Is Security a Matter of State or Government? A study about how international and domestic variables produced different International Security Policies in Latin America

Abstract

The Latin American (LA) security environment comprises several factors, from borders and territorial claims to natural disasters, from transnational crimes to ideological divergences among heads of state. Latin American countries have been addressing these issues through what I call International Security Policies (ISP), based on the conjugation of foreign and defense policies towards answering international security issues. While some scholars have been attempting to analyze these policies, it is possible to say that few efforts have been made to produce comparative, empirical, and generalizable analyses about these policies. Therefore, in this article, I intend to contribute to understanding the Latin American security environment and the International Security Policies implemented in the region. Using a quantitative large-n research design, I investigate all the Latin American States, from 1980 to 2012, to answer the following question - what explains the varying international security policies adopted by these countries during this period? Based on the Neoclassical Realism, I hypothesize that the interaction among systemic factors, such as global polarity and interstate rivalry (the end of the Cold War, for example), and domestic factors, such as the president’s ideologies and domestic institutions (e.g., the re-democratization wave in the 1980s, and the Left Turn in the 2000s) can account for the implementation of different ISP among time and cases. Contrary to conventional wisdom, identifying the effect of domestic variables on these policies leads us to that ISP is a matter of government, varying according to who is in charge of it. Using Factor Analysis, I build indexes based on several indicators to assess these policies along two continuums - (1) aggressive-deterrent and (2) comprehensive-traditional. Then, using Ordinary Least Squares (OLS) regressions applied to longitudinal data, I investigate the effects of the mentioned variables on Latin American countries’ ISPs, using these indexes as proxies for these policies.