UNIVERSITY OF CALGARY. Playing in the Margins: Collaboration Between Local Party Organizations in the Canadian Party. System. Robin M.

Transcription

1 UNIVERSITY OF CALGARY Playing in the Margins: Collaboration Between Local Party Organizations in the Canadian Party System by Robin M. Currie-Wood A THESIS SUBMITTED TO THE FACULTY OF GRADUATE STUDIES IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTERS OF ART GRADUATE PROGRAM IN POLITICAL SCIENCE CALGARY, ALBERTA AUGUST, 2016 Robin M. Currie-Wood 2016

2 ABSTRACT Collaboration between local party organizations is a peculiar organizational trait for Canadian political parties. It is peculiar because local party organizations were once conceived to be solely concerned with electoral activities in their own district, while central parties helped local campaigns in strategically important ridings. Collaboration also goes against the prevailing theory that political parties will centralize operations under new party finance law. Despite this some local party organizations in the Conservative Party of Canada began helping campaigns in other ridings. Financial data are assessed for the Conservative, Liberal, and New Democratic Party between 2007 and 2011 in order to determine why parties at the constituency level mobilize their own resources in other constituencies. Playing in the Margins reveals that collaboration occurs in order to better utilize campaign effort made in competitive and uncompetitive ridings, and may even be a new component of intra-party contests. ii

3 ACKNOWLEDGEMENTS No written work is fully the product of the author. I owe a great deal of gratitude to my supervisor, Dr. Anthony M. Sayers. Anthony has been interested in this project since I began at the University of Calgary and he championed my education during grad school. He forced me to bolster my analytical skills by addressing my shortcomings with writing. Anthony was also very relatable outside of the classroom, which was most welcome and kept me motivated over the three-year project. I also owe special thanks to Dr. Melanee Thomas and Dr. David K. Stewart. Melanee and David introduced me to quantitative methods in social science, political processes, and remained involved during the thesis stage of grad school. Special thanks to Dr. Marc-Andre Bodet for making available his data-set that measures riding-level competition. I now know how difficult constructing a data-base can be and appreciate Dr. Bodet s willingness to share his work. There are many other anonymous academics, political activists and party volunteers who have helped out enormously along the way. Playing in the Margins has been strengthened by the contributions of each person. Any problems in this thesis remain my own. I also wish to thank the Political Science Department administrators, Judi, Bonnie, and Ella. They provided a great deal of help navigating the university bureaucracy. Finally, I wish to thank the taxpayers of Alberta for generously funding a portion of my research through the Queen Elizabeth II Scholarship. This funding was awarded by the Department of Political Science at the University of Calgary. iii

4 DEDICATION For my Mom and Dad, the social democrats; For Grammy and Grandaddy Wood, the Tories; and, For Grandma and Grandpa Currie, the agrarian populists. iv

5 Table of Contents ABSTRACT...ii ACKNOWLEDGEMENTS... iii DEDICATION... iv Table of Contents... v Chapter 1 Introduction... 1 Political Parties as Vote Gathering Institutions... 5 Renegotiating the Bargain? Reforming the Party Finance Regime... 6 Playing in the Margins: Local Party Organizations and Inter-association Funding... 9 The Institutionalist Approach, the Canadian Party System, and Case Selection Thesis Layout Structure Chapter 2 Harnessing Campaign Effort with Local Party Organizations: Continuity and Change in the Canadian Party System The Logic of Electoral Contests in an Single-Member District Electoral Systems The Canadian Party System: Table 2.1: Percentage of Popular Vote Won and Seat Translation in each Province by Political Party Stratarchically Organized Parties: Central Party Offices and Local Party Organizations Local Party Organizations and Direct and Indirect Campaign Effort Reforming Party Finance Law, Altering Electoral Competition Conclusion Chapter 3 Collaboration between Local Party Organizations: The Conservative Advantage in Election Campaigns Methodology: Follow the Money Local Party Fundraising Table 3.1: Election Year EDA Fundraising Table 3.2 : Electoral District Association Fundraising Table 3.3: Local Party Fundraising Averages by Province Local Party Savings Table 3.4: Electoral District Association Savings v

