Abstract
In recent years, there has been increased focus in the scholarly research on the effect of natural resources on the duration of civil wars. Some scientists argue that countries dependent on natural resources, measured as the primary commodity export ratio to the gross domestic product, are more prone to the onset of civil war. Additionally, resource dependence is claimed to prolong the duration of civil war, as both rebels and the central government do not find incentives to settle for peace. The reason is that the cost of settling for peace is higher than the alternative as long as resource extraction can take place. Other scientists emphasize the need to classify natural resources by lootability and obstructability. This argument claims that unobstructable, lootable natural resources, referring to resources that are easily extracted by unskilled labourers and easy to smuggle, will most likely benefit the rebels and prolong the duration of civil wars than compared with unlootable resources.
This thesis applies the lootability and obstructability approach and the resource dependence approach in studying the effect of natural resources on civil war. Results from the quantitative analysis indicates that resource dependence shorten civil war duration: if primary commodity export increases with 1%, the risk of civil war termination increases with 2.1%. Moreover, there is indirect support for the hypothesis that lootable resources prolong civil war duration. Furthermore, unlootable resources increase the risk of termination with 3.0% when exports of this resource group increase with 1%. The results also show that the estimates for obstructability of natural resources are positive, and similar, but not statistically significant, which implies that the obstructability approach is not a suitable method. This is highlighted in the chapter about the Democratic Republic of the Congo, a country abundant with natural resources. Although natural resources can be said to have fuelled the civil war in the DRC, it is nevertheless not the cause of it. However, most of the results are insignificant, and the results are generally weak, indicating that more precise is needed in order to firmly establish a correlation between natural resource and civil war duration.