The Political Party in Canada 9780774868235

Citation preview

The Political Party in Canada

This page intentionally left blank

Political Parties and Their Place in Canadian Democratic Life

Political parties are often characterized as a fundamental political institution, perhaps even the fundamental political institution that defines con temporary democratic politics. Minimalist definitions of parties focus on the fact that they nominate candidates for election, as this activity is what generally separates parties from other politically oriented organizations such as interest or advocacy groups. In reality, however, political parties do so much more than simply nominate candidates. Indeed, the list of functions that political parties perform is significant and includes the organization of government, the formation of policy, the aggregation and articulation of interests, the recruitment and selection of party personnel ( candidates, leaders, etc.), the integration and mobilization of individuals into tl1e political system, and the structuring of the vote during elections (King 1969). By engaging in these many functions, political parties have the ability to offer meaningful opportunities for citizens (and even non-citizen residents in some cases) to participate in formal, national level, democratic politics through activities such as candidate nomination, leadership selection, election campaign volunteering, and, to a lesser degree, the policy development process. In short, political parties make modern democracy possible and provide p articipatory opportunities for members and non-members alike. In the Canadian case, parties are often said to have additional functions, particularly brokerage and the fostering of national unity (see Meisel and Mendelsohn 2001). No small feat in a country that is as geographically, linguistically, and ethnically diverse as Canada.

4 The Political Party in Canada

Let's consider, for a moment, just how central political parties truly are to Canadian democratic life, with a few illustrative examples. A useful starting point regarding the primacy of political parties can be seen in who is ultimately elected to the House of Commons after each general election. In the first fifteen general elections since party labels were first listed on ballots, in 1972, only eleven independent candidates were successful in their electoral bids, just three of whom were non-incumbents (Marland 2020, 278). Stated differently, only 0.2 percent of the possible 4,474 seats have gone to non-party candidates over the course of the last five decades. During the 2019 general election, for instance, more than 94 percent of some 2,100 candidates who contested the election ran under a party banner (Small and Philpott 2020). While independents continue to contest elections and while it is true that some party candidates are nothing more than names on paper (Sayers 1999), voters continuously show a clear preference for party people when choosing a local representative. As a result, parties dominate elections, and it is their candidates who routinely capture the vast majority of votes and seats. Moreover, the empirical evidence suggests that even though voters are selecting from a range of individual candidates, party cues and party identification play a crucial role in their decisionmaking process (Blais et al. 2003; Johnston et al. 1992; Roy and Alcantara 2015; Stevens et al. 2019). The primacy of parties is seen not only in the preferences of voters but is enshrined in Canadian law as well. A clear example can be seen in the election-financing regime. While those participating in election campaigns must adhere to strict financing regulations (Young 2015; Young and Jansen 2011), the limits are far more generous for party actors. According to Elections Canada, if an election were to have been held in November 2020, the average candidate would be legally permitted to spend $115,825.68 in a district, with registered parties being able to spend an additional $88,169.82 on average. Registered third parties (i.e., persons or groups other than candidates, registered political parties, or electoral district associations seeking to participate in elections), by contrast, would have been limited to just $4,473 in a district during the same election period (Elections Canada 2020) .1 The result is that parties and their candidates are highly visible during elections ( able to purchase television advertising, mail cam paign literature, etc.), while non-party actors are sidelined. This, of course,

Political Parties and Their Place in Canadian Democratic Life

is no accident. Lawlor and Crandall (2020, 3) note that the legislative intent of the financing regime is, indeed, to prioritize political parties over other political actors; ((candidates and parties are the primary actors within campaigns, and other actors (specifically non-parties) ought to remain secondary:'2 Third-party spending limits are not only enshrined in legislation, but they have also been upheld by the courts on numerous occasions. In Libman v. Quebec (AG), (1997] 1 S.C.R. 569, 3 for example, the Supreme Court noted that while we recognize their right to participate in the electoral process, independe11t individuals and groups cannot be subject to the same financial rules as candidates or political parties and be allowed the same spending limits. Although what tl1ey have to say is important, it is the candidates and political parties tl1at are run11ing for election. Limits on independent spending must therefore be lower than those imposed on candidates and political parties.

Parties do not simply monopolize voter decision-making processes and election campaigns, they a1·e also central to how our legislatures operate. Once the election is over, parties remain dominant actors in the House of Commons, which is almost completely organized around them: Members of Parliament (MPs) sit with their respective parties; government is formed and operated along party lines (single party government having been the 11orm in Canada); and resources (such as the ability to participate during Question Period and to serve on committees) are allocated and distributed based on party standings. Nowhere is the importance of parties in the House more obvious than in the voting behaviour of MPs. High levels of party cohesion are a defining characteristic of the Canadian Parliament (Godbout 2020; Kam 2009; Malloy 2003; Marland 2020), with elected members rarely casting dissenting ballots. As Malloy (2003, 116) writes, ((The Canadian House of Commons has very disciplined parties even by Westminster standards:' During the 42nd Parliament (2015- 19), for example, the average MP voted with their party 99. 7 percent of the time, and even the most ((rebellious" MP did so 96.6 percent of the time (Thomas, Petit-Vouriot, and Morden 2020). While not as consistent, similar patterns can be found in the Senate as well: cc independent» senators

5

6 The Political Party in Canada

appointed by Pri1ne Minister Justin Trudeau voted with the Liberal government 94.5 percent of the time in the years following the expulsion of Liberal senators from the party's caucus (Grenier 2017 a). While brief, these examples should highlight the centrality of parties to democratic politics. The idea that political parties are important is not particularly controversial, 4 nor is it new. The importance of political parties is captured in Schattschneider's (1942, 1) classic argument that "political parties created democracy and modern democracy is unthinkable save in terms of the parties:' This importance has been recognized in the Canadian literature time and time again. Cross (2004, 1), for example, writes that «political parties are the central players in Canadian democracy:' He goes on to suggest that "parties are so central to our democratic life that if they are not participatory our politics cannot be participatory, if they a.r e not inclusive our politics cannot be inclusive, and if they are not responsive then our politics cannot be responsive:' Likewise, Carty et al. (2000, 1) note that «making sense of Canada has always meant making sense of its party politics. That is inevitable in a country first put together, and then kept together, by party politicians." More recently, Wiseman (2020, xiv) has referred to parties as the «essential vehicles of Canada's parliamentary democracy:' Despite their clear centrality and importance, scholarship on Canadian parties is som ewhat uneven. Indeed, the last truly comprehensive examinations of the inner workings of Canadian parties appear to be almost two decades old with the publication of Cross's Political Parties, in 2004; and Carty, Cross, and Young's Rebuilding Canadian Party Politics, in 2000. To be sure, mucl1 has been written about Canadian political parties since then, especially on specific topics such as party integration (Esselment 2010; Pruysers 2014; Thorlakson 2013), candidate nomination (Cheng and Tavits 2011; Cross and Young 2013; Pruysers and Cross 2016a; Thomas and Bodet 2013; Tolley 2019), leadership selection and tenure (Cross and Blais 2012a; Cross et al. 2016; O'Neill, Pruysers, and Stewart 2021; Thomas 20 18), party membership (Cross 2015a; Cross and Young 2004, 2008), personalization (Bittner 2011; Cross et al. 2020; Cross and Young 2015; Pruysers and Cross 20 16b), and election campaigning (Carty and Eagles 2005; Cross and Young 2011; Currie-Wood 2020; Sayers 1999). Furthermore, there have been a number of more recent book-length studies of

Political Parties and Their Place in Canadian Democratic Life

Canadian parties, including Flanagan's (2009) Harper's Team: Behind the Scenes in the Conservatives Rise to Power, Koop's (2011) Grassroots Liberals: Organizing/or Local and National Politics, Carty's (2015) Big Tent Politics: The Liberal Party's Long Mastery of Canada's Public Life, Lewis and Everitt's

(2017) The Blueprint: Conservative Parties and Their Impact on Canadian Politics, McGrane's (2019) The New NDP: Moderation, Modernization, and Political Marketing, Wiseman's (2020) Partisan Odysseys: Canada's Political Parties, and Jeffrey's (2021) Road to Redemption: The Liberal Party of Canada, 2006-2019.

While this scholarship is both impressive and important, it tends to have two limitations that we seek to address in this book. First, the books noted above generally only consider a single political party within the Canadian political landscape ( Carty and Koop focus on the Liberal Party, Flanagan on the Conservative Party, McGrane on the New Democratic Party, etc.). This limitation has been a recurring theme in the literature on Canadian political parties for decades. 5 The exclusion of multiple parties in the literature obscures potential differences between the parties and limits the generalizability of the findings. Comparison, which is often missing, is therefore crucial. Second, the broader literature (which we have contributed to and are therefore guilty of perpetuating) tends to examine only a narrow slice of what parties do and, in doing so, cannot provide a detailed and comprehensive account of what Canadian parties are, the functions they perform, and how they perform them. The result being that while we know a lot about various specific functions of Canadian party organizations, we have a much less coherent picture of how the various components of the party interact with one another, who populates the party at various levels, and so on. This lacuna, of course, raises an important question: Who (or what) is the political party? For some authors, the political party might be the MPs (parliamentary party) who represent the party in elected office. For others, it might be the dues-paying party members who constitute the grassroots, although the introduction of party supporters may complicate this somewhat. For others still, the party might comprise the 338 electoral district associations (EDAs) that organize party life across the country and maintain the party during the inter-election period. Political parties are complex and multifaceted, and the answer to "who is the party'' encompasses all the above and more. 6

7

8 The Political Party in Canada

As Carty (1991, 227) suggested three decades ago, attempting to describe our political parties reminds one of the old fable about people trying to describe an elephant by using only their sense of touch: «after touching only the tail, the tusk, the trunk, the legs or the ears, no one was able to provide much of an account of the beast:' In this sense, continuously writing about party members separate from the EDAs that recruit, retain, and engage with them is unsatisfactory, just as writing about parliamentary candidates separate from the party members who select them and the donors who financially support them is equally problematic. A comprehensive examination of Canadian parties - the central political institution in Canadian political life - is long overdue: to continue the elephant metaphor, we need to describe the tail, tusk, trunk, legs, and ears of our political parties simultaneously. We need to provide an anatomy of our parties, which is precisely the purpose of this book: to provide a rich, detailed, and comprehensive account of who our political parties are and what they do. In other words, we seek to fully understand the «beast:' ANIMATING THEMES

In providing a more fulsome account of who and what the political party is, this book is guided by a number of animating themes that run throughout the chapters. Discussed in more detail below, these themes include 1) stratarchy and the organizational complexity of Canadian parties; 2) representation, inclusiveness, and the makeup of Canadian parties; and 3) personalization within Canadian parties. Importantly, all these themes build on, and relate to, one another. If Canadian parties are complex and multifaceted due to the adoption of a form of stratarchical organizational design, as we argue, then we ought to be concerned with the makeup of the party at all levels, not just with the elected representatives at the centre of the party apparatus. Since power is not only allocated at the centre of parties, we should also be interested in who occupies positions throughout them at various levels (e.g., members, candidates, EDA executives, etc.). Our concern with representation, therefore, flows directly from our first theme of complexity and an understanding that parties have multiple sites of power and influence. Likewise, a focus on representation and the makeup of parties logically leads to a consideration of the influence of these individuals (personalization) and the question of whether it matters who

Political Parties and Their Place in Canadian Democratic Life

occupies these positions. In this sense, our third theme flows from the first and second ones. While these are the organizing themes of the book, it is important to note that the book is not overtly structured around them in the sense that each chapter has the same subheadings addressing these specific issues. Rather, these themes animate the content of each chapter where appropriate and often in different ways. The Complexity of Canadian Political Parties

Canadian parties, while often portrayed as singular, coherent, and unitary actors, are really anything but. The first animating theme of this book, therefore, is that Canadian parties, like their counterparts in other parliamentary democracies around the world, are complex and multifaceted organizations. In order to fully understand our political parties, they must be disaggregated rather than treated as single monolithic actors. To be sure, this idea is not a new one. Katz and Mair (1993, 594) convincingly argue that political parties are not singular actors so much as they comprise three important, and often competing, components; ((rather than analyzing parties according to a simple parliamentary versus extra-parliamentary dichotomy, or a simple leaders-versus-followers hierarchy (no matter how finely subdivided), it is more productive to consider parties as comprising a number of different elements, or faces, each of which potentially interacts with all of the others:' The different "faces" of the party identified by Katz and Mair include the p arty in public office (e.g., those members in Parliament or government), the party on the ground (e.g., party members, activists, etc.), and the party in central office (e.g., national organizational leadership distinct from the party in public office). The identification of these three faces of political parties is important as it challenges the longstanding view of parties as hierarchical and unified organizations. As Carty (2004, 6) explains, the traditional conception of a political party "as a single identifiable organization that some group can capture and command" may no longer accurately describe how parties actually organize. While parties were once organized in a hierarchical fashion (Michels 191 5/2001),7 stratarchy is a more apt description of how many modern parties organize. The defining features of a stratarchical design are multiple sources of power and a number of mutually autonomous organizational subunits with considerable independence within their

9

10 The Political Party in Canada

respective spheres of competency/authority. Perhaps one of the first to recognize the stratarchical nature of parties, Eldersveld (1964, 10) has written that con trary to the bureaucratic and authoritarian models of social organization, the party is not a precisely ordered system of authority and influence from the top down, tl1ough as a "paper,, structure it may give this appearance. The organization does not fu11ction through the issuance of directives from the top which are obeyed without question. Rather, there is a tolerance of autonomy, local initiative, local inertia.

Developing the stratarchy argument further and linking it back to the work of Katz and Mair, Carty (2002; 2004) and Carty and Cross (2006) suggest that stratarchy is an organizational bargain that is made between the competing faces of the party. Applying it to the Canadian context, this bargain occurs between the party on the ground and the party in public office as Canadian parties have never had a particularly strong party in central office. The party on the ground has long been an important player in Canadian party politics. The country's single-m ember plurality electoral system divides the country into 338 distinct electoral districts (sometimes referred to as constituencies or ridings) . Given that electoral victory is achieved by winning individual districts and not necessarily the most votes, Canadian parties have long had an incentive to create active and strong local grassroots party organizations (Carty 1991; Carty, Cross, and Young 2000; Carty and Eagles 2005; Cross 2016; Currie-Wood 2020; Pruysers 2015a; Sayers 1999). The stratarchical bargain that manifests in Canada has also been referred to as the "franchise" model of party organization (Carty 2002): local constituency organizations (EDAs) act as franchisees where the central organization establishes the brand, m anages the marketing campaign, provides leadership, and determines policy. Despite central party oversight, the local franchise retains autonomy over local affairs and exercises local decision-making authority. In the Canadian case, the stratarchical bargain is said to provide the party on the ground (specifically local constituency associations) with a significant degree of autonomy and independence over the management a.n d direction of local party affairs, including the selection

Political Parties and Their Place in Canadian Democratic Life

of the party's local candidate for the general election. This bargain, therefore, allows the local party apparatus to organize and conduct their business with minimal interference from the central party. In exchange for this near autonomy, the party in public office is provided disciplined support and the freedom to develop policy and manage the party's brand (Carty and Cross 2006, 97; Marland 2016). More recently, however, Cross (2018) has argued that mutual interdependence, rather than mutual autonomy, is a more accurate description of how stratarchy is actually practised. He argues that there is «little evidence of the commonly presented model of stratarchy as mutual autonomy for each level within discrete areas of competency" and that «both the party on the ground and in the centre share authority ... resulting in a pattern of mutual interdependence rather than mutual autonomy" (205). The concept of shared authority, rather than discrete and autonomous authority, is therefore a better way of understandin g the relationships, interactions, and interconnectedness of Canadian party organizations. Take, for example, the n ature of candidate selection. This essential party function is almost always portrayed as being an inherently local task. After all, EDAs recruit candidates, and it is local party members who vote in candidate nominations. While this portrayal is true, a variety of other party actors are also actively involved in the process (see Pruysers and Cross 2016a, Rabat and Cross 2018). Party leaders are required to "sign -off" on all candidates and even h ave the ability to veto local candidates. Central party officials recruit candidates, set the timing of nominations, and rigorously vet potential candid ates before their names can appear on the ballot. Candidate selection is by no means unique. Other party functions such as developing policy, selecting party leaders, and waging election campaigns are all characterized by shared authority and are much more complex and multilayered than they are generally portrayed to be. Stratarchy, and shared authority more generally, means that we cannot simply focus our attention on the "centre" of the party but must, therefore, adopt a more expansive approach to the study of parties. The stratarchical bargain that defines Canadian party organizations also means that there is considerable variation not only between political parties but also within them. The freedom enjoyed by the party on the ground, although not absolute, means that different local party organizations within the same

11

12 The Political Party in Canada

political party ( of which there are hundreds) will vary in any number of ways, including in their capacity (e.g., electoral strength, membership, vibrancy, etc.), their relationship to the centre, their ability to implement and follow standardized rules set by the national party, the nature of the strategies they implement during elections, and so on. The complexity of Canadian parties is further complicated by the federal nature of the country (Cross 20 l Sb). Each of the three principal parties has adopted a different formal organizational approach to address the issue of multilevel politics, with the New Democrats typically described as the most "integrated;' the Liberals as "semi-integrated;' and the Conservatives as "truncated" (see Pruysers 2014). While Canadian parties, especially the Liberals and Conservatives, have removed many of the formal organizational connections that once unified federal and provincial parties (Dyck 1991; Esselment 2010; Koop 2011), many informal, or behavioural, linkages remain (Pruysers 2014; 2015a; 2016). Throughout the chapters we pay attention not only to how the various faces of the party interact with one another, but how various elements of the party may also interact with their provincial counterparts. These connections include, for example, multilevel party membership, campaign support, and career paths of elected representatives. The complexity of our parties, however, is more than just a result of stratarchy and federalism. Another key part of what makes our parties complex is the fact that they are not static organizations. The Greek philosopher Heraclitus is said to have once remarked that you cannot step in the same river twice; the river is in a constant state of flux, ever-changing as the water ebbs and flows. Our political parties, like the river, are in a constant state of change. Party leaders come a11d go, often bringing with them an entirely new entourage and staff (Noel 2007). MPs experience exceptionally high levels of electoral turnover such that the composition of the House changes considerably after each and every election (Docherty 1997). Party membership declines in non-election years and swells in the lead up to general elections and personnel selection contests as new members are mobilized into the party (Carty 1991; Pruysers and Cross 2016a). Grassroots activism , including life within EDAs, becomes somewhat more dormant in the inter-election period and springs to life during election season as candidates are nominated and local campaigns are waged ( Cross

Political Parties and Their Place in Canadian Democratic Life

2016). Lastly> the number of national party staff (including professional campaign operatives) balloons as the election nears in an effort to conduct a coordinated and sophisticated national campaign (Marland and Giasson 2020) . All this is to say that the nature and composition of political parties is far from static: the party of the inter-election period is not the same as the party of the election period, and the party of 2019 is not the party it was in 2015. 8 Disaggregating Canadian parties, therefore, is more than just a partyon-the-ground versus party-in-the-centre dichotomy> and it requires further consideration beyond how parties operate during election season. Indeed, as tl1e chapters in this book demonstrate, the three faces of the party identified by Katz and Mair ( 1993) are themselves complex, and can and should be disaggregated further. This is why> for example, this book does not feature a single chapter about the party on the ground but instead considers financial donors, party members, activists> and EDAs separately. Likewise, we disaggregate the party in public office to explore candidates who run for election, those who ultimately serve in Parliament, and those party professionals who support these individuals behind the scenes. Doing so provides a much more complete picture of how Canadian parties truly appear. Some of the questions we address along the theme of stratarchy/ complexity include: Who does what within the party? How (and how well) do they fulfil their functions? How does the nature and composition of party actors change as we move from the inter-election to election period and vice versa? How do the various faces of the party apparatus work together and how much tension is there between them? Representation within Canadian Pol itical Parties

Many conceptions of political parties, according to Katz and Mair (1995, 7), portray parties as ccthe (not an) essential linkage between citizens and the state:' Chandler and Siaroff ( 1991, 192), for instance, have written that «parties operate as the crucial intermediaries linking rulers and ruled. The most basic party function is that of representation involving the translation of public opinion to political leaders:>Likewise, Cross (2019, 15) writes that «as political parties play a key role in structuring and executing election campaigns, in organizing our legislatures and governments and in playing a gatekeeping role to political office, it is important to consider

13

14 The Political Party in Canada

who constitutes the political party and whether they are inclusive of the various segments that together form civil society:' In linking civil society to the state, political parties have the ability to provide both descriptive and substantive representation - incorporating marginalized voices not only in the political party but in the policy process more broadly. Crotty (1968, 260), for example, explains the importance of candidate selection in relation to representation in the following way: The party in recruiting candidates determines the personnel and, more symbolically, the groups to be represented among the decision-making elite. Through recruitment, tl1e p arty indirectly influences the types of policy decisions to be enacted and the interests most likely to be heard. Candidate recruitment then represents one of the key linkages between the electorate and the policy-making process.

