An Experimental Test of Overconfidence as a Cause of Over- and Underinvestment in Conflict:

Sammanfattning: Overconfidence has been proposed as a cause of conflicts. However, there is little empirical support for this claim. We test overconfidence as a cause of over- and underinvestment in conflicts in a computerized experiment. First, we derive a model where overconfidence explains over- and underinvestment in conflict. We then test the predictions of this model in a laboratory experiment, both on a group level and with individual level regressions. Even though true overconfidence was successfully induced in the experiment, we do not find support for overconfidence as a cause of overinvestment in conflict, but weak support for overconfidence as a cause of underinvestment.