An Empirical Study of the Causes of Military Coups and the Consequences of Military Rule in the Third World: 1960-1985

This study analyzed the causes of military coups and the consequences of military rule in the Third World during the 1960-1985 period. Using a coup d" etat score, including both successful and unsuccessful coups, as a dependent variable and collecting data for 109 developing nations from the World Handbook of Political and Social Indicators, The New York Times Index, and public documents, sixteen hypotheses derived from the literature on the causes of military coups were tested by both simple and multiple regression models for the Third World as a whole, as well as for four regions (Sub-Saharan Africa, Asia, Latin America, and the Middle East and North Africa) and in two time periods (1960-1970 and 1971-1985). Similarly, three models of military rule (progressive, Huntington's, and revisionist models) were analyzed to assess the consequences of military rule. The results of the study concerning the causes of military coups suggest four conclusions. First, three independent variables (social mobilization, cultural homogeneity, and dominant ethnic groups in the society) have stabilizing consequences. Second, six independent variables (previous coup experience, social mobilization divided by political institutionalization, length of national independence, economic deterioration, internal war, and military dominance) have destabilizing consequences. Third, multiple regression models for each region are very useful; most models explain more than 50% of the variance in military coups. Fourth, the time period covered is an important factor affecting explanations of the causes of military coups. In the analysis of the consequences of military rule, this study found that military governments did not differ significally from civilian governments in terms of economic, education, health, and social performances. However, the study found that military rule decreased political and civil rights. Its findings are thus very consistent with the best of the literature