If we can accept that parties fulfil important democratic functions ( organizing government, structuring the vote, etc.) and make important democratic decisions (selection of party leaders, creation of platforms, etc.), as we have already argued in this chapter, then how parties are constituted and who makes these decisions also matters. The second animating theme of the book, therefore, focuses on representation within our political parties. Much of the empirical research regarding the representativeness of political parties has focused on the party in public office. That is, most studies have been concerned with the makeup of members of the elected legislature (for example, Caul 1999; Paxton 1997; Paxton and Kunovich 2003; Reynolds 1999; Tremblay 2007). Diverse parliamentary party groups, and by extension diverse parliaments, are important and have many benefits like challenging stereotypes (Bashevkin 2009), enhancing civility (Young 2009), and increasing substantive (in addition to descriptive) representation (Tremblay 1998). In the previous section, however, we made the case that political parties are not hierarchical and singular actors. Instead, they are complex organizatio11s with multiple sites of power and authority. Given the stratarchical nature of Canadian parties, we cannot be satisfied with simply documenting who makes up the "top» positions (i.e., MPs) within the party organization. If power is distributed throughout the party, then

Political Parties and Their Place in Canadian Democratic Life

it becomes crucial to understand who makes up the various faces throughout the party. A desire to understand the composition of other faces of the party is already underway in the party literature. Indeed, there is a growing literature on the representativeness of party members ( Cross 2015a; Cross and Young 2004; Gauja 2013; van Haute and Gauja 2015; see also Heidar and Wauters 2019), general election candidates (Black and Erickson 2006; Erickson 1991; Pruysers and Cross 2016a; Thomas and Bodet 2013), political aspirants (Cross 2016; Cross and Pruysers 2019a; Pruysers and Blais 2019; Tolley 2019), and party leaders (Bashevkin 2010; Cross 2014; Cross and Blais 2012a; Cross and Pilet 2015; O'Brien et al. 2015; O'Neill, Pruysers, and Stewart 2021; Wauters and Pilet 2015). While this is an important first step, this literature rarely speaks to one another.9 Studies of members are typically kept separate from studies on candidates, studies of candidates separate from party leaders, and so on. The result is that no clear picture on the representativeness of the party as a whole can be found. This siloing is problematic, especially in light of emerging research which documents that representativeness at one level of the party is connected to representativeness at other levels (Cross 2016; Pruysers et al. 2017). Cross and Pruysers (2019a) and Tolley (2019) explore nomination contestants rather than general election candidates. In doing so, and building on the work of Cheng and Tavits (2011), these authors have found compelling evidence that who enters the race (i.e., the representativeness of nomination contestants) is itself influenced by the representativeness of the local party organization (see also Tremblay and Pelletier 2001). Cross and Pruysers (2019a), for example, find that an EDA is considerably more likely to have a woman contestant for the nomination when the association president is also a woman. Likewise, Tolley (2019) finds that racialized party gatekeepers play a crucial role in the emergence of visible minority candidates. These recent findings highlight the importance of understanding the makeup of the party at all levels, not just within the central ranks. As such, we carefully document who belongs to and holds positions of authority within multiple levels of the party organization. The party literature also reveals what areas we know comparatively little about. For example, one area of the party that is often overlooked is the party in central office, that is, the individuals who orbit around the

15

16 The Political Party in Canada

party leader, often working in leadership campaigns, national party offices, and parliamentary offices (such as the Prime Minister,s Office), and orchestrating national election campaigns. While there has been recent attention devoted to campaign professionals, such as pollsters and media and communications specialists (for example, Carty, Cross, and Young 2000; Marland 2016; Marland and Giasson 2020; Marland, Giasson, and Lawlor 2018), far less attention has been paid to what we call the leader's entourage. Similarly, the existing literature pays almost no attention to the parties' Funds, which have authority over their financing and have recently used this to exert influence over leadership in both the Conservative and Green parties. We make an important contribution in this book by paying attention to these understudied party officials and by comparing them to other, better documented, layers of the party apparatus. Personalization and Canadian Party Politics

As democratic political institutions, political parties clearly matter. So too, however, do individual party actors. As Pruysers, Cross, and Katz (2018, 1) write, «Despite the centrality of political parties as organizations ... individual political actors have been, and will always be, crucially important to democratic politics:' Whether it is the leader and their entourage, the MPs in Parliament, the candidates running in general elections, the donors funding campaigns through their financial contributions, or the dedicated stalwarts meeting locally month after month to keep the party alive between elections, people are at tl1e core of what a political party is.10 To be sure, organizational rules and structures are essential, but people are crucial to the way parties actually function. Thus, part of understanding the complexity of Canadian parties, part of disaggregating the various faces of the broader organization, is understanding who the individuals that populate our parties are, what they do, and how they interact with other actors in the party. Trends over the last number of decades toward more candidate/ leader-centred politics (Cross and Blais 2012a; McAllister 2015; Pruysers and Cross 2016b; Wattenberg 1991; Zittel 2015), increasingly individualized campaigns (Cross and Young 2015; D e Winter and Baudewyns 2015; Zittel and Gschwend 2008), and the growing authority of party leaders and chief executives (Malloy 2018; Poguntke and Webb 2005; Savoie 1999a)

Political Parties and Their Place in Canadian Democratic Life

are all part of a broader phenomenon that has been labelled the personalization of politics. Rahat and Sheafer (2007, 65) define personalization as a process in which the «political weight of the individual actor in the political process increases over time, while the centrality of the political group (i.e., political party) declines." Likewise, Karvonen (2010, 4) explains personalization as "the notion that individual political actors have become more prominent at the expense of parties and collective identities. The central concept denotes a process of change over time: at t politics was less perso11alised than at t+ 1:' While these definitions accurately capture the concept of personalization, Pruysers, Cross, and Katz (2018) warn against placing too much emphasis on the «change over time" element, and suggest that by focusing longitudinally, we often lose sight of how important party actors are at any given time.11 Instead, they argue we should not discount the more static form of personalized politics ( or "personalism"). We adopt this latter theoretical approach when examining questions related to our third animating theme of the book, the personalized nature of Canadian party politics. While the personalization and related literatures have tended to focus on party leaders and prime ministers (Bittner 2011; Lobo and Curtice 2015; Poguntke and Webb 2005; Stewart 2018), personalization can, and does, occur at multiple levels within the party. Many party scholars have demonstrated that personalization is common at the candidate level, something labelled «decentralized personalization" (Balmas et al. 2014; Cross and Young 2015; Cross, Currie-Wood, and Pruysers 2020; Pruysers and Cross 2018; Zittel and Gschwend 2008). More recently, others have argued that personalization occurs not only in a dichotomous centralized (leaders) versus decentralized (candidates) fashion. Indeed, Gauja (2018) demonstrates how even party membership is also becoming increasingly person alized. In this sense, and consistent with our argument regarding the need to disaggregate parties, personalization occurs throughout party organizations at a variety of different levels. While a seemingly abstract concept, personalization has a number of important democratic implications. Ongoing demands for relaxed party discipline and more frequent free votes (Loat and MacMillan 2014), for example, can be seen as an example of personalization inside the legislature: an effort to shift the balance of power and provide more authority

17

18 The Political Party in Canada

and autonomy to individual party actors as opposed to the party itself. Members of Parliament, even when toeing the party line, often express their individuality through the questions they ask in the House of Commons or the private members' legislation they introduce for debate, subtly raising their own profile or highlighting their own district (Loewen et al. 2014; Chiru 2018). As the final theme of the book, we consider questions related to personalization, explore the degree to which various levels of the party are personalized, and consider the implications of such personalism on our democratic politics. A NOTE ON DATA

To establish a deeper understanding of our parties, who they are, and what they do, this book utilizes a mixed-methods approach that draws on both quantitative (i.e., surveys) and qualitative (i.e., interviews, party documents) data. 12 In doing so, we have assembled a truly impressive body of original data that spans a wide number of sources and methodological approaches. Triangulating data in this manner allows us to tell a rich and comprehensive story about Canadian political parties and to be more confident in our findings. By no means an exhaustive list, data for the book include original surveys across two recent election cycles (party members, donors, candidates, and electoral district associations), an original dataset on the career trajectories of central party staff, interviews with party officials, analyses of party documents, coding of party websites/social media usage, financial and nomination reports made available by Elections Canada, candidate biographies, and other biographical information on members of Parliament made available by the Parliament of Canada. While we focus our attention primarily on the three principal parties - the Liberals, Conservatives, and New Democrats 13 - we occasionally leverage our data to provide details about other parties including the Greens and Bloc Quebecois. In terms of temporal scope, we focus our analysis on the contemporary period over two election cycles (2015-19), which allows us to capture changes in the party organization during and between elections. With that said, we do cover major party changes that predate this period, such as the Liberal Party's movement toward institutionalizing party «supporters" as a replacement for traditional dues-paying members. Finally, while our primary focus is on the federal level, a recurring theme

Political Parties and Their Place in Canadian Democratic Life

throughout the chapters is the informal multilevel nature of Canadian parties and their relationship to provincial counterparts (Pruysers 2015a; Stephenson et al. 2019), whether it is in terms of m embership, activism, or general cooperation. While data are documented throughout the various chapters, we highlight some of their basic details here (launch dates, response rates, etc.) together with our other data sources: • 201 5 Candidate Survey. In cooperation with tl1e federal parties, the 2015 candidate survey was conducted online between November 2015

and January 2016. The dataset includes 440 candidates who contested the election across five political parties: Conservative (138), Liberal (78), New Democratic (11 3), Green (100), and Bloc Quebecois (11). • 2016 EDA Survey. The 2016 Electoral District Association survey was

conducted online during February of 2016. The dataset includes 256 respondents across five political parties: Conservative (39), Liberal (91), New Democratic (81), Green (44), and Bloc Quebecois (1). • 2016 Members Survey. The members survey was conducted online

between February and April of 2016. The dataset includes 932 party members. 14 The breakdown of survey respondents is as follows: Liberal (345), Conservative (330), New Democratic (156), Green (41), Bloc Quebecois (45), and 6 respondents reported belonging to a minor party. Nine respondents failed to provide their party. • 2019 Candidate Survey. In cooperation with the federal parties, the

2019 candidate survey was conducted online between November 2019 and January 2020. The d ataset includes 255 candidates who contested the election across three parties: Conservative (8 1), Liberal (67), and New Democratic ( 107). • 2019 EDA Survey. The 2019 Electoral District Association survey was conducted online between June and July of 2019. The dataset incudes

367 respondents from four political parties: Conservative (83), Liberal (93), New Democratic (98), and Green (77), and 16 who declined to identify their party. • 2020 Don.or Survey. The 2020 Donor Survey was conducted online in

November 2020. The dataset includes 256 respondents who previously reported donating to a Canadian political party, candidate, or EDA

19

20

The Political Party in Canada

•

•

•

•

between 2015 and 2019. 15 We supplement this with 2019 Canadian Election Study data from party donors. Elite Interviews. Semi-structured interviews were conducted via Zoom with six current a11d former senior party officials in early 2021. The purpose of these interviews was to gather information on the party in central office and to confirm data collected through our online and media research. Party Constitutions and Organizational Documents. Official party documents were accessed and reviewed to reveal the parties' «formal" organizational details including their composition and the relationship between different strata of the parties. Elections Canada. Financial reports of parties, general election candidates, leadership contestants, and electoral district associations were accessed and reviewed to analyze income, expenditures, and contribution sources. Online searches. Candidates' campaign biograpl1ies, parliamentarians' profiles, career and demographic information of central political operatives, and candidate and electoral district association web presence were all documented through online searches of sources such as Linkedln, the Parliament of Canada, party web pages and media sources.

In total, then, the book draws on original survey responses from 695 candidates, 623 EDAs, and more than 1,100 p arty members/donors. When combined with the other data we have collected - such as interviews, candidate biographies, profiles of party professionals, social media presence, party rules and statutes, Elections Canada reports, and much more - we are able to provide a rich and detailed account of Canadian parties with a specific focus on who they are and what they do at multiple levels. STRUCTURE O F THE BOOK

In the remainder of the book we provide one of the m ost comprehensive descriptions of the n1akeup and activities of Canadian political parties ever conducted. Parties, we argue, are more than the sum of their parts. As such, it is important to examine various components of the party organization/ apparatus simultaneously in order to establish a meaningful understanding of one of our most important democratic institutions. We start from the

Political Parties and Their Place in Canadian Democratic Life

grassroots and move toward the centre of the party ranks. Chapter 2 provides an analysis of party members and committed party activists. Still exploring the party on the ground, Chapter 3 provides an account of electoral district associations (EDAs) both in terms of makeup and activity. Chapter 4, on candidates, and Chapter 5, on financial donors, bridge the party on the ground and the other faces of tl1e party. Chapter 6 1noves to a particularly underresearched component of the party: the party officers and central party staff who work behind the scenes supporting campaigns, crafting policy platforms, and managing party affairs. Chapter 7 explicitly focuses on the parliamentary party, and here we are primarily interested in the House of Commons and its members. Finally, in Chapter 8, we are able to draw out some important comparisons across the various faces of the party. Careful observers will note that there is no single chapter devoted to leaders. Party leaders are remarkably important in contemporary democratic politics, and this is indeed tl1e case in Canada as well ( Cross and Blais 2012a; Cross et al. 2016; Cross, Katz, and Pruysers 2018; O'Neill, Pruysers, and Stewart 2021). The decision not to include a separate chapter on leaders is not a rebuke of tl1eir importance. Instead of a standalone chapter on party leaders (of which there are so few federally), party leadership is addressed throughout the book. In the EDA chapter, for instance, we discuss the role of local associations in leadership selection, and in the chapter on party members we consider the reasons members join parties, including factors like the ability to participate in a leadership contest. Chapter 6 includes a discussion of how central party committees shape the rules surrounding leadership selection. FiI1ally, in the chapter on the party in public office, we consider questions of party discipline (imposed by the party leader), the importance of caucus endorsements in leadership selection, the profile of party leaders since 1965, and a variety of other themes that directly touch on party leaders. In this sense, party leaders are integrated into many chapters, an approach that highlights the interdependence of the various faces of the party.