6 Table 3.5: Electoral District Association Savings Entering an Election Year Table 3.6: Local Party Savings Averages by Province Intra-Party Funding and the Importance of Inter-Association Fund-Transfers in the Conservative Party of Canada Table 3.7: Central vs. Local Intra-Party Fund-transfer Ratio Table 3.8: Inter-association Fund-Transfers Table 3.9: Conservative Party of Canada Riding Association Fund-Transfers Constituency Campaigning in 2008 and Table 3.12: Constituency Campaign Spending Totals by Party and Election Year Table 3.13: The Ratio of Campaign Spending against District Expense Limit by Party and Province Table 3.14: Ratio of Average Constituency Campaign Spending Against District Expense Limit Compared to District Competitiveness Discussion Chapter 4 Local Party Collaboration: Candidate Quality, Electoral Predictability, and Interassociation Funding Modelling Collaboration between Local Party Organizations Figure 4.1: Modelling Inter-association Fund-Transfers Case Selection Removing Support off the Top: Outflow Fund-transfers from Stronghold Ridings Figure 4.2 Campaign Stats: Calgary Southwest Figure 4.3 Campaign Stats: Calgary Centre-North Figure 4.4 Campaign Stats: Calgary Southeast Figure 4.5 Intra-Party Funding: Calgary Southwest Figure 4.6 Intra-Party Funding: Calgary Centre-North Figure 4.7 Intra-Party Funding: Calgary Southeast Table 4.1: Outflow Fund-transfers Received and Sent by Conservative Riding Associations 114 Receiving Party Support from Other Ridings: Inflow Fund-transfers to Battleground Ridings Figure 4.8 Campaign Stats: Desnethe-Missinippi-Churchill River Figure 4.9 Campaign Stats: Fleewood-Port Kells vi

7 Figure 4.10 Campaign Stats: Welland Figure 4.11 Intra-Party Funding: Desnethe-Missinippi-Churchill River Figure 4.12Intra-Party Funding: Fleewood-Port Kells Figure 4.12Intra-Party Funding: Fleewood-Port Kells Table 4.2: Inter-Association Fund-transfers Received and Sent by Top Three 2008 Inflow Conservative Riding Associations Inter-association Funding to and from Other Party Strongholds Figure 4.14 Campaign Stats: Random-Burin St. George's Figure 4.15 Campaign Stats: Vancouver Kingsway Figure 4.16 Campaign Stats: Vancouver Quadra Figure 4.17 Intra-Party Funding: Random-Burin St. George's Figure 4.18 Intra-Party Funding: Vancouver Kingsway Figure 4.19 Intra-Party Funding: Vancouver Quadra Table 4.3: Inter-association Fund-transfers Received and Sent by Conservative Riding Associations in Other Party Strongholds Discussion Chapter 5 Conclusion Playing in the Margins Implications for Political Parties as Stratarchical Institutions Bibliography Appendix A: Conservative Party Registered Association Fund-Transfer Reports Appendix B: Local Party Organization Case Profiles vii

8 Chapter 1 Introduction 1

9 Local party organizations are vital to winning elections in Canada. As the component at the riding-level, local party organizations are the political party apparatus directly responsible for engaging citizens and mobilizing voters on election day. Political parties cannot rely solely on their leaders or even multi-million dollar advertising buys to win elections, and they cannot depend upon state subsidies alone to finance electoral operations. Campaign effort must also be made on the ground. Campaign resources in the form of labour, money, and votes originate with individuals in civil society who must be drawn together to direct, finance, and deliver campaigns. Generating campaign resources and making campaign effort remains the responsibility of political parties at the riding-level, the local party organization. There is a divide among local party organizations between those that have an overage (or surplus) of campaign resources and those that do not. Overages are a surplus above the amount necessary for winning the local contest. Overages can occur with votes or campaign resources. Those in districts where the party is relatively assured of success may have access to plentiful resources, whether resources are votes, volunteers, money or a combination of the three. They may be unable to make full use of their endowments because party support is already very high. Alternatively, local party organizations located in highly-competitive, closely contested ridings may not have the resources needed to maximize their campaign effort and chance of success. The Conservative Party of Canada has responded to this divide by shifting resources from local party organizations that have an overage of resources to those that have a shortage. The collaboration between local parties and campaigns is achieved through inter- 2

10 association funding. 1 Inter-association funding is a new organizational feature made possible due to regulatory changes in the party finance regime that have exposed overages of party support at the riding-level. This thesis analyzes the distribution of electoral support and campaign resources throughout local party organizations in Canada. I argue the Conservative Party of Canada responded to the modern party finance regime by shifting resources between riding associations in order to improve their chances of electoral success at the local-level, and through this heightened focus on grassroots campaigning, increased their electoral fortunes at the national level. A combination of national party directives and local party initiatives has evolved into a broad network of collaborating grassroots campaign teams in the Conservative Party. When combined with the intensified campaign effort required in highly-competitive ridings, some speculate that local party organizations enhance their electoral competitiveness by depending on campaign support from central party offices (for example, Carty and Young 2012; Coletto and Eagles 2011). Until now, data on this has been limited because of the absence of evidence of collaboration between local party organizations. Evidence collected here demonstrates the importance of inter-association transfers for overcoming the disequilibrium between have and have not local party organizations. The idea that local campaign teams collaborate came from my time as an active volunteer for a provincial campaign. A conversation occurred about whether money should be 1 Inter-association funding can also be conceived as inter-district funding. The latter terminology implicitly conceives electoral district associations as different than local campaign organizations. Although both organizations have the same purpose to elect candidates of their party banner the people in these organizations can differ. 3

