Design of Platform Reputation Systems: Optimal Information Disclosure

Reputation systems play a central role in the elimination of information asymmetry between sellers and consumers in a variety of marketplaces. Despite their crucial role, the design of reputation systems remains a complicated issue. Some platforms disclose as much product information as possible by encouraging consumers to leave reviews or ratings through the reputation system, whereas others disclose only partial product information to the public. Platforms, as the designers of such reputation systems, are thus an additional player in the seller-consumer game. This paper studies the amount of information disclosure in a reputation system that optimizes the platform’s profit. In a three-player game, sellers of products with different base qualities decide whether to invest in quality improvement; the platform decides how many ratings to disclose to prospective consumers; and prospective consumers make purchase decisions after observing the available ratings as indicators of product quality. We model consumers as Bayesian learners and specify conditions under which withholding some information (i.e., partial disclosure instead of complete disclosure) can maximize the platform’s profit. Two main effects favor partial disclosure: First, a quality-improving effect makes sellers more willing to invest in quality. Although complete disclosure can eliminate information asymmetry, it can decrease the seller’s incentive to invest in quality, which ultimately hurts the platform’s profit. Second, a sales-increasing effect leads to more transactions and thus higher profit for the platform.