xvi Preface Research Council of the UK (Research Grant 0000-23-1991) and conducted by Pippa Norris and [oni Lovenduski. The Canadian Candidate Project 1993 study was funded by the Social Sciences and Humanities Research Council of Canada. Lynda Erickson would also like to thank Jane Dyson, Gonzalo Ceresco, Susan Fonseca, Christopher Kam, JeanFrancois Labell, Judy Morrison, Ken Stewart and Jeffrey Yip for their assistance with the project, Ken Carty for advice, Elizabeth Arnold, Cynthia Cusinatto, Rosemary Dolman, Sheila Gervais, Tom Flanagan and the national offices of the Canadian political parties for their participation in the project. The New Zealand Study was funded by the Foundation for NewZealand Research Science and Technology and conducted by Peter Aimer, Helena Cart, Raymond Miller, [ack Vowles and Jim Lamare. Lastly this book would not have been possible without the cooperation of all the 'parliamentary candidates who generously set aside time in a busy schedule to participate in the surveys, and all the party officials who provided assistance, information and advice. We thank you all for your help. PIPpA NORRIS Harvard University 1 Introduction: theories ofrecruitment Pippa Norris Competitive democratic elections offer citizens a choice of alternative partiesy governments and policies. But, equally important, campaigns provide voters with a choice of candidates for office, The nature of the ':'.ballot may vary: Dutch voters are offered a lang list of names under a ··\.::common party banner; Canadians face the choice of one candidate per ?::.partyin their riding; in American primary elections citizens can pick from ~:'.:,riv-al nominees within a parry. But in all cases voters are selecting political ::~::leaderswho may determine the future oftheir country. \X1hich candidates get on the ballot, and therefore who enters legislative office, depends on .the prior recruitment process. "'-, The concept of legislative recruitmentrefers to the critical step as indi-::'::viduals move from lower levels into parliamentary careers. TIle chapters .~,~:, this book work within a common conceptual framework which ,;'::,~sslUUes that all such recruitment involves four levels of analysis (see ;,iFigure 1.1): the politicalsystem, notably the legal regulations, party system and electoral system, which structure candidate opportunities in the political market-place; 'the -recruitment. process) particularly the degree of internal democracy within party organisations and the rules governing candidate selection; the supply of candidates willing to pursue elected office, due to their motivation and political capital; and lastly, the demands of gatekeepers (whether voters, party members, financial supporters or political leaders) who select some from the pool of aspirants. can be understood as nested, in a (funnel ofcausality', so that .0 ipply and demand works within party recruitment processes, which in '-'Ie.. are shaped by the broader political system, The core question '):rrsned in subsequent chapters concerns how individual actors interact vithin differem institutional contexts. By comparing how the recruitment process works in a range of advanced industrialised democracies we can explore how far variations in the institutional setting have a major impact upon the outcome.. The book compares established democracies including nineteen advanced industrialised societies in North America, Western Europe, Scandinavia and the Pacific. The research design is based upon a contextual analysis of the recruitment process within the major parties in each country combined with elite-level surveys of individual parliamentary candidates in elections held during the early 1990s, Mas' chapters deal with the recruitment process in general elections while me conclusion covers candidates running for election to the European Parliament. Similar, or functionally equivalent, questions were asked in each survey, allowing comparison offactors such as the social background and political experience of candidates. This approach provides significant crossnational variations in the institutional context, while allowing a richer and denser understanding ofthe specific process ofrecruitment within parties in each country. The conclusion seeks to test theories of recruinnent in a systematic manner across member states of the European Union. The aim of this introduction is to suggest why political recruitment matters, to map out common perspectives in the literature and to outline the 3Introduction: theories of recruitment approach adopted by this book, and to identify the core features of the recruitment process which will be covered by subsequent chapters. l17hat is the'impactofrecruitment onpoliticalcareers? Many previous studies have analysed political careers, where recruitment represents the first step in a lifetime's parliamentary service (Mezey 1979; Jewell 1985; Buck 1963; Blondel 1973; King 1981; Riddell 1993; "" Schlesinger 1966, 1991), Through recruitment people are choosing their ., leaders. In the long run 'who gets into the legislature, perhaps rising during a-twenty- or thirty-year career into the highest offices of state, may have <'--more important repercussions for the future of the country than other .' electoral choice. In many countries recruitment into parliament is a ",::,'filtering mechanism which determines who is eligible for government Jiffice. Same who pursue legislative careers will ultimately rise to become "'-G'abiner Ministers, parry leaders and heads ofstate. There are alternative ',~pathways into political elites, including the military] media or bureau\~acYJ but experience of elected office remains the most common route in ""ost democracies (Blonde11987, 1995), As Kazee (1994, 165) argues, . e effcctiveness of government in any society depends in large part upon equality ofthe leaders who seek office. The personal