid observations.
      The table demonstrates that there are only two aspects for which a share of about 40% of all local chairs indicates receiving (very) much support: training for candidates or staff and staffing and organization. The three following items are in the range of around 30% and more. For those remaining, less than 20% of respondents perceive that level of support. Still, these types of organisational linkage are present to a certain extent in most local branches. At least 85% of all chairs perceive at least some support for each of the resources and services listed. The aspects on top and at the bottom of that list are also the ones where, respectively, the lowest and the highest share of local chairs indicate having received no support.4xRespectively a bit more than 3% (training of candidates or staff) and 14% (voter research).
      The measures of central tendency align with the distribution of support and suggest substantial variation within the population of local chairs. Their perception of support can be summarised further. A principal component analysis of the responses shows all items load high on one factor, explaining a bit more than 59% of the variance. On that basis, we are able to develop a reliable (Cronbach’s alfa = 0.899) sum-scale (0-40) expressing the overall national headquarters support for local branches (mean = 20.7; standard deviation = 8.3).

      4.2 Upward Influence

      Table 2 Local chapter influence on national party headquarters functioning5xOriginal question: Local branches can try to influence different aspects of their national party headquarters functioning. Please indicate how much influence your local chapter actually has on each of the following aspects.
      Aspect(Very) Much influence (%)Mean (0-5)Standard deviationN
      The selection of candidates for the regional or federal elections 12.3 2.1 1.2 346
      The composition of the party manifesto for the regional or federal elections 12.3 2.0 1.2 343
      The selection of the national party leader 9.3 1.7 1.3 340
      The acting of party representatives in the regional or federal political bodies 7.1 1.7 1.2 339
      The functioning of the national party executive 6.0 1.4 1.1 329
      The selection of the national party executive 5.5 1.6 1.2 339

      To grasp the upward influence that local branches have on the functioning of their national headquarters, we focus on various intra-party arenas of decision-making. These relate to different faces of parties (Katz & Mair, 1995) with a keen eye on their functions as agents of the state shaping democratic party governance (Müller & Narud, 2013). Table 2 considers the extent of that influence (on a 0-5 scale) perceived by the local chairs for different aspects of the functioning of their national headquarters. The table displays the percentage of chairs indicating (very) much influence, the mean and the standard deviation for each aspect as well as the number of valid observations.
      The table demonstrates that there are only two aspects of their national headquarters functioning where more than 10% of all local chairs indicate exercising (very) much influence upon. Both pertain to party functions in the contestation of supralocal elections: the selection of candidates and the composition of the party manifesto. In other domains, the share of local chairs claiming that level of influence is even lower. Thus, only a limited number of local branches have a rather large amount of influence in these arenas of decision-making. However, this form of integration is also present to a certain extent in most local branches. At least 80% or more of all chairs perceive at least some influence on each of the aspects listed. The second aspect and the one at the bottom of that list are the ones where, respectively, the lowest and the highest shares of local chairs indicate having no influence.6xRespectively a bit less than 8% (composition of the party manifesto) and 20% (functioning of the national party executive).
      The measures of central tendency align with the distribution of influence and suggest substantial variation within the population of local chairs. However, their perception of influence can be further summarised. A principal component analysis of the responses shows all items load high on one factor that explains a bit more than 69% of the variance. On that basis, we are able to develop a reliable (Cronbach’s alfa = 0.910) sum-scale (0-30) expressing the overall influence of local branches on the functioning of their national headquarters (mean = 10.7; standard deviation = 5.9).

      4.3 Advantages and Disadvantages

      Table 3 Advantages and disadvantages of integration into national headquarters7xOriginal question: Integration between the local chapter and the national headquarters can have a number of advantages or disadvantages. To what extent do you agree with the following statements regarding your local chapter? Advantages and disadvantages were presented mixed to the respondents and not pre-categorised as visualised in the table.
      (Very) Much agree (%)Mean (1-5)Standard deviationN
      Advantages
      Support of the national headquarters is crucial for the organisational functioning of our local chapter 62.8 3.6 1.0 342
      Integration helps to defend the interests of our local chapter vis-à-vis the national headquarters 61.9 3.6 0.8 339
      Integration helps to mediate in case of conflicts between our local chapter and the national headquarters 46.7 3.4 0.9 334
      Disadvantages
      Integration implies that in case of conflicts the goals and the interests of national headquarters will prevail over those of our local chapter 25.5 2.9 1.0 337
      Integration into the national headquarters limits our local branches to adequately react to place-bound circumstances 16.4 2.5 1.0 339
      Integration into the national headquarters limits our local chapter to make its own strategic choices 13.6 2.4 0.9 338

