From High Bar to Uneven Bars: The Impact of Information Granularity in Quality Certification

Abstract

Quality certification is ubiquitously used in markets with asymmetric information, but little is known about the impact of the number of certification tiers on market outcomes. Exploiting a field experiment on a large e-commerce marketplace with an existing two-tier certification, we study the impact of introducing a new certification tier that is less history-dependent and less demanding than the existing top tier. Consistent with the theoretical literature on information design, we find that the change in information granularity among the previously non-certified sellers increases demand for high-quality young sellers, incentivizes their quality provision, and increases their chance of eventually obtaining the top-tier certification. Moreover, the three-tier certification structure prompts established sellers with higher effort cost to reduce their effort, opt out of the top tier, and adopt the new tier instead. Lastly, we provide evidence that the net impact of introducing the new certification tier on seller effort depends on how history-dependent the certification requirements are in a market.

Keywords: Quality certification, reputation systems, information granularity, e-commerce

JEL Classification: D47, D82, L86