Unraveling Bias in International Mediation: Using Survey Experiments

Abstract

The article aims to address the question of whether neutral or biased mediators are more effective in resolving international conflicts and to help overcome the neutral- bias debate over mediators. A large number of previous research have focused on the bargaining problems (the problem of information uncertainty and the commitment problem) and have argued that biased mediators are effective. However, why does the neutral-bias debate still persist?

First, previous research has lumped together a variety of biases in their analyses. Second, previous research using observational data such as quantitative analysis and case studies, has endogeneity problems including confounding factors and selection bias. This article, therefore, focuses on biases in military and economic relationships, which are considered theoretically important. As an analytical method, this research uses a survey experiment, which is relatively new in international mediation research but is considered superior in identifying causal relationships to other methods.

The experimental results show that mediators with economic-relational bias are more effective than neutral mediators. On the other hand, mediators with military relational bias are more effective than neutral mediators under the problem of information uncertainty. The two main contributions of this article are as follows. First, it presents the importance of distinguishing the nuances of bias with respect to the specific roles of the mediator to overcome the neutral-bias debate. Second, this research shows that experimental methods are effective in addressing endogeneity problems.