      Finally, local chairs could assess a number of advantages or disadvantages of multilevel linkage. These were captured in statements reflecting particular benefits or costs for the local chapter of connections with their national headquarters. Table 3 regards the extent of agreement (on a 1-5 scale) expressed by the local chairs with each of the propositions. The table includes the percentage of chairs indicating (very) much agreement, the mean and the standard deviation for each statement as well as the number of valid observations.
      The table shows there is (very) much agreement among local chairs on the idea that integration supports the organisational functioning of the local chapter and helps to defend its interests towards the national headquarters (more than 60% for both). Although still a substantial number, a minority of local chairs (very) much agrees with multilevel linkage as a means of conflict mediation. Additional inspection of the data demonstrates that at most a bit more than 15% of all chairs (very) much disagree with these advantages. The measures of central tendency align with the distribution of agreement on the items and suggest some variation within the population of local chairs. However, their extent of agreement with the advantages listed can be further summarised. A principal component analysis of the responses shows all items load high on one factor that explains a bit less than 64% of the variance. On that basis, we are able to develop a reliable (Cronbach’s alfa = 0.713) sum-scale (3-15) expressing the overall agreement on the advantages of linkage (mean = 10.6; standard deviation = 2.1).
      Local chairs appear to agree to a lesser extent with some of the alleged disadvantages of multilevel linkage. One out of four respondents (very) much does so with the assertion that the goals and the interests of the national headquarters will prevail in case of conflict. Less than 20% agrees to the same extent that integration limits the discretion of local branches to respond to place-bound circumstances or to make their own strategic choices. In turn, at least 39% of respondents (very) much disagree with these disadvantages. Similar patterns as for the advantages can be discerned regarding the measures of central tendency. Here also, agreement with the disadvantages can be captured in one underlying factor explaining a bit less than 63% of the variance. Again, this gives way to a relatively reliable (Cronbach’s alfa = 0.696) sum-scale (3-15), articulating the overall agreement on the disadvantages of linkage (mean = 7.8; standard deviation = 2.3).
      Further analysis shows a significantly negative relationship between overall agreement on advantages and disadvantages (Pearson’s correlation coefficient = –0.259; p < 0.01). The more local chairs agree on the benefits of multilevel linkage, the less they do on its costs. To discern the balance, we have subtracted the score on the scale of agreement with the disadvantages from those of that with the advantages. This renders a scale (from –12 to 12) of the surplus of multilevel linkage (mean = 2.7; standard deviation = 3.5). A bit more than 11% of local chairs see the overall benefits of integration equal the costs. The same share considers the latter outweigh the former. A bit more than one-fifth of the population thus actually discerns a levelling out or a deficit in multilevel linkage. The remaining 78% identifies a surplus.

      4.4 Local-National Linkage Condensed: Scope and Surplus

      Many aspects of local-national linkage can thus be summarised in three composite measures: the extent of 1) support local branches receive from their national headquarters; 2) influence they exercise upon it and 3) surplus discerned out of local-national integration. Although the literature emphasises the multidimensionality of linkage in general (Fabre, 2010; Moon & Bradberg, 2010; Thorlakson, 2009, 2011), we explored the relationship between our more particular indicators. The result can be found in Table 4. It displays the strength and the significance of Pearson’s correlation coefficient for each pair of measures. To compare the latter more properly, standardised versions were calculated.
      The table shows the three measures are significantly and positively correlated. The more support received from the national party headquarters, the more influence exercised upon it and/or surplus discerned by the chairs of local branches. For support and influence, the coefficient is above the rule-of-thumb 0.400 threshold. This hints at a latent measure of linkage confirmed by a factor analysis. Both support and influence load high (0.866) on a principal component explaining 75% of the variance found. This ultimately allows us to calculate a relatively reliable (Cronbach’s alfa = 0.642) scale (0-100) expressing the scope of local-national linkage (mean = 38.9; standard deviation = 15.4). Together with the standardised one on surplus (mean = 61.4; standard deviation = 14.5), this measure will be used as a dependent variable in the subsequent explanatory section of the article.

      Table 4 Correlations between standardised composite measures of local-national linkage
      InfluenceSurplus
      Support 0.500*** 0.268***
      Influence 0.270***

      Significance: *** p < 0.001

    • 5 Local-National Linkage Explained

      To account for the apparent variation in linkage, three levels of explanation are derived from the literature and accommodated to the local-national nexus in Belgium. All take the local branches as their unit of analysis. They will be denoted as macro (affecting the branches of all parties in a designated context), meso (differentiating branches of one or more parties from those of others) or micro (opposing some branches to others in the same party or else). Owing to the exploratory nature of this article and for reasons of parsimony, we formulate similar hypotheses for scope and surplus for each level.

