Resource Dependence, Economic Performance, and Political Stability

Existing literature on resource curse provide economic explanations. The author argues that political foundations of resource curse should be studied as well. The article emphasizes that political decisions matter because they determine whether resource wealth will result in political stability or not. Dunning develops a dynamic game-theoric model. According to Dunning, scholars should consider certain variables as conditional in their models. 
In the article, Dunning seeks to account for the variance between political stabile and politically unstable states that depend on resource wealth. Why do some countries achieve diversification of their economies with the help of resource wealth, while some other countries fail to do so? He argues that three key factors shaped the incentive of political elites to encourage diversification: the volatility of resource revenue, the degree of societal opposition to incumbent elites, and the prior development of the nonresource private sectors. The article tests the model with three case studies: Indonesia, Democratic Republic of Congo (former Zaire), and Botswana. 
The model developed in the article is about a resource-rich political economy in which social conflict takes place between two groups: people with political power, which is political elite, and people without political power, which is nonelite. Under some conditions nonelite actors stage a coup or rebel against the elite. 
According to the model the game is played in the following way. The country realizes the revenue of the resource in the first period. Elites decide whether to invest in the public good and the surplus is spent by the elite. In the second period, if the public good is not funded in the previous period, elites remain in power; consume the revenue and the period ends. If the elites decide to invest in public good, two scenarios follow: a) if the price of the resource is high in the international market, revolt is unlikely because it is assumed that people will not rebel against the ruling class during the good times. b) if the price of the resource is low in the international market, non-elite group can prefer to revolt and assume the cost whether the revolt succeeds or fails.  
Finally, Dunning tests the model with three case studies: Botswana, Democratic Republic of Congo, and Indonesia.