Triangulating horizontal inequality : Toward improved conflict analysis

Abstract

Does economic inequality cause civil war? Deviating from individualist measures of inequality such as the Gini coefficient, recent studies have found a statistical link between group-level inequalities and conflict onset. Yet, this connection remains controversial, not least because of the difficulties associated with conceptualizing and measuring group-level differences in development. In an effort to overcome weaknesses afflicting specific methods of measurement, we introduce a new composite indicator that exploits the strengths of three sources of data. The first step of our method combines geocoded data from the G-Econ project with night lights emissions data from satellites. In a second step, we bring together the combined spatial values with survey estimates in order to arrive at an improved measure of group-level inequality that is both more accurate and more robust than any one of the component measures. We evaluate the effect of the combined indicator and its components on the onset of civil violence. As expected, the combined index yields stronger results as more information becomes available, thus confirming the initial hypothesis that horizontal economic inequality does drive conflict in the case of groups that are relatively poor compared to the country average. Furthermore, these findings appear to be considerably more robust than those relying on a single data source.