We study the impact of consumer quality-based self-selection on online reviews. Consumers differ in their expertise, which has two effects. First, expertise is instrumental to choice: Experts purchase better products than Non-Experts. Second, because of their superior choices, Experts endogenously form higher reference points, which leads them to post harsher ratings for a given quality. Combined, these two facts imply a bias against higher-quality products. When this bias gets large, ratings are non-monotonic in quality: lower-quality products can obtain higher ratings than superior alternatives, thanks to the lower standard they are held to.

I study the impact of quality disclosures on buyers' ratings using data from an online recommender system. Disclosures may alter expectations on sellers' quality and affect buyers' rating behavior. In particular, if buyers' utility depends on their expectations, a positive disclosure of quality such as an award may lead to buyers' disappointment, negatively influencing their ratings. I identify the disappointment effect in moviegoers' ratings originated from the rise in expectations due to movies' nominations for the Academy of Motion Picture Arts and Sciences awards. I control for the selection of moviegoers who watch and rate movies before or after nominations with a non-parametric matching technique. After nominations, ratings for nominated movies significantly drop relative to ratings for movies that were not nominated. This short-term disappointment effect reduces the rating premium of nominated movies by more than five percent.