Natural resources, rent seeking and civil war

The availability of large amounts of natural resources is sometimes a curse, sometimes a boon. There are plenty of countries where the discovery or exploitation of, say, oil or diamonds, leads to corruption or civil war, that is counterproductive attempts at grabbing the rents. But there are others where not such conflict occurs, rather various factions are appeased with the distribution of these rents, which may be a second best but is still much better than corruption and conflict. What triggers in which equilibrium a country falls in?

Kjetil Bjorvatn and Alireza Naghavi build a model where the government cannot credibly commit to transfers. When there are few rents to distribute, there is no conflict. When there are a lot of resource rents, the government has a lot to lose from conflict and is credible when it promises to redistribute, for example through patronage employment, and continue doing so. But when there is an intermediate level of resources rents, the government's future policy is not credible, as redistribution leaves it with fewer rents than if it were to try to keep them violently.