This is the 15th in our series of posts by job market candidates.
Conflict has shaped societies for centuries, creating deep grievances that persist for generations and often perpetuate the cycles of violence. Despite these devastating consequences, Rohner (2017) shows that the number of active conflicts worldwide continues to rise. This suggests that we still lack a understanding of the causes and consequences of conflict the factors that can help de-escalate it.
In my job market paper, I examine one widely debated yet empirically understudied factor in conflict de-escalation: the gender of leaders during conflict. Drawing from theories in multiple disciplines, I test whether female leaders can play a unique role in mitigating the harmful effects of armed conflict.
I explore this question by looking at female leadership within the Colombian civil conflict. Colombia offers a unique setting for two key reasons. First, the conflict involved many groups, many of which have already laid down their arms, making it easier to see how female leaders operated on different sides. Second, the communist ideology of the guerrilla groups led to a higher-than-usual number of women in their ranks, which resulted in occasional—but meaningful—instances of female leadership.
Background
Colombia’s armed conflict dates to the 1950s and involved three main actors: left-wing guerrillas, right-wing paramilitaries, and government forces. The guerrillas, primarily peasant and communist organizations founded during the Cold War, aimed to overthrow what they saw as a government "captured by the elites" and replace it with one representing "the rule of the people." This ideological drive created an environment where women could relatively thrive, with some sources estimating that women made up as much as 35% of their ranks.
Although mayors in Colombia have limited authority over security forces, they often serve as the highest representatives of the state in their municipalities. This is especially true in remote areas where the government’s presence has historically been weak, allowing illegal armed groups to establish a lasting foothold. As a result, part of the "job description" for these mayors involves engaging with all actors in their communities, including illegal armed groups. This engagement makes them key figures in securing and maintaining local peace.
Key Findings
To estimate the causal effects of female leadership, I analyzed closely contested mayoral elections, comparing outcomes in municipalities where a female mayor narrowly won against those where she narrowly lost. Comparing outcomes in municipalities where women won (or lost) by a narrow margin allows me to identify the effect of female political leadership on violence under the assumption that all other characteristics (of municipalities and mayors) are similar on both groups. In my paper, I provide evidence that this is the case for the municipalities in the sample. Additionally, I constructed a unique dataset to track the approximate location of guerrilla units and the gender of their commanders. These data allow me to study violence levels in municipalities with and without guerrilla presence—and, more importantly, with and without female-led guerrilla units (See Figure 1)—as well as to observe changes within the same municipality over time. Together, these comparisons allow me to quantify the impact that female leadership had on the intensity of conflict violence in Colombia.
Figure 1. Intensity of guerrilla violence between 1996 and 2016
The core finding of my paper is that female leaders—whether municipal mayors or guerrilla commanders—play a crucial role in reducing violence in a municipality. Figure 2 illustrates this effect, showing that municipalities with a female mayor (depicted on the right side of the plot) experienced significantly fewer instances of guerrilla violence during the period studied. This decline in violence amounts to a 60% reduction in the average number of guerrilla attacks, translating to approximately 4 or 5 fewer armed attacks per 100,000 inhabitants. The effect is 2 to 3 times larger in magnitude than what Dube and Vargas (2013)—one of the most influential studies on the Colombian conflict—documented following the drop in coffee prices between 1997 and 2003. Notably, it is even more pronounced when both the mayor and the guerrilla commander are women, suggesting that female leadership on both sides creates a powerful synergy that further decreases violence.
Figure 2. The effect of female mayors on guerrilla violence – Regression discontinuity
Mechanisms
Before exploring how female leaders contribute to a reduction in conflict violence, it is essential to consider why women might opt for less violence in the first place. To explore this, I analyzed the campaign manifestos of female mayors using text-analysis techniques. The results show that female candidates use less aggressive language when discussing the conflict and illegal groups. On the commanders' side, I leverage high-frequency conflict data to show that guerrilla units with female commanders retaliate less against civilians following government crackdowns. These findings suggest that female mayors and commanders prefer peace over violence when given the choice.
To understand how female leaders influence conflict, it is essential to account for various relevant patterns in the data. These include the persistent presence of violence, no significant effect on extortion rates, and the guerrillas' reluctance to scale back activities in strategically important areas. Using a simple theoretical model, I show that these patterns, along with the reduction in violence, can be explained by a gender gap in negotiation skills. While this result is novel in the context of conflict, it aligns with earlier laboratory studies, such as those by Niederle and coauthors, showing that women tend to favor cooperation over conflict in competitive settings.
Additional evidence points in the same direction: female leadership fosters higher community building and engagement. In the sample, municipalities led by female mayors demonstrated increased social capital, which likely contributed to reducing violence. Finally, the guerrillas were unwilling to compromise in areas of high strategic importance—such as regions critical to smuggling routes—regardless of leadership. This suggests that while female leaders are effective in promoting peace, their influence may be limited in contexts where strong economic incentives to sustain violence persist
Conclusion
Female leadership, whether in government or guerrilla movements, plays a crucial role in reducing violence during conflict. In my job market paper, I provide empirical evidence supporting their effectiveness as agents of conflict de-escalation. However, these findings come with important caveats. Leaders, particularly women, do not emerge randomly; they often face significant barriers that may result in a distinct set of traits among those who rise to leadership positions. While my analysis directly addresses this selection issue, it remains possible that women who attain leadership roles possess unique, unobservable qualities that influence their choice to avoid violence. Furthermore, although this pattern is observed across two distinct forms of leadership, caution is needed when applying these findings to other conflict regions, as factors such as religion, culture, and resource availability may shape or constrain the impact of female leaders in armed conflict.
Francisco Eslava is a Job Market candidate at the University of British Columbia.
Join the Conversation