She Persisted: Gender,Electoral Loss,and the Decision to Run Again

Abstract

Losing elections is a political reality for most candidates. While much fanfare was made of the historic number of women who won in 2018, three times as many women ran and lost. And although we know much about the reasons that motivate women to run in the first, far less is known about the decision-making processes of women who lose and run again. How do women who were once motivated rebound for another campaign and election? We address this question, working with a unique, exploratory data set of interviews with women who ran for Congress in 2018 and lost. Our analysis offers some important insights on how women who rebound think about their loss, and their decision-making processes. We find compelling evidence that women’s paths to office continue to differ from men’s, reminding us of the shortcomings of a one-size-fits-all model for candidate emergence or reemergence. We conclude with the foundation of a gendered theory of electoral persistence.

Keywords

candidate persistence women candidates congressional elections

I’ll be back. I’ve heard that you’re not truly a politician until you lose. Because I’ll tell what what—there’s a lot more lessons in losing than there are in the [campaign] process.

—Female congressional candidate who lost her 2018 primary.

Losing elections is a political reality for most candidates. While much fanfare was made of the historic number of women who won in 2018, three times as many women ran and lost. And although we know much about the reasons that motivate women to run in the first place—policy concerns (Carroll and Sanbonmatsu 2013; Schneider et al. 2016); positive role models (Silva and Skulley 2019); someone asked her (Carroll and Sanbonmatsu 2013)—far less is known about the decision-making processes of women who lose and run again. How do women who were once motivated rebound for another campaign and election?

We address this question, working with a unique, exploratory data set of interviews with women who ran for Congress in 2018 and lost. Our focus on losing candidates is important for building a more complete picture of women’s experiences on the campaign trail. We argue that scholarship that focuses solely on the winners potentially paints a skewed portrait of female candidates’ perspectives and experiences. Just about everything we know about female candidates comes from analyzing those who won or through making inferences about the qualities of those who emerged victorious (Carroll and Sanbonmatsu 2013; Palmer and Simon 2012; Pearson and McGhee 2013). Given most women are unsuccessful, we know little about how they view their initial run, their thoughts about running again, or how this influences their paths to office.

Below, we begin with a review of the literature on candidate persistence and identify a number of theoretical possibilities for predicting which women will decide to run again after losing and which women are less likely to do so. We then turn to an analysis of in-depth qualitative interviews with 53 women who lost in 2018 to tease out some key distinctions between these two groups of women. Given the limitations of our data, our conclusions are suggestive, but lead us to development of a theory of gendered electoral persistence that we will continue to refine as we move on to the 2020 elections.

Literature Review

An extensive literature compares political ambition between men and women and either concludes that under certain circumstances women are less politically ambitious than men (Bledsoe and Herring 1990; Costantini 1990; Fulton et al. 2006), or that gender has little effect on candidate ambition (Carroll 1985). The literature on candidate emergence, on the other hand, concludes that women perceive themselves to be less ready, less qualified, and less likely to win office (Fox and Lawless 2004, 2011).

Little is known about how electoral loss influences political ambition, and even less is known about women who lose elections. Extrapolating from previous research in political science, behavioral economics, and psychology, however, we identify three potential models for explaining when women might bounce back and run for office after an initial loss. Given the scarcity of models that focus on women who decide to run again after an electoral loss, a priori we do not presume to differentiate between the three theories offered above, nor do we posit the theories are mutually exclusive. Rather, we employ the models to probe multiple components of women’s campaigns and develop a more nuanced understanding of the woman who loses and decides to run again.

Rational Choice Model

The first model follows rational choice theories of political ambition—a simple cost/benefit analysis of the election, and an assessment of her likelihood of winning (e.g., see Jacobson and Kernell 1983; Taylor and Boatright 2005). This model suggests an objective assessment of the first loss when potential candidates look at the objective data—Was I out-fundraised? Were the margins of loss close? Was 2018 a unique opportunity not soon to be repeated? Can I capitalize on the momentum I created in the first race?

There are a few scholarly examples of testing the rational choice model as it affects electoral persistence. Carsey et al. (2003) do not examine gender as an explanatory variable but find that repeat candidates are more likely to have been serious contenders in the first race, and value the seat. More recently, Thomsen (2018) investigates gender differences specifically and finds little evidence of a depressive effect of loss when the electoral context—vote margins, open seats, fundraising—is taken into account.

