Search
Program Calendar
Browse By Day
Search Tips
Virtual Exhibit Hall
Personal Schedule
Sign In
This research focuses on the control activity of Separation of Duties (SOD). The purpose of this study is to discern the cost-benefit tradeoffs for a firm’s SOD policy in terms of how many employees should be involved in any given task (i.e., degree of separation for a given task group). We develop mathematical models of theoretical economic expectations for the number of fraud incidents and the cost of fraud, with and without SOD, as well as the cost effectiveness of various degrees of separation for controlling errors and fraud. Counter-intuitively, we find that for a task group within the firm (1) SOD is likely to increase the expected number of incidents and cost of fraud. Furthermore, we find that (2) degrees of separation should be minimized for maximum cost effectiveness; and (3) SOD is always cost effective for controlling errors, but most degrees of separation are not cost effective for controlling fraud because of the collusion probability. These last two results are relatively insensitive to the probabilities of fraud or error detection, the dollar amount of fraud/errors, and to measurements of productivity when operational efficiency is affected by SOD.
Roberta Ann Barra, University of Hawaii-Hilo
Arline Savage, University of Alabama-Birmingham
Eric Im, University of Hawaii at Hilo