Gravitating toward War: Preponderance May Pacify, but Power Kills

Countries have better abilities and stronger incentives to engage in militarized conflicts the larger and more powerful they are. The article applies Zipf’s notion of a ‘‘gravity model’’ to the risk of interstate conflict and argues that the empirical relationship between size and distance and conflict is stronger than any other identified in dyadic statistical studies of interstate conflict. Most empirical studies of interstate conflict fail to take size properly into account. The article shows that controlling for size variables improves the estimation of other variables of interest, and it explores the impact of omitting size variables for the investigation of the power preponderance versus power parity debate. The results indicate that even though a power capability ratio variable suggests asymmetric dyads are less conflict-prone, the risk-increasing effect of power itself means that a unilateral increase of power in one country increases the risk of conflict. Keywords: interstate conflict; gravity model; country size; power preponderance; power parity; model specification; omitted variable bias