Together, these works provide some evidence of an objective assessment of likelihood of winning to be suggestive of a future run. We expect women who performed “well” in their first run—were successful in fundraising, came close to winning the primary, outperformed their party in the general election—will be more likely to run again.

Psychological Model

The second model mirrors the recent research on gendered candidate emergence and focuses on the psychological effects of competition and loss. A number of articles have examined gender differences in experimental settings and labor markets. For example, Buser (2016) looks at the gender differences in winner-take-all experiments, which he argues mimics situations like competitive job searches and promotions, where there are clear winners and losers. Buser (2016) finds that after facing a setback in these types of competitions, men react by picking a more challenging target, while women react by setting their sights lower. Gill and Prowse (2014) similarly find that women and men react differently to winning and losing during competition. Women in particular are more likely to react negatively when forced to compete, and when they encounter random setbacks unrelated to performance. They conclude that this risk aversion may explain why women enter into competitions less often and why women may be less inclined to pursue careers that involve multiple rounds of competition for advancement.

Recent research by Wasserman (2018) and Vallejo (2019) find some support for this psychological model in electoral settings. In her quantitative analysis of local and state offices in California, Wasserman (2018) concludes that losing has a depressive effect on women’s electoral persistence and theorizes that the differential discouragement women receive from external feedback—That is, other’s questioning their qualifications—is particularly detrimental. Vallejo (2019) examines Brazilian municipal elections and finds losing has a negative effect on political ambition regardless of gender, but that the effect is stronger for women, positing that “women tend to attribute failure to internal factors . . . and success to external factors, while men tend to do the opposite” (Vallejo 2019, 4).

In this paper, we focus on the role of doubt in particular. Research on candidate emergence finds many would-be candidates doubt their ability to run and win and are labeled weaker candidates by others (Fox and Lawless 2004). Rebound candidates, however, may experience this doubt differently. On the one hand, we know they overcame this doubt the first time they ran; on the other hand, those feelings of doubt may have been confirmed—they lost. We posit, however, that women who decide to run again will conceptualize that doubt differently than those who are less willing. In particular, we expect women who entertain another run will be more likely to attribute their loss to external rather than internal factors.

Relationally Embedded Model

Our third model builds upon the literature that identifies the importance of relationally embedded candidacies for women. Carroll and Sanbonmatsu (2013, 45) develop the theory and argue that compared with men

women’s decision making about office holding is more likely to be influenced by the beliefs and reactions, both real and perceived, or other people and to involve considerations of how candidacy and office holding would affect the lives of others with whom the potential candidate has close relationships.

Carroll and Sanbonmatsu (2013) examine this theory comparing women and men’s initial decisions to run and find women were more likely to run when their children were older and they felt supported in their family relationships. Moreover, their ambition to run and their subsequent candidacy was most often attributed to someone asking them—family members, friends, or political leaders.

We posit that these relational considerations could play a key role in determining rebound candidacies as well, particularly as women think about their loss. Specifically, we expect women will look to their relationships as key in their decision-making and will be more likely to run when they feel supported by strong networks of relationships. These networks might include their family, campaign team, political party, and voters.

Methods

To better understand the experiences of the women who ran and lost in the 2018 congressional elections, we attempted to talk to them directly. Between August 2018 and February 2019, we emailed all female primary and general election losers, excluding incumbents, for whom we could find email addresses (n = 274), inviting them to share their story with us.¹ In all, we interviewed 57 women (54 House and 3 Senate candidates) who lost in either the primary or general election and include 53 women for whom we have usable data here.² We promised all of the women confidentiality and replace their names with codes throughout the paper.³

We asked each woman approximately ten open-ended questions about their experiences running in 2018, beginning with a question about her motivations to run, and subsequently probing for factors others have identified as barriers to women entering the race, including previous experience and apprehensions about running for office, party support and encouragement, and family responsibilities. We also asked specifically about future plans, inquiring about whether they intended to run again at some point in the future (see Online Appendix A for interview instrument).

Qualitative interviews are ideal for connecting campaign experiences to future ambitions as they enable us to uncover rich details of how each woman processed lessons from her own campaign and used that information to weigh a future candidacy. By asking open-ended questions, we invited each woman to identify the most important pieces of her story and to tell us in her own words how she interpreted and processed the campaign experience and felt about running again. We also had the opportunity to ask follow-up questions and finished each interview with an invitation to the interviewees to tell us anything else they deemed important but not yet covered in our conversation. Our interviews ranged from 15 to 111 minutes, averaging forty-one minutes.

