Resource Depletion and Conflict: Experimental Evidence
Publisher: Journal of Economic Behavior & Organization
Author(s): Karolina Safarzynska and Marta Sylwestrzak
Date: 2021
Topics: Conflict Causes, Renewable Resources
Understanding how resource scarcity affects violent conflict over resources has been one of the central topics in economics and other social sciences for decades. Yet, we know surprisingly little about how resource scarcity affects individual behaviour that can spur inter-group conflict. To address this gap, we propose the common-pool experiment with renewable resources to study how incentives to engage in conflict change with resource depletion. Moreover, we examine how the possibility of conflict affects extraction. Each round, after harvesting, the players vote on whether to engage in conflict with the opponent group, in which they can annex a part of the opponent group's resource. The probability of winning varies between treatments. In the “50:50” treatment, to which we refer to as a baseline, the probability of winning is exogenous and equal for each party. In the “conflict-by-greed” treatment, the probability is proportional to the difference in resources between groups. Finally, in the “resource wars” treatment, the likelihood of winning depends on the number of tokens invested by each group in conflict. Experimental results show that “conflict-by-greed” promotes resource conservation. Subjects extract less resources compared to the baseline so as to increase chances of winning in war, which supports the parochial altruism hypothesis. Here, resource depletion is conducive to conflict, which distinguishes “conflict-by-greed” from other treatments. In addition, we find that “resource wars” are extremely destructive. Not only is conflict expenditure above the equilibrium, but also investing in conflict makes group members increase their extraction. As a result, the probability of resource exhaustion in the “resource wars” treatment is twice as high as in the baseline. We discuss how differences in uncertainty and the incentive structure between treatments explain these findings.