Supply Chain Networks, Stock Return Dependence, and Extreme Losses

Summary

This study examines whether supplier and customer links help explain dependence between listed companies’ returns. It constructs a directed company network from Bloomberg relationship data, resolving cases with links in both directions by choosing the direction with the larger reported amount or, when amounts are absent, the more frequent direction. An expanded network also connects firms that share a third-party customer or supplier. Modularity optimization identifies communities in the largest connected component.

The reported analysis finds higher return correlations among directly or indirectly connected firms than among unrelated pairs, including comparisons within the same sector or region. Correlation varies by relationship type, with shared-supplier and shared-customer pairs reported as strongest. Extreme daily losses also cluster, and the timing analysis suggests that connected firms’ extreme declines often precede those of a focal firm. These findings motivate supply-chain data as an input to dependence and risk models, but the document does not provide the underlying figures or a trading backtest. The lead-lag pattern is presented as evidence of possible predictive value, not proof of a robust forecasting strategy.

Key ideas

• The study builds a directed supplier-customer network and expands it through shared third parties.
• Community detection on the largest connected component reveals groups that differ from common sector classifications.
• Returns of connected firms are reported as more correlated than returns of unrelated pairs.
• Shared-supplier and shared-customer relationships show the strongest average correlations among the listed link types.
• Extreme declines cluster within network communities, and connected firms’ declines often precede those of a focal firm.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.