Political Science ETDs

Publication Date



This dissertation examines the issue of presidential failure — any instance in which a president fails to complete his/her term in office without a break in the democratic regime. South America stands as an anomaly for having faced an uncommonly high rate of presidential failure, as eleven elected presidents have failed to complete their terms in office since the third wave of democratization. This phenomenon presents an interesting puzzle for scholars because it allows for inquiries into governmental stability as well as executive accountability. I evaluate the causes of presidential failure in South America through a multi-method approach that looks at the phenomenon from three different levels of analysis. First, I examine the cross-national trends that explain why presidents are removed from office in South America. Various scholars have analyzed the reasons that presidents fail in Latin America. This assessment builds on those past arguments in order to perform a comprehensive analysis of South American presidential failure. I focus on variables that have not consistently been utilized in the past. More than that, this analysis uses a new technique, survival modeling, to identify those factors that increase or decrease the likelihood that a president will complete his or her term in office. Through this analysis, I identify the importance of minority legislative support, inflation, prolonged recession, executive wrongdoing, and protest in increasing the likelihood that a president will fall. Second, I perform two national assessments of repeated presidential failure. The cross-national statistical analysis shows that certain factors put presidents at risk. The qualitative analyses of Argentina and Ecuador, two countries with repeated failure, allow for process-tracing in order to identify how presidents are removed. I draw from the importance of protest and legislative opposition, which are found significant in the survival model, to explain presidential failure in these two countries. I show that political actors with poder de convocatoria (power to convoke/rallying power) can use that power to challenge a president when he/she faces other performance-related issues, like scandal or economic problems. The study of Argentina highlights how the Peronist Party maintains this power through a variety of connections to its organized base. This relationship to base support changes over time and is visible in each instance of presidential failure. On the other hand, Ecuador demonstrates the importance of the indigenous movement in explaining mobilization against presidents. The Argentine story shows how presidential failure flows from an actor with power, a top-down process. Ecuadorian presidential failure shows the bottom-up path of failure that occurs when actors who have lacked traditional access to politics oppose presidents. Third, I analyze survey responses from Argentine and Ecuadorian citizens in order to identify why individuals in these countries choose to protest. I assess what demographic, organizational, and attitudinal factors influence the likelihood of failure. As protest is instrumental to the process of failure, this final assessment demonstrates the importance of civil society organizations and unions in pushing individuals into the street, supporting the findings of the previous two sections. Thus, the causal mechanisms of failure can be witnessed at cross-national, national, and individual levels of analysis. In conclusion, I discuss the important implications of this research for Latin American politics. I provide predictions for the future of presidential stability in the region, and I assess how recent protests differ from those of the past.

Degree Name

Political Science

Level of Degree


Department Name

Political Science

First Committee Member (Chair)

Goldfrank, Benjamin

Second Committee Member

Hansen, Wendy




Presidential Failure, Democracy, South America, Argentina, Ecuador, Social Movements, Political Parties

Document Type