Certain societal groups are significantly overrepresented within politics, including men, ethnic majority groups, and socioeconomic elites. This has fueled debates regarding meritocracy within political recruitment. While meritocracy is desirable, its definition and measurement are contested. The criteria used in theoretical and empirical academic studies differ from those of political parties and voters, as discussed below. Furthermore, there is bias in favor of the male status quo, with all groups preferring the qualities of existing elites. The definition of a “good” politician is therefore highly subjective. Nonetheless, political traditionalists claim that the “best” candidate should be selected even if this leads to significant gender imbalance in parliaments. Yet, if political recruitment is biased in favor of social elites, the overrecruitment of men may not derive solely from merit (Murray Reference Murray2014). Without definitive criteria for evaluating prospective candidates and judging those already elected, we cannot determine whether political recruitment is meritocratic.
Criteria in Political Theory
Political theorists question whether representatives should be trustees exercising independent decision-making or delegates following their constituents’ demands (Eulau et al. Reference Eulau, Wahlke, Buchanan and Ferguson1959). An ideal representative would think independently and make wise decisions, while also consulting with constituents and including their perspectives. Limitations are imposed by the widespread ignorance and heterogeneity of public opinion. Nonetheless, the imperative of reelection stops politicians from ignoring their constituents. A politician can be an authentic representative only through understanding others’ perspectives enough to act on their behalf and with their consent.
Pitkin (Reference Pitkin1967) explores descriptive, substantive, and symbolic representation. Descriptive representation is conceptualized at the group level if representatives’ demographic traits mirror those of society. Individual representatives descriptively represent constituents sharing their demographic features. Substantive representation entails representing interests rather than identities, although the latter frequently shape the former. Descriptive representation can facilitate substantive representation, although it is neither necessary nor sufficient (Reingold Reference Reingold, Wolbrecht, Beckwith and Baldez2008; Swers Reference Swers2005). Symbolic representation ensures that all groups feel included.
Mansbridge (Reference Mansbridge2003) introduces gyroscopic and surrogate representation. Gyroscopic representatives base decisions on their own beliefs and experiences. As external influences seldom shape their opinions, this concept of representation would be problematic without descriptive representation. Surrogate representatives may speak for groups outside their districts who share their characteristics, such as African Americans (Brookman Reference Brookman2013). Both approaches favor diversity and balance across the whole legislature.
Dovi (Reference Dovi2007) provides a normative framework for assessing the quality of representatives, arguing that a “good representative” should display fair-mindedness, critical trust building, and good gatekeeping. Together, these “virtues” are effective for evaluating the democratic efficacy of representatives. However, their benefit is limited when defining meritocratic candidate selection criteria, as they measure the performance of those already in office, and it is harder to demonstrate these qualities prior to election.
Overall, insights from theorists focus on latent attributes that are measurable only indirectly through evaluating observable activity. This disadvantages unelected outsider groups, including women, who lack opportunities to demonstrate their competence.
Criteria in Empirical Studies
Empirical studies focus on concrete qualities that can be measured easily and objectively but are inaccurate predictors of legislator quality. Numerous studies claim that better educated politicians will be more intelligent, knowledgeable, and will make better policies (Baltrunaite et al. Reference Baltrunaite, Bello, Casarico and Profeta2014; Besley and Reynal-Querol Reference Besley and Reynal-Querol2011; Galasso and Nannicini Reference Galasso and Nannicini2011). However, educational attainment may derive from socioeconomic privilege and may also be gendered. The knowledge obtained through advanced qualifications can be provided by aides, whereas true understanding of deprivation will be absent if legislatures are composed only of elites. This awareness gap reduces the efficacy and legitimacy of legislatures, especially if representatives use gyroscopic representation that reflects their own experiences rather than those of their constituents. Likewise, income is a poor proxy for candidate competence. Irrespective of skills, women collectively are paid less than men, and earned income may also stem from privilege and social capital.
Other measures of qualification for office include prior political and professional experience (Murray Reference Murray2010a; O'Brien Reference O'Brien, Franceschet, Krook and Piscopo2012; Verge Reference Verge2011; Weeks and Baldez Reference Weeks and Baldez2015). Prior political experience demonstrates commitment, knowledge, and a track record, though it may reflect longevity rather than talent. It also favors men, as women often avoid entering politics during their child-rearing years. Professional experience may again favor those with elite backgrounds, and certain professions benefit political careers through expedience rather than inherently qualifying someone for office. These measures of merit therefore require refinement.
