Using a novel data set, we present three findings about the rankings that business schools choose to display on their websites. First, the data strongly rejects patterns predicted by classic models of voluntary disclosure. In contrast with the traditional unraveling hypothesis, top schools are least likely to display their rankings. Second, schools that do poorly in the U.S. News rankings are more likely to disclose their Princeton Review certification, suggesting that schools treat different certifications as substitutes. Third, conditional on displaying a ranking, the majority of schools coarsen information to make it seem more favorable.