Companies do not exist in isolation but are connected to their customers, suppliers, competitors, and joint ventures. In this novel piece of research, we study the FactSet Revere supply chain database and show how portfolio managers can utilize unstructured customer-supplier data to generate alpha.
Supply chain data is unstructured, incomplete, and highly complex. A single shock at one company may be transmitted to other connected firms. When a subject company raises its earnings guidance (or increases dividends or beats earnings expectations), its suppliers (and to a lesser extent, customers) also tend to benefit. Furthermore, the performance of a company’s customers and suppliers is predictive of its own stock returns and fundamentals. We also find that stock selection signals based on supply chain data contain significant alpha.
Inspired by algorithms in social networks and Internet searches such as Google PageRank, we analyze the supply chain web network as a whole to unlock a new differentiated alpha source. We follow goods as they move upstream and downstream through the full length of the supply chain. This allows us to identify key suppliers, customers and other companies of systemic importance to the fulfillment process. Our analysis shows that key companies within the supply chain network contain strong alpha on the long side, after controlling for other factors and risk measures.