We study a broad class of asset pricing models in which the stochastic discount factor (SDF) can be factorized into an observable component and a potentially unobservable, model-specific, one. Exploiting this decomposition we derive new entropy bounds that restrict the admissible regions for the SDF and its components. Without using this decomposition, to a second order approximation, entropy bounds are equivalent to the canonical Hansen-Jagannathan bounds. However, bounds based on our decomposition have higher information content, are tighter, and exploit the restriction that the SDF is a positive random variable. Our information-theoretic framework also enables us to extract a non-parametric estimate of the unobservable component of the SDF. Empirically, we find it to have a business cycle pattern, and significant correlations with both financial market crashes unrelated to economy-wide contractions, and the Fama-French factors. We apply our methodology to some leading consumption-based models, gaining new insights about their empirical performance.
