Collier, Paul and Anke Hoeffler. "Greed and Grievance in Civil War". Oxford Economic Papers, Vol.56, No.4 (2004): 563-595.

Collier, Paul and Anke Hoeffler. "Greed and Grievance in Civil War". Oxford Economic Papers, Vol.56, No.4 (2004): 563-595.

• This paper was written to update a previous 1998 paper by the duo, which claimed that wars with largest expected payoffs post-war would last longer, as that could be used to cover war costs. The authors recognize now that most rebel groups cover their costs during wartime, and do not anticipate a post-war windfall (563-564).
• The authors note that two theories exist explaining the general existence of violent civil war: a political science theory focusing on grievances justifying violent resistance, and a economics theory which views rebels as rational actors engaging in a 'war industry' through looting and extortion (564).
• The authors measure a number of factors as proxies for opportunistic, greed-based, rebellions. These rebellions should be associated with high levels of wartime resource exploitation, high remittance levels from diasporas during wartime, funding from foreign powers, mountainous terrain, dispersed populations, and low levels of other economic opportunities for rebels (565-570).
• The authors measure four factors as proxies for grievance-based rebellions: high levels of ethnic and religious tension, political repression as measured by the lack of civil liberties, economic inequality, and presence of ethnic exclusion from government (570-572).
• Log regression analysis shows that civil wars are more likely to occur in countries with large populations, those that are significantly dependent on primary commodities for export, and with large diasporas to finance the conflict. Civil wars are less likely when economic growth is high and many are enrolled in secondary education, as more young men have opportunities beyond combat (573-574).
• The measure of the diaspora shows that larger diasporas are correlated to longer and recurrent conflicts, but this may not be a causal relationship. It could also be that longer conflict result in more refugees and thus larger diasporas, questioning the effect of the diaspora on the length of wars (575).
• Regression analysis shows that many of the theorized factors for grievance do not actually relate to conflict, as neither levels of ethnic and religious tension nor ethnic exclusion are associated with increased likelihood for conflict. Inequality in land ownership and economic inequality are also insignificant in predicting the outbreak of civil war. The only grievance-based factor associated with civil war is democracy, as autocracies are more likely to face rebellions (576-577).
• The authors find that the model based on economic greed as an explanatory factor is generally better at predicting the outbreak of civil war, although factors for economic inequality are found to be significant together with this greed-based model (577).
• This combined model finds that the most significant factors were the amount of commodities exports as a percentage of the economy, the number of economic and educational opportunities available outside of war, the concentration of the population, and the presence of overwhelming ethnic dominance (580-581).
• The model had some difficulties accounting for revolutions in Romania in 1989, and Iran during the Islamic Revolution. By removing these atypical incidences from the data set, the model now works much better (582).
• "Opportunity as an explanation of conflict risk is consistent with the economic interpretation of rebellion as greed-motivated. However, it is also consistent with grievance motivation as long as perceived grievances are sufficiently widespread to be common across societies and time. Opportunity can account for the existence of either for-profit, or not-for-profit, rebel organizations" (589).