Because we were able to cover so much ground with our interviewees, we find these data very well suited to interrogating the three theories discussed above. And as mentioned previously, we do not assume that the theories are mutually exclusive, but rather can guide us in identifying which factors, or which combination of factors, are present or absent in these women’s stories and future plans.

Analysis

To distinguish between our interviewees, we draw primarily on their response to our standard question about their future intentions to run. We code as “yes” all of those who told us they had already declared their next candidacy, those who had not yet declared but indicated they were definitely planning to run again, and those who initially answered they were considering the possibility, and then went on to share information about how they were actively preparing to run. Nearly 60 percent of our interviewees fall into this category. The other 40 percent of our interviewees expressed greater reluctance to run again, often indicating that they had not ruled out the possibility but had no current plans. Of these women, about 20 percent fell into the not likely to run category, explicitly telling us they were not planning to run again or were highly unlikely to do so.

To tease out the factors shaping their decision for or against a future candidacy, we began by analyzing the entirety of their response to our question about future plans, including any follow-up responses that emerged after we probed with clarifying questions. Some of the women shared details about their future plans before we asked, and so we include these passages and follow-up questions as well. Others returned to the question later in the interview, saying things like “The next time I run, I’ll make sure to start fundraising earlier” or “I forgot to mention just how supportive my husband was.” Finally, various other questions elicited in-depth responses about factors that weighed heavily on their future plans, including lessons learned on the campaign trail, any necessary skills or qualifications they deem important for female candidates, and apprehensions they confronted before deciding to run. As such, we carefully coded the entirety of the transcripts to ensure that we were fully capturing the various dynamics that factored into their decisions about future plans.

Rational Choice Model

We begin with the rational choice model, investigating the standard electoral factors that should theoretically distinguish between those who intend to run again from those who are more reluctant. As previously discussed, the rational choice model posits that strategic candidates will assess their electoral situation as a cost/benefit analysis and run again when the benefits outweigh the costs. For our purposes, we employ various competitiveness measures to gauge the likely benefits and costs of running again: all else equal, we expect that women who fared best in their previous races should be most interested and willing to run again.

In particular, we consider the specific electoral circumstances in our female candidates’ districts, comparing and contrasting their performance against those with whom they competed in the primary and/or general election. For the candidates who did not advance beyond the primary, we gauge their competitiveness with three measures: the vote and financial gaps that separated them from the eventual primary winners and how close they came to something we refer to as their “expected” vote in the primary.

For the first measure, we simply compute the raw difference between each woman’s percentage of the vote received and the percentage of the vote received by the primary winner in her party. All else equal, we expect that smaller gaps will reflect a more competitive election. For the financial gap, we calculate a similar measure: the difference between the financial expenditures of our interviewee’s campaign compared with those of the eventual primary winner. Again, we expect that smaller gaps will indicate a more competitive election. Our third measure roughly gauges whether our candidate did better or worse than might have been expected. To compute this measure, we divide hundred by the number of candidates competing in the primary to develop a baseline for “expected vote.”⁴ For example, the “expected vote” in a primary with five candidates would be 20 percent for each candidate, all else equal. Using the baseline expected vote, we subtract the actual vote received in order to determine whether our candidate did better or worse than the baseline, and by what magnitude.

As Table 1 shows, there are no significant differences between our two groups of candidates on any of these measures. However, the substance and direction of the differences provides some evidence that women who are interested in running again performed slightly better than those who are less certain about their future plans, as we expected. Focusing first on their actual vote share in the primary, women who intend to run again did slightly better in their primary elections, averaging about 6 percent more of the vote (21.4% vs. 15.4%) and trailing the eventual winner by a smaller gap (by 22% and 27%). Looking at their expected vote share, the pattern is similar: women who are keen to run again averaged about two percentage points better than we might have expected while those in the not so certain category performed about three percentage points worse than expected.

Table 1.

Electoral Competitiveness Statistics, By Decision to Run Again.

