Contracts and sensitivity analysis in a retailer dominant supply chain system under symmetric and asymmetric information


Student thesis: Doctoral Thesis

View graph of relations


  • Chang SU

Related Research Unit(s)


Awarding Institution
  • Gang HAO (Supervisor)
  • Hon Shiang LAU (Supervisor)
Award date3 Oct 2012


This thesis investigates different scenarios of a two-echelon supply chain in which a manufacturer wholesales a product to a Stackelberg dominant retailer who delivers the product to end customers. In a retailer dominant situation, the retailer can have centralized control of the chain and take the lead to draw the contract terms such as ordering price and policies. But this may lead to some unexpected dissensions. For example, the dominated manufacturer may not be inclined to participate in the deal if its minimal profit is not guaranteed. Such dissensions could be more serious when the retailer does not have accurate or full information of the manufacturer's production cost. In reality, we have seen that the growing dominance of many large retailers has altered the traditional channel behavior and mechanism. In literature, however, few studies have devoted to this area. This study aims to contribute to providing a better understanding of the contracting problem in a retailer dominant supply chain by investigating how a dominant retailer should design a purchase contract under different scenarios. This thesis examines, from a dominant retailer's perspective, the above-mentioned contract dispute problem under both symmetric and asymmetric information environments, taking into consideration the manufacturer's nonparticipation. We propose several effective contract schemes to maximize the retailer's profit while guaranteeing the manufacturer the least profit it requires. Firstly, in a price-sensitive newsvendor model with symmetric information, a linear price-discount sharing (PDS) contract and a novel integrative return policy is developed. Second order and unsold products' return is permitted in our assumption. Instead of assigning the right to all pricing decisions to the dominant retailer, wholesale price is negotiated between the two players, i.e. the retailer and the manufacturer, following the rule that one with more marketing power has larger bargaining power. The results show that it is better for a dominant retailer to leave some pricing discretion to the dominated manufacturer as well. Secondly, we explore the issues in a similar model with asymmetric information, where the manufacturer has private knowledge of its production cost, while the retailer can only estimate manufacturing cost. By combining regular markup contract, revenue sharing contract and slotting fee contract with volume discount contracts, several workable contract schemes and some extensions are developed, based on the well-known menu of contracts (MC). We prove that the MC is the optimal form of volume discount contracts but it may not be practical in real implementation. The numerical results show that all these contract schemes perform well in both traditional and consignment situations. Thirdly, to find out how the magnitude of asymmetric of information affects decisions and profits of supply chain players, we analyze sensitivity of profit of asymmetry of information (measured by the variance). Going by intuition, the informed player should always wish the other player to be as ignorant as possible and the uninformed player should always try to obtain more accurate forecasting of the asymmetric information. To some extent, the result turns out to be counter-intuitive in that the uninformed retailer's expected profit is a convex function of the accuracy level of asymmetry of information, and the informed manufacturer's real profit function is concave when he accepts the retailer's contract. In other words, if the dominant retailer is able to control or choose the accuracy level of forecasting information, its best choice can be at two extremes, i.e. either very high or very low. The manufacturer would not be willing to help the retailer improve forecasting accuracy unless forecasting by the retailer is poor (very low accuracy). This surprising result could be explained by the theory of information rent. We also compare our results with Corbett et al. (2004), who studied a similar problem in an upstream player dominant channel but achieved different results. The study shows that our investigation is robust in both retailer dominant and manufacturer dominant scenarios.

    Research areas

  • Business logistics, Sensitivity theory (Mathematics), Management, Mathematical models