When studying the problem of the emergence of superstars, scholars face great difficulties in measuring talent, obtaining confidential data on earnings, and finding econometric techniques that lead to results that are robust to the presence of outliers (superstars). In this paper we use an original dataset from the Pokemon trading card game in which (i) there is no unidentifiable heterogeneity and (ii) all characteristics of individuals are public domain. To prevent the results to be distored by the presence of outliers, we estimate the «fair» price of each individual, using the robust «Least Trimmed of Squares» regression technique in a hedonic prices framework, and check the effective price at which they are sold. This allows to identify superstars, i.e. individuals that are sold at a price which represents several times their intrinsec value. We find that the two main theories of superstars developed by Rosen (1981), who awards a central importance to talent, and by Adler (1985), who awards more importance to the need of consumers to share a common culture are complementary and not mutually exclusive as is often claimed.
