The Feminist Paradox:How Labels Keep Women Candidates from Equal Representation

Abstract

Although theories of descriptive representation often posit that citizens are drawn to individuals from groups with which they identify, and despite the fact that women remain largely politically underrepresented in the United States, female voters do not favor ingroup - female - candidates. Why is shared gender between voters and candidates such a weak attractive force compared to other group attachments? I argue that the female electorate is divided into two gender subgroup identities: feminists and non-feminists. Feminists are a homogeneous group and share a feminist gender ideology that focuses on fostering gender equality. In contrast, non-feminists are a very diverse group that includes three distinct types characterized by differing gender ideologies: 1) non-labelers, who endorse core egalitarian values and principles of feminism (i.e., feminist gender ideology) yet eschew the feminist label, because they are wary of the social stigma attached to it, 2) gender individualists, who acknowledge gender inequality but support individualism and self-determination rather than collectivism and public policies to promote gender equality and combat discrimination, and 3) gender inegalitarians, who deny the existence of gender inequality and discrimination and/or oppose both individual and collective efforts to address them. I test my hypotheses on the role played by gender subgroups by relying on both quantitative and qualitative evidence from two surveys, the 2016 American National Election Study and a survey fielded on YouGov in 2018, and two experiments, one conducted on Amazon’s Mechanical Turk in 2017 and the other conducted on CloudResearch/TurkPrime in 2019. My results demonstrate that support for gender equality, which both feminists and non-labelers share, is much more widespread than acceptance of the feminist label. The choice to self-label as a feminist vs. a non-feminist has important political implications. Among both men and women, feminist identity is largely and significantly associated with considering the election of more women to political office as important. In contrast, non-feminist identity strongly undermines support for women in political office, all else equal. In the context of the 2016 U.S. presidential election, a mayoral election, and a legislative primary, I show that feminist voters in both the Democratic and the Republican Party largely favor ingroup (feminist) candidates over outgroup (non-feminist) candidates of both genders, but they also favor non-labeler over individualist candidates. Despite their group heterogeneity, non-feminists of both parties display highly coherent preferences: they favor individualist candidates over non-labeler and, particularly, feminist candidates, regardless of candidate gender. In sum, labels related to gender subgroups divide the U.S. public much more than the actual idea of gender equality, particularly the female electorate. The divides among feminists, non-labelers, individualists, and inegalitarians are distinct from and cut across those based on gender group membership, gender identification, and political ideology. They influence how individual voters perceive and evaluate candidates, interpret the political spectrum, and decide to cast their ballot. Explicit association with the feminist label on the campaign trail thus puts feminist candidates at a severe electoral disadvantage and is likely to be riskier and costlier than touting a gender egalitarian policy agenda. The label appeals to one rather small gender subgroup, feminists, but alienates a larger and more diverse one, non-feminist voters. It is the aversive reaction to the label, rather than a rejection of the substance of their politics, that drives many women away from candidates of their own gender group.