Political Candidate Selection

Who Wins, Who Loses, and Under-Representation in the UK

Jeanette Ashe

• Publisher: Routledge
• Year: 2019
• ISBN:
  9781351707831
• Edition: 1
• Topic: Politics & International Relations
• Subtopic: Politics
• Index: Politics & International Relations

The "secret garden of politics", where some win and others lose their candidate selection bids, and why some aspirant candidates are successful while others fail have been enduring puzzles within political science. This book solves this puzzle by proposing and applying a universally applicable multistage approach to discover the relationship between selection rules, selectors' biases, aspirants' attributes, and selection outcomes.

Rare party and survey data on winning and losing candidates and insider views on what it takes to win a selection contest at multiple selection stages are compared and used to reveal the inner workings of the secret garden. With a primary focus on the British Labour party over several elections, the findings challenge many long-held assumptions about why some aspirant candidate types are successful over others and provides real-world and controversial solutions to addressing women's and other marginalised groups' descriptive underrepresentation. As such, it provides a much-needed fresh look at party selection processes and draws new conclusions as to why political underrepresentation occurs and should inform policies to remedy it.

This text will be of key interest to scholars and students of gender and ethnicity in politics, political parties and candidate selection, and more broadly to the study of political elites, comparative politics, sociology, labour studies, gender, race, and disability studies, and to practitioners.

1 Study overview

There is no shortage of highly publicised stories retelling candidate selection contests as if they were made-for-Netflix dramas, replete with allegations of vote rigging, insider meddling, and unexpected endings. The British Labour party – a comparative star amongst parties in liberal democracies trying for the increasingly coveted part “most diverse candidate slate” – is no exception. Recent controversies include Vauxhall’s selection for the 2019 snap election where social media has been alighted with heated exchanges over whether the central party should designate it an all women shortlist (AWS) seat or implement all black shortlists (ABS) to ensure a Black, Asian, minority, or ethnic (BAME) candidate replaces Labour MP Kate Hoey. Other standouts include Liverpool Walton, one of Labour’s safest seats, where, following the selection for the 2017 snap election, six Labour officials resigned after the central party’s candidate, a politically inexperienced Dan Carden, was selected over the local party’s choice, Liverpool’s mayor Joe Anderson. Suspicious sign ups prevailed as the theme in the lead-up to the 2015 General Election in Falkirk, where selection voting irregularities linked to trade union Unite’s efforts to get Kate Murphy selected led to an internal investigation and eventually to Murphy standing down. Questions abound in Erith and Thamesmead, another safe seat, where local party members selected the relatively unknown Teresa Pearce as their 2010 candidate over the party-establishment-backed Georgia Gould and six other women aspirants on the AWS. This result, like other selection outcomes, left the losing aspirant candidates scratching their heads and wondering why they lost and, more to the point, why another aspirant won.

Selection process outcomes illustrate three important and connected points concerning representation and party selection processes explored throughout this study. First, even though some selected candidates go on to secure House of Commons seats, they sit in a legislature which does not reflect the demographic composition of the population from which it is drawn. As is the case with all liberal representative democracies, the British Parliament’s elected lower house does not mirror population characteristics such as sex, gender, race, physical ability, age, or class. Rather, it is disproportionately composed of members belonging to social groups generally associated in the scholarly and popular literature with political success, and, as argued in this study, the ideal aspirant candidate type. From party selectors’ choices, the ideal aspirant candidate type tends to be white, professional, middle-aged men and men without disabilities, and tends not to be women, BAME, working class, young, or disabled people. Further, the ideal aspirant candidate type often possesses several other attributes similarly linked in the scholarship to political success. With this in mind, this study evaluates eight categories of variables commonly associated with the ideal candidate type: social background, party experience, political experience, political networks, personal networks, appropriate political attitudes, political ambition, and voter mobilisation methods.

Truly representative legislatures should be composed of the same proportion of social groups found in the population (Phillips, 1995). In other words, they should be a microcosm of society on the premise that policy decisions are not seen as legitimate unless voters from marginalised groups see themselves in the social makeup of their legislative bodies (Mansbridge, 2010; Phillips, 1995). Increasing the descriptive representation of social groups leads to substantive changes in public policies on grounds that they better represent members of the same social groups (Childs, 2007, p. 85; Lovenduski, 2005; Norris and Lovenduski, 1995, p. 135, p. 224). If a population is composed of 51 percent women, 12 percent BAME people, and 10 percent disabled people, there should be at least the same proportion of legislators elected from each of these social groups (Bird, 2005; Charlton and Barker, 2013, p. 324; Mansbridge, 2010; Phillips, 1995).