	Yes, will run again(n = 30)	Maybe/not likely to run again(n = 23)
Primary election losers
Vote percentage	21.4	15.4
Vote gap from primary winner	−21.97	−27.2
Gap in money spent, compared with primary winner	−$1.5 million	−$805,000
Relative vote performance (difference from expected vote share)	+2.13	−2.7
General election losers
Vote percentage	33.8	38.6
Vote gap from eventual winner	−30.5	−22.1
Relative performance (difference from expected two-party vote)	−0.2	+5
Gap in money spent, compared with general election winner	−$683,000	−$3.8 million

But turning to the money gap, the findings are paradoxical: women who are more certain about running again in the future lagged further behind in the money race than did those who are less certain, being outspent, on average, by $1.5 million versus $805,000 in their primaries.⁵ When we examine the ratios by which our interviewees were outspent, the findings remain much the same. Women who intend to run again were outspent, on average, by a remarkable ratio, with the primary winner outspending them by a ratio of more than 100 to 1. On the other hand, women who are less certain about their future plans were outspent, on average, by a ratio of 18 to 1.⁶

Turning to the women who won their primaries, we develop similar district-specific measures of competitiveness to assess their general election performance: their vote and spending gaps in comparison with the eventual election winner, and an estimate of competitiveness based on the expected two-party vote in the district. Again, we expect smaller vote and spending gaps will indicate more competitive circumstances. For the expected vote, we begin with each candidate’s general election vote share and subtract the percentage of the vote her party’s presidential candidate received in the district in 2016.⁷ Those who outperformed their party score positively on this measure, while those who underperformed score negatively.

Beginning with the vote share measures, we find that performance in the general election is not a particularly good predictor of intentions to run again. In fact, the women uncertain about launching future candidacies actually garnered a larger share of the vote in the general election than those who intend to run again (38.6% vs. 33.6%), and so their corresponding vote gap from the winner is smaller (22.1% vs. 30.5%). In addition, these women, on average, outperformed their party’s presidential candidate and did better, on average, than those women who have decided to run again. Women on the fence outperformed their party by over five percentage points in the general election while those who are planning to run again slightly underperformed their party’s presidential candidate (by less than 1%).

However, the women with future intentions to run did much better in terms of fundraising. Although they were outspent by nearly $700,000 in the general election, such a figure is a much less daunting gap than the $3.8 million average deficit for women who are less certain about running again.

In sum, these findings paint a mixed picture in terms of how overall competitiveness shapes these women’s future decisions to run. Among those who lost their primary, those who intend to run again score higher in actual and expected vote shares yet lag behind in fundraising totals. For the general election, the reverse is true: those who intend to run again appear more competitive when it comes to fundraising but perform less well when it comes to actual and expected vote share. As such, their electoral experiences and competitiveness of their races do not neatly predict women’s future intentions to run, suggesting that additional factors are at play in shaping their future plans.

Our interviews provide alternative evidence that these female candidates did evaluate the costs and benefits of running again but focused on indicators not historically included in the rational choice model, which has most often been tested among men. For example, when we examine our interviewees’ response to our question about their future plans, the most striking difference between the two groups is the extent to which they emphasize either pros or cons in their responses. Fewer than 10 percent of the women intending to run again focused on the downside or cons of running, while more than half of the less certain women emphasized costs that they were not sure they wanted to face again or that seemed insurmountable. Their initial responses to the question about their future plans were much more likely to include discussion of some of the obstacles they encountered or reasons they would think twice before running again: the impact on their family, an out of touch party, or a hopelessly gerrymandered district. In contrast, the women who intend to run again often spoke at length about the benefits, or positive forces propelling their next run. As such, we believe throwing out the rational choice model may be premature, especially with the nature of our sample. Rather, we consider an expansion to the rational choice model later in the paper, after first examining the other models.

Psychological Model

The second model of rebound candidacies focuses on the psychological impact of loss. In particular, we look at the role that doubt—both internal and external—plays on the likelihood of running again. One of the primary explanations for women’s reluctance to run is that they often doubt their own qualifications or spend far more time amassing experience in lower level offices before making a bid for an office like Congress (Bledsoe and Herring 1990; Fox and Lawless 2004).

But how might an electoral loss shape the subsequent decision to run? The psychological model predicts that internal self-doubt will have more of a depressive effect than external doubt since women who look inward and find fault will be less inclined to want to repeat the experience. On the other hand, women who attributed their losses to external factors, such as their status as newcomers navigating uncharted territory or being greatly outspent by an opponent, should be more open to the possibility of running again.

