While episodes of financial distress are followed by large and persistent drops in economic activity, structural time series analyses point to relatively mild and transitory effects of financial market disruptions. We argue that these seemingly contradictory findings are due to the asymmetric effects of financial shocks, which have been predicted theoretically but not taken into account empirically. We estimate a model designed to identify the (possibly asymmetric) effects of financial market disruptions, and we find that a favorable financial shock—an easing of financial conditions—has little effect on output, but an adverse shock has large and persistent effects. In a counterfactual exercise, we find that over two-thirds of the gap between current US GDP and its 207 precrisis trend was caused by the 2007–2008 financial shocks.