Rethinking the Franchise Party: Adding the Ideological Dimension—The Irish Case

Rethinking the Franchise Party: Adding the Ideological Dimension—The Irish Case Abstract Attributing the label ‘franchise’ to parties that possess certain organisational characteristics, such as centralized authority at the national level combined with substantial autonomy at the local level, has become increasingly common. These franchise characteristics have been identified in parties in a wide variety of country cases, in party systems as diverse as those of Canada, Australia, Israel and Ireland. Drawing upon evidence from three original parliamentary surveys in the Republic of Ireland, this article seeks to expand our understanding of franchise parties by focusing on the ideological behaviour adopted by franchise parties at both the national and the candidate levels. An analysis of the Irish case suggests that, in addition to previously studied organisational characteristics, candidates from franchise parties enjoy an unusual degree of flexibility to vary their policy positions. 1. Introduction Parties of all types modify their organisational structures in the face of increasingly competitive elections and more fluid social environments. One type of organisational response is for parties to become more stratarchical, whereby distinct levels within a party operate more or less independently of each other, resulting in parties that behave like franchise systems. Similar to franchises in business, the concept of a franchise party implies that the overall brand is determined at the national level, including selecting strategies, policies, marketing and overall management of the party, while party candidates and organisations at the local level are relatively free to deal with the interests of the communities they represent. Franchise parties find ways to combine growing centralisation at the national level with substantial autonomy at the local level where local party officials can often designate candidates of their own choice and can tailor local campaigns to address the particular interests of local constituents. This franchise model of party organisation seeks to reconcile the competitive imperative for greater centralised authority at the national level with the need to manage diverse demands from constituents (Carty, 2002, 2004; Wolinetz, 2015, pp. 73–75). For nearly two decades, scholars have been highlighting the mounting trend of parties to behave more like franchise parties in contexts as diverse as Canada, Australia, Israel and Ireland (Katz and Mair, 1995; Carty, 2002, 2004; Wolinetz, 2015). Parliamentary systems appear to be especially conducive to franchise arrangements because they allow ideologically cohesive parties to vary their ‘product’ to match the tastes of consumers in local markets. Similarly, electoral systems that place a premium on the personal popularity of individual candidates provide additional inducements for parties to behave in a franchise-like fashion, allowing local personalities opportunities to capitalise on their embedded social networks. It is not necessary to view the empirical manifestation of the franchise party in terms of a binary category, but more usefully can be seen along a spectrum of more ‘franchise-like’ or less. Although scholars have helpfully attributed the label ‘franchise party’ to a growing number of parties and have identified certain crucial organisational characteristics that the parties adopt, such as nominating candidates to run for office at the local level, few scholars have focused on the ideological attributes and behaviours of franchise parties. When ideology is taken into account, the focus is usually on the national level and analysing party manifestoes or roll call votes in parliament where a high degree of party discipline is often observed. Yet, a richer and fuller understanding of how franchise parties employ programmatic appeals can be achieved by complementing the attention to organisational adaptation with increased attention to adaptations within the ideological dimension. Specifically, what has been missing in the literature on franchise parties has been an awareness of the relationship between the party platform at the national level and the policy positions adopted by local members of parliament. Conceiving parties as franchise systems illustrates how parties may benefit competitively from allowing their individual candidates to vary their ideological positions, either to enhance candidates’ specific electoral appeal or to retain broad support from the political elite and attract high-quality candidates. Tolerating a degree of relative autonomy regarding policy positions helps parties appeal to voters with diverse, high-quality candidates and in diverse settings. However intentional or unintentional, by turning a blind eye when candidates adopt positions different than the overall party platform, national party organisations stand to gain electoral ground. Although members of parliament may need to toe the party line in parliament while in office, they can, and often do, vary their policy positions during campaigns and in private intra-party discussions. This article underscores the relative autonomy of franchise party candidates to adopt programmatic appeals opportunistically during elections. By focusing on the relationship between the programmatic commitments articulated in the party platform at the national level and the often divergent positions adopted by franchise party candidates at the local level, I examine a largely neglected dimension of the behaviours characteristic of franchise parties, that is, the ideological dimension. I argue that, in addition to tell-tale organisational adaptations characteristic of the franchise party, a fuller account of their behaviours can be achieved by focusing on the relatively greater degree of ideological autonomy enjoyed by franchise party candidates. The evidence to support my argument is drawn largely from the quantitative and qualitative data derived from three originally designed and implemented parliamentary surveys conducted in the Republic of Ireland. These surveys asked Irish MPs (known in Ireland as Teachta Dála or TDs) to place themselves and their party on a 0-10 continuum on six policy issues, and the results reveal that considerable ideological dispersion exists within parties. Although the surveys do not provide direct evidence to link what the TDs said in private during the face-to-face interviews to how they campaigned in their constituency (and there is limited space to explore comprehensively other campaign literature), the surveys explicitly asked TDs to reflect on their experience during the campaign. More specifically, the TDs were asked to locate themselves, their party and (in the 2016 survey) the average person in their constituency for each policy in the ideology section. The survey asked TDs to reflect on their attitudes on issues. The considerable intra-party variation across the issues shows that individual candidates differ from their parties on various issues, which has strategic benefits for a franchise party. This intra-party ideological variation underscores the flexibility individual candidates enjoy to adopt slightly different positions from their party. Fully explaining this variation is beyond the scope of this article. However, highlighting the prevalence of this intra-party ideological variation suggests that TDs are able to adopt their ideological positions—for a variety of personal reasons—to enhance their appeal. This is particularly relevant despite the fact that there is strict party discipline within Irish parties once in parliament, a majority of Irish voters are still attracted to candidates for non-ideological factors, and the overall ideological spectrum is comparatively narrow. The quantitative evidence from the surveys is reinforced by qualitative evidence from hundreds of interviews to suggest that individual candidates do seek to attract support from voters based on the ideological positions they adopt—even if this means differing slightly from the official party position. A recurring theme from the overall set of interviews is that Irish candidates have to be ‘all things to all people’ and must do what it takes to gain an edge over other candidates within the constituency, and sometimes their programmatic views provide this difference, especially on salient local issues. Since Ireland’s independence in 1922, social or economic ideologies were never substantial issues in Irish politics. The main cleavage was nationalism. In that context there was always a healthy and sometimes subliminal tension between national party positions and what candidates proposed locally. This was further bound up with the fact that national politicians were also candidates in local elections. A good example of this tension in elections is the issue of local hospitals. In 1973, reporters commented on how the status of local hospitals dominated that election: ‘Paradoxically, the answers they [candidates] give, which are most likely to please many of the local pressure groups, would not always result in an improvement in health care ….’ nor did they necessarily coincide with national party position.1 More recently, candidates often determine whether they adopt their party’s position depending on whether that policy offers specific versus general benefits. During the 2007 election, Michael Fitzpatrick was a member of the historically dominant party Fianna Fáil, which was in power for the majority of the Irish state's existence and routinely won over 40% of the vote at every general election up to 2011. Although the Fianna Fáil government had announced that there would be only one national children’s hospital located in Dublin in 2006, Fitzpatrick decried that decision on local radio, arguing that the children services in his constituency’s hospital desperately needed to be enhanced. He defended his position against that of the party arguing that he thought someone in the party may be persuaded by his arguments. Yet, when asked about his policy stance on other policy positions, such as the nurses’ strike, he evinced incredulity that one might take a position in opposition to that of the party: ‘We are the government party. We will always be supportive of the government line (M. Fitzpatrick, interview, July 2007).’ In Irish politics, candidates seek to maximise both party and individual support. When there is a conflict, candidates often adhere to the party positions when more general benefits are at stake, such as wages, taxes, etc. But when push comes to shove in a matter of local benefits to the constituency, local candidates will readily jettison national party positions so as to not expose themselves to attacks from their local opponents. In a highly competitive and narrow electoral arena, candidates can ill-afford to lose out on such appeals; hence, Fitzpatrick. When one examines the policy appeals of the other candidates, they, too, seek to mobilise voters on the basis of both strong party support and specific policy benefits connected to hospitals, schools and other services to the local constituency. As this example shows, the Irish case is a useful context to explore the behaviour of franchise parties because several of its major parties behave as franchise systems that, especially during electoral campaigns, combine strong national party organisations with highly mobilised local candidates that embody the party brand. There has been an explosion of literature (Kavanagh, 2015; Murphy, 2016; Marsh et al., 2017; Reidy and White, 2017) assessing the changing nature of Irish politics since Ireland’s earthquake election in 2011, which saw Fianna Fáil decimated as it fell to 17% of the vote and from 78 seats to 20 in the national parliament. For many, Ireland may finally be shifting from Civil War politics to a context where ideology plays a more significant role. This article examines previously understudied dynamics within Irish politics and illustrates how ideological flexibility at the level of individual parliamentarian is one way in which parties can adapt to an evolving electoral context and can behave more like franchise systems. The article is organised in the following way. First, I provide a brief overview of the Irish party system and offer a rationale for why Ireland serves as a useful case study for examining the broader phenomenon of franchise parties. Secondly, I describe the prevalence of intra-party ideological diversity within major Irish parties and argue that this set of adaptive behaviours is indicative of franchise parties. Parties can utilise internal ideological diversity to bolster their attractiveness locally, a trait especially useful in more volatile electoral environments. The article then turns to an analysis of the factors that shape the ideologically diverse positions taken by individual Irish politicians within parties. By analysing the linkage between the ideological positions adopted by individual candidates during elections and comparing these with how the party markets itself at the national level, I seek to add to our understanding of the multiple levels on which franchise parties employ programmatic appeals to broaden their attractiveness to voters. The findings suggest that Irish parliamentarians, especially among major parties, possess considerable autonomy to select their policy appeals during elections, an autonomy which is shaped by the individual parliamentarian’s demographic background and electoral and constituency context. The article concludes with a discussion of how the approach adopted here can augment our understanding of how parties compete more generally. 2. The Irish case Unlike many other Western European party systems, the Irish party system did not emerge from sharp social divisions such as class, but rather from a split emanating from within the original Sinn Féin party, which was the national Irish political party prior to the Irish Civil War (1922–1923). It has some organisational continuity with the contemporary Sinn Féin party, but also with Fine Gael and Fianna Fáil. Sinn Féin in the contemporary context represents a combination of the former political wing of the Irish Republican Army (IRA) and left-leaning politicians who seek to create a United Ireland. Key founding conditions strengthened the cross-class nature of partisan ideological appeals and contributed to the overall weakness of the left in Irish electoral politics (Marsh et al., 2008, p. 32). Since the consolidation of the party system in the 1930s until 2016, Fianna Fáil, Fine Gael and Labour have managed to attract an average of over 80% of the total vote. Much of this predominance is due to the major parties’ use of non-ideological appeals, organisational strength and their ability to effectively employ both partisan and personalistic appeals within Ireland’s Proportional Representation-Single Transferable Vote (PR-STV) electoral system. Ireland’s parties now face a more demanding electoral environment than they faced in elections prior to the 1980s. In recent years, the underlying cleavage structure in Ireland, as well as the political value system and the issue dimensions relevant for elections, have shifted in critical ways. The demographic composition of the electorate has dramatically changed as Ireland has transitioned from one of the poorest to one of the wealthiest European countries in less than a generation. Historic cleavage patterns have largely lost their hold. This turbulent environment has resulted in lower turnout, lower party attachment and increased numbers of floating voters. For example, partisan identities have declined precipitously: the percentage of voters in the Irish electorate reporting that they lack a close relationship with any party has more than doubled from 34% in 1978 to 79% in 2016 (Marsh et al., 2008, pp. 59–79; Marsh and McElroy, 2016, p. 179). A rapidly changing Irish society and an increasingly competitive electoral marketplace have combined to increase the prospects for electoral volatility. Taken together with the complex and technocratic character of the issues facing modernising Ireland, parties have been forced to adapt to these challenges or suffer electoral defeat. As a result, major party organisations have become more centralised, professionalised and bureaucratised. These trends, in turn, have carried with them important consequences, resulting in an enhanced capacity for party elites at the national level to coordinate efforts. In addition, these developments have necessitated a streamlining of how parties select candidates and leaders, raise money and communicate policy priorities. At the same time, the imperative of efficient local delivery of goods and services and the personalised nature of electoral appeals incentivise parties to field candidates who can simultaneously sustain multiple linkages to government offices at every level. Successful candidates must simultaneously maintain linkages at the national party level and simultaneously engage successfully in Ireland’s complicated civil service protocols to deliver goods, services and information locally. The parties’ ability to field strong candidates that work effectively at both the national and local levels has been crucial for their longer term electoral success. Although it is has been 50 years since Chubb (1963) famously suggested that the role of Irish TDs is to ‘go about persecuting civil servants’, the demand for unusually robust constituency service and the expectation that TDs will serve as brokers between voters and the state continue to dominate Irish politics. On average, 40% of voters claim to choose their TDs primarily based on the candidate’s ability to look after the constituency’s needs. Additionally, people from all sectors of Irish society, including educated and upper-income citizens, continue to use their elected representatives in a brokerage capacity to access goods and services from the state. Prior to the 1990s, it was easier for elected officials to maintain well-oiled local machines and to deliver the goods. Before 2003, Irish TDs conveniently held both the parliamentary post and a local council office, called the ‘dual mandate’. Given that more than half of all government spending, although funded by central government transfers, is allocated locally; local councils provided parliamentarians a rich source of patronage. In addition, locally elected councillors hold dispositive authority regarding issues pertaining to planning permission/zoning, roads, medical cards, waste management and so on. The elimination of the ‘dual mandate’ deprived members of parliament of an important source of patronage, rendering it more difficult after 2003 for TDs to confer favours and deliver the goods to local constituents. With little perceived programmatic difference among Ireland’s parties during elections, party organisational capacity and personalistic appeals become essential mechanisms by which candidates and parties establish and maintain linkages and make their case with voters. However, whereas in the past Irish TDs may have relied primarily on non-ideological appeals to attract voter support, circumstances have changed. As Irish society has become more fluid and urban, politicians have had to develop a broader array of tools to compete in this increasingly competitive environment. While constituency service remains the principal factor driving vote choice (Marsh et al., 2008, pp. 220–238), there has been a growing influence of policy position on vote choice according to elections surveys. Between 1977 and 2016, approximately 39% of voters claimed that constituency service most influenced their vote choice compared with 26% who reported that party policy matters most. The 2011 (41%) and 2016 (33%) election polls indicated above average results for those who reported voting based on policy considerations. Thus, a growing insistence among voters for more sophisticated policies requires parties and candidates to respond accordingly. One important way in which political entrepreneurs have adapted to this changing set of circumstances is to tailor ideological appeals to fit voter profiles within their local constituencies. Data drawn from parliamentary surveys suggest that the policy positions adopted by individual TDs can diverge from the positions adopted by their respective parties at the national level. This is particularly true of major party (i.e. Fianna Fáil, Fine Gael and to a lesser extent Labour) TDs. Even though major parties are perceived as holding very similar programmatic views at the aggregate level and TDs typically adhere to strict party discipline in legislative votes, the high degree of internal ideological differences observed during elections reflects significant levels of autonomy that individual TDs have to adapt in their policy. The analysis of campaign material and speeches, while interesting, is outside the scope of this article. Therefore, the responses TDs share in surveys provide indirect evidence for how they shaped programmatic appeals to voters during election campaigns. Nevertheless, it is critical to note that the Irish TDs interviewed in the three waves of the survey were asked to identify their individual positions and those of their party on the same policy continuums during each respective election. Although there is some inevitable variation in whether a specific TD construes their party’s position as a ‘6’ or a ‘7’, if a TD places himself one or two points to the left or right of the party on an issue, there is some reason why they did not place themselves equal to their party. Although it can be helpful for Irish TDs to announce their ideological positions in very public ways, it is widely known that the primary way candidates interact with individual voters is during door-to-door canvassing during election campaigns. In surveys, most TDs surveyed reported that door-to-door canvassing was the single most effective means of connecting with voters—with 72% in 2007, 68% in 2011 and 86% in 2016, highlighting the importance of this critical personal contact. Because these personal interactions constitute the grit and gristle of Irish politics, candidates seek to establish a strong personal connection with the voters during these encounters on the doorstep. Candidates are exposed to and listen to whatever issues voters identify as being important to them and share, in turn and when time permits, their support of these views. The overall impression from doing over 300 interviews is that the views TDs report in the parliamentary surveys represent their sincere positions on key issues facing the electorate. 