We find modest support for the psychological model. A telling delineator between women planning another run and those who were more uncertain was how they understood their loss. For women intending to run again, not a single one appeared to internalize the loss. Rather, the way in which they characterized challenges faced on the campaign trail suggests a healthy ego and ability to externalize setbacks rather than internalize as their own shortcomings. They pointed to external factors such as getting into the race too late in the game, not having sufficient campaign staff or funds, and recognizing their party had essentially given up on the district. In other words, it was them, not her. One such response highlighting the role of external factors came from a first-time candidate who lost in her primary:

I’m still coming to terms with the grief of losing and the grief knowing that ultimately, in my heart and in my mind, I know that it was money that played a hand in who was victorious. (M2)

To be fair, most of the women expressing greater uncertainty about their future plans likewise focused on external factors contributing to their losses, mentioning many of the same factors listed above. In addition, they pointed to dirty tricks waged against them as giving them pause about running again. For example, one candidate’s social media was hacked just as she was rising in the polls in a crowded primary. Not only did she spend precious time attempting to recover her account but had to reboot her entire communications strategy with a rapidly approaching primary. In reflecting upon the loss, she was clear about the devastating impact of being hacked on the eventual outcome of the race:

I’m also very frustrated because I feel wronged. My formula was so good and in [a previous race], it worked perfectly . . . And it should have worked except for that somebody figured me out and just shut me down. (K3)

In contrast, when we focus on the small number of women who indicated they would not run again, we find the strongest support for the impact of internal doubt. One first time candidate confided that she was reluctant to run again after losing because her loss demonstrated that she wasn’t, in her words, “a proven winner.” As she continued, her “not a proven winner” mindset made it difficult to ask people for money, an essential part of campaigning. She went on to share how she had successfully raised funds for a non-profit organization but found it much easier to ask people to give for a cause, not to her personally. And in the wake of a loss, she explained the task of asking for funds would not be psychologically easier for her.

Another woman, who was also the target of dirty tricks by her own party, had this to say about a future run:

the brutality of my race was such that I think people run for president and go through less than I was put through. And I don’t, I don’t think I have that in me, you know? I didn’t want it enough and I’m not mean enough. (L1)

As such, these two women’s misgivings about running again provide some support for the psychological model: they appear to internalize their loss and own shortcomings in deciding against future candidacies.

Yet on the whole, and contrary to the experimental evidence on women’s loss (Buser 2016; Gill and Prowse 2014), we mostly find evidence that even in the face of loss, women had positive self-evaluations of their performance, and greater motivation to be better the next time. Many of these women mentioned the things they did well on the campaign trail that helped shore up their confidence: they became better fundraisers and public speakers, learned to trust their own instincts, performed better than expected, and sometimes recognized that they had underestimated their own abilities when beginning the campaign. As one said,

I had just about every odd against me and still managed to come in third out of seven or eight candidates. So we surprised a lot of people. And one of the things that I heard . . . at the candidate forum, it was pretty consistent people were saying I was winning over votes and outshining everyone. (N1)

Despite plentiful comments about facing double standards on the campaign trail, these women were able to brush off such annoyances. More than half of these women shared stories of being underestimated or dismissed by voters, media and members of their own party but seemed to take it in stride. In fact, one remarked that such gendered treatment only strengthened her resolve:

to be a woman doing what I’m doing in a very rural area . . . you know, sexual harassment, condescending communication. I’ve seen all of it and it’s frustrating. But that stuff just empowered me to keep going. (J2)

Others explained that they emerged stronger than before, fearless, and ready for the next challenge. As one woman explained, she initially worried about her personal appearance and how photos might portray her as an “ugly woman.” She admitted having a “deep, deep concern . . . that I would appear to be this angry, ugly, wrinkled, down at the mouth, serious, angry woman.” Yet, much to her delight, the campaign dispelled such worries and led to increased self-confidence:

I’m incredibly comfortable in my own skin now . . . I am photogenic. I take beautiful photos, even when I’m angry, when I’m pointing my finger, or I’m sneering . . . I am a very talented and very intelligent person. I had a deep understanding of the issues that were involved . . . [and people] respected my knowledge. (W1)

In sum, we find evidence that an electoral loss has psychological effects on women candidates, but not in the way expected by previous research. Losing the election could have been interpreted as justification of the doubts women dealt with the first time around, and for a couple of women who decided not to run again, this was true. However, for the vast majority of the women we spoke to, the loss was mobilizing, and they saw themselves as better prepared for the next battle.

Relationally Embedded Model

Last, we investigated how women situated their run within their relationships to their “team,” and their community. If support from their networks is key to women’s initial decisions to run, then perhaps the promise of continued support, or the enduring quality of relationships built along the way will make all the difference in deciding to run again. In particular, we assessed how the women discussed their relationships with voters, their communities, and their families.