21

The Parties' Members

Similar to those in most democracies, Canada's political parties have been membership organizations for many decades. Organized through a network of local associations found in each of the country's electoral districts, the parties have typically welcomed supporters to join through these community-based groups. Members provide parties with several key benefits (see van Haute and Gauja 2015). Members offer evidence of support from the electorate, thus, large numbers strengthen claims to legitimacy for a party's activities and decision-making between elections. They provide a linkage between voters and party officials as the "eyes and ears" of the party on the ground in communities across the nation. They provide a pool of ready ccfoot soldiers" for the party to activate during election campaigns, particularly for their all important get-out-the-vote activities. Members are a source of financing for the parties, particularly in the Canadian case, with a prohibition on corporate and trade union contributions and a recent decrease in state-financing (see Chapter 5). Finally, they also provide a pool of potential general election candidates ( see Chapter 4). Members, as the face of the party in local communities, facilitate many of the activities assigned to parties for a functioning and healthy democratic system (see King 1969; Meisel and Mendelsohn 2001). If, as suggested in Chapter 1, political parties form a conduit between civil society and the state (Dalton, Farrell, and Ian McAllister. 2011; Kitschelt 2000; Mueller and Katz 1997), then party members are an essential link in this connection. Given this importance, comparative researchers observed with

The Parties' Members

concern an erosion of membership rates in many parliamentary democracies beginning in the 1960s (Mair and van Biezen 2001; Whiteley 2011). Alongside membership decline, parties became increasingly reliant on state subsidies, which could be used to replace volunteer labour with paid staff. Some feared that parties would welcome and encourage this transition, viewing the cost of facilitating memberships to exceed the benefits (see Katz 1990; Katz and Mair 1995) simply because staff are easier to manage than members with policy goals and other objectives that may be incongruent with the party leadership (for example, May 1973). Access to public funding not only created an opportunity for party elites to replace members with paid workers but also made member-based financing less important. While these concerns have proven to have some merit, they have not fully materialized. For example, while parties have professionalized to a certain extent (Flanagan and Jansen 2009; Panebianco 1988), research is beginning to show that parties still seek to maintain robust memberships even when they have access to substantial public funding (van Biezen and Kopecky 2017, 100). Moreover, those at the centre of party politics, both in our interviews and in emergi11g comparative literature, generally credit rank-and-file members as "assets, not liabilities" (Bale, Webb, and Poletti 2019, 166; see also Gerber and Green 2000). In fact, the demand for grassroots members has encouraged parties to adopt new types of partisan affiliation. In a variety of parliamentary democracies, parties are offering informal forms of ''membership" that encourage community organizing and message dissemination - key tasks often traditionally performed by full-fledged members (Gauja 2015; Scarrow 2015). These affiliations range from financial donors to non-member campaigners to social media influencers and beyond. Indeed, engaging in politics online has fuelled some of the changes to the way parties organize (Dommett, Kefford, and Power 2020), especially because various forms of affiliation, what Scarrow (2015) calls "multi-speed" memberships, are thought to be more appealing to certain groups, such as youth, who are less likely than older and affluent individuals to formally join parties (see also Cross and Young 2008; Widfeldt 1995). Some Canadian parties have created light memberships to increase their rolls, the most notable example being the Liberal Party of Canada's enfranchisement of "registered supporters" alongside party members

23

24 The Political Party in Canada

during its 2013 leadership contest and their subsequent replacement of formal membership with Registered Liberals.1 Party members have the right to participate in internal party decisions whereas other affiliates typically do not. Members, and Liberal supporters, in Canadian parties, play a key role in selecting the parties' general election candidates and the party leader. Participation in these personnel selection contests, along with relatively low membership fees, create strong incentives to join parties because there is a clear material benefit (Young and Cross 2002a). The result is highly cyclical membership life cycles. It is common for party membership rolls to balloon as "tourists» join to participate in personnel selection before deflating back to long-time activists thereafter (Carty 2015; Carty and Cross 2006; Cross 2015a). In this sense, members are important gatekeepers to elected office and exercise significant influence over the types of individuals who serve in Parliament and as prime minister. Members also play a role in determining the organizational parties' policy positions, and, while these may have limited influence in their parliamentary party's policy priorities, the views of party members do provide some incentives and constraints ( Cross 2007). For all these reasons, in our effort to understand the composition of the political party, it is appropriate to start with a consideration of their members. In this chapter we pursue several key questions relating to party membership, including the mechanisms through which one joins a party, the sociodemographic and ideological profiles of party members, their motivations for joining the party, their degree of activis1n in party affairs, the types of activities they engage in, and, finally, their degree of satisfaction with their role in the party. To answer these questions, we draw on an original survey and dataset of 932 party members collected after the 2015 general election. Importantly, we are able to compare this analysis to earlier studies of party members (Cross and Young 2004) in order to contextualize our results and understand the extent to which party membership is changing. We begin with a consideration of how the parties organize and structure their me1nberships. THE ORGANIZATION OF PARTY MEMBERSHIP

Party membership in Canada is easily accessible and open to a wide group. Eligibility rules are determined by the parties themselves and are more

TI1e Parties' Members

expansive than those adopted by the state for general elections. None of the parties restrict membership to Canadian citizens, though residency is generally required. And, while voting in general elections is restricted to those eighteen and older, all the parties allow younger members. A typical minimum age is fourteen for membership with full voting rights. The result is that many who are i11eligible to vote in a general election are able to participate in deciding who their party,s candidates will be in those elections and to vote for the selection of the party leader - effectively their candidate for prime minister. For the Conservatives and Liberals, their membership rules are set in their national constitutions and bylaws as adopted by the party in convention (Conservative Party of Canada 2015; 2021a, Sec. 4; Liberal Party of Canada 2020a; 2021a, Sec. 7, 10). The New Democratic Party (2018a, Article 3. 1) national constitution stipulates that membership requisites are set by its affiliated provincial parties. A product of its "federated,, structure, members of the NDP automatically belong to both the federal party and its counterpart in their home province. Parties argue that their more expansive membership eligibility provides a vehicle for entry and orientation to the political sphere for those not yet eligible to vote and in that sense encourages broader democratic participation. As discussed above, the Liberal Party moved to a system of party "supporters;, as opposed to members, following constitutional revisions adopted in 2016. The primary difference realized through this change is that supporters are not required to pay a fee but, otherwise, generally have rights and responsibilities similar to those previously ascribed to members. Three con1mon restrictions on party membership are the requirement that a prospective member not belong to another federal party, that they agree with the basic principles of the party, and that they pay an annual m embership fee. The fee in the Canadian case is comparatively small and typically ranges between $10 and $25 with reduced fees for groups such as youth and those purchasing a multi-year membership. While the fees are modest, they do provide a significant source of party funding in the aggregate - particularly in years when there is growth in membership numbers resulting from leadership and candidate selection contests. For example, the Conservative Party reported receiving membership fees totalling $1,157,642 in 2016, $3,506,818 in 2017, and $2,419,800 in 2018. 2

25

26

The Political Party in Canada

These numbers illustrate the flux in membership numbers with significant mobilization associated with the party's 2017 leadership election. The Liberals raised $2,221,085 in membership fees in the election year of 2015 and $644,289 in 2016, before replacing members with supporters who are not required to pay a fee. With the expansion of the online world, obtaining party membership has become easier than ever as all the parties now offer membership acquisition through their websites. A prospective member need only answer a few basic questions confirming their eligibility, contact information, and credit-card details. One caveat is that members must pay the membership fee themselves, not have it bulk-purchased as was a common past practice employed by leadership and candidate selection campaigns. There are then very few barriers to membership in Canadian parties. And after joining, members quickly are eligible to fully participate in party affairs. As opposed to apprentice or activism requirements that some parties in other countries impose as a requisite to obtaining voting rights on party decisions (Cross and Gauj a 2014), any waiting periods in Can adian parties typically relate only to administrative efficiency. For example, parties will impose relatively short membership cut-off dates for leadership and candidate selection votes for the primary purpose of allowing time to verify the accuracy of the information, provide candidates with a voters' list, and distribute ballots. Typically weeks or a couple of months in duration, these waiting periods are driven primarily by administrative concerns as opposed to a desire to restrict participation to long- or medium-term members. 3 Notwithstanding the easy accessibility of membership, the parties have not been successful in attracting large numbers of their voters to their membership rolls. Most of the literature on the subject suggests that between one and two percent of voters belong to a federal political party at any given time (for example, Carty, Cross, and Young 2000; Cross 2015a) . Our data are consistent with these earlier findings (see Chapter 3, Figure 3.1) and, as suggested above, we find significant variation dependent on the electoral season . Membership numbers increase, often dramatically, in election years, primarily in response to candidate selection contests, and in years in which a party selects a new leader. The reason behind this growth is that participation in these contests h as traditionally been restricted to those holding a current membership in the party. Indeed, the primary

The Parties' Me1nbers

benefit of party membership is the ability to vote in one of these contests. As discussed in subsequent chapters, while these are internal party votes, they have much broader implications as they largely determine the composition of the House of Commons and greatly reduce Canadians, choice for both the prime ministership and their local representative. It is worth noting that when general population surveys ask respondents whether they belong to a party, the numbers answering «yes» are significantly higher than the numbers released by the parties themselves (for example, Statistics Canada 2015). This rnay result from financial con tributors, campaign volunteers, and strong partisans considering themselves a ((member» without having formally joined the party. This speaks to the evolving definition of member and to the Liberal Party's move to having registered supporters as opposed to members. The parties take very different approaches to organizing segregated groups within their membership. The New Democrats have the largest number of internal «commissions;, six, to which members with particular social/physical characteristics may belong. The Liberals formally recognize four groups in this manner; the Greens, two groups; and the Bloc Quebecois, one. The Conservatives, since their inception in 2003 and unlike the earlier Progressive Conservatives, have explicitly and controversially rejected recognition of this form of membership organization. Debates over whether the party should facilitate groups for youth and women members have been ongoing. Most common among the parties is the organization of youth and women members, commissions and we find these in the Liberals, New Democrats, and Greens (the Bloc has a youth but not a women's group). The Liberals and New Democrats also have recognized groups for Indigenous persons; the Liberals, a group for seniors; and the ND P, a group for those with physical challenges, members of the LGBTQ2+ community, and racialized persons (Liberal Party of Canada 2018a; 2021a, Sec. 31-2; NDP 2018a, Article 10). Among those parties with a formalized youth wing, our data reveal that one in five current members report that at some poi11t they belonged to such a group. Similarly, among women members in those parties with a dedicated group, approximately 18 percent have at some point belonged to their party,s internal women,s group. As discussed in Chapter 6, these

27

28

The Political Party in Canada

groups are typically provided with representational guarantees on party decision-making bodies. Cheng and Tavits (2011), Kittilso11 (2013), and Pruysers et al. (2017) suggest that the presence of these groups and the representational guarantees accompanying them are important mechanisms for ensuring that otherwise underrepresented voices, particularly those of women, are present in party decision-making structures, and Curtin (2008) shows how they can also play a role in ensuring various interests are incorporated in policy debates. SOCIODEMOGRAPHIC CHARACTERISTICS OF MEMBERS

Who are the parties' members? While there are many dimensions to the answer, the logical star ting point is a straightforward description of basic demographic attributes. These include members' gender, age, educational background, income, and visible minority status. We examine these traits for two groups: first, for all members, and second, when appropriate, for members who indicate some level of activism in the party. 4 It is often important to separate out these two groups as it is the active party mem hers who have the most influence in those decision areas devolved to the membership. When illustrating differences by party, we restrict analysis to the three largest parties; our number of Bloc Quebecois and Green Party respondents is too small to allow for confident generalizations. In the 2000 study of party members, Cross and Young found that on almost all measures the membership was not representative of the Canadian population, with members being disproportionately male, old, well-educated, born in Canada, and economically well off (Cross 2004; Cross and Young 2004). Our findings, from data collected almost two decades later, are nearly identical. Party members continue to be overwhelmingly male. As illustrated in Table 2.1, almost two-thirds of all members are men. This has not changed over time; Cross and Young reported almost exactly the same numbers in 2000. There is som e difference among the parties, but none approacl1es an equitable representation of women: 69 .0 percent of Conservative members are men, as are 62. 7 percent of Liberals and 56.9 percent of New Democrats. Consideration of gender by age cohort also suggests that little change may be on the horizon. Men comprise 63.7 percent of the youngest fifth of members (those aged 45 and under). A slightly more encouraging finding

The Parties' Me1nbers

TABLE 2 .1 Sociodemographic characteristics of party members (%)

Conservative

Liberal

New Democratic

All

69.0

62.7

56.9

63.6

300

316

144

839

4.0

5.4

4.9

4.7

30-44

14.1

14.0

17.3

14.8

45 -59

24.0

23 .8

22.9

23.4

60 and older

57.9

56.8

54.9

57.1

Mean

59.9

58.0

57.8

58.6

n

299

315

144

836

High school or less

13.1

7.3

7.7

9.7

Trade/vocational

11.4

10.2

7.0

9.8

College/university

56.4

56.5

60.8

57.6

Postgraduate

19.1

26.0

24.5

22.9

298

315

143

835

2.7

11.3

8.5

7.8

$30,000-59,999

21.4

22.3

35.5

25.2

$60,000-99,999

33.9

30.3

41.8

33.0

$100,000 and more

42.0

36.1

14.1

34.0

295

310

141

817

81.6

81.6

87.9

82.8

293

310

140

818

Gender (male)

n Age 29 and younger

Education

n Family income Less than $30,000

n Born in Canada

n

Note: "All" column includes Bloc Quebecois and Green Party members.

is that the group of members who joined within the past five years is somewhat more balanced, with women comprising 40.9 percent of this cohort, still remarkably distant from gender parity. The picture is less encouraging when we examine the relationship between gender and party activism. Male members are significantly more likely than their female counterparts to be active in party affairs (55.4% versus 46.7%). The result is that 67 .4 percent of active members are men.

29

30 The Political Party in Canada

In terms of age, again the findings are consistent with the 2000 data: the mean age of a party member was 58.6 years in 2016, compared with the earlier finding of 59 years. There is no meaningful difference among parties, as the means range from 57.8 to 59.9 years. Wl1ile there may well be a respondent-distortion skewing data toward older members, this certainly cannot fully account for the extreme underrepresentation of young Canadians in the parties' memberships. Only 5 percent of members are under the age of 30, while almost six in ten are 60 or older. The situation is only marginally better among active party members: the mean age for this cohort is 56.8 years. Party membership seems to be a middle-to-old age avocation among Canadians. Party members are also significantly better educated than Canadians generally. Slightly fewer than one in ten members has only a high school education, compared with approximately one in four voters, while almost one in four members has a postgraduate university degree, compared with one in ten voters (Cross 2019). There are significant interparty differences in this regard. Conservative members are almost twice as likely as Liberals and New Democrats to have only a high school education, while approximately one in four Liberals and New Democrats has a postgraduate degree, compared with just under one in five Conservative members. These findings are again consistent with the 2000 data, with Cross (2004, 21) reporting that «non -university attendees are significantly underrepresented in all of the parties:' Not surprisingly, higher education levels translate i11to high family incomes. One-third of members report family incomes in excess of one hundred thousand dollars, compared with one in five voters (Cross 2019), while fewer than one in ten members has an annual income of less than thirty thousand dollars. There are significant differences among the parties in this regard. New Democrats are significantly less likely to report family incomes in excess of one hundred thousand dollars: 14. l percent compared to approximatately four in ten Liberals and Conservatives. Conservatives are significantly less likely to have family incomes in the lowest category, with 2. 7 percent reporting incomes of less than thiry thousand dollars, contrasted with approximately ten percent of other members. There is no meaningful difference between active and non-active members on income and education levels.

30 The Political Party in Canada

In terms of age, again the findings are consistent with the 2000 data: the mean age of a party member was 58.6 years in 2016, compared with the earlier finding of 59 years. There is no meaningful difference among parties, as the means range from 57.8 to 59.9 years. Wl1ile there may well be a respondent-distortion skewing data toward older members, this certainly cannot fully account for the extreme underrepresentation of young Canadians in the parties' memberships. Only 5 percent of members are under the age of 30, while almost six in ten are 60 or older. The situation is only marginally better among active party members: the mean age for this cohort is 56.8 years. Party membership seems to be a middle-to-old age avocation among Canadians. Party members are also significantly better educated than Canadians generally. Slightly fewer than one in ten members has only a high school education, compared with approximately one in four voters, while almost one in four members has a postgraduate university degree, compared with one in ten voters (Cross 2019). There are significant interparty differences in this regard. Conservative members are almost twice as likely as Liberals and New Democrats to have only a high school education, while approximately one in four Liberals and New Democrats has a postgraduate degree, compared with just under one in five Conservative members. These findings are again consistent with the 2000 data, with Cross (2004, 21) reporting that «non -university attendees are significantly underrepresented in all of the parties:' Not surprisingly, higher education levels translate i11to high family incomes. One-third of members report family incomes in excess of one hundred thousand dollars, compared with one in five voters (Cross 2019), while fewer than one in ten members has an annual income of less than thirty thousand dollars. There are significant differences among the parties in this regard. New Democrats are significantly less likely to report family incomes in excess of one hundred thousand dollars: 14. l percent compared to approximatately four in ten Liberals and Conservatives. Conservatives are significantly less likely to have family incomes in the lowest category, with 2. 7 percent reporting incomes of less than thiry thousand dollars, contrasted with approximately ten percent of other members. There is no meaningful difference between active and non-active members on income and education levels.

The Parties' Me1nbers

One characteristic where the parties, membership reflects Canadians generally is in the percentage born in Canada. Eighty-three percent of members were born in Canada, which is consistent with the overall population. However, those members born outside the country are not reflective of new immigration patterns. Statistics Canada (2016) reports that the top five countries of origin for those immigrating to Canada between 2011 and 2016 are the Philippines (15.2% of all immigrants), India (12.1 %), China (10.6%), Iran (3.5%), and Pakistan (3.4%). While these countries cumulatively comprised 44.8 percent of the recent immigrant population, they represent only 9.2 percent of the parties' foreign-born membership. The largest share of members from these countries is 5.7 percent originating from India, and no members report being from the Philippines or Pakistan. Instead, non-Canadian born members are primarily from the United States and Western Europe. The United Kingdom (17.0%) is the most common birthplace for these members, followed by the United States (14.2%), the Netherlands (7.1 %), and Germany and France (both 5.0%). Thus, cumulatively, these five countries represent 48.3 percent of foreign-born members. Only three of the five countries are in the top ten birth countries of recent immigrants: the United States (2.7%), and the United Kingdom and France (both 2.0%). There is no significant difference between active and non-active members in this regard. Thus, while the parties have a representative number of non-Canadian born members, this cohort is not reflective of Canadian society, as very few members are born in the countries sending the largest numbers of recent immigrants to Canada. 5 Party members are also unrepresentative on many other characteristics. For example, only 1.9 percent are students, only one in five has a child under age nineteen, 43.3 percent are retired from the workplace, and among employed members, almost one-third are public sector employees, compared with 22 percent of working Canadians (Statistics Canada 2021). Public sector employees are disproportionately represented in the ND P, comprising 41.5 percent of their membership. The lack of a membership reflecting Canadian society raises several important challenges for parties. As Heidar and Wauters (2019, 5) write, "Descriptive representation is also of utmost importance when it concerns party members since it has implications for party politics, candidate selection and democratic legitimacy:' Members make important personnel

31

32 The Political Party in Canada

and policy decisions within parties, and different cohorts of voters have different priorities and interests. When some of these groups are not present in party memberships, their views are not provided equitable voice. Gauja (2018) suggests this also makes it more difficult for parties to play an integrative role in shaping divergent group interests into cohesive national policies, long identified as a crucial function of Canada's «broker» . age parties. It is important to note that Canada's parties are not unique in their failure to have a representative membership on many sociodemographic traits. Studies of parties in many Western democracies in recent decades have reached similar conclusions (for example, Whiteley 2009; Widfeldt 1995). As Gauja and van Haute (2015, 195) conclude: «Our comparative analysis of the profile of party members confirms the fact that they are not socially representative of tl1e larger population:' PARTY MEMBERS AND THEIR COMMUNITIES

Beyond the descriptive characteristics of party members, we are interested in how embedded members are in their communities. Traditionally, one of the core benefits parties ascribe to having members is their serving as a conduit of information both between the party and local voters and vice versa. If members are to act as local «ambassadors" of the party, they should not only be representative of the local population but also be engaged in civic life. We asked members whether they belonged to a wide variety of community organizations ranging from neighbourhood and school groups to youth, seniors, hobby, and cultural groups. Overall, twothirds of respondents report belonging to at least one such organization. There is a significant difference in the level of community involvement between active and non-active party members. The active members would potentially be most effective as representatives of the party within their community (and vice versa), and we find these members are significantly more connected to community activities than are passive members. Close to three-quarters of active party members belong to a community group, and 27.5 percent belong to three or more such groups. These connections to other engaged local residents provide m embers both the opportunity to learn about the issues of importance to their fellow citizens and to champion the merits of their party.