3. Intra-party ideological diversity It should come as no surprise that Irish parties evince a high degree of internal ideological diversity. Political parties in a wide range of comparative contexts experience considerable heterogeneity of ideological positions among their individual politicians. Parties can, and do, in fact, capitalise on the diverse views held by their elected representatives to win votes, even if representatives generally follow the party line when voting in parliament. This intra-party ideological diversity occurs not only in Ireland, but also in countries with traditions of broader and more clearly defined ideological competition. Data from the Comparative Candidates Survey (CCS) show that party affiliation explains between 60% and 80% of the variance in candidates’ ideology among a broad number of country cases, including Germany, Hungary, Sweden, Switzerland and Portugal. For example, in Germany following the 2005 election, 62% of the variance in MPs’ ideological positions was between-party variance, while 38% was within-party variance. In Ireland, by contrast, less than 15% of the variance is explained by party affiliation and the remainder by internal party differences (Courtney, 2015). The prevalence of ideological diversity within parties represents an additional tool parties have at their disposal to increase their vote in highly competitive contexts where there is a need to attract votes from various segments of society and from among unattached voters. Although it is widely known that many parties evince intra-party ideological diversity, what explains this diversity? Why does it matter? At first glance, internal party heterogeneity appears to matter little to Ireland’s political parties and to large portions of the electorate. Parties at the national level in Ireland establish party policy positions and enjoy rigorous party discipline. Even one vote against the party within parliament generally results in immediate expulsion. In a comparison of 15 developed democracies, Irish parliamentarians voted programmatically with their party more than any other country in the survey (van Vanno et al., 2014, p. 124). As a consequence, what individual politicians profess in terms of their personal programmatic preferences may not matter to the national party. Nevertheless, evidence from my 2016 Parliamentary Survey indicated that across all parties, strong majorities of TDs were ‘likely’ or ‘very likely’ to express opinions privately within their party—whether to party leadership, to fellow TDs, or in parliamentary meetings. TDs were less willing to broadcast disagreements outside the party. A narrow majority (52%) said they were likely to express opinions privately to constituents—although Sinn Féin, Green and left party TDs were less likely to do so, compared with nearly two-thirds of major party TDs. This suggests that the beliefs of TDs may impose a potential constraint on the behaviour of party leadership. Although personality and local services matter a great deal in shaping voter preferences, there is still some room for policy to be important. Internal party heterogeneity, and the ideological flexibility it provides at the local level, does seem to matter to individual Irish politicians who must find any way possible to distinguish themselves in an ever more competitive electoral environment. The importance of a candidate's personal interests and reputation increases in systems like Ireland’s where co-partisans compete with each other for votes and seats (Carey and Shugart, 1995, pp. 430–431). Ireland's PR-STV electoral system provides voters with a requirement to choose among both candidates and parties. This electoral formula incentivises behaviours, which highlight intra-party distinctions among co-partisans as they compete, not only against candidates from competing parties, but also against members of their own party. To succeed in this competitive market with fewer attached voters and increased competition from other parties and independents, candidates have been driven to develop highly personalized organisations—or local franchises—to gain their party’s nomination and to win their seat in parliament. Because non-ideological factors, candidate quality and organisational strength are crucial for candidates’ success in the Irish context, parties have an incentive to tolerate ideological autonomy from TDs, as long as TDs vote as instructed when in office. This ideological autonomy is tolerated regardless of whether TDs use it to advance locally popular arguments or advocate for personal beliefs and causes. Parties cannot afford to alienate effective local TDs or foist locally unpopular positions upon them, which means programmatic diversity matters for Irish franchise parties’ strategy. 4. Franchise parties in the Irish context Ideological competition in the Republic of Ireland underscores how the franchise party operates. Irish parliamentarians combine extremely coherent ideological party positions in parliament with high levels of constituency service and varied programmatic appeals during elections (Farrell et al., 2015; Wolinetz, 2015, p. 87). Behaving as franchise parties allows Irish parties to reconcile these apparently contradictory imperatives. Individual politicians can vary their ideological positions to enhance their appeal for several reasons. First, while candidates may want to offer distinct policy appeals from their opponents, they may also choose a position closer to their opponent, thereby de-emphasising the importance of policy and increasing relative importance of his or her valence advantage. In effect, such an approach encourages the median voter to conclude that if candidates are so similar on policy, then they will cast their vote based on other, more personal characteristics and abilities (Groseclose, 2001, pp. 862–863). This returns the advantage to those who are better positioned to serve local constituency needs once policy concerns are neutralised. A second reason to adopt diverse or contrary views from the party might reside in an effort to ‘gain points’ in line with Mayhew’s position-taking strategy (Mayhew, 1974). According to one senior elected Irish politician, and reiterated by scores of others, TDs regularly attempt to persuade constituents that they hold deep sympathies for positions most favoured by local constituents, and indeed argued strenuously for such positions during parliamentary party meetings, but claim to have been outvoted. These same politicians then protest that they must remain within the party, and in power, if they are to better serve the local constituency (anonymous interview with Fianna Fáil politician, 2 June 2016, Dublin). In this way, hundreds of TDs reported how they must present themselves as being simultaneously sensitive to local concerns and loyal to the party, which they convince voters help them to serve constituents more effectively over the long run. A third reason individual TDs may feel free to advance divergent views during elections as a means of establishing a deeper connection with voters is because they can reasonably gamble that they may never be forced to vote on the contended issues in parliament. Previous scholarship has demonstrated that many of Ireland’s more controversial issues have in fact been displaced from the electoral and parliamentary arenas to non-electoral institutional domains, and that as a result, many salient issues facing Irish society are simply side-stepped by parliament (McGraw, 2015). The practice of regularly displacing highly controversial issues from the electoral arena to non-parliamentary domains provides individual TDs greater degrees of freedom to take positions on these same potentially divisive issues that most resonate with their constituents. Finally, TDs may hold positions for personal rather than strategic reasons. Given the importance of personal appeal in Irish voting behaviour, parties have a strong incentive to recruit appealing local candidates. The need for locally appealing candidates may override the need for policy and ideological orthodoxy, as long as these candidates ultimately follow the party line in parliament. Franchise parties are likely to tolerate ideologically diverse candidates TDs regardless of the reasons for this diversity, so long as it does not threaten party discipline. 5. Measuring intra-party ideological diversity I designed three original surveys of the Irish Parliament to capture party positions, and in particular, to analyse how individual parliamentarians vary their programmatic appeals. This allows us to look at how major Irish parties behave as franchise systems that operate differently at the national versus the local level. The surveys highlight both aggregate party positions and individual TD attitudes on key issues in Ireland’s 2007, 2011 and 2016 general elections. While earlier research highlights ideological flexibility at the level of the party, this article focuses on this same behaviour at the level of individual politicians. Each of the surveys involved face-to-face interviews where elected TDs were asked to reflect on their experience in that election. My research team interviewed 102 of the 166 Irish TDs after the 2007 general election, 115 of 166 TDs after the 2011 general election, and 97 of 158 TDs after the 2016 general election. The three surveys occurred in very different political contexts: the roaring ‘Celtic Tiger’ economy in 2007, the aftermath of the worst economic crisis in Ireland’s history in 2011 and modest recovery by 2016. These three elections saw unusually high turnover and represent a virtual earthquake in the Irish electoral landscape: Fianna Fáil secured a strong mandate to govern in 2007, dropped from 78 to 20 seats at the 2011 elections, and partially recovered in 2016; Sinn Féin incrementally increased its number of seats, and small leftist parties also made marked gains. The survey sample was representative of key social and political demographics, such as age, gender, religious denomination, education levels, party, type of geographic constituency and previous experience in office (see Supplementary Materials for additional detail). Similar to other recent studies, evidence from these parliamentary surveys also confirms that there is more attitudinal variance within Irish parties than between them. Courtney found that 87% of ideological variance occurred within parties compared to 74% in my surveys (Courtney, 2015, p. 185). Given these results, it is instructive to look beneath the aggregate level of the party, where high levels of programmatic convergence are commonly observed, to the level of individual politicians, where a more nuanced picture of the array of ideological attitudes emerges. The surveys asked TDs to place themselves and their parties on a 0–10 scale within a number of policy areas: left–right position, and policy attitudes towards Northern Ireland, European integration, abortion, fiscal policy and environmental policy. (Specific information regarding the survey, including precise question wording, is available in Supplementary Materials.) These issues measure the most salient policy dimensions within contemporary Irish politics and society (Sinnott, 1995, pp. 279–297; Shu, 2003). To test what determines individual MPs’ attitudes, three sets of variables were included in this analysis. Party affiliation is the most common, and usually strongest, determinant of position. Key electoral and constituency level variables may also encourage politicians to vary their positions to meet personal or strategic needs. These latter variables test whether geography, internal party competition, levels of electoral competitiveness and other constituency factors influence TD ideological positions. The assumption is that key competitive dynamics could encourage candidates to alter their ideological positions to offset their competitors. Including these variables allows us to examine how changes in Irish society and the increasingly competitive electoral environment may be affecting ideological competition. Demographic variables are also likely to affect candidates’ positions, as they reflect the life experiences that sway personal political views. If demographic variables play a large role in determining party position after controlling for other factors, this suggests that TDs’ personal views and parties’ needs for attractive candidates drives the ideological diversity of Irish franchise parties. Thus, I test whether TDs adopt locally varied positions for personal demographic reasons, and/or whether TDs adopt locally varied positions for strategic reasons. The findings are somewhat inconclusive, which suggests that both factors may be at work. Even though there is no systematic explanation for why individual politicians choose their positions, the results suggest that Irish politicians enjoy an unusual degree of flexibility to employ ideology strategically to complement an array of other strategies. As previously noted, in contrast to the ideological convergence that is reflected at the aggregate level of the party, the level of ideological heterogeneity is greater at the level of the individual politician, reflecting a broader spread of policy positions among individual MPs. Table 1 shows that Irish TDs maintain a wide diversity of views on key issues. Table 1 Standard deviation of TD attitudes in 2016, by party (See Supplementary Tables S1 and S2 for similar results in 2007 and 2011) FF FG Lab Gr SF Ind Left Brown–Forsythe test statistic (Number of cases) 36 22 3 2 16 7 10 United Ireland 1.9 1.7 1.2 0 0.6 1.9 3.4 3.9** Left–Right 1.1 1.5 1.5 1.4 1.4 1.8 1.8 0.4 Abortion 2.5 3.1 1.7 0 2.9 3.8 2.1 1.4 EU 1.7 1.1 1.5 0.7 1.6 3.4 2.9 4.4** Fiscal policy 1.1 1.7 1.2 1.4 1.1 1.6 2.0 0.7 Environment 1.4 1.9 2 0 1.6 1.6 1.9 1.0 FF FG Lab Gr SF Ind Left Brown–Forsythe test statistic (Number of cases) 36 22 3 2 16 7 10 United Ireland 1.9 1.7 1.2 0 0.6 1.9 3.4 3.9** Left–Right 1.1 1.5 1.5 1.4 1.4 1.8 1.8 0.4 Abortion 2.5 3.1 1.7 0 2.9 3.8 2.1 1.4 EU 1.7 1.1 1.5 0.7 1.6 3.4 2.9 4.4** Fiscal policy 1.1 1.7 1.2 1.4 1.1 1.6 2.0 0.7 Environment 1.4 1.9 2 0 1.6 1.6 1.9 1.0 Significant at **p ≤ 0.01; *p ≤ 0.05;  ^ = .10, * = .05, ** = .01. Table 1 Standard deviation of TD attitudes in 2016, by party (See Supplementary Tables S1 and S2 for similar results in 2007 and 2011) FF FG Lab Gr SF Ind Left Brown–Forsythe test statistic (Number of cases) 36 22 3 2 16 7 10 United Ireland 1.9 1.7 1.2 0 0.6 1.9 3.4 3.9** Left–Right 1.1 1.5 1.5 1.4 1.4 1.8 1.8 0.4 Abortion 2.5 3.1 1.7 0 2.9 3.8 2.1 1.4 EU 1.7 1.1 1.5 0.7 1.6 3.4 2.9 4.4** Fiscal policy 1.1 1.7 1.2 1.4 1.1 1.6 2.0 0.7 Environment 1.4 1.9 2 0 1.6 1.6 1.9 1.0 FF FG Lab Gr SF Ind Left Brown–Forsythe test statistic (Number of cases) 36 22 3 2 16 7 10 United Ireland 1.9 1.7 1.2 0 0.6 1.9 3.4 3.9** Left–Right 1.1 1.5 1.5 1.4 1.4 1.8 1.8 0.4 Abortion 2.5 3.1 1.7 0 2.9 3.8 2.1 1.4 EU 1.7 1.1 1.5 0.7 1.6 3.4 2.9 4.4** Fiscal policy 1.1 1.7 1.2 1.4 1.1 1.6 2.0 0.7 Environment 1.4 1.9 2 0 1.6 1.6 1.9 1.0 Significant at **p ≤ 0.01; *p ≤ 0.05;  ^ = .10, * = .05, ** = .01. Whereas TDs within the three historic parties exhibit a relatively wide dispersion of views on almost every issue, TDs within more policy-driven parties exhibit far fewer internal party differences on their primary policy concerns. However, even policy-driven parties show substantial internal differences on more peripheral issues. These non-major parties evince less ‘franchise-like’ behaviours than the major parties, but still tolerate ideological diffusion on some policies. Although caution must be exercised in interpreting these results given the small sample size for the minor parties, the more cohesive results for these parties suggest that they are more ideologically coherent on core issues. Consider the case of Sinn Féin. Naturally, Sinn Féin TDs offered clearer, more unified positions on its core policy issue related to United Ireland, holding essentially identical positions on that issue. It also held the most coherent set of positions on the left–right scale. However, Sinn Féin TDs hold increasingly diverse views the further one’s moves from the party’s core issues. Sinn Féin TDs were less cohesive on spending and the environment, and the party experienced growing diversity among its TDs in terms of abortion. In fact, Peadar Tobin, TD, was temporarily suspended from Sinn Féin for his vote on the same 2013 abortion bill that divided Fine Gael. Interestingly, Tobin was reinstated six months later in February 2014 with little fanfare and Tobin stated he was delighted to be back in his political home (Irish Independent, February 5, 2014). Results from the United Left Alliance in 2011 display a similar pattern to Sinn Féin. Also, the Green TDs held reported identical positions on half of the issues. Figure 1 charts TD attitudes on three fundamental issues within Irish politics for the 2007, 2011 and 2016 elections: the founding united Ireland cleavage, the left–right scale and abortion. A diversity of positions exists at the level of the individual TD along various policy dimensions, irrespective of whether there is political contention over an issue (e.g. abortion), it is the perspective of the system as a whole (e.g. left–right), or uncertainty as to how to achieve longer term goals (e.g. United Ireland). Figure 1 View largeDownload slide TD attitudes on salient issues in 2007, 2011 and 2016 elections. Figure 1 View largeDownload slide TD attitudes on salient issues in 2007, 2011 and 2016 elections. These findings underscore an essential feature of Irish politics: individual TDs and candidates from major parties exercise a considerable degree of personal discretion whether to embrace—or distance themselves—from the policy positions taken by their co-partisans. The utility of a high degree of personal discretion becomes apparent when the dynamics of the Ireland’s PR-STV electoral system are again considered. Because candidates from the major parties do not rely on any one social group or policy appeal for their votes, they necessarily rely on greater ideological flexibility. It is easier for these latter candidates to emphasise policies that increase their appeal within the constituency, or to emphasise policies that fulfil their personal ideological and policy goals. At the national level, the major parties are able to highlight their official party position, while at the local level their candidates are relatively free to offer more targeted appeals. By contrast, candidates from more policy-driven parties naturally have less incentive to jettison or modify the party line on their core issues since policy goals are central to their very candidacy (Tavits, 2007). 6. What explains intra-party ideological diversity? What explains the intra-party ideological variation that exists within Irish political parties? Or, in other words, what are the factors that determine the diverse policy positions adopted by individual parliamentarians? In addition to party affiliation, there are two main mechanisms that shape TDs’ political views. First, TDs hold ideologically diverse views, as a matter of course, because they are recruited from an ideologically representative cross-section of society and hold personal political leanings informed by demographic and family background (Byrne and O’Malley, 2012). Moreover, TDs are influenced by the views held by their neighbours and fellow constituents, which, in turn, shape the specific electoral challenge they face. The ability of TDs to adopt programmatic positions opportunistically may be highly consequential for their electoral prospects, for it allows them to maximise their appeals among blocs of voters within their local constituency which can engender voter loyalty. A series of linear regression models are employed to test the way in which these three sets of factors—party affiliation; demographic and family background; and local context—shape MPs’ positions on policy issues. These factors hold the potential to explain the high levels of intra-party ideological diversity among the major parties observed in the previous section, while simultaneously maintaining a high degree of ideological consensus at the aggregate level. As dependent variables, the models use TD position on the six policy areas described previously. I seek to predict policy positions using three sets of independent variables. First, there is a notable but relatively low-magnitude relationship between party and MP’s ideological positions. When controlling for party, I then show how background and local context supplement party as predictors of TD positions. Because the surveys of TDs necessarily covered a relatively small number of cases, I test background and local context via a series of separate models, rather than aggregating them into a single model. This allows us to test a large number of independent variable specifications without greatly limiting the degrees of freedom and undermining the explanatory power of the findings. Most of the models presented are ordinary least squares linear regressions. I used clustered standard errors to account for TDs who completed multiple surveys, as well as a fixed effect for a year. Because Ireland uses multi-member electoral districts, the analysis of constituency-level variables was conducted using hierarchical models that nested individuals within electoral districts and years. The analysis of some specifications for electoral marginality uses the difference between TD and party position as a dependent variable; because this variable takes a small number of values (usually a 0-, 1- or 2-point difference), I treat the dependent variable as categorical and use an ordered logit model. 