In Conflict Resolution situations where two parties with opposed preferences need to make a number of decisions simultaneously, we propose a simple mechanism that endows agents with a certain number of votes that can be distributed freely across issues. This mechanism allows parties to trade off their voting power across issues and extract gains from differences in the intensities of their preferences. The appealing properties of such a mechanism may be negated by strategic interactions among individuals. We test its properties using controlled laboratory experiments. We observe that equilibrium play increases over time and truthful/honest play decreases over time. The subjects almost reach the welfare predicted by the theory even when their behaviour is far from equilibrium. The fact that deviations from equilibrium do not do much damage to its welfare properties is a further argument in favour of the use of this mechanism in the real world.