experiences, politcalattitudes and abilities which politicians bring to public life can vary ubstantially across different political systems, depending upon the qual'cations which are regarded as relevant for elected office. Unlike those "ho'become physicians or civil engineers, there are no standardised and iternationally recognised qualifications to be a politician. Unlike vacan'.ies:in executive management] there are no specified and well-defined job __~scriptions. Politicians can adopt multiple legislative roles (see Searing -"·,94). Whether the recruitment process favours those who can raise 'dependent financial resources, those who have worked their way up the ty ladder, or those with considerable experience in local government, '. yproduce very different types of legislators. In turn, this may influence ective governance. .As ideal types] routes into political careers can be classified as relatively.: I Normative concerns about political recruitInent The recruitment process represents one of the basic functions of every ~. political system which has long raised normative concerns about tile con',',., .,,~,equences ofthe recruitment process for political careers, the social diver• n . sity of legislative elites, and the democratic distribution of power within parties. Figure 1.1 Model of recruitment. mocrer ot recruitment Pippa Norris2 5Introduction: theories of recruitment Does recruitmentproduce. diverse leaders? ierecruitment process also determines the composition of parliaments] .o'·gets into power, and therefore whether legislatures reflect society at ··:'·ge.- This process has long raised concerns about the legitimacy of presentative bodies. In comparing the social composition of parlia~~)nts we can draw a distinction between the larger pool of aspirants who ~'¢dnterested in pursuing elected office, the smaller group of candidates ~~~~o-:are nominated to stand, and the smallest group oflegislatorswho are ected into parliaments. Like a game of musical chairs) some fall by the ~"yside at every stage ofthe process. Ifthe recruitment process involves a tally neutral competition for office, then parliaments will perfectly 'C'" or the supply of aspirants who come forward. But unless .MPs are ~B:tted "purely at randomv the recruitment process filters some over ~:thers.:i on a systematic basis. Some candidates fail while others succeed, cohesiveness and permeability of the legislatures. Most importantly, many established democracies have experienced an increased professionalisation. of legislativecareers (Buck 1963; King 1981). To use the Weberian distinction, in many countries amateurs who live for politics have been increasingly replaced by professionals who live from politics (Weber 1958). This shift signifies a move from amateurs who may enter public service as a temporary step) perhaps at the end ofa long and distinguished life in business, the law or journalism, towards a full-time life-long career with its own training, qualifications artd rewards, Tills trend means that more and more representatives tend to be experienced politicians) adept .: campaigners and skilled legislators] with many years of public service. ··:~>-oponents ofresponsible party government argue that democracy works ::~dst'.effectivelYwhere parties provide an alternative set of programmes ~#i;fue major issues facing the country, voters choose parties based an .~~eir.. policies and performance, and free and fair elections are held at gular intervals to allow alternation of the parties in power. As such ~ocracy provides'all citizens with an opportunity to hold parties col,qtively responsible for their actions. In contrast advocates of "strong' or \rrticipatory' democracy argue that the choice of parties in elections ..ce every four or five years provides only limited opportunities for ,.Jizen control over their leaders. To supplement this system, it is sugSweden (Esaiassori and Holmberg 1996: 31-48). Gender differences :~':among legislators have been found to be significant predictors of their '~:;attitudes in Britain (Norris and Lovenduski 1995). Scandinavia "'cKarvonen and Per Selle 1995) and tile United SIaIeS (Thomas 1994). ~s suggests that the development of a more diverse legislature may .influence not JUSt its legitimacy but also its dominant policy agenda. and ~~::p-erhapsits style of politics. 6 Pippa Norris depending upon factors such as their party service] formal qualifications, legislative experience, speaking abilities, financial resources, political connections, name-recognition, group networks, organisational skills] ambition for office or incumbency status, The criteria which are relevant for success can vary from one country to another. The central concern here is that as a result of this filtering process legislators .are often atypical of the electorate. One long tradition in the literature has traditionally focussed on political elites) notably the socio-economic background of leaders in government] the civil service, business and industry, and the military (Thomas 1939; Ross 1955; Aaronovitch 1961; Bottomore 1964; Parry 1969; Scott 1991; Mellors 1978; Putnam 1976; Loewenberg and Patterson .1979; Aberbach, Putnam and Rockman 1981). These studies established that legislatures worldwide include more of the affluent than the less weli-off, more men than women, more middle-aged than young, and more white-collar professionals than blue-cellar workers. Moreover, over time the paucity of working-class MPs has been exacerbated, with the growth of representatives from a professional background like lawyers, businessmen and journalists (Norris 1996a). In recent decades traditional issues about social class have received less attention than concern about the persistent under-representation of women and ethnic minorities (Lovenduski and Norris 1993; Randall 19B7). Worldwide women are 9 per cent of parliamentarians, and 