Essays on the Candidate Entry Process for LGBTQ+ Individuals in Brazil

Abstract

This dissertation comprises three essays on the LGBTQ+ candidate entry process in Brazil. The first essay seeks to gauge how individuating information about a gay candidate may affect approval of gay people’s right to run. I find that positive individuating information does not increase approval of a gay candidate’s right to run relative to the level of approval expressed for gay people as a group absent any individuating information. I also show heterogeneous effects by religion. The second essay uses a grounded-theory analysis of semi-open interviews to examine how LGBTQ+ candidates talk about barriers in the electoral process. I find that LGBTQ+ candidates share some common barriers with other non-traditional candidates, but other barriers are more specific to the LGBTQ+ electoral experience. LGBTQ+ candidates as a group report more barrier-overcoming tactics than barriers themselves, with transgender candidates indicating more reliance on their internal sense of self-efficacy than on encouragement from parties and voters. The third essay uses the 2020 local elections in Brazil to investigate whether and the extent to which party elites withhold non-financial support from LGBTQ+ candidates, by way of the mnemonic quality of candidate-specific ballot codes. Drawing from a microcensus of LGBTQ+ candidates, I find that LGBTQ+ candidates compete with less memorable and idiosyncratically derogatory ballot codes, with the least memorable codes being assigned to transgender candidates. The fourth essay, a coda, assesses faith-based homophily's effect on married and unmarried gay candidates’ electability. Conjoint-experimental evidence shows that voters generally penalize gay candidates and suggests that faith-based homophily does not increase religion-branded gay candidates’ electability relative to non-religious gay candidates.