11 given to neighbouring ridings that were currently held by opposing parties. One thought was the money should be spent locally. Even though there was more than enough money, the concern was money given away would be squandered and no new seats would be won. The other side advocated these seats could be won, if only past teams had the money to be competitive. Money was ultimately given. The party claimed both new seats on election night. Collaboration between local campaigns was especially fascinating to me because I had not read about it as a specific trait of parties or party systems in other studies (for example, Carty and Eagles 2005; Sayers 1999). I set out to determine whether local teams collaborated at the federal level. Elections Canada financial reports document a range of financial behaviour; however, the design of the web system prohibited scraping data with a digital program. I found myself copying and pasting thousands of pages of financial reports submitted by riding associations and candidates. The result of this task, taking me over 300 hundreds hours over ten weeks, was a robust database containing 1,848 cases of local party organizations and their finances in the Canadian party system. The aggregate data confirmed that some local party organizations in only one federal party collaborated with each other. Playing in the Margins fills the gap about the collaboration that may occur between local party organizations during campaigns. The combination of singledistrict voting and the enduring variability of organizational competitiveness across ridings activated collaboration when a new party finance regime was installed. The Conservative Party of Canada enhanced its competitiveness by moving money from well-resourced associations to under-funded teams in very competitive ridings. 4

12 Political Parties as Vote Gathering Institutions Political parties provide the organizational framework to accumulate resources and mobilize campaign effort. In single-member plurality electoral systems, resources are accumulated and consumed in order to attract a sufficient number of votes to win a district, sending a representative to parliament. Political parties hope to win enough districts to form government. The process for accumulating resources and using them for campaign effort results over time in predictable behaviour. Formal organizations are born from institutional, social, and competitive features of the party system in each district (Ware 1996). The level of institutionalization in the party organization can be measured by responsiveness (internal reorganization in the extra-parliamentary body) to change in the party system (Panebianco 1988). One enduring feature of parties in Canada is the need to develop semi-autonomous organizations at multiple levels. Political parties must reduce support variation between ridings if they are to navigate the logic of the electoral system. The combination of the regionalized, ethnolinguistically diverse electorate and single-member plurality voting system establishes a sequence which requires parties to concentrate support at local-levels to win national contests (Carty 2002). Unlike parties competing in proportional based systems, Canadian parties can generate support overages at the local-level that create seat shortages at the national level. This has the effect to limit the number of seats votes a party has in parliament. A political party may have a large percentage of the national vote that disproportionately translates into a small percentage of parliamentary votes. Similarly, a political party may generate modest national support that translates into zero parliamentary votes because support is thinly spread 5

13 across ridings. One way political parties navigate the logic of the electoral system and mitigate regional diversity is by organizing and campaigning in each riding (for example, Sayers 1999). Similar to a business franchise, Canadian parties deliver operations in the most suitable way through the local party organization and coordinate operations through a central party office (Carty 2002). Features at the riding-level interact with party ideology to form the narrative of the constituency campaign and from them the regional contests across the nation (Sayers 1999). The relationship between local party organizations and the central party office evolves over time in response to changes in the party system. Identifying the extent to which party organizations balance local party responsiveness or sensitivity to local concerns with central party control or discipline over campaign machinery reveals a great deal about the structure of parties and how they enhance electoral competitiveness in Canadian elections (Carty 1992). Renegotiating the Bargain? Reforming the Party Finance Regime Present scholarship suggests political parties are centralizing their organizational structure in the twenty-first century. This conclusion is drawn from institutional, social, and competitive changes to the Canadian party system. First, party finance reforms (Bill C-24, 2003; Accountability Act 2006) increased spending power at the centre by creating state funded quarterly allowances (Coletto et. al 2011). Central party offices now have predictable incomes (Flanagan and Jansen 2009), while local party organizations have unprecedented restrictions on allowable income sources. Second, central parties are capable of gathering individual voter information and communicating directly to voters in a way that was previously done by local 6

