What Determines Parties’ Choice of Incumbent-Renomination Methods? the Case of the UK Labour Party, 1979–2019

ABSTRACT

The UK Labour Party changed its incumbent-renomination process six times between 1979 and 2019. Using Labour as an influential case, I propose a two-part theory explaining parties’ choice of incumbent-renomination methods: who controls the party, and whether they believe MPs are bound by dictates of the parliamentary leadership or the extra-parliamentary party. Renomination is central to intra-party candidate selection affecting the functioning of parties, legislatures, governments and oppositions. Findings expand our understanding of intra-party organisational dynamics and incumbent-renomination processes.

KEYWORDS:

• Incumbent renomination
• candidate selection
• intra-party democracy

Introduction

Selecting parliamentary candidates is a key function of political parties in established democracies. Once selected, candidates determine much of parties’ nature and characteristics and influence legislatures, governments, and oppositions (Norris & Lovenduski, 1995). As part of pre-election candidate selection, parties also face a decision about their incumbent members of parliament (MPs). Most parties’ renomination processes are automatic (or practically automatic),¹ but some make it more difficult, using various methods to decide whether incumbents will stand for the party in an upcoming general election. For example, the Argentinian Radical Civil Union asked incumbents to gain two-thirds support in intra-party primaries (Field, 2006). The Austrian People’s Party made similar demands of four-term incumbents (Müller, 1992). In Israel, Labor and the Liberals asked two-term incumbents to win at least 60 per cent of parties’ Central Committee votes (Goldberg & Hoffmann, 1983). In the UK, Liberal Democrats were renominated if most local members endorsed them, and Conservative incumbents needed approval by their association’s Executive Council. Labour incumbents (as of 2008) needed affirmative nominations from a majority in their local constituency (Ashiagbor, 2008).

Numerous scholars study incumbent renomination, mainly focusing on renomination rates, incumbents’ renomination prospects, and various implications of renomination (Asquer, Golden, & Hamel, 2020; Baumann, Debus, & Klingelhöfer, 2017; Gallagher & Marsh, 1988; Marino & Diodati, 2017; Müller, 2000; Somit, Wildenhamman, Boll, & Römmele, 1994; White & Ypi, 2020; Yildirim, Kocapinar, & Yüksel, 2019). Much less attention has been paid to determines of incumbent-renomination methods. This question is essential because renomination and the debate surrounding it are not merely technical. At their heart is a debate over the nature of intra-party democracy, the relationship between MPs and constituents, and the relationship among parties’ different parts. All of these are vital for parties’ functioning and development. While several scholars examine determinants of general candidate selection methods (e.g. Cross & Gauja, 2014; Lundell, 2004; Shomer, 2014), this paper is the first attempt, to my knowledge, to examine the determinants of incumbent-renomination methods.

Using Katz and Mair’s party-faces approach to party organisation (1993), my main argument is that to explain parties’ choice of incumbent-renomination methods, we need to examine who controls parties and their conception of MPs’ partisan representative style. This is key, I argue, because this conception differs between the party in public office (party in parliament or government) and the party on the ground (extra-parliamentary party) and leads to a different incumbent-renomination method. The former sees partisans as bound by dictates of parliamentary-party leaders, leading to automatic (or almost automatic) renomination. The latter sees partisans as bound by dictates of the extra-parliamentary party, leading to a method exerting greater extra-parliamentary control over renomination.

Publicly available data on incumbent-renomination methods are hard to obtain, with information primarily found in accounts of general candidate selection methods or studies of intra-party processes. In this paper, I propose to examine the empirical relevance of my theoretical framework using the UK Labour Party as an influential case. Labour serves as a good building block for developing a theory to explain what determines incumbent-renomination methods for several reasons. First, between 1979 and 2019, Labour adopted six different incumbent-renomination methods. The 1980 adoption of Mandatory Reselection is the most well-known example of a party denying incumbents automatic renomination (Hazan & Rahat, 2010).

Table 1. Incumbent Renomination Mechanisms, UK Labour Party, 1979–2019.

Founding to 1979	Practically automatic renomination. Denying renomination involved three special GMC, a meeting of all locally affiliates to mandate GMC delegates to decide renomination and either ratification or rejection by the NEC. In 1970 the process was amended slightly but remained effectively automatic.
Before 1983 General Election (GE)	First Change. 1980 Conference adopted Mandatory Reselection: all incumbents seeking renomination faced constituency reselection procedures within three years of the last general election. MPs retained automatic right to be shortlisted.
Before 1992 GE	Second Change. 1987 Conference adopted local electoral collages: affiliated organisations retained up to 40 per cent of the total vote and individual rank-and-file members in CLPs cast the rest. Third Change. 1991 Conference adopted ‘trigger’ ballots: gave CGCs power to decide on a contested renomination rather than initiating it automatically in all constituencies
Before 1997 GE	Fourth Change. 1993 Conference adopted OMOV and unopposed renomination of incumbents receiving at least two-thirds of intra-CLP nominations.
Before 2001 GE	Fifth Change. definitive discontinuation of Mandatory Reselection. Adoption of ‘affirmative nomination’: number of constituency nominations required to avoid contested renomination reduced from two-thirds to half.
Before 2019 GE	Sixth Change. 2018 Conference increased number of constituency nominations required to avoid a contested renomination to two-thirds.

