War as a Redistributive Problem

The forthcoming article “War as a Redistributive Problem” by Jason Sanwalka Davis is summarized by the author below. 

Does the distribution of costs and benefits from war across domestic groups influence whether war is likely to occur? The idea that this might matter has certainly been a part of the popular narratives of conflict. Concerns about the influence of oil and defense contractors loomed over discussions of the genesis of the Iraq War. Similarly, during World War I, many countries were rife with internal domestic tensions over the distribution of the burdens of the war, with the wealthiest accused of profiting while middle-class soldiers suffered the costs of fighting. Going back further, Kant’s Perpetual Peace centered its theory of “republican” peace on the idea that leaders of republics – unlike autocracies – would be forced to care about the groups subject to war’s most significant harms. 

Despite this, the rationalist conflict literature has usually sought to explain war via features of interstate bargaining: generally, commitment problems and information asymmetries. Recently, however, international relations scholars have begun to explore the possibility that war might instead emerge from agency problems – i.e. situations where the incentives of decision makers responsible for entering a war differ from the incentives of the population more broadly. Importantly, this creates space for domestic distributive politics to have an impact on war onset, as it may not matter if the country as a whole benefits from a peaceful bargain if the key constituencies do not. 

Or does it? This paper develops a model that shows that the conditions under which domestic distributive politics can matter are more tightly circumscribed than previously understood. Indeed, their impact is conditional on the existence of redistributive frictions within a country. If war is to be the result of state capture by a pro-war constituency, it must be the case that the state-level agent lacks other cost-free means of “buying off” this group via redistribution from other parties; otherwise, even an agent who cares little for those who experience the burdens of war would be better served by using these costless tools to satisfy the pro-war group, while avoiding the destructive costs of conflict. It should not then matter if this group is 10 times or 1,000 times more influential than those who face the burdens of war, as the “inefficiency puzzle” at the heart of interstate conflict would simply have been relocated intrastate and left unresolved. 

However, when these redistributive frictions exist, both their magnitude and the structure of internal distributive politics become essential to understanding the origins of interstate conflict, in a way that requires they be considered jointly. Decisions about conflict should be biased towards both those who are most influential and those from whom value can be most easily extracted, but the effect of one is conditioned by the other; for instance, political influence matters less for determining war outcomes when costs to redistribute are low. Put differently, if war is to be thought of as a redistributive political tool, one needs to think carefully about the costs of the substitute tools available for redistribution. 

About the Author: Jason Sanwalka Davis is Postdoctoral Fellow, Browne Center for International Politics at University of Pennsylvania. His research “War as a Redistributive Problem” is now available in Early View and will appear in a forthcoming issue of the American Journal of Political Science.