Dynamic and contemporaneous causality in a supply chain: An application of the US beef industry

Causal relationships are used to investigate information flows and directions of control in a decentralized multi-echelon supply chain where no central authority has system level control over optimizing decisions. We use secondary time-series data representing the US beef industry to investigate dynamic and contemporaneous causality based on outof- sample Granger causality and Direct Acyclic Graphs (DAGs). Results indicate: (i) the US beef supply chain experienced a significant structural change in late 1996 and early 1997 that may be attributed to a weather induced production shock and an apparent turnaround of the cattle cycle; (ii) contemporaneous causalities appear to be stronger and dynamic causalities appear to be weaker after the structural change, suggesting faster and more effective information transmission along the supply chain after the structural change; (iii) contemporaneous information appears to flow from upstream to downstream tiers in the supply chain before the structural break, which reverses after the structural break, suggesting a shift in control from upstream to downstream firms; and (iv) co-use of spot market and contracts to procure strategic inputs by processors appears to allow processors some control of spot price through contract purchases in the post-break period. Our approach could be readily used to investigate other multi-echelon systems.