Second, Labour’s internal debates over renomination methods date back to its founding and are well documented (e.g. Criddle, 1984, 1988, 1992, 1997, 2002, 2005, 2010; Quinn, 2004; Russell, 2005; Seyd, 1987; Shaw, 1988, 1994, 1996; Young, 1983), providing excellent ground for this examination. The rich historical analysis based on these accounts will serve as a strong basis for theory-building and future comparative works. Finally, the six changes were adopted across four decades and ten general elections, allowing for an intra-party comparative analysis of connections between those controlling the party, their prevailing conception of MPs’ representative style, and the chosen incumbent-renomination method.

The paper proceeds as follows. After a brief literature review on incumbent renomination, I detail the theoretical framework and use the Labour Party to showcase its empirical relevance. The historical analysis supports my argument of a connection between prevailing conceptions of MPs’ partisan representative style among those controlling parties and choice of incumbent-renomination method. Findings shed light on incumbent-renomination methods, with implications for understanding intra-party power dynamics and intra-party democracy. The last section discusses these and suggests avenues for future research.

Incumbent Renomination

To choose which candidates to field in an upcoming general election, parties use various methods (Hazan & Rahat, 2010). While parties may select new candidates using the same methods that decide incumbents’ fate, these are two distinct decisions that merit separate analyses because incumbents hold unique intra-party positions. Unlike new candidates, incumbents are familiar with their constituents and intra-party dynamics, have more legislative and electoral experience, more intra-party influence, and are more likely to reach leadership positions. Incumbent renomination is the subject of various scholarly explorations. For example, Asquer et al. (2020), Marino and Diodati (2017), and Yildirim et al. (2019) examined incumbents’ renomination prospects, while Baumann et al. (2017) and Müller (2000) examined implications of renomination for intra-party competition and party discipline. In related literatures, Matland and Studlar (2004), Kerby and Bildook (2011), and Gherghina (2015) examined legislative turnover (a consequence of incumbent renomination) and White and Ypi (2020) examined the use of renomination as a tool of direct democracy. But the question of what determines the methods parties use to renominate incumbents is under-researched. Several scholars examine what determines general candidate selection methods, including electoral systems, district magnitude, and party ideology (Cross & Gauja, 2014; Lundell, 2004; Shomer, 2014). However, these studies do not distinguish incumbent renomination from the general candidate selection process and do not comment on the determinants of the former.

The starting point for evaluating these determinants is with the few accounts dealing specifically with incumbent renomination as a process of intra-party candidate selection. Thus, Dickson (1975) and Hazan and Rahat (2010) suggest that most parties automatically or almost automatically renominate incumbents due to several advantages. First, assuming parties’ goal is to win elections, renomination is logical because incumbents have already shown they can win election(s). Moreover, incumbents likely have a strong constituency support base, further increasing electoral victory chances. Third, incumbents tend to be strong intra-party actors who know how to take care of constituents’ interests because renomination is their goal. Fourth, incumbents’ automatic renomination renders unnecessary potential fights against constituency challenges before each election, protecting parties’ vital interest to minimising internal conflict so close to general elections. Fifth, constituencies’ dissatisfaction with MPs’ performance will usually not be severe enough to deny renomination. Instead, it would be communicated to MPs and the matter would be solved internally. Sixth, despite claims to the contrary, parties’ national executives frequently exert significant ‘behind the scenes’ influence to ensure renomination. Finally, according to Dickson (1975), the most powerful motivation to renominate is the ‘nature of traditional beliefs defining the relationship between the MP and his constituency party … [that] constituency parties should not act as a delegating or controlling body’ (63).²

But, as mentioned above, some parties pose obstacles to incumbents’ renomination. Why might that be? Hazan and Rahat (2010) propose that automatic renomination may impart a static image upon parties that could hurt their chances in the general election. An internal change can be perceived by voters as an attempt at self-renewal and regeneration, resulting in a favourable electoral outcome. And yet, this explanation does not satisfy the question of why parties adopt and adapt incumbent-renomination methods, nor do Hazan and Rahat (2010) or, to the best of my knowledge, other studies offer a systematic evaluation of this question. To fill this lacuna, I propose to focus on two explanatory factors: (1) who controls the party—the party in public office or the party on the ground; and (2) whether those in control perceive MPs as bound by the dictates of the parliamentary party leadership or the extra-parliamentary party.

