Working Paper: CEPR ID: DP10617
Authors: Paul Castaeda Dower; Victor Ginsburgh; Shlomo Weber
Abstract: We introduce two societal polarization measures that, unlike standard approaches based on time invariant and arbitrary divisions of a society into ethnolinguistic or religious groups, take into account how a society's history can alter inter-group relations. One of our measures allows for different inter-group divisions due to different experiences in the colonial era, while the other allows these divisions to change as a result of violence throughout the conflict episode. By examining the protracted war in Sri Lanka and applying these indices to a data set describing victims of terrorist attacks by district and year, we find that, for each of our polarization indices, there is a positive effect on the number of victims from terrorist attacks. The historical underpinnings of our indices allow us to demonstrate in a quantitative and concrete way the relevance of the historical path for understanding patterns of civil conflict.
Keywords: colonial legacy; conflict; linguistic disenfranchisement; polarization
JEL Codes: D74; F54; O15
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| colonial legacy (F54) | societal polarization (Z13) |
| societal polarization (Z13) | conflict intensity (D74) |
| colonial legacy (F54) | disenfranchisement of Tamil population (J15) |
| disenfranchisement of Tamil population (J15) | increased tensions (F52) |
| increased tensions (F52) | higher number of victims in terrorist attacks (H84) |
| polarization (C46) | number of victims from terrorist attacks (H84) |