The previous three chapters have focused on the group level and explored how political exclusion and economic inequalities may generate widespread grievances conducive to ethnonationalist mobilization and armed conflict with the state. In this chapter, we take a step back to the country level and examine how particular configurations of political and economic privileges among groups in a country give rise to differences in the risk of civil war.

Given our focus on disaggregation and the importance of a dyadic approach to civil war, the move back up to the country level might at first seem puzzling. However, comparing the causes of civil war across different levels of analysis is helpful for a number of reasons. Many existing analyses of civil war have been conducted at the country level, which makes it difficult to compare their findings directly with our group-level results. By aggregating the properties of groups and other actors up to the country level, we are able to explicitly explore the relationship between group-level and country-level findings and whether these may differ due to potential scaling effects. Moreover, the number and size of ethnic groups varies greatly between countries. Many existing country profile measures, such as ethno-linguistic fractionalization indices, depend on atheoretical population-weighting procedures that disregard the political status of ethnic groups, and where small groups by construction will carry little weight in the measure. However, as we will show, it is possible to construct more informative country-level measures, reflecting group-level characteristics in a more theoretically informed manner. As such, a country-level analysis allows us to compare the explanatory power of our new proposed measures of horizontal inequalities with existing proxies for societal grievances related to restrictions on political participation and uneven distributions of wealth among individuals.