VAR Analysis and the Great Moderation

Abstract

Most analyses of the U.S. Great Moderation have been based on structural VAR methods, and have consistently pointed towards good luck as the main explanation for the greater macroeconomic stability of recent years. Based on an estimated New-Keynesian model in which the only source of change is the move from passive to active monetary policy, we show that VARs may misinterpret good policy for good luck. First, the policy shift is sufficient to generate decreases in the theoretical innovation variances for all series, and decreases in the variances of inflation and the output gap, without any need of sunspot shocks. With sunspots, the estimated model exhibits decreases in both variances and innovation variances for all series. Second, policy counterfactuals based on the theoretical structural VAR representations of the model under the two regimes fail to capture the truth, whereas impulse-response functions to a monetary policy shock exhibit little change across regimes. Since these results are in line with those found in the structural VARbased literature on the Great Moderation, our analysis suggests that existing VAR evidence is compatible with the 'good policy' explanation of the Great Moderation.

Keywords: Great Moderation, DSGE models, indeterminacy, vector autoregressions

JEL Classification: E38, E52