Theory

MPs perform multiple functions as parliamentarians (Blomgren & Rozenberg, 2012; Searing, 1994), including representation. Examining the various elements involved in MPs’ roles as representatives, Eulau et al. (1959) distinguish two primary elements: focus of representation and style of representation. The first concerns whom MPs represent, traditionally differentiating constituency voters and the nation en masse. The second concerns MPs’ manner of representation, how they ‘think about their representational role and what source(s) they use, or claim to use, when making decisions’ (Önnudóttir, 2016, p. 734). Together, these elements produce the classic concepts of MPs as trustees or delegates. Trustees focus on the nation (though a specific geographic constituency might elect them) and base decisions on their judgment rather than (local) constituents’ opinions. MPs are thought of as free agents who follow their understanding of what is right or just when making decisions (Judge, 1999). Delegates focus on a specific group, the local constituency, and ‘perceive themselves as mouthpieces of constituency demands’ (Zittel, 2012, p. 101). They are agents of their constituents, elected to express their opinions irrespective of the MPs’ views (Önnudóttir, 2016; Riemer, 1967).

As illuminating as this distinction is, note that it originated before the development of political parties and the Responsible Party Model (RPM) (Schattschneider, 1942, 1975). At the core of RPM is the idea that unified political parties are the source of policy decision-making. Parties decide on policies that MPs are expected to adhere to and be accountable for through their parties.³ RPM meant that the original trustee/delegate distinction was inadequate for examining MPs’ representative roles in the age of political parties. Now, MPs need to decide between local, national, and party interests. In response, scholars introduced a third concept—MPs as partisans. Like delegates, partisans focus on representing a specific group rather than the nation en masse and consider the group’s opinions when making decisions. The crucial difference between delegates and partisans is that for the latter, the specific group is not the geographic constituency from which MPs are returned but their political party (Önnudóttir, 2016; Riemer, 1967).

This paper seeks to develop a theoretical framework to explain the determines of incumbent-renomination methods, which is an intra-party matter. Accepting RPM’s argument that parties act in cohesion to arrive at a set of policies to which MPs are meant to adhere, and thus that the relevant conception of MPs’ representative style is that of partisans, I set aside MPs’ focus of representation (that deals with whom MPs represent) and devote my attention to their style of representation—the way they represent. Second, accepting that the concepts of trustees or delegates are insufficient to account for MPs’ representative styles in the age of political parties, I emphasise the third style—MPs as partisans. Doing so reveals that its existing formulation—that MPs are meant to adhere to their parties’ policy positions—does not provide a satisfactory answer to the question: who within the party decides what those policy positions are? Or, more broadly, who controls the party and, therefore, where does MPs’ intra-party accountability lay? This last question is especially important because it bears on parties’ choice of incumbent-renomination methods—those who control the party will also decide which method(s) to use.

To start, let’s think of parties through the lenses provided by Katz and Mair (1993). They argued that parties are complex organisations with multiple parts operating separately and in interaction with one another, dividing parties into three faces: (1) party in public office; (2) party on the ground ; and (3) party in central office.

The party in public office (PPO)—the party in parliament or government—comprises individuals who won general elections. As elected representatives, whose main goal is gaining office, they depend on the electorate’s support to keep their posts (Katz & Mair, 1993). Frequently, these representatives also rely on a selectorate’s intra-party support in pre-general-election processes of candidate selection or renomination (Hazan & Rahat, 2010).

The party on the ground (POG)—the extra-parliamentary party—comprises party activists and members. When parties have a formal membership, the extra-parliamentary party is composed of individual members. But even people who are not formally party members are part of the extra-parliamentary party: activists, members through affiliated organisations, financial supporters, and, sometimes, even loyal voters (Katz & Mair, 1993). If elected representatives are primarily motivated by re-election, the ‘primary incentives for members of the party on the ground are public purposive (policy), symbolic, and solidaristic. Thus, making and adhering to formal statements of party policy and identity are likely to be of great significance … ’ (598). Throughout the paper, I use both the PPO/the parliamentary party and the POG/the extra-parliamentary party interchangeably.

The third face, party central office, is composed of permanent staff, offices, and institutions. It is primarily the bureaucratic face, connecting the POG and the PPO, coordinating national election campaigns, fund-raising, policy research, and other administrative tasks.

