Social Learning and Strategic Pricing with Rating Systems

Despite widespread use in online transactions, rating systems only provide summary statistics of buyers' diverse opinions at best. To investigate the consequences of this coarse form of information aggregation, we consider a dynamic lemons market in which buyers share their evaluations anonymously through a rating system. When the buyers have diverse preferences, the value of a good rating depends endogenously on the seller's pricing strategy, which in turn creates complicated dynamic interactions and results in stochastic price fluctuations. Occasional flash sales induced by the rating system yield a non-trivial welfare effect that stands in sharp contrast to standard adverse selection models: all buyers are weakly better off with information asymmetry than without. Incentivizing buyers to leave ratings may backfire by exacerbating the seller's strategic pricing incentives.