The Parties' Members

A significant number of party members are also active in municipal politics. Focusing again on active party members, almost one in five (18.7%) have run for a local public office (mayor, municipal, regional councillor, or school board), and 8.0 percent have served in one of these offices. NDP members are considerably more likely to have sought municipal office, though they have a lower electoral success rate than other members who run less often but serve at similar rates. Federal party members are also active in provincial politics; more than six in ten also belong to a provincial political party. Among active federal party members this number rises to 71.4 percent, with no significant differences among the parties. It is interesting that approximately one-quarter of NDP members do not claim provincial party membership even though, by virtue of the party's integrated structure, they automatically belong at both levels. McGrane (2019, 70) explains that now that members can join directly on the federal party's website, "the federal party was no longer even obliged to inform prospective members that they were simultaneously becoming provincial party members:' The data suggest that many are unaware of this dual affiliation. Two-thirds of active members report belonging to a religion. The large majority of those who belong are Christian, with 43.1 percent belonging to a Protestant faith; 33.5 percent, Roman Catholic; and 6.1 percent, Jewish. No other faith is listed by more than 2 percent of respondents. NDP members are significantly less likely to belong to an organized religion than are Liberals or Conservatives, a pattern we find repeated when examining other party cohorts in later chapters. Among those who do belong to a religion, 35.4 percent attend services at least once a week, while close to another third "hardly ever" attend. Public opinion surveys estimate that approximately 13 percent of Canadians attended religious services on a weekly basis (Hiemstra and Stiller 2016), suggesting that active party members, on the whole, are more ardent parishioners than the general population (see Table 2.2). There are significant differences here among the parties. Almost three-quarters of Conservatives belong to a religion compared with twothirds of Liberals and half of New Democrats. Roman Catholicism is the most common faith for Liberals and New Democrats, while Conservatives are twice as likely to be Protestant than Catholic. Recent studies suggest that the Conservative Party has been particularly successful in

33

34 The Political Party in Canada

TABLE 2.2 Community involvement levels of "active" party members(%)

Conservative

Liberal

New Democratic

At least one

69.7

73.1

72.0

71.8

Three or more

20.6

31.7

29.3

27.5

155

145

75

425

11.3

17.7

31.2

18.7

151

141

77

418

6.7

8.5

8.6

8 .0

164

153

81

452

69.0

68.5

72.2

71.4

158

149

79

440

72.3

66.4

53.2

66.2

159

146

77

432

40.9

30.9

37.5

35.4

115

94

40

280

All

Belong to a community organization

n Ran for local elected office (yes) n Served in local elected office (yes) n Belong to a provincial party (yes) n Belong to a rel igion (yes) n Religiosity of congregants (attend religious service at least once a week) n

Note: " All" column includes Bloc Quebecois and Green Party members.

winning support among evangelical Protestants (Malloy 2013; Rayside, Sabin, and Thomas 2017). Consistent with this, among those who belong, Conservatives are also the most likely to attend services regularly; more than four in ten Conservatives do so. Slightly more than one-third of religious New Democrats and three in ten Liberals attend services at least once a week. The higher degree of religiosity among Conservatives is apparent in their support for more traditionally conservative social policies as described below (see Rayside, Sabin, and Thomas 2017). Taken together, these data present a picture of active party members comprising a group of citizens who are connected to their communities. They are more likely tl1an their fellow citizens to be active in community affairs, far more likely to have sought a local public office and to be active in provincial politics, and more likely to attend religious services regularly. While the numbers of these individuals may be fewer than the parties

The Parties' Members

would like, they do provide them with potential tentacles into local communities both for transmission of local citizens, sentiments upwards within the party and for advocacy of the party and its positions within their communities. These engaged party members also provide a pool of potential general election candidates, and, as discussed in the next chapter, community involvement is highly valued by local party leadership when seeking a general election candidate. PATHS TO MEMBERSHIP

In examining the paths to party membership, we consider members' reasons for joining the party, whether they were asked or recruited to join, and the influence of family members on the decision. One of the challenges for parties in attracting members, and particularly younger and active members, is convincing voters that party activism is an effective way of achieving political goals. Generally speaking, members view parties positively in this regard. When asked to rank the effectiveness of six different activities geared toward achieving political change, 55. 7 percent ranked joining a party first. Views on this question have a modest relationship with party activism, as 53.0 percent of those who rank parties as most effective are active members compared with 46.8 percent who name another activity as most effective. The largest divide in this regard, however, is among different age cohorts. Specifically, younger party members are considerably less likely to see membership as the most effective way of achieving their political goals. As shown in Table 2.3, while 58. 7 percent of members over the age of forty rank parties as the most effective means, only 37.6 percent of those under forty share this view. In their earlier study of young political activists, Cross and Young (2008) found a stark divide on this dimension between those youths who belonged to parties and those who belonged to advocacy groups. Those who participated in group activity were far less likely to see joining a political party as an effective means of achieving change. Considering that membership levels are relatively low, it does appear that there is a significant intrafamily influence on the decision to join a party. Overall, 37.2 percent of respondents report that a member of their immediate family (parent, partner, or child) belonged to a party prior to their decision to join. For example, 24.8 percent report their father was an

35

36 The Political Party in Canada

TABLE 2.3 Efficacy of various political activities for political change, by age(%)

18-39 years old

40 years and older

Join a political party

37.6

58.7

Join a lawful demonstration

15.1

10.9

Join a boycott

15.1

6.3

Join an illegal strike

10.8

2.3

6 .5

3.6

15.1

18.2

93

605

Occupy a building/factory Sign a petition n Note: Percentage ranking activity most effective.

earlier member; 19.8 percent, their mother; and 21.5 percent, their spouse or partner. There is a gender difference here: four in ten women members report that a family member belonged before them compared witl1 onethird of men members. There is also a significant difference by party: 29 .1 percent of New Democrats followed a close relative to party membership compared with 37 .0 percent of Conservatives and 43.5 percent of Liberals. The effect does not stop when current members join: 31.7 percent of respondents report their spouse or partner joined after them and 16.6 percent report that one of their children did likewise. The result is that party membership among family members is dramatically higher than it is among the general population. As shown in Table 2.4, party membership very much appears to be a family affair. A majority of members (55.9%) join their party through their local EDA. This speaks to the importance of parties having vibrant and accessible local associations in order to attract and sustain membership. A positive indicator in this direction is that eight in ten members report that it was easy to contact their local association, and close to nine in ten members had at some point done so. (The parties' local associations are considered fully in Chapter 3.) When members are asked whether they were recruited to join the party, 63.0 percent report that they joined of their own initiative. The experiences of men and women members differ significantly in this regard. Almost seven in ten men join of their own accord compared with 53.5 percent of

The Parties' Me1nbers

TABLE 2.4 Members with a family relative who belonged to a party before the member joined(%)

Father belonged first

n Mother belonged first

n Spouse/partner belonged first n

Child belonged first

n Any of above family members belonged first

n

Conservative

Liberal

New Democratic

All

23.4

29.0

22.2

24.8

320

328

153

888

18.2

23.5

16.4

19.8

318

327

152

884

22.3

24.1

15.3

21.5

314

324

150

876

6.4

11.6

8.1

8.8

312

319

148

864

37.0

43.5

29.1

37.2

311

317

148

861

Note: "All" column includes Bloc Quebecois and Green Party members .

women . Wom en party memb ers are twice as likely as men to have been asked by a family m ember to join their party. Overall, however, when m embers are recruited to the party, it is most often by a candidate for the party's local n o1n ination. This is the case for one in ten members, while 6.6 percent were asked by a fam ily member; 5.8 percent by a friend or coworker; 5.3 percent by a local party official; 3.7 percent by a par ty leadership campaign; and 2.5 percent by a m em ber of th eir provincial or federal legislature. Consistent with these findings, it has long been hypoth esized that party personnel contests, particularly local n omination and n ation al leadership elections, are im portant triggers for local partisans' decisions to join th eir party (Cross et al. 2016; Pruysers and Cross 2016a; Young and Cross 2002a) . As discussed above, participation in these contests h as tradition ally been restricted to party members, and this is one of the central «perks" associated with m embership. The importan ce of these intraparty contests to m embers' decisions to join the party is evident in our data. Three-quarters of respondents say th at supporting a candidate for their local party's nomination was important to their decisio11 to join, and m ore than six in ten say the same for supporting a candidate for th e party's leadership. 6

37

38 The Political Party in Canada

Table 2.5 Importance of material incentives for joining political party(%)

Helpful to getting a government job n

Helpful to my career n

Conservative

Liberal

New Democratic

All

7.3

12.2

9.8

9.8

302

312

144

841

11.9

16.9

10.9

13.7

303

312

146

844

Note: Percentage reporting motivation was "somewhat or very important" to decision to join the party. "All" column includes Bloc Quebecois and Green Party members.

Interestingly, members also see some general benefit to joining their party in terms of personal development. Fifty-six percent say that gaining a better understanding of politics was important to their decision to join, and almost one in four say they were looking to acquire new skills through party membersl1ip. The social aspects ofbelonging are also important. Half of members tl1ought joining would be a good way of m eeting like-minded people, and one-third of members ranked the social activities associated with m embership as an irnportant incentive to joining. As illustrated in Table 2.5, few members joined for prospective "material" gain. Fewer than one in ten thought m embership would help them gain a government job, and 13.7 percent thought joining would help their career. This is a significant shift from the party politics of a century ago when membership was required for many of the vestiges of a more patronage-based system (Carty 1992). Likely reflecting their status as the governing party and thus their greater ability to provide patronage rewards to their supporters, Liberals are more likely to consider these incentives as important to their decision to join. There are two important findings concerning the relationship between the decision to join and members' policy views. First, half of members say that a desire to influen ce policy on a specific issue was important to their decision to join. This is interesting as it suggests that, at least at the time of joining, members viewed party membership as a way of influen cing policy. One can infer from this that these individuals believed members had the potential to influence party policy and that party policy had the ability to influen ce public policy. This is significant both when we consider

The Parties' Members

members' levels of satisfaction with their role, later in this chapter, and when we consider the work of EDAs, in Chapter 3, and particularly the amount of time devoted to public policy discussion. The second related finding is that support for what a party "stands for" appears to be a constraint on the pool of potential members. More than nine in ten members say that support for a party's positions was important to their decision to join. When put together with the findings regarding personnel recruitment, we suggest that these intraparty contests are the trigger for joining but that the pool of potential recruits is largely limited to those who share a belief in the party's general policy perspective. These findings are generally consistent with those of Young and Cross (2002a, 549) who conclude that "simply put, individuals are available to be recruited into party affairs via social networks only when they already sympathize with the party's ideology:' MEMBERS' IDEOLOGY AND POLICY VIEWS

The ideological disposition of party members is consistent with the broad policy positions staked out by their parties. Traditionally, the Canadian party system was described as only weakly ideological with an emphasis on brokerage or accommodative-style politics at the expense of parties taking rigidly consistent left/right positions both across policy fields and over time (Carty 1992; Johnston 1992). With the demise of the Progressive Conservative Party at the end of the twentieth century and the rise of Reform/Canadian Alliance and, subsequently, the new Conservative Party, there was evidence that the emerging system was more saliently defined by ideological differences among the parties and their supporters (Carty, Cross, and Young 2000; Cross and Young 2002). Of the three national parties, the New Democrats are generally perceived to be the furthest left of centre; the Liberals, a centre-left party; and the Conservatives falling to the right of centre (Christian and Campbell 1983), and these perceptions have been empirically validated with evidence from campaign policy commitments (Cochrane 2010; Koop and Bittner 20 13, 320). An unresolved question for many is whether the Conservative Party continues to occupy a centre-right placement, consistent with the Progessive Conservative tradition, or has moved into a consistent hard-right position more closely resembling the Reform/ Canadian Alliance side of the party. The views of

39

40

The Political Party in Canada

TABLE 2.6 Members' placement of themselves and their party on 0-10 left-right scale

Self-placement n

Party placement n

Mean difference between self and party n

Conservative

Liberal

New Democratic

7.41

4.80

3.06

294

299

147

7.31

4.95

3.74

296

279

145

1.03

0.96

1.36

292

272

145

party members suggest that the latter may be the case resulting in increased ideological distance between the m embers of the Liberals and NDP on the left and the Conservatives on the right. We asked members to place themselves on an eleven-point scale reflecting their general political views, with zero being furthest left and ten furthest right. Not surprisingly, as shown in Table 2.6, New Democrats place themselves furthest left with a 1nean score of 3.06. Liberal members place themselves almost precisely at the centre point with a mean score of 4.80. Conservative members on average place themselves at 7.41 on the scale. Several observations can be made from these findings. First, the general ideological disposition of the membership of each party reflects the place of the party in the party system. At least in a general sense, ideological consistency between a party and its members exists. This is consistent with our earlier suggestion that the pool of potential members available to the parties is limited to those supporting their ideological disposition. Second, the Conservative membership is the furthest, on average, from the scale's centre point. While New Democrats, on average, place themselves slightly less than two points to the left of centre, the Conservatives' mean is almost two-and-one-half points to the right of centre. Third, there is considerable ideological distance covered by the parties, with a total of 4.35 points, representing 40 percent of the scale, separating New Democrats from Conservatives. Fourth, the Liberals continue to be a centre-seeking party in terms of their members' views.

The Parties' Members

Further examination within each party,s data illustrates additional findi11gs relating to the ideological cast of their members. Among Liberals we find the vast majority, 75.6 percent, place themselves on one of the five centre points of the scale ( ranging from three to seven); 13 .1 percent place themselves on the far left (0-2), and 11.4 percent on the far right (8-10). Liberal members do not see themselves as significantly ideologically distinct from their party. We asked members to place their party on the same scale, and, on average, Liberals place themselves 0.96 of a point away from the party. Slightly less than half of members place themselves and their party at the same point, while one-third place themselves one point away, 13.6 percent two points away and slightly fewer than one in ten three or more points distant. Approximately one in five members place the party to their left and one-third to their right. The mean score for the party of 4.95 was 0.15 of a point to the right of members' average placement. There is somewhat more variance among Conservative members, reflecting the different ideological traditions of the two predecessor parties. While the largest cohort, 53.7 percent, place themselves on the far right (8-10), just less than half, 44.6 percent of members, place themselves on one of the scale's more moderate points (three to seven) . There is evidence of significant ideological «extremism' in this membership: 10.9 percent placed themselves at ten and 13.9 percent at nine. There is consistency between how members view the party's general ideological placement and that of themselves. On average, members place themselves 1.03 points away fro1n their party. Thirty-eight perce11t of members place themselves and the party at the same point, one-third place themselves one point away, one in five Conservative members place themselves two points away, and 8.2 percent three or more points distant. One-third of Conservative members see the party as slightly to their left and three in ten see the party to their right. The mean score for the party, 7.31, is 0.10 of a point to the left of the members, average placement. New Democrats report the largest chasm between their own views as compared to those of their party. New Democrats place their party at 3. 74 on the scale, and on average place themselves 1.36 points distant from their party. The NDP's membership includes a large cohort of ideologically «extreme" members, with just shy of half, 45.6 percent, placing themselves

41

42

The Political Party in Canada

at one of the three furthest left scores (O to 2), while 47.6 percent place themselves toward the centre, between three and seven. Twenty-eight percent of members place themselves and the party at the same position, one-third of NDP members place themselves one point away from the party; 22.1 percent, two points distant; and 16.5 percent place themselves three or more points apart from the party. Half of members place themselves to the left of the party contrasted with fewer than one-quarter who perceive the party's positions to be to their left. The mean placement of the party, at 3.74, is 0.68 of a point to the right of members' self-placement. It is important here to keep in mind that the data were collected after the 2015 election, when the NDP's campaign and its leader positioned themselves more toward the centre in an attempt to appear as a credible governing alternative (McGrane 2019, Chapter 4). This move was believed to be unpopular with many of the more ideological - and longserving - members of the party, particularly, in retrospect, as the party did not achieve the anticipated electoral breakthrough. The divide between the party's newer and longstanding members is apparent in the data. Using the 2008 election as a cut-off point, 56.9 percent of NDP respondents joined the party before this election and 43 .1 percent afterward. The first group placed themselves considerably to the left of the party's newest members, at 2. 77 compared to 3.44. Both groups placed the party at a similar position (3.70 compared to 3.81), but the earlier members see themselves almost a full point to the left of their party contrasted with newer members, who placed themselves just over onethird of a point to the left of the party. This suggests that those joining in the run-up to the 2015 election were somewhat more ideologically moderate, consistent with the then-leader's push to make the party an acceptable governing alternative to more voters. The difference for Conservatives is less stark but also illustrates that longer-term members of the party are more ideological. Those who joined in 2008 or earlier placed themselves at 7.59 compared to 7.11 for newer members. The latter group views themselves as very slightly to the left of the party (0.04) while longer-term members see themselves 0.17 points to the party's right. The difference between members' own views and their views of their party's position is also apparent when we ask members about their level

The Parties' Members

of satisfaction with the ideological orientation of their party. On a fourpoint scale, with one being not at all satisfied and four being very satisfied, New Democrats have the lowest mean score at 2.71 compared with Conservatives at 2.91 and Liberals at 3.15. We find that gender matters differently among the parties. There is no difference between the general ideological predisposition of Liberal men and women members. There are substantial gender differences among New Democrats and Conservatives. Women NDP members place themselves to the left of their male counterparts, at 2.85 compared with 3.22. This may be somewhat attributable to the higher rate of union membership among men, as these members place themselves slightly more than twotenths of a point to the right of non-union members. Among Conservatives, women place themselves one-third of a point to the left of their male counterparts (7.18 versus 7.51). While the data are not fully comparable, there is much consistent here with the earlier findings from the 2000 party members survey conducted by Young and Cross (they used a ten-point scale ranging from one to ten). The national party system at that time was more fragmented, with the presence of both the Reform/Canadian Alliance (CA) and the Progressive Conservative (PC) parties. NDP members were again significantly to the left, placing themselves on average at 3.47 (with a midpoint of 5.5), while Liberals were solidly in the middle, at 5.3. The division on the right between the PCs and the Reform/CA members was apparent with PC members placing themselves at 6.77; Reformers, at 7.33. Again, the most significant difference between self and party placement was among New Democrats, with members placing their party a full point to the right of themselves (4.48). This appears to be an ongoing tension in the party between a more left-leaning membership and its perception of the party as significantly more ideologically moderate (for more on ideological divides within the NDP, see Erickson and Zakharova 2014; McGrane 2019). Comparison with the earlier data also indicates that the ideological predisposition of current Conservative Party members is far more consistent with that of earlier Reform/CA party members than of Progressive Conservatives. In fact, current Conservatives are even slightly to the right of Reform/CA members on average, and one in four place themselves on the far right (9 or 10) compared with one in five Reformers.