As the two political elements of party organisations, and arguably the two main sides of the coin, I focus on the relationship between the first and second faces. The intra-party power balance between them varies across time and parties. In classic conceptions of mass parties, the POG is regarded as in control because the party originated outside parliament. In contrast, in the catch-all party, power shifts away from the POG to the PPO (Katz & Mair, 1993).⁴ The point of emphasis is this: in thinking of parties as composed of different faces, each vying for control, the conception of MPs as partisans should reflect the intra-party power balance between the faces. In parties where the two faces fundamentally agree, the partisan conception is straightforward—MPs are accountable to a party that operates in harmony, with no conflict over the intra-party locus of accountability. But, in parties where the faces struggle for power, there will also be disagreement over MPs’ intra-party accountability because each face holds that MPs are accountable to it and not to the other. This disagreement should also affect the choice of an incumbent-renomination method since the threat to deny renomination is a powerful tool to ensure accountability (Gallagher & Marsh, 1988; Müller, 2000). Thus, if leaders of the extra-parliamentary party hold that MPs-as-partisans ought to adhere to their dictates, and this face is dominant, incumbent renomination should reflect this by giving the extra-parliamentary party control over the process. In contrast, if the parliamentary party leaders hold that MPs as partisans ought to adhere to their dictates instead, and this face is dominant, incumbent renomination should be under their control and likely automatic or almost automatic.

Note that this theoretical framework does not assume the necessity or existence of unanimity of opinion among all face members regarding the conception of MPs as partisans. Such disagreements over the issue may very well exist within the party face. The emphasis here is on the conception of MPs’ representative style held by the leading actors of the PPO or the POG and on which of the faces is dominant. That is, the combination of both conditions is necessary for explaining incumbent-renomination methods choice. It should further be emphasised that within the paper’s constraints, this argument is necessarily a stylised account of intra-party power dynamics. However, as the analysis of the UK Labour Party below shows, this framework nevertheless sheds light on parties’ choice of incumbent-renomination methods and the influence upon it of the power balance between the PPO and the POG.

The UK Labour Party

Between 1979 and 2019, the UK Labour Party changed the way it renominated incumbents six times. Table 1 summarises these developments. In what follows, I analyse these changes and the intra-party debate surrounding them.

Since the first sizeable number of Labour MPs entered Parliament in 1906, the Party has debated the relationship between its MPs and extra-parliamentary bodies. These debates reveal a long-standing disagreement between the POG and the PPO over the nature of intra-party democracy and MPs’ representative role (Seyd, 1987). Early in Labour’s history, when its main objective was ensuring parliamentary representation for the working class as a group, the POG, with the trade unions at its core, was at least formally the dominant face (Birch, 1964; Judge, 1999; Pilkington, 1997; Quinn, 2004). Then, Labour ostensibly rejected the idea of MPs as trustees favouring the partisan conception of MPs as ‘servants of the movement’ and thus subject to extra-parliamentary control through Conference (Young, 1983). However, the extra-parliamentary version of the partisan representative style posed a problem for a party aspiring to government, especially after 1929, because it ‘makes no allowances for the government’s need to adjust to changing circumstances; or for the fact that the party’s conference delegates may not see the world as it really is … ’ (A. Young, 1983, p. 28). These two conceptions of MPs’ roles, combined with tensions over intra-party power balance, led to disputes ‘between the ‘Bevanite’ left and more ‘moderate’ or right-wing elements in the 1950s … between the ‘revisionists’ and more traditional elements into the 1960s’ (Russell, 2005, p. 13), and again when Labour was in government in the 1970s (Judge, 1999). As Young (2000) argues, ‘Labour Left invariably sought to maximise leadership accountability to party conference, whereas Labour right-wingers (including sections of its parliamentary leadership) typically sought to minimize the influence [of] activist groups … ’ (8).

While the stylised account in this paper necessitates a crude distinction between Labour’s internal factions, ‘Labour Left’ and ‘Labour Right’ are often used to discuss Labour’s history and merit an explanation. According to Young (2000), Labour Left organised both within and outside Labour and can be broadly divided into three types. It supported a socialist political agenda, using different tactics to influence party policy and its internal workings. The first type, the ‘extreme-left,’ included the Militant Tendency (est. 1964). Supporting a radical socialist program, these groups mainly pursued ‘covert strategies of entryism … becom[ing] active party members in order to subvert … the direction of party policy … ’ (R. Young, 2000, p. 38), a strategy that was exposed and stopped in the early 1980s when Militant members were expelled from Labour. The second type, the ‘hard-left,’ including the Campaign for Labour Party Democracy (CLPD, est. 1973 by grassroots activists), ‘favoured a range of radical political objectives, in particular the extension of grassroots party democracy, public ownership … and constitutional reform’ (39). Its most prominent member was Tony Benn, with other leading figures, including Dennis Skinner, Jeremy Corbyn, and Diane Abbott.

