Which group constitutes Californias most underrepresented minority in terms of elected officials

Any discussion about descriptive representation in the United States is guided by the Voting Rights Act of 1965 (and associated laws and amendments), which establishes single-member districts as the standard for ensuring representation of racial and ethnic minorities.4 In many contexts, single-member districts were an improvement over the multi-member systems used before them (Bullock and MacManus, 1993; Casellas, 2009; Engstrom and McDonald, 1981; Grofman et al., 1986; Trebbi et al., 2008; Welch, 1990; Richardson and Cooper, 2003). In part, the positive impact of single-member districts was a legacy of the use in the American South of a non-proportional multi-member system (block voting, often called “at-large” voting in the U.S. local context) to limit African-American representation. Under block voting, a cohesive plurality of voters can elect candidates to all seats up for election. By contrast, in single-member systems like plurality or majority runoff, districts can be drawn to ensure a minority population makes up a majority (or a sizeable plurality) in at least one district, thereby affording a greater opportunity for that population to elect a candidate of choice (assuming that population votes as a bloc). As the decades pass, studies suggest that the efficacy of single-member districts in the U.S. has declined (Bullock and MacManus, 1993; Welch, 1990; Bullock and MacManus, 1993; Welch, 1990) and may be limited for some racial and ethnic groups, such as Latinos (Casellas, 2009; Karnig and Welch, 1979; Casellas, 2009; Karnig and Welch, 1979), yet single-member systems remain accepted as the standard means by which to protect minority representation in the United States (Brischetto, 1995).

By contrast, numerous studies of women's representation, internationally and in the United States, show that more women are elected to legislative office in multi-member electoral systems, whether proportional or not, than in single-member systems (Barkman, 1995; Kaminsky and White, 2007; Kenworthy and Malami, 1999; King, 2002; Paxton et al., 2010; Rule, 1987; Trounstine and Valdini, 2008; Crowder-Meyer et al., 2015; Fortin-Rittberger and Rittberger, 2014; Matland, 1993). At least one study found the electoral system to be the most determinative predictor of the percentage of women in legislative office (Rule, 1987).

Cross-national studies generally conclude that more women are elected to chambers elected using party-list proportional systems than chambers using non-proportional multi-member systems, like the block vote. Party recruitment strategies appear to be key to explaining why more women are elected in party-list proportional systems, with evidence indicating that, in many electoral contexts (especially partisan elections in which voters vote for their party rather than for a candidate), women appearing on the ballot are not electorally disadvantaged (Darcy and Slavin Schramm, 1977; Darcy et al., 1994; Lefteris, 2016; Dolan, 2014). Rather, the underrepresentation of women tends to stem from a dearth of female candidates either choosing to run in winnable seats or being recruited and nominated by political parties in winnable seats (Lane, 1995; Lawless and Fox, 2012; Matland and Studlar, 1996).

Internationally, there is debate as to whether a proportional electoral formula or higher district magnitude is more important to the election of women (Welch and Studlar, 1990; Matland and Brown, 1992). In the United States, only a handful of the more than 10,000 election jurisdictions use proportional systems for their elections, but hundreds of state and local jurisdictions (including about 450 county commissions and boards of supervisors) do use multi-member districts. In the U.S., it is widely accepted, by academics, activists, and practitioners, that more women tend to be elected in non-proportional multi-member districts than in single-member districts (Darcy et al., 1994; Hogan, 2001; Kaminsky and White, 2007; King, 2002; Matland and Brown, 1992; Paxton et al., 2010; Pyeatt and Yanus, 2016; Trounstine and Valdini, 2008; Vermont Legislative Research Service, 2011; Darcy et al., 1985; Zimmerman, 1994).5 King (2002 at 163) notes that “[t]here is little question regarding the relationship between the use of MMDs [multi-member districts] and the representation of women in [U.S.] state legislatures.” This relationship exists even within states that elect a legislative chamber using a mix of multi-member and single-member districts, with a greater proportion of women winning in the multi-member districts within a chamber than in the single-member districts (Darcy et al., 1994 and King, 2002).

Party recruitment practices may explain part of this relationship. The decentralized, county-based political parties that recruit candidates for U.S. state legislative office may adopt differing strategies in the multi-member districts within their jurisdictions than in the single-winner districts. This explanation notwithstanding, the relationship between multi-member districts and more women being elected is widely accepted in the United States, with Crowder-Meyer et al. (2015) offering a rare counterpoint.