43

44 The Political Party in Canada

Given the divide between active and non-active party members, we are interested in considering whether there is a relationship between activism and the distance one perceives between their ideological positioning and that of their party. We might expect higher levels of activism among those who have greater ideological affinity with their party. Here we compare the absolute difference between where active and inactive members place themselves and their party on the left/right scale. Our data suggest no significant relationship between a member's perceived difference between their and their party's ideological positioning and whether they are active in party affairs. Non-active members on average place themselves 1.04 points away from their party while active members place themselves 1.10 points away. This is consistent with Webb, Bale, and Poletti (2020), who find that those who are motivated by «ideological impulses,, to join UK parties are not particularly high-intensity activists. Beyond left/right perceptions, we asked members for their views on a series of policy questions relating to the environment, social tolerance, the economy, and federalism. On the first three dimensions, for which we included ten separate questions, the parties consistently line up the same way: the NDP furthest to what might be considered the «progressive" position, the Liberals in the middle, but typically closer to the NDP than to the Conservatives, and the Tories consistently the most «conservative." Examples of these questions and the responses by party are reported in Table 2.7. These data support the argument that members understand the general policy positioning of their party, and, as discussed above, this is an important factor/constraint in deciding to join a particular party. On the environment we find overwhelming support for stronger protective measures a1nong New Democrats and Liberals while Conservatives are split 011 the question. Similarly, a strong majority of New Democrats and Liberals support a carbon tax while only 14.9 percent of Conservatives agree. There is a common perception that the Conser vatives suffered an electoral penalty for failure to propose a strong environmental-protection policy in recent elections, and at their 2021 policy convention, members, to the chagrin of many party leaders, declined to support a stronger position on climate change (see Tasker 2021a) . On questions of social tolerance there is over whelming support for a women's right to choose abortion among Liberals and New Democrats,

Table 2.7 Examples of the range of party members' policy views(%) Strongly agree

Somewhat agree

Neutral

Immigrants should be required to adapt to Canadian Customs

25.9

7 .8

Liberal

59.5 30.7

36.3

17.1

New Democratic

20.0

28.0

27.3

Conservative

Stronger measures should be taken to protect the environment Conservative

10.3

38.4

29.4

Liberal

44.9

38.4

11.8

New Democratic

67.3

24.0

7.3

23.1

12.0

22.1

Liberal

5.6

6.8

16.1

New Democratic

7.3

3 .3

7.3

21.9 40.5

18.7

Liberal

9.4 18.1

25 .9

New Democratic

48.0

34.0

12.7

Liberal

42.7 21.8

36.6 29.6

New Democratic

13.3

26.7

15.5 25 .9 21.3

Conservative

40.6

26.1

11.0

Liberal

65.2 70.7

19.6 14.0

8 .7 10.7

Same-sex marriage should be prohibited by law Conservative

Income and wealth shou ld be redistributed toward ordinary people Conservative

People who break the law should be given tougher sentences Conservative

Women should be free to decide on matters of abortion

New Democratic

Somewhat di.

46 The Political Party in Canada

while support among Conservatives is almost twenty points lower. Similarly, while New Democrats and Liberals express strong support for samesex marriages, more than one-third of Conservatives believe they should be prohibited. The disproportionate number of men in the party,s membership is important here, as Conservative men are 20 percent more likely to oppose same-sex marriages and 57 percent more likely to oppose a woman's right to choose than are women members. The divide among Conservatives, with sizable support among members for socially conservative positions, has been apparent in recent leadership contests (Thomas and Sabin 2019). Support for these positions appears to be stronger among members than it is among the party,s 2019 general election supporters. Liberal and New Democratic members are divided on whether there should be tougher criminal sentencing laws, while only one in twenty Conservatives disagrees with this proposition. The one social tolerance issue where all three parties are distinct from each other is their view on whether new immigrants should be required to adapt to Canadian customs. Support for this proposition is strongest among Conservatives at 85.4 percent compared with agreement from two-thirds of Liberals and half of New Democrats. Similar patterns are found regarding economic policy. On the general question of whether government should promote redistribution of wealth in favour of ordinary people, there is substantial difference among the three parties. The New Democrats are most in favour of this sort of economic intervention with 82.0 percent expressing support contrasted with 58.6 percent of Liberals and slightly fewer than one-third of Conservatives. This pattern is largely consistent throughout this issue domain. For example, one-third of Conservatives believe the government should abstain from intervening in the economy compared with one in six Liberals and fewer than one in ten New Democrats. Differences are also found in attitudes toward free trade. Three-quarters of Conservatives support government promotion of free trade and globalization compared with 55.5 percent of Liberals and one in four New Democrats. The New Democrats are also most likely to support measures to protect Canadian businesses from foreign competition, at 69.3 percent, followed closely by the Liberals, at 63.4 percent, compared with half of Conservatives.

The Parties' Members

The fourth dimension, federalism, does not fit neatly on a left/right continuum, and this is evident in the members' views. Conservatives are most likely to support devolution of more powers to the provinces (50.2% ), but here they are followed by the New Democrats, at 42.0 percent, and the Liberals, at 37 .1 percent. This is not surprising given the Liberals' traditional dominance of federal politics (Carty 201 5) and their resulting support for a strong national government. The Conservatives are, however, least likely to support recognition of Quebec as a distinct society (20.0%) compared to one-third of Liberals and 45.3 p ercent of New Democrats. MEMBERS AND PARTY ACTIVISM

Beyond considering how members come to belong to a political party and who they are in terms of their demographic characteristics and ideological perspective, it is important to examine the scope and n ature of the party activities they engage in. Here we separate out their activities during and between election campaigns. The overarching finding is that many members are largely inactive in party affairs between election campaigns. When asked how many hours they devote to party activity in an average month outside of campaign periods almost h alf of members report zero while another three in ten members spend some time but less than one hour on average. Accordingly, fewer than one in four members commits one hour or more of their time per month to party activity on an ongoing basis. Similarly, we asked m embers how many party functions they attended in the year leading up to the general election, and 35.0 percent responded they h ad not attended any functions . At the other end of the spectrum, one in five m embers h ad attended four or more party meetings during that twelve-month period. The findings are similar in regard to specific non-campaign activities of m embers in their local party associations. Slightly fewer than four in ten have ever participated in the election of their EDA executive, and only three in ten h ave ever participated in a policy-related discussion within their party. Perhaps the most committed members are the 22.3 percent who have held an elected position in their local party association. Considering the motivations for membership, it is not surprising that the one activity that draws the greatest participation levels among members

47

48 The Political Party in Canada

is the selection of their local candidate. Half of members report that they have attended a local party nomination 1neeting. Leadership contests also draw greater participation: four in ten members have voted in one of these contests. An important difference here is that participation in nomination contests requires members to attend a party meeting outside of the home whereas voting in leadership contests is now primarily by post and online. In this sense it is encouraging that nomination meetings draw the highest participation rates. That membership votes are largely determinative of nomination outcomes likely influences this. Members are more likely to participate in events where their participation is consequential, such as leadership and candidate selection, as opposed to local party policy discussions that they often view as having little impact on party and, ultimately, public policies. The one activity that a strong majority of party members participate in is contributing financially to their party. We asked members whether they donated, beyond paying their membership fee, and 65.9 p ercent responded affirmatively. Many of these members are otherwise inactive in party affairs and their contributions are often solitary acts, as only four in ten contributors report having attended a party fundraising activity. The large number of «cheque-book" members, who contribute financially but devote little or no time to the party and that form a significant cohort of party members, may account for the energies the parties expend on trying to attract and retain members. In Chapter 5, we examine those who contribute to the parties, but at this point we simply confirm the earlier finding of Jansen , Thomas, and Young (2012) that a large proportion of the parties' contributors comes from their membership rolls. Members are generally a «loyal" group. While more than nine in ten members voted for the party they belonged to in the 2015 election, there is variance between the parties. Among the three largest parties, the NDP experienced the most slippage, losing the votes of 10.8 percent of their members compared with 7.1 percent for the Liberals and 5.2 percent for the Conservatives. New Democrats in this election seemed to have suffered from «strategic voting" as most of their «defectors" moved to the Liberals.7 Liberal campaigns regularly make a late-campaign pitch to voters on the left to coalesce around their candidates as the only plausible

The Parties' Me1nbers

alternative to a Conservative victory (Clarkson 2005). In addition to this slippage at the polls, loyalty is not absolute in intraparty affairs either. Given the mobilization that occurs during personnel selection contests, Cross and Pruysers (2019b) reveal that «nomination losers;' those members who supported a losing nomination candidate, are less likely to report that they will still be a member in three years or engage in high-intensity activism during the subsequent general election campaign. This speaks to the high degree of personalization found in candidate selection, as candidates recruit large numbers of new members who have stronger affinity to themselves than to the party. Members, overall, are more active during election campaigns when local party associations and the associated campaigns of the local candidate come to life. When asked how much time they devote to party activity duri11g election campaigns, the number of inactive members drops to three in ten, and one in four reports spending three or more hours per week on campaign activity. As illustrated in Table 2.8, we asked members

Table 2 .8 2015 campaign activities engaged in by party members(%)

Persuaded others to vote for the party

n Distributed campaign materials

n Worked in a party phone bank

n Canvassed voters on behalf of party

n Attended a party rally

n Contributed financially to the party

n

Conservative

Liberal

New Democratic

58.9

59.2

67.6

61.2

299

316

145

843

21.3

23.2

39.0

25.4

300

315

146

844

10.1

15.6

16.3

13.3

298

315

147

844

19.7

25.1

36.1

25.8

300

315

147

844

38.2

41.8

49.7

41.6

301

316

147

847

60.8

61.3

87.0

67.2

301

315

146

845

Note: "All" column includes Bloc Quebecois and Green Party members.

All

49

50

The Political Party in Canada

whether they engaged in a battery of election-related activities, and nine in ten report having participated in at least one activity, while almost half l1ave engaged in at least five of these activities. While contributing financially remains the most common method of participation, it is followed closely, at 61.2 percent, by members working to persuade others to vote for their party. This finding highlights the importance of members being engaged in their communities as these activities increase the number of potential voters that members come into meaningful contact with. This speaks to the community-based form of campaign activity that Gauja (2018) characterizes as increasing in prominence in other Western democracies. Other activities that a sizable number of party members engage in during campaigns include posting a yard sign, attending a party meeting, canvassing on behalf of the party, and distributing campaign materials. Many members are also actively supporting their party on social media during campaigns; four in ten members have posted campaign materials supporting their party online. (For a general discussion of online political activity, see Small et al. 2014.) While there is little interparty difference in terms of activism rates outside of election campaigns, we do find that New Democratic m embers are more likely to be engaged in campaign activities than are their Liberal and Conservative cou.n terparts. For example, while more than three in ten Liberals and Conservatives devoted no time in the average week to campaign activities, the same is true for only approximately one in five New Democrats. Similarly, almost half of New Democrats (48.3%) participated in six or more of the 2015 campaign activities they were asked about. This contrasts with 37.3 percent of Liberals and 29.0 percent of Conservatives. For every one of the activities included in Table 2.8, during the 2015 election campaign ND P members participated in the highest numbers, consistently followed by the Liberals and then the Conservatives. For New Democrats this reflects their highly competitive position at the outset of the campaign. Having formed the Official Opposition after the previous campaign, the party entered this election with a lead in public opinion surveys, and a general consensus that 2015 represented their best chance to date of forming a national government. This unique competitive position, we suspect, encouraged members to volunteer in high numbers. The Conservatives, on the other hand, having been in government for

The Parties' Members

nearly a decade, were seeing the first signs of a withering of enthusiasm at the grassroots level. Cross (2016) chronicles a relative decline in membership numbers and :financial contributions to Conservative riding associations in the lead up to the 2015 election and attributes this to a growing weariness within the party's grassroots and a sense that the party on the ground was neglected during the Conservative's tenure in government. This is a common malady suffered by parties at the end of a relatively long run in office. (For more on this, see Chapter 3.) MEMBERS' VIEWS OF PARTY DEMOCRACY

Given the relatively low levels of party membership, it is important to consider members' levels of satisfaction with their experiences in the party and their views toward the influence they have in party decision-making. Our findings are rather mixed. On the whole, members report modest levels of satisfaction with their experiences. However, when asked how the role assigned to members compares with other groups within the party, their views indicate that they do not perceive members as being sufficiently influential. Given this :finding, it is not surprising that there is a difference between the power they view members having in the party and the degree of influence they believe members should have. To provide a general view, members were asked how their experiences measured up against the expectations they held when they first joined. On a four-point scale, with one indicating that their experience exceeded expectations and four representing membership not at all meeting their expectations, we find that while most members are generally satisfied, a not insignificant number are dissatisfied. Six in ten members placed themselves at two on the scale, which was described as membership generally meeting their expectations, while approximately one in three indicated that their experiences only partially met or completely failed to meet their initial expectations. With an overall mean of 2.28, there is not significant difference among the parties. To probe more deeply, we asked members their opinions on statements regarding decision-making within the party. Overall, we find that while there is variance within each party, the data suggest that members on the whole are not strongly dissatisfied with party democracy. For example, on a five-point scale, with one being strongly disagree and five

51

52

The Political Party in Canada

strongly agree, men1bers fall slightly on the disagree side (2.64, with a midpoint of 3) of the statement that in their party «a few people have all the power and the rest of us have no say." There is a difference here by party, with Conservatives most likely to agree (2.93) compared to Liberals (2.60) and New Democrats (2.46). The somewhat more negative opinions of the Conservatives may be a result of the party serving in government for most of the decade prior to the survey and having only recently returned to the opposition benches. It is common for parties to become more centralized in decision-making during long runs in government and for grassroots members to see this concentration of power as a cause of subsequent electoral defeat (Cross and Blais 2012a, Chapter 5; Harmel and Janda 1994). Members' general, though not overwhelming, satisfaction with their role in the party is also evident when asked about their views regarding the rights and obligations of members. On a scale from one to four, with four representing very satisfied, respondents average 2.79 with no significant differences among the parties. Finally, we asked respondents whether they expect to still be members in three years, and only one in ten answered "no:' while seven in ten indicated they expect to still belong to their party. Generally, it appears that members are moderately satisfied with their role in the party. This is interesting given that much of the literature suggests that stagnant rates of party membership are, at least in part, a result of members' dissatisfaction with their role in party decision-making (for discussion of this, see Cross and Katz 2013). The argument is that this leads to high levels of membership exit and a declining number of new recruits. The data reported above do not necessarily support this hypothesis. An important caveat is that these data capture the views of those who have retained their membership in the party and not those who have joined and subsequently chosen to exit. On average, members report having belonged to their party for fifteen years, with one-half having joined prior to the 2006 election and one-third after the 2011 election. It is highly conceivable that there exists a separate group of former party members who were more disgruntled with the role assigned to members and have left their party and thus are not captured in these data. Our final consideration concerns the views of party members toward the roles and degrees of influence various actors exercise within the party. As described in Chapter 1, parties are not monolithic organizations, and

The Parties' Members

''power» and authority a.r e distributed among different groups of individuals. This can be reflected in a centralization versus decentralization struggle between local and national party associations, often described in the literature on stratarchy in the Canadian case (Carty 2004; Carty and Cross 2006). It can also potentially exist among different groups at the same level, for example, a party's national executive and its national policy convention. Different groups of individuals might also contest for power within the party, such as MPs and political professionals including pollsters and campaign consultants. We test for members' views concerning how influential they believe each of these groups is and their views as to the appropriateness of the degree of influence the different groups wield within the party. We also asked about the perceived influence of three demographic groups that, as described above, parties often recognize with dedicated subgroups or representational guarantees on party decision-making bodies: women, youth, and visible minorities. Members were asked to rank the influence each group has on a scale ranging from one to ten, with one representing "very little influence" and ten meaning "very influential:' As displayed in Table 2.9, members rate the party leader as having the most influence, and this is consistent across all parties. Members' views in tl1is regard are in line with a great deal of literature on the Canadian party system finding parties to be leader-centric (for example, Godbout 2020; Marland 2016; Savoie 1999b). Comparative studies of political party organization across Western democracies have also regularly found Canadian leaders to be among the most powerful ( O'Malley 2007). Interestingly, the top four positions in terms of degree of influence are identical across the parties, with the national executive viewed as the second most influential followed by MPs and national party conventions. It is important to note that each of these are national groupings representing centralized power (with the caveat that party caucuses, conventions, and executives are largely composed of local and regional representatives). Members do not see themselves as having significant influence in the party compared with these other groups. Liberals rank members twelfth of twelve; New Democrats, eleventh; and Conservatives placed themselves in tenth position.

53

Table 2.9 Members' views of "influence" of various groups within party Conservative

Liberal

New Democrat

Influence they have

Influence they should have

Influence they have

Influence they should have

Influence they have

Party leader

8.00

7.80

8.38

8.06

8 .14

National executive

7.33

7 .36

7.71

7.45

7 .57

Party conventions

6.76

7 .27

7.21

7.22

7 .03

MPs

6.88

7 .63

7.44

7.62

7 .26

EDAs

5.98

6.92

6.38

6.94

5.94

Party members

5.03

6.69

5.54

6.79

5.08

Experts/Consultants

5.96

5.44

6.31

5.85

6.04

Youth

4.70

5.45

5.64

6.48

5.14

Women

5.76

6 .16

6.26

6.79

5.87

Visible minorities

5.51

5.56

5.95

6.56

5.55

Business

6.63

6.06

6.77

5.94

4.70

Unions

4.94

3 .83

6.01

5.43

6.64

n

282

273

299

300

138

lnfluenc shoul

Influence Differential Party leader

0.24

0.33

National executive

0.00

0.28

Party conventions

-0.50

0.00

MPs

-0.72

-0.16

EDAs

-0.94

-0.56

Party members

-1.66

-1.30

0.54

0.47

Youth

-0.76

-0.83

Women

-0.45

-0.51

Visible minorities

-0.05

-0.59

Business

0.62

0.84

Unions

1.18

0.64

n

265

291

Experts/Consultants

Note: Negative influence differentials reflect members' views that group has less influence than it should have. "All" column inc Green Party members.