Most importantly, the hard-left preferred ‘Labour’s traditional socialist orthodoxy over the pragmatic and revisionist electoralism of party leaders, and to defend the importance of grassroots membership activism as ‘champions’ of intra-party democracy’ (40). Finally, the third type, the ‘soft-left,’ emerged during the early-1980s as an alternative to the CLPD and took in groups like Tribune (est. 1964). The most pragmatic of the left-leaning elements, the soft-left ‘was of particular strategic importance to Neil Kinnock’s leadership … [providing] support and legitimacy for the process of party-level modernisation and renewal’ (40).⁵

Unlike the groups associated with Labour Left, Labour Right’s natural affiliation with the trade-unions’ non-ideological elements meant that it tended to organise less publicly, and that organisation on the ground was not as necessary. Labour’s close relationship with unions also meant that Labour Right (almost) always controlled the National Executive Committee (NEC—the party central office) and the Parliamentary Party (PLP). Thus, ‘the ethos of this group was synonymous with respect for and loyalty to leaders … [implying] a pragmatic approach … ’ (Russell, 2005, p. 20). But Labour Left’s position as the perpetual minority in both PLP and NEC changed when trade unions began drifting left in the 1960s.⁶ Labour Left’s strength peaked by the early 1980s, and ‘it became the Labour Party in the sense that this faction predominated within much of the Party’ (Seyd, 1987, p. 17 emphasis added). As Labour Left and Right ‘subscribed to radically different conceptions of the Party’s role and purposes … ’ (Shaw, 1996, p. 162), the struggle between them for party control was fierce. How do these power struggles help us understand Labour’s choice of incumbent-renomination methods?

Power Balance, Debate, and Reform

Technically, Labour’s extra-parliamentary party always had control over incumbent renomination as CLPs had the right to dismiss incumbents since the Party’s founding. However, before 1970, this involved a prolonged and complicated process of three General Management Committee (GMC) meetings specifically called to discuss deselection, a meeting of all locally affiliated bodies to mandate GMC delegates to decide on deselection, and finally, either ratification or rejection by the NEC. Under these rules, renomination was, in practice, automatic. In 1970, the process was amended slightly to make it easier for CLPs to dismiss MPs. Still, because the NEC retained the right to make the final decision, renomination remained practically automatic.

After Labour lost power in 1970, factional tensions mounted, leading to conflicts between the POG and the PPO. Left-wing activists were dissatisfied with Wilson’s government’s apparent refusal to abide by Labour’s manifesto and conference decisions, attempting to break with Labour’s natural supporters—the working class (Judge, 1999). Arguing that the extra-parliamentary party bounded MPs as partisans, left-wing activists called for candidate selection reform, including incumbent-renomination, to gain control over the PPO (Armstrong, 1991). In that spirit, The CLPD called for structural reforms to ensure greater PLP accountability to the POG, including Mandatory Reselection of all incumbents. These reforms aimed to weaken and even terminate PLP independence by shifting power away from the PLP to activists and unions, thus asserting de facto the authority they already held in principle (Webb, 2000). In the background of this call to empower the extra-parliamentary party was the left’s dominance among trade union branches (Seyd, 1987) and Constituency General Committees (CGCs) responsible for renomination. The percentage of votes for successful Labour Left candidates in the CLP section of the NEC, which grew steadily from 55 in 1970 to65 per cent in 1979, illustrates this dominance, as does Seyd’s (1987) observation that ‘between thirty and one hundred local parties were willing to cast one of their seven votes for a Militant Group candidate sometime between 1971 and 1982’ (52-53). The PLP’s right-wing, and especially the leadership, who did not think that MPs as partisans were accountable to the extra-parliamentary party, objected to Mandatory Reselection. Instead, they argued that ‘MPs were properly accountable to the voters, not to a narrow party caucus … [because] MPs were far more sensitive to the currents of public opinion’ than Labour activists (Shaw, 1994, p. 16).

The 1980s

Calls for intra-party reform further intensified following the Callaghan Government’s defeat in 1979. They were ‘accused of abandoning the 1974 manifesto, flouting Conference and treated the wider Party with contempt … ’ with the lesson ‘[being] that whatever the formal rules, in practice it was the parliamentary party elite, which … determined policy’ (Shaw, 1996, p. 163). For the extra-parliamentary party, the solution to this problem was to change the intra-party power balance by changing the rules, which succeeded when, over the leadership’s and most of the PLP’s profuse objections, Mandatory Reselection was adopted by the 1980 Conference (Cronin, 2004). Under the new rules, all renomination-seeking incumbents would face a constituency reselection procedure within three years of the last general election, though MPs retained an automatic right to be shortlisted. During the 50 years that preceded Mandatory Reselection, a total of 28 incumbents were deselected. By 1983 eight incumbents were deselected—the largest number of MPs formally deselected before any single general election in Labour’s history (Butler & Butler, 2011).

Mandatory Reselection was adopted as part of an intra-party power struggle when the extra-parliamentary party, dominated by Labour Left, sought to assert its control over Labour. The importance of this power-balance shift is bolstered by the fact that the number of deselected incumbents before 1983 could have been much higher but for two things. First, numerous MPs anticipated tough renomination battles in CLPs controlled by Labour Left and retired. Second, 28 mainly right-wing MPs defected to the Social Democratic Party, citing Mandatory Reselection as a catalyst (Criddle, 1984; Seyd, 1987). The intra-party power struggle was so fierce, and disagreements over the proper relationship between the PLP and the extra-parliamentary party so strong that Labour split.