      5.1 Macro Level: Institutional Environment

      The macro level refers to differences in the institutional environment in which a local chapter functions. These pertain to opportunities and constraints providing an incentive structure for particular courses of action. So far, research has emphasised cross-national variation in the design and structure of the state in terms of the distribution of competences and allocation of resources between levels. The more policymaking requires multilevel coordination and cooperation, the more incentives for linkage (Deschouwer, 2003; Fabre, 2010; Reiser & Vetter, 2011; Thorlakson, 2009). As mentioned in the introduction, this is the default in the intergovernmental relations of the federal Belgium. Given the previous focus on the regional-national dyad, the literature has been less concerned with institutional differences within states such as those at the level of subnational polities. These might in turn affect the need to sustain local-national relations, particularly where regions are the main point of reference for the municipal tier, as is the case in Belgium.
      Our data allow one to compare local branches of the Dutch-speaking parties in the Flemish and the Brussels Capital Region. The difference between both most pertinent to our research is in the regional party system (Dandoy, Amjahad & Schrobiltgen, 2013). In Brussels, Dutch-speaking parties clearly represent a linguistic minority. Their local branches often face difficulties in autonomously acquiring a critical mass of voters in municipal elections. Therefore, they tend to form pre-electoral alliances with the French-speaking counterparts of their party family (who can target the linguistic majority). This strategy holds for all parties but the regionalists and the radical right. Owing to the specific institutional arrangements of the Capital Region, this is beneficial to both language groups and more than just a means to pick up additional votes or seats. To guarantee the bilingual representation in the executive, municipalities can decide to appoint one additional alderman of a designated language group, if all others come from the other. Invariably, this means a Dutch speaker (an additional incentive to include candidates from that language group in the alliance). The complex Brussels’ reality is thus typically the frame of reference. Therefore, we expect that Dutch-speaking local branches in Brussels are less oriented towards their own national party headquarters but more to their French-speaking regional counterpart. Thus, we surmise: Local branches in the Flemish Region have a larger scope of linkage (H1a) and surplus of integration (H1b) than those in the Brussels Capital Region.

      5.2 Meso Level: Party Genes

      The meso level involves differences between various types of parties. These are often punctuated by the so-called genetic approach arguing contemporary parties are moulded by their historical template through processes of path dependency (Krouwel, 2006; Panebianco, 1988). This could also affect local-national linkage in two ways. On the one hand, integration might reflect the characteristics of the party type dominant in a certain era, juxtaposing the branches of traditional descendants of the mass type (where local branches were foundational and integration organisationally paramount) with their newer counterparts. On the other, ideologically defined party families could oppose the members of one or more of these to those of other(s) in terms of how local branches relate to the national headquarters (Bardi, Bartolini & Trechsel, 2014; Bolleyer, 2012). Hence, for some parties we might expect more and for others less linkage. In this article, we focus most explicitly on the former.
      For one thing, socialist parties traditionally thrived on an extensive membership base structured in branches combined with a central leadership characteristic of the mass party type. On that footing, we might still expect stronger integration of local branches into the national headquarters (Thorlakson, 2009). Thus, we argue: Local branches of the socialist family have a larger scope (H2a) of linkage and surplus of integration (H2b) than those of others. For another, we also surmise more linkage for the local branches of the newer regionalist party. As a successor of a dissolved ancestor with some place-bound rooting, the national headquarters deliberately invested in the further development of its branches to facilitate the electoral sustainability and organisational institutionalisation of the party (Beyens, Deschouwer, van Haute & Verthé, 2017). The literature suggests this strategy of territorial penetration aligns with a more centralised party structure (Panebianco, 1988). Hence, we expect: Local branches of the regionalist family have a larger scope (H3a) of linkage and surplus of integration (H3b) than those of others.
      Alternatively, and each for its own reasons, traditional parties such as the liberals (with a cadre party blueprint as a caucus of notables) or the Christian-democrats (owing to the principles of subsidiarity) and newer parties such as the greens (as anti-establishment outfits emphasising basic democracy) could be envisaged to demonstrate weaker integration (Delwit, Pilet & Van haute, 2011; Reiser & Vetter, 2011). Therefore, we hypothesise: Local branches of a particular set of party families (i.e. liberals, Christian-democrats and greens) have a smaller scope of linkage (H4a) and surplus of integration (H4b) than those of others. We include the local branches of the remaining national parties without theoretical expectations. These concern new parties at opposing fringes of the ideological spectrum, i.e. the purifying radical left and the prophetic radical right (Beyens, Lucardie & Deschouwer, 2016). In both parties, the municipal tier is the relative laggard of their multilevel organisation.

