Who Has Room for Error? Scandal and Accountability Among Minority Candidates

Much literature has separately shown that i) voters hold politicians accountable for misbehavior, and ii) Black, gay, and/or women candidates are at a disadvantage on a campaign trail. Using an original vignette experiment (N=4129), we studied how Black, gay, and/or women candidates were sanctioned for financial and moral scandals compared to their white, straight and/or men counterparts. In other words, we were interested in a mechanism of discrimination: What drives the lessened electability when it comes to voter accountability? Are marginalized candidates penalized more for the same transgressions, or is there a different mechanism driving disparate outcomes? In a review of the literature, we find mixed explanations. Some previous literature on direct discrimination would indicate that Black, gay, and/or women candidates would just be penalized for identity, regardless of whether they have any scandals (Nivem 1998; Terkildsen 1993). However, other studies have even indicated that women candidates (conditional on other gender-conforming practices) are in fact considered more electable than men candidates (Teele, Kalla, and Rosenbluth 2018). This alludes to a strand of literature on activated discrimination, wherein i) interactive candidate qualities (e.g., a Black candidate who is also lesbian) and ii) particular political conditions (e.g., a candidate who is accused of scandal) might instead spark latent prejudices that might not be otherwise stimulated had these traits, behaviors, and/or circumstances been operating on their own. Through our study, we attempt to add to this existing literature on discrimination by: 1) extending literature on direct discrimination to better understand LGBTQ+ candidates; 2) expanding literature on activated discrimination to consider candidates of intersectional marginalized identities (i.e., considering Black women candidates as opposed to considering Black candidates and women candidates separately); and 3) understanding activated discrimination against minority candidates through allegations of scandal. Notably, we find that gay candidates—across gender and race—are less likely to win elected office after a scandal. However, contrary to current literature, we find that this is not due to heightened penalties, but instead lowered baselines. Since returns to vote share are non-linear and subject to thresholds (i.e., the loss in expected win probability for a candidate who sees a drop from 55% to 45% in vote share is greater than from 65% to 55%), candidates with lower baselines can expect to receive worse outcomes, even with uniform consequences. We term this the “Room for Error” Hypothesis, since minority candidates have lessened ability to make mistakes. We find significant evidence of direct discrimination against gay candidates, with gay scandal-free candidates being 5.14 percentage points lower in vote shares. We also find evidence of activated discrimination: For instance, straight, Black women candidates accused of a sex scandal were significantly more likely to be elected than gay, Black, men candidates accused of a sex scandal. This revealed to us again the importance of studying multiple intersecting identities at once. In all, our study adds to the literature on theories of discrimination by manipulating voter accountability and studies the mechanisms by which candidates of intersecting identities (and particularly queerness) are penalized.