Subverting Process-Based Controls: Oscillation in Automotive Recalls and a Simulation on Opportunism Within a Network

Published In

Decision Sciences

Document Type


Publication Date



The increase in the frequency and impact of automotive recalls has had broad reaching economic and social consequences. Recent examples of automotive recalls suggest that some are due to deliberate actions of firms to subvert processes designed to ensure quality. Such controls are dependent on partners acting in good faith in the relationship and abiding by relational norms and so do not recognize the risks of firms lying, falsifying data, or intentionally circumventing process‐based controls. Data on automotive recalls shows that recalls exhibit a statistically significant oscillatory pattern with an estimated period of around 3.6 years. We argue that the cyclical pattern in automotive recalls is due, in part, to opportunistic behavior within a network, which is likely to spread and be reciprocated. To test whether opportunism can lead to similar network effects, we develop an agent‐based simulation to model opportunism within a network of connected firms. The simulation is based on an extension of the prisoner's dilemma where the cooperate/defect decision is based on a dyadic relational model driven by trust, knowledge, and dependence levels within each relationship. The results from the simulation suggest that cyclical patterns, similar to those in automotive recalls, emerge as well as “behavioral clustering” within the network, where connected firms exhibit highly similar behaviors that tend to cycle within small clusters. Therefore, opportunistic behavior in supply networks might be an important, relational determinant of product recalls. Our dynamic modeling approach differs from current perspectives on understanding product recalls, contributing to the current literature.


© 2020 Decision Sciences Institute



Persistent Identifier