Uncertainty and Monetary Policy in the US: A Journey into Non-Linear Territory
Melbourne Institute Working Paper No. 06/17
This paper estimates a non-linear Interacted VAR model to assess whether the real effects of monetary policy shocks are milder during times of high uncertainty. In a novel way, uncertainty, i.e., the conditioning indicator discriminating “high” and “low” uncertainty states, is modelled endogenously in the VAR and is found to reduce after an expansionary shock. Generalized Impulse Response Functions à la Koop, Pesaran and Potter (1996) suggest that monetary policy shocks are significantly less powerful during uncertain times, with the peak reactions of a battery of real variables being about two-thirds milder than those during tranquil times. Among the theoretical explanations proposed by the literature, real option effects and precautionary savings appear the ones supported by our results.
- Monetary policy shocks, Non-Linear Structural Vector Auto-Regressions, Interacted VAR, Generalized Impulse Response Functions, uncertainty