Most analyses of the US Great Moderation are based on structural VARs, and point toward good luck as the main explanation for the recent macroeconomic stability. Based on an estimated New-Keynesian model where the only source of change is the move from passive to active monetary policy, we show that (i) the theoretical VAR innovation variances for all series decrease across regimes; (ii) VAR-based counterfactuals assign a minor role to improved policy; and (iii) VAR impulse-response functions to a monetary shock exhibit little variation across regimes. Our analysis suggests that existing VAR evidence is also compatible with the “good policy” hypothesis. (JEL C32, C52, E13, E52, N12)