International Security Policies in Latin America: a new framework for analysis

• Autor(a) principal: Thales Leonardo de Carvalho
• Data de Publicação: 2022
• Tipo de documento: Tese
• Idioma: eng
• Título da fonte: Repositório Institucional da UFMG
• Texto Completo: http://hdl.handle.net/1843/50815
• topic: International Security; Latin America; Foreign Policy; National Defense Policy; Multimethod Research; Ciência política - Teses; Segurança internacional - Teses

The Latin American security environment is not as peaceful as frequently seen. Transnational problems, such as organized crime, join interstate issues such as border disputes and political disagreements in a region with one of the highest numbers of militarized interstate disputes in the world. Scholars often have good information about these dynamics because of several descriptive works about them. However, we still have no theory to explain international security policies (ISP) in the region, although such a theory could help us in more than understanding the past and preparing for the future. Producing such a theory is thus the objective of this dissertation by answering: what explains the varying ISP adopted by Latin American states? I build such a theory in seven steps. First, I rely on the existing literature to propose a definition for ISP, as well as to operationalize the concept for the analysis, showing different axes through which it can be assessed. Second, I provide a typology for ISP in Latin America based on five policy types: Anti-Hegemonic, Balance of Power, Coexistence, Pro-Democracy Security Cooperation, and Transnational Threats. Third, I introduce my theory to explain these policies based on three variables: great power influence, the role of leaders, and domestic institutions. Fourth, I measure different ISP axes by using statistical techniques, such as a Principal Component Analysis and a Structural Topic Model. Fifth, I conduct statistical tests using panel data analysis to investigate the role of my theory in explaining each of the ISP axes. Sixth, I use cluster analysis to classify Latin American ISP, considering the proposed typology and test how my theory fits in explaining and predicting ISP types. Seventh, I combine a within-case analysis with the synthetic control method to conduct three case studies in order to corroborate my causal claims and show causal mechanisms. Finally, I introduce some concluding remarks.