We present a model of competition between ESG raters who acquire information about multiple unrelated categories and sell ratings. Raters specializing in different categories maximizes the amount of information transmitted and surplus, and can be the equilibrium outcome. When investors place a high value on ESG performance across multiple categories, the unique equilibrium is for the raters to generalize — splitting their effort among the categories, resulting in less informative ratings. Greenwashing by firms can make generalization the only equilibrium. We also demonstrate that specialization maximizes ratings disagreement and, thus, empirical measures of disagreement may be poor measures of surplus.