Fixing feedback revision rules in online markets

Abstract

Feedback withdrawal mechanisms in online markets aim to facilitate the resolution of conflicts during transactions. Yet, frequently used online feedback withdrawal rules are flawed and may backfire by inviting strategic transaction and feedback behavior. Our laboratory experiment shows how a small change in the design of feedback withdrawal rules, allowing unilateral rather than mutual withdrawal, can both reduce incentives for strategic gaming and improve coordination of expectations. This leads to less trading risk, more cooperation, and higher market efficiency.

Behavioral market design

“…Bolton, Greiner, and Ockenfels (2013) show how this can be achieved in feedback systems by changing the rules by which feedback information flows through the market, leading to more accurate reputation information, more trust, and more efficient trading. In response to such insights, eBay supplemented its old two-sided feedback system with a one-sided system (called “detailed seller ratings”), Airbnb created a blind feedback system where transaction partners cannot see each other's feedback until they have left their own, and Uber makes it difficult for passengers to identify a particular feedback giver (e.g., Bolton, Greiner, & Ockenfels, 2018, 2020, 2023; Fradkin, Grewal, & Holtz, 2019).…”