      5.3 Micro Level: Chapter Autonomy and Strength

      The micro level pertains to differences between specific local branches potentially also within the same national party. Broadly speaking, “large, rich or important units may well have a level of independence and power not shared by smaller or less vital units” (Carty, 2004, p. 10). The former units likely identify less scope and surplus of linkage. Here we scrutinise the impact of three factors specifying that idea: the financial autonomy, organisational and electoral strength of the local chapter. The first builds on “the assumption … that there is a trade-off between … integration and party autonomy” (Thorlakson, 2011, p. 715). The capacity of a party level to act without constraint or interference in terms of policy or interests is likely conditioned by its discretion in means, particularly those financial (Fabre, 2010). More financial autonomy then implies less linkage. For local branches, this autonomy refers to the share of their income coming from their own means vis-à-vis that transferred by their national headquarters. Based on the distribution of that share in our data, we oppose local branches that are fully autonomous to those that are only partial.8xA majority of just under 57% of local branches are fully financial autonomous from their national headquarters. Among the remaining that share partial autonomy, only less than 10% derive more than half of their income from the national headquarters. For financial autonomy, we argue: Local branches with partial financial autonomy have a larger scope of linkage (H5a) and surplus of integration (H5b) than those with full.
      The second factor concerns the institutionalisation of the party branch as an organisation. While this might alternatively refer to its duration, paid staff or activities, this article focuses on the size of the local chapter executive and its membership (Thorlakson, 2011). Both could be seen as indicators of the permanence and strength of the local chapter central office and on the ground. To compare the local branches in our data set and given the distribution of the raw data, we categorised the extent of the executive and the membership in three equal groups relative to the whole of the sample obtained: low, medium and high.9xBased on self-reporting by the local chairs and cut-off points for three equal groups. For the local branch executive, this means a number of 7 or less (low), 8-14 (medium) and 15 or more (high). For members, it implies a number of 56 or less (low), 57-149 (medium) and 150 or more (high). We assume that locally less institutionalised branches might benefit more from being integrated into their national headquarters and will see this bearing over the cost. The others are sufficiently viable to act autonomously. Therefore, in terms of organisational strength, we argue that: Local branches with a low number of executives will have a larger scope of linkage (H6a) and surplus of integration (H6b) than those with a medium or high. Also: Local branches with a low number of members will have a larger scope of linkage (H7a) and surplus of integration (H7b) than those with a medium or high.
      Third, and finally, linkage can be affected by the electoral strength of the party branch (Thorlakson, 2009). Here also, two indicators in the electoral life cycle of the local chapter are scrutinised. On the one hand, representation, as the number of seats the chapter obtained in the preceding municipal elections. Analogous to the reasoning in the previous paragraph, this number is categorised as low, medium or high.10xBased on self-reporting by the local chairs and cut-off points for three equal groups: a number of 3 or less (low), 4-8 (medium) and 9 or more (high). It should be noted that less than 5% of all local branches did not obtain any seats. Local branches with a low number of seats are likely to depend more on the support of their national headquarters and to perceive this as advantageous. Thus, local branches with a low number of seats will have a larger scope of linkage (H8a) and surplus of integration (H8b) than those with a medium or high. On the other hand, whether or not the local chapter participated in the governing majority as a consequence of those elections. Given the partisan nature of local decision-making along the majority-opposition divide in Belgium (Steyvers & Heyerick, 2017), we expect the branches of the latter to turn more to their national headquarters to compensate their relatively marginalised local position. Therefore, local branches in municipal opposition will have a larger scope of linkage (H9a) and surplus of integration (H9b) than those partaking in the governing majority.

      5.4 Results: Variegated for Scope and Surplus

      To empirically scrutinise similarities and differences in local-national linkage along the lines outlined in the foregoing, we conducted a number of multivariate analyses (OLS regressions). Therein, scope and surplus of local-national linkage are the respective dependent variables. In turn, local chapter characteristics function as independent variables in line with their hypothesised effects. The regressions help us to estimate the direction, significance and strength of these effects if other variables are simultaneously taken into consideration. For variables with more than two categories, the theoretically most pertinent is taken as a reference. Regarding the second level of explanation, we opt for the socialists. This allows one to compare the effect of membership of other families with that of a traditional party along the centralised mass model template expected to show a larger scope and surplus of linkage (especially vis-à-vis that of designated others). At the third level, a low number of executives, members or seats are the respective references. For each of those categories, we expected a larger scope and surplus of linkage in comparison with medium or high.
      The results of the regressions can be found in Table 5. It reports beta coefficients for specific variables and indicates their level of significance through asterisks. The overall explanatory power of our models is expressed by the adjusted R2 at the bottom. To gauge what the variables at the respective levels add thereto, each model represents a step in the analysis. Only the third is discussed more extensively, as this considers the conjoint effect of all variables and levels.

      Table 5 Multivariate analysis of scope and surplus of local-national linkage (OLS regressions)
      Dependent variables
      Local national-linkage
      Independent variablesScopeSurplus
      Local chapter characteristic Model 1Model 2Model 3Model 1Model 2Model 3
      Institutional environment Beta Beta Beta Beta Beta Beta
      Flemish region 0.089 0.006 -0.054 0.088 0.048 0.149*
      Party family
      Ref. cat. = Socialists
      Radical left 0.165** 0.099 0.036 -0.083
      Greens 0.095 0.072 -0.041 -0.117
      Christian-democrats 0.029 0.068 0.033 0.079
      Liberals -0.164* -0.173* -0.147* -0.168*
      Regionalists 0.297*** 0.368*** 0.098 0.089
      Radical right 0.235*** 0.199* 0.047 -0.013
      Financial autonomy
      Partial -0.003 0.079
      Organizational strength
      Executive
      Ref. cat. = low
      Medium -0.084 -0.084
      High 0.020 0.070
      Members
      Ref. cat. = low
      Medium 0.134 0.056
      High 0.086 0.111
      Electoral strength
      Seats
      Ref. cat. = low
      Medium -0.025 -0.158*
      High -0.088 -0.019
      Government participation
      Opposition -0.055 0.252***
      Adjusted R2 0.004 0.177*** 0.197*** 0.005 0.031* 0.148***

