Disruptions can occur within manufacturing organizations resulting in negative economic consequences rippling upstream and downstream. Publicly traded firms often employ various account giving strategies as a means of offering stakeholders additional information about the causality associated with the disruption. To properly mitigate disruptions, organizations and stakeholders need to design policies based upon how disruptions along supply chains and account giving affect share prices over time. The purpose of this study is to us Artificial Neural Networks to model nonlinear stock recovery patterns after supply chain disruption events, examining how a firm’s position in the supply chain and account giving strategy influences time to recover and resilience.
This study uses a sample of 161 publicly announced supply chain disruptions to model stock price returns over time. Using Artificial Neural Networks (ANNs), this study utilizes daily observations which allow us to capture nonlinear patterns over the first quarter post-disruption. This study also codes the position in the supply chain and the specific account giving strategy that the firm utilized after the disruption. This offers insight into how companies, investors and local community groups can more accurately execute strategies to minimize losses and possibly turn the disruptions into gains.
This study finds that the ANNs methodology is more accurate than comparable approaches in predicting stock returns. Using time-series returns one quarter after the supply chain disruption, this study identifies the time-to-recover (TTR) for different account giving strategies and supply chain network partners. Based upon attribution literature, this study finds that it is advantageous for supply chain partners to proactively agree to either stay silent (no account) or to minimize responsibility immediately after the event when the causes for the disruption are not yet fully understood. As part of a resilience policy, companies should not be tempted to blame network partners, possibly rupturing long-term supply chain relationships, especially when the TTR is only 11 weeks. Such a resilience policy could explicitly articulate lessons for pandemic-related shocks or global disruptions due to trade-wars.
This study uses ANNs to model time series share price patterns after a supply chain disruption occurs. As opposed to previous studies, looking at only snapshots in time, this study captures non-linear patterns of returns over one quarter offering investors and policymakers new insights. Furthermore, this study investigates how supply chain structural factors such as where the disruption occurs in the supply chain and how the firm responds to the crisis impact stock price returns over time.