On balance, the literature suggests that in the United States (where party-list proportional systems do not have widespread acceptance), single-member systems advance descriptive representation for people of color while multi-member systems, including the block vote, advance descriptive representation for women (see Trounstine and Valdini, 2008)). Designing American electoral systems to simultaneously ensure descriptive representation for women and racial and ethnic minorities appears difficult. Increasing representation of minority women, one of the most underrepresented groups, is especially complicated (Trounstine and Valdini, 2008; Lien, 2015). Research shows that minority men tend to benefit most from single-member districts, with some studies showing that minority women fare better in multi-member systems (Darcy et al., 1994; Trounstine and Valdini, 2008). Further complicating matters, electoral systems appear to affect men and women from different minority groups differently (Swain and Lien, 2017; Trounstine and Valdini, 2008).

As part of an effort to disentangle these strands, we explore the impact on descriptive representation of the reform of the electoral formula used in non-partisan single-member systems at the municipal level. In these non-partisan elections, political parties and their recruitment practices ought to have less impact on candidate supply, recruitment, and victory and the other effects of electoral systems reform should be clearer. Taking the opportunity afforded by the adoption of the alternative vote (AV) in several California Bay Area cities in the 2000s, we use a differences-in-differences design to explore whether the recent adoption of AV caused an increase in female, minority, and minority female candidacy and whether those candidates were more likely to win than under the previous plurality or majority runoff electoral formulas. As an ordinal system, in which voters rank candidates in order of preference, and a system that is more resistant to vote splitting and the spoiler effect than other single-member systems, AV might encourage different candidate campaign strategies or create environments that reward different candidate attributes.

Given that AV is compatible with the entrenched system of single-member districts in the U.S., it is valuable to know what effect the adoption of AV had on the representation of women, minorities, and minority women. Our results show that reform of the electoral formula used in single-member systems can have a significant impact on descriptive representation. AV never caused a decline in descriptive representation. For women and minority women, in particular, AV appears to have improved descriptive representation, with more female and female minority candidates winning elective office.

In light of our findings, we suggest that greater consideration should be given to the electoral formula when studying electoral systems and descriptive representation. While AV did not produce a large improvement in descriptive representation for all studied groups, fine-grained study of different electoral formula may uncover a system that does.

Studying electoral formula reform

The apparent dichotomy between multi-member districts being associated with the election of more women and single-member districts being associated with the election of more ethnic and racial minorities may, in part, be caused by a limited set of comparisons and a lack of precision about the electoral systems, and their formulas, being contrasted (see: Richardson and Cooper, 2003). American literature tends to contrast one type of multi-member system, the block vote, against two forms of

The alternative vote in American elections

A cluster of four cities in the California Bay Area use AV in their non-partisan municipal elections, with a total of fifty-two offices in the Bay Area being elected by AV. All of these offices are elected in non-partisan elections, where political parties are not strong gatekeepers to candidacy. Indeed, in line with a 1986 state constitutional amendment initiated by the people, political parties are prohibited from formally endorsing any candidate.7

Data and methods

To test the effect of the introduction of AV on descriptive representation, this study draws on a dataset of contests for city council and executive office seats in eleven California cities. Each observation in the dataset is a seat that was up for election between 1995 and 2014 in the selected cities.

The eleven cities include the four California cities that adopted AV during the period of the study (the “treatment” cities), as well as seven control cities that did not adopt AV (the “control”

AV and running for election

Adopting AV was associated with a 9-point increase in the predicted percentage of candidates who were from racial or ethnic minority groups, holding all else constant (p < 0.05; Table 3). When treatment cities adopted AV, the predicted percentage of racial or ethnic minority candidates increased from 17.2 to 25.6 (p < 0.05; Fig. 1); during that same time period, the predicted percentage of racial or ethnic minority candidates in control cities decreased slightly from 12.0 to 11.4 (Table 3). We

Discussion and conclusion

Our study of the adoption of AV for local elections in the Bay Area shows that the introduction of AV was associated with a 9 point increase in the percentage of minority candidates, a 0.16 increase in the predicted probability that a female candidate will win office and a 0.14 increase in the predicted probability that a female minority candidate will win office. We did not find a significant effect of AV on candidacy for women and minority women or on the probability of victory for minority

Funding

This work was supported by the Women Donors Network and the Democracy Fund (grant number G-201505-12530/Governance).

Conflict of interest declaration

We wish to draw the attention of the Editor to the following facts which may be considered as potential conflicts of interest and to significant financial contributions to this work.

At the time the work on this submission was done, Sarah John and Haley Smith were employed by FairVote, a 501(c)(3) nonprofit that describes its purpose as “a nonpartisan champion of electoral reforms that give voters greater choice, a stronger voice, and a representative democracy that works for all Americans.”

Acknowledgements

The authors wish to thank Kathryn Gansler, Viviana Gonzalez and Michael Patison for their research assistance.

Cited by (11)

Recommended articles (6)