While many authors attempt to explain what causes unbalanced legislatures, answers remain elusive. However, as getting elected generally requires first getting selected as a party candidate, it would appear a close examination of party selection processes is the key to understanding the social imbalance in any legislature. Party selection processes are the focus of this study, given the potential influence they have upon the final candidate pool presented to the general electorate. Party selectors’ quest for the so-called ideal aspirant candidate described earlier is influenced by the homo politicus model, leading to a system

designed to select a standard model candidate who is articulate, well-educated and typically employed in a professional career, in business as an executive or manager, in education as a school teacher or university lecturer or in the law as a practicing barrister or lawyer. . .. By defining the appropriate qualifications for a career in politics in such a way then certain types of candidates will tend to be successful. As a result, women, working class candidates and those from the ethnic minorities will tend to be consistently disadvantaged.

(Norris and Lovenduski, 1989, p. 94)

It is proposed here that the homo politicus model and selector preference for the ideal aspirant candidate type disadvantage aspirants from marginalised groups such as women, people who identify as BAME, and people with disabilities.

Secondly, a complete picture of a party’s selection process must not only portray who wins but also who loses. For example, Pearce’s and other candidates’ wins only become intriguing and counterintuitive when other aspirant candidates are included in the story. Using the logic presented earlier, a party’s aspirant candidate pool – comprised of all individuals who put forward their names for candidacy and either won or lost their bids for candidacy – should proportionally reflect the social characteristics of its final candidate pool – comprised of individuals who are selected as candidates. For example, if a party’s aspirant candidate pool is composed of 30 percent women, 10 percent BAME people, and 2 percent disabled people, its final candidate pool should be at least the same proportion of aspirant candidates who came forward from each of these social groups. Any disproportionality between the aspirant candidate and candidate pools suggests the presence of structural or intentional biases either toward the standard, ideal aspirant candidate type or against the non-standard, non-ideal aspirant candidate type (Phillips, 1998, p. 229).

Third, selections such as Erith’s and Thamesmead’s also demonstrate how selection processes are multistage. While being chosen as the candidate by party members at the final selection meeting receives the most attention, aspirants win candidacies only after submitting their curriculum vitae to the Party, applying for seats, acquiring nominations from key party organisations, and making it onto the party’s final shortlist. They must not only survive the final selection vote, but they must also survive several other elimination rounds of the selection process. In this sense, the social imbalance in candidate pools is not solely determined by the party’s selection rules and selection participants at the final selection stage; it is also influenced by the rules and participants at preceding stages. For example, if an aspirant does not receive the nominations required to get onto the local party’s shortlist, their name is not considered by party members for candidacy. The selection process, then, is a multistage filter with ideal aspirant candidates potentially being more successful than non-ideal candidates at different stages throughout the process.

At its core, this study seeks to better understand the representational imbalance by developing a methodology toward explaining why some and not other types of aspirant candidates secure party candidacies. Women’s underrepresentation is highlighted throughout this study for several reasons. Concern with women’s underrepresentation has exponentially grown in prominence amongst the academic community, interest groups, the media, and the public, not least because recent viral campaigns such as #MeToo and #SheShouldRun are reinforcing the link between women’s political presence and women’s social justice. Further, the suggestion that women candidates offer parties an electoral advantage has led many parties to track more closely this group’s progress over others’ (Kenny and Verge, 2016; Russell, 2005, pp. 111–112). Given this incentivization, if parties collect any data on aspirant candidates, they are more likely to first collect it on sex, resulting in more robust data on women aspirants than, for example, on lesbian, gay, bisexual, transgendered, and queer (LGBTQ) aspirants. Considerable attention is also given to the underrepresentation of two other marginalised groups, BAME people and people with disabilities, whose underrepresentation, although gaining attention, is relatively underexplored within the context of party selection processes.¹

This study uses the well-established framework of supply and demand and a multistage approach to test the extent to which party rules and party selector preferences impact selection process outcomes using a diverse range of data on almost 20 years of Labour party selection processes. These include rare Labour party census data on all aspirant candidates for the 2001, 2005, and 2010 General Elections, party data on aspirant candidates for the 2015 General Election, original survey data of aspirant candidates for the 2005 and 2010 General Elections, interviews with party officials and aspirant candidates from 2010 to 2017, and analyses of parliamentary selection rule books. ‘Given the expedited nature of the 2017 snap election, the central party did not systemically collect usable selection data and thus they are absent from this study (but see Campbell, Childs, and Hunt, 2018). In addition, this book was at press when the 2019 snap election was called, making data analysis impossible for this election.’ These available data sources from elections held between 2001 to 2015 help illuminate an underexplored yet critical part of the larger legislative recruitment process concerning the micro-steps individuals take at the beginning of their parliamentary careers (Norris, 1997a, p. 1).² In doing so, this in-depth and intensive examination of a single party within a single political system over four general elections uses a wide range of data to shed light on the immediate case at hand and also to test the validity of long-held theories and provide generalisable methodological advances.³