14 party organizations (see Carty 1992: 565; Siegfried 1966: ). The adoption of computer database technology and widespread use of the Internet enable central parties to circumvent local party organizations as communication mechanisms (Flanagan 2003; Delacourt 2013). Third, parties interact in a highly-competitive system of election contests. The creation of the Conservative Party of Canada reduced the number of effective parties in the system in This led to three consecutive minority parliaments between 2004 and Political instability forced parties to be in a constant state of election readiness (Flanagan 2014, Chapter 6). Changes in the party system correspond to concentrated campaign effort made in highlycompetitive ridings. These election contests are decided by very few votes. Some predict that party finance reforms empowered central party offices by endowing them with money that can be used to command and control effort made by local party organizations in strategically important ridings (Carty and Young 2012; Coletto et. al 2011; Jansen and Lambert 2013). However, predictions that political parties will exert more control over riding-level campaign operations under new party finance law are drawn from aggregate data. The data do not account for regional variations in party support or campaign effort made at the riding-level. This has led to an imperfect understanding of local party organizations and constituency campaigns under the modern party finance regime. Moreover, presumptions regarding party centralization rely heavily on the financial inequality produced by per-vote subsidies paid to central party offices (Carty and Young 2012; Coletto et. al 2011; Coletto and Eagles 2011; Jansen and Lambert 2013). Subsequent amendments to the party finance regime in 2013 eliminated the quarterly allowance (Fair Elections Act 2013). If we are to understand Canadian parties and elections in the twenty-first century, we must also understand how local party 7

15 organizations accumulate campaign resources and consume campaign effort under the modern party finance regime created in 2004 and Consider this: the Canadian Broadcast Corporation (2013) reported that (then) cabinet minister Jason Kenney s riding association, the Calgary Southeast Conservative Electoral District Association (EDA), earned half its income from donors outside Alberta. Elections Canada financial reports (2013) also show his riding association had the highest income of any other in Despite his large campaign war chest, Kenney spent only half of the district expense limit campaigning for re-election in the He handily won re-election with just over seventy-five percent of the district vote. How was the rest of the money spent? The Calgary Southeast Conservative EDA transferred roughly $100, to sixteen riding associations in other districts. Kenney sent money to six Conservative campaigns in Ontario, six in British Columbia, three in Quebec, and one in Nova Scotia (Appendix A). The resources sent to other ridings had the effect of building campaign effort in other ridings. In this sense, Kenney s local party organization transcended the riding boundaries of Calgary Southeast. This thesis reconsiders how political parties react to recent party finance reforms by analyzing the financial behaviour of local party organizations between 2007 and It begins with two observations. The first is that most analysis suggests that the party finance regime should favour more central party focus over local campaigns. The second is that there is a pattern of substantial transfers between local party organizations, notably around election time. In general, I adopt an institutionalist approach in order to explain the impact of the party 8

16 finance regime on the role of local party organizations in the current party system. This begins by treating political parties as vote gathering organizations to answer the following questions: 1) Why do local party organizations move resources into other constituencies? This is counter-intuitive because local party organizations are assumed to be riding specific instruments for winning constituency campaigns (for example, Carty 2002). 2) Does each effective federal party exhibit collaboration between local party organizations, or do parties incorporate distinct features that derive from their own institutionalized structures? 3) To what extent is collaboration between local party organizations the product of district-level features? Conclusions have implications for the ongoing centrality of local politics in federal elections. Canadian parties remain indebted to the electoral innovations made by riding associations and constituency campaigns. Playing in the Margins: Local Party Organizations and Inter-association Funding Local party organizations retain their importance within political parties under the modern party finance regime because they have a unique ability to redirect party support overages into ridings where marginal improvements to campaign effort may translate into an electoral victory. Some local party organizations have an exceptional competency for carrying out partisan functions that shape their overall character. These organizations can be identified by the amount of money they raise, which suggests they also have the capacity to grow their membership, organize events, engage activists between elections, and prosecute successful election campaigns. While having a receptive audience in the district may help the local organization carry out these functions, it is the well-organized network of activists inside the district that demarcates highly competent organizations in very supportive ridings from incompetent 9

17 organizations in very supportive ridings. Well-organized local party organizations in very supportive ridings are more likely to select high quality candidates who may go on to serve in important parliamentary or executive roles. Some well-organized local party organizations may also be situated in less-competitive or very uncompetitive ridings; yet, even though they cannot elect their candidate, the local party remains exceptional at raising large amounts of money. This reveals that a very active and competent team exists on the ground. I define these highly competent partisan networks as mature local party organizations. Mature local party organizations can afford, if they choose, to move locally held resources to other party components without depleting the vitality of their own organization. Alternatively, some local organizations face greater challenges at raising money and this suggests less capacity to carry out partisan functions, like selling memberships, organizing events, engaging activists between elections, and prosecuting successful election campaigns. These organizations may select lower quality candidates, as indicated by their modest success at raising money which points to lower competency when compared to mature local party organizations. I define these party networks as developing local party organizations. Developing local party organizations in highly competitive or very uncompetitive ridings may still be building campaign machinery. Opposite to mature local parties, developing local party organizations may be situated in highly supportive constituencies. However, a higher chance of electoral victory may reduce incentives for the local team to carry out many partisan functions, such as membership drives, sustainer events, fundraising, or vibrant campaigns. In these cases the local organization may be less capable of carrying out the same activities found in mature organizations, even if their party is dominant in the riding. 10

