This paper assesses the extent to which rare-disaster models explain the equity premium when calibrated with subjective disaster probabilities from the Survey of Professional Forecasters (SPF). In each quarter, I interpret the SPF’s lowest GDP forecast bin and its associated probability as a conservative upper-bound proxy for disaster risk and incorporate the resulting time series into canonical Barro-type models. While these models capture elevated equity premia during crisis episodes, they fail to account for equity premia in normal periods, implying an average premium of approximately 3.3\%, roughly half the historical U.S. average. This suggests that standard disaster models overlook important mechanisms or risk channels relevant for asset pricing, casting doubt on the sufficiency of disaster risk as a standalone explanation for the equity premium.