Despite the existence in Latin America of a highly developed norm to collectively promote and defend democracy, there is variation in its application during democratic crises in the region. What explains the variation in responses to democratic crises? This dissertation makes three major arguments. The first is that the variation in responses is a result of the clarity and severity of the democratic crisis. Crises that present a clear threat to the constitutional underpinnings of democracy receive the strongest responses, while threats that are neither clearly undemocratic nor democratic receive the weakest. The second argument I make is that when there is variation to threats that are not clear, the factors of time, polarization within the organization, information, and uncertainty regarding the consequences of responding determine which cases receive responses and which do not. Finally, I demonstrate empirical evidence for the hypothesis that regional intergovernmental organizations in Latin America can be characterized as “clubs of presidents.” These findings have implications for our understanding of the various threats to the consolidation and quality of democracy. The data covers 31 cases of presidential crises from 1990-2012. The project uses cross-tabulations as well as in-depth case studies drawing from personal interviews with the former presidents, ambassadors of the Organization of American States, and ambassadors of member-states of IGOs directly involved in the crises.