Theory of civil war under asymmetric information

In this paper, we consider a model that suggests that the theory of exchange with asymmetric information seems suitable to provide a possible explanation model of occurrence and duration of civil wars. We show that although civil conflicts are not Pareto optimal ex post they may be Pareto optimal ex ante, in the way that every alternative leaves either the government or the rebellion worse off in some contingency which cannot be ruled out on the basis of the information which is common to both sides. Therefore, the critical determinants of the occurrence of civil conflicts appear to be arising as consequences of asymmetries in the amount of information about some relevant variables available to bargaining parties. Indeed, the civil war commitment is used as device of division of rents accruing to rebellion and the ruling government elite.