Monetary-Fiscal Mix and Inflation Performance: Evidence from the U.S.

There has been a lot of interest recently in developing small scale rule-based empirical macro models for the analysis of monetary policy. These models, based on the conventional view that inflation stabilization should be a concern of monetary policy only, have typically neglected the role of fiscal policy. We start with the evidence that a baseline VAR-augmented Taylor rule can deliver recurrent mispredictions of inflation in the U.S. before 1987. We then show that a fiscal feed-back rule, in which the primary deficit reacts to both the output gap and the government debt, can well characterize the behavior of fiscal policy throughout the sample. However, by employing Markov-switching methods, we find evidence of substantial instability across fiscal regimes. Yet this precisely happens QTR{it}{before 1987}. We then augment the monetary VAR with a fiscal policy rule and control for the endogenous regime switches for both rules. We find that only over time windows belonging to the pre-1987 period the model based on the two rules can predict the behavior of inflation better than the one based just on the monetary policy rule. QTR{it}{After 1987}, when fiscal policy is estimated to switch to a regime of fiscal discipline, the monetary-fiscal mix can be appropriately described as a regime of monetary dominance. Over this period a monetary policy rule based model is always a better predictor of the inflation behavior than the one comprising both a monetary and a fiscal rule.