6.1 Party affiliation While party affiliation is not an overwhelming predictor of TDs’ attitudes, it is still relevant. The R2 values indicate how much of the variance in TDs’ attitudes is explained by party affiliation, along with a fixed effect for year-to-model shifts in ideology between the 2007 and 2011 elections and between the 2011 and 2016 elections (Table 2). For example, 45% of the variation in attitudes towards the European Union (EU) is explained by party affiliation and year, while 55% remains unexplained. Party affiliation and year explained less than half of the variation in TD attitudes on the other salient issues as well. Labour, Sinn Féin and the United Left Alliance TDs tend to show differences from Fianna Fáil (the base category), but Fine Gael and Fianna Fáil TDs are rarely distinct. The explanatory power of the models, measured by R2, also dropped substantially when minor parties were excluded from the model—confirming that party identity does more to explain policy positions among minor party MPs. For example, the R2 fell to 0.36 for left–right, 0.05 for the EU, 0.21 for fiscal policy, 0.05 for environmental policy, 0.13 for Northern Ireland and 0.36 for abortion when the minor parties were excluded. Table 2 Regression of policy attitudes on party ID United Ireland Left–Right Abortion EU Fiscal policy Environment Party: Fine Gael 1.55 (0.35)** 0.82 (0.25)** −0.73 (0.36)* 0.36 (0.24) 0.73 (0.27)** 0.21 (0.26) Party: Labour 1.71 (0.42)** −1.84 (0.27)** −4.00 (0.54)** 0.35 (0.30) −1.68 (0.33)** −0.55 (0.35) Party: Sinn Fein −3.01 (0.28)** −2.38 (0.29)** −3.40 (0.68)** −3.28 (0.44)** −2.90 (0.28)** −1.67 (0.35)** Party: Green 0.39 (0.42)^ −1.01 (0.42)* −1.40 (0.82)^ −0.40 (0.68) −1.08 (0.48)* −3.37 (0.84)** Party: ULA 0.66 (0.95) −3.84 (0.47)** −6.43 (0.62)** −3.86 (0.71)** −3.95 (0.52)** −2.90 (0.64)** Party: Independent 0.34 (0.70) −1.15 (0.54)* −2.57 (0.91)** −2.30 (0.78)** −1.68 (0.33)** −0.82 (0.62) Year: 2011 −0.45 (0.28) 0.13 (0.18) 0.16 (0.30) −0.65 (0.21)** −0.48 (0.26)^ 0.10 (0.23) Year: 2016 −0.58 (0.30)* 0.13 (0.18) −0.81 (0.35)* −0.76 (0.23)** −0.42 (0.23)^ −0.28 (0.23) Intercept 3.81 (0.28)** 4.63 (0.15)** 8.12 (0.27)** 8.82 (0.17)** 5.38 (0.22)** 5.28 (0.19)** R2 0.32 0.48 0.36 0.45 0.43 0.23 United Ireland Left–Right Abortion EU Fiscal policy Environment Party: Fine Gael 1.55 (0.35)** 0.82 (0.25)** −0.73 (0.36)* 0.36 (0.24) 0.73 (0.27)** 0.21 (0.26) Party: Labour 1.71 (0.42)** −1.84 (0.27)** −4.00 (0.54)** 0.35 (0.30) −1.68 (0.33)** −0.55 (0.35) Party: Sinn Fein −3.01 (0.28)** −2.38 (0.29)** −3.40 (0.68)** −3.28 (0.44)** −2.90 (0.28)** −1.67 (0.35)** Party: Green 0.39 (0.42)^ −1.01 (0.42)* −1.40 (0.82)^ −0.40 (0.68) −1.08 (0.48)* −3.37 (0.84)** Party: ULA 0.66 (0.95) −3.84 (0.47)** −6.43 (0.62)** −3.86 (0.71)** −3.95 (0.52)** −2.90 (0.64)** Party: Independent 0.34 (0.70) −1.15 (0.54)* −2.57 (0.91)** −2.30 (0.78)** −1.68 (0.33)** −0.82 (0.62) Year: 2011 −0.45 (0.28) 0.13 (0.18) 0.16 (0.30) −0.65 (0.21)** −0.48 (0.26)^ 0.10 (0.23) Year: 2016 −0.58 (0.30)* 0.13 (0.18) −0.81 (0.35)* −0.76 (0.23)** −0.42 (0.23)^ −0.28 (0.23) Intercept 3.81 (0.28)** 4.63 (0.15)** 8.12 (0.27)** 8.82 (0.17)** 5.38 (0.22)** 5.28 (0.19)** R2 0.32 0.48 0.36 0.45 0.43 0.23 Base Categories: Fianna Fáil/2007 Dáil; Standard errors in parentheses. Significant at **p ≤ 0.01; *p ≤ 0.05; ^ = .10, * = .05, ** = .01. Table 2 Regression of policy attitudes on party ID United Ireland Left–Right Abortion EU Fiscal policy Environment Party: Fine Gael 1.55 (0.35)** 0.82 (0.25)** −0.73 (0.36)* 0.36 (0.24) 0.73 (0.27)** 0.21 (0.26) Party: Labour 1.71 (0.42)** −1.84 (0.27)** −4.00 (0.54)** 0.35 (0.30) −1.68 (0.33)** −0.55 (0.35) Party: Sinn Fein −3.01 (0.28)** −2.38 (0.29)** −3.40 (0.68)** −3.28 (0.44)** −2.90 (0.28)** −1.67 (0.35)** Party: Green 0.39 (0.42)^ −1.01 (0.42)* −1.40 (0.82)^ −0.40 (0.68) −1.08 (0.48)* −3.37 (0.84)** Party: ULA 0.66 (0.95) −3.84 (0.47)** −6.43 (0.62)** −3.86 (0.71)** −3.95 (0.52)** −2.90 (0.64)** Party: Independent 0.34 (0.70) −1.15 (0.54)* −2.57 (0.91)** −2.30 (0.78)** −1.68 (0.33)** −0.82 (0.62) Year: 2011 −0.45 (0.28) 0.13 (0.18) 0.16 (0.30) −0.65 (0.21)** −0.48 (0.26)^ 0.10 (0.23) Year: 2016 −0.58 (0.30)* 0.13 (0.18) −0.81 (0.35)* −0.76 (0.23)** −0.42 (0.23)^ −0.28 (0.23) Intercept 3.81 (0.28)** 4.63 (0.15)** 8.12 (0.27)** 8.82 (0.17)** 5.38 (0.22)** 5.28 (0.19)** R2 0.32 0.48 0.36 0.45 0.43 0.23 United Ireland Left–Right Abortion EU Fiscal policy Environment Party: Fine Gael 1.55 (0.35)** 0.82 (0.25)** −0.73 (0.36)* 0.36 (0.24) 0.73 (0.27)** 0.21 (0.26) Party: Labour 1.71 (0.42)** −1.84 (0.27)** −4.00 (0.54)** 0.35 (0.30) −1.68 (0.33)** −0.55 (0.35) Party: Sinn Fein −3.01 (0.28)** −2.38 (0.29)** −3.40 (0.68)** −3.28 (0.44)** −2.90 (0.28)** −1.67 (0.35)** Party: Green 0.39 (0.42)^ −1.01 (0.42)* −1.40 (0.82)^ −0.40 (0.68) −1.08 (0.48)* −3.37 (0.84)** Party: ULA 0.66 (0.95) −3.84 (0.47)** −6.43 (0.62)** −3.86 (0.71)** −3.95 (0.52)** −2.90 (0.64)** Party: Independent 0.34 (0.70) −1.15 (0.54)* −2.57 (0.91)** −2.30 (0.78)** −1.68 (0.33)** −0.82 (0.62) Year: 2011 −0.45 (0.28) 0.13 (0.18) 0.16 (0.30) −0.65 (0.21)** −0.48 (0.26)^ 0.10 (0.23) Year: 2016 −0.58 (0.30)* 0.13 (0.18) −0.81 (0.35)* −0.76 (0.23)** −0.42 (0.23)^ −0.28 (0.23) Intercept 3.81 (0.28)** 4.63 (0.15)** 8.12 (0.27)** 8.82 (0.17)** 5.38 (0.22)** 5.28 (0.19)** R2 0.32 0.48 0.36 0.45 0.43 0.23 Base Categories: Fianna Fáil/2007 Dáil; Standard errors in parentheses. Significant at **p ≤ 0.01; *p ≤ 0.05; ^ = .10, * = .05, ** = .01. 6.2 Demographic factors Historically, party loyalty has been passed down through families in Ireland based on the founding cleavage regarding the 1921 treaty and the establishment of the independent Irish state. Anti-treaty families support Fianna Fáil, while pro-treaty families support Fine Gael. However, family background does not appear to have an impact on policy positions. Supplementary Table S3 reflects the finding that family background is not a significant predictor of TD attitudes after controlling for the family’s party affiliation. In other words, family background—one of the key factors in sorting political elites into parties—is almost entirely non-ideological. Although TDs’ family background displays no discernible relation with the positions TD espouse, TDs’ personal demographic qualities reflect modest impacts on policy positions (Table 3). Once controlling for party affiliation and other variables, age is rarely a significant determinant of TD attitudes. However, gender and religiosity have somewhat stronger effects. In terms of gender and ideology, female TDs are markedly more inclined to abandon the demand for a united Ireland, and they are somewhat further to the left on environmental issues, which means they report believing that Ireland should protect the environment even if it damages economic growth as opposed to those that believe that Ireland should encourage economic growth even if this damages the environment. The strength of gender effects evaporates when party is included, which suggests that gender informs one’s party choice more so than policy positions (see Supplementary Tables S4 and S5). Table 3 Regression of policy attitudes on demographics United Ireland Left–Right Abortion EU Fiscal Policy Environment Age: 45–59 (years) −0.12 (0.30) −0.44 (0.22)^ −0.07 (0.37) 0.01 (0.26) −0.36 (0.25) 0.13 (0.25) Age: 60 +  (years) 0.20 (0.34) −0.41 (0.27) 0.32 (0.40) 0.07 (0.28) −0.54 (0.29)^ −0.15 (0.31) Religiosity: 1–5 scale 0.05 (0.11) −0.09 (0.07) −0.32 (0.13)* 0.03 (0.10) 0.00 (0.08) 0.02 (0.10) Gender: Female 1.68 (0.38)** −0.11 (0.28) −0.88 (0.48)^ 0.23 (0.30) −0.14 (0.26) −0.45 (0.26)^ Party: Fine Gael 1.37 (0.35)** 0.75 (0.25)** −0.68 (0.37)^ 0.29 (0.23) 0.66 (0.28)* 0.24 (0.28) Party: Labour 1.13 (0.39)** −1.70 (0.28)** −3.62 (0.54)** 0.16 (0.32) −1.57 (0.35)** −0.44 (0.36) Party: Sinn Fein −3.25 (0.32)** −2.31 (0.34)** −3.00 (0.71)** −3.31 (0.46)** −2.92 (0.30)** −2.02 (0.38)** Party: Green 0.34 (0.46) −0.63 (0.30)* −0.76 (1.05) −0.96 (0.61) −0.72 (0.44)^ −2.89 (1.06)** Party: ULA 0.39 (0.76) −3.96 (0.52)** −6.60 (0.59)** −3.90 (0.76)** −3.82 (0.62)** −3.18 (0.65)** Party: Independent −0.03 (0.69) −1.13 (0.55)* −2.56 (0.90)** −2.56 (0.80)** −1.94 (0.49)** −0.80 (0.63) Year: 2011 −0.40 (0.28) 0.10 (0.18) 0.28 (0.30) −0.64 (0.21)** −0.56 (0.27)* 0.12 (0.23) Year: 2016 −0.78 (0.29)** 0.14 (0.19) −0.70 (0.39)^ −0.69 (0.23)** −0.48 (0.25)^ −0.45 (0.25)^ Intercept 3.55 (0.49)** 5.24 (0.31)** 8.97 (0.51)** 8.75 (0.35)** 5.78 (0.38)** 5.25 (0.41)** R2 0.36 0.48 0.39 0.45 0.42 0.27 United Ireland Left–Right Abortion EU Fiscal Policy Environment Age: 45–59 (years) −0.12 (0.30) −0.44 (0.22)^ −0.07 (0.37) 0.01 (0.26) −0.36 (0.25) 0.13 (0.25) Age: 60 +  (years) 0.20 (0.34) −0.41 (0.27) 0.32 (0.40) 0.07 (0.28) −0.54 (0.29)^ −0.15 (0.31) Religiosity: 1–5 scale 0.05 (0.11) −0.09 (0.07) −0.32 (0.13)* 0.03 (0.10) 0.00 (0.08) 0.02 (0.10) Gender: Female 1.68 (0.38)** −0.11 (0.28) −0.88 (0.48)^ 0.23 (0.30) −0.14 (0.26) −0.45 (0.26)^ Party: Fine Gael 1.37 (0.35)** 0.75 (0.25)** −0.68 (0.37)^ 0.29 (0.23) 0.66 (0.28)* 0.24 (0.28) Party: Labour 1.13 (0.39)** −1.70 (0.28)** −3.62 (0.54)** 0.16 (0.32) −1.57 (0.35)** −0.44 (0.36) Party: Sinn Fein −3.25 (0.32)** −2.31 (0.34)** −3.00 (0.71)** −3.31 (0.46)** −2.92 (0.30)** −2.02 (0.38)** Party: Green 0.34 (0.46) −0.63 (0.30)* −0.76 (1.05) −0.96 (0.61) −0.72 (0.44)^ −2.89 (1.06)** Party: ULA 0.39 (0.76) −3.96 (0.52)** −6.60 (0.59)** −3.90 (0.76)** −3.82 (0.62)** −3.18 (0.65)** Party: Independent −0.03 (0.69) −1.13 (0.55)* −2.56 (0.90)** −2.56 (0.80)** −1.94 (0.49)** −0.80 (0.63) Year: 2011 −0.40 (0.28) 0.10 (0.18) 0.28 (0.30) −0.64 (0.21)** −0.56 (0.27)* 0.12 (0.23) Year: 2016 −0.78 (0.29)** 0.14 (0.19) −0.70 (0.39)^ −0.69 (0.23)** −0.48 (0.25)^ −0.45 (0.25)^ Intercept 3.55 (0.49)** 5.24 (0.31)** 8.97 (0.51)** 8.75 (0.35)** 5.78 (0.38)** 5.25 (0.41)** R2 0.36 0.48 0.39 0.45 0.42 0.27 Base Categories: Male/45 or younger/Fianna Fáil/2007 Dáil; Standard errors in parentheses. Significant at **p ≤ 0.01; *p ≤ 0.05; ^p ≤ 0.10. Table 3 Regression of policy attitudes on demographics United Ireland Left–Right Abortion EU Fiscal Policy Environment Age: 45–59 (years) −0.12 (0.30) −0.44 (0.22)^ −0.07 (0.37) 0.01 (0.26) −0.36 (0.25) 0.13 (0.25) Age: 60 +  (years) 0.20 (0.34) −0.41 (0.27) 0.32 (0.40) 0.07 (0.28) −0.54 (0.29)^ −0.15 (0.31) Religiosity: 1–5 scale 0.05 (0.11) −0.09 (0.07) −0.32 (0.13)* 0.03 (0.10) 0.00 (0.08) 0.02 (0.10) Gender: Female 1.68 (0.38)** −0.11 (0.28) −0.88 (0.48)^ 0.23 (0.30) −0.14 (0.26) −0.45 (0.26)^ Party: Fine Gael 1.37 (0.35)** 0.75 (0.25)** −0.68 (0.37)^ 0.29 (0.23) 0.66 (0.28)* 0.24 (0.28) Party: Labour 1.13 (0.39)** −1.70 (0.28)** −3.62 (0.54)** 0.16 (0.32) −1.57 (0.35)** −0.44 (0.36) Party: Sinn Fein −3.25 (0.32)** −2.31 (0.34)** −3.00 (0.71)** −3.31 (0.46)** −2.92 (0.30)** −2.02 (0.38)** Party: Green 0.34 (0.46) −0.63 (0.30)* −0.76 (1.05) −0.96 (0.61) −0.72 (0.44)^ −2.89 (1.06)** Party: ULA 0.39 (0.76) −3.96 (0.52)** −6.60 (0.59)** −3.90 (0.76)** −3.82 (0.62)** −3.18 (0.65)** Party: Independent −0.03 (0.69) −1.13 (0.55)* −2.56 (0.90)** −2.56 (0.80)** −1.94 (0.49)** −0.80 (0.63) Year: 2011 −0.40 (0.28) 0.10 (0.18) 0.28 (0.30) −0.64 (0.21)** −0.56 (0.27)* 0.12 (0.23) Year: 2016 −0.78 (0.29)** 0.14 (0.19) −0.70 (0.39)^ −0.69 (0.23)** −0.48 (0.25)^ −0.45 (0.25)^ Intercept 3.55 (0.49)** 5.24 (0.31)** 8.97 (0.51)** 8.75 (0.35)** 5.78 (0.38)** 5.25 (0.41)** R2 0.36 0.48 0.39 0.45 0.42 0.27 United Ireland Left–Right Abortion EU Fiscal Policy Environment Age: 45–59 (years) −0.12 (0.30) −0.44 (0.22)^ −0.07 (0.37) 0.01 (0.26) −0.36 (0.25) 0.13 (0.25) Age: 60 +  (years) 0.20 (0.34) −0.41 (0.27) 0.32 (0.40) 0.07 (0.28) −0.54 (0.29)^ −0.15 (0.31) Religiosity: 1–5 scale 0.05 (0.11) −0.09 (0.07) −0.32 (0.13)* 0.03 (0.10) 0.00 (0.08) 0.02 (0.10) Gender: Female 1.68 (0.38)** −0.11 (0.28) −0.88 (0.48)^ 0.23 (0.30) −0.14 (0.26) −0.45 (0.26)^ Party: Fine Gael 1.37 (0.35)** 0.75 (0.25)** −0.68 (0.37)^ 0.29 (0.23) 0.66 (0.28)* 0.24 (0.28) Party: Labour 1.13 (0.39)** −1.70 (0.28)** −3.62 (0.54)** 0.16 (0.32) −1.57 (0.35)** −0.44 (0.36) Party: Sinn Fein −3.25 (0.32)** −2.31 (0.34)** −3.00 (0.71)** −3.31 (0.46)** −2.92 (0.30)** −2.02 (0.38)** Party: Green 0.34 (0.46) −0.63 (0.30)* −0.76 (1.05) −0.96 (0.61) −0.72 (0.44)^ −2.89 (1.06)** Party: ULA 0.39 (0.76) −3.96 (0.52)** −6.60 (0.59)** −3.90 (0.76)** −3.82 (0.62)** −3.18 (0.65)** Party: Independent −0.03 (0.69) −1.13 (0.55)* −2.56 (0.90)** −2.56 (0.80)** −1.94 (0.49)** −0.80 (0.63) Year: 2011 −0.40 (0.28) 0.10 (0.18) 0.28 (0.30) −0.64 (0.21)** −0.56 (0.27)* 0.12 (0.23) Year: 2016 −0.78 (0.29)** 0.14 (0.19) −0.70 (0.39)^ −0.69 (0.23)** −0.48 (0.25)^ −0.45 (0.25)^ Intercept 3.55 (0.49)** 5.24 (0.31)** 8.97 (0.51)** 8.75 (0.35)** 5.78 (0.38)** 5.25 (0.41)** R2 0.36 0.48 0.39 0.45 0.42 0.27 Base Categories: Male/45 or younger/Fianna Fáil/2007 Dáil; Standard errors in parentheses. Significant at **p ≤ 0.01; *p ≤ 0.05; ^p ≤ 0.10. To measure religiosity, a simple index was constructed that was comprised of the average of five religiosity variables from the TD survey: how frequently TDs reported praying, how frequently they reported attending church services, how religious or secular they considered themselves, how religiously traditional or liberal they believed they were and how important religion was in their home while growing up (on all these variables, 1 indicates the least religious and 5 is the most religious). As Table 3 (above) and Figure 2 show, the TDs with the highest levels of religiosity were on average one point further to the right and four points more pro-life than the most secular TDs. Perhaps not surprisingly, religiosity appears to have less influence in determining positions on other issues. Figure 2 View largeDownload slide Marginal effects of religiosity on policy attitudes (2007, 2011, 2016). Figure 2 View largeDownload slide Marginal effects of religiosity on policy attitudes (2007, 2011, 2016). Borderline evidence also suggested that those with higher levels of religiosity held slightly more supportive attitudes towards environmental protection. Thus, among the demographic variables analysed, religiosity and gender were the background variables that seem to shape TD ideological positions the most. Overall, TDs views’ do seem affected by their idiosyncratic personal experiences, which provides initial evidence for the hypothesis that TDs adopt some non-strategic policy stances at odds with the national party. 6.3 Local context Given the limited degree to which demographic factors explain MPs’ policy positions, I turn to analysing the importance of local constituency factors. As seen earlier, demographics appear to influence mostly non-economic policy areas, and even there, a great deal of variance in TDs’ attitudes is left unexplained. In addition to demographics, structural factors appear to play a role in explaining intra-party ideological variation. Given the high level of personalised engagement within the constituency, TDs unsurprisingly respond to some important characteristics of their local constituencies, which suggests that individual Irish TDs are responsive to local needs and sentiments. Given that levels of party membership are low among Irish voters and there exists an increasingly high percentage of ‘unattached’ voters in Ireland, candidates and parties alike must make a more concerted effort to be responsive to local interests (Marsh et al., 2008, pp. 59–79; Marsh, 2008). Evidence from the surveys also reveals a consequential divide between urban, rural and commuter constituencies. Table 4 shows that differences between urban, rural and commuter constituency TDs are largest, and become borderline statistically significant, on ‘post-materialist’ issues that do not easily fall into traditional left–right or pro-treaty/anti-treaty divisions. Table 4 Regression of policy attitudes on constituency type United Ireland Left–Right Abortion EU Fiscal Policy Environment Constituency: Commuter −0.47 (0.31) −0.22 (0.26) 0.65 (0.44) 0.24 (0.28) 0.10 (0.31) −0.23 (0.28) Constituency:Rural −0.02 (0.30) −0.09 (0.20) 0.79 (0.36)* 0.03 (0.25) 0.08 (0.23) 0.47 (0.23)* Party: Fine Gael 1.56 (0.34)** 0.81 (0.25)** −0.64 (0.35)^ 0.36 (0.24) 0.74 (0.27)** 0.28 (0.25) Party: Labour 1.66 (0.42)** −1.88 (0.28)** −3.70 (0.55)** 0.38 (0.31) −1.64 (0.34)** −0.42 (0.34) Party: Sinn Fein −3.05 (0.27)** −2.41 (0.30)** −3.22 (0.65)** −3.25 (0.45)** −2.88 (0.29)** −1.61 (0.34)** Party: Green 0.26 (0.44) −1.08 (0.43)* −1.06 (0.72) −0.33 (0.69) −1.03 (0.49)* −3.32 (0.78)** Party: ULA 0.56 (0.94) −3.92 (0.51)** −5.92 (0.62)** −3.79 (0.73)** −3.89 (0.55)** −2.72 (0.64)** Party: Independent 0.28 (0.71) −1.18 (0.53)* −2.44 (0.93)** −2.26 (0.78)** −1.90 (0.49)** −0.82 (0.61) Year: 2011 −0.43 (0.28) 0.15 (0.18) 0.03 (0.30) −0.66 (0.21)** −0.50 (0.26)^ 0.03 (0.23) Year: 2016 −0.56 (0.27)* 0.14 (0.18) −0.88 (0.34)* −0.77 (0.22)** −0.43 (0.23)^ −0.30 (0.22) Intercept 3.93 (0.34)** 4.72 (0.20)** 7.65 (0.34)** 8.75 (0.24)** 5.33 (0.27)** 5.15 (0.23)** R2 0.33 0.48 0.37 0.45 0.43 0.25 United Ireland Left–Right Abortion EU Fiscal Policy Environment Constituency: Commuter −0.47 (0.31) −0.22 (0.26) 0.65 (0.44) 0.24 (0.28) 0.10 (0.31) −0.23 (0.28) Constituency:Rural −0.02 (0.30) −0.09 (0.20) 0.79 (0.36)* 0.03 (0.25) 0.08 (0.23) 0.47 (0.23)* Party: Fine Gael 1.56 (0.34)** 0.81 (0.25)** −0.64 (0.35)^ 0.36 (0.24) 0.74 (0.27)** 0.28 (0.25) Party: Labour 1.66 (0.42)** −1.88 (0.28)** −3.70 (0.55)** 0.38 (0.31) −1.64 (0.34)** −0.42 (0.34) Party: Sinn Fein −3.05 (0.27)** −2.41 (0.30)** −3.22 (0.65)** −3.25 (0.45)** −2.88 (0.29)** −1.61 (0.34)** Party: Green 0.26 (0.44) −1.08 (0.43)* −1.06 (0.72) −0.33 (0.69) −1.03 (0.49)* −3.32 (0.78)** Party: ULA 0.56 (0.94) −3.92 (0.51)** −5.92 (0.62)** −3.79 (0.73)** −3.89 (0.55)** −2.72 (0.64)** Party: Independent 0.28 (0.71) −1.18 (0.53)* −2.44 (0.93)** −2.26 (0.78)** −1.90 (0.49)** −0.82 (0.61) Year: 2011 −0.43 (0.28) 0.15 (0.18) 0.03 (0.30) −0.66 (0.21)** −0.50 (0.26)^ 0.03 (0.23) Year: 2016 −0.56 (0.27)* 0.14 (0.18) −0.88 (0.34)* −0.77 (0.22)** −0.43 (0.23)^ −0.30 (0.22) Intercept 3.93 (0.34)** 4.72 (0.20)** 7.65 (0.34)** 8.75 (0.24)** 5.33 (0.27)** 5.15 (0.23)** R2 0.33 0.48 0.37 0.45 0.43 0.25 Base Categories: Urban constituency/Fianna Fáil/2007 Dáil; Standard errors in parentheses. Significant at **p ≤ 0.01; *p ≤ 0.05; ^p ≤ 0.10. Table 4 Regression of policy attitudes on constituency type United Ireland Left–Right Abortion EU Fiscal Policy Environment Constituency: Commuter −0.47 (0.31) −0.22 (0.26) 0.65 (0.44) 0.24 (0.28) 0.10 (0.31) −0.23 (0.28) Constituency:Rural −0.02 (0.30) −0.09 (0.20) 0.79 (0.36)* 0.03 (0.25) 0.08 (0.23) 0.47 (0.23)* Party: Fine Gael 1.56 (0.34)** 0.81 (0.25)** −0.64 (0.35)^ 0.36 (0.24) 0.74 (0.27)** 0.28 (0.25) Party: Labour 1.66 (0.42)** −1.88 (0.28)** −3.70 (0.55)** 0.38 (0.31) −1.64 (0.34)** −0.42 (0.34) Party: Sinn Fein −3.05 (0.27)** −2.41 (0.30)** −3.22 (0.65)** −3.25 (0.45)** −2.88 (0.29)** −1.61 (0.34)** Party: Green 0.26 (0.44) −1.08 (0.43)* −1.06 (0.72) −0.33 (0.69) −1.03 (0.49)* −3.32 (0.78)** Party: ULA 0.56 (0.94) −3.92 (0.51)** −5.92 (0.62)** −3.79 (0.73)** −3.89 (0.55)** −2.72 (0.64)** Party: Independent 0.28 (0.71) −1.18 (0.53)* −2.44 (0.93)** −2.26 (0.78)** −1.90 (0.49)** −0.82 (0.61) Year: 2011 −0.43 (0.28) 0.15 (0.18) 0.03 (0.30) −0.66 (0.21)** −0.50 (0.26)^ 0.03 (0.23) Year: 2016 −0.56 (0.27)* 0.14 (0.18) −0.88 (0.34)* −0.77 (0.22)** −0.43 (0.23)^ −0.30 (0.22) Intercept 3.93 (0.34)** 4.72 (0.20)** 7.65 (0.34)** 8.75 (0.24)** 5.33 (0.27)** 5.15 (0.23)** R2 0.33 0.48 0.37 0.45 0.43 0.25 United Ireland Left–Right Abortion EU Fiscal Policy Environment Constituency: Commuter −0.47 (0.31) −0.22 (0.26) 0.65 (0.44) 0.24 (0.28) 0.10 (0.31) −0.23 (0.28) Constituency:Rural −0.02 (0.30) −0.09 (0.20) 0.79 (0.36)* 0.03 (0.25) 0.08 (0.23) 0.47 (0.23)* Party: Fine Gael 1.56 (0.34)** 0.81 (0.25)** −0.64 (0.35)^ 0.36 (0.24) 0.74 (0.27)** 0.28 (0.25) Party: Labour 1.66 (0.42)** −1.88 (0.28)** −3.70 (0.55)** 0.38 (0.31) −1.64 (0.34)** −0.42 (0.34) Party: Sinn Fein −3.05 (0.27)** −2.41 (0.30)** −3.22 (0.65)** −3.25 (0.45)** −2.88 (0.29)** −1.61 (0.34)** Party: Green 0.26 (0.44) −1.08 (0.43)* −1.06 (0.72) −0.33 (0.69) −1.03 (0.49)* −3.32 (0.78)** Party: ULA 0.56 (0.94) −3.92 (0.51)** −5.92 (0.62)** −3.79 (0.73)** −3.89 (0.55)** −2.72 (0.64)** Party: Independent 0.28 (0.71) −1.18 (0.53)* −2.44 (0.93)** −2.26 (0.78)** −1.90 (0.49)** −0.82 (0.61) Year: 2011 −0.43 (0.28) 0.15 (0.18) 0.03 (0.30) −0.66 (0.21)** −0.50 (0.26)^ 0.03 (0.23) Year: 2016 −0.56 (0.27)* 0.14 (0.18) −0.88 (0.34)* −0.77 (0.22)** −0.43 (0.23)^ −0.30 (0.22) Intercept 3.93 (0.34)** 4.72 (0.20)** 7.65 (0.34)** 8.75 (0.24)** 5.33 (0.27)** 5.15 (0.23)** R2 0.33 0.48 0.37 0.45 0.43 0.25 Base Categories: Urban constituency/Fianna Fáil/2007 Dáil; Standard errors in parentheses. Significant at **p ≤ 0.01; *p ≤ 0.05; ^p ≤ 0.10. This geographic variation is indicative of TDs exercising personal discretion to adapt their ideological appeals based on their specific electoral context (Farrell, 2001, pp. 121–152). Citizens living in urban, rural and commuter constituencies are likely to hold different positions given their differing life contexts. Although constituency level opinion polls are not available in Ireland, elements of certain constituency attitudes can be gleaned using results from the 2002 right-to-life referendum and 2009 Lisbon Treaty (EU) referendum. Not surprisingly, legislators in constituencies where referendum results were more anti-abortion also tend to take stronger personal stances against abortion. A similar effect may exist regarding the EU, but it narrowly missed attaining statistical significance. Table 5 shows a regression analysis of the relationship between constituency public opinion and TDs’ self-placement on abortion. Table 5 Regression of TD attitudes on constituency referenda results Abortion Reg.abort.hlm02 EU reg.eu.hlmlisbon2 Vote share in 2002 right to life referendum 