Men and Women Candidates Are Similarly Persistent After Losing Elections
In the U.S., female candidates are no more likely to quit politics after suffering an electoral loss than their male counterparts.
Authors
This article was written by and , and was originally published by Proceedings of the National Academy of Sciences in June of 2021Despite public perceptions that women don’t run for office, after the 2016 presidential election, the number of first-time female candidates in the United States soared. These women lost their races at higher rates, however, and the extent to which the increased number of women running for office will translate into an increased number of women in office in the long run remains unclear.
Prior research suggests that women might be more discouraged by losses than men. Some scholars attribute this difference to higher levels of risk aversion among women, while others point to differing levels of rejection sensitivity, contending that women are more likely to opt out of future competition after a loss. Research on women's rejection sensitivity, however, remains inconclusive: the question of whether women are less willing to re-enter a competition after a loss is still unanswered.
In this study, the authors examine whether women are more likely to leave politics after suffering an electoral loss than men. They compare the track records of male and female candidates who narrowly won or lost elections at the state or local level between 1950 and 2018, and use a design-based inference strategy to determine the impact of an electoral loss on a candidate's likelihood of running for office again.
Candidates who lose an election are less likely to run again than candidates who win an election, but this effect does not differ significantly between female and male candidates.
Candidates who win an election are more likely to run again than candidates who lose an election.
- Candidates who win an election are more likely to run again by a margin of 51 percentage points at the state legislative level countrywide, 18 percentage points at the local level in California, and 47 percentage points at the mayoral level countrywide.
Female and male candidates who lose an election have similar levels of likelihood of running again.
- Male candidates who lose an election are less likely to run again by a margin of 38 percentage points at the state legislative level countrywide, 13 percentage points at the local level in California, and 20 percentage points at the mayoral level countrywide.
- Female candidates who lose an election are less likely to run again by a margin of 39 percentage points at the state legislative level countrywide, 13 percentage points at the local level in California, and 32 percentage points at the mayoral level countrywide.
These findings indicate that the difference between women and men's decision to participate in politics occurs at the first point of entry, rather than the point of re-entry. They imply that efforts to increase the number of women in politics should focus on earlier points in the pipeline, while not overlooking candidates who lost their first race as potential recruits for subsequent elections, as women who do choose to run may have a higher risk acceptance and lower rejection sensitivity than women in the general population.
The dataset used for this study was compiled from 141,116 state legislative elections countrywide, 23,466 local elections in California, and 9,132 mayoral elections countrywide, representing a total of 212,805 candidates who ran for office across 22,473 jurisdictions between 1950 and 2018.
The measures for three variables were established. First, the authors established a measure of the margin by which candidates won or lost a race, enabling the identification of winning and losing candidates in elections for both single- and multi-member posts, which have more than one winnable seat. Second, the authors coded candidates' likely gender based on data from the U.S. Social Security Administration's baby name database. Third, the authors measured whether a candidate ran in another election by tracing candidates' track record in both primary and general elections throughout the dataset.
A regression discontinuity design was used to compare candidates who barely won an election to those who barely lost, allowing the authors to identify the effects of winning or losing a race on participation and success in subsequent elections.
