Signaling Cost Uncertainty: Some Experimental Evidence

This dissertation examines the effects on decision making of providing a signal of cost uncertainty in managerial financial reporting. It describes an experimental study used to evaluate these effects, as well as the process of using the z-Tree software to program this experiment. The z-Tree software allowed for a very easy administration of the experiment. Results of the study suggest that providing decision makers with a cost uncertainty signal leads to reduced decision performance, as measured using several different methods. Such decision makers choose more frequently to forego sales opportunities, compared to decision makers not provided with the uncertainty signal. However, no evidence is found to suggest that the difference in decision performance is driven by loss-aversion behavior or risk-seeking behavior, based on different prospect conditions. The effects of the uncertainty signal do appear to dampen in later iterations of a repeated decision task, primarily as a result of decision makers with the signal adapting their behavior to be more like decision makers without the signal (rather than the other way around, as predicted).