Abstract Explaining persistent gender asymmetry in women’s descriptive representation necessitates identifying the barriers to women’s opportunities and willingness to contest elections. I extend recent scholarship examining the gendered consequences of political violence for women’s political participation and descriptive representation. I specifically argue that political violence observed in the months preceding an election introduces distortions into the candidate selection process, thereby reducing women’s share of the pool of candidates competing for seats in the legislature. I empirically evaluate this hypothesis by combining a new cross-national dataset on women’s candidacy rates in 555 multiparty national legislative elections held in 136 countries between 1990 and 2022 with temporally fine-grained violent events data. Even after accounting for multiple confounding factors and addressing unobserved heterogeneity, the results demonstrate a robust negative association between political violence occurring during the candidate selection process and the rate at which women contest national legislative elections. Additional exploratory analyses suggest that these effects are driven principally by violent events occurring within the context of civil conflicts.
Reed M. Wood (Tue,) studied this question.