Political glass cliffs arise when candidates from low-status groups disproportionately run for less-winnable seats. The burden of these worse odds has been shown to negatively impact election outcomes, slowing progress toward fair political representation. Relying on research suggesting signaling motives for glass cliff appointments, we investigated the potential of these political party decisions to persuade voters in the context of evolving social norms. We hypothesized that party differences in the signaling context underlie variation in the magnitude, impact, and dynamic evolution of elective glass cliff conditions over time, leading to more rapid improvements in the representation of women and ethnic, racial, and immigrant (ERI) minorities in left-leaning versus right-leaning parties. We examined glass cliff candidacies in elections for the French National Assembly from 2002 to 2017. Relying on three measures of seat winnability, we adopted a multiple group structural equation approach to investigate whether variation in glass cliff conditions and their effect on outcomes differed by election year and party belonging. We found larger glass cliff disadvantages for right-leaning women and ERI candidates compared to left-leaning. While the magnitude of glass cliffs for women decreased over time as representation increased, this link for ERI candidates was less clear. Outcomes demonstrate that dynamic glass cliff conditions can be a major obstacle on the road to representational fairness in politics. We argue that because the impact of glass cliffs can depend on party-dependent variation in the signaling value of women and ERI minorities, it is essential to focus more on this issue for socially conservative political parties and for all political parties in elective contexts where low-status candidates remain largely underrepresented.