56 The Political Party in Canada

In terms of tl1e influence of our three demographic groups, they rank toward the bottom of the list with women ranked eighth, visible minorities ranked ninth, and youth in twelfth place. There is little varia.n ce in this regard, as members of all three parties rank women ahead of visible minorities, followed by youth in terms of their degree of intraparty influence. The presence of internal commissions for these sociodemographic groups within the Liberal Party and the NDP, and their accompanying representational guarantees, does not result in members believing they have significantly greater influence within the party when compared to the Conservative Party, in which there are no such formal internal groupings. There is a dramatic, though not unexpected, difference among the parties in terms of the influence wielded by business groups and trade unions. New Democrats rank unions as the fifth most powerful group and businesses twelfth, while the Liberals and Conservatives both rank business groups the fifth most influential and place trade unions toward the bottom of the list. Members were also asked how much influence each of these groups "should have:' The overall ranki11gs are not dramatically different from the influence that members perceive the various groups actually holding. The party leader is again ranked first overall, though second by New Democrats. The leader is followed by MPs for Liberals and Conservatives, with New Democrats expressing the view that MPs should be most influential. The national executive is ranked third and the party convention remains in fourth position. The most significant differences are in the placement of party members and political professionals such as pollsters and consultants. In all three parties, political professionals are viewed as the sixth or seventh most powerful group while members placed them in eleventh position in terms of the influence they should possess. It is members themselves where the largest difference exists. While overall they place themselves eleventh of the twelve groups in actual influence, they place themselves sixth in terms of how much power they should have. When we consider the influence differential, a value representing the score on the scale for actual power (out of ten) minus the score for desired power, it is greatest for "members" in all three parties. This is the only •

The Parties' Me1nbers

grouping for which the score differential exceeds one full point across all parties. The differentials also indicate consistency in viewing local party associations as well as youth, women, visible minorities, and MPs as having less influence than they should in party affairs. It is somewhat surprising that Conservative Party members who have consistently resisted the creation of special intraparty groups for women, youth, and minority members believe them to have less influence in party decision-making than they should have. In addition to the leader, three groups are consistently viewed as exercising more influence than desirable. These are trade unions, business and political professionals. It is interesting that New Democrats view trade unions, one of the party's core constituent groups, as exercising too much influence. There has been considerable debate in recent years within the party regarding the appropriate role of organized labour in party influence (Jansen and Young 2009; McGrane 2019), with dissension emerging between those seen as representing a new left concerned primarily with «postmodern" issues such as identity and the environment and those prioritizing organized labour's traditional concerns. There is a significant divide in the party's membership in this regard with union members relatively satisfied with the role of organized labour, with a differential score of 0.18 compared to 0.98 for members who do not belong to a union. CONCLUSION

Young and Cross (2002b) suggest that both the organization and practice of decision-making in Canada's political parties has shifted from what they term a representational-based model to a plebiscitarian form. The earlier model prioritized the aggregation of members into groups to ensure representation within central decision-making bodies of what each party considered key civil society segments. In practice, this meant «providing representation for groups through election of representatives to participate in semi-deliberative bodies such as national leadership and policy conventions" (Young and Cross 2002b, 676). There were three main characteristics to this model: first, a reliance on representative, as distinct from direct, internal democracy; second, aggregation of members into groups based on ascriptive characteristics; and third, a differentiated conception

57

58 The Political Party in Canada

of party membership. The result was that members' participation was largely mediated through various recognized and formal subgroups within the party (based on, for example, age, gender, geography, trade union membership). There was little or no direct relationship between the member and the national party and its decision-making processes. Party conventions and leadership electorates in all three major parties during this period included mandated numbers of representatives of these groups (see Carty, Cross, and Young. 2000). The plebiscitarian model is based on the conception of egalitarian membership with all members having equal voice in party decisionmaking. It prioritizes direct participation and individual equality over representational considerations and in this sense is consistent with a personalization of politics. Consistent with this is a rejection of group-based representation and a hesitancy to organize members along ascriptive characteristics. Methods of direct democracy replace deliberative gatherings of representatives, and members have direct, individualized roles rather than having their participation mediated through subgroups within the party (Seyd 1999; Young and Cross 2002b). While the shift toward the plebiscitarian model is not complete, it is evidenced in things such as the federal parties' universal adoption of onemember-one-vote forms of leadership selection. This is also apparent in the Conservative's and Bloc's unwillingness to structure their internal organizations around segmented groups of members and reluctance to provide representational guarantees in this way. The Liberals have also moved in this direction through their "opening up" of national policy conventions and personnel selection contests to any interested supporter - thus, no longer ensuring that women, youth, and the regions have "equitable" representation. Similarly, the New Democrats have eliminated provisions ensuring that votes from trade-unionists have significant influence in leadership contests. At the same time, largely through the expansion of the Internet, the national parties have dramatically increased their direct, unmediated com munication with members. Emails are sent frequently, often daily, to individual members with the parties' message of the day and often asking members to provide feedback on central-party-initiated questions (often dealing with issues of public policy). And many now join the party directly

The Parties' Me1nbers

through the national offices' websites. All this is consistent with a «personalization" of party membership that Gauja (2018) has identified as occurring in many Western democracies. When the membership as a whole is empowered to make decisions directly, which in many ways may be a preferable form of intraparty democracy, the degree of representativeness of that body takes on greater importance. Our data suggest that parties are failing in this regard. As discussed in this chapter, membership in all the parties is skewed toward older, well-educated, economically privileged Canadians largely of Western ancestry. The parties' memberships do not reflect Canadian society generally, and thus many voices are left unheard when they choose general election candidates and their leaders, when they meet to debate and adopt public policy planks, and when they communicate with voters in communities across the country.

59

The Parties' Me1nbers

through the national offices' websites. All this is consistent with a «personalization" of party membership that Gauja (2018) has identified as occurring in many Western democracies. When the membership as a whole is empowered to make decisions directly, which in many ways may be a preferable form of intraparty democracy, the degree of representativeness of that body takes on greater importance. Our data suggest that parties are failing in this regard. As discussed in this chapter, membership in all the parties is skewed toward older, well-educated, economically privileged Canadians largely of Western ancestry. The parties' memberships do not reflect Canadian society generally, and thus many voices are left unheard when they choose general election candidates and their leaders, when they meet to debate and adopt public policy planks, and when they communicate with voters in communities across the country.

59

The Parties' Local Associations

Electoral district associations or EDAs (also known as constituency or riding associations) are the organizational apparatus that allow national political parties to have a formal, organized presence in each of the single-member districts across the country. National parties establish associations in each electoral district in order to remain connected to their local members, to ensure a presence across the country, and to support local candidates in the general election. Taken together, these grassroots associations make up the fundamental building blocks of party life in Can ada. As Carty and Eagles (2005, 2) write, "Constituencies and the political organizations and processes that they frame, stand at the very core of Canada's political life:' Most partisans and activists engage with their respective parties only at the local level and, therefore, through the EDA. While there are, of course, opportunities to participate at other levels (at a national convention, for instance), the vast majority of activist activity occurs at the grassroots, through candidate nominations, policy discussions, and election volunteering. What parties do at the grassroots, how they organize, and their makeup is therefore critical to the success of the broader party. With that said, EDAs have typically been overlooked in traditional analyses of Canadian party and electoral politics (Cross 2016). For much of Canada's electoral history, local constituency associations were outside the purview of state regulatio11. These local organizations were viewed as private associations that had no reporting obligations and no requirement to be registered with state agencies ( unlike the national or provincial branch of the party). While local candidates have b een subject

The Parties' Local Associations

to spending limits and other requirements for decades (campaign spending limits, for example, were introduced in 1974), constituency associations did not become legally recognized entities until 2004 with the introduction of Bill C-24. This new legislation requires local associations to register with Elections Canada, limits who can contribute financially and how much, and necessitates proper financial reporting and transparency. What exactly is an EDA? According to Elections Canada, an electoral district association is simply defined as "an association of members of a political party in an electoral district,, (Elections Canada 2018). In order to perform essential tasks such as accepting contributions, issuing tax receipts, and transferring funds, the association must formally register with Elections Canada. To register with Elections Canada an association is required to have a financial agent and an auditor, and must submit a signed declaration from the party leader certifying that the association is indeed the party's EDA in the district (a party can only have one registered EDA per district). While necessary to perform its duties, registering with Elections Canada places an administrative burden on EDAs, as we will see later in the chapter. Reporting, for example, includes notification of any changes to the ED.A's makeup (financial agents, president, etc.), extensive accounting statements of assets and liabilities, reports regarding any nom ination contests, details on all financial transactions, and data of contributions to the EDA (also sent to Canada Revenue Agency). Although these associations serve as the fundamental organizational building blocks for all Canadian parties, and while all federal associations are subject to the same legal requirements, not all associations function exactly the same. While they typically share a similar organizational structure with three major components (membership, executive, and president; Koop 2010, 2011), they vary considerably in terms of diversity and inclusion, wealth and size, health, vibrancy, professionalization, and relationship to the central party. Carty (1991), for example, finds that the number of members in local associations ranges from two to three in some s1nall ridings to thirteen thousand regular members in others. Additionally, some local associations have "virtually no money" while others have incomes of "over $40,000 and nest eggs approaching $100,000,, (Carty and Eagles 2005, 17-18). The stratarchical bargain that provides authority and (some) autonomy to local party associations ( Carty 2002; Carty and Cross 2006)

61

62

The Political Party in Canada

means that there is no single template for what an EDA will look like, how it will interact with the national party, or how it manages its local business and affairs (despite party bylaws providing rules and guidelines on many of these matters). This means that there is great variation not only between parties in Canada but also within parties. As a result, the Liberal Party association in Ottawa South may be very different from the Liberal Party association in Victoria. This chapter explores the great diversity that exists among party EDAs while considering the makeup, responsibilities, and activities of local party organizations within the context of Canadian party politics.1 Data for this chapter are drawn from a variety of sources including, though not limited to, two original surveys ofEDAs (2016 and 2019), formal party documents, an original dataset of 11omination contests, and reports filed with Elections Canada. WHO MAKES UP AN EDA?

EDA Executives

As Koop (20 I 0, 896) explains, EDA executives are "small groups of local party activists who provide leadership for constituency associations and conduct their month-to-month business:' Each party sets out how its EDA executive (or board of directors) will be constituted. Liberal Party bylaws, for example, state that a chair (president), vice-ch air, secretary, organization chair, policy chair, up to six directors at large, the current Liberal MP in the district (if there is one), the party's nominated candidate in the district, and one representative from each recognized commission (e.g., women, youth, etc., as discussed in Chapter 2) shall comprise an EDA. Additionally, the board includes a number of non-voting members, specifically, a treasurer and fundraising chair (see Liberal Party of Canada 2020b ). Parties often have more specific rules regarding EDA composition as well. The Conservatives, for instance, allow one in five members of the EDA board to be a resident of another district and set a maximum size of an EDA board at thirty members (Conservative Party 2021a). Likewise, the NDP Peterborough-Kawartha EDA has two youth members as executive members, with specific rules regarding one being male and one being female (New Democratic Party of Canada 2015). These NDP statutes also state that "at least one of the President or Vice-President shall be a woman:'

The Parties' Local Associations

While party statutes can reveal important details regarding the formal structure, they tell us little about how local parties operationalize the rules or the kinds of individuals who occupy these positions. While scholars are slowly turning their atte11tion to the importance of EDAs ( Cross 2016; Cross and Pruysers 2019a; Currie-Wood 2020; Koop 2011; Pruysers 2016; Tolley 2019), we still know very little about their makeup. Our survey data suggest that the average EDA had eleven positions on their executive in 2016 and nine in 2019. The Liberals report the largest executives (13.7 members in 2016 and 10.0 in 2019) whereas the Conservatives report the smallest executives (8.8 members in 2016 and 7.9 in 2019). The NDP reported an average of about ten executive members in both years (9.8 and 9.3). The average size by party, however, masks a lot of variation. Some associations report having upwards of thirty-seven members on their executives. In fact, in 2016, nearly one in five EDAs in our sample reported having twenty or more executive members. Others, by contrast, report having only two members, something that would seemingly contradict party bylaws regarding the expected makeup of local associations and highlights the apparent weakness of some local party organizations. While the number of positions on a local executive is interesting and provides a glimpse into the organizational capacity of the party at the grassroots level, who holds the positions is also worth exploring. In order to provide a better understanding ofwho holds positions of authority inside ofEDAs, we coded the gender of the president for each available EDA in the five parties represented in Parliament (as of the end of2020). 2 1he data (Table 3.1), which are derived from reports filed with Elections Canada, reveal that local executives can be important sites for descriptive representation, as a third of EDA presidents across the country are women. While far below parity, women's presence is higher here than it is at some other levels of the party. As expected, there are important party differences. Parties on the left (Bloc, Green, and NDP) have the highest percentage of wom en sitting atop of the EDA. In each of these parties the percentage is 37.0 percent or more. The Liberals and Conservatives, by contrast, fare much worse. For the Liberals the figure is 30.7 percent, and for the Conservatives it is just 23.1 percent. Even in the NDP, with clear guidelines requiring that a woman be in the position of president or vice-president,

63

64 The Political Party in Canada

TABLE 3.1 Percentage of women EDA presidents by party Party

Woman EDA president(%)

Number of associations

Bloc Quebecois

43.8

32

Conservative

23.1

338

Green

42.2

249

Liberal

30.7

336

New Democratic

37.0

235

All

32.5

1,190

Note: Coded from Elections Canada reports. Data include all associations registered with Elections Canada as of 2020. Since some individuals are listed as the president for multiple EDAs, we eliminated these duplicates in order to get a sense of how many unique women and men each party had in this position. This is especially common in the NOP and is why then is lower for the party. Bloc Quebecois numbers incl ude twenty deregistered associations.

we see that achieving parity at the EDA level is difficult. In the case of the NDP, with 37.0 percent local female party presidents, women are overrepresented in the vice-president position and underrepresented at the very top. Our survey data allow us to drill down a little deeper and consider the EDA executive more broadly (see Table 3.2). 3 In 2016, less than half of positions, an average of 4.8 positions out of 11.3, were held by women. In 20 19 the average ratio of women on local executives was essentially iden tical, 3.9 of 9.1 positions. In both 2016 and 2019, we find that NDP executives are the only ones to approach ge11der parity on average, no small feat in organizations that have long been male-dominated. Not surprisingly, and consistent with their share of female EDA presidents, the Conservatives lag behind the other parties on this metric by ten or more percentage points. Our data also include information about the executive beyond gender. Local party executives tend to have an average of slightly less than two visible minority EDA board members. While two members from visible minority groups may seem somewhat low, this generally reflects the percentage of the general population who identifies as a visible minority. Given that the size of Conservative EDA executives is slightly smaller than their NDP counterparts, this translates into a slightly higher percentage ofvisible minority representation. 4 Overall, local executives are relatively diverse

TABLE 3.2 Makeup of local EDA executives, 2016 and 2019

Conservative n

'°

...-i

0

N

Liberal n

New Democratic n

Conservative n 0\ ...-i

Liberal

0

N

n

New Democratic n

Mean number of executive positions

Mean number of women

Mean percentage of female executives (%)

8.8

3.0

32.6

39

39

39

13.7

5.6

41.6

90

86

86

9.8

4.7

48.2

79

79

78

7.9

2.8

30.2

64

63

62

10.0

4.3

42.8

60

62

60

9.3

4.4

46.4

74

75

74

Mean numt visible mino

66 The Political Party in Canada

and inclusive organizations (at least in terms of women and visible minorities), especially by Canadian party standards. EDA Presidents

An emerging literature has revealed that who EDA presidents are as individuals matters for the kinds of decisions that local party organizations ultimately make. To some extent, this speaks to the personalization literature and the growing importance of individual actors inside of party organizations. The gender and ethnicity of association presidents, for instance, has been shown to be related to the kinds of candidates recruited and selected: associations with a woman president are significantly more likely to nominate a woman to carry the party,s banner (Cheng and Tavits 2011; Cross 2016; Cross and Pruysers 2019a), and associations with a visible minority president are more likely to nominate visible minorities (Tolley 2019). Outside of gender and ethnicity, however, we know very little about EDA presidents. As individuals who sit atop of the local party hierarchy, it is worth exploring who EDA presidents are, including factors like their level of formal education, citizenship, age, and previous political experience. Are these local party leaders reflective of the broader society, the party membership, or neither? As revealed in Table 3.1, the majority of EDA presidents are men. In terms of their background, every respondent in our 2016 survey indicated that they were a Canadian citizen, and more than 80 percent of association presidents in each party were born in Canada. The same general pattern applies in 2019 as well, with more than eight in ten association presidents being born in Canada. Consistent with the findings relating to party members and candidates (see Chapters 2 and 4), however, many of those born outside of Canada do not come from the countries that account for much of Canadas recent immigration but rather from the United States and Western Europe (Germany, UK, Italy, etc.). Also, like party m embers more broadly, EDA presidents are older in age. The average age in the Liberal, Conservative, and New Democratic parties was 58.7, 56.7, and 60.4 years old respectively in 2016. These numbers are virtually identical in 2019 at 57.5, 55.9, and 51.7, albeit slightly younger for the NDP. The vast majority of EDA presidents are well educated, though there is considerable variation among the parties. In 2016, Liberal presidents

The Parties' Local Associations

had the highest levels of formal education: nearly nine in ten (86.9%) had completed university, college, or a postgraduate degree. This figure is fully ten percentage p oints lower in the NDP where 76.3 percent had a university or postgraduate degree. Conservative EDA presidents reported the least formal education: 62.1 percent had a university, college, or postgraduate degree (25 percentage points lower than in the Liberals). The 2019 figures are remarkably similar to 2016, though slightly higher overall: 85.7 percent of Liberals, 71.8 percent of New Democrats, and 68.8 percent of Conservatives report obtaining a university or postgraduate degree. Despite the clear variation that can be found between the parties, EDA presidents tend to b e highly educated individuals overall. Given their education, it is not surprisi11g that these individuals have above average incomes. Half of the Liberal (49.0%) and Conservative (50.0%) presidents report a household incom e of $110,000 or more. This figure, however, is considerably lower in the NDP (36.2%). To put this in perspective, Statistics Canada (2020) reports that the median household income (families and unattached individuals) in 2018 was $61,400. Overall, a fairly clear and predictable portrait of EDA presidents is emerging in terms of their demographic profile. These individuals tend to be highly educated, older, m ale, Canadian citizens, and born in Canada, and they tend to have above average household incomes - not particularly reflective of the broader Canadian population but not entirely dissimilar to party members. What about characteristics beyond their sociodem ographics? What kinds of political backgrounds do these individuals have? What motivated them to join their executive? Do they have higher political ambitions beyond serving as EDA president? Before we consider their partisan activities, however, it is worth con sidering their basic ideological outlook. In terms of their self-reported ideology, are EDA presidents more extreme in their ideology or do they place themselves in predictable positions along the ideological spectrum? What we find is that EDA presidents do not appear to differ all that much in their ideology compared to the ordinary party members discussed i11 the previous chapter. Moreover, the data for 2016 and 2019 are remarkably similar, revealing very little movement within the three parties. In both surveys, the Conservatives are furthest to the right, with mean scores of 7.4 and 6.9 in 2016 and 20 19 respectively. The Liberals, by contrast, are

