Recent large-N studies of civil war conclude that inequality does not increase the risk of internal armed conflict. This paper argues that such conclusions may be premature because existing studies, which usually measure inequality as income inequality between individuals, neglect the group aspect of inequality and social organization. Case studies suggest that what matters for conflict is so-called ‘horizontal inequalities’ – systematic inequalities that coincide with ethnic cleavages – and that in addition to economics one should also consider other dimensions of inequality. The main objective of this paper is to explore whether horizontal inequalities affect the probability of civil conflict when tested quantitatively across several cases. Using data from the Demographic and Health Surveys (DHS) for 33 developing countries, I construct aggregated macro-indicators on inequality between the two largest ethnic groups in each country, along three dimensions: social, economic and health-related. The main finding is that social horizontal inequalities seem to be positively related to outbreak of civil conflict.