0.06 (0.016)** – Vote share in 2009 Lisbon Treat (EU) referendum – 0.023 (0.019) Party: Fine Gael −0.73 (0.40) 0.23 (0.26) Party: Labour −3.79 (0.53)** 0.17 (0.34) Party: Sinn Fein −3.43 (0.59)** −3.17 (0.40)** Party: Green −0.61 (1.17) −0.54 (0.71) Party: ULA −5.76 (0.79)** −3.64 (0.51)** Party: Independent −3.01 (0.66)** −2.07 (0.44)** Year: 2011 0.28 (0.30) −0.55 (0.19)** Year: 2016 −0.76 (0.32)* −0.67 (0.20)** Intercept 5.20 (0.92)** 7.24 (1.25)** Conditional R2 (Nakagawa & Schielzeth) 0.71 0.77 Abortion Reg.abort.hlm02 EU reg.eu.hlmlisbon2 Vote share in 2002 right to life referendum 0.06 (0.016)** – Vote share in 2009 Lisbon Treat (EU) referendum – 0.023 (0.019) Party: Fine Gael −0.73 (0.40) 0.23 (0.26) Party: Labour −3.79 (0.53)** 0.17 (0.34) Party: Sinn Fein −3.43 (0.59)** −3.17 (0.40)** Party: Green −0.61 (1.17) −0.54 (0.71) Party: ULA −5.76 (0.79)** −3.64 (0.51)** Party: Independent −3.01 (0.66)** −2.07 (0.44)** Year: 2011 0.28 (0.30) −0.55 (0.19)** Year: 2016 −0.76 (0.32)* −0.67 (0.20)** Intercept 5.20 (0.92)** 7.24 (1.25)** Conditional R2 (Nakagawa & Schielzeth) 0.71 0.77 Base Category: Fianna Fáil/2007 Dáil; Standard errors in parentheses. Significant at **p ≤ 0.01; *p ≤ 0.05; ^p ≤ 0.10;  ^ = .10, * = .05, ** = .01. Table 5 Regression of TD attitudes on constituency referenda results Abortion Reg.abort.hlm02 EU reg.eu.hlmlisbon2 Vote share in 2002 right to life referendum 0.06 (0.016)** – Vote share in 2009 Lisbon Treat (EU) referendum – 0.023 (0.019) Party: Fine Gael −0.73 (0.40) 0.23 (0.26) Party: Labour −3.79 (0.53)** 0.17 (0.34) Party: Sinn Fein −3.43 (0.59)** −3.17 (0.40)** Party: Green −0.61 (1.17) −0.54 (0.71) Party: ULA −5.76 (0.79)** −3.64 (0.51)** Party: Independent −3.01 (0.66)** −2.07 (0.44)** Year: 2011 0.28 (0.30) −0.55 (0.19)** Year: 2016 −0.76 (0.32)* −0.67 (0.20)** Intercept 5.20 (0.92)** 7.24 (1.25)** Conditional R2 (Nakagawa & Schielzeth) 0.71 0.77 Abortion Reg.abort.hlm02 EU reg.eu.hlmlisbon2 Vote share in 2002 right to life referendum 0.06 (0.016)** – Vote share in 2009 Lisbon Treat (EU) referendum – 0.023 (0.019) Party: Fine Gael −0.73 (0.40) 0.23 (0.26) Party: Labour −3.79 (0.53)** 0.17 (0.34) Party: Sinn Fein −3.43 (0.59)** −3.17 (0.40)** Party: Green −0.61 (1.17) −0.54 (0.71) Party: ULA −5.76 (0.79)** −3.64 (0.51)** Party: Independent −3.01 (0.66)** −2.07 (0.44)** Year: 2011 0.28 (0.30) −0.55 (0.19)** Year: 2016 −0.76 (0.32)* −0.67 (0.20)** Intercept 5.20 (0.92)** 7.24 (1.25)** Conditional R2 (Nakagawa & Schielzeth) 0.71 0.77 Base Category: Fianna Fáil/2007 Dáil; Standard errors in parentheses. Significant at **p ≤ 0.01; *p ≤ 0.05; ^p ≤ 0.10;  ^ = .10, * = .05, ** = .01. Supplementary Figure S1 includes visual plots which estimate the relationship between referenda voting and average TD opinion in a constituency in the 2002 right-to-life referendum and the 2009 Lisbon Treaty that illustrate these same results. The degree of political competition that TDs face also appears to have some relationship with the local policy position adopted by individual TDs. For example, TDs who are more electorally secure are moderately less pro-Europe and tend to follow the party line more closely on Northern Ireland and the EU. TDs’ electoral marginality was measured by looking at what fraction of a quota (the number of votes divided by number of seats) each received on the first vote count. This measure, although limited in many ways, was selected because it is used by political practitioners and pollsters themselves and is, therefore, likely to reflect TDs’ own understanding of how electorally marginal they are. TDs who receive nearly a full quota are quite likely to be elected, while those who receive less than half a quota are rarely elected. There are geographic and party differences with this measure because rural candidates generally must earn a higher quota to be elected than those in urban areas and candidates with running mates tend to earn lower quotas than those who do not have a running mate. Nevertheless, this measure provides a proxy to test if there is any difference in ideological positions based on whether one is elected quite handily compared to those that are the last to be elected. Although results show that candidates who follow the party line, or deviate towards the centre, are more successful electorally, that evidence is limited. As Supplementary Table S6 shows, no evidence was found that electorally marginal candidates were further left or right. More importantly, the difference between a TD’s position, and his or her placement of their own party, produced significant bivariate relationships on European integration and Northern Ireland. This can be seen in analysis of variance (ANOVA) tests (Table 6) and bivariate regressions (Supplementary Table S7). Table 6 Mean quotas received on first count (2007, 2011, 2016; by TDs’ distance from party) N. Ireland* Left–Right Abortion EU* Fiscal Environment 0-point difference 0.76 (0.01) 0.78 (0.02) 0.80 (0.02) 0.76 (0.02) 0.77 (0.02) 0.76 (0.02) 1-point difference 0.79 (0.04) 0.74 (0.02) 0.75 (0.03) 0.80 (0.03) 0.75 (0.02) 0.77 (0.02) 2-point difference 0.83 (0.04) 0.76 (0.03) 0.76 (0.03) 0.81 (0.05) 0.73 (0.03) 0.80 (0.04) >2-point difference 0.68 (0.02) 0.74 (0.04) 0.72 (0.02) 0.68 (0.03) 0.77 (0.04) 0.72 (0.03) ANOVA p-value 0.01 0.56 0.12 0.04 0.60 0.45 N. Ireland* Left–Right Abortion EU* Fiscal Environment 0-point difference 0.76 (0.01) 0.78 (0.02) 0.80 (0.02) 0.76 (0.02) 0.77 (0.02) 0.76 (0.02) 1-point difference 0.79 (0.04) 0.74 (0.02) 0.75 (0.03) 0.80 (0.03) 0.75 (0.02) 0.77 (0.02) 2-point difference 0.83 (0.04) 0.76 (0.03) 0.76 (0.03) 0.81 (0.05) 0.73 (0.03) 0.80 (0.04) >2-point difference 0.68 (0.02) 0.74 (0.04) 0.72 (0.02) 0.68 (0.03) 0.77 (0.04) 0.72 (0.03) ANOVA p-value 0.01 0.56 0.12 0.04 0.60 0.45 Standard errors in parentheses. Significant at **p ≤ 0.05; *p ≤ 0.10;  ^ = .10, * = .05, ** = .01. Table 6 Mean quotas received on first count (2007, 2011, 2016; by TDs’ distance from party) N. Ireland* Left–Right Abortion EU* Fiscal Environment 0-point difference 0.76 (0.01) 0.78 (0.02) 0.80 (0.02) 0.76 (0.02) 0.77 (0.02) 0.76 (0.02) 1-point difference 0.79 (0.04) 0.74 (0.02) 0.75 (0.03) 0.80 (0.03) 0.75 (0.02) 0.77 (0.02) 2-point difference 0.83 (0.04) 0.76 (0.03) 0.76 (0.03) 0.81 (0.05) 0.73 (0.03) 0.80 (0.04) >2-point difference 0.68 (0.02) 0.74 (0.04) 0.72 (0.02) 0.68 (0.03) 0.77 (0.04) 0.72 (0.03) ANOVA p-value 0.01 0.56 0.12 0.04 0.60 0.45 N. Ireland* Left–Right Abortion EU* Fiscal Environment 0-point difference 0.76 (0.01) 0.78 (0.02) 0.80 (0.02) 0.76 (0.02) 0.77 (0.02) 0.76 (0.02) 1-point difference 0.79 (0.04) 0.74 (0.02) 0.75 (0.03) 0.80 (0.03) 0.75 (0.02) 0.77 (0.02) 2-point difference 0.83 (0.04) 0.76 (0.03) 0.76 (0.03) 0.81 (0.05) 0.73 (0.03) 0.80 (0.04) >2-point difference 0.68 (0.02) 0.74 (0.04) 0.72 (0.02) 0.68 (0.03) 0.77 (0.04) 0.72 (0.03) ANOVA p-value 0.01 0.56 0.12 0.04 0.60 0.45 Standard errors in parentheses. Significant at **p ≤ 0.05; *p ≤ 0.10;  ^ = .10, * = .05, ** = .01. For example, TDs who shared a position with their party on Northern Ireland won 76% of the first preference votes necessary to win a seat, while TDs who diverged from their party by two or more points on this issue won 68% of the first-preference votes needed to win seat. However, once party and year were controlled for in an ordered logistic regression model, the relationship lost significance (see Supplementary Table S9). An alternative specification for the TDs’ positioning measured whether TDs were more centrist or more extreme than their parties. Bivariate logistic regression findings (Supplementary Table S8) suggest that TDs who adopted more extreme positions than their parties on abortion and possibly the EU paid an electoral price, while those who adopted more centrist positions than their parties were successful. (However, bivariate ANOVA tests (Supplementary Table S10) and multivariate regressions (Supplementary Table S11) failed to adequately confirm these findings, though (Supplementary Table S10) provides somewhat confirmatory evidence for EU policy position.) Regardless of whether they were electorally marginal candidates, those TDs who faced intra-party competitors in their constituency also took more critical stances towards the EU, breaking with their generally more pro-European parties as Table 7 shows. Table 7 Regression of policy attitudes on intra-party competition United Ireland Left–Right Abortion EU Fiscal Policy Environment Intra-party competitor in district −0.01 (0.25) 0.04 (0.18) −0.38 (0.367) −0.34 (0.23) 0.01 (0.24) −0.20 (0.24) Party: Fine Gael 1.55 (0.35)** 0.82 (0.25)** −0.72 (0.36)* 0.37 (0.24) 0.73 (0.27)** 0.21 (0.26) Party: Labour 1.71 (0.43)** −1.85 (0.28)** −3.87 (0.53)** 0.46 (0.30) −1.68 (0.32)** −0.48 (0.35) Party: Sinn Fein −3.01 (0.29)** −2.40 (0.31)** −3.18 (0.69)* −3.08 (0.44)** −2.89 (0.28)** −1.56 (0.40)** Party: Green 0.39 (0.45) −1.03 (0.45)* −1.08 (0.87) −0.17 (0.74) −1.07 (0.51)* −3.20 (0.86)** Party: ULA 0.66 (1.01) −3.87 (0.49)** −6.15 (0.68)** −3.61 (0.73)** −3.94 (0.54)** −2.76 (0.68)** Party: Independent 0.35 (0.70) −1.18 (0.56)* −2.28 (0.95)* −2.04 (0.79)* −1.91 (0.48)** −0.66 (0.66) Year: 2011 −0.45 (0.28) 0.13 (0.18) 0.16 (0.30) −0.65 (0.21)** −0.48 (0.26)^ 0.10 (0.23) Year: 2016 −0.58 (0.28)* 0.13 (0.18) −0.78 (0.35)* −0.73 (0.23)** −0.42 (0.24)^ −0.26 (0.22) Intercept 3.81 (0.28)** 4.62 (0.15)** 8.17 (0.28)** 8.87 (0.18)** 5.38 (0.22)** 5.30 (0.19)** R2 0.32 0.48 0.36 0.46 0.43 0.23 United Ireland Left–Right Abortion EU Fiscal Policy Environment Intra-party competitor in district −0.01 (0.25) 0.04 (0.18) −0.38 (0.367) −0.34 (0.23) 0.01 (0.24) −0.20 (0.24) Party: Fine Gael 1.55 (0.35)** 0.82 (0.25)** −0.72 (0.36)* 0.37 (0.24) 0.73 (0.27)** 0.21 (0.26) Party: Labour 1.71 (0.43)** −1.85 (0.28)** −3.87 (0.53)** 0.46 (0.30) −1.68 (0.32)** −0.48 (0.35) Party: Sinn Fein −3.01 (0.29)** −2.40 (0.31)** −3.18 (0.69)* −3.08 (0.44)** −2.89 (0.28)** −1.56 (0.40)** Party: Green 0.39 (0.45) −1.03 (0.45)* −1.08 (0.87) −0.17 (0.74) −1.07 (0.51)* −3.20 (0.86)** Party: ULA 0.66 (1.01) −3.87 (0.49)** −6.15 (0.68)** −3.61 (0.73)** −3.94 (0.54)** −2.76 (0.68)** Party: Independent 0.35 (0.70) −1.18 (0.56)* −2.28 (0.95)* −2.04 (0.79)* −1.91 (0.48)** −0.66 (0.66) Year: 2011 −0.45 (0.28) 0.13 (0.18) 0.16 (0.30) −0.65 (0.21)** −0.48 (0.26)^ 0.10 (0.23) Year: 2016 −0.58 (0.28)* 0.13 (0.18) −0.78 (0.35)* −0.73 (0.23)** −0.42 (0.24)^ −0.26 (0.22) Intercept 3.81 (0.28)** 4.62 (0.15)** 8.17 (0.28)** 8.87 (0.18)** 5.38 (0.22)** 5.30 (0.19)** R2 0.32 0.48 0.36 0.46 0.43 0.23 Base Categories: Urban constituency/Fianna Fáil/2007 Dáil; Standard errors in parentheses. Significant at **p ≤ 0.01; *p ≤ 0.05; ^p ≤ 0.10. Table 7 Regression of policy attitudes on intra-party competition United Ireland Left–Right Abortion EU Fiscal Policy Environment Intra-party competitor in district −0.01 (0.25) 0.04 (0.18) −0.38 (0.367) −0.34 (0.23) 0.01 (0.24) −0.20 (0.24) Party: Fine Gael 1.55 (0.35)** 0.82 (0.25)** −0.72 (0.36)* 0.37 (0.24) 0.73 (0.27)** 0.21 (0.26) Party: Labour 1.71 (0.43)** −1.85 (0.28)** −3.87 (0.53)** 0.46 (0.30) −1.68 (0.32)** −0.48 (0.35) Party: Sinn Fein −3.01 (0.29)** −2.40 (0.31)** −3.18 (0.69)* −3.08 (0.44)** −2.89 (0.28)** −1.56 (0.40)** Party: Green 0.39 (0.45) −1.03 (0.45)* −1.08 (0.87) −0.17 (0.74) −1.07 (0.51)* −3.20 (0.86)** Party: ULA 0.66 (1.01) −3.87 (0.49)** −6.15 (0.68)** −3.61 (0.73)** −3.94 (0.54)** −2.76 (0.68)** Party: Independent 0.35 (0.70) −1.18 (0.56)* −2.28 (0.95)* −2.04 (0.79)* −1.91 (0.48)** −0.66 (0.66) Year: 2011 −0.45 (0.28) 0.13 (0.18) 0.16 (0.30) −0.65 (0.21)** −0.48 (0.26)^ 0.10 (0.23) Year: 2016 −0.58 (0.28)* 0.13 (0.18) −0.78 (0.35)* −0.73 (0.23)** −0.42 (0.24)^ −0.26 (0.22) Intercept 3.81 (0.28)** 4.62 (0.15)** 8.17 (0.28)** 8.87 (0.18)** 5.38 (0.22)** 5.30 (0.19)** R2 0.32 0.48 0.36 0.46 0.43 0.23 United Ireland Left–Right Abortion EU Fiscal Policy Environment Intra-party competitor in district −0.01 (0.25) 0.04 (0.18) −0.38 (0.367) −0.34 (0.23) 0.01 (0.24) −0.20 (0.24) Party: Fine Gael 1.55 (0.35)** 0.82 (0.25)** −0.72 (0.36)* 0.37 (0.24) 0.73 (0.27)** 0.21 (0.26) Party: Labour 1.71 (0.43)** −1.85 (0.28)** −3.87 (0.53)** 0.46 (0.30) −1.68 (0.32)** −0.48 (0.35) Party: Sinn Fein −3.01 (0.29)** −2.40 (0.31)** −3.18 (0.69)* −3.08 (0.44)** −2.89 (0.28)** −1.56 (0.40)** Party: Green 0.39 (0.45) −1.03 (0.45)* −1.08 (0.87) −0.17 (0.74) −1.07 (0.51)* −3.20 (0.86)** Party: ULA 0.66 (1.01) −3.87 (0.49)** −6.15 (0.68)** −3.61 (0.73)** −3.94 (0.54)** −2.76 (0.68)** Party: Independent 0.35 (0.70) −1.18 (0.56)* −2.28 (0.95)* −2.04 (0.79)* −1.91 (0.48)** −0.66 (0.66) Year: 2011 −0.45 (0.28) 0.13 (0.18) 0.16 (0.30) −0.65 (0.21)** −0.48 (0.26)^ 0.10 (0.23) Year: 2016 −0.58 (0.28)* 0.13 (0.18) −0.78 (0.35)* −0.73 (0.23)** −0.42 (0.24)^ −0.26 (0.22) Intercept 3.81 (0.28)** 4.62 (0.15)** 8.17 (0.28)** 8.87 (0.18)** 5.38 (0.22)** 5.30 (0.19)** R2 0.32 0.48 0.36 0.46 0.43 0.23 Base Categories: Urban constituency/Fianna Fáil/2007 Dáil; Standard errors in parentheses. Significant at **p ≤ 0.01; *p ≤ 0.05; ^p ≤ 0.10. It is unclear whether TDs feel they need to offer more extreme positions because they are marginal candidates, or whether TDs who diverge from their parties tend to suffer electorally. Ultimately, TDs do appear to adopt locally varied positions for strategic reasons albeit the evidence is not overwhelming. 7. Conclusion This study seeks to add to our understanding of the behaviour of franchise parties by focusing on adaptations within the ideological, rather than the organisational, domain. A key attribute of franchise, or franchise-like, parties is their ability to reconcile competing electoral imperatives at both the national and local levels. The degree to which a party conforms to a franchise-like model, the party must create a strong national organisation that can manage and control party strategies and overall brand identity. This becomes especially imperative in a highly volatile electoral environment. However, they must simultaneously find ways to be sufficiently nimble to secure support (from voters and high-quality candidates) within diverse local contexts. While a number of studies have focused on the organisational mechanisms parties employ to function as a franchise party, this study has sought to expand the utility of the concept of franchise parties by offering an analysis of how Irish political entrepreneurs from an array of parties vary their ideological appeals, either to enhance their attractiveness or pursue personal beliefs and agendas once in office. I have argued that, in the Irish case, although a wide variety of parties evince these behaviours, the larger parties exhibit measurably more franchise-like behaviour than the non-major parties. To provide evidence for the analysis, the article employs original quantitative and qualitative data from three longitudinal parliamentary surveys designed to examine how candidates compete ideologically. It finds that candidates from Ireland’s larger parties in particular take advantage of the flexibility of the franchise model to adapt their programmatic appeals, which in turn enhances the electoral chances of the party as a whole. Specifically, I examined how party affiliation, demographic and family background, and local context shape MPs’ positions on policy issues. Not surprisingly, party affiliation explains individual politicians’ ideological positions more so than these other social and environmental factors. Demographic factors such as age and religiosity, or whether a TD lives in an urban or rural constituency, shape party affiliation, but the strength of these factors lessens when party is included in our statistical tests. Nevertheless, the findings confirm that even though party does influence policy positions, individual Irish politicians enjoy considerable flexibility and autonomy to adapt policy positions at variance from the national platform during elections. As the franchise model implies, this strategy is left unchallenged by party leaders at the national level because the party maintains overall control of policy and strategy at the national level, but party leaders also recognise that individual candidates are best suited to know the set of policy offerings that will result in maximising local support. Supplementary Data Supplementary data are available at Parliamentary Affairs online. Footnotes 1 ‘What to Ask Your Candidate about Health and Medical Services’, Irish Times (1973, February 19). Acknowledgements I would like to thank several colleagues for their generous and thoughtful feedback on this paper, including Ken Carty, Eoin O’Malley, Jaimie Bleck, Karrie Koesel, Mike Hoffman, Lauren Honig, and especially Tim Scully. Ben Mainwaring and Steve Ponisciak provided invaluable assistance with the quantitative analysis. I would also like to thank the three reviewers for their thoughtful and thorough comments which helped make this a stronger article. Any errors are my sole responsibility. The author reports no other declarations of interest. Funding This work was supported by internal grants at the University of Notre Dame by the Kellogg Institute for International Studies and the Nanovic Institute for European Studies. Conflict of interest No conflict of interest to disclose. References Byrne K. , O'Malley E. ( 2012 ) ‘ Politics with Hidden Bases: Unearthing Party System's Deep Roots’, British Journal of Politics and International Relations , 14 , 613 – 629 . Google Scholar CrossRef Search ADS Carey J. M. , Shugart M. S. ( 1995 ) ‘Incentives to Cultivate a Personal Vote: A Rank Ordering of Electoral Formulas’ , Electoral Studies , 14 , 417 – 439 . Google Scholar CrossRef Search ADS Carty R. K. ( 2002 ) ‘ The Politics of Tecumseh Corners: Canadian Political Parties as Franchise Organizations’, Canadian Journal of Political Science , 35 , 723 – 745 . Google Scholar CrossRef Search ADS Carty R. K. ( 2004 ) ‘ Parties as Franchise Systems: The Stratarchical Organizational Imperative’, Party Politics , 10 , 5 – 24 . Google Scholar CrossRef Search ADS Chubb B. ( 1963 ) ‘ Going about Persecuting Civil Servants: The Role of the Irish Parliamentary Representative’, Political Studies , 11 , 272 – 286 . Google Scholar CrossRef Search ADS Courtney M. ( 2015 ) ‘ Social Background and Intra-Party Attitudes in Ireland’, Irish Political Studies , 30 , 178 – 198 . Google Scholar CrossRef Search ADS Farrell D. M. ( 2001 ) Electoral Systems: A Comparative Introduction , Basingstoke, Hampshire , Palgrave MacMillan . Farrell D. M. , Mair P. , Ó Muineacháin S. , Wall M. ( 2015 ) ‘Courting, But Not Always Serving: Perverted Burkeanism and the Puzzle of Irish Parliamentary Cohesion’. In Johnston R. , Sharman C. (eds) Parties & Party Systems: Structure and Context , Vancouver , University of British Columbia Press , pp. 92 – 107 . Groseclose T. ( 2001 ) ‘ A Model of Candidate Location When One Candidate Has a Valence Advantage’, American Journal of Political Science , 45 , 862 – 886 . Google Scholar CrossRef Search ADS Katz R. , Mair P. ( 1995 ) ‘ Changing Models of Party Organization and Party Democracy: The Emergence of the Cartel Party’, Party Politics , 1 , 5 – 28 . Google Scholar CrossRef Search ADS Kavanagh A. P. ( 2015 ) ‘ An End to “Civil War Politics”? The Radically Re-shaped Political Landscape of Post-Crash Ireland’, Electoral Studies , 38 , 71 – 81 . Google Scholar CrossRef Search ADS Marsh M. , Farrell D. M. , McElroy G. (eds) ( 2017 ) A Conservative Revolution? Electoral Change in Twenty-First Century Ireland , Oxford : Oxford University Press . Google Scholar CrossRef Search ADS Marsh M. , G McElroy . ( 2016 ) ‘Voting Behaviour: Continuing De-alignment’. In Gallagher M. , Marsh M. (eds) How Ireland Voted 2016: The Election That Nobody Won , Cham, Switzerland , Palgrave Macmillan , pp. 159 – 184 . Google Scholar CrossRef Search ADS Marsh M. , Sinnott R. , Garry J. , Kennedy F. ( 2008 ) The Irish Voter: The Nature of Electoral Competition in the Republic of Ireland , Manchester , Manchester University Press . Marsh M. ( 2008 ) ‘Explanations for Party Choice’. In Gallagher M. , Marsh M. (eds) How Ireland Voted 2007: The Full Story of Ireland’s General Election , Basingstoke, Hampshire , Palgrave Macmillan , pp. 105 – 131 . Google Scholar CrossRef Search ADS McGraw S. ( 2015 ) How Parties Win: Shaping the Irish Political Arena , Ann Arbor , Michigan University Press . Google Scholar CrossRef Search ADS Murphy G, ( 2016 ) Electoral Competition in Ireland Since 1987: The Politics of Triumph and Despair , Manchester , Manchester University Press . Google Scholar CrossRef Search ADS Reidy T. , White T. J. ( 2017 ) ‘Political Transformation in Ireland: Boom, Bust and Beyond’ , Eire-Ireland , 52 , 101 – 123 . Google Scholar CrossRef Search ADS Shu M. ( 2003 ) Cope with Two-Dimensional Cleavage Structure: Party Politics in Referendums on European Integration, Paper presented at Workshop 19, ‘Cleavage Development: Causes and Consequences’, ECPR Joint Sessions, Edinburgh, 28 March–2 April 2003. Sinnott R. ( 1995 ) Irish Voters Decide: Voting Behaviour in Elections and Referendums Since 1918 , Manchester , Manchester University Press . Tavits M. ( 2007 ) ‘ Principle vs. Pragmatism: Policy Shifts and Political Competition’, American Journal of Political Science , 51 , 151 – 165 . Google Scholar CrossRef Search ADS van Vanno C. M. C. , Malka R. I. , Depauw S. , Hazan R. Y. , Andeweg R. B. ( 2014 ) Agreement, Loyalty, and Discipline: A Sequential Approach to Party Unity , Oxford , Oxford University Press . Wolinetz S. B. ( 2015 ) ‘Franchising the Franchise Party: How Far Can a New Concept Travel’. In Johnston R. , Sharman C. (eds) Parties & Party Systems: Structure and Context , Vancouver , University of British Columbia Press , pp. 72 – 91 . © The Author 2017. Published by Oxford University Press on behalf of the Hansard Society; all rights reserved. For Permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Parliamentary Affairs Oxford University Press

Rethinking the Franchise Party: Adding the Ideological Dimension—The Irish Case

Parliamentary Affairs , Volume Advance Article (3) – Dec 20, 2017

Loading next page...
