Criminal violence is one of the most serious challenges in contemporary Latin America. While the drug economy -- which sparked much of the violence -- developed later in the 20th century, the institutions which shape contemporary vulnerability have deep historic roots. In this dissertation, I study three informal institutions: systematic corruption in the security sector, traditional governance institutions, and patriarchal norms, all of which have consequences for contemporary violence and vulnerability. In combining these essays, I aim to uncover some of the historical and social origins of contemporary variation in criminal violence and vulnerability. The first paper examines systematic corruption in the security sector. I ask: why did Mexico's strong and enduring civilian autocratic regime fail to reform a military riddled with corruption? I argue that the regime's reliance on the military for political control and repression created openings for the military to act corruptly when the center state was faced with political threats. I use an original data-set of military-landlord paramilitaries under the Cárdenas administration to show that where the regime faced greater political threat, military officials abused their power to profit from collusion with landed elites. Tracing these dynamics through to Mexico's dirty war, I find that the presence of these militaries in the 1970s is associated with higher levels of excess repression, suggesting enduring consequences of these collusive agreements for military professionalization. The second paper examines collaborative governance institutions -- created in Mexico's land reform -- as a source of variation in contemporary vulnerability to criminal violence. Each major land reform in Latin America was accompanied by the creation of collective institutions to administer redistributed land and govern beneficiary communities of land reform. However, little is known about the long term consequences of these administrative institutions. I advance a theory which argues that administrative institutions which enable the preservation of indigenous governance traditions can facilitate collective action capacity, which yields security dividends by empowering communities to respond strategically and collectively to criminal threat. I leverage insight from two months of in-depth, interview-based fieldwork in Michoacán, Mexico, combined with a difference-in-difference design to uncover the consequences of institutional variation in Mexico's land reform for vulnerability to criminal violence and criminal presence. In line with theoretical expectations, I find that land reform communities which preserve traditions of indigenous governance generate security in the context of Mexico's drug war. These findings have important implications for a vast literature which studies the relationships between violence and property rights, as well as for studies of rural security. The final paper studies the relationship between social inequality and criminal victimization, focusing on hierarchies created and upheld by patriarchal norms. I advance a theory of intersectional vulnerability to criminal violence, arguing that the same traditional structures which enable high collective action and social control of criminal violence can also lead to the preservation of stronger patriarchal norms. I suggest that these strong patriarchal norms lead to more criminal victimization of women relative to men. In patriarchal contexts, women's relative vulnerability is increased by community failure to apply social control to protect women from criminal violence, and exacerbated by women's lack of recourse due to their political exclusion. I test this theory empirically in the context of the Mexican drug war. I use an original measure of patriarchal norms drawn from household surveys on gender roles to identify empirical associations between traditional social structures and higher levels of patriarchal norms pre-drug war. Exploiting the shock of the onset of the drug war, I find that higher levels of patriarchy pre-drug war lead to substantially greater increases in women's victimization relative to men's following the onset of the war. I find strong evidence that this victimization is non-domestic, reflecting how community control of violence fails to protect women from criminal victimization, and that women are most at risk when they are politically excluded. These findings speak to how social and political inequalities shape vulnerability to criminal violence, particularly in contexts where the state fails to provide security. Together, these papers highlight the complex layering of informal institutions which shape contemporary welfare in the context of widespread criminal violence.

Villagisation and large-scale agricultural investments in Gambella region has been a major concern of human right groups. The Ethiopian government argues that Villagisation program is voluntary and part of Growth and Transformation Plan (GTP) which attempts to bring development to indigenous communities and nothing to do with large-scale agricultural investment. On the contrary, human right groups and local civil society organizations claim that the Ethiopian government is forcefully relocating indigenous people from their ancestral land under the disguise of development while the true motive of the government is to expand agricultural investment in the region at the expense of the livelihood of the local communities. This research is an attempt to investigate the controversial Villagisation and large-scale agricultural investment in Gambella regional state by looking into the link between large scale agricultural investment and Villagisation. The main focus of the research is to examine the impacts of agricultural investment and Villagisation in Gambella region the light of the Ethiopian government policy in the region and the alleged development induced human right violations. The research is based on a qualitative method to capture data from 32 Villagisation sites using in-depth interviews, focus group discussions and field observations. For the purpose of this study, 241 participants were selected from the study sites to participate in the research. Out of the 241 participants 75 of them were participated in in-depth interviews and the rest were included in focus group discussions and informal discussions based on the participants’ knowledge, views, experience and feelings associated with Villagisation and large-scale agricultural investment in the region. The findings of this study show no indication of involuntary Villagisation, no significant relationship between Villagisation and investment, or no evidence of previously occupied land being leased to investors. However, the study reveals that there has been serious lack of communication and misinformation from the government side in the process of planning and implementing the Villagisation program. Owing to this, suspicion and lack of trust between government officials and the local communities characterized implementation of the Villagisation project. / Geography / M.A. (Geography)