Or rather: Where previous research has either suggested a link or sought to explain it by an indirect effect through resource abundance tending to corrupt weak governments Lujala sets out to look closer at the influence of the more exact locations of the resources (something I believe for example Le Billon did already).
Lujala is confident enough to state in the introduction that "new data on localities of hydrocarbon fields throughout the world, shows that crude oil and natural gas directly affect rebel movements [...] easily extractable resources, such as gemstones, have an effect on rebel groups." The relationship is quantified - for example:
- "If resources are located inside the actual conflict zone, the duration of conflict is doubled." And...
- "oil and gas reserves have this effect on duration regardless of whether there has been production or not."
- Regarding the initiation of rebellion "onshore [not offshore] oil production increases the risk of conflict onset by 50%."
- And "secondary diamond production increases the risk of conflict onset by more than 40%."
One of the things I have been pondering is the simple distinguishing of states of societies in just two categories: in peace or in war. It is insufficient for intelligent discussion. Lujala is adding words to this argument: "natural resources, especially those that are easily exploited, provide motivation and means for rebel uprisings." In other words, rebellions may just be organized crime with political agendas attached. Of course, even more or less "legitimate" rebel groups could be driven to resource exploitation if their financial resources are exhausted in a prolonged conflict.
Hydrocarbons have traditionally been categorized as non-lootable as production usually requires substantial international industrial activities. But Lujala finds oil reserves even not in production does have effects and explains it by rebels potentially being "willing to engage in a long conflict [...] if the future price is large enough". While regarding reserves in production exceptions are seen in Nigeria where large scale looting occurs and in Columbia where rebels extort oil companies. However, note that regarding gas "production has no effect on conflict onset." Gas probably appears too hard to loot?
A look at the statistical analysis reveals how the data set was checked for influences other than profitable natural resources. The same tests are run for presence of mountainous regions, forest cover, language differences and difficult weather. It is worth noting that separatist conflicts (over territory rather than government power) and conflicts with one or more sides being democratic tends to last longer, while countries suffering from poverty and overpopulation are more prone to conflict onset.
Lujala, P. (2010). The spoils of nature: Armed civil conflict and rebel access to natural resources Journal of Peace Research, 47 (1), 15-28 DOI: 10.1177/0022343309350015
Also see Ross, 2004 which is mentioned by Lujala several times.