Working Paper: CEPR ID: DP7101
Authors: Timothy J. Besley; Torsten Persson
Abstract: This paper studies the incidence of civil war over time. We put forward a canonical model of civil war, which relates the incidence of conflict to circumstances, institutions and features of the underlying economy and polity. We use this model to derive testable predictions and to interpret the cross-sectional and times-series variations in civil conflict. Our most novel empirical finding is that higher world market prices of exported, as well as imported, commodities are strong and significant predictors of higher within-country incidence of civil war.
Keywords: commodity prices; conflict; natural resources; political institutions
JEL Codes: D74; F52; O11; Q34
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Higher world market prices for exported commodities (F69) | Higher incidence of civil war (N41) |
| Higher world market prices for imported commodities (F69) | Higher incidence of civil war (N41) |
| Higher prices of exported commodities (F14) | Higher natural resource rents (Q33) |
| Higher natural resource rents (Q33) | Higher likelihood of conflict (D74) |
| Higher prices of imported commodities (F69) | Lower wages (J31) |
| Lower wages (J31) | Lower opportunity cost of fighting (H56) |
| Lower opportunity cost of fighting (H56) | Higher probability of civil war (D74) |
| Higher world market prices for commodities (Q02) | Variations in resource rents (Q31) |