      Significance: * p < 0.05; ** p < 0.01 and *** p < 0.001

      The table demonstrates that after including variables of the three levels, the overall explanatory power of our model for the scope of linkage is significant but relatively weak (accounting for a bit less than 20% of the variance). Moreover, the effects of particular variables are all situated at the meso level of party families. In comparison with local branches of the socialist party, those of the liberals display a significantly smaller scope of linkage, while those of the regionalists and the radical right a significantly larger. This brings nuance to H2a. As an established party along a historic template, the socialists differ significantly from only one other traditional party and some newer ones (all to its ideological right). In comparison with the latter, the effects are even opposite to those expected. Belonging to the regionalist party is the strongest predictor for an increase in scope. The strategy of territorial penetration appears to translate into a more integrated outlook among local chairs (as support received from and influence exercised upon the national party headquarters). Being part of the radical right is the second strongest predictor. This could suggest that the hitherto lower level of presence and success in the local arena is compensated by developing a larger scope of national integration.
      Neither the institutional environment (macro level) nor differences within parties in terms of financial autonomy, organisational or electoral strength (micro level) matter significantly. It can be noted that adding the third level alters the effect of the variables at the second (inflating that of some whilst deflating that of the party families at the extremes of the ideological spectrum which are likely to have less organisational and electoral strength at the local level). After the multivariate analyses for scope, H2a (although holding for only one pair of party families), H3a and H4a (equally for only one pair) can thus not be refuted.
      In addition, the table shows that after including variables of the three levels, the overall explanatory power of our model for the surplus of linkage is equally significant but relatively weak (accounting for a bit less than 15% of the variance). Significant effects of particular variables are situated at the three levels differentiating the explanation of surplus from that of scope. For one, variation in the institutional environment matters as expected. Local branches operating in the Flemish Region see significantly more surplus of integration than their counterparts in Brussels. For another, the effect of party family at the second level is less proliferated albeit not absent and alike in one. Compared with local branches of the socialists, again those of the liberals see significantly less surplus in integration.
      At the third, and final, level, electoral strength matters, twice as expected (although once in part). Compared with local branches with a low number of seats, those with a medium indicate significantly less surplus in integration. Those in opposition discern significantly more. This is the strongest predictor for an increase in surplus. This is in line with the presumption that integration with the national headquarters is positively valued to counterbalance the relative marginalisation of the local chapter in the municipal arena. There is no such effect for financial autonomy or organisational strength. For surplus, H1b, H2b (in part), H4b (in part), H8b (in part) and H9b can thus not be refuted.
      As a robustness test, we included a number of individual-level variables in a fourth model (not displayed in Table 5): whether the chair was female, of middle (45-57) or older (more than 57) age, of non-Belgian origin, was a medium (equivalent to more than one but less than three legislatures) or high (more than three legislatures) numbers of years active in local politics or had a (non-)executive local elected mandate. For scope, including these variables led to a slight increase in the overall explanatory power of the final model (0.225***), while for surplus to a slight decrease (0.146***). For both, none of the individual variables were significant in the final model. Equally, adding these variables did not substantially alter the effects found in model 3 (only for scope, the negative effect of a local chapter belonging to the liberal party changed to non-significant).

