A simulation analysis of interactions among intended biases in costing systems and their effects on the accuracy of decision‐influencing information

On the one hand, management accounting systems increase in complexity, and the frequency and intensity of use of the provided information for decision‐making are rising. On the other hand, there is evidence that the provided information is inaccurate, which in most cases incurs economic costs. This paper analyzes the impact of biases in raw accounting data on the accuracy of the provided decision‐influencing information. The simulation study presents results concerning single and multiple biases which are intendedly entered into traditional costing systems. The presented results give insights into interactions among biases and indicate that multiple input biases do not necessarily affect the information quality negatively. Surprisingly, in some setups interactions among biases lead to a mitigation or even a compensation among themselves. Furthermore, the findings can constitute the basis for generating efficient organizational data quality policies, i.e., results indicate where (not) to tolerate biases and how to prioritize actions regarding information quality with respect to accuracy and cost of accuracy.