Title: Sign restrictions in smooth transition VAR models
Authors: Martin Bruns - University of East Anglia (United Kingdom) [presenting]
Michele Piffer - Queen Mary, University of London (United Kingdom)
Abstract: A way of identifying Smooth Transition Vector Autoregressive models using sign restrictions is developed. In so doing, we offer an alternative to the recursive identification typically used for these models. What makes sign restrictions tractable despite the nonlinearities of the model is the use of a Bayesian approach. We then use the model and argue that monetary policy shocks have a smaller effect on output when the economy is experiencing relatively high levels of uncertainty. The results cannot be driven by the higher price flexibility documented by the literature in periods when uncertainty is high. In fact, we find that not only output but also prices display a limited response to monetary shocks when uncertainty is high. The results suggest that the effectiveness of monetary policy is driven by mechanisms that go beyond price stickiness.