How Does Natural Resource Wealth Influence Civil War? Evidence from 13 Case Studies

Several large-N studies have found that a country’s natural resource dependence is correlated with either the likelihood it will suffer from civil war, or the duration of a civil war once it begins. What accounts for these natural resource-civil war correlations? The causal mechanisms behind the correlations are neither obvious, nor wellexplained by the large-N studies. It is important to identify the correct causal mechanisms, both to rule out endogeneity and spurious correlations, and because different causal mechanisms have different policy implications. This paper uses a novel “medium-N” approach to examine the causal mechanisms behind the natural resource-civil war correlation. It begins by explaining the rationale for the unusual research design. It then describes nine hypotheses about how resources may influence a conflict; specifies the observable implications of each; and reports which of these implications can be observed in a sample of 13 civil wars. The findings suggest that the natural resource-civil war correlation found by the large-N studies may be accounted for by a variety of different mechanisms, some of which influence the onset of conflict, some of which influence the duration of conflict, and some of which influence the intensity of conflict. One of the most widely cited causal mechanisms, suggested by Collier and Hoeffler, is not observed in any of the 13 wars.