Arms Races and Conflict: Experimental Evidence

Abstract

We study escalation and aggression in an experimental first-strike game in which two participants play multiple rounds of a money-earning task. In each round, both players can spend money to accumulate weapons. The player with more weapons can spend money to strike against the other player, which almost totally eliminates the victim's earnings potential and removes their capacity to strike. Weapons can serve as a means of deterrence. In four treatments, we find that deterrence is strengthened if weapon stocking cannot be observed, that a balance of power is effective in maintaining peace, and that mutually beneficial trade decreases the risk of confrontation, but not necessarily the likelihood of costly arms races.

Equilibria of Attacker-Defender Games

“…Vice versa, those with a higher endowment than their Defender would be more inclined to aggress and thereby amplify wealth and power differences between Attacker and Defender (viz. Conflict Spiral Theory;de Dreu, 1995;Deutsch, 1973;Kydd, 2000;Abbink et al, 2021;Bacharach and Lawler, 1981). On the other hand, Attackers with a lower endowment have relatively much to gain from winning the contest.…”