 
/lp/ou_press/rethinking-the-franchise-party-adding-the-ideological-dimension-the-9ziwjzpOcX
Publisher
Oxford University Press
Copyright
© The Author 2017. Published by Oxford University Press on behalf of the Hansard Society; all rights reserved. For Permissions, please e-mail: journals.permissions@oup.com
ISSN
0031-2290
eISSN
1460-2482
D.O.I.
10.1093/pa/gsx046
Publisher site
See Article on Publisher Site

Abstract

Abstract Attributing the label ‘franchise’ to parties that possess certain organisational characteristics, such as centralized authority at the national level combined with substantial autonomy at the local level, has become increasingly common. These franchise characteristics have been identified in parties in a wide variety of country cases, in party systems as diverse as those of Canada, Australia, Israel and Ireland. Drawing upon evidence from three original parliamentary surveys in the Republic of Ireland, this article seeks to expand our understanding of franchise parties by focusing on the ideological behaviour adopted by franchise parties at both the national and the candidate levels. An analysis of the Irish case suggests that, in addition to previously studied organisational characteristics, candidates from franchise parties enjoy an unusual degree of flexibility to vary their policy positions. 1. Introduction Parties of all types modify their organisational structures in the face of increasingly competitive elections and more fluid social environments. One type of organisational response is for parties to become more stratarchical, whereby distinct levels within a party operate more or less independently of each other, resulting in parties that behave like franchise systems. Similar to franchises in business, the concept of a franchise party implies that the overall brand is determined at the national level, including selecting strategies, policies, marketing and overall management of the party, while party candidates and organisations at the local level are relatively free to deal with the interests of the communities they represent. Franchise parties find ways to combine growing centralisation at the national level with substantial autonomy at the local level where local party officials can often designate candidates of their own choice and can tailor local campaigns to address the particular interests of local constituents. This franchise model of party organisation seeks to reconcile the competitive imperative for greater centralised authority at the national level with the need to manage diverse demands from constituents (Carty, 2002, 2004; Wolinetz, 2015, pp. 73–75). For nearly two decades, scholars have been highlighting the mounting trend of parties to behave more like franchise parties in contexts as diverse as Canada, Australia, Israel and Ireland (Katz and Mair, 1995; Carty, 2002, 2004; Wolinetz, 2015). Parliamentary systems appear to be especially conducive to franchise arrangements because they allow ideologically cohesive parties to vary their ‘product’ to match the tastes of consumers in local markets. Similarly, electoral systems that place a premium on the personal popularity of individual candidates provide additional inducements for parties to behave in a franchise-like fashion, allowing local personalities opportunities to capitalise on their embedded social networks. It is not necessary to view the empirical manifestation of the franchise party in terms of a binary category, but more usefully can be seen along a spectrum of more ‘franchise-like’ or less. Although scholars have helpfully attributed the label ‘franchise party’ to a growing number of parties and have identified certain crucial organisational characteristics that the parties adopt, such as nominating candidates to run for office at the local level, few scholars have focused on the ideological attributes and behaviours of franchise parties. When ideology is taken into account, the focus is usually on the national level and analysing party manifestoes or roll call votes in parliament where a high degree of party discipline is often observed. Yet, a richer and fuller understanding of how franchise parties employ programmatic appeals can be achieved by complementing the attention to organisational adaptation with increased attention to adaptations within the ideological dimension. Specifically, what has been missing in the literature on franchise parties has been an awareness of the relationship between the party platform at the national level and the policy positions adopted by local members of parliament. Conceiving parties as franchise systems illustrates how parties may benefit competitively from allowing their individual candidates to vary their ideological positions, either to enhance candidates’ specific electoral appeal or to retain broad support from the political elite and attract high-quality candidates. Tolerating a degree of relative autonomy regarding policy positions helps parties appeal to voters with diverse, high-quality candidates and in diverse settings. However intentional or unintentional, by turning a blind eye when candidates adopt positions different than the overall party platform, national party organisations stand to gain electoral ground. Although members of parliament may need to toe the party line in parliament while in office, they can, and often do, vary their policy positions during campaigns and in private intra-party discussions. This article underscores the relative autonomy of franchise party candidates to adopt programmatic appeals opportunistically during elections. By focusing on the relationship between the programmatic commitments articulated in the party platform at the national level and the often divergent positions adopted by franchise party candidates at the local level, I examine a largely neglected dimension of the behaviours characteristic of franchise parties, that is, the ideological dimension. I argue that, in addition to tell-tale organisational adaptations characteristic of the franchise party, a fuller account of their behaviours can be achieved by focusing on the relatively greater degree of ideological autonomy enjoyed by franchise party candidates. The evidence to support my argument is drawn largely from the quantitative and qualitative data derived from three originally designed and implemented parliamentary surveys conducted in the Republic of Ireland. These surveys asked Irish MPs (known in Ireland as Teachta Dála or TDs) to place themselves and their party on a 0-10 continuum on six policy issues, and the results reveal that considerable ideological dispersion exists within parties. Although the surveys do not provide direct evidence to link what the TDs said in private during the face-to-face interviews to how they campaigned in their constituency (and there is limited space to explore comprehensively other campaign literature), the surveys explicitly asked TDs to reflect on their experience during the campaign. More specifically, the TDs were asked to locate themselves, their party and (in the 2016 survey) the average person in their constituency for each policy in the ideology section. The survey asked TDs to reflect on their attitudes on issues. The considerable intra-party variation across the issues shows that individual candidates differ from their parties on various issues, which has strategic benefits for a franchise party. This intra-party ideological variation underscores the flexibility individual candidates enjoy to adopt slightly different positions from their party. Fully explaining this variation is beyond the scope of this article. However, highlighting the prevalence of this intra-party ideological variation suggests that TDs are able to adopt their ideological positions—for a variety of personal reasons—to enhance their appeal. This is particularly relevant despite the fact that there is strict party discipline within Irish parties once in parliament, a majority of Irish voters are still attracted to candidates for non-ideological factors, and the overall ideological spectrum is comparatively narrow. The quantitative evidence from the surveys is reinforced by qualitative evidence from hundreds of interviews to suggest that individual candidates do seek to attract support from voters based on the ideological positions they adopt—even if this means differing slightly from the official party position. A recurring theme from the overall set of interviews is that Irish candidates have to be ‘all things to all people’ and must do what it takes to gain an edge over other candidates within the constituency, and sometimes their programmatic views provide this difference, especially on salient local issues. Since Ireland’s independence in 1922, social or economic ideologies were never substantial issues in Irish politics. The main cleavage was nationalism. In that context there was always a healthy and sometimes subliminal tension between national party positions and what candidates proposed locally. This was further bound up with the fact that national politicians were also candidates in local elections. A good example of this tension in elections is the issue of local hospitals. In 1973, reporters commented on how the status of local hospitals dominated that election: ‘Paradoxically, the answers they [candidates] give, which are most likely to please many of the local pressure groups, would not always result in an improvement in health care ….’ nor did they necessarily coincide with national party position.1 More recently, candidates often determine whether they adopt their party’s position depending on whether that policy offers specific versus general benefits. During the 2007 election, Michael Fitzpatrick was a member of the historically dominant party Fianna Fáil, which was in power for the majority of the Irish state's existence and routinely won over 40% of the vote at every general election up to 2011. Although the Fianna Fáil government had announced that there would be only one national children’s hospital located in Dublin in 2006, Fitzpatrick decried that decision on local radio, arguing that the children services in his constituency’s hospital desperately needed to be enhanced. He defended his position against that of the party arguing that he thought someone in the party may be persuaded by his arguments. Yet, when asked about his policy stance on other policy positions, such as the nurses’ strike, he evinced incredulity that one might take a position in opposition to that of the party: ‘We are the government party. We will always be supportive of the government line (M. Fitzpatrick, interview, July 2007).’ In Irish politics, candidates seek to maximise both party and individual support. When there is a conflict, candidates often adhere to the party positions when more general benefits are at stake, such as wages, taxes, etc. But when push comes to shove in a matter of local benefits to the constituency, local candidates will readily jettison national party positions so as to not expose themselves to attacks from their local opponents. In a highly competitive and narrow electoral arena, candidates can ill-afford to lose out on such appeals; hence, Fitzpatrick. When one examines the policy appeals of the other candidates, they, too, seek to mobilise voters on the basis of both strong party support and specific policy benefits connected to hospitals, schools and other services to the local constituency. As this example shows, the Irish case is a useful context to explore the behaviour of franchise parties because several of its major parties behave as franchise systems that, especially during electoral campaigns, combine strong national party organisations with highly mobilised local candidates that embody the party brand. There has been an explosion of literature (Kavanagh, 2015; Murphy, 2016; Marsh et al., 2017; Reidy and White, 2017) assessing the changing nature of Irish politics since Ireland’s earthquake election in 2011, which saw Fianna Fáil decimated as it fell to 17% of the vote and from 78 seats to 20 in the national parliament. For many, Ireland may finally be shifting from Civil War politics to a context where ideology plays a more significant role. This article examines previously understudied dynamics within Irish politics and illustrates how ideological flexibility at the level of individual parliamentarian is one way in which parties can adapt to an evolving electoral context and can behave more like franchise systems. The article is organised in the following way. First, I provide a brief overview of the Irish party system and offer a rationale for why Ireland serves as a useful case study for examining the broader phenomenon of franchise parties. Secondly, I describe the prevalence of intra-party ideological diversity within major Irish parties and argue that this set of adaptive behaviours is indicative of franchise parties. Parties can utilise internal ideological diversity to bolster their attractiveness locally, a trait especially useful in more volatile electoral environments. The article then turns to an analysis of the factors that shape the ideologically diverse positions taken by individual Irish politicians within parties. By analysing the linkage between the ideological positions adopted by individual candidates during elections and comparing these with how the party markets itself at the national level, I seek to add to our understanding of the multiple levels on which franchise parties employ programmatic appeals to broaden their attractiveness to voters. The findings suggest that Irish parliamentarians, especially among major parties, possess considerable autonomy to select their policy appeals during elections, an autonomy which is shaped by the individual parliamentarian’s demographic background and electoral and constituency context. The article concludes with a discussion of how the approach adopted here can augment our understanding of how parties compete more generally. 2. The Irish case Unlike many other Western European party systems, the Irish party system did not emerge from sharp social divisions such as class, but rather from a split emanating from within the original Sinn Féin party, which was the national Irish political party prior to the Irish Civil War (1922–1923). It has some organisational continuity with the contemporary Sinn Féin party, but also with Fine Gael and Fianna Fáil. Sinn Féin in the contemporary context represents a combination of the former political wing of the Irish Republican Army (IRA) and left-leaning politicians who seek to create a United Ireland. Key founding conditions strengthened the cross-class nature of partisan ideological appeals and contributed to the overall weakness of the left in Irish electoral politics (Marsh et al., 2008, p. 32). Since the consolidation of the party system in the 1930s until 2016, Fianna Fáil, Fine Gael and Labour have managed to attract an average of over 80% of the total vote. Much of this predominance is due to the major parties’ use of non-ideological appeals, organisational strength and their ability to effectively employ both partisan and personalistic appeals within Ireland’s Proportional Representation-Single Transferable Vote (PR-STV) electoral system. Ireland’s parties now face a more demanding electoral environment than they faced in elections prior to the 1980s. In recent years, the underlying cleavage structure in Ireland, as well as the political value system and the issue dimensions relevant for elections, have shifted in critical ways. The demographic composition of the electorate has dramatically changed as Ireland has transitioned from one of the poorest to one of the wealthiest European countries in less than a generation. Historic cleavage patterns have largely lost their hold. This turbulent environment has resulted in lower turnout, lower party attachment and increased numbers of floating voters. For example, partisan identities have declined precipitously: the percentage of voters in the Irish electorate reporting that they lack a close relationship with any party has more than doubled from 34% in 1978 to 79% in 2016 (Marsh et al., 2008, pp. 59–79; Marsh and McElroy, 2016, p. 179). A rapidly changing Irish society and an increasingly competitive electoral marketplace have combined to increase the prospects for electoral volatility. Taken together with the complex and technocratic character of the issues facing modernising Ireland, parties have been forced to adapt to these challenges or suffer electoral defeat. As a result, major party organisations have become more centralised, professionalised and bureaucratised. These trends, in turn, have carried with them important consequences, resulting in an enhanced capacity for party elites at the national level to coordinate efforts. In addition, these developments have necessitated a streamlining of how parties select candidates and leaders, raise money and communicate policy priorities. At the same time, the imperative of efficient local delivery of goods and services and the personalised nature of electoral appeals incentivise parties to field candidates who can simultaneously sustain multiple linkages to government offices at every level. Successful candidates must simultaneously maintain linkages at the national party level and simultaneously engage successfully in Ireland’s complicated civil service protocols to deliver goods, services and information locally. The parties’ ability to field strong candidates that work effectively at both the national and local levels has been crucial for their longer term electoral success. Although it is has been 50 years since Chubb (1963) famously suggested that the role of Irish TDs is to ‘go about persecuting civil servants’, the demand for unusually robust constituency service and the expectation that TDs will serve as brokers between voters and the state continue to dominate Irish politics. On average, 40% of voters claim to choose their TDs primarily based on the candidate’s ability to look after the constituency’s needs. Additionally, people from all sectors of Irish society, including educated and upper-income citizens, continue to use their elected representatives in a brokerage capacity to access goods and services from the state. Prior to the 1990s, it was easier for elected officials to maintain well-oiled local machines and to deliver the goods. Before 2003, Irish TDs conveniently held both the parliamentary post and a local council office, called the ‘dual mandate’. Given that more than half of all government spending, although funded by central government transfers, is allocated locally; local councils provided parliamentarians a rich source of patronage. In addition, locally elected councillors hold dispositive authority regarding issues pertaining to planning permission/zoning, roads, medical cards, waste management and so on. The elimination of the ‘dual mandate’ deprived members of parliament of an important source of patronage, rendering it more difficult after 2003 for TDs to confer favours and deliver the goods to local constituents. With little perceived programmatic difference among Ireland’s parties during elections, party organisational capacity and personalistic appeals become essential mechanisms by which candidates and parties establish and maintain linkages and make their case with voters. However, whereas in the past Irish TDs may have relied primarily on non-ideological appeals to attract voter support, circumstances have changed. As Irish society has become more fluid and urban, politicians have had to develop a broader array of tools to compete in this increasingly competitive environment. While constituency service remains the principal factor driving vote choice (Marsh et al., 2008, pp. 220–238), there has been a growing influence of policy position on vote choice according to elections surveys. Between 1977 and 2016, approximately 39% of voters claimed that constituency service most influenced their vote choice compared with 26% who reported that party policy matters most. The 2011 (41%) and 2016 (33%) election polls indicated above average results for those who reported voting based on policy considerations. Thus, a growing insistence among voters for more sophisticated policies requires parties and candidates to respond accordingly. One important way in which political entrepreneurs have adapted to this changing set of circumstances is to tailor ideological appeals to fit voter profiles within their local constituencies. Data drawn from parliamentary surveys suggest that the policy positions adopted by individual TDs can diverge from the positions adopted by their respective parties at the national level. This is particularly true of major party (i.e. Fianna Fáil, Fine Gael and to a lesser extent Labour) TDs. Even though major parties are perceived as holding very similar programmatic views at the aggregate level and TDs typically adhere to strict party discipline in legislative votes, the high degree of internal ideological differences observed during elections reflects significant levels of autonomy that individual TDs have to adapt in their policy. The analysis of campaign material and speeches, while interesting, is outside the scope of this article. Therefore, the responses TDs share in surveys provide indirect evidence for how they shaped programmatic appeals to voters during election campaigns. Nevertheless, it is critical to note that the Irish TDs interviewed in the three waves of the survey were asked to identify their individual positions and those of their party on the same policy continuums during each respective election. Although there is some inevitable variation in whether a specific TD construes their party’s position as a ‘6’ or a ‘7’, if a TD places himself one or two points to the left or right of the party on an issue, there is some reason why they did not place themselves equal to their party. Although it can be helpful for Irish TDs to announce their ideological positions in very public ways, it is widely known that the primary way candidates interact with individual voters is during door-to-door canvassing during election campaigns. In surveys, most TDs surveyed reported that door-to-door canvassing was the single most effective means of connecting with voters—with 72% in 2007, 68% in 2011 and 86% in 2016, highlighting the importance of this critical personal contact. Because these personal interactions constitute the grit and gristle of Irish politics, candidates seek to establish a strong personal connection with the voters during these encounters on the doorstep. Candidates are exposed to and listen to whatever issues voters identify as being important to them and share, in turn and when time permits, their support of these views. The overall impression from doing over 300 interviews is that the views TDs report in the parliamentary surveys represent their sincere positions on key issues facing the electorate. 