67

68

The Political Party in Canada

closer to the middle, though somewhat left of centre, with mean scores of 4.2 and 4.3. Finally, New Democratic EDA presidents place themselves furthest to the left of the spectrum with mean scores of 2.2 and 2.3. Our data suggest that those serving as local party executives have considerable party backgrounds. On the one hand, EDA presidents have a long history with their party as ordinary members. When asked when they first joined, the average EDA president joined their party roughly twenty-one years ago. This, of course, is partly owing to their older age in general. Nonetheless, these individuals have been party members for son1e time (though we cannot be certain that they have been members continuously). Beyond ordinary membership, federalism provides many additional opportunities for party activism. In each party, more than onethird to half of all respondents report that they have previously served on a provincial party executive. Not surprisingly, this is highest in the New Democrats (55.l %), a party that has an integrated multilevel party structure that connects federal and provincial parties, and often their local associations, to one another in meaningful ways (Pruysers 2014, 2016). While a smaller percentage than the NDP, 37.0 percent of Liberals and 38.7 percent of Conservatives also report prior provincial executive experience. Consistent with this pattern, a majority of EDA presidents in each party also belong to a provincial political party (50.9% for the Liberals, 83.9% for the Conservatives, and 90.0% for the NDP). What's more, party membership is highly consistent across levels: 100, 96.4, and 86.0 percent for the New Democratic, Liberal and Conservative parties respectively. While some may find the Conservative figures surprising given the fact that federal and provincial conservative parties are organizationally truncated, evidence suggests a great deal of informal cooperation occurs in these parties, nonetheless (Pruysers 2016, 2018). As individuals holding leadership positions within the party organization, we should expect to find that association presidents have more connections to their local community than do ordinary party members. Recall that in Chapter 2 we revealed that about two-thirds of party members belong to a local community group (i.e., neighbourhood association, women's association, etc.). This figure is indeed higher for association presidents. In 2016, 81.5 percent of EDA presidents belonged to one such

The Parties' Local Associations

group, with 45.0 percent reporting that they belong to three or more groups. In 2019 the figure was nearly identical, with 79.5 percent belonging to at least one and 39.7 percent reporting membership in three or more. These individuals are active not only within their respective parties (having taken on a leadership role in the local organization) but also within their local communities. We do find some interesting party-level variation. In 2016, 29.7 percent of Liberal EDA presidents reported that they did not belong to any of the listed associations whereas this number is considerably lower for the NDP (9.9%) and Conservatives (10.3%). By 2019 the figures were unchanged for the Liberals (31.0%) and NDP (11.1 %) while considerably more Conservatives reported not belonging to a local group compared to 2016 (23.2%). Overall, however, Conservative and New Democratic EDA presidents appear to have consistently stronger civil society connections than their Liberal counterparts. While executive members often join to support a particular candidate (discussed below), few outside of the New Democrats have actually run for office themselves. Only 5.6 percent of Liberal and 1.6 percent of Conservative EDA presidents have sought a federal party nomination in the past. This is compared to one in five ND P presidents. The same pattern can be seen provincially, where 5.6 percent of Liberals and 8.1 percent of Conservatives have previously sought a provincial party nomination compared to one in four New Democrats. More than multilevel party connections, this may reflect the NDP's perennial third-place status and need to find candidates for even unwinnable ridings. In these cases, the party may simply rely on EDA presidents to step forward and fill the gap. With that said, these individuals do express a considerable degree of political ambition; a majority reported that prior to having served on their party executive, they thought in the future they may run for elected office. As with previous candidate experience, the New Democrat presidents report the highest levels of ambition (74.6%), with the Liberals (58.9%) and Conservatives (50.0%) exhibiting considerably lower levels of interest in an elected political career. While there is a considerable gender gap i11 terms of ambition among EDA presidents (64.2% of men compared to 54.9% of women), the differences are much smaller than what is found in other samples of ordinary citizens (Blais, Pruysers, and Chen 2019; Pruysers

69

70

The Political Party in Canada

and Blais 2017, 2019). In these studies, it is not uncommon to find a persistent gender gap oftwenty or more percentage points on questions related to whether participants have thought about running for office, how qualified they feel they are for elected office, and whether they think they would win an election if they did run. Interestingly, having served on an EDA executive appears to actually suppress political ambition among presidents rather than foster it. When asked if, now that they have served on their local executive, they thought they would run for elected office, there is a noticeable drop in every party. Interest for New Democrats drops from 74.6 to 45.1 percent (-29.5 percentage points), from 58.9 to 41 . l percent for the Liberals (-17.8 percentage points), and 50.0 to 31.1 percent for the Conservatives (- 18.9 percentage points). In none of the parties do a majority of EDA presidents say that they would run for office after having served on their executive. This is a finding that deserves future attention, as it is unclear if ( 1) local party presidents become disillusioned with party politics after serving on their local EDA; (2) their time on their local EDA has simply fulfilled their political ambitions; or (3) serving as president provides insight into what it takes to be a11 MP, and this knowledge influences their ambitions. While we cannot uncover the precise mechanism, it appears that membership on the executive decreases future ambition to seek elected office by a considerable degree. While there is a growing "incentives to membership" literature as demonstrated in the previous chapter (Clark and Wilson 1961; Cross and Young 2008; Heidar and Kosiara-Pedersen 2019; Scarrow 2015; Seyd and Whiteley 1992; You11g and Cross 2002a), less is known about why individuals seek leadership positions within the party, especially at the grassroots. To that end, we asked EDA presidents in our 2019 survey why they originally joined their local party association (see Table 3.3). Perhaps due to the more ideological nature of their parties, Conservative and New Democratic presidents are considerably less likely tha11 Liberals to say that meeting new people and extending their social life was very important to their initial decision to seek a position on the executive. Whereas fewer than one in ten Conservatives and New Democrats report that this was a very important factor, slightly more than one in five Liberals report the same. Acquiring new skills is also not particularly important for our respondents:

The Parties' Local Associations

TABLE 3.3 Motivation to join an EDA executive(%)

Conservative n

Liberal n

New Democratic n

Extend social life/ meet new people

Acquire new skills

Set up next step in political career

Influence party policy

Support a candidate in their nomination

8.2

11.5

13.3

49.2

51.6

61

61

60

61

62

21.8

13.0

9.4

33.3

54.5

55

54

53

54

55

8.7

14.5

13.0

38.6

31.4

69

69

69

70

70

Note: Percent reporting "very important" to their decision to join.

just 13.0 percent of Liberals, 11.5 percent of Conservatives, and 14.5 percent of New Democrats report that learning new skills was very important to their decision. This is somewhat surprising given the administrative and managerial demands that are placed on local EDAs - by both the central party and Elections Canada. Given these demands we might have expected individuals to see a position on the local party executive as an opportunity to acquire new skills. The decision to become part of the local executive does not appear to be driven by politically ambitious individuals seeking to further their career either. The percentage of respondents reporting that joining the executive was an important next step in their political career is 13 percent or lower in all three parties. Consistent with the broader ambition literature, men are considerably more likely to acknowledge the importance of their political ambitions (Cross and Young 2013; Pruysers and Blais 2018, 2019). While participating in social activities, acquiring new skills, and furthering their political careers do not appear to be driving factors in the decision to join a local executive, two factors do stand out as being relevant. First, one-third to nearly a half of all EDA presidents report that having a greater influence in party policy was very important to their decision to join the executive. This is consistent with evidence presented in Chapter 2 regarding the incentives to party membership more broadly. A desire to be involved in the policy process and to shape what the party

71

72

The Political Party in Canada

stands for is a motivating factor for many individuals. As we will see later in this chapter, however, policy discussion is something that many EDAs actually spend relatively little time on. What is even more important than policy, however, is the ability to support a candidate in the nomination process. More than half of Liberals and Conservatives report that supporting a candidate was very important to their initial decision to join. While lower, nearly a third of New Democrats also indicate that this was very important in their decision. The importance of supporting a candidate speaks to the privileged role that local EDAs play in personnel selection. It also highlights the increasingly personalized nature of intraparty politics, where candidates draw on their own networks for support rather than necessarily relying on long-term party members (Cross, Katz, and Pruysers 2018). This often takes the form of potential candidates maneuvering to ensure that trusted allies are in positions of leadership within the party (see Cross et al. 2016). A particularly telling (and recent) example of this comes from the local Conservative riding association in the riding of Milton. After some minor disagreements between the local association executive and the successfully nominated ca11didate, the candidate used the riding association's annual general meeting to remove all ten executive members who were up for re-election (including former deputy party leader, Lisa Raitt) , replacing them with his own slate of preferred executive members (Levitz and Boutilier 2021). INTER-ELECTION MAINTENANCE

Thus far we have considered who occupies positions within EDAs but not what EDAs do. Canada's single -member plurality electoral system divides the country into 338 distinct electoral districts. Given that electoral victory is acl1ieved by winning the most individual districts, and not necessarily the most votes, Canadian parties have always had an incentive to create active and strong local grassroots party organizations (Carty 1991; Carty, Cross, and Young 2000; Carty and Eagles 2005; Pruysers 20 l Sa; Sayers 1999). As Koop (2011, 9) suggests, ''an impressive national campaign is useless if the party cannot win in more ridings than the other parties, and doing so requires both competent candidates and local organizations to support them:, Similar sentiments regarding the importance of

The Parties' Local Associations

a healthy grassroots are echoed by senior campaign operatives as well. In a postmortem on the 2015 general election, for instance, Conservative campaign chair Jenni Byrne noted how the Conservatives benefited greatly from Liberal weakness on the ground. Byrne (2016) wrote that in the wake of successive Conservative victories was a "weakened and divided Liberal party that had allowed local riding associations to atrophy and die on the vine, with no money and no organization:' What parties do at the grassroots is therefore important (though often overlooked). EDAs vary, often considerably, in terms of their health and vibrancy. Some have few members while others have thousands; some rarely meet while others have routine meetings and social events; some engage in inter-election maintenance while others seem to become empty shells in between elections. In this section we highlight the wide variance that can be found in terms of healthy local orga11izations. We do so by considering a number of indicators such as membership recruitment, party meetings, communication, and fundraising. Before we consider these indicators of EDA health, however, it is worth examining the basic organizational capacity of Canada's EDAs. To do so, we consider the number of registered EDAs each party had in 2021, roughly eighteen months after the 2019 general election. Each of the three principal parties maintains 338 registered EDAs between elections - one in each district across the country. While this tells us little about the strength of these EDAs, it does provide us with a glimpse into the basic organizational capacity of the party on the ground. Maintaining a registered EDA requires parties to have a minimal level of continued organization in the riding insofar as EDAs require a governance structure (president, executive, etc.) and must complete basic administrative tasks (completing Elections Canada documents, etc.) to remain registered. While a continued EDA presence is the norm for the Liberals, Conservatives, and New Democrats, it is much more of a struggle for the Bloc and Greens. The Bloc, for instance, only had 32 EDAs at the beginning of 2021. This represents less than half of all possible EDAs the party could have in the province of Quebec. What's more, 20 of these EDAs had been deregistered with Elections Canada leaving just 12 actively registered EDAs for the party. The Greens fare better in this regard, but struggle to maintain a continued national presence nonetheless. The party had 255 EDAs in 2021; however, 28 had been

73

74 The Political Party in Canada

deregistered. This left the party with actively registered EDAs in two-thirds of the country's districts. One other basic metric worth considering is the number of EDAs that list a unique «CEO" (riding president) with Elections Canada. Having multiple EDAs that list the same person as CEO might suggest that while the party has some capacity in the district it is struggling to maintain its presence. For the most part, this is not a common occurrence. As of 2021, The Bloc, Conservatives, Liberals, and Greens respectively listed zero, one, two, and four EDAs with the same CEO as another district. In other words, the vast majority of EDAs in these parties have unique local party presidents, suggesting a minimal level of organizational capacity in the districts where an EDA is present. The NDP, by contrast, stands out. Ninety-nine NDP EDAs listed a local party CEO who was also listed in another district. While this trend spans multiple provinces, it appears to be particularly common in Quebec. Considering the party's history in the province, this is not entirely surprising. Despite having some recent success, the NDP has been historically quite weak in Quebec. Furthermore, without a provincial NDP wing to fall back on, the federal party has less administrative and organizational support than it has in other parts of the country. Membership Recruitment

An essential component of any EDA is inter-election maintenance (Carty and Eagles 2005; Cross 2016; Koop 2010; Pruysers 2016; Sayers 1999), that is, ensuring the party is healthy and vibrant in the ((off season" and is prepared for the next election campaign. This includes fundraising, community engagement, policy development, and a variety of other activities that engage the local base of supporters. One of the main tasks is to maintain and recruit new party members. Party members serve as an important source of local party income (membership dues and donations) as well as an important resource for campaign labour. As such, it is important for parties to ensure their local membership does not fade away in between elections (see Cross 2016). Internal party statutes increasingly recognize the importance of ongoing grassroots maintenance. The operational targets for the Liberal Party's nomination rules, for instance, outline a variety of benchmarks that should be met before the opening of a nomination contest. This includes ( among other administrative requirements) evidence of

The Parties' Local Associations

fundraising equal to "at least fifty percent ( 50%) of the anticipated election expense limit:' "7,500 voter and/or door-knocking attempts:' and "at least 300 Registered Liberals" (Liberal Party of Canada [Liberal Party] 2019a). 5 As indicated in Chapter 2, a majority of members join their party through a local association. Local associations, therefore, are essential for membership recruitment and retention. Three-quarters of EDAs in 2016 reported that they had made special efforts in the previous three years to attract new party members. There is very little variation among the parties in this regard, with 79.8 percent of Liberals, 75.6 percent of New Democrats, and 73.7 percent of Conservatives reporting efforts to recruit new members. Perhaps due to the changing nature of party m em bersl1ip (Scarrow 2015; see also Chapter 2 in this volume), many EDAs are finding it more difficult to recruit members than they had in the past. When asked if it is more difficult to recruit new members than it was five years ago, a slight majority (51.5%) of EDAs in our 2016 survey agreed. On this question there is some variation, with tl1e Liberals reporting the least trouble recruiting new members (59.6% disagreed it is getting more difficult) whereas the NDP and Conservatives reported considerably more trouble (41 .0% and 37.8% respectively disagreeing that it is becoming more difficult). This may speak to the considerable efforts that the Liberals put in to grassroots revitalization following their 2011 electoral defeat (Cross 2016) . In an internal policy document, «Roadmap to Renewaln (Liberal Party 2011, 2), the party clearly recognized that healthy and vibrant local EDAs were an organizational imperative moving forward: "The future health of the Party depends primarily and directly on the health of its Electoral District Associations (EDAs) and of its provincial and territorial party affiliates. As a national party, our first priorities must therefore be (i) rejuvenating and rebuilding LPC's inactive or underperforming EDAs with vastly increased grassroots engagement:' Given the party's recent emphasis on grassroots health and growth, it is not surprising that Liberal EDAs report the least difficulty in recruiting new members. Interestingly, efforts to recruit new members, as well as difficulty in recruiting tl1em, are not limited to associations without an incumbent member of Parliament. In fact, EDAs witl1 and without an incumbent report nearly identical efforts to attract new members (79.6% and 76.2% respectively have made a special effort in the last three years) and EDAs

75

76

The Political Party in Canada

FIGURE 3.1 Average local membership by party and year 1,400 ~ - - - - - - - - - - - - - - - - - - - - - - - ~ ,_ ] V)

1,200

E

E 1.000 ,_ Q)

-

-

Conservative Liberal NDP

800 ,,

.0

E

:l

600

C

a,

ll.0

400

cu ,_ Q)

~

, , ,

, , ,

, , ,

---- ---------- ------ ----

200

o_,____ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ____, 2013

2014

2015

2016

2017

2018

2019

with an incumbent are actually more likely to report difficulty recruiting new members compared to districts without an incumbent (58.5% compared to 49.0%) . Figure 3.1 combines data from our 2016 and 2019 surveys in order to track party membership at the grassroots level over a number of years (2013-19). The data, which contain seven data points, reveal some interesting patterns. First, on average, all parties witnessed grassroots growth over this period. That is, the average EDA did not have fewer members in 2019 than they did in 2013. Second, the rate and extent of growth varies, often dramatically, by party. We see the steady rise of the Liberal Party's grassroots strength after their historically poor 2011 election results and their commitment to renewed EDA health. Average Liberal EDA membership grew from approximately 220 members in 2013 to more than 1,000 members in 2019, an increase of more than 400 percent. Of course, some of this can also be attributed to the party's changing membership structure; moving from formal dues-paying members to a broader supporter category has likely fuelled this growth and 1nakes direct comparisons to the other parties less straightforward (see Chapter 2). Nonetheless, the party has clearly witnessed sustained growth at the local level over the past two election cycles. Conservative EDA membership growth, by contrast, is much less steady and is marked by a spike in new members in 2017 (leadership contest) and during the lead up to the 2019 general election (candidate nomination),

The Parties' Local Associations

with minimal movement in the years in between. In fact, the 2013 and 2018 numbers are remarkably similar. Thus, while the ch ange in Conservative membership is an increase of 149 percent during this period, it is the post-2018 period where much of this change can be found. Overall growth for the NDP associations is somewhat masked by the significant increases of the Liberals and Conservatives. 6 Nonetheless, NDP associations experienced growth during this period: the average NDP EDA grew by approximately 100 members (37.5%) between 2013 and 2019. What's interesting is that the NDP witnessed the least growth of all the parties, even less than the Greens, and has been on a slight downward trajectory since 2016. The NDP's struggles may be the result of the Liberals outflanking them on the left during the 2015 election, as well as yet another electoral "loss» in 2019. How active are members in local party affairs according to the EDAs? In 2016, EDAs reported that on average, about one-fifth of their members were active in local party affairs (i.e., attended meetings and routi11ely participated in activities). This varies by party (see Table 3.4), with the Liberals and Conservatives reporting the highest percentage of active members (27.5% and 26.1 % respectively) and tl1e New De1nocrats reporting considerably fewer active members (17.1 %). The 2019 data is not too dissimilar, though the Conservatives and Liberals have seen a decline in how active their membership is at the local level. While the Liberals are seeing consistent growth in their membership numbers, their activism appears to be on the decline.