To start, many of our interviewees discussed the value of engaging with and being responsive to people in their community. And for many of the women who intend to run again, focusing on the unmet policy needs of their constituents is a key component of that commitment. They mentioned various issues like the opioid crisis, economic inequality, education, immigration, and health care. As one woman shared,

I’ve seen policies take place over the last few years that are killing this district and harming the people that live here . . . I don’t know that there’s anything I can do to stop it, but by God I have to give it a shot. (S1)

In addition, several of these women expressed a sense of obligation to continue the work they started. Nearly one-third of the women who are intending to run again voiced such a sense of commitment (30%), while none of the less certain women did so. Beyond feeling a responsibility to get off of the sidelines to fight for things they valued, this sentiment was often couched in terms of being responsive to various people in their lives, of not letting down those who believed in them and were counting on them. As one said,

I think there’s a heavy responsibility that comes on someone when they’ve had this kind of money invested in raising their name ID . . . I went from 1% name ID to almost winning a congressional race . . . We’re looking at $1.2 million invested [by others] in raising my profile. So I think that’s a heavy responsibility that the party just invested in me. (B2)

And others voiced similar responsibility to their campaign volunteers, staff, and voters. For example,

We have mobilized people who had never volunteered, never contributed, never been involved in any sort of grassroots organizing or political campaign. And so I feel obligated to make good use of that . . . they’re just encouraged. They don’t want to stop. (M1)

When we turn to the women who are less certain about running again, fewer highlight the importance of continuing to work on policy needs and none of them mention a similar sense of responsibility to those who supported them. Some of these women highlight the power of others’ continued faith in them, but the tone of their comments is slightly different: rather than voicing a sense of responsibility or obligation to run, these women generally note that the support is gratifying rather than determinative.

Another area where the two groups diverge sharply is in their characterizations of the role and value of their campaign team. While both groups gave credit to their team when reflecting back upon the lessons they learned, the women who are planning to run again appeared to build deeper connections with their team, and many indicated that these supporters remained ready and willing to stay engaged in order to launch the next campaign. As two general election losers with future plans to run explained,

[M]y team wants to stay together. We enjoy working together and we’ve already established the network in the district. We need to grow it. But we feel like we’ve got such a good investment already made that we can build on it. We have two years to do that. (C1)

I may run in 2020 again for Congress because of the groundwork that I have. I have a great volunteer base that is really encouraging me to run again. (M4)

With a support network in place, the decision to run again appears to come much more easily.

For those less inclined to run again, the value of relationships also weighed heavily, but many of these women emphasized the negative impact their candidacy had upon their families and loved ones. At least three women talked about the ways in which the campaign was difficult on their families. One woman noted that she had barely seen her children over the past year, another mentioned that her husband was not crazy about the idea of another campaign, and another explained that her husband’s business was targeted by fellow partisans during the primary. As she said,

Your whole life comes when you see your family being destroyed, you know? It’s like, is it really worth it. (L1)

But for women who intend to run again, mentions of family were more likely to be a positive than negative force propelling their candidacy. One first-time candidate who narrowly lost her primary explained,

I’m a 41-year old woman who’s got two little boys and I’m very fearful of the direction that we’re going in. So I’m still going to be involved. I’m going to run again because now I feel, I’m obligated to. (J2)

In more ways than one, we find that women’s aspirations to run again subsequent to losing are shaped considerably by their political and personal relationships. For women intending to run again, their own words highlight just how greatly they care about the relationships they built throughout the process and how much that support and sense of responsibility shapes their decisions to run again: they want to continue inspiring and speaking for others, address their policy concerns, make good on personal obligations built along the way, and draw on their existing well of support by running again. Their political and family relationships are positive forces encouraging them to run.

For the less certain women, those relationships appear to be of a different nature. None of these women spoke about their obligation to others in making another run, they were far less likely to mention the value of their campaign team in reflecting upon their races and generally found the campaign process more difficult on their families and loved ones.

Data Limitations

This project gives voice to the women who ran and lost an election and examines how their stories fit within our broader understanding of women’s paths to elected office. With so few studies of electoral losers, and even less of women who lose, our project begins to fill an important gap. That said, we acknowledge data limitations and do not seek to make more generalized claims.

In particular, our data trends more Democratic and more white than the field of women candidates who ran in 2018. Although we do not believe partisanship theoretically influences the outcome of electoral loss, it is possible that the recruitment and training by the Democrats is different than the Republicans, and that those differences become meaningful when women consider a second bid for office. Similarly, given the literature that finds women of color often face more racism and sexism than their white counterparts (Carew 2016; Doherty et al. 2019), we have reason to believe race and ethnicity might play a role in future bids for office. As we detail in the conclusion, we hope to remedy these data limitations with additional data from the 2020 elections.

