Imperfect Credibility Versus No Credibility of Optimal Monetary Policy

Abstract

When the probability of not reneging commitment of optimal monetary policy under quasi-commitment tends to zero, the limit of this equilibrium is qualitatively and quantitatively di¤erent from the discretion equilibrium assuming a zero probability of not reneging commitment for the classic example of the new-Keynesian Phillips curve. The impulse response functions and welfare are di¤erent. The policy rule parameter have opposite signs. The in ‡ation auto-correlation parameter crosses a saddlenode bifurcation when shit…ng to near-zero to zero probability of not reneging commitment. These results are obtained for all values of the elasticity of substitution between goods in monopolistic competition which enters in the welfare loss function and in the slope of the new-Keynesian Phillips curve.

Ramsey Optimal Policy versus Multiple Equilibria with Fiscal and Monetary Interactions

“…If q = 0, the policy maker knows that he is replaced next period. He does static optimization ((βq) 0 = 1) of his current period quadratic loss function subject to a static transmission mechanism where he considers the expectations terms related to the next period policy maker as exogenous intercepts (Chatelain and Ralf (2019b)).…”

Hopf Bifurcation From New-Keynesian Taylor Rule to Ramsey Optimal Policy

“…The discretionary equilibrium where the central bank re-optimizes continuously without any credibility (q = 0) is a point of measure zero with respect to the continuous interval q ∈ ]0, 1]. Since Ramsey optimal policy under quasi-commitment includes the case of very low credibility such as q = 10 −7 the discretionary equilibrium is not a relevant equilibrium (Chatelain and Ralf (2018b)).…”