3. Intra-party ideological diversity It should come as no surprise that Irish parties evince a high degree of internal ideological diversity. Political parties in a wide range of comparative contexts experience considerable heterogeneity of ideological positions among their individual politicians. Parties can, and do, in fact, capitalise on the diverse views held by their elected representatives to win votes, even if representatives generally follow the party line when voting in parliament. This intra-party ideological diversity occurs not only in Ireland, but also in countries with traditions of broader and more clearly defined ideological competition. Data from the Comparative Candidates Survey (CCS) show that party affiliation explains between 60% and 80% of the variance in candidates’ ideology among a broad number of country cases, including Germany, Hungary, Sweden, Switzerland and Portugal. For example, in Germany following the 2005 election, 62% of the variance in MPs’ ideological positions was between-party variance, while 38% was within-party variance. In Ireland, by contrast, less than 15% of the variance is explained by party affiliation and the remainder by internal party differences (Courtney, 2015). The prevalence of ideological diversity within parties represents an additional tool parties have at their disposal to increase their vote in highly competitive contexts where there is a need to attract votes from various segments of society and from among unattached voters. Although it is widely known that many parties evince intra-party ideological diversity, what explains this diversity? Why does it matter? At first glance, internal party heterogeneity appears to matter little to Ireland’s political parties and to large portions of the electorate. Parties at the national level in Ireland establish party policy positions and enjoy rigorous party discipline. Even one vote against the party within parliament generally results in immediate expulsion. In a comparison of 15 developed democracies, Irish parliamentarians voted programmatically with their party more than any other country in the survey (van Vanno et al., 2014, p. 124). As a consequence, what individual politicians profess in terms of their personal programmatic preferences may not matter to the national party. Nevertheless, evidence from my 2016 Parliamentary Survey indicated that across all parties, strong majorities of TDs were ‘likely’ or ‘very likely’ to express opinions privately within their party—whether to party leadership, to fellow TDs, or in parliamentary meetings. TDs were less willing to broadcast disagreements outside the party. A narrow majority (52%) said they were likely to express opinions privately to constituents—although Sinn Féin, Green and left party TDs were less likely to do so, compared with nearly two-thirds of major party TDs. This suggests that the beliefs of TDs may impose a potential constraint on the behaviour of party leadership. Although personality and local services matter a great deal in shaping voter preferences, there is still some room for policy to be important. Internal party heterogeneity, and the ideological flexibility it provides at the local level, does seem to matter to individual Irish politicians who must find any way possible to distinguish themselves in an ever more competitive electoral environment. The importance of a candidate's personal interests and reputation increases in systems like Ireland’s where co-partisans compete with each other for votes and seats (Carey and Shugart, 1995, pp. 430–431). Ireland's PR-STV electoral system provides voters with a requirement to choose among both candidates and parties. This electoral formula incentivises behaviours, which highlight intra-party distinctions among co-partisans as they compete, not only against candidates from competing parties, but also against members of their own party. To succeed in this competitive market with fewer attached voters and increased competition from other parties and independents, candidates have been driven to develop highly personalized organisations—or local franchises—to gain their party’s nomination and to win their seat in parliament. Because non-ideological factors, candidate quality and organisational strength are crucial for candidates’ success in the Irish context, parties have an incentive to tolerate ideological autonomy from TDs, as long as TDs vote as instructed when in office. This ideological autonomy is tolerated regardless of whether TDs use it to advance locally popular arguments or advocate for personal beliefs and causes. Parties cannot afford to alienate effective local TDs or foist locally unpopular positions upon them, which means programmatic diversity matters for Irish franchise parties’ strategy. 4. Franchise parties in the Irish context Ideological competition in the Republic of Ireland underscores how the franchise party operates. Irish parliamentarians combine extremely coherent ideological party positions in parliament with high levels of constituency service and varied programmatic appeals during elections (Farrell et al., 2015; Wolinetz, 2015, p. 87). Behaving as franchise parties allows Irish parties to reconcile these apparently contradictory imperatives. Individual politicians can vary their ideological positions to enhance their appeal for several reasons. First, while candidates may want to offer distinct policy appeals from their opponents, they may also choose a position closer to their opponent, thereby de-emphasising the importance of policy and increasing relative importance of his or her valence advantage. In effect, such an approach encourages the median voter to conclude that if candidates are so similar on policy, then they will cast their vote based on other, more personal characteristics and abilities (Groseclose, 2001, pp. 862–863). This returns the advantage to those who are better positioned to serve local constituency needs once policy concerns are neutralised. A second reason to adopt diverse or contrary views from the party might reside in an effort to ‘gain points’ in line with Mayhew’s position-taking strategy (Mayhew, 1974). According to one senior elected Irish politician, and reiterated by scores of others, TDs regularly attempt to persuade constituents that they hold deep sympathies for positions most favoured by local constituents, and indeed argued strenuously for such positions during parliamentary party meetings, but claim to have been outvoted. These same politicians then protest that they must remain within the party, and in power, if they are to better serve the local constituency (anonymous interview with Fianna Fáil politician, 2 June 2016, Dublin). In this way, hundreds of TDs reported how they must present themselves as being simultaneously sensitive to local concerns and loyal to the party, which they convince voters help them to serve constituents more effectively over the long run. A third reason individual TDs may feel free to advance divergent views during elections as a means of establishing a deeper connection with voters is because they can reasonably gamble that they may never be forced to vote on the contended issues in parliament. Previous scholarship has demonstrated that many of Ireland’s more controversial issues have in fact been displaced from the electoral and parliamentary arenas to non-electoral institutional domains, and that as a result, many salient issues facing Irish society are simply side-stepped by parliament (McGraw, 2015). The practice of regularly displacing highly controversial issues from the electoral arena to non-parliamentary domains provides individual TDs greater degrees of freedom to take positions on these same potentially divisive issues that most resonate with their constituents. Finally, TDs may hold positions for personal rather than strategic reasons. Given the importance of personal appeal in Irish voting behaviour, parties have a strong incentive to recruit appealing local candidates. The need for locally appealing candidates may override the need for policy and ideological orthodoxy, as long as these candidates ultimately follow the party line in parliament. Franchise parties are likely to tolerate ideologically diverse candidates TDs regardless of the reasons for this diversity, so long as it does not threaten party discipline. 5. Measuring intra-party ideological diversity I designed three original surveys of the Irish Parliament to capture party positions, and in particular, to analyse how individual parliamentarians vary their programmatic appeals. This allows us to look at how major Irish parties behave as franchise systems that operate differently at the national versus the local level. The surveys highlight both aggregate party positions and individual TD attitudes on key issues in Ireland’s 2007, 2011 and 2016 general elections. While earlier research highlights ideological flexibility at the level of the party, this article focuses on this same behaviour at the level of individual politicians. Each of the surveys involved face-to-face interviews where elected TDs were asked to reflect on their experience in that election. My research team interviewed 102 of the 166 Irish TDs after the 2007 general election, 115 of 166 TDs after the 2011 general election, and 97 of 158 TDs after the 2016 general election. The three surveys occurred in very different political contexts: the roaring ‘Celtic Tiger’ economy in 2007, the aftermath of the worst economic crisis in Ireland’s history in 2011 and modest recovery by 2016. These three elections saw unusually high turnover and represent a virtual earthquake in the Irish electoral landscape: Fianna Fáil secured a strong mandate to govern in 2007, dropped from 78 to 20 seats at the 2011 elections, and partially recovered in 2016; Sinn Féin incrementally increased its number of seats, and small leftist parties also made marked gains. The survey sample was representative of key social and political demographics, such as age, gender, religious denomination, education levels, party, type of geographic constituency and previous experience in office (see Supplementary Materials for additional detail). Similar to other recent studies, evidence from these parliamentary surveys also confirms that there is more attitudinal variance within Irish parties than between them. Courtney found that 87% of ideological variance occurred within parties compared to 74% in my surveys (Courtney, 2015, p. 185). Given these results, it is instructive to look beneath the aggregate level of the party, where high levels of programmatic convergence are commonly observed, to the level of individual politicians, where a more nuanced picture of the array of ideological attitudes emerges. The surveys asked TDs to place themselves and their parties on a 0–10 scale within a number of policy areas: left–right position, and policy attitudes towards Northern Ireland, European integration, abortion, fiscal policy and environmental policy. (Specific information regarding the survey, including precise question wording, is available in Supplementary Materials.) These issues measure the most salient policy dimensions within contemporary Irish politics and society (Sinnott, 1995, pp. 279–297; Shu, 2003). To test what determines individual MPs’ attitudes, three sets of variables were included in this analysis. Party affiliation is the most common, and usually strongest, determinant of position. Key electoral and constituency level variables may also encourage politicians to vary their positions to meet personal or strategic needs. These latter variables test whether geography, internal party competition, levels of electoral competitiveness and other constituency factors influence TD ideological positions. The assumption is that key competitive dynamics could encourage candidates to alter their ideological positions to offset their competitors. Including these variables allows us to examine how changes in Irish society and the increasingly competitive electoral environment may be affecting ideological competition. Demographic variables are also likely to affect candidates’ positions, as they reflect the life experiences that sway personal political views. If demographic variables play a large role in determining party position after controlling for other factors, this suggests that TDs’ personal views and parties’ needs for attractive candidates drives the ideological diversity of Irish franchise parties. Thus, I test whether TDs adopt locally varied positions for personal demographic reasons, and/or whether TDs adopt locally varied positions for strategic reasons. The findings are somewhat inconclusive, which suggests that both factors may be at work. Even though there is no systematic explanation for why individual politicians choose their positions, the results suggest that Irish politicians enjoy an unusual degree of flexibility to employ ideology strategically to complement an array of other strategies. As previously noted, in contrast to the ideological convergence that is reflected at the aggregate level of the party, the level of ideological heterogeneity is greater at the level of the individual politician, reflecting a broader spread of policy positions among individual MPs. Table 1 shows that Irish TDs maintain a wide diversity of views on key issues. Table 1 Standard deviation of TD attitudes in 2016, by party (See Supplementary Tables S1 and S2 for similar results in 2007 and 2011) FF FG Lab Gr SF Ind Left Brown–Forsythe test statistic (Number of cases) 36 22 3 2 16 7 10 United Ireland 1.9 1.7 1.2 0 0.6 1.9 3.4 3.9** Left–Right 1.1 1.5 1.5 1.4 1.4 1.8 1.8 0.4 Abortion 2.5 3.1 1.7 0 2.9 3.8 2.1 1.4 EU 1.7 1.1 1.5 0.7 1.6 3.4 2.9 4.4** Fiscal policy 1.1 1.7 1.2 1.4 1.1 1.6 2.0 0.7 Environment 1.4 1.9 2 0 1.6 1.6 1.9 1.0 FF FG Lab Gr SF Ind Left Brown–Forsythe test statistic (Number of cases) 36 22 3 2 16 7 10 United Ireland 1.9 1.7 1.2 0 0.6 1.9 3.4 3.9** Left–Right 1.1 1.5 1.5 1.4 1.4 1.8 1.8 0.4 Abortion 2.5 3.1 1.7 0 2.9 3.8 2.1 1.4 EU 1.7 1.1 1.5 0.7 1.6 3.4 2.9 4.4** Fiscal policy 1.1 1.7 1.2 1.4 1.1 1.6 2.0 0.7 Environment 1.4 1.9 2 0 1.6 1.6 1.9 1.0 Significant at **p ≤ 0.01; *p ≤ 0.05;  ^ = .10, * = .05, ** = .01. Table 1 Standard deviation of TD attitudes in 2016, by party (See Supplementary Tables S1 and S2 for similar results in 2007 and 2011) FF FG Lab Gr SF Ind Left Brown–Forsythe test statistic (Number of cases) 36 22 3 2 16 7 10 United Ireland 1.9 1.7 1.2 0 0.6 1.9 3.4 3.9** Left–Right 1.1 1.5 1.5 1.4 1.4 1.8 1.8 0.4 Abortion 2.5 3.1 1.7 0 2.9 3.8 2.1 1.4 EU 1.7 1.1 1.5 0.7 1.6 3.4 2.9 4.4** Fiscal policy 1.1 1.7 1.2 1.4 1.1 1.6 2.0 0.7 Environment 1.4 1.9 2 0 1.6 1.6 1.9 1.0 FF FG Lab Gr SF Ind Left Brown–Forsythe test statistic (Number of cases) 36 22 3 2 16 7 10 United Ireland 1.9 1.7 1.2 0 0.6 1.9 3.4 3.9** Left–Right 1.1 1.5 1.5 1.4 1.4 1.8 1.8 0.4 Abortion 2.5 3.1 1.7 0 2.9 3.8 2.1 1.4 EU 1.7 1.1 1.5 0.7 1.6 3.4 2.9 4.4** Fiscal policy 1.1 1.7 1.2 1.4 1.1 1.6 2.0 0.7 Environment 1.4 1.9 2 0 1.6 1.6 1.9 1.0 Significant at **p ≤ 0.01; *p ≤ 0.05;  ^ = .10, * = .05, ** = .01. Whereas TDs within the three historic parties exhibit a relatively wide dispersion of views on almost every issue, TDs within more policy-driven parties exhibit far fewer internal party differences on their primary policy concerns. However, even policy-driven parties show substantial internal differences on more peripheral issues. These non-major parties evince less ‘franchise-like’ behaviours than the major parties, but still tolerate ideological diffusion on some policies. Although caution must be exercised in interpreting these results given the small sample size for the minor parties, the more cohesive results for these parties suggest that they are more ideologically coherent on core issues. Consider the case of Sinn Féin. Naturally, Sinn Féin TDs offered clearer, more unified positions on its core policy issue related to United Ireland, holding essentially identical positions on that issue. It also held the most coherent set of positions on the left–right scale. However, Sinn Féin TDs hold increasingly diverse views the further one’s moves from the party’s core issues. Sinn Féin TDs were less cohesive on spending and the environment, and the party experienced growing diversity among its TDs in terms of abortion. In fact, Peadar Tobin, TD, was temporarily suspended from Sinn Féin for his vote on the same 2013 abortion bill that divided Fine Gael. Interestingly, Tobin was reinstated six months later in February 2014 with little fanfare and Tobin stated he was delighted to be back in his political home (Irish Independent, February 5, 2014). Results from the United Left Alliance in 2011 display a similar pattern to Sinn Féin. Also, the Green TDs held reported identical positions on half of the issues. Figure 1 charts TD attitudes on three fundamental issues within Irish politics for the 2007, 2011 and 2016 elections: the founding united Ireland cleavage, the left–right scale and abortion. A diversity of positions exists at the level of the individual TD along various policy dimensions, irrespective of whether there is political contention over an issue (e.g. abortion), it is the perspective of the system as a whole (e.g. left–right), or uncertainty as to how to achieve longer term goals (e.g. United Ireland). Figure 1 View largeDownload slide TD attitudes on salient issues in 2007, 2011 and 2016 elections. Figure 1 View largeDownload slide TD attitudes on salient issues in 2007, 2011 and 2016 elections. These findings underscore an essential feature of Irish politics: individual TDs and candidates from major parties exercise a considerable degree of personal discretion whether to embrace—or distance themselves—from the policy positions taken by their co-partisans. The utility of a high degree of personal discretion becomes apparent when the dynamics of the Ireland’s PR-STV electoral system are again considered. Because candidates from the major parties do not rely on any one social group or policy appeal for their votes, they necessarily rely on greater ideological flexibility. It is easier for these latter candidates to emphasise policies that increase their appeal within the constituency, or to emphasise policies that fulfil their personal ideological and policy goals. At the national level, the major parties are able to highlight their official party position, while at the local level their candidates are relatively free to offer more targeted appeals. By contrast, candidates from more policy-driven parties naturally have less incentive to jettison or modify the party line on their core issues since policy goals are central to their very candidacy (Tavits, 2007). 6. What explains intra-party ideological diversity? What explains the intra-party ideological variation that exists within Irish political parties? Or, in other words, what are the factors that determine the diverse policy positions adopted by individual parliamentarians? In addition to party affiliation, there are two main mechanisms that shape TDs’ political views. First, TDs hold ideologically diverse views, as a matter of course, because they are recruited from an ideologically representative cross-section of society and hold personal political leanings informed by demographic and family background (Byrne and O’Malley, 2012). Moreover, TDs are influenced by the views held by their neighbours and fellow constituents, which, in turn, shape the specific electoral challenge they face. The ability of TDs to adopt programmatic positions opportunistically may be highly consequential for their electoral prospects, for it allows them to maximise their appeals among blocs of voters within their local constituency which can engender voter loyalty. A series of linear regression models are employed to test the way in which these three sets of factors—party affiliation; demographic and family background; and local context—shape MPs’ positions on policy issues. These factors hold the potential to explain the high levels of intra-party ideological diversity among the major parties observed in the previous section, while simultaneously maintaining a high degree of ideological consensus at the aggregate level. As dependent variables, the models use TD position on the six policy areas described previously. I seek to predict policy positions using three sets of independent variables. First, there is a notable but relatively low-magnitude relationship between party and MP’s ideological positions. When controlling for party, I then show how background and local context supplement party as predictors of TD positions. Because the surveys of TDs necessarily covered a relatively small number of cases, I test background and local context via a series of separate models, rather than aggregating them into a single model. This allows us to test a large number of independent variable specifications without greatly limiting the degrees of freedom and undermining the explanatory power of the findings. Most of the models presented are ordinary least squares linear regressions. I used clustered standard errors to account for TDs who completed multiple surveys, as well as a fixed effect for a year. Because Ireland uses multi-member electoral districts, the analysis of constituency-level variables was conducted using hierarchical models that nested individuals within electoral districts and years. The analysis of some specifications for electoral marginality uses the difference between TD and party position as a dependent variable; because this variable takes a small number of values (usually a 0-, 1- or 2-point difference), I treat the dependent variable as categorical and use an ordered logit model. 