The new power balance between the POG and the PPO and between Labour Left and Labour Right did not last long. As early as 1981, Labour Left began to fracture after Tony Benn lost the deputy-leadership election to Roy Hattersley (of the centre-right), a defeat blamed on members of the Tribune Group who failed to endorse him. In 1982, tensions within the Left further intensified with the founding of the pro-Benn Campaign Group (a rival group to Tribune) and several losses in that year’s NEC election (R. Young, 2000). In the 1983 election run-up, the leadership publicly distanced itself from some of the more radical left-wing elements launching an inquest into Militant Tendency activities.

After Labour’s landslide defeat in 1983 and subsequent selection of Neil Kinnock as Party Leader, Labour Left was weakened further by ongoing internal disputes, exemplified by the hard-left’s protesting Militant members’ expulsion. The weakness of Labour Left and a desire to avoid another electoral defeat put significant pressure on Kinnock to modernise Labour and change incumbent renomination. Mandatory Reselection, it was argued, rendered many MPs far too dependent on local activists’ goodwill, eroding security of tenure and parliamentary discipline. In response, Kinnock proposed ‘One Member One Vote’ (OMOV), transferring renomination decisions from the CGCs to rank-and-file members. The proposal meant to essentially keep renomination in line with the conception of partisan MPs as responsible to the POG but shift the base of the extra-parliamentary party away from unions and activists to rank-and-file members. It aimed to decrease friction between CLPs and incumbents (passive members were considered more ‘moderate’ than activists) and reassert the PPO control over Labour. But, with the extra-parliamentary party still dominant and with left-wing unions and activists still dominant within it, the 1984 Conference rejected the proposal (Shaw, 1996; R. Young, 2000).

Despite the failure to pass OMOV, continued weakening of Labour Left meant that the gradual return to the previous power balance—de facto dominance of the PLP over the extra-parliamentary party—continued. This, combined with Labour’s third consecutive general election defeat, resulted in the 1987 Conference reforming incumbent renomination by adopting local electoral colleges as part of a larger modernisation plan. The plan sought to amend party policy and other intra-party power structures to strengthen the PPO. For incumbent renomination, adopting local electoral colleges meant affiliated organisations—mainly unions—retained up to 40 per cent of the total vote, and individual rank-and-file members in CLPs cast the rest.⁷ Unlike the early 1980s, by the end of the decade Labour Left was much weaker, and many trade union leaders strongly desired replacing the Conservative government. Coupled with growing strength of centrists and right-wingers within Labour, the modernisation plan encountered some resistance, but it and further reforms to incumbent renomination, could not be stopped.

The 1990s and ‘New Labour’

With intra-party power balance shifting back to the PLP and the centre- right-wing, two more changes to incumbent renomination were made in the early 1990s. In the run-up to the 1992 election, the 1991 Conference adopted ‘trigger’ ballots, giving CGCs the right to decide whether to initiate a reselection process rather than initiating it automatically in all constituencies (Criddle, 1992). Second, the 1993 Conference adopted OMOV and unopposed reselection of incumbents receiving at least two-thirds of all nominations within their respective CLPs (Criddle, 1997). While MPs still spent time soliciting constituency support to get two-thirds of the nominations (which arguably pushed them to adhere to constituents’ wishes to some degree), renomination was more automatic now than under Mandatory Reselection.

Between these two reforms, Labour lost yet another election. Reflecting continued shifts in intra-party power balance between Labour Left and Labour Right over the past decade, Tony Blair was selected Party Leader following John Smith’s sudden death in May 1994. Blair firmly believed that to win power Labour had to be dominated by the PPO and further diminished activists’ power by extending individual members’ balloting from personnel selection to voting on party policies (Russell, 2005). By appealing directly to Labour’s at-large membership over party activists’ heads, Blair was the first to rewrite Clause IV of Labour’s Constitution (Seyd & Whiteley, 2002). Furthermore, while intra-party democracy remained a prominent feature of Labour’s structure, Blair and his fellows in the leadership continued to assert the PPO’s control over Labour. Like Kinnock’s push to adopt OMOV, Blair supported MPs being accountable to the party at large rather than party activists (Judge, 1999, p. 85), as it would make it easier to control the party.

With that spirit among Labour’s leadership and after 18 years in opposition, Labour won the 1997 election in a historic landslide, leading to the first time incumbent renomination would be an affirmative process implemented while Labour was in government. Criddle (1997) predicted ‘Labour activists could well by then [the next general election] be disaffected and seek to threaten their sitting MP[s]’ (207). He was wrong. Instead, the 1997–2001 inter-election period saw the definitive end of Mandatory Reselection and the fifth change of incumbent renomination. The return to easier renomination came at the height of Labour Right’s control over the PLP and the latter’s intra-party dominance. The new method, officially termed ‘affirmative nomination,’ reduced the number of constituency nominations required to avoid a OMOV ballot from two-thirds to half, making renomination more automatic. Under those rules during Labour’s 13 years in government (1997-2010), CLPs deselected four MPs (Criddle, 2002, 2005, 2010).

