We examine whether ambiguity is priced in the cross-section of expected stock returns. Using the cross-sectional dispersion in real-time forecasts of real GDP growth as a measure for ambiguity, we find that high ambiguity beta stocks earn lower future returns relative to low ambiguity beta stocks. This negative predictive relation between the ambiguity beta and future returns is consistent with theory that predicts the marginal utility of consumption rises when ambiguity is high. A long-short portfolio formed on the ex-ante measure of the ambiguity beta generates an ambiguity premium that is statistically and economically significant. We also find that time variation of the ambiguity premium is systematically related to changing economic conditions, but is not related to the investor sentiment index. Our results are robust to controlling for stock characteristics that are known to predict cross-section returns.

JEL classification: G12, G14

Keywords: Ambiguity; Dispersion of beliefs; Stock returns