While many political leaders garner support by providing public goods, still many others use violence and repression. A common explanation for this variation is institutional differences, with some scholars positing that democracy limits repression. However, many political decisions remain difficult to explain on this basis. Motivated by this, this paper investigates non-institutional sources of variation in leader strategies. Specifically, the authors focus on Colombian paramilitary commanders and explore how their social backgrounds—particularly childhood socialization in rural communities—affect their strategic behavior in providing public goods vs. using violence.
The authors utilize a unique dataset compiled from Colombian transitional justice documents, examining paramilitary groups that operated from 1997 to 2006. The analysis focuses on 72 paramilitary fronts, with detailed data on their commanders’ backgrounds, including whether they were raised in rural (“peasant”) environments. The researchers employ a spatial regression discontinuity: An experimental design that estimates causal effects using geographic or spatial boundaries, where units on either side of the boundary experience different treatments design to isolate the effects of commanders’ social backgrounds on their strategies, comparing neighboring fronts led by commanders of different social backgrounds. The methodology allows for a causal interpretation of the relationship between a commander’s rural upbringing and their likelihood to provide public goods or commit violence. They find the following:
- Commanders with peasant origins were 1.5 times more likely to provide public goods such as roads, schools, and healthcare facilities. This behavior is linked to reciprocal social preferences developed in rural communities, where social exchanges are common.
- Peasant commanders were also nearly twice as likely to commit massacres. The authors theorize that this is due to a culture of reciprocity, common among peasants, that leads commanders to offer people public goods in exchange for contributions and punish those who do not reciprocate with violence.
- Fronts led by peasant commanders tended to last longer, with an average of 1.5 additional years in power compared to non-peasant-led fronts. This longevity is attributed to the combination of public good provision and violence, which helped maintain control over territories.
The study highlights the significant role that non-institutional factors, such as childhood socialization and reciprocal preferences, play in shaping leader strategies. These findings suggest that the social backgrounds of leaders can lead to substantial variation in political strategies, including the provision of public goods and the use of violence. By focusing on the social preferences of leaders, this paper provides new insights into the factors that drive political decisions, extending beyond traditional institutional explanations.