Abstract:

Online sellers often face holdup problems that occur when a platform abuses data advantages to compete against them in the product market, leaving them with lower incentives to invest in product innovation. This paper explores the role of relational contracts in solving such conflicts. In each period, a seller decides whether to remain on a platform and, if so, how much to invest in upgrading their product, which will depreciate without continuous investment from the seller. Then, the platform decides whether to copy the seller's product. If the platform chooses to copy, both parties engage in competition; otherwise, they share the monopoly profit. We show that the efficiency of a relational contract depends critically on the seller's outside option and the product depreciation factor. Specifically, the outside option has discontinuous and nonmonotonic effects on efficiency, suggesting that higher outside options, including higher offline market values, may not benefit online sellers. Furthermore, the product depreciation factor can exert opposite effects on efficiency, depending on whether the platform copies the seller off-path.

Abstract:

The recommendation system employed by content platforms serves as a pivotal avenue through which content reaches viewers, thereby exerting a substantial influence on creators’ behavior. Observations suggest that platforms take into account both viewer surplus and total revenue when making content recommendations. Our analysis focuses on the orientation of a platform-designed recommendation system toward the revenue relative to the surplus for viewers. We study a game-theoretic model to explore the optimal design of the recommendation system for a monopolistic content platform. The platform engages in revenue sharing with creators while also considering viewer surplus. Our study shows: 1) There may not be a monotonic relationship between the creator's effort level and the orientation of the recommendation system, depending on the distribution of viewer preferences. 2) If the platform has control over the effort level of the creator, then the platform’s optimal orientation of the recommendation system toward revenue should equal the commission ratio. 3) If the platform lacks control over the effort level, to incentivize the creator to exert a higher effort level, the platform has the motivation to strategically adjust the orientation of the recommendation system. Specifically, the optimal orientation should be either lower or higher than the commission ratio, depending on the relationship between orientation and effort.