Looking Forward: A Theory of Gendered Electoral Persistence

Our analysis of these women who lost in 2018 lead to a number of suggestive conclusions about the factors that predict when women will run again. Yet together, they do suggest that women may react to loss differently than men. Given the data limitations discussed above, we acknowledge that this proposed theory of gendered electoral persistence requires additional data and analysis for further refinement, but also provides the foundation for future work on women and electoral loss.

First, our theory of gendered electoral persistence posits women will look to electoral as well as relational factors when considering whether or not to mount another run. Because we found mixed evidence when it comes to the role of electoral factors like money raised or votes garnered in the primary or general election in shaping future decisions to run, we are reluctant to dismiss the rational choice model at this point. All else equal, we expect women who performed well in their first election will be slightly more likely to run than candidates who were less successful.

We add to the traditional cost/benefit analysis an assessment of the good or harm women consider it would do to (1) themselves (psychological model) and (2) others (relationally embedded model). The theory of gendered electoral persistence expects women who found the initial loss to be harmful to themselves to be less likely to run again. Here, we see the perceptions of doubt reinforced by the first loss, and consistent with earlier research that finds women to be less likely to compete after a loss. In contrast, women who were energized and mobilized by the loss and were able to attribute the loss to external factors will be more likely to run again. For these women, the loss did not harm them, but provided them with improved tools for the next election.

In addition, our theory of gendered electoral persistence posits that women will consider relationships with their voters, community, campaign team, and families in their decision, as argued by Carroll and Sanbonmatsu (2013) in their relationally embedded model of candidate emergence. The benefits to their network will be weighed similarly to the benefits to themselves, as will the costs. A rational candidate will think about the decision to run as consequential to not just herself, but to the commitments she has made to others. We argue that these relationships are key in shaping women’s perception of the costs and benefits involved in running: if a support network is already in place, the perceived costs become more navigable and less weighty. All else equal, those with strong networks of relationships are easier able to bear the significant costs associated with campaigning. Without such a network in place, the anticipated costs may be prohibitive.

In sum, our theory of gendered electoral persistence argues women act rationally in determining whether to run again but emphasizes that these rational factors are more expansive than existing models suggest. Carroll and Sanbonmatsu (2013) uncover compelling evidence that women’s paths to office continue to differ from men’s, reminding us of the shortcomings of a one-size-fits-all model for candidate emergence or reemergence. They identify “persistent gender differences in pathways to office” (p. 37), giving us reason to believe gender similarly shapes the cost–benefit calculus in deciding to run after loss and suggesting our existing models focus too heavily on men’s experiences. We believe our interviews suggest as much. For women, doing well the first time around should matter, as should an assessment of how much of the work already invested can be used again. But in addition to this, women consider the psychological impact on themselves and act on what is best for them and what is best for their communities.

Conclusion

I’m so glad I did it. I lost some friends along the way. I picked up so many more . . . If I think about just the campaign process and going through it again, I’m like, “Oh God, no.” But when I think about the reasons why I ran in the first place, I don’t want to just sit on the sideline, I probably will [run again]. (Female congressional candidate who lost general election in 2018)

The 2020 election is shaping up to be another historic year, suggesting 2018 was not a blip. As of this writing, 590 women have announced their congressional candidacies, and of these, ninety-two are rebound candidates.⁸ In this project, we argue that the women who lost, and consider running again, have much to tell us about the larger questions of candidate emergence and pathways to political office. Very little is known about candidates who lose and run again, and even less about women who do so. We added to this scant literature by examining a unique data set of women who lost in 2018, who offered important context on how gender may matter in electoral decision making. While preliminary, our analysis offers some important insights on how women who rebound think about their loss, and their decision-making processes.

We thank the women who ran in 2018 for telling us their stories, and as the 2020 elections unfold, we look forward to more fully developing our new theory of gendered electoral persistence and adding to our understanding of rebound female candidates.

Supplemental Material

A_and_B – Supplemental material for She Persisted: Gender, Electoral Loss, and the Decision to Run Again

Supplemental material, A_and_B for She Persisted: Gender, Electoral Loss, and the Decision to Run Again by Julie Dolan and Paru Shah in Political Research Quarterly

Footnotes

Author Note

Anonymized data are available at .

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Paru Shah

Supplemental Material

Supplemental materials for this article are available with the manuscript on the Political Research Quarterly (PRQ) website.

