In a series of papers, Benhabib, Schmitt-Grohé and Uribe (2001a, 2001b, 2001c and 2002) have shown that active feedback interest rules may lead to unexpected consequences in a DGEM (Dynamic General Equilibrium Model): indeterminacy, deflation traps, large cyclical instability, and even lead to chaotic dynamics under standard sets of parameter values. The specification of the feedback rule used implies that the nominal interest rate is strictly positive and strictly increasing in the inflation rate. This paper concentrates in particular on the (2002b) model and puts forward three basic points: (i) the model developed by Benhabib et al. seems to display an undesirable characteristic which apparently has been unnoticed so far: the more aggressive the central bank becomes towards the fighting of inflation, the more unstable the economy will be; (ii) the model does not take into account that, given the particular nature of the dynamics that it comes up with (chaotic dynamics), tiny perturbations to the nominal interest rate may actually stabilize the economy without producing modifications to the original model, apart from locally changing its type of stability; (iii) ignorance about the true state of initial conditions are not a serious impediment to stabilize the endogenous dynamics in the model. We argue that the conventional view of economic policy in nonlinear general equilibrium models ?? when endogenous fluctuations exist in optimizing models, the associated policy advice should be laissez--faire ?? seems to be based on a misconception of nonlinear chaotic dynamics in general.

--

Optimal Monetary Policy, Interest Rate Rules, Endogenous Fluctuations and Stabilization