    • 6 Discussion and Conclusion

      This article scrutinised linkage between local branches and their national party headquarters. Incorporating the branch stratum added to the understanding of territorial power relations in multilevel parties. By drawing on a survey of their chairs in Belgium it adopted an innovative informal and bottom-up empirical approach.
      Multilevel linkage has been described through the dimensions and aspects of vertical integration and influence central in the literature and adapted to the local-national nexus. Regarding the former, the article focused on the downward provision of resources and services as a form of support for local branches. Pertaining to the latter, it considered the upward impact of local branches in key arenas of national headquarters decision-making. The article added an assessment of the perceived costs and benefits of linkage. Results showed that four-fifths or more of all local branches received at least some support from their national party headquarters and exercised a ditto influence on it. Although the extent varied with the domain or arena under study, both mother-daughter support and daughter-mother influence could be summarised into one composite measure. Further analysis pointed out that these scales are in turn related and can be captured in one factor expressing the scope of linkage. In the eyes of their chairs, local branches receiving more support are also likely to exercise more influence on their national party headquarters. Some daughters thus see more exchange with their mothers than others. These findings on local branches in Belgium nuance existing insights emphasising the distinction between vertical integration and influence at other levels of linkage. Both seem to refer to the intensity of the local-national relation.
      Furthermore, the largest part of local chairs agreed with most of the alleged advantages, while a smaller one did for the disadvantages. Based on the two underlying components demonstrated, we also developed a standardised measure indicating the surplus of linkage. Almost four-fifths of all chairs perceived the benefits of integration outweigh the costs. The contribution of linkage to the functioning of the chapter as an organisation, a means to defend interests vis-à-vis the national party headquarters or to mediate conflicts thus level out its limits to react to local circumstances, making own strategic choices or mother ultimately getting her way in case of conflict. These results add to our grasp on the attitudes towards linkage, which seem quite positive for local branches in Belgium.
      Our findings appear to align with the conception of linkage as interdependence, in which downward flows of support strengthen upward streams of influence. The relationship between branches and their headquarters is neither one of mere central hierarchy along the mass party type nor one of exclusive decentral independence as in the cartel thesis (Katz & Mair, 1995; Mair, 1994). Moreover, the branch stratum values the cooperation and collaboration within the party franchise (Carty, 2004).
      Apart from these general tendencies, the descriptive analysis demonstrated substantial variation in the scope and surplus of local-national linkage. In line with the literature, integration appears as a matter of degree (Bolleyer, 2012). To account for that, the article probed into three levels of explanation: macro (between regions as institutional environments), meso (between parties) and micro (within parties). The analysis enabled an explanation of a significant proportion of the variance in both variables. However, it showed that scope and surplus of the mother-daughter relationship were dependent on a different set of factors.
      As it comes to scope, differences between parties trump those within them. The combination of downward integration and upward influence appears not to be a function of the autonomy, organisational or electoral strength of a specific chapter but more of the general family approach of the national headquarters. Hence, in the eyes of their offspring mothers do not tend to treat some of their daughters differently than others. Instead, some mothers do for all of their daughters compared with others. Certain genes come with a particular style of parenting.
      For surplus, specific differences between parties as well as within them matter. The balance of benefits and costs tilts with the institutional environment in which local branches operate, some of the particular families they belong to and their preceding electoral strength. Here, in the eyes of their offspring some mothers treat some of their daughters differently than others. Particular genes matter, but the appreciation of parenting also aligns with the circumstances the descendants are functioning in. In some regions, the surplus is more apparent. This also holds for those facing difficulties in stepping up the ladder of political life into the partisan adulthood of representation and government participation.
      As often, the answer to the guiding question like mother, like daughter is it depends on for what and for whom. Further research should corroborate that conclusion. For one, it could deepen it by trying to account for the ample extent of variance left unexplained in the Belgian context bringing other factors than those hitherto considered and/or more refined versions thereof into the equation. For another, by scrutinising other aspects of the dimensions studied and particularly linking these more extensively to patterns of multilevel autonomy. There is also room for improvement through expansion. The bottom-up and/or informal could, for instance, be more systematically integrated with the top-down and/or formal approach (Poguntke et al., 2016). To complement causal effects with ditto mechanisms, more in-depth qualitative research is also necessary. The robustness of the findings could equally benefit from more cross-national research beyond one informative context. Given the multilevel nature of Belgian politics and the substantial presence and success of branches of national parties in local elections, the general patterns might be representative for the study of linkage in similarly integrated polities but not elsewhere (Bolleyer, 2012; Fabre, 2010; Thorlakson, 2009). From a party organisation and systemic point of view, almost the full range of families is included. The context harbours both mainstream catch-all as well as newer niche parties demonstrating specific tendencies in linkage. To take up such an ambitious agenda for further inquiry, this article serves as an open invitation to join the extended research family.

    • References
    • Ackaert, J., De Winter, L., Dumont, P. & Hennau, S. (2013). Met man en macht. Over mensen, macht en beleid bij lokale coalitievorming. In K. Deschouwer, T. Verthé & B. Rihoux (Eds.), Op zoek naar kiezers. Lokale partijafdelingen en de gemeenteraadsverkiezingen van oktober 2012 (pp. 147-164). Brussels: Academic and Scientific Publishers.

    • Allern, E. H. & Saglie, J. (2012). Inside the black box: Parties as multi-level organisations in a unitary state. West European Politics, 35(5), 947-970.

    • Bardi, L., Bartolini, S. & Trechsel, A. (2014). Party adaptation and change and the crisis of democracy. Party Politics, 20(2), 151-159.

    • Beyens, S., Deschouwer, K., van Haute, E. & Verthé, T. (2017). Born again, or born anew: Assessing the newness of the Belgian New-Flemish Alliance (N-VA). Party Politics, 23(4), 389-399.

    • Beyens, S., Lucardie, P. & Deschouwer, K. (2016). The life and death of new political parties in the low countries. West European Politics, 39(2), 257-277.