6.1 Party affiliation While party affiliation is not an overwhelming predictor of TDs’ attitudes, it is still relevant. The R2 values indicate how much of the variance in TDs’ attitudes is explained by party affiliation, along with a fixed effect for year-to-model shifts in ideology between the 2007 and 2011 elections and between the 2011 and 2016 elections (Table 2). For example, 45% of the variation in attitudes towards the European Union (EU) is explained by party affiliation and year, while 55% remains unexplained. Party affiliation and year explained less than half of the variation in TD attitudes on the other salient issues as well. Labour, Sinn Féin and the United Left Alliance TDs tend to show differences from Fianna Fáil (the base category), but Fine Gael and Fianna Fáil TDs are rarely distinct. The explanatory power of the models, measured by R2, also dropped substantially when minor parties were excluded from the model—confirming that party identity does more to explain policy positions among minor party MPs. For example, the R2 fell to 0.36 for left–right, 0.05 for the EU, 0.21 for fiscal policy, 0.05 for environmental policy, 0.13 for Northern Ireland and 0.36 for abortion when the minor parties were excluded. Table 2 Regression of policy attitudes on party ID United Ireland Left–Right Abortion EU Fiscal policy Environment Party: Fine Gael 1.55 (0.35)** 0.82 (0.25)** −0.73 (0.36)* 0.36 (0.24) 0.73 (0.27)** 0.21 (0.26) Party: Labour 1.71 (0.42)** −1.84 (0.27)** −4.00 (0.54)** 0.35 (0.30) −1.68 (0.33)** −0.55 (0.35) Party: Sinn Fein −3.01 (0.28)** −2.38 (0.29)** −3.40 (0.68)** −3.28 (0.44)** −2.90 (0.28)** −1.67 (0.35)** Party: Green 0.39 (0.42)^ −1.01 (0.42)* −1.40 (0.82)^ −0.40 (0.68) −1.08 (0.48)* −3.37 (0.84)** Party: ULA 0.66 (0.95) −3.84 (0.47)** −6.43 (0.62)** −3.86 (0.71)** −3.95 (0.52)** −2.90 (0.64)** Party: Independent 0.34 (0.70) −1.15 (0.54)* −2.57 (0.91)** −2.30 (0.78)** −1.68 (0.33)** −0.82 (0.62) Year: 2011 −0.45 (0.28) 0.13 (0.18) 0.16 (0.30) −0.65 (0.21)** −0.48 (0.26)^ 0.10 (0.23) Year: 2016 −0.58 (0.30)* 0.13 (0.18) −0.81 (0.35)* −0.76 (0.23)** −0.42 (0.23)^ −0.28 (0.23) Intercept 3.81 (0.28)** 4.63 (0.15)** 8.12 (0.27)** 8.82 (0.17)** 5.38 (0.22)** 5.28 (0.19)** R2 0.32 0.48 0.36 0.45 0.43 0.23 United Ireland Left–Right Abortion EU Fiscal policy Environment Party: Fine Gael 1.55 (0.35)** 0.82 (0.25)** −0.73 (0.36)* 0.36 (0.24) 0.73 (0.27)** 0.21 (0.26) Party: Labour 1.71 (0.42)** −1.84 (0.27)** −4.00 (0.54)** 0.35 (0.30) −1.68 (0.33)** −0.55 (0.35) Party: Sinn Fein −3.01 (0.28)** −2.38 (0.29)** −3.40 (0.68)** −3.28 (0.44)** −2.90 (0.28)** −1.67 (0.35)** Party: Green 0.39 (0.42)^ −1.01 (0.42)* −1.40 (0.82)^ −0.40 (0.68) −1.08 (0.48)* −3.37 (0.84)** Party: ULA 0.66 (0.95) −3.84 (0.47)** −6.43 (0.62)** −3.86 (0.71)** −3.95 (0.52)** −2.90 (0.64)** Party: Independent 0.34 (0.70) −1.15 (0.54)* −2.57 (0.91)** −2.30 (0.78)** −1.68 (0.33)** −0.82 (0.62) Year: 2011 −0.45 (0.28) 0.13 (0.18) 0.16 (0.30) −0.65 (0.21)** −0.48 (0.26)^ 0.10 (0.23) Year: 2016 −0.58 (0.30)* 0.13 (0.18) −0.81 (0.35)* −0.76 (0.23)** −0.42 (0.23)^ −0.28 (0.23) Intercept 3.81 (0.28)** 4.63 (0.15)** 8.12 (0.27)** 8.82 (0.17)** 5.38 (0.22)** 5.28 (0.19)** R2 0.32 0.48 0.36 0.45 0.43 0.23 Base Categories: Fianna Fáil/2007 Dáil; Standard errors in parentheses. Significant at **p ≤ 0.01; *p ≤ 0.05; ^ = .10, * = .05, ** = .01. Table 2 Regression of policy attitudes on party ID United Ireland Left–Right Abortion EU Fiscal policy Environment Party: Fine Gael 1.55 (0.35)** 0.82 (0.25)** −0.73 (0.36)* 0.36 (0.24) 0.73 (0.27)** 0.21 (0.26) Party: Labour 1.71 (0.42)** −1.84 (0.27)** −4.00 (0.54)** 0.35 (0.30) −1.68 (0.33)** −0.55 (0.35) Party: Sinn Fein −3.01 (0.28)** −2.38 (0.29)** −3.40 (0.68)** −3.28 (0.44)** −2.90 (0.28)** −1.67 (0.35)** Party: Green 0.39 (0.42)^ −1.01 (0.42)* −1.40 (0.82)^ −0.40 (0.68) −1.08 (0.48)* −3.37 (0.84)** Party: ULA 0.66 (0.95) −3.84 (0.47)** −6.43 (0.62)** −3.86 (0.71)** −3.95 (0.52)** −2.90 (0.64)** Party: Independent 0.34 (0.70) −1.15 (0.54)* −2.57 (0.91)** −2.30 (0.78)** −1.68 (0.33)** −0.82 (0.62) Year: 2011 −0.45 (0.28) 0.13 (0.18) 0.16 (0.30) −0.65 (0.21)** −0.48 (0.26)^ 0.10 (0.23) Year: 2016 −0.58 (0.30)* 0.13 (0.18) −0.81 (0.35)* −0.76 (0.23)** −0.42 (0.23)^ −0.28 (0.23) Intercept 3.81 (0.28)** 4.63 (0.15)** 8.12 (0.27)** 8.82 (0.17)** 5.38 (0.22)** 5.28 (0.19)** R2 0.32 0.48 0.36 0.45 0.43 0.23 United Ireland Left–Right Abortion EU Fiscal policy Environment Party: Fine Gael 1.55 (0.35)** 0.82 (0.25)** −0.73 (0.36)* 0.36 (0.24) 0.73 (0.27)** 0.21 (0.26) Party: Labour 1.71 (0.42)** −1.84 (0.27)** −4.00 (0.54)** 0.35 (0.30) −1.68 (0.33)** −0.55 (0.35) Party: Sinn Fein −3.01 (0.28)** −2.38 (0.29)** −3.40 (0.68)** −3.28 (0.44)** −2.90 (0.28)** −1.67 (0.35)** Party: Green 0.39 (0.42)^ −1.01 (0.42)* −1.40 (0.82)^ −0.40 (0.68) −1.08 (0.48)* −3.37 (0.84)** Party: ULA 0.66 (0.95) −3.84 (0.47)** −6.43 (0.62)** −3.86 (0.71)** −3.95 (0.52)** −2.90 (0.64)** Party: Independent 0.34 (0.70) −1.15 (0.54)* −2.57 (0.91)** −2.30 (0.78)** −1.68 (0.33)** −0.82 (0.62) Year: 2011 −0.45 (0.28) 0.13 (0.18) 0.16 (0.30) −0.65 (0.21)** −0.48 (0.26)^ 0.10 (0.23) Year: 2016 −0.58 (0.30)* 0.13 (0.18) −0.81 (0.35)* −0.76 (0.23)** −0.42 (0.23)^ −0.28 (0.23) Intercept 3.81 (0.28)** 4.63 (0.15)** 8.12 (0.27)** 8.82 (0.17)** 5.38 (0.22)** 5.28 (0.19)** R2 0.32 0.48 0.36 0.45 0.43 0.23 Base Categories: Fianna Fáil/2007 Dáil; Standard errors in parentheses. Significant at **p ≤ 0.01; *p ≤ 0.05; ^ = .10, * = .05, ** = .01. 6.2 Demographic factors Historically, party loyalty has been passed down through families in Ireland based on the founding cleavage regarding the 1921 treaty and the establishment of the independent Irish state. Anti-treaty families support Fianna Fáil, while pro-treaty families support Fine Gael. However, family background does not appear to have an impact on policy positions. Supplementary Table S3 reflects the finding that family background is not a significant predictor of TD attitudes after controlling for the family’s party affiliation. In other words, family background—one of the key factors in sorting political elites into parties—is almost entirely non-ideological. Although TDs’ family background displays no discernible relation with the positions TD espouse, TDs’ personal demographic qualities reflect modest impacts on policy positions (Table 3). Once controlling for party affiliation and other variables, age is rarely a significant determinant of TD attitudes. However, gender and religiosity have somewhat stronger effects. In terms of gender and ideology, female TDs are markedly more inclined to abandon the demand for a united Ireland, and they are somewhat further to the left on environmental issues, which means they report believing that Ireland should protect the environment even if it damages economic growth as opposed to those that believe that Ireland should encourage economic growth even if this damages the environment. The strength of gender effects evaporates when party is included, which suggests that gender informs one’s party choice more so than policy positions (see Supplementary Tables S4 and S5). Table 3 Regression of policy attitudes on demographics United Ireland Left–Right Abortion EU Fiscal Policy Environment Age: 45–59 (years) −0.12 (0.30) −0.44 (0.22)^ −0.07 (0.37) 0.01 (0.26) −0.36 (0.25) 0.13 (0.25) Age: 60 +  (years) 0.20 (0.34) −0.41 (0.27) 0.32 (0.40) 0.07 (0.28) −0.54 (0.29)^ −0.15 (0.31) Religiosity: 1–5 scale 0.05 (0.11) −0.09 (0.07) −0.32 (0.13)* 0.03 (0.10) 0.00 (0.08) 0.02 (0.10) Gender: Female 1.68 (0.38)** −0.11 (0.28) −0.88 (0.48)^ 0.23 (0.30) −0.14 (0.26) −0.45 (0.26)^ Party: Fine Gael 1.37 (0.35)** 0.75 (0.25)** −0.68 (0.37)^ 0.29 (0.23) 0.66 (0.28)* 0.24 (0.28) Party: Labour 1.13 (0.39)** −1.70 (0.28)** −3.62 (0.54)** 0.16 (0.32) −1.57 (0.35)** −0.44 (0.36) Party: Sinn Fein −3.25 (0.32)** −2.31 (0.34)** −3.00 (0.71)** −3.31 (0.46)** −2.92 (0.30)** −2.02 (0.38)** Party: Green 0.34 (0.46) −0.63 (0.30)* −0.76 (1.05) −0.96 (0.61) −0.72 (0.44)^ −2.89 (1.06)** Party: ULA 0.39 (0.76) −3.96 (0.52)** −6.60 (0.59)** −3.90 (0.76)** −3.82 (0.62)** −3.18 (0.65)** Party: Independent −0.03 (0.69) −1.13 (0.55)* −2.56 (0.90)** −2.56 (0.80)** −1.94 (0.49)** −0.80 (0.63) Year: 2011 −0.40 (0.28) 0.10 (0.18) 0.28 (0.30) −0.64 (0.21)** −0.56 (0.27)* 0.12 (0.23) Year: 2016 −0.78 (0.29)** 0.14 (0.19) −0.70 (0.39)^ −0.69 (0.23)** −0.48 (0.25)^ −0.45 (0.25)^ Intercept 3.55 (0.49)** 5.24 (0.31)** 8.97 (0.51)** 8.75 (0.35)** 5.78 (0.38)** 5.25 (0.41)** R2 0.36 0.48 0.39 0.45 0.42 0.27 United Ireland Left–Right Abortion EU Fiscal Policy Environment Age: 45–59 (years) −0.12 (0.30) −0.44 (0.22)^ −0.07 (0.37) 0.01 (0.26) −0.36 (0.25) 0.13 (0.25) Age: 60 +  (years) 0.20 (0.34) −0.41 (0.27) 0.32 (0.40) 0.07 (0.28) −0.54 (0.29)^ −0.15 (0.31) Religiosity: 1–5 scale 0.05 (0.11) −0.09 (0.07) −0.32 (0.13)* 0.03 (0.10) 0.00 (0.08) 0.02 (0.10) Gender: Female 1.68 (0.38)** −0.11 (0.28) −0.88 (0.48)^ 0.23 (0.30) −0.14 (0.26) −0.45 (0.26)^ Party: Fine Gael 1.37 (0.35)** 0.75 (0.25)** −0.68 (0.37)^ 0.29 (0.23) 0.66 (0.28)* 0.24 (0.28) Party: Labour 1.13 (0.39)** −1.70 (0.28)** −3.62 (0.54)** 0.16 (0.32) −1.57 (0.35)** −0.44 (0.36) Party: Sinn Fein −3.25 (0.32)** −2.31 (0.34)** −3.00 (0.71)** −3.31 (0.46)** −2.92 (0.30)** −2.02 (0.38)** Party: Green 0.34 (0.46) −0.63 (0.30)* −0.76 (1.05) −0.96 (0.61) −0.72 (0.44)^ −2.89 (1.06)** Party: ULA 0.39 (0.76) −3.96 (0.52)** −6.60 (0.59)** −3.90 (0.76)** −3.82 (0.62)** −3.18 (0.65)** Party: Independent −0.03 (0.69) −1.13 (0.55)* −2.56 (0.90)** −2.56 (0.80)** −1.94 (0.49)** −0.80 (0.63) Year: 2011 −0.40 (0.28) 0.10 (0.18) 0.28 (0.30) −0.64 (0.21)** −0.56 (0.27)* 0.12 (0.23) Year: 2016 −0.78 (0.29)** 0.14 (0.19) −0.70 (0.39)^ −0.69 (0.23)** −0.48 (0.25)^ −0.45 (0.25)^ Intercept 3.55 (0.49)** 5.24 (0.31)** 8.97 (0.51)** 8.75 (0.35)** 5.78 (0.38)** 5.25 (0.41)** R2 0.36 0.48 0.39 0.45 0.42 0.27 Base Categories: Male/45 or younger/Fianna Fáil/2007 Dáil; Standard errors in parentheses. Significant at **p ≤ 0.01; *p ≤ 0.05; ^p ≤ 0.10. Table 3 Regression of policy attitudes on demographics United Ireland Left–Right Abortion EU Fiscal Policy Environment Age: 45–59 (years) −0.12 (0.30) −0.44 (0.22)^ −0.07 (0.37) 0.01 (0.26) −0.36 (0.25) 0.13 (0.25) Age: 60 +  (years) 0.20 (0.34) −0.41 (0.27) 0.32 (0.40) 0.07 (0.28) −0.54 (0.29)^ −0.15 (0.31) Religiosity: 1–5 scale 0.05 (0.11) −0.09 (0.07) −0.32 (0.13)* 0.03 (0.10) 0.00 (0.08) 0.02 (0.10) Gender: Female 1.68 (0.38)** −0.11 (0.28) −0.88 (0.48)^ 0.23 (0.30) −0.14 (0.26) −0.45 (0.26)^ Party: Fine Gael 1.37 (0.35)** 0.75 (0.25)** −0.68 (0.37)^ 0.29 (0.23) 0.66 (0.28)* 0.24 (0.28) Party: Labour 1.13 (0.39)** −1.70 (0.28)** −3.62 (0.54)** 0.16 (0.32) −1.57 (0.35)** −0.44 (0.36) Party: Sinn Fein −3.25 (0.32)** −2.31 (0.34)** −3.00 (0.71)** −3.31 (0.46)** −2.92 (0.30)** −2.02 (0.38)** Party: Green 0.34 (0.46) −0.63 (0.30)* −0.76 (1.05) −0.96 (0.61) −0.72 (0.44)^ −2.89 (1.06)** Party: ULA 0.39 (0.76) −3.96 (0.52)** −6.60 (0.59)** −3.90 (0.76)** −3.82 (0.62)** −3.18 (0.65)** Party: Independent −0.03 (0.69) −1.13 (0.55)* −2.56 (0.90)** −2.56 (0.80)** −1.94 (0.49)** −0.80 (0.63) Year: 2011 −0.40 (0.28) 0.10 (0.18) 0.28 (0.30) −0.64 (0.21)** −0.56 (0.27)* 0.12 (0.23) Year: 2016 −0.78 (0.29)** 0.14 (0.19) −0.70 (0.39)^ −0.69 (0.23)** −0.48 (0.25)^ −0.45 (0.25)^ Intercept 3.55 (0.49)** 5.24 (0.31)** 8.97 (0.51)** 8.75 (0.35)** 5.78 (0.38)** 5.25 (0.41)** R2 0.36 0.48 0.39 0.45 0.42 0.27 United Ireland Left–Right Abortion EU Fiscal Policy Environment Age: 45–59 (years) −0.12 (0.30) −0.44 (0.22)^ −0.07 (0.37) 0.01 (0.26) −0.36 (0.25) 0.13 (0.25) Age: 60 +  (years) 0.20 (0.34) −0.41 (0.27) 0.32 (0.40) 0.07 (0.28) −0.54 (0.29)^ −0.15 (0.31) Religiosity: 1–5 scale 0.05 (0.11) −0.09 (0.07) −0.32 (0.13)* 0.03 (0.10) 0.00 (0.08) 0.02 (0.10) Gender: Female 1.68 (0.38)** −0.11 (0.28) −0.88 (0.48)^ 0.23 (0.30) −0.14 (0.26) −0.45 (0.26)^ Party: Fine Gael 1.37 (0.35)** 0.75 (0.25)** −0.68 (0.37)^ 0.29 (0.23) 0.66 (0.28)* 0.24 (0.28) Party: Labour 1.13 (0.39)** −1.70 (0.28)** −3.62 (0.54)** 0.16 (0.32) −1.57 (0.35)** −0.44 (0.36) Party: Sinn Fein −3.25 (0.32)** −2.31 (0.34)** −3.00 (0.71)** −3.31 (0.46)** −2.92 (0.30)** −2.02 (0.38)** Party: Green 0.34 (0.46) −0.63 (0.30)* −0.76 (1.05) −0.96 (0.61) −0.72 (0.44)^ −2.89 (1.06)** Party: ULA 0.39 (0.76) −3.96 (0.52)** −6.60 (0.59)** −3.90 (0.76)** −3.82 (0.62)** −3.18 (0.65)** Party: Independent −0.03 (0.69) −1.13 (0.55)* −2.56 (0.90)** −2.56 (0.80)** −1.94 (0.49)** −0.80 (0.63) Year: 2011 −0.40 (0.28) 0.10 (0.18) 0.28 (0.30) −0.64 (0.21)** −0.56 (0.27)* 0.12 (0.23) Year: 2016 −0.78 (0.29)** 0.14 (0.19) −0.70 (0.39)^ −0.69 (0.23)** −0.48 (0.25)^ −0.45 (0.25)^ Intercept 3.55 (0.49)** 5.24 (0.31)** 8.97 (0.51)** 8.75 (0.35)** 5.78 (0.38)** 5.25 (0.41)** R2 0.36 0.48 0.39 0.45 0.42 0.27 Base Categories: Male/45 or younger/Fianna Fáil/2007 Dáil; Standard errors in parentheses. Significant at **p ≤ 0.01; *p ≤ 0.05; ^p ≤ 0.10. To measure religiosity, a simple index was constructed that was comprised of the average of five religiosity variables from the TD survey: how frequently TDs reported praying, how frequently they reported attending church services, how religious or secular they considered themselves, how religiously traditional or liberal they believed they were and how important religion was in their home while growing up (on all these variables, 1 indicates the least religious and 5 is the most religious). As Table 3 (above) and Figure 2 show, the TDs with the highest levels of religiosity were on average one point further to the right and four points more pro-life than the most secular TDs. Perhaps not surprisingly, religiosity appears to have less influence in determining positions on other issues. Figure 2 View largeDownload slide Marginal effects of religiosity on policy attitudes (2007, 2011, 2016). Figure 2 View largeDownload slide Marginal effects of religiosity on policy attitudes (2007, 2011, 2016). Borderline evidence also suggested that those with higher levels of religiosity held slightly more supportive attitudes towards environmental protection. Thus, among the demographic variables analysed, religiosity and gender were the background variables that seem to shape TD ideological positions the most. Overall, TDs views’ do seem affected by their idiosyncratic personal experiences, which provides initial evidence for the hypothesis that TDs adopt some non-strategic policy stances at odds with the national party. 6.3 Local context Given the limited degree to which demographic factors explain MPs’ policy positions, I turn to analysing the importance of local constituency factors. As seen earlier, demographics appear to influence mostly non-economic policy areas, and even there, a great deal of variance in TDs’ attitudes is left unexplained. In addition to demographics, structural factors appear to play a role in explaining intra-party ideological variation. Given the high level of personalised engagement within the constituency, TDs unsurprisingly respond to some important characteristics of their local constituencies, which suggests that individual Irish TDs are responsive to local needs and sentiments. Given that levels of party membership are low among Irish voters and there exists an increasingly high percentage of ‘unattached’ voters in Ireland, candidates and parties alike must make a more concerted effort to be responsive to local interests (Marsh et al., 2008, pp. 59–79; Marsh, 2008). Evidence from the surveys also reveals a consequential divide between urban, rural and commuter constituencies. Table 4 shows that differences between urban, rural and commuter constituency TDs are largest, and become borderline statistically significant, on ‘post-materialist’ issues that do not easily fall into traditional left–right or pro-treaty/anti-treaty divisions. Table 4 Regression of policy attitudes on constituency type United Ireland Left–Right Abortion EU Fiscal Policy Environment Constituency: Commuter −0.47 (0.31) −0.22 (0.26) 0.65 (0.44) 0.24 (0.28) 0.10 (0.31) −0.23 (0.28) Constituency:Rural −0.02 (0.30) −0.09 (0.20) 0.79 (0.36)* 0.03 (0.25) 0.08 (0.23) 0.47 (0.23)* Party: Fine Gael 1.56 (0.34)** 0.81 (0.25)** −0.64 (0.35)^ 0.36 (0.24) 0.74 (0.27)** 0.28 (0.25) Party: Labour 1.66 (0.42)** −1.88 (0.28)** −3.70 (0.55)** 0.38 (0.31) −1.64 (0.34)** −0.42 (0.34) Party: Sinn Fein −3.05 (0.27)** −2.41 (0.30)** −3.22 (0.65)** −3.25 (0.45)** −2.88 (0.29)** −1.61 (0.34)** Party: Green 0.26 (0.44) −1.08 (0.43)* −1.06 (0.72) −0.33 (0.69) −1.03 (0.49)* −3.32 (0.78)** Party: ULA 0.56 (0.94) −3.92 (0.51)** −5.92 (0.62)** −3.79 (0.73)** −3.89 (0.55)** −2.72 (0.64)** Party: Independent 0.28 (0.71) −1.18 (0.53)* −2.44 (0.93)** −2.26 (0.78)** −1.90 (0.49)** −0.82 (0.61) Year: 2011 −0.43 (0.28) 0.15 (0.18) 0.03 (0.30) −0.66 (0.21)** −0.50 (0.26)^ 0.03 (0.23) Year: 2016 −0.56 (0.27)* 0.14 (0.18) −0.88 (0.34)* −0.77 (0.22)** −0.43 (0.23)^ −0.30 (0.22) Intercept 3.93 (0.34)** 4.72 (0.20)** 7.65 (0.34)** 8.75 (0.24)** 5.33 (0.27)** 5.15 (0.23)** R2 0.33 0.48 0.37 0.45 0.43 0.25 United Ireland Left–Right Abortion EU Fiscal Policy Environment Constituency: Commuter −0.47 (0.31) −0.22 (0.26) 0.65 (0.44) 0.24 (0.28) 0.10 (0.31) −0.23 (0.28) Constituency:Rural −0.02 (0.30) −0.09 (0.20) 0.79 (0.36)* 0.03 (0.25) 0.08 (0.23) 0.47 (0.23)* Party: Fine Gael 1.56 (0.34)** 0.81 (0.25)** −0.64 (0.35)^ 0.36 (0.24) 0.74 (0.27)** 0.28 (0.25) Party: Labour 1.66 (0.42)** −1.88 (0.28)** −3.70 (0.55)** 0.38 (0.31) −1.64 (0.34)** −0.42 (0.34) Party: Sinn Fein −3.05 (0.27)** −2.41 (0.30)** −3.22 (0.65)** −3.25 (0.45)** −2.88 (0.29)** −1.61 (0.34)** Party: Green 0.26 (0.44) −1.08 (0.43)* −1.06 (0.72) −0.33 (0.69) −1.03 (0.49)* −3.32 (0.78)** Party: ULA 0.56 (0.94) −3.92 (0.51)** −5.92 (0.62)** −3.79 (0.73)** −3.89 (0.55)** −2.72 (0.64)** Party: Independent 0.28 (0.71) −1.18 (0.53)* −2.44 (0.93)** −2.26 (0.78)** −1.90 (0.49)** −0.82 (0.61) Year: 2011 −0.43 (0.28) 0.15 (0.18) 0.03 (0.30) −0.66 (0.21)** −0.50 (0.26)^ 0.03 (0.23) Year: 2016 −0.56 (0.27)* 0.14 (0.18) −0.88 (0.34)* −0.77 (0.22)** −0.43 (0.23)^ −0.30 (0.22) Intercept 3.93 (0.34)** 4.72 (0.20)** 7.65 (0.34)** 8.75 (0.24)** 5.33 (0.27)** 5.15 (0.23)** R2 0.33 0.48 0.37 0.45 0.43 0.25 Base Categories: Urban constituency/Fianna Fáil/2007 Dáil; Standard errors in parentheses. Significant at **p ≤ 0.01; *p ≤ 0.05; ^p ≤ 0.10. Table 4 Regression of policy attitudes on constituency type United Ireland Left–Right Abortion EU Fiscal Policy Environment Constituency: Commuter −0.47 (0.31) −0.22 (0.26) 0.65 (0.44) 0.24 (0.28) 0.10 (0.31) −0.23 (0.28) Constituency:Rural −0.02 (0.30) −0.09 (0.20) 0.79 (0.36)* 0.03 (0.25) 0.08 (0.23) 0.47 (0.23)* Party: Fine Gael 1.56 (0.34)** 0.81 (0.25)** −0.64 (0.35)^ 0.36 (0.24) 0.74 (0.27)** 0.28 (0.25) Party: Labour 1.66 (0.42)** −1.88 (0.28)** −3.70 (0.55)** 0.38 (0.31) −1.64 (0.34)** −0.42 (0.34) Party: Sinn Fein −3.05 (0.27)** −2.41 (0.30)** −3.22 (0.65)** −3.25 (0.45)** −2.88 (0.29)** −1.61 (0.34)** Party: Green 0.26 (0.44) −1.08 (0.43)* −1.06 (0.72) −0.33 (0.69) −1.03 (0.49)* −3.32 (0.78)** Party: ULA 0.56 (0.94) −3.92 (0.51)** −5.92 (0.62)** −3.79 (0.73)** −3.89 (0.55)** −2.72 (0.64)** Party: Independent 0.28 (0.71) −1.18 (0.53)* −2.44 (0.93)** −2.26 (0.78)** −1.90 (0.49)** −0.82 (0.61) Year: 2011 −0.43 (0.28) 0.15 (0.18) 0.03 (0.30) −0.66 (0.21)** −0.50 (0.26)^ 0.03 (0.23) Year: 2016 −0.56 (0.27)* 0.14 (0.18) −0.88 (0.34)* −0.77 (0.22)** −0.43 (0.23)^ −0.30 (0.22) Intercept 3.93 (0.34)** 4.72 (0.20)** 7.65 (0.34)** 8.75 (0.24)** 5.33 (0.27)** 5.15 (0.23)** R2 0.33 0.48 0.37 0.45 0.43 0.25 United Ireland Left–Right Abortion EU Fiscal Policy Environment Constituency: Commuter −0.47 (0.31) −0.22 (0.26) 0.65 (0.44) 0.24 (0.28) 0.10 (0.31) −0.23 (0.28) Constituency:Rural −0.02 (0.30) −0.09 (0.20) 0.79 (0.36)* 0.03 (0.25) 0.08 (0.23) 0.47 (0.23)* Party: Fine Gael 1.56 (0.34)** 0.81 (0.25)** −0.64 (0.35)^ 0.36 (0.24) 0.74 (0.27)** 0.28 (0.25) Party: Labour 1.66 (0.42)** −1.88 (0.28)** −3.70 (0.55)** 0.38 (0.31) −1.64 (0.34)** −0.42 (0.34) Party: Sinn Fein −3.05 (0.27)** −2.41 (0.30)** −3.22 (0.65)** −3.25 (0.45)** −2.88 (0.29)** −1.61 (0.34)** Party: Green 0.26 (0.44) −1.08 (0.43)* −1.06 (0.72) −0.33 (0.69) −1.03 (0.49)* −3.32 (0.78)** Party: ULA 0.56 (0.94) −3.92 (0.51)** −5.92 (0.62)** −3.79 (0.73)** −3.89 (0.55)** −2.72 (0.64)** Party: Independent 0.28 (0.71) −1.18 (0.53)* −2.44 (0.93)** −2.26 (0.78)** −1.90 (0.49)** −0.82 (0.61) Year: 2011 −0.43 (0.28) 0.15 (0.18) 0.03 (0.30) −0.66 (0.21)** −0.50 (0.26)^ 0.03 (0.23) Year: 2016 −0.56 (0.27)* 0.14 (0.18) −0.88 (0.34)* −0.77 (0.22)** −0.43 (0.23)^ −0.30 (0.22) Intercept 3.93 (0.34)** 4.72 (0.20)** 7.65 (0.34)** 8.75 (0.24)** 5.33 (0.27)** 5.15 (0.23)** R2 0.33 0.48 0.37 0.45 0.43 0.25 Base Categories: Urban constituency/Fianna Fáil/2007 Dáil; Standard errors in parentheses. Significant at **p ≤ 0.01; *p ≤ 0.05; ^p ≤ 0.10. This geographic variation is indicative of TDs exercising personal discretion to adapt their ideological appeals based on their specific electoral context (Farrell, 2001, pp. 121–152). Citizens living in urban, rural and commuter constituencies are likely to hold different positions given their differing life contexts. Although constituency level opinion polls are not available in Ireland, elements of certain constituency attitudes can be gleaned using results from the 2002 right-to-life referendum and 2009 Lisbon Treaty (EU) referendum. Not surprisingly, legislators in constituencies where referendum results were more anti-abortion also tend to take stronger personal stances against abortion. A similar effect may exist regarding the EU, but it narrowly missed attaining statistical significance. Table 5 shows a regression analysis of the relationship between constituency public opinion and TDs’ self-placement on abortion. Table 5 Regression of TD