Vote or Fight?

Abstract Why do some nations allocate power through voting while others do it through fighting? When political power is indivisible, voting is a substitute for fighting—provided the losing side accepts the outcome. We study a theoretical model of this substitution, and use a country-level panel dataset to assess empirically whether the economic factors influencing fighting also shape voting. They do. We contribute several theoretical and empirical innovations. First, we apply a recently developed method for analysing conflict resolution functions to derive robust theoretical results. Second, we introduce a new explanatory variable—productive efficiency as measured by income relative to the global frontier—and we explain the theoretical and empirical relevance of this variable. Finally, we show that absolute income levels do not matter, whereas oil wealth and ethnic divisions do—though their influence is less important than that of productive efficiency in explaining patterns of fighting and voting. A key implication of our analysis is that reducing global inequality is crucial for decreasing conflict and fostering democracy.