Dynamic Supplier Contracts under Asymmetric Inventory Information

In this paper, we examine a supply chain in which a single supplier sells to a downstream newsvendor-type retailer. We make two assumptions that enrich this simple and well-understood model. First, we consider a multi-period model, in which the sequence of events is as follows. In a period, t, the supplier offers a contract to the retailer, and the retailer makes her purchasing decision in anticipation of the demand. The demand then unravels and the retailer carries over any excess inventory (we assume a lost sales model) to the next period. In period t 1, the supplier designs a new contract based on his belief of the retailer's inventory and the game is played dynamically. We assume that short-term contracts are used, i.e., the contracting is dynamic and is done at the beginning of each period. Second, we assume that the inventory position of the retailer before ordering is not observed by the supplier. This setting describes scenarios in which the downstream retailer does not share inventory/sales information with the supplier. For instance, it captures the phenomenon of retailers distorting past sales information to secure better contracting terms from their suppliers. In this setting, under certain assumptions, we characterize and evaluate the supplier's optimal contract. To do so, we cast our problem as an adverse selection model with dynamic contracting. We then analyze the performance of the optimal contract with respect to various useful benchmarks and quantify the value of optimal contracting and the value of inventory information to the entire system. Dynamic adverse selection models which are Markovian (that is, the action in a period affects the hidden state in the subsequent period) are recognized as being theoretically difficult and are thus relatively less understood. We believe that in our analysis we provide a framework for analyzing such models under short-term contracting and thus take an important first step towards understanding such models