Decision-making Procedures: A General Theory and Its Field Experimental Test

It is a persistent finding in psychology and experimental economics that people's behavior is not only shaped by outcomes but also by decision-making procedures. In this paper we develop a general framework capable of modelling these procedural concerns. Within the context of psychological games we define procedures as mechanisms that influence the probabilities of reaching different endnodes. We show that for such procedural games a sequential psychological equilibrium always exists. Applying this approach within a principal-agent context we show that the way less attractive jobs are allocated is crucial for the effort exerted by agents. This prediction is tested in a field experiment, where some subjects had to type in data, whereas others had to verify the data inserted by the typists. The controllers' wage was 50% higher than that of the typists. In one treatment the less attractive typists' jobs were allocated directly, whereas in the other treatment the allocation was done randomly. As predicted, random allocation led to higher effort levels of the typists than direct appointment.