This paper investigates the puzzling negative empirical relationship between tail risk and expected return. Using Expected Shortfall as a measure of tail risk, this study decomposes it into elemental systematic and idiosyncratic components which allow for a deep probing of the relationship. The evidence suggests that while Expected Shortfall is an important determinant of expected returns, it earns a negative risk premium in stark contradiction of theory. After verifying empirically the negative tail risk premium anomaly, the paper investigates how the systematic and idiosyncratic components of tail risk influence expected returns and challenges prevailing explanations of tail risk premia. The negative tail risk premium anomaly is driven mainly by idiosyncratic Expected Shortfall. Moreover, contradicting recent findings in the literature which document that systematic tail risk has a positive impact on expected returns, the systematic Expected Shortfall is either negative or at times insignificantly positive. This is a new and puzzling finding. These findings contribute to a deeper understanding of the drivers behind tail risk anomalies and hold implications for both investment strategies and the interpretation of tail risk-return relationships.