Post ‘New Labour’—2010-2019.

After three terms in government, Labour lost the 2010 election, and Gordon Brown resigned as Leader. His successor, Ed Miliband, was selected on an ‘anti-New-Labour’ platform, having declared that ‘’the era of New Labour is over’ [adding] in private … that he wanted to bury New Labour’ (Cowley & Kavanagh, 2016, p. 70). Miliband advocated for reviving class politics on Labour’s agenda (Bale, 2015) and was supported by left-wing union leaders like Unite’s Len McCluskey. For Miliband and his supporters, the term class was used as ‘a metaphor for more radical policies … after 13 years of … compromises and disappointments of government’ reflecting ‘a revival of [intra-party] factional conflict long suppressed in the governing years’ (Criddle, 2016, p. 342). Following the 2015 election defeat, Labour Left’s increase in strength resulted in the founding of Momentum—an extra-parliamentary left-wing organisation—and the selection of Jeremy Corbyn, a long-time Labour Left MP as Leader.

From 2015 to 2017, Momentum’s and other Labour Left elements like Unite’s strength continued to grow in the POG, reflected again in a power shift within the NEC (Pope, 2016). However, like four decades ago, while the left’s power in the extra-parliamentary party grew, significant right-wing presence in the PLP led to strong disagreements over the proper relationship between the party faces. In the run-up to the 2017 election, debate over renomination and reinstatement of Mandatory Reselection was reignited. Both Momentum and Unite strongly advocated for it because they wanted ‘to ensure democratic accountability and the rights of party members to select candidates that reflect their views … ’ (Forester, 2016). While the snap election in 2017 did not leave time to amend the incumbent renominate process, calls for reform and its justifications reflected differences between conceptions of MPs as partisans held by different intra-party elements, each vying for control. The POG demanded MPs be more accountable to it, achievable by direct and firm control of MPs.

In contrast, strong elements in PPO resisted, arguing for the importance of MPs’ independence, and articulating a fear of reprisal over a disagreement with Corbyn. However, after losing the 2017 election, the strength and influence of Labour Left continued to increase (Bloom, 2018), and with it the call to reform incumbent renomination (Rodgers, 2008). At the 2018 Conference, supporters of greater accountability succeeded in passing the sixth change to incumbent renomination, reducing to one-third the constituency nominations threshold needed to trigger a reselection process (Watson, 2018). Interestingly, Corbyn also supported the move (Waugh, 2016), illustrating the more nuanced aspect of the proposed theoretical framework: some members of the PPO may hold different conceptions of MPs as partisans, as was the case in the 1980s with the members of the Campaign Group.⁸ But I would argue that the POG dominance within the party and the prevailing conception of MPs as partisans among the POG’s leading actors determined the incumbent-renomination method adopted by the 2018 Conference.

Conclusions

The paper aimed to develop a theoretical framework to explain the determinants of incumbent-renomination methods, using the UK Labour Party as an influential case to illustrate its empirical relevance. The argument was that incumbent-renomination methods are determined by two factors: which party face controls the party, the PPO or the POG, and the prevailing conception of MPs’ representative style as partisans among the leading actors within the controlling face. Because, generally speaking, the PPO sees partisans as bound by the dictates of the parliamentary-party leadership, when that face is dominant, incumbent renomination would be automatic (or practically so). In contrast, because the POG, again generally speaking, sees partisans as bound by the dictates of the extra-parliamentary party, when that face is dominant, there would be more extra-parliamentary control over renomination.

With the frequent changes to its renomination system and the years-long intra-party debate on the issue, analysing the Labour Party serves as a building block for developing a theory to explain the determinants of incumbent-renomination methods. The proposed theoretical framework sheds light on the six methods by which the party renominated incumbents between 1979 and 2019. The debates surrounding the changes and their respective timing show that incumbent-renomination methods were shaped by the prevailing conception of MPs’ representative roles as partisans among the leading actors in the face controlling the party. Mandatory Reselection was adopted when the extra-parliamentary party sought to assert control over Labour in the early 1980s; the final discontinuation of Mandatory Reselection came when the PLP was in firm control in the late 1990s and early 2000s; and the latest change to incumbent renomination in 2018, making it easier to deselect incumbents, came when the extra-parliamentary party was again in control.

