Are civil wars really caused by political repression, inequality, or religious and ethnic differences? What roles do factors other than grievance play in rebellion? This paper for the World Bank looks at the causes of civil war, using a new data set of wars during 1960-99.
Civil wars are now more common than international conflict. Of 25 armed conflicts in 2000, 23 were internal. Rebellion needs both motivation and opportunity. Political science explains conflict in terms of motive. When grievances are sufficiently acute, here is violent protest. Such grievances include inequality, oppression, religious and ethnic tensions. A much smaller literature from economic theory models rebellion as an industry that generates profit from looting. Greed, not grievance, is the driving force, and opportunity is more important than motive. The authors develop an econometric model for predicting the incidence of civil wars, resting on the following factors:
- Availability of finance. Dependence on primary commodity exports creates a high risk of civil strife as rebels have opportunities for extortion. Diasporas provide funding for rebel groups, and increase the risk of conflict renewal.
- The cost of rebellion. Low income means there is less to lose by fighting. Male secondary education enrolment, per capita income and growth rates have statistically significant effects on conflict risk.
- Military advantage. Dispersed populations, and mountainous terrain increase the risk of conflict.
- The risk of conflict is proportional to population. Opportunities increase with population. Diversity also increases, but measurable aspects of this were not found to contribute to civil war.
- Ethnic tension. In the case of domination by one ethnic group, conflict risk increases. This is the only grievance-based cause. Without a dominant ethnic group, rebel cohesion becomes more costly and conflict risk goes down.
Considering the opportunities for rebellion is more useful than using political, grievance-based approaches. The economic model gives objectively verifiable indicators, and is robust to a range of tests for outliers, alternative variable definition, and variations in estimation method. Policy implications of this are implied, not stated:
- Analysis of funding of rebels is crucial. Donors also need to consider how to deal with and include diasporas to prevent conflict renewal.
- Programmes that target male secondary education will reduce conflict risk by increasing the costs of war.
- Macro, meso, and micro level economic programmes will also reduce conflict risk in this way. The reports findings may imply that wealth creation is more significant than poverty reduction.
- Where there is ethnic diversity, donor programmes should reduce or negate dominance by one group.