Studying deception without deceiving participants: An experiment of deception experiments

Like avoiding labor protection laws via foreign subcontractors, banning deception in economic experiments does not exclude experiments with participants in the role of experimenters who, similar to properly incentivized subcontractors, can gain by deceiving those in the role of proper participants. We compare treatments with and without possible deception by 'experimenter-participants' in a dictator experiment and test whether participants in the role of experimenters engage in deception and whether deception affects the behavior of 'participant-participants.' We find that most participants in the role of experimenters engage in deception and that there is no difference in the behavior of participant-participants between treatments, even when repeating the experiment without deception after debriefing. Our results can be viewed as a contribution to studying the effects of unethical behavior via outsourcing it to subcontractors, by letting them do the harm.