Employee reporting dishonesty is a significant area of concern for firms. In this study, we investigate how providing information about their prosocial actions, such as organizational citizenship behaviors, affects the extent of employee reporting dishonesty. We distinguish prosocial actions whose welfare effects are mutually beneficial (i.e., that help others and the employee), which are common in business practice, from those that are selfless in nature (i.e., that help others at a personal cost to the employee). In addition to examining the effect of the type of prosocial action on the extent of employee reporting dishonesty, we also examine the effect of construal (the manner in which individuals perceive and interpret the action). Using an experiment, we find that participants with high moral identity are less dishonest when they describe their selfless prosocial actions than when they describe their mutually beneficial prosocial actions, but only when they abstractly construe this information. However, we do not find evidence that the reporting dishonesty of participants with low moral identity is influenced by the type of prosocial action they provide information about or the construal of that information. We discuss implications of these results for theory and practice.