As an integrated part of a supply contract, trade credit has intrinsic connections with supply chain coordination and inventory management. Using a model that explicitly captures the interaction of firms’ operations decisions, financial constraints, and multiple financing channels (bank loans and trade credit), this paper attempts to better understand the risk-sharing role of trade credit—that is, how trade credit enhances supply chain efficiency by allowing the retailer to partially share the demand risk with the supplier. Within this role, in equilibrium, trade credit is an indispensable external source for inventory financing, even when the supplier is at a disadvantageous position in managing default relative to a bank. Specifically, the equilibrium trade credit contract is net terms when the retailer’s financial status is relatively strong. Accordingly, trade credit is the only external source that the retailer uses to finance inventory. By contrast, if the retailer’s cash level is low, the supplier offers two-part terms, inducing the retailer to finance inventory with a portfolio of trade credit and bank loans. Further, a deeper early-payment discount is offered when the supplier is relatively less efficient in recovering defaulted trade credit, or the retailer has stronger market power. Trade credit allows the supplier to take advantage of the retailer’s financial weakness, yet it may also benefit both parties when the retailer’s cash is reasonably high. Finally, using a sample of firm-level data on retailers, we empirically observe the inventory financing pattern that is consistent with what our model predicts. This paper was accepted by Vishal Gaur, operations management.

Multistage stochastic linear programs model problems in financial planning, dynamic traffic assignment, economic policy analysis, and many other applications. Equivalent representations of such problems as deterministic linear programs are, however, excessively large. This paper develops decomposition and partitioning methods for solving these problems and reports on computational results on a set of practical test problems.