While it is difficult to reach generalisable conclusions based on a single case study, this paper is a necessary first step in developing this area of research. Our existing knowledge of incumbent renomination primarily rests on descriptions of overall candidate selection (Hazan & Rahat, 2010; Ranney, 1965), studies of their implications for intra-party competition (Baumann et al., 2017), party discipline (Gallagher & Marsh, 1988; Müller, 2000), MPs’ renomination prospects (Asquer et al., 2020; Marino & Diodati, 2017; Yildirim et al., 2019), and legislative turnover (Gherghina, 2015; Kerby & Bildook, 2011; Matland & Studlar, 2004). And while Dickson (1975) and Hazan and Rahat (2010) briefly discuss what might motivate parties to either automatically renominate incumbents or pose obstacles, this topic remains significantly under-researched theoretically and empirically. As mentioned above, publicly available data on incumbent-renomination methods are scarce, with analysis in this paper heavily relaying in existing accounts of either general candidate selection or party histories. Part of the difficulty lies in existing accounts not making clear distinctions between overall candidate selection and incumbent renomination or only providing technical details. For example, Criddle (1984, 1988, 1992, 1997, 2002, 2005, 2010) describes the varying degree of local associations’ involvement in the UK Conservative Party’s pre-general-election selection process over the years. Still, the decision about new candidates is not distinguished from that of incumbents. Ashiagbor (2008) provides comparative technical information on incumbent-renomination methods for parties in South Africa, Canada, and Ireland (among others). Still, extensive archival research is needed to evaluate the intra-party debates over the issue.

Building a theoretical framework and using an in-depth analysis of secondary materials of an influential case provides a foundation for future work. Such an agenda can shed important light on intra-party dynamics, exposing how internal power struggles and different understandings of the appropriate relationship between parties’ various parts shape intra-party processes. Future scholars can evaluate the theory’s generalisability using other case studies and pose alternative or complementary explanations for what determines incumbent-renomination methods.

While such efforts are beyond this paper’s scope, I conclude with one such possible alternative explanation using the example of the UK Labour Party again. Like the determinants of parties’ choice of general candidate selection methods (Shomer, 2014), electoral systems, ideology, size, or government status may affect parties’ choice of incumbent-renomination methods. The analysis of the Labour Party indicates that there might very well be a relationship between changes to incumbent renomination, intra-party power dynamics, and a party’s government status. During the Wilson and Callaghan governments, Labour experienced internal tensions between its Left and Right factions and between the POG and the PPO. Labour Left’s rise in power in the late 1970s coincided with the Party’s transitioned to the opposition following the 1979 defeat. As the leadership of the extra-parliamentary party had argued that MPs as partisans ought to adhere to the dictates of the POG, what followed was a change to incumbent renomination—the adoption of Mandatory Reselection. Most MPs’ deselections under Mandatory Reselection happened while Labour was in opposition. When Labour returned to power in 1997, Labour Left was weak, and the Party was dominated by the PPO, which was, in turn, dominated by Labour Right. The PLP leadership argued that party policies should be decided via debate between the party’s faces, but that ‘it was the [PLP] leadership which was to set the parameters of the debate’ (Judge, 1999, p. 85). Therefore, it is reasonable to surmise that the PLP leadership thought of MPs as partisans who ought to adhere to their dictates rather than those of the POG. Labour’s return to power also saw a definitive end of Mandatory Reselection, while the latest change, in the opposite direction, came when Labour was in opposition again, making it easier for the POG to deselect incumbent MPs. This is a crude account of Labour’s history, but it does lead us to ask: what are the relationships between parties’ government status, internal power struggles, and incumbent renomination? If there is a relationship, is it causal? And if so, is it directional? All these questions are thought-provoking and await future research.

Acknowledgments

I thank Richard S. Katz and Tristan Klingelhöfer for their valuable comments and advise on earlier versions on this work.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Alona O. Dolinsky

Alona O. Dolinsky is a PhD candidate in the Department of Political Science at Johns Hopkins University, Baltimore. She previously worked as a student research assistant to Gideon Rahat on the Political Party Database (PPDB) at the Hebrew University in Jerusalem, Israel. Her research interests lie in the field of comparative party politics focusing on text-as-data and on questions of group appeals, candidate selection, representation in political parties, and coalition politics.

Notes

1 For example: the Irish Fianna Fail (Gallagher, 1988) and the Japanese Liberal Democrats (Shiratori, 1988).

2 While Dickson (1975) refers to individual constituencies’ motivations for renomination rather than what determines parties’ incumbent-renomination methods, they illuminate party-level decisions on renomination.

3 More on the Responsible Party Model in Klingemann et al. (1994), and Mair (2008).

4 These are ideal types used to illustrate the argument, and in the real world, parties’ features may combine ideal types.

5 More on the three types of groups composing Labor Left in Young (2000).

6 Illustrated by Minkin (1978) who showed that an overwhelming volume of annual conference resolutions and amendments between 1956 and 1970 were left of official party policy in both rhetoric and sentiment.

7 Though the reform passed with overwhelming support, it drew much criticism for being too expensive and difficult to operate and was discontinued in 1993 (Criddle, 1988).

8 I thank the anonymous reviewers for this comment.