18 The overage of resources at the riding-level allows mature local party organizations to shift some of their campaign effort into other ridings. This is possible because, while votes are geographically confined by district boundaries, money is a fungible commodity that can be easily transmitted to local party organizations in other districts. This method of party finance is defined as inter-association funding. Mature local party organizations may transfer party support in the form of inter-association fund-transfers to other local party organizations that require additional resources to increase campaign effort. Local teams receiving money are more likely to be developing a competent party organization. They may require interassociation funding to make capital intensive campaign effort on the ground. Parties vying to elect the most candidates to parliament rely on local party organizations as a source of flexibility and focus in distributing campaign effort to maximize effect. The modern party finance regime born from electoral reforms in 2004 and 2006 has two critical features that make inter-association funding possible. First, local party organizations cannot hide their wealth in the party finance regime. Transparency provisions extended to riding associations revealed how much money accumulated at the riding-level. The inclusion of riding associations in election law made local parties fully transparent organizations. Riding associations are compelled to report financial holdings or face dissolution by the state (Bill C-24, 2003). Second, the new party finance regime does not restrict intra-party fund-transfers (Bill C- 24, 2003), but it does restrict financial inputs and outputs (Accountability Act 2006; Bill C ). Registered party components can transfer unlimited quantities of money to other registered components, whether central or local. When combined with income limitations and election expense limits, local party organizations with overages of non-financial resources (such 11

19 as candidate quality and volunteer labour) can transfer financial resources to other campaigns. Local party organizations become a vital funding source for those unable to generate enough party support inside their riding. The regulatory environment that expanded transparency and limited income sources for local party organizations is the catalyst that encourages parties to employ inter-association funding. The Conservative Party of Canada (formed out the 2003 merger between the Canadian Alliance and Progressive Conservatives) was the only party to employ inter-association funding in any significant way between 2007 and In responding to a changed regulatory regime, elements of the Conservative Party recognized the opportunity to improve local campaign effectiveness and through this overall campaign efficiency by moving wealth between local party organizations. This is consistent with the institutionalist literature on parties and party systems (Carty 1992; Duverger 1966; Panebianco 1988). The massive amount of savings accumulated by Conservative local party organizations enabled inter-association funding. This is evident from aggregate data analyzed at the provincial level. Conservative local party organizations situated in highly-competitive ridings made more campaign effort than those in less competitive electoral areas of the country. The Conservative Party enhanced its electoral competitiveness in the Canadian party system by playing in the margins. The Institutionalist Approach, the Canadian Party System, and Case Selection Institutionalists attribute state institutions and rules as the primary cause of party organizations and party behaviours (for example, Duverger 1996; Panebianco 1988). The institutionalist approach to understanding party systems begins with identifying how political 12

20 organizations form around state imposed rules and institutions. Parties reorganize in idiosyncratic ways when alterations are made to election rules. Carty (1992) demonstrates this best in his analysis of three party systems defined in part from three sets of election rules. Conclusions in my work are consistent with institutionalists because I find a change to party finance rules led to changes in party organization and party behaviour. This thesis joins the work of other institutionalists like Ken Carty (1991; 1992; 2002; 2004; 2015), Royce Koop (2010; 2011; 2012), Anthony Sayers (1999), David Smith (1981), and Lisa Young (1998). Modifications to the electoral law produce changes to the organizational makeup of the Conservative Party of Canada and ultimately the party system. I argue the modern party finance regime, born from amendments to the Canada Elections Act in Bill C-23 (2003) and the Accountability Act (2006), is the primary cause of this change. The modern party finance regime banned corporate and union contributions to political parties. This lost source of income was replaced with per-vote subsidies for central parties, but local party organizations were forced to adopt new funding methods as no additional state subsidy was made directly available to them specifically. Two minor amendments became critical for encouraging collaboration between local parties. First, the modern party finance regime allows unlimited movement of financial resources inside political parties. Second, the regime requires all organizational components of the political party to publish their financial holdings annually. The recognition of riding associations in election law eliminated any possibility for candidates and campaign teams to hide money from other components of the party. Fleshing out the impact these regulatory changes have on parties may lead to new insights about the outcomes of recent federal elections. 13

21 Political parties that were effective at the national level between 2007 and 2011 have been selected for analysis. The Conservatives, Liberals, and New Democrats demonstrate their effectiveness by 1) fielding candidates in nearly every riding for the 2008 and 2011 federal elections; and, 2) forming either government or official opposition after the 2008 or 2011 federal elections. The Bloc Quebecois and the Green Party were also effective, to an extent, in both 2008 and 2011; however, they were not selected for analysis here because they do not meet both criteria established above. The Bloc did not field candidates outside Quebec and the Greens did not have enough influence in parliament to manifest coalition potential or blackmail potential against other parties (see Sartori 2005: 108-9). The main focus is on the Conservative Party of Canada. The Conservative Party is where collaboration between local party organizations is observed most. This party had the most extensive integration of local party organizations through inter-association funding than either other party. As a remnant from the Reform Party, the new Conservative Party of Canada incorporated past ideological predispositions into its present organizational structure by building up local party organizations with non-local resources generated by other grassroots teams. It is as if stereotypical Conservative activist s trepidation of an over-domineering, centralized authority led the modern party to build up competitive grassroots campaigns with the help of other local teams. The Conservative Party is also most suitable because it was the incumbent party in government throughout the period of analysis. The period between 2007 and 2011 is chosen for the time of investigation because regulatory changes in 2004 and 2006 are most likely to have altered each party as vote 14