    • Bolleyer, N. (2012). New party organization in western Europe: Of party hierarchies, stratarchies and federations. Party Politics, 18(3), 315-336.

    • Buelens, J., Rihoux, B. & Deschouwer, K. (Eds.) (2008). Tussen kiezer en hoofdkwartier. De lokale partijafdelingen en de gemeenteraadsverkiezingen van 2006. Brussels: VUB Press.

    • Carty, R. K. (2004). Parties as franchise systems: The stratarchical organizational imperative. Party Politics, 10(1), 5-24.

    • Copus, C. & Erlingsson, G. (2012). Parties in local government: A review. Representation, 48(2), 235-247.

    • Dandoy, R., Amjahad, A. & Schrobiltgen, M.-H. (2013). Les élections communales à Bruxelles. In R. Dandoy, J. Dodeigne, G. Matagne &, M. Reuchamps (Eds.), Les élections communales de 2012 en Wallonie (pp. 115-127). Bruges: Van Broele.

    • Dandoy, R., Dodeigne, J., Matagne, G. & Reuchamps, M. (Eds.). (2013). Les élections communales de 2012 en Wallonie. Brugge: Vanden Broele.

    • De Rynck, F. & Wayenberg, E. (2010). Belgium. In E. Page & M. Goldsmith (Eds.), Changing Government Relations in Europe: From Localism to Intergovernmentalism (pp. 14-29). London: Sage.

    • Delwit, P., Pilet, J.-B. & Van haute, E. (Eds.). (2011). Les partis politiques en Belgique. Bruxelles: Editions de l’Université de Bruxelles.

    • Deschouwer, K. (2003). Political parties in multi-layered systems. European Urban and Regional Studies, 10(3), 213-226.

    • Deschouwer, K. (2006). Political parties as multi-level organizations. In R. Katz & W. Crotty (Eds.), Handbook of Party Politics (pp. 291-300). London: Sage.

    • Deschouwer, K., Rihoux, B. & Verthé, T. (2013). Op zoek naar de lokale partijvoorzitter. In K. Deschouwer, T. Verthé & B. Rihoux (Eds.), Op zoek naar kiezers. Lokale partijafdelingen en de gemeenteraadsverkiezingen van oktober 2012 (pp. 11-26). Brussels: Academic and Scientific Publishers.

    • Deschouwer, K., Pilet, J.-B. & Van Haute, E. (2017). Party families in a split party system. In K. Deschouwer (Ed.), Mind the Gap. Political Participation and Representation in Belgium (pp. 91-112). London: Rowman & Littlefield.

    • Detterbeck, K. (2012). Multi-Level Party Politics in Western Europe. Houndmills: Palgrave.

    • Devos, C., Verlet, D. & Reynaert, H. (2007). Spreken lokale afdelingen van Vlaamse partijen uit één mond? Res Publica, 49(1), 89-110.

    • De Winter, L., Erzeel, S., Vandeleene, A. & Wauters, B. (2013). Nationale bemoeienis of lokale autonomie? Het lijstvormingsproces bij gemeenteraadsverkiezingen. In K. Deschouwer, T. Verthé & B. Rihoux (Eds.), Op zoek naar kiezers. Lokale partijafdelingen en de gemeenteraadsverkiezingen van oktober 2012 (pp. 93-122). Brussels: Academic and Scientific Publishers.

    • Egner, B., Gendzwill, A., Swianiewicz, P. & Pleschberger, W. (2018). Mayors and political parties. In H. Heinelt, A. Magnier, M. Cabria & H. Reynaert (Eds.), Political Leaders and Changing Local Democracy. The European Mayor (pp. 327-358). London: Palgrave.

    • Fabre, E. (2010). Measuring party organization: The vertical dimension of the multi-level organization of state-wide parties in Spain and the UK. Party Politics, 17(3), 343-363.

    • Geser, H. (1999). The local party as an object of interdisciplinary comparative study. Some steps toward theoretical integration. In M. Saiz & H. Geser (Eds.), Local Parties in Political and Organizational perspective (pp. 3-43). Oxford: Westview Press.

    • Katz, R. S. & Mair, P. (1995). Changing models of party organization and party democracy. The emergence of the cartel party. Party Politics, 1(1), 5-28.

    • Krouwel, A. (2006). Party models. In R. Katz & W. Crotty (Eds.), Handbook of Political Parties (pp. 249-269). London: Sage.

    • Lehmbruch, G. (1979). Der Januskopf der Ortsparteien. In H. Köser (Ed.), Der Bürger in der Gemeinde (pp. 320-334). Hamburg: Hoffmann und Campe.

    • Mair, P. (1994). Party organizations: From civil society to the state. In R. S. Katz & P. Mair (Eds.), How Parties Organize. Change and Adaptation in Party Organizations in Western Democracies (pp. 1-22). London: Sage.