Notes

References

Bledsoe

Timothy

Herring

Mary

. 1990. “Victims of Circumstances: Women in Pursuit of Political Office.” American Political Science Review 84 (1): 213–23.

Buser

Thomas

. 2016. “The Impact of Losing in a Competition on the Willingness to Seek Further Challenges.” Management Science 62 (12): 3439–49.

Carew

Jessica D. Johnson

. 2016. “How Do You See Me? Stereotyping of Black Women and How It Affects Them in an Electoral Context.” In Distinct Identities, 111–31.

Carroll

Susan J.

1985. Women as Candidates in American Politics. Bloomington: Indiana University Press.

Carroll

Susan J.

Sanbonmatsu

Kira

. 2013. More Women Can Run: Gender and Pathways to the State Legislatures, edited by Brown

Nadia

Gershon

Sarah Allen

. New York: Oxford University Press; Taylor and Francis Publishing.

Carsey

Thomas M.

Berry

William D.

Forrest

Walter

2003. “‘If At First You Don’t Succeed...’: Repeat Challengers in State Legislative Elections.” Paper presented at the Third Annual State Politics and Policy Conference, Tucson, AZ.

Costantini

Edmond

. 1990. “Political Women and Political Ambition: Closing the Gender Gap.” American Journal of Political Science 34 (3): 741–70.

Doherty

David

Dowling

Conor M.

Miller

Michael G.

2019. “Do Local Party Chairs Think Women and Minority Candidates Can Win? Evidence from a Conjoint Experiment.” The Journal of Politics 814: 1282–97.

Fox

Richard L.

Lawless

Jennifer L.

2004. “Entering the Arena? Gender and the Decision to Run for Office.” American Journal of Political Science 48 (2): 264–80.

10.

Fox

Richard L.

Lawless

Jennifer L.

2011. “Gendered Perceptions and Political Candidacies: A Central Barrier to Women’s Equality in Electoral Politics.” American Journal of Political Science 55 (1): 59–73.

11.

Fulton

Sarah A.

Maestas

Cherie D.

Maisel

L. Sandy

Stone

Walter J.

2006. “The Sense of a Woman: Gender, Ambition, and the Decision to Run for Congress.” Political Research Quarterly 59 (2): 235–48.

12.

Gill

David

Prowse

Victoria

. 2014. “Gender Differences and Dynamics in Competition: The Role of Luck.” Quantitative Economics 5 (2): 351–76.

13.

Jacobson

Gary C.

Kernell

Samuel

. 1983. Strategy and Choice in Congressional Elections. New Haven: Yale University Press.

14.

Palmer

Barbara

Simon

Dennis Michael

. 2012. Women and Congressional Elections: A Century of Change. Boulder: Lynne Rienner.

15.

Pearson

Kathryn

McGhee

Eric

. 2013. “What It Takes to Win: Questioning ‘Gender Neutral’ Outcomes in U.S. House Elections.” Politics & Gender 9 (4): 439–62.

16.

Schneider

Monica C.

Holman

Mirya R.

Diekman

Amanda B.

McAndrew

Thomas

. 2016. “Power, Conflict, and Community: How Gendered Views of Political Power Influence Women’s Political Ambition.” Political Psychology 37 (4): 515–31.

17.

Silva

Andrea

Skulley

Carrie

. 2019. “Always Running: Candidate Emergence among Women of Color over Time.” Political Research Quarterly 72:342–359.

18.

Stripp

Semilla B.

2019. “The Purple Wave: Gender and Electoral Outcomes in the 2018 Midterms.” Macalester College Political Science Honors Projects. https://digitalcommons.macalester.edu/poli_honors/79/.

19.

Taylor

Andrew J.

Boatright

Robert G.

2005. “The Personal and the Political in Repeat Congressional Candidacies.” Political Research Quarterly 58 (4): 599–607.

20.

Thomsen

Danielle M.

2018. “Gender Differences in Candidate Reemergence.” Boston. https://daniellethomsen.files.wordpress.com/2019/08/thomsen_apsa2018.pdf.

21.

Vallejo

Agustin

. 2019. “Nevertheless, She Persisted? Gender Differences in the Effect of Losing an Election on Political Ambition.” Paper presented at the Midwest Political Science Association Annual Meeting, Chicago, IL.

22.

Wasserman

Melanie

. 2018. Gender Differences in Politician Persistence. SSRN Scholarly Paper. Rochester: Social Science Research Network. https://papers.ssrn.com/abstract=3370587.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.14 MB