This study addresses the question of what makes ethnic quota systems in parliament work to manage ethnopolitical violence. By a reconstruction of Arend Lijphart’s theory on consociationalism, two causal mechanisms are identified. The first mechanism levels the power balance of contending groups by permanent inclusion in parliament. The second mechanism reduces the number of conflict issues to be agreed on jointly, by decentralization of decision-making to the respective ethnic groups. According to the logic of consociationalism, ethnic quotas in parliament are expected to prevent violence by levelling the power balance in parliament.

The study includes an investigation of the ethnic quota systems of the world. Two cases which challenge Lijphart’s theory in two different ways, are selected for in-depth analysis. Contrary to the predictions of consociational theory, Cyprus as a typical consociational case has failed in conflict management, whereas New Zealand as the prime example of non-consociational cases has succeeded in promoting peace.

The essence of consociational theory is reconstructed in a two-player game which is applied to the cases of ethnic relations in Cyprus and New Zealand. The conclusion is that ethnic quotas can contribute to changes in the actor’s ranking order of preferences by upgrading the value of cooperation. Only under the condition that the actors appreciate the mutual benefits of such cooperation, can ethnic quotas contribute to viable peace.