This note reports a replication study of Falk and Kosfeld's (2006) medium control treatment. In the experimental game, an agent has an endowment of 120 experimental currency units and decides how much to transfer to a principal. For every unit that the agent gives up, the principal receives two...

We report three repetitions of Falk and Kosfeld's (2006) C5 and C10 treatments whose results largely conflict with those of the original study. We mainly observe hidden costs of control of low magnitude which lead to low-trust principal-agent relationships. We also report an extension where...