    • Moon, D. & Bradberg, O. (2010). Conceptualising the multi-level party: Two complementary approaches. Politics, 30(1), 52-60.

    • Müller, W. & Narud, H. (Eds.). (2013). Party governance and party democracy. Berlin: Springer.

    • Panebianco, A. (1988). Political parties: Organization and power. Cambridge: Cambridge University Press.

    • Pedahzur, A. & Britcha, A. (2000). The attractiveness of the seemingly unattractive: An Israeli case. Local Government Studies, 19(3), 55-70.

    • Pilet, J.-B. & Wauters, B. (2014). The selection of party leaders in Belgium. In J. Pilet & W. Cross (Eds.), The selection of political party leaders in contemporary parliamentary democracies (pp. 30-46). London: Routledge.

    • Poguntke, T., Scarrow, S. E. Webb, P. D., Allern, E. H., Aylott, N., Biezen, I. V., … Verge, T. (2016). Party rules, party resources and the politics of parliamentary democracies. How parties organize in the 21st century. Party Politics, 22(6), 661-678.

    • Pruysers, S. (2018). Party integration at the grassroots: Evidence from Canada. Territory, Politics and Governance, 6(1), 104-117.

    • Reiser, M. & Vetter, A. (2011). Multi-level parties and local interest representation: First approaches towards a more encompassing research design. Paper presented at the ECPR General Conference 2011, University of Reykjavik, Reykjavik, Iceland.

    • Scarrow, S., Webb, P. & Poguntke, T. (Eds.). (2017). Organizing political parties. Representation, participation, and power. Oxford: Oxford University Press.

    • Steyvers, K. & Heyerick, A. (2017). Fifty shades of Rokkan? Reconceiving local party system nationalisation in Belgium. Croatian and Comparative Public Administration, 17(4), 509-538.

    • Swenden, W. & Maddens, B. (Eds.). (2009). Territorial Party Politics in Western Europe. Houndmills: Palgrave.

    • Thorlakson, L. (2006). Party systems in multi-level contexts. In D. Hough & C. Jeffery (Eds.), Devolution and electoral politics (pp. 37-52). Manchester: Manchester University Press.

    • Thorlakson, L. (2009). Patterns of party integration, influence and autonomy in seven federations. Party Politics, 15(2), 157-177.

    • Thorlakson, L. (2011). Measuring vertical integration in parties with multi-level systems data. Party Politics, 19(5), 713-734.

    • Vandeleene, A., De Winter, L. & Baudewyns, P. (Eds.). (2018). Candidates, parties and voters in the Belgian partitocracy. London: Palgrave.

    • Wayenberg, E., De Rynck, F., Steyvers, K. & Pilet, J.-B. (2012). Belgium: A tale of regional divergence? In J. Loughlin, F. Hendriks & A. Lidström (Eds.), The Oxford handbook of local and regional democracy in Europe (pp. 72-95). Oxford: Oxford University Press.

    Noten

    • * The author is very grateful to the two anonymous reviewers who provided valuable comments and suggestions to an earlier version of the manuscript.
    • 1 See, https://www.politicalpartydb.org.

    • 2 This is the general assertion coming out of aggregated research. As pointed out in a review of a first version, the national party headquarters might keep a close watch on local coalition formation in large cities or be willing to secure the so-called cordon sanitaire against the radical right Vlaams Belang. National parties may also provide guidance in candidate selection or list composition.

    • 3 Original question: National party headquarters can support their local branches with different resources and services. Please indicate how much support your local chapter receives for each of the following aspects of its functioning.

    • 4 Respectively a bit more than 3% (training of candidates or staff) and 14% (voter research).

    • 5 Original question: Local branches can try to influence different aspects of their national party headquarters functioning. Please indicate how much influence your local chapter actually has on each of the following aspects.

    • 6 Respectively a bit less than 8% (composition of the party manifesto) and 20% (functioning of the national party executive).

    • 7 Original question: Integration between the local chapter and the national headquarters can have a number of advantages or disadvantages. To what extent do you agree with the following statements regarding your local chapter? Advantages and disadvantages were presented mixed to the respondents and not pre-categorised as visualised in the table.

    • 8 A majority of just under 57% of local branches are fully financial autonomous from their national headquarters. Among the remaining that share partial autonomy, only less than 10% derive more than half of their income from the national headquarters.

    • 9 Based on self-reporting by the local chairs and cut-off points for three equal groups. For the local branch executive, this means a number of 7 or less (low), 8-14 (medium) and 15 or more (high). For members, it implies a number of 56 or less (low), 57-149 (medium) and 150 or more (high).

    • 10 Based on self-reporting by the local chairs and cut-off points for three equal groups: a number of 3 or less (low), 4-8 (medium) and 9 or more (high). It should be noted that less than 5% of all local branches did not obtain any seats.

The author is very grateful to the two anonymous reviewers who provided valuable comments and suggestions to an earlier version of the manuscript.

Print dit artikel