Antonio Ciccone argues that a 5% income shock (say by drought) raises the likelihood of civil conflict by 15 percentage points.
To see whether impoverishment causes the onset of civil conflict, I take a detailed look at data on rainfall levels in years before the outbreak of civil conflicts in Sub-Saharan African countries between 1980 and 2006. It is well known that living standards in these countries tend to be below trend in drought years and above trend when rainfall levels are above average. If civil conflict is triggered by sudden impoverishment, civil conflict onset in Sub-Saharan Africa should therefore have been more likely following drought years.
…If civil conflict onset is partly driven by sudden impoverishment, conflict outbreak in Sub-Saharan Africa should be more likely following below-average rainfall years. I find this to be the case. This result, combined with the effect of rainfall on income, allows me to estimate the effect of sudden impoverishment on the probability of civil conflict onset. My estimates indicate that a negative 5% income shock raises the likelihood of civil conflict by 15 percentage points.
More here
This is consistent with Fisman and Miguel’s argument that in Africa an income drop of 5% increases the risk of civil conflict in the following year to nearly 30%.
However, Simeon Djankov and Marta Reynal-Querol disagree by arguing that poverty does not have an effect on civil wars.
In Nepal, during the Maoists rebellion (1996-2006), the GDP growth rate averaged 4.1% which is barely different from the average growth rate of one decade before. However, head count poverty rate declined by more than 11 percentage points. This means that conflict did not substantially affect GDP growth rate. This can be explained by the rise in remittances inflow, even during peak time of conflict. This means declining poverty did not affect the likelihood of conflict in Nepal during 1996-2006. The effect of remittances far outweighed the negative effect of conflict on poverty and growth (though growth rate did plunged to –0.1% in 2001).
The figure below shows the increase in remittances:
