Description:

Christian Wolf (MIT): "Evaluating Policy Counterfactuals: A 'VAR-Plus' Approach" with Tomas E. Caravello and Alisdair McKay

Abstract: In a general family of linearized structural macroeconomic models, the counterfactual evolution of the economy under alternative policy rules is fully pinned down by: (i) reduced-form projections with respect to a large information set; and (ii) the causal effects of policy on macroeconomic aggregates. This identification result motivates a three-step approach. First, the researcher constructs (i) directly from time-series data. Second, she partially estimates (ii) using semi-structural methods for policy shock analysis. Third, if needed, she uses structural models of policy transmission to extrapolate beyond those observed policy experiments, giving the missing parts of (ii). We document attractive robustness properties of this procedure, and use it to study U.S. business-cycle fluctuations under alternative assumptions on the conduct of monetary policy.