Optimal design of sales contracts under information asymmetry

We consider a supply chain where one manufacturer sells products to local markets through dealers. Each dealer faces a geographically dispersed market with uncertain demand, and is able to observe demand signal and exert effort to enhance sales. Both signal and effort affect the sales outcomes probabilistically, and cannot be observed by the manufacturer. Furthermore, dealers’ signals are correlated across markets. We characterize sufficient conditions for the manufacturer to extract full surplus from dealers in this scenario. Surprisingly, when there are multiple dealers and signals are sufficiently informative, full surplus extraction can be implemented by a simple combination of linear contracts and lotteries. The linear contracts induce the dealers to exert the optimal effort, and the lotteries eliminate their incentives to misreport their signals. A linear contract plus lottery and nonlinear sales-contingent contracts can be used alternately to ensure full surplus extraction. Additionally, we show that using just linear contracts can leave large surplus for dealers. When the conditions for full surplus extraction fail to hold, we show that only a limited number of types receive information rent. We provide a two-stage procedure to compute the payment scheme. This scheme, as well, can be implemented using a linear payment scheme and a lottery.