5.6 per cent ofcabinet ministers (United N ations 1995). The proportion ofwomen MPs has declined in recent years, following the abandonment of quotas in Central and Eastern Europe. Identifying the differences and similarities in the pathways to power which cause this pattern, and analysing their consequences, is one of the primary aims of this book. One major stream of literature has been concerned to understand the consequences of the composition of elites for political representation (Birch 1964, 1971, 1993; Pitkin 1967; .Penock and Chapman 1968; Eulau and WaWke 1978; Converse and Pierce 1986; Esaiasson and Holmberg 1996). In the older literature, based on the responsible party model of representative democracy, it was commonly assumed that what members stood for, particularly their party affiliation, was more important than where they came from (Edinger and Searing 1967; Schleth 1971; Matthews 1985:45). Yet a growing body of work has demonstrated that the social background of legislators may matter not just for the symbolic legitimacy of elected bodies, but also for the attitudes and behaviour of representatives. Studies have found that the class, generation. gender and education of elected members produces attitudinal differences within parties in Germany (Wessels 1985: 50-72) and I !q I i'i, , ! I I gested, core party activists in local areas or all grassroots parry members need robe able to exercise influence over their leaders through internal party mechanisms, including determining the selection of party candidates, leaders and policy platforms. This debate poses unresolved issues about whether representatives should be accountable to the whole electorate, to grassroots party members, to a smaller group of party activists, or to the party leadership. The new institutionalism research design The process of recruitment therefore raises significant normative concerns about how the process should operate according to rival conceptions of democracy, and empirical issues about how the process does operate in practice. While the importance of recruitment is widely acknowledged, there have been few systematic studies into the shadowy pathways to power prior to election in most countries. Comparative studies comparing the process in different countries, using a common theoretical framework, remain even scarcer (for detailed reviews of the literature see Matthews 1985; Czudncwski 1975; and comparative studies by Loewenberg and Patterson 1979; Mezey 1979; Norris 1996a; and Gallagher and Marsh 1988). This means that although we have welldeveloped theories of voting behaviour and elections, which have been examined and replicated in many different national contexts, as a result of this neglect it sometimes appears as if candidates are bam by miraculous conception, politically fully clothed, the day the campaign is announced. We lack powerful and well-tested theories which could unify comparative research on candidacies. Building on the literature which is available, this book seeks to develop our theoretical and empirical knowledge by exploring routes to power in a wide range ofparties in advanced industrialised democracies. The core questions explored by subsequent chapters concern who becomes a candidate, how, and why this happens. A thorough review ofthe literature (Fowler 1993) highlights the variety of theoretical and methodological approaches to understanding legislative recruitment, and the fragmentation ofthe field. We have already discussed some ofthe predominant perspectives which have focussed on the insights recruitment provides into political careers, issues of social representation, and the process of party politics. Each approach has provided important clues to understand part of the puzzle of legislative recruit- .~" merit. While establishing a rich foundation on which to build, these approaches need to be melded into a more integrated and comprehensive theoretical framework. The book works within a new institutionalismperspective, an increasingly popular approach which suggests that the atti- 9Introduction: theories of recruitment TIleanalyticalframeuorli oflegislative recruitment '0; echapters which follow work within a broadly common framework, though the stress on different components varies, as befits each particu":;,::tudes and behaviour of individual actors need to be understood within ~~t'tpeir broader institutional context (March and Olsen 1989; Powell and ',imaggio 1991). Many studies have focussed on the formal recruitment . ,focess as set out in legal regulations, constitutional conventions and ,;oifid a1party rules (see, for example, Rush 1969; Ranney 1965). These .sPJdies often assume that the formal processes determine the outcome. ':[be obvious weakness of this approach is that formal rules may have little ,'bearing on u:rormal practices, Constitutions may exercise dejure, not de l!:,cto, authority, The focus on party ~tructures neglects the attitudes, . iriorities and concerns of selectors, whether party leaders, members, oters or non-party financial supporters, or interest groups. Moreover .~ tirutional approaches have also paid little attention to the motivation .~ d experience of candidates. In contrast behavioural approaches have .:S'cd surveys ofelites to understand the attitudes ofparty selectors or can,:dates (see, for example, Gallagher and Marsh 1988; Bochel and Denver ·'~9.83).Yet the micro-behavioural perspective assumes that these attitudes '~~e .static, and generalisable irrespective of the broader context. For ~.~·ample,it assumes that selectors are looking for the same qualifications ~;candidates irrespective of the type ofseat, type ofparry, or type of rules -" verning the process. Yet a change in the procedures, for example a [arty's adoption of affirmative action quotas to boost women's repre?-~tation,may encourage more women to aspire for office, and may alter e: