The Risk of Civil Conflicts as a Determinant of Political Institutions

This paper proposes a specific mechanism to explain differences in political institutions based on the asymmetric and uncertain costs of civil conflicts. Asymmetry implies that the net benefit of fighting an insurgency is not shared equally by members of the elite. But uncertainty implies that these benefits are more evenly distributed ex-ante. The members of the elite face a commitment problem: they would like to commit in advance to a strong response to insurgencies, but ex-post they have the incentives to block any response if the conflict mainly affects other members of the elite. One way of solving this is empowering the executive so he may react forcefully to conflicts, despite the opposition of some fraction of the elite. In the model this group has to decide on the constraints imposed on the executive. Fewer constraints lead to a higher risk of expropriation. But more constraints lead to a suboptimal response to conflicts. The main prediction is that, conditional on asymmetric and uncertain costs, the higher is the likelihood of a civil conflict in the future, the lower are the constraints imposed on the executive. The paper empirically validates this implication using two types of evidence. First, it uses a sample of former colonies that became independent after WWII and geographic variables to identify the exogenous component of the likelihood of civil conflicts at the moment of the independence. Second, the model is used to explain the political events in the Americas after independence. Countries less prone to internal conflicts were the ones that imposed more constraints on the executive during the second half of the nineteenth century.