TABLE 3.4 Percentage of local membership characterized as "active" by year and party(%)

Conservative

n Liberal

n New Democratic

n

2016

2019

Change

26.1

21.1

-5.0

36

63

99

27.5

19.8

- 7.7

90

60

150

17.1

17.3

+0.2

77

75

152

77

78

The Political Party in Canada

Party Meetings and Communication

Inter-election maintenance, however, is more than simply recruiting and maintaining party members. Figure 3.2 reveals that all the parties began meeting more frequently as the 2015 general election approached. The average number of yearly meetings within the Liberals, for instance, grew from approximately 10 in 2013 to 17 in tl1e election year of 2015. Looking at the m ean number of meetings does, however, m ask the variance within each party. Each party had multiple EDAs that reported holding zero meetings in 20 13. 7 As the election neared, fewer EDAs reported holding zero meetings: only 5 percent in 2014, and less than 2 percent in 2015. Unsurprisingly, membership and activity (meetings) are highly correlated in 2013, indicating that EDAs with more m embers meet more often than EDAs with fewer members (and vice versa). This correlation fades in 2014 and 2015, suggesting that even poorly populated EDAs need to meet for election planning purposes eventually. Data for 2016- 18 show that average meeting frequency quickly dropped after the election and remained relatively stable for a number of years. 8 Our 2019 survey was fielded prior to the 2019 general election, and as such we are unable to capture the full extent of EDA activity in that particular year. However, given that the average EDA had already met seven times when the survey was fielded (just halfway through 2019), it is fair to assume that we would have seen a considerable spike in 2019, as there was still a number of months of meetings left to be conducted. FIGURE 3.2 Meeting frequency by party and year 18 ~ - - - - - - - - - - - - - - - - - - - - - - - - ~ VI

ll0

C +'

·-

Cl) Cl)

-... E 0

Cl)

.a

E

:l C

16 14 12 10 8 6

Conservative Liberal

Cl)

ll0

4

-

1

2

- - NOP

...ro

o~------------------------~ 2013

2014

2015

2016

2017

2018

The Parties' Local Associations

Wl1at do EDAs do at these meetings? While members often join a party to engage in the policy process (see Chapter 2), many associations do not engage in frequent policy discussions. Our 2016 survey found that nearly one in four (23.3%) associations rarely or never engage in discussions of public policy issues. The Conservatives are the most likely to discuss policy, with only 12.9 percent reporting that they rarely or never discuss policy. The New Democrats, by contrast, are considerably less likely to discuss policy, with 30.1 percent reporting that they rarely or never do so. This, of course, is surprising given the typical depiction of the NDP as an ideologically driven mass party that provides ample opportunities for its members to engage and discuss policy at the grassroots level (Sayers 1999). While Liberal associations engage in policy discussion more frequently than the NDP, one in five also report rarely or never holding policy discussions. Overall, it is clear that policy discussion is not a top priority for many associations. It is important to note, however, that policy discussions do not just happen at the grassroots. EDAs have formal representation at national conventions as well. In the Conservative Party, for example, each EDA can send ten delegates, the president of the association, and the official candidate in the district to the national party convention. Conservative EDAs can also propose amendments to the party constitution (Conservative Party 2021a, Sec. 16.2.2). Likewise, NDP EDAs are entitled to send one delegate for their first fifty party members or fewer, one additional delegate for each additional fifty members, and one youth delegate to their party's national convention. The NDP also allows EDAs to call "special conventions" if they are requested by a majority of EDAs in a majority of provinces and territories (New Democratic Party of Canada 2018a, Article 5.l [b]). These formal and institutionalized mechanisms ensure that local party organizations are not completely shut out of policy discussions. This is important as little policy discussion takes place locally. It is not just policy discussions that rarely happen. Even more general communication between the EDA and party members occurs relatively infrequently. Nine in ten associations, for instance, report that they communicate with their local members by post less than once a month. Similarly, 93.8 percent of EDAs report communicating with their members

79

80

The Political Party in Canada

by phone less than once a month. This, of course, may not be particularly surprising given changes in communication technology in recent decades. We might expect, for example, phone and postal communication to be replaced with email and online forms of communication. Even with email, however, 60.9 percent of EDAs report communicating online less than once a month with their members. The Liberals tend to have the most communication online, with 46.0 percent of associations reporting at least monthly communication, whereas the Conservatives and New Democrats sit at 36.1 and 32.9 percent respectively. This is consistent with a similar, though more general, question asked in 2019. Here a majority (62.3%) of EDAs in the three parties report that they communicate with members less than once a month. As above, the Liberals tend to have the most communication: 45.6 percent report communicating with members less than once a month compared to 75.0 percent for the NDP and 62.9 percent for the Conservatives. The lack of communication is surprising given that the vast majority (87.7%) of associations report that having a vibrant local association is very important. The lack of communication with members may be, at least partly, the result of the central party playing an active role in sending emails to the membership, releasing press releases, and l1aving a somewhat robust online presence (Marland 2016; Marland, Giasson, and Lawlor 2018). Constituency associations, while the «closest» level of the party to ordinary citizens and voters, are not particularly public facing. A 2013 report conducted by the Samara Centre for Democracy, for instance, found that more than one in five associations did not have a website. 9 As expected, there is wide variation between the parties. None of the Bloc Quebecois associations, for example, had a website, and only one-third of Conservative associations had one. Fully 100 percent of NDP associations and 95 percent of Liberal associations, by contrast, had websites. Of the associations that had a public-facing website, very few provided important information for those interested in the local party organization. Less than half (48%) provided the president's name, 17 percent offered the names of the broader executive, 21 percent provided information on candidates and past candidates, and less than one in five ( 19%) had information on the electoral district. Perhaps most concerning, Samara's report finds that just 4 percent of association websites contained information on meetings

The Parties' Local Associations

(i.e., dates, locations, etc.) and less than a quarter listed upcoming events. Where the parties fared much better is in providing details regarding the national party (74% had information on the party leader and 78% provided a link to the national party platform) as well as information on how to sign up for a newsletter (75%), how to donate (96%), and how to become a volunteer (77%). We collected new data as of 2020 and updated this to include activity on social media in addition to whether or not the EDA has a public-facing website. Table 3.5 reveals the percentage ofEDAs we found to have a webpage attached to the national party,s website, a website independent of the natio11al party, a Facebook page, a Twitter account, and/or another social media profile (YouTube or Instagram) .10 Both the Conservative and Liberal parties include an EDA page for each party association as part of their national party website. The NDP, by contrast, does not. Independent EDA websites are far less common than pages attached to the national website. Our search revealed that less than half of Conservative EDAs have their own independent website compared to just one in seven ND P associations. Liberal Party EDAs rely completely on their national pages and appear to have no easily identifiable i11dependent websites. Social media accounts are somewhat more common, though this varies considerably by platform. According to our search, a majority of EDAs in each party has a public Facebook account. From the accounts we were able to identify, this ranges from 61.8% of Conservative EDAs to 75.7% of Liberal EDAs. While a Twitter presence is far less common, a sizeable minority of associations have an account nonetheless. Consistent with their relatively low presence on Facebook, the Conservatives have the fewest Twitter accounts (24.3%) and the Liberals have the most (41.1 %). Other platforms such as YouTube and Instagram are even less common. As Facebook is the largest social media platform - with more than 2 billion users worldwide - it is not surprising to find the parties more active on the platform relative to others. It is also the platform that serves the oldest demographics (Auxier and Anderson 2021), something that would be consistent with membership patterns inside of parties (see Chapter 2). While EDAs do not always have a highly vibrant web presence, it is useful to consider the kinds of information that members and interested individuals can find online. Considering webpages on the national party

81

82

The Political Party in Canada

TABLE 3.5 Online presence of EDAs (%) Conservative

Liberal

New Democratic

100

100

0.0

Independent EDA website

47.0

0.0

15.4

Facebook

61.8

75 .7

66.3

Twitter

24.3

41.1

28.7

9.2

11.8

15.4

Webpage attached to national party website

Other social media account (YouTube, lnstagram) Note: Data cu rrent as of July 2020.

website, independent websites, and social media accounts, Table 3.6 provides information on the percentage of EDAs that provide public-facing information regarding the EDA president, executive, local candidate, district, party platform, upcoming events, or a poll regarding party policy. 11 The data reveal that very little information about local party associations is available online. Information about the members of local executives, for instance, is scant. Less than 1 percent of Liberal EDAs provide the details of the broader local party executive. Similarly, ver y few EDAs provide interested individuals with information about their district (ranging from about 1 to 5%). Far 1nore likely to be available is information about the candidate or upcoming events. Associations that do have a web presence almost universally provide contact information for the EDA, as well as information about how to join, donate, and volunteer. In this sense, EDAs l1ighlight their role as electoral machines, and their web presence is, primarily, a means of gathering funds and volunteers. Thus far we have considered the extent to which local associations attempt to recruit new members, hold meetings, discuss policy, communicate with existing party members, and provide a public-facing online presence for their constituents. This, however, only begins to scratch the surface of the various activities that local party organizations have to engage in on a regular basis. To get a better sense of what EDAs spend their time on, we asked EDA presidents in our 2019 survey what percentage of their association's activity was spent on a variety of activities in the previous twelve months. Table 3. 7 reports the mean percentage of time spent on

TABLE 3.6 Percentage of information available online (all EDAs) (%) Party

EDA president

EDA executive

Local candidate

District

Party platform

Conservative

100

5.6

34.3

4.7

3.3

Liberal

99.1

0.6

52.1

0.9

99.4

9 .2

7 .4

15.1

3.3

26.6

New Democratic Note: Data current as of July 2020 .

TABLE 3.7 Percentage of association time spent on various activities(%)

Conservative n

Liberal n

New Democratic

n

Routine administration

Election planning/ campaigning

Community outreach/ Recruitment

Policy discussion

26.8

30.4

13.5

11.7

60

61

57

54

26.5

29.1

19.7

11.9

58

56

55

54

30.4

35.7

15.6

13.8

71

69

60

61

84 The Political Party in Canada

administration, election planning, community outreach, policy discussion, social events, and other activities (totalling to roughly 100%). Given that it was an election year, it is not surprising to see that considerable time was spent on election planning. When election planning is combined with routine administration (i.e., :financial reporting for Elections Canada), more than half of an average EDXs time is spent on these two activities alone. Once outreach and recruitment are accounted for, this leaves only a quarter of the associations time to be devoted to policy discussion and social eve11ts for the membership combined. The amount of time spent on the various activities listed in Table 3.7 is reflected in the perceived importance of them. When asked to rate the importance of various actives (0 being not at all important and 10 being very important), EDAs consistently rate election planning (mean = 8.7) and administration for Elections Canada (mean = 8.0) as being much more important than activities such as developing party policy (mean= 6.2) and hosting social events for the local membership (mean = 6.7). 12 The portrait that emerges from the data suggests that EDAs are not the social and participatory party organizations they are often portrayed to be. They do not actively engage in policy discussions, they host few social events for their members, and they are in infrequent communication with their local base of support. They do, however, spe11d much of their time on administration required by the national party and Elections Canada and on planning for upcoming election campaigns. To some extent, EDAs appear more like the local campaign organization for national parties tl1an they do their own grassroots organizations. Federal associations also engage in some multilevel party events that are consistent with the cross-level membership and activism of ordinary members. Not surprisingly given the structure and history ofNDP associations (Dyck 1991; Pruysers 2014), a majority (57.6%) report that they hold joint social events at least occasionally with a provincial party association. Such multilevel activities are not completely uncommon for the other parties, as 39.6 percent of Liberal EDAs and 36.8 percent of Conservative EDAs also report occasionally holding joint social events with their provincial counterparts. Parties also meet in a multilevel fashion to discuss policy. However, given how infrequently parties already engage in policy discussions in general, these forms of multilevel policy discussions

84 The Political Party in Canada

administration, election planning, community outreach, policy discussion, social events, and other activities (totalling to roughly 100%). Given that it was an election year, it is not surprising to see that considerable time was spent on election planning. When election planning is combined with routine administration (i.e., :financial reporting for Elections Canada), more than half of an average EDXs time is spent on these two activities alone. Once outreach and recruitment are accounted for, this leaves only a quarter of the associations time to be devoted to policy discussion and social eve11ts for the membership combined. The amount of time spent on the various activities listed in Table 3.7 is reflected in the perceived importance of them. When asked to rate the importance of various actives (0 being not at all important and 10 being very important), EDAs consistently rate election planning (mean = 8.7) and administration for Elections Canada (mean = 8.0) as being much more important than activities such as developing party policy (mean= 6.2) and hosting social events for the local membership (mean = 6.7). 12 The portrait that emerges from the data suggests that EDAs are not the social and participatory party organizations they are often portrayed to be. They do not actively engage in policy discussions, they host few social events for their members, and they are in infrequent communication with their local base of support. They do, however, spe11d much of their time on administration required by the national party and Elections Canada and on planning for upcoming election campaigns. To some extent, EDAs appear more like the local campaign organization for national parties tl1an they do their own grassroots organizations. Federal associations also engage in some multilevel party events that are consistent with the cross-level membership and activism of ordinary members. Not surprisingly given the structure and history ofNDP associations (Dyck 1991; Pruysers 2014), a majority (57.6%) report that they hold joint social events at least occasionally with a provincial party association. Such multilevel activities are not completely uncommon for the other parties, as 39.6 percent of Liberal EDAs and 36.8 percent of Conservative EDAs also report occasionally holding joint social events with their provincial counterparts. Parties also meet in a multilevel fashion to discuss policy. However, given how infrequently parties already engage in policy discussions in general, these forms of multilevel policy discussions

The Parties' Local Associations

are more rare than joint social events. Four in ten NDP associations meet jointly at least occasionally to discuss policy compared to 23 .1 percent of Liberals and 21.1 percent of Conservatives. 13 While such multilevel party connections are often hidde11 at the central party level, considerable amounts of informal multilevel cooperation and communication clearly occurs at the grassroots. Fund raising

An important function of local EDAs is to raise money to support the party in the next election. As documented in Chapter 5, election campaigns are expensive endeavours. The need for money, however, extends beyond fighting election campaigns. Without adequate funds, local associations may find it difficult to hold social events, engage the community on a regular basis, and stay active in the inter-election period i11 general. Raising money is particularly important in light of changes to legislation that have limited the financing of parties and candidates (including the amount and who is entitled to contribute) as well as more recent changes that phased out the state-sponsored per-vote subsidy that parties received. As Cross, Currie-Wood, and Pruysers (2020) describe them, EDAs are tl1e party apparatus that financially sustain local partisan life between elections. Even though the local party association is a separate legal entity from the candidate, it constitutes the inter-election period apparatus of the party, and it is meant to support the election of party nominees. Consistent with its efforts to recruit members and build its grassroots organization, Liberal EDAs attracted significantly more donors than their NDP and Conservative counterparts during 2015. As shown in Table 5.4, examining the entire universe of EDA financial reports submitted to Elections Canada 14 reveals that in 2015, the mean number of contributors to a Liberal EDA was 235. The Conservatives, by contrast, had an average of 113 donors per EDA, 122 fewer donors on average than the Liberals had. The NDP fared only slightly better with an average of 133 donors per EDA, 20 more than the Conservatives had but 102 fewer than the Liberals had on average. Despite having more contributors on average, the mean dollar amount raised by NDP EDAs was actually lower than the Conservatives raised (which in turn was considerably lower than the Liberals raised). The mean dollar amount raised by Liberal EDAs in 2015 was

85

86

The Political Party in Canada

$55,151, compared to $34,163 by the Conservatives and just $18,791 by the NDP. Perhaps because of their grassroots strength, Liberal EDAs received considerably less in financial transfers from other party branches than both the Conservative and NDP EDAs. The average Liberal association received nearly $8,000 in transfers compared to about $20,000 for the Conservatives and $13,000 for the New Democrats. 15 While there is considerable variation between the parties, there is also variation within parties. Each party, for example, has associations that reported zero donors (despite being an election year), while a number of other associations reported upwards of 1,400 contributors. In terms of dollars raised, some associations had $0 (not surprising given that some had zero contributors) while other associations raised more than $200,000. As you would expect, EDAs located in districts where their candidates were elected were able to attract significantly more donors. EDAs that had their candidate elected had an average of 259 donors whereas EDAs with a losing candidate attracted 113 donors (56% fewer) . This translated into a substantial $36,000 difference on average between EDAs with and without a successful candidate. 16 Winning EDAs received more financial transfers than losing EDAs; however, the gap is not as substantial (approximately $11,000 for losers compared to $17,000 for winners) as for fundraising more generally. 17 In this regard, other party branches appear to ensure that even less competitive associations had the financial resources needed to support their local candidates. At the same time, there may be other, less egalitarian, motives behind financial transfers from the centre. Cross, Currie-Wood, and Pruysers (2020) reveal that local campaigns receiving a greater percentage of their funds from the central party tended to place significantly more emphasis on their party's brand while those receiving a smaller percentage of their income from the central party placed significantly more emphasis on themselves. In this sense, central party transfers may be a means of combatting personalization and ensuring that local candidates stay on script. Money, therefore, could be viewed as a means of exerting authority over local campaigns. How does this compare to 2018, a non-election year? The first observation is that, unsurprisingly, EDAs report considerably fewer donors in non-election years. The Liberal EDAs reported an average of 129 donors;

The Parties' Local Associations

Conservative EDAs, 77 donors; and NDP EDAs, an abysmal 25 donors. The number of contributors ranged from Oto 1,266. Although the Liberals maintained their grassroots advantage, especially over the ND P, the Conservatives narrowed the gap to some degree. This is reflected in not only the number of contributors but also in the dollar amounts raised. The average Liberal EDA raised $25,485 compared to $14,376 for the Conservatives and $3,591 for the NDP. This again speaks to the Liberals' commitment to grassroots revitalization. The ND P's weakness in this regard is also quite apparent. Consistent with 2015, Liberal EDAs received relatively minor transfers from other party branches (mean transfers of less than $1,000). Also consistent with 2015 is the Conservatives' more handson approach in transferring funds across the country, with the average Conservative EDA receiving about $8,000. Unlike 2015, where NDP EDAs received considerable transfers, in 2018 they received an average of just a few hundred dollars. Electoral strength seems to matter even in non-election years. Associations with an incumbent were considerably better fundraisers in 2018 than associations whose candidate lost the 2015 election, though part of this is the result of the Liberals (who had the best fundraising record at the time) having the most incumbents. Associations that won the 2015 general election had an average of 165 donors in 2018 compared to just 35 donors for those associations that lost. In terms of amount raised, EDAs with an incumbent raised $33,316 on average compared to $5,473 (a difference of $27,843). At least one explanation for the better fundraising performance of incumbents are the incentives built in to candidate nomination procedures. Parties routinely include benchmarks for incumbents to achieve in order to be automatically re-nominated without challenge. In the Liberal Party, for instance, one of these benchmarks requires incun1bents to secure "an increase of at least ten (10) additional Victory Fund donors" (Liberal Party 2019a). For illustratio11 purposes, Figure 3.3 compares the average number of contributors per EDA in 2015 and 2018. It is easy to see how parties are much more effective at mobilizing contributions in election years. With that said, tl1e Liberals were able to attract more contributors per EDA i11 2018 (non-election year) than the Conservatives attracted in 2015 (election year). This is consistent with evidence that the Conservatives ignored the

87

88 The Political Party in Canada

FIGURE 3.3 Average number of contributors per EDA by party and year 250

....Ill 0

C

0 -0

D D

200 -

It] NOP

'+-

0

....QI

Conservative Liberal

150 -

.0

E

::J C

100 -

QI

tl,O

....QI (0

50 -

>