attitudes on constituency referenda results Abortion Reg.abort.hlm02 EU reg.eu.hlmlisbon2 Vote share in 2002 right to life referendum 0.06 (0.016)** – Vote share in 2009 Lisbon Treat (EU) referendum – 0.023 (0.019) Party: Fine Gael −0.73 (0.40) 0.23 (0.26) Party: Labour −3.79 (0.53)** 0.17 (0.34) Party: Sinn Fein −3.43 (0.59)** −3.17 (0.40)** Party: Green −0.61 (1.17) −0.54 (0.71) Party: ULA −5.76 (0.79)** −3.64 (0.51)** Party: Independent −3.01 (0.66)** −2.07 (0.44)** Year: 2011 0.28 (0.30) −0.55 (0.19)** Year: 2016 −0.76 (0.32)* −0.67 (0.20)** Intercept 5.20 (0.92)** 7.24 (1.25)** Conditional R2 (Nakagawa & Schielzeth) 0.71 0.77 Abortion Reg.abort.hlm02 EU reg.eu.hlmlisbon2 Vote share in 2002 right to life referendum 0.06 (0.016)** – Vote share in 2009 Lisbon Treat (EU) referendum – 0.023 (0.019) Party: Fine Gael −0.73 (0.40) 0.23 (0.26) Party: Labour −3.79 (0.53)** 0.17 (0.34) Party: Sinn Fein −3.43 (0.59)** −3.17 (0.40)** Party: Green −0.61 (1.17) −0.54 (0.71) Party: ULA −5.76 (0.79)** −3.64 (0.51)** Party: Independent −3.01 (0.66)** −2.07 (0.44)** Year: 2011 0.28 (0.30) −0.55 (0.19)** Year: 2016 −0.76 (0.32)* −0.67 (0.20)** Intercept 5.20 (0.92)** 7.24 (1.25)** Conditional R2 (Nakagawa & Schielzeth) 0.71 0.77 Base Category: Fianna Fáil/2007 Dáil; Standard errors in parentheses. Significant at **p ≤ 0.01; *p ≤ 0.05; ^p ≤ 0.10;  ^ = .10, * = .05, ** = .01. Table 5 Regression of TD attitudes on constituency referenda results Abortion Reg.abort.hlm02 EU reg.eu.hlmlisbon2 Vote share in 2002 right to life referendum 0.06 (0.016)** – Vote share in 2009 Lisbon Treat (EU) referendum – 0.023 (0.019) Party: Fine Gael −0.73 (0.40) 0.23 (0.26) Party: Labour −3.79 (0.53)** 0.17 (0.34) Party: Sinn Fein −3.43 (0.59)** −3.17 (0.40)** Party: Green −0.61 (1.17) −0.54 (0.71) Party: ULA −5.76 (0.79)** −3.64 (0.51)** Party: Independent −3.01 (0.66)** −2.07 (0.44)** Year: 2011 0.28 (0.30) −0.55 (0.19)** Year: 2016 −0.76 (0.32)* −0.67 (0.20)** Intercept 5.20 (0.92)** 7.24 (1.25)** Conditional R2 (Nakagawa & Schielzeth) 0.71 0.77 Abortion Reg.abort.hlm02 EU reg.eu.hlmlisbon2 Vote share in 2002 right to life referendum 0.06 (0.016)** – Vote share in 2009 Lisbon Treat (EU) referendum – 0.023 (0.019) Party: Fine Gael −0.73 (0.40) 0.23 (0.26) Party: Labour −3.79 (0.53)** 0.17 (0.34) Party: Sinn Fein −3.43 (0.59)** −3.17 (0.40)** Party: Green −0.61 (1.17) −0.54 (0.71) Party: ULA −5.76 (0.79)** −3.64 (0.51)** Party: Independent −3.01 (0.66)** −2.07 (0.44)** Year: 2011 0.28 (0.30) −0.55 (0.19)** Year: 2016 −0.76 (0.32)* −0.67 (0.20)** Intercept 5.20 (0.92)** 7.24 (1.25)** Conditional R2 (Nakagawa & Schielzeth) 0.71 0.77 Base Category: Fianna Fáil/2007 Dáil; Standard errors in parentheses. Significant at **p ≤ 0.01; *p ≤ 0.05; ^p ≤ 0.10;  ^ = .10, * = .05, ** = .01. Supplementary Figure S1 includes visual plots which estimate the relationship between referenda voting and average TD opinion in a constituency in the 2002 right-to-life referendum and the 2009 Lisbon Treaty that illustrate these same results. The degree of political competition that TDs face also appears to have some relationship with the local policy position adopted by individual TDs. For example, TDs who are more electorally secure are moderately less pro-Europe and tend to follow the party line more closely on Northern Ireland and the EU. TDs’ electoral marginality was measured by looking at what fraction of a quota (the number of votes divided by number of seats) each received on the first vote count. This measure, although limited in many ways, was selected because it is used by political practitioners and pollsters themselves and is, therefore, likely to reflect TDs’ own understanding of how electorally marginal they are. TDs who receive nearly a full quota are quite likely to be elected, while those who receive less than half a quota are rarely elected. There are geographic and party differences with this measure because rural candidates generally must earn a higher quota to be elected than those in urban areas and candidates with running mates tend to earn lower quotas than those who do not have a running mate. Nevertheless, this measure provides a proxy to test if there is any difference in ideological positions based on whether one is elected quite handily compared to those that are the last to be elected. Although results show that candidates who follow the party line, or deviate towards the centre, are more successful electorally, that evidence is limited. As Supplementary Table S6 shows, no evidence was found that electorally marginal candidates were further left or right. More importantly, the difference between a TD’s position, and his or her placement of their own party, produced significant bivariate relationships on European integration and Northern Ireland. This can be seen in analysis of variance (ANOVA) tests (Table 6) and bivariate regressions (Supplementary Table S7). Table 6 Mean quotas received on first count (2007, 2011, 2016; by TDs’ distance from party) N. Ireland* Left–Right Abortion EU* Fiscal Environment 0-point difference 0.76 (0.01) 0.78 (0.02) 0.80 (0.02) 0.76 (0.02) 0.77 (0.02) 0.76 (0.02) 1-point difference 0.79 (0.04) 0.74 (0.02) 0.75 (0.03) 0.80 (0.03) 0.75 (0.02) 0.77 (0.02) 2-point difference 0.83 (0.04) 0.76 (0.03) 0.76 (0.03) 0.81 (0.05) 0.73 (0.03) 0.80 (0.04) >2-point difference 0.68 (0.02) 0.74 (0.04) 0.72 (0.02) 0.68 (0.03) 0.77 (0.04) 0.72 (0.03) ANOVA p-value 0.01 0.56 0.12 0.04 0.60 0.45 N. Ireland* Left–Right Abortion EU* Fiscal Environment 0-point difference 0.76 (0.01) 0.78 (0.02) 0.80 (0.02) 0.76 (0.02) 0.77 (0.02) 0.76 (0.02) 1-point difference 0.79 (0.04) 0.74 (0.02) 0.75 (0.03) 0.80 (0.03) 0.75 (0.02) 0.77 (0.02) 2-point difference 0.83 (0.04) 0.76 (0.03) 0.76 (0.03) 0.81 (0.05) 0.73 (0.03) 0.80 (0.04) >2-point difference 0.68 (0.02) 0.74 (0.04) 0.72 (0.02) 0.68 (0.03) 0.77 (0.04) 0.72 (0.03) ANOVA p-value 0.01 0.56 0.12 0.04 0.60 0.45 Standard errors in parentheses. Significant at **p ≤ 0.05; *p ≤ 0.10;  ^ = .10, * = .05, ** = .01. Table 6 Mean quotas received on first count (2007, 2011, 2016; by TDs’ distance from party) N. Ireland* Left–Right Abortion EU* Fiscal Environment 0-point difference 0.76 (0.01) 0.78 (0.02) 0.80 (0.02) 0.76 (0.02) 0.77 (0.02) 0.76 (0.02) 1-point difference 0.79 (0.04) 0.74 (0.02) 0.75 (0.03) 0.80 (0.03) 0.75 (0.02) 0.77 (0.02) 2-point difference 0.83 (0.04) 0.76 (0.03) 0.76 (0.03) 0.81 (0.05) 0.73 (0.03) 0.80 (0.04) >2-point difference 0.68 (0.02) 0.74 (0.04) 0.72 (0.02) 0.68 (0.03) 0.77 (0.04) 0.72 (0.03) ANOVA p-value 0.01 0.56 0.12 0.04 0.60 0.45 N. Ireland* Left–Right Abortion EU* Fiscal Environment 0-point difference 0.76 (0.01) 0.78 (0.02) 0.80 (0.02) 0.76 (0.02) 0.77 (0.02) 0.76 (0.02) 1-point difference 0.79 (0.04) 0.74 (0.02) 0.75 (0.03) 0.80 (0.03) 0.75 (0.02) 0.77 (0.02) 2-point difference 0.83 (0.04) 0.76 (0.03) 0.76 (0.03) 0.81 (0.05) 0.73 (0.03) 0.80 (0.04) >2-point difference 0.68 (0.02) 0.74 (0.04) 0.72 (0.02) 0.68 (0.03) 0.77 (0.04) 0.72 (0.03) ANOVA p-value 0.01 0.56 0.12 0.04 0.60 0.45 Standard errors in parentheses. Significant at **p ≤ 0.05; *p ≤ 0.10;  ^ = .10, * = .05, ** = .01. For example, TDs who shared a position with their party on Northern Ireland won 76% of the first preference votes necessary to win a seat, while TDs who diverged from their party by two or more points on this issue won 68% of the first-preference votes needed to win seat. However, once party and year were controlled for in an ordered logistic regression model, the relationship lost significance (see Supplementary Table S9). An alternative specification for the TDs’ positioning measured whether TDs were more centrist or more extreme than their parties. Bivariate logistic regression findings (Supplementary Table S8) suggest that TDs who adopted more extreme positions than their parties on abortion and possibly the EU paid an electoral price, while those who adopted more centrist positions than their parties were successful. (However, bivariate ANOVA tests (Supplementary Table S10) and multivariate regressions (Supplementary Table S11) failed to adequately confirm these findings, though (Supplementary Table S10) provides somewhat confirmatory evidence for EU policy position.) Regardless of whether they were electorally marginal candidates, those TDs who faced intra-party competitors in their constituency also took more critical stances towards the EU, breaking with their generally more pro-European parties as Table 7 shows. Table 7 Regression of policy attitudes on intra-party competition United Ireland Left–Right Abortion EU Fiscal Policy Environment Intra-party competitor in district −0.01 (0.25) 0.04 (0.18) −0.38 (0.367) −0.34 (0.23) 0.01 (0.24) −0.20 (0.24) Party: Fine Gael 1.55 (0.35)** 0.82 (0.25)** −0.72 (0.36)* 0.37 (0.24) 0.73 (0.27)** 0.21 (0.26) Party: Labour 1.71 (0.43)** −1.85 (0.28)** −3.87 (0.53)** 0.46 (0.30) −1.68 (0.32)** −0.48 (0.35) Party: Sinn Fein −3.01 (0.29)** −2.40 (0.31)** −3.18 (0.69)* −3.08 (0.44)** −2.89 (0.28)** −1.56 (0.40)** Party: Green 0.39 (0.45) −1.03 (0.45)* −1.08 (0.87) −0.17 (0.74) −1.07 (0.51)* −3.20 (0.86)** Party: ULA 0.66 (1.01) −3.87 (0.49)** −6.15 (0.68)** −3.61 (0.73)** −3.94 (0.54)** −2.76 (0.68)** Party: Independent 0.35 (0.70) −1.18 (0.56)* −2.28 (0.95)* −2.04 (0.79)* −1.91 (0.48)** −0.66 (0.66) Year: 2011 −0.45 (0.28) 0.13 (0.18) 0.16 (0.30) −0.65 (0.21)** −0.48 (0.26)^ 0.10 (0.23) Year: 2016 −0.58 (0.28)* 0.13 (0.18) −0.78 (0.35)* −0.73 (0.23)** −0.42 (0.24)^ −0.26 (0.22) Intercept 3.81 (0.28)** 4.62 (0.15)** 8.17 (0.28)** 8.87 (0.18)** 5.38 (0.22)** 5.30 (0.19)** R2 0.32 0.48 0.36 0.46 0.43 0.23 United Ireland Left–Right Abortion EU Fiscal Policy Environment Intra-party competitor in district −0.01 (0.25) 0.04 (0.18) −0.38 (0.367) −0.34 (0.23) 0.01 (0.24) −0.20 (0.24) Party: Fine Gael 1.55 (0.35)** 0.82 (0.25)** −0.72 (0.36)* 0.37 (0.24) 0.73 (0.27)** 0.21 (0.26) Party: Labour 1.71 (0.43)** −1.85 (0.28)** −3.87 (0.53)** 0.46 (0.30) −1.68 (0.32)** −0.48 (0.35) Party: Sinn Fein −3.01 (0.29)** −2.40 (0.31)** −3.18 (0.69)* −3.08 (0.44)** −2.89 (0.28)** −1.56 (0.40)** Party: Green 0.39 (0.45) −1.03 (0.45)* −1.08 (0.87) −0.17 (0.74) −1.07 (0.51)* −3.20 (0.86)** Party: ULA 0.66 (1.01) −3.87 (0.49)** −6.15 (0.68)** −3.61 (0.73)** −3.94 (0.54)** −2.76 (0.68)** Party: Independent 0.35 (0.70) −1.18 (0.56)* −2.28 (0.95)* −2.04 (0.79)* −1.91 (0.48)** −0.66 (0.66) Year: 2011 −0.45 (0.28) 0.13 (0.18) 0.16 (0.30) −0.65 (0.21)** −0.48 (0.26)^ 0.10 (0.23) Year: 2016 −0.58 (0.28)* 0.13 (0.18) −0.78 (0.35)* −0.73 (0.23)** −0.42 (0.24)^ −0.26 (0.22) Intercept 3.81 (0.28)** 4.62 (0.15)** 8.17 (0.28)** 8.87 (0.18)** 5.38 (0.22)** 5.30 (0.19)** R2 0.32 0.48 0.36 0.46 0.43 0.23 Base Categories: Urban constituency/Fianna Fáil/2007 Dáil; Standard errors in parentheses. Significant at **p ≤ 0.01; *p ≤ 0.05; ^p ≤ 0.10. Table 7 Regression of policy attitudes on intra-party competition United Ireland Left–Right Abortion EU Fiscal Policy Environment Intra-party competitor in district −0.01 (0.25) 0.04 (0.18) −0.38 (0.367) −0.34 (0.23) 0.01 (0.24) −0.20 (0.24) Party: Fine Gael 1.55 (0.35)** 0.82 (0.25)** −0.72 (0.36)* 0.37 (0.24) 0.73 (0.27)** 0.21 (0.26) Party: Labour 1.71 (0.43)** −1.85 (0.28)** −3.87 (0.53)** 0.46 (0.30) −1.68 (0.32)** −0.48 (0.35) Party: Sinn Fein −3.01 (0.29)** −2.40 (0.31)** −3.18 (0.69)* −3.08 (0.44)** −2.89 (0.28)** −1.56 (0.40)** Party: Green 0.39 (0.45) −1.03 (0.45)* −1.08 (0.87) −0.17 (0.74) −1.07 (0.51)* −3.20 (0.86)** Party: ULA 0.66 (1.01) −3.87 (0.49)** −6.15 (0.68)** −3.61 (0.73)** −3.94 (0.54)** −2.76 (0.68)** Party: Independent 0.35 (0.70) −1.18 (0.56)* −2.28 (0.95)* −2.04 (0.79)* −1.91 (0.48)** −0.66 (0.66) Year: 2011 −0.45 (0.28) 0.13 (0.18) 0.16 (0.30) −0.65 (0.21)** −0.48 (0.26)^ 0.10 (0.23) Year: 2016 −0.58 (0.28)* 0.13 (0.18) −0.78 (0.35)* −0.73 (0.23)** −0.42 (0.24)^ −0.26 (0.22) Intercept 3.81 (0.28)** 4.62 (0.15)** 8.17 (0.28)** 8.87 (0.18)** 5.38 (0.22)** 5.30 (0.19)** R2 0.32 0.48 0.36 0.46 0.43 0.23 United Ireland Left–Right Abortion EU Fiscal Policy Environment Intra-party competitor in district −0.01 (0.25) 0.04 (0.18) −0.38 (0.367) −0.34 (0.23) 0.01 (0.24) −0.20 (0.24) Party: Fine Gael 1.55 (0.35)** 0.82 (0.25)** −0.72 (0.36)* 0.37 (0.24) 0.73 (0.27)** 0.21 (0.26) Party: Labour 1.71 (0.43)** −1.85 (0.28)** −3.87 (0.53)** 0.46 (0.30) −1.68 (0.32)** −0.48 (0.35) Party: Sinn Fein −3.01 (0.29)** −2.40 (0.31)** −3.18 (0.69)* −3.08 (0.44)** −2.89 (0.28)** −1.56 (0.40)** Party: Green 0.39 (0.45) −1.03 (0.45)* −1.08 (0.87) −0.17 (0.74) −1.07 (0.51)* −3.20 (0.86)** Party: ULA 0.66 (1.01) −3.87 (0.49)** −6.15 (0.68)** −3.61 (0.73)** −3.94 (0.54)** −2.76 (0.68)** Party: Independent 0.35 (0.70) −1.18 (0.56)* −2.28 (0.95)* −2.04 (0.79)* −1.91 (0.48)** −0.66 (0.66) Year: 2011 −0.45 (0.28) 0.13 (0.18) 0.16 (0.30) −0.65 (0.21)** −0.48 (0.26)^ 0.10 (0.23) Year: 2016 −0.58 (0.28)* 0.13 (0.18) −0.78 (0.35)* −0.73 (0.23)** −0.42 (0.24)^ −0.26 (0.22) Intercept 3.81 (0.28)** 4.62 (0.15)** 8.17 (0.28)** 8.87 (0.18)** 5.38 (0.22)** 5.30 (0.19)** R2 0.32 0.48 0.36 0.46 0.43 0.23 Base Categories: Urban constituency/Fianna Fáil/2007 Dáil; Standard errors in parentheses. Significant at **p ≤ 0.01; *p ≤ 0.05; ^p ≤ 0.10. It is unclear whether TDs feel they need to offer more extreme positions because they are marginal candidates, or whether TDs who diverge from their parties tend to suffer electorally. Ultimately, TDs do appear to adopt locally varied positions for strategic reasons albeit the evidence is not overwhelming. 7. Conclusion This study seeks to add to our understanding of the behaviour of franchise parties by focusing on adaptations within the ideological, rather than the organisational, domain. A key attribute of franchise, or franchise-like, parties is their ability to reconcile competing electoral imperatives at both the national and local levels. The degree to which a party conforms to a franchise-like model, the party must create a strong national organisation that can manage and control party strategies and overall brand identity. This becomes especially imperative in a highly volatile electoral environment. However, they must simultaneously find ways to be sufficiently nimble to secure support (from voters and high-quality candidates) within diverse local contexts. While a number of studies have focused on the organisational mechanisms parties employ to function as a franchise party, this study has sought to expand the utility of the concept of franchise parties by offering an analysis of how Irish political entrepreneurs from an array of parties vary their ideological appeals, either to enhance their attractiveness or pursue personal beliefs and agendas once in office. I have argued that, in the Irish case, although a wide variety of parties evince these behaviours, the larger parties exhibit measurably more franchise-like behaviour than the non-major parties. To provide evidence for the analysis, the article employs original quantitative and qualitative data from three longitudinal parliamentary surveys designed to examine how candidates compete ideologically. It finds that candidates from Ireland’s larger parties in particular take advantage of the flexibility of the franchise model to adapt their programmatic appeals, which in turn enhances the electoral chances of the party as a whole. Specifically, I examined how party affiliation, demographic and family background, and local context shape MPs’ positions on policy issues. Not surprisingly, party affiliation explains individual politicians’ ideological positions more so than these other social and environmental factors. Demographic factors such as age and religiosity, or whether a TD lives in an urban or rural constituency, shape party affiliation, but the strength of these factors lessens when party is included in our statistical tests. Nevertheless, the findings confirm that even though party does influence policy positions, individual Irish politicians enjoy considerable flexibility and autonomy to adapt policy positions at variance from the national platform during elections. As the franchise model implies, this strategy is left unchallenged by party leaders at the national level because the party maintains overall control of policy and strategy at the national level, but party leaders also recognise that individual candidates are best suited to know the set of policy offerings that will result in maximising local support. Supplementary Data Supplementary data are available at Parliamentary Affairs online. Footnotes 1 ‘What to Ask Your Candidate about Health and Medical Services’, Irish Times (1973, February 19). Acknowledgements I would like to thank several colleagues for their generous and thoughtful feedback on this paper, including Ken Carty, Eoin O’Malley, Jaimie Bleck, Karrie Koesel, Mike Hoffman, Lauren Honig, and especially Tim Scully. Ben Mainwaring and Steve Ponisciak provided invaluable assistance with the quantitative analysis. I would also like to thank the three reviewers for their thoughtful and thorough comments which helped make this a stronger article. Any errors are my sole responsibility. The author reports no other declarations of interest. Funding This work was supported by internal grants at the University of Notre Dame by the Kellogg Institute for International Studies and the Nanovic Institute for European Studies. Conflict of interest No conflict of interest to disclose. References Byrne K. , O'Malley E. ( 2012 ) ‘ Politics with Hidden Bases: Unearthing Party System's Deep Roots’, British Journal of Politics and International Relations , 14 , 613 – 629 . Google Scholar CrossRef Search ADS Carey J. M. , Shugart M. S. ( 1995 ) ‘Incentives to Cultivate a Personal Vote: A Rank Ordering of Electoral Formulas’ , Electoral Studies , 14 , 417 – 439 . Google Scholar CrossRef Search ADS Carty R. K. ( 2002 ) ‘ The Politics of Tecumseh Corners: Canadian Political Parties as Franchise Organizations’, Canadian Journal of Political Science , 35 , 723 – 745 . Google Scholar CrossRef Search ADS Carty R. K. ( 2004 ) ‘ Parties as Franchise Systems: The Stratarchical Organizational Imperative’, Party Politics , 10 , 5 – 24 . Google Scholar CrossRef Search ADS Chubb B. ( 1963 ) ‘ Going about Persecuting Civil Servants: The Role of the Irish Parliamentary Representative’, Political Studies , 11 , 272 – 286 . Google Scholar CrossRef Search ADS Courtney M. ( 2015 ) ‘ Social Background and Intra-Party Attitudes in Ireland’, Irish Political Studies , 30 , 178 – 198 . Google Scholar CrossRef Search ADS Farrell D. M. ( 2001 ) Electoral Systems: A Comparative Introduction , Basingstoke, Hampshire , Palgrave MacMillan . Farrell D. M. , Mair P. , Ó Muineacháin S. , Wall M. ( 2015 ) ‘Courting, But Not Always Serving: Perverted Burkeanism and the Puzzle of Irish Parliamentary Cohesion’. In Johnston R. , Sharman C. (eds) Parties & Party Systems: Structure and Context , Vancouver , University of British Columbia Press , pp. 92 – 107 . Groseclose T. ( 2001 ) ‘ A Model of Candidate Location When One Candidate Has a Valence Advantage’, American Journal of Political Science , 45 , 862 – 886 . Google Scholar CrossRef Search ADS Katz R. , Mair P. ( 1995 ) ‘ Changing Models of Party Organization and Party Democracy: The Emergence of the Cartel Party’, Party Politics , 1 , 5 – 28 . Google Scholar CrossRef Search ADS Kavanagh A. P. ( 2015 ) ‘ An End to “Civil War Politics”? The Radically Re-shaped Political Landscape of Post-Crash Ireland’, Electoral Studies , 38 , 71 – 81 . Google Scholar CrossRef Search ADS Marsh M. , Farrell D. M. , McElroy G. (eds) ( 2017 ) A Conservative Revolution? Electoral Change in Twenty-First Century Ireland , Oxford : Oxford University Press . Google Scholar CrossRef Search ADS Marsh M. , G McElroy . ( 2016 ) ‘Voting Behaviour: Continuing De-alignment’. In Gallagher M. , Marsh M. (eds) How Ireland Voted 2016: The Election That Nobody Won , Cham, Switzerland , Palgrave Macmillan , pp. 159 – 184 . Google Scholar CrossRef Search ADS Marsh M. , Sinnott R. , Garry J. , Kennedy F. ( 2008 ) The Irish Voter: The Nature of Electoral Competition in the Republic of Ireland , Manchester , Manchester University Press . Marsh M. ( 2008 ) ‘Explanations for Party Choice’. In Gallagher M. , Marsh M. (eds) How Ireland Voted 2007: The Full Story of Ireland’s General Election , Basingstoke, Hampshire , Palgrave Macmillan , pp. 105 – 131 . Google Scholar CrossRef Search ADS McGraw S. ( 2015 ) How Parties Win: Shaping the Irish Political Arena , Ann Arbor , Michigan University Press . Google Scholar CrossRef Search ADS Murphy G, ( 2016 ) Electoral Competition in Ireland Since 1987: The Politics of Triumph and Despair , Manchester , Manchester University Press . Google Scholar CrossRef Search ADS Reidy T. , White T. J. ( 2017 ) ‘Political Transformation in Ireland: Boom, Bust and Beyond’ , Eire-Ireland , 52 , 101 – 123 . Google Scholar CrossRef Search ADS Shu M. ( 2003 ) Cope with Two-Dimensional Cleavage Structure: Party Politics in Referendums on European Integration, Paper presented at Workshop 19, ‘Cleavage Development: Causes and Consequences’, ECPR Joint Sessions, Edinburgh, 28 March–2 April 2003. Sinnott R. ( 1995 ) Irish Voters Decide: Voting Behaviour in Elections and Referendums Since 1918 , Manchester , Manchester University Press . Tavits M. ( 2007 ) ‘ Principle vs. Pragmatism: Policy Shifts and Political Competition’, American Journal of Political Science , 51 , 151 – 165 . Google Scholar CrossRef Search ADS van Vanno C. M. C. , Malka R. I. , Depauw S. , Hazan R. Y. , Andeweg R. B. ( 2014 ) Agreement, Loyalty, and Discipline: A Sequential Approach to Party Unity , Oxford , Oxford University Press . Wolinetz S. B. ( 2015 ) ‘Franchising the Franchise Party: How Far Can a New Concept Travel’. In Johnston R. , Sharman C. (eds) Parties & Party Systems: Structure and Context , Vancouver , University of British Columbia Press , pp. 72 – 91 . © The Author 2017. Published by Oxford University Press on behalf of the Hansard Society; all rights reserved. For Permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Parliamentary AffairsOxford University Press

Published: Dec 20, 2017

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off