Title: On the fundamentalness of nonfundamentalness in DSGE models
Authors: Marco Maria Sorge - University of Salerno (Italy) [presenting]
Abstract: The role of indeterminacy (e.g. equilibrium non-uniqueness) is explored for the VAR-based analysis of dynamic stochastic general equilibrium (DSGE) models. By means of simple examples, we first show that determinate models may well entail nonfundamental (i.e. noninvertible) equilibrium reduced forms - which only (and uniquely) depend on the structural (fundamental) shocks - whereas indeterminate ones may actually deliver fundamental representations in the observables. We then investigate conditions for the emergence of nonfundamentalness in a general class of (linearized) DGSE models, which nests the prototypical New Keynesian framework as a special case. Monte Carlo simulations show that an indeterminate equilibrium model need not engender nonfundamentalness, even when its determinate counterpart always proves nonfundamental. As a main implication, detecting a causal representation of the data cannot be interpreted as evidence of determinacy.