MICROCON’s leading academic sextuplet, Maarten Voors, Eleonora Nillesen, Philip Verwimp, Erwin Bulte, Robert Lensink and Daan van Soest recently tried to figure out how conflict changes long-term behavior. Using experimental data from 35 randomly selected villages in Burundi, the team finds that besides systematic violence and destruction, war isn’t that bad in the long-term:
The main objective of this paper is to examine the causal effect of exposure to violence on behavior in economic experiments in which payoffs vary between choices across three dimensions: timing, riskiness, and social consequences. Do victims of conflict behave more pro-socially, do they have a higher propensity to invest in the future, and are they more prone to taking risks? We try to answer this important question by pulling together survey and new experimental data from Burundi.
Our results strongly suggest that exposure to violence affects behavior – possibly via altering preferences. We find that individuals who have either experienced violence themselves, or who live in communities that have been violently attacked, display more altruistic behavior, are more risk seeking, and act less patiently. Our results are robust across several specifications, and are obtained for both experimental and observational data. We believe they shed important new light on post-war recovery processes by speaking against overly pessimistic views on the destructive long-term consequences of civil war.
The paper does an excellent job of challenging assumptions about how violence alters behavior and is a great read for anybody working in post-conflict reconstruction and development.