There are several reasons to focus on the British Labour party. For some time the party has been considered an exemplar within liberal democracies with majoritarian single member plurality electoral systems and a party that has a “contagion effect” upon other parties (Seyd, 1999, p. 387). While it has a long tradition of internal party democracy, only in the early 1990s did Labour extend the right to select parliamentary candidates to all party members. There have since been many reforms to its selection process. Over the last 30 years, Labour has exhibited a higher degree of commitment to increasing the diversity of its candidate slates than any other party in the United Kingdom (UK) as well as any other party in countries with similar electoral systems, such as the United States (US) and Canada. The British Labour party is the only party in the UK using equality guarantees for women in the form of all women shortlist seats (AWS) and equality quotas for women and BAME aspirant candidates.⁴ These reforms contributed to the party having a higher percentage of women and BAME candidates and MPs than other major parties, including the British Conservative party, Liberal Democrats, and Scottish National Party.⁵ Thus, this study offers a fresh look at the Labour party’s selection rules and the way they influence the behaviour of selectors and the success of aspirant candidates at multiple stages of the selection process. Additionally, not only does this study seek to better understand the internal workings of the Labour party but also to use the lessons learned as a starting point by which to better understand the internal workings of other parties’ candidate selection processes and outcomes in the UK and in other countries.

The remainder of this chapter introduces the study and its key components. After a brief justificatory and definitional section, the chapter outlines the various steps in the selection process. It then explains how others have sought to understand why some and not other types of aspirant candidates successfully move through the selection process to secure candidacies, with special emphasis on detailing the dominant supply and demand framework and associated approaches and testing regimes.

Legislative imbalance and candidate selection

Understanding why legislatures do not currently reflect their host populations is one of the enduring mysteries in politics. From a normative perspective, this imbalance is of critical concern as a matter of democratic justice. According to Phillips (1998), it is “patently and grotesquely unfair” for any one group to “monopolise representation” (p. 229). Many studies on this topic draw attention to longstanding patterns of persistent descriptive underrepresentation of women and other social groups, such as BAME, disabled, young, and LGBTQ people, with some using explicitly intersectional analyses to do so (Bird, 2011; Childs and Dahlerup, 2018; Evans, 2012; Everitt and Camp, 2014; Freidenvall, 2016; Norris and Lovenduski, 1995; Phillips, 1995; Shah, 2013; Tolley, 2019; Wagner, 2019).⁶

The underrepresentation of women remains a significant problem despite scholarly efforts to understand why it occurs and the corrective efforts of some political parties and legislatures. While the percentage of women legislators is higher than it was 20 years ago, recent gains have been small. Current data indicating women still comprise under 25 percent of MPs in lower houses demonstrate the widespread and persistent nature of their political underrepresentation (IPU, 2019).⁷ In 39th place out of 193 countries, the British Parliament’s standing in the Inter-parliamentary Union’s ranking of women in national parliaments has improved since the last election; still, there is a long way to go before women achieve sex parity in the House (see The Fawcett Society, 2019).⁸ In the span of nine years, the overall percentage of women MPs has increased by 10 percentage points from 22 percent in the 2010 General Election to 32 in the 2017 snap election. However, this historic high is only a 3 percentage point increase from the 2015 General Election, suggesting progress has stalled on this front and diminishing the symbolic effect of a woman prime minister, Theresa May, at the helm (Campbell, Childs, and Hunt, 2018).

In terms of BAME peoples’ representation, cross-national research, although increasing, is rarer than on women’s representation (Black, 2017; Bird, 2014, 2011; Norris and Lovenduski, 1995; Ruedin, 2013; Shah, 2013; Tolley, 2019).⁹ A 2006 study reveals there is at least one BAME legislator in 75 percent of all democratically elected national legislatures (Ruedin, 2009). Another study shows BAME legislators exceed 5 percent of legislators in only 2 of 15 established democracies despite comprising a much higher proportion of the general population (Bird, 2005).¹⁰ By 2019, BAME MPs in the British Parliament comprised 8 percent of all MPs – still, this falls short of their 13 percent representation in the general population (Operation Black Vote, 2019; Of...

Table of contents

• Cover
• Half Title
• Series Title
• Title
• Copyright
• Dedication
• Contents
• List of tables
• List of figures
• Acknowledgements
• List of abbreviations
• 1 Study overview
• 2 The selection process puzzle and ideal candidate types
• 3 Data and initial supply and demand tests
• 4 Centralisation and the Labour party's candidate selection process
• 5 Assessing centralisation in the British Labour party's selection process
• 6 Assessing early stage selector preference for "ideal" candidates
• 7 Do local party members select "ideal" candidates?
• 8 Conclusion
• Appendix: British Labour party candidate survey
• Bibliography
• Index