Party Criteria
Party candidate selection practices vary considerably by culture, internal democracy, ideology, and nationality. Practice may also vary within the party, even when codified guidelines on candidate selection exist. Nonetheless, research has revealed some commonalities in party selectorates’ expectations of prospective candidates, including contributing campaigning resources (Hazan and Rahat Reference Hazan and Rahat2010; Kittilson Reference Kittilson2006). As men typically enjoy greater free time and higher incomes than women, this criterion indirectly favors men. Parties also expect candidates to support their message and demonstrate loyalty. Voters may appreciate party loyalty, as this enables them to vote on national party manifestos and hold governments accountable. However, some constituents may be frustrated if a candidate's loyalties lie with the party rather than the district.
Additional desirable attributes include charisma, eloquence, public speaking skills, good media presence, intelligence, and good networks (Murray Reference Murray2010b). The ideal candidate unites the party faithful while drawing in a wider electorate. These seemingly neutral attributes conceal socially constructed biases against women that manifest themselves through gendered evaluations of candidates, negative media coverage of women, and less (self) belief in women's strengths (Lawless and Fox Reference Lawless and Fox2010; Norris and Lovenduski Reference Norris and Lovenduski1995). Networks may also work against women, especially when men collude together (Bjarnegård Reference Bjarnegård2013; see also contributions by Bjarnegård and Kenny, and Verge in this volume).
Voter Criteria
The qualities sought by voters are somewhat different. Experimental studies reveal interesting voter preferences for certain types of candidates. The perceived sex of the candidate influences how voters evaluate candidates’ traits and policy strengths, with women considered warmer and more honest, while men are seen as tougher and more rational (Alexander and Andersen Reference Alexander and Andersen1993; Huddy and Terkildsen Reference Huddy and Terkildsen1993). Other studies have identified strong voter preference for candidates who are local (Campbell and Cowley Reference Campbell and Cowley2014a; Górecki and Marsh Reference Górecki and Marsh2012), indicating that voters want geographical descriptive representation. Yet, emphasizing localism can been used to close ranks against “outsiders” such as female or ethnic minority candidates (Cowley Reference Cowley2013).
High levels of income, education, and high-status professions do not actually improve candidates’ status with voters (Campbell and Cowley Reference Campbell and Cowley2014a, Reference Campbell and Cowley2014b). Voters do like representatives to spend ample time in the constituency (Campbell and Lovenduski Reference Campbell and Lovenduski2014; Vivyan and Wagner Reference Vivyan and Wagner2015), indicating that voters want their representatives to be accessible delegates rather than trustees. Voters also want someone “like them,” sharing their experiences and interests, as well as supporting more diversity (Campbell and Cowley Reference Campbell and Cowley2014b; Cowley Reference Cowley2013). Hence, descriptive representation may better reflect voter preferences than the elitist conceptions of candidate quality advanced by scholars. Shared backgrounds and genuine understanding of constituents’ lives may better qualify candidates for office than elite status.
CONCLUSION
Within politics, unlike other professions, being the “best” representative may therefore entail communicating honestly and effectively what it is to be “average.” If we seek talented representatives who can also speak authentically for us, this requires politicians whose backgrounds and experiences are common rather than exceptional. Conventional measures of merit matter less than personal qualities, though these are subjective and difficult to measure, entailing two particular risks. First, pervasive stereotyping may lead certain types of candidate (such as women or ethnic minorities) to be evaluated less highly. Second, criteria for evaluating candidates are shaped by path dependence and societal norms, both of which favor existing elites (Bacchi Reference Bacchi1996; Young Reference Young2000). When we think of a (good) representative, we usually imagine a wealthy white male (Norris and Lovenduski Reference Norris and Lovenduski1995). By internalizing norms of class, race, and gender, we unconsciously favor the status quo.
For example, we might prefer a successful businessman to a childcare worker, even though the latter might have more expertise on education, health care, welfare, and family policy. By focusing on managing the economy rather than thinking about the full policy spectrum, we subscribe to the dominant masculinist model of politics. We prioritize the policy areas associated with wealthy men and assume that wealthy men can best advance these issues. As Taylor (Reference Taylor1991, 233) notes, “one way discrimination is perpetuated is by the dominance of elite white men over … what counts as merit” (original emphasis), leading us to overlook important alternative qualities found in women (Franceschet, Krook, and Piscopo Reference Franceschet, Krook and Piscopo2012, 11).
The search for a valid, unbiased, and measurable definition of meritocracy remains incomplete. Theoretical qualities are abstract while empirical measurements are arbitrary, and neither provides sufficient measures of candidate merit. Education, income, and profession are imperfect proxies for latent qualities such as charisma, eloquence, or intelligence, and they cannot disentangle attainment from privilege. In contrast, parties and voters use subjective judgments, increasing the risk of bias. A possible agenda for future research is to decouple theory from current practice and return to first principles. Focusing on what representation is and what it really requires of a candidate would be a first step toward a more objective definition of “merit.”