22 gathering organizations. The primary reason for analyzing parties between 2007 and 2011 is the modern party finance regime does not change over two different elections. This allows for isolating how regulatory changes made in 2004 and 2006 impact party organization and the party system during the 2008 and 2011 federal elections. After the implementation of the Accountability Act in 2006, local party organizations could only solicit donations from individuals and contributions could not exceed $1,000. The secondary reason for analyzing parties between 2007 and 2011 is no other study presents financial data for this period. Readers can refer to Carty and Young (2012), Coletto, Jansen, and Young (2011), and Coletto and Eagles (2011) for a discussion about parties and elections between 2004 and Riding-level financial characteristics reveal four features of the Canadian party system between 2007 and 2011 that suggest each of the major federal political parties responded differently to party finance reforms. First, riding-level fundraising for the Conservative Party of Canada (CPC), the Liberal Party of Canada (LPC), and the New Democratic Party of Canada (NDP) reveal that the greatest electoral competition was primarily between the Conservatives and Liberals. Regardless of election results in 2011, fundraising shows that the Conservatives and Liberals had roughly equal capacity to mobilize activists in both election years. Thus it is less remarkable that the Liberal Party moved from third party status to a majority government in 2015, given that the party maintained competitive organizations at the grassroots. The second distinguishing feature of the party system was the Conservatives ability to sustain election resources over time. Conservative local party organizations had demonstrably larger savings entering election years, totalling more than the Liberals and the NDP combined. 15

23 The third feature was that Conservative local party organizations redirected their resource overages to other ridings through inter-association transfers. This feature is a departure from that of the past twenty years of analysis of local party organizations (see Carty 1991; Carty and Eagles 2005; Coletto and Eagles 2011). The final feature was that the Conservative Party had a better distribution of campaign effort than the Liberals or NDP. On average, Conservative local party organizations spent the most in highly-competitive ridings, followed by uncompetitive safe ridings and uncompetitive unsafe ridings. The spirit of my argument is to question claims that parties centralize campaign operations under party finance laws from 2004 and The financial characteristics of local party organizations between 2007 and 2011 remain the product of candidates, campaign teams, and party activists in each constituency. Diverging characteristics indicate that interassociation funding is the result of a variety of features produced at the riding-level. There is little financial evidence at the riding-level to suggest political parties and constituency campaigns are the product of central party offices. This is not to suggest local party organizations diverge completely from central party offices during campaigns. But local party organizations retain control over money, and subsequently maintain their independence from the central party office. Political parties and election campaigns remain embedded at the grassroots under new election law. Thesis Layout Structure This thesis is a preliminary analysis of parties and elections in Canadian constituencies under the new party finance regime. The primary focus of this thesis is on how local party 16

24 organizations transfer money among themselves in order to increase the number of competitive local campaigns. In this way, local decisions regarding where money might best be spent improve the overall competitiveness of the party. This points to the capacity of local organizations to respond rationally to the incentives embedded in the party finance regime, creating a secondary market for financial resources that allows them to move from ridings where resources are wasted to those where they might improve electoral outcomes. I incorporate an inductive logic to population data, isolate inter-association fundtransfers as the most notable characteristic, and analyze case studies of ridings where interassociation funding is observed. Chapter 2 explores the scholarship on the electoral system, political parties, and election law to establish incentives underlying the current party system. The logic of the electoral system encourages parties to have flexibility and focus at the ridinglevel in order to win the most seats in a highly regionalized, highly-competitive, carefully financed party system. Chapter 3 analyzes financing for the population of local party organizations between 2007 and The data show that the Conservative Party of Canada maximized the use of money by concentrating spending in highly-competitive districts, while the Liberals and NDP spent more campaigning to defend their own strongholds. Chapter 4 investigates case studies that reveal the local factors influencing collaboration between wellresourced campaigns and under-resourced campaigns located in highly-competitive ridings. The final chapter concludes with a research agenda to guide future studies of parties and elections. The Conservative Party of Canada is used as the reference point for parties and elections in many instances throughout this thesis. This is because the Conservative Party had the most 17

25 notable organizational structure and electoral tactics during the period of analysis. But at the same time, using the Conservatives as a reference point is preferential because the party has been under-explored by scholarship (see Carty 2015: ). There are two reasons for this. First, the Conservative Party of Canada is a young party; it has only existed for thirteen years at the time of writing (2016). And second, conservative parties in Canada have historically placed second to the Liberal Party of Canada. As Koop (2011: 11-12) points out, the Liberal Party won fifty-eight percent of the elections held since Confederation and this ratio increased to sixty seven percent after the enfranchisement of women in This pattern has produced five party systems that begin with Conservative interludes and finish with Liberal dynasties (Leduc et. al 2010). The most recent transition from majority Conservative parliament in 2015 certainly highlights the prowess of the dynasties and interludes theory, especially given that the new majority Liberal government was elected immediately after being the third party in parliament. Understanding how the Conservative Party of Canada is organized and identifying its interactions with other parties at the constituency-level may help make sense of Canadian parties and party systems into the twenty-first century. 18

26 Chapter 2 Harnessing Campaign Effort with Local Party Organizations: Continuity and Change in the Canadian Party System 19

27 Local party organizations are vital components of political parties because they are mechanisms for resolving support variations across the Canadian electoral system (Sayers 1999). The uneven distribution of party support across ridings and regions is a longstanding feature of Canadian party systems (for example, Cairns 1968). Local party organizations enable national parties to overcome support variations at the regional level. By concentrating campaign effort in strategically important ridings, political parties can improve electoral prospects at the districts level, and from them increase overall chances at forming government. Local party organizations that are capable of generating adequate election resources are also more likely able to generate votes (Coletto and Eagles 2011: 124). This may be problematic for political parties seeking to expand their number of parliamentary seats. Disequilibrium exists between where resources are generated and where they are most needed for campaigning. One way of resolving resource disequilibrium is by empowering local party organizations to generate support in accordance with district-level features. However, under the modern party finance regime (Bill C ; Accountability Act 2006) most local party organizations may only generate campaign resources from individuals within the riding or from other components within the political party. 2 Scholarship predicts that local party organizations will become increasingly dependent and submissive to central party offices in exchange for campaign resources generated at the centre (Carty and Young 2012: 236; Coletto et. al 2011; Coletto and Eagles 2011; Jansen and 2 There are exceptions like high profile politician Jason Kenney, whose prowess enabled him to raise money from individuals in ridings other than Calgary Southeast. 20

28 Lambert 2013). 3 But this theory assumes that only central party offices collaborate with local party organizations. In fact, local party organizations may also collaborate with other local party organizations (CBC 2013; Hill Times 2015). Collaboration allows local party organizations to resolve uneven support distributions by making indirect campaign effort beyond their own riding. It is likely that collaboration at the local-level will become an imperative for resolving resource scarcity, especially since subsequent party finance reforms eliminated per-vote subsidies paid to central party offices (Fair Elections Act 2013). Modifications to the party finance regime in 2004 and 2006 altered patterns of fundraising and campaigning done by local party organizations in the Conservative Party of Canada. The Logic of Electoral Contests in an Single-Member District Electoral Systems Political parties must incorporate the logic of the electoral system into their party organizations and campaigns if they are to be effective (see Sartori 2005). The logic of the single-member plurality (SMP) electoral system requires parties to convince voters to cast their ballot in a particular district on a particular day. Political parties must convince voters to cast ballots locally for national party brands. Parties with concentrated local and regional appeal may win office while others with thinly-spread national appeal may not. Local, regional, and national votes remain the outcome of ballots counted inside each district. This is why local campaigns continue to matter: parties need to convince voters they are the best choice locally and nationally. 3 The centralization theory is consistent with technological advancements, such as computer databases and internet communications, which may empower central party offices to override the functions of local party organizations (Flanagan 2003; Delacourt 2014). 21

29 Canadian political parties contest federal elections in a single-member plurality electoral system (SMP). In SMP systems, a single candidate wins election as the district representative by gaining more votes than any other candidate. A candidate does not need a majority of votes to win the district in a two party race. The first-place candidate only needs more votes than the second-place candidate. Political parties are elected to govern by winning the most riding-level contests. Similar to the riding-level formula, the winning party does not need to win more than fifty percent of the seats in parliament to win power. The party-in-power may govern with a minority of parliamentary seats, supplemented by the support of opposition parties during votes of confidence. 4 The SMP electoral system has become known as the first-past-the-post system (FPTP) because candidates and parties win election by placing first among the plurality of contestants. Parties may experience an overage of votes at the district-level. An overage occurs when there is a commodity surplus above a required quantity. For example, Candidate A gets 50.01% of the votes to win election against one other candidate in a two party system. After the ballots are counted, Candidate A actually receives 65% of the votes. Candidate A receives an overage of 14.99% of votes cast. The Canadian electoral system has the potential to create higher ballot overages because elections are usually contested by at least three parties in every district. Candidate A can actually win with only 34% of the vote if all three parties are competitive. The opposite is true for Candidate B and C. Losing candidates can suffer from a ballot shortage of one percent that prevents either from winning. The zero sum contest in the constituency 4 Minority governments were formed this way in 2004, 2006, and The plurality convention has lasted 90 years. The rules of responsible government enable the opposition parties to unseat a minority government; however, such a coalition has never successfully formed inside Parliament. 22

30 means that it does not matter whether a candidate wins by one vote or ten thousand votes; winning is winning. Parties may exploit ballot overages by identifying where they exist and incorporating that into campaign strategy. The potential for balloting overages and shortages which arise from the logic of SMP systems influences district-level voting behaviour. Cox (1999) argues that SMP inspires more local coordination among candidates and strategic voting among voters than occurs under PR systems. Candidates may enter or exit the local contest if they perceive a strategic advantage over other candidates. A candidate may enter an election race if she perceives that other candidates already running have left certain policies out of their platforms. Alternatively, a candidate may exit the contest if she perceives that her own platform overlaps with another, more competitive candidate. Strategic entry (or exit) occurs because the electorate votes for the candidate closest to their own policy preferences. Cox hypothesizes local coordination is greater in SMP systems because candidates and voters seek to achieve electoral outcomes closest to their own policy preferences. Strategic entry is intended to increase vote splitting, while strategic exit is intended to reduce vote-splitting. Candidate coordination is meant to produce district representatives closest to the policy preferences held by the plurality of voters. SMP systems have a psychological effect on voters and parties that rationalize the logic of the election contest: only one candidate wins the district and only one party forms government. This is the psychological logic of SMP voting. The Green Party of Canada is a contemporary example of the logic introduced by Cox. Green Party candidates enter riding-level contests because they believe others do not have 23

31 appropriate environmental policy proposals. 5 As a group, the Green Party has received upwards of 6.8 percent of the national vote during campaigns (Elections Canada 2013). However, Green Party voter support is spread too thinly across the Canadian electoral system for the party to be effective. The majority of voters cast their ballots for other parties at the riding-level. As of 2015, the dispersion of support across electoral districts creates an entry barrier to parliament for all but one Green Party candidate. The logic of SMP aggregates into regional contests in the Canadian case (Cairns 1968). Parties may have concentrated support at the district-level that translates into winning an entire region of Canada. Cairns (1968: 78) argues, Sectionalism in the party system is unavoidable as long as there are significant differences between the distribution of party voter support in any one section and the distribution in the country as a whole. According to Cairns, the uneven distribution of party support across the regionalized electorate shapes electoral competition between parties. Multiple parties can be effective in Canadian party systems because regionally-concentrated support and single-member plurality voting enables parties to become effective, even if it is in only one section of the electoral system. Parties may become effective by maximizing just enough voter support at the riding-level to be ranked first by voters in each electoral district of a given region. This is the mechanical logic of SMP in Canada. The regionally and ethnolinguistically-diverse Canadian electorate potentially magnifies the variability of electoral support because parties may enjoy highly stable support in one region and volatile support in others. For example, the Liberal Party of Canada dominated 5 The Green Party (2015) campaigns to ensure that the planet, the biosphere and our part of it, survive and thrive for our children and theirs and theirs... 24

32 federal politics during the 20 th century by winning the plurality of votes and the majority of seats in Quebec and Ontario (Johnston 1992). The Liberals managed to straddle conflicting regional and ethnolinguistic concerns by brokering disputes internally inside an easily accessible big tent political party (Carty 2015). Johnston (1992: 619) argues Liberal party electoral success throughout the 20 th century originates from its electability or balloting overages in Quebec ridings, like a trump card against other big tent parties such as the Progressive Conservatives (PC). Even though support stability in Quebec led to balloting shortages in Western Canada, the Liberal Party prolonged its duration in power by benefitting from balloting overages in seat-rich regions. In another example, the Reform Party of Canada sprang into Parliament by displacing PC candidates in western Canada. The PCs enjoyed nearly four decades of electoral predictability in Western Canada (Johnston 1992: 610). The Reform Party emerged as a western protest to policies approved by the Western Canadian PC caucus. Rather than ignoring regional conflicts, Reforms electoral strategy was to articulate regional conflicts by championing Western Canadian concerns. Reform Party candidates entered the electoral contest and gained a strategic advantage by filling position space left void by the PC party. The Reform Party initially sought to push the PCs out of Western Canada by invading from the margin to win a plurality of votes in western ridings (Flanagan 2009). The Reform strategy worked in the 1993 federal election; with the equivalent amount of national vote, the Reform Party (19%) won fifty-two ridings while the PCs (16%) only won two (see Carty et. al 2000: 32-3). 25