ESRI Discussion Paper Series No.312
Monetary Policy Regime Shifts Under the Zero Lower Bound:
An Application of a Stochastic Rational Expectations Equilibrium to a Markov Switching DSGE Model
Abstract
I extend a simple new Keynesian model with the Markovswitchingtype Taylor rule introduced by Davig and Leeper (2007 ) by incorporating the constraint of the zero lower bound (ZLB), using the concept and algorithms of the stochastic rational expectations equilibrium proposed by Billi (2013). According to the calibration, when an economy does not face the ZLB constraint, there is no gap in the fluctuation of output and inflation between stochastic expectations and perfect foresight because of the linear policy functions. In contrast, once negative aggregate demand shocks make the nominal interest rate hit the ZLB under stochastic expectations, unlike perfect foresight, intensifying uncertainty plays an important role in further declines of the output and price level even in response to the same shock, regardless of the monetary policy regime adopted. The calibration also indicates the possibility that the steady states of a model, in the absence of the ZLB, are underestimated in periods of deflation, since the means often used as estimates of the steady states are biased downward from these. The analysis sheds light on an exit strategy from the zero interest rate policy, since a passive policy regime reduces the expected interest rate and induces both the expected output and the inflation to increase under the ZLB.
Structure of the whole text(PDFFormat 1 File)

Full Text(PDFFormat 720 KB)

page1Abstract

page21 Introduction

page42 Motivation

page6Figure 1: Perfect foresight vs. Stochastic Rational Expectationsy

page7Figure 2: Expected values of Markovswiching DSGE model under stochastic rational expectations


page83 Model

page83.1 Setup

page103.2 Stochastic Rational Expectation Equilibrium (SREE)

page153.3 Application of an SREE to an MSDSGE Model


page174 Calibration

page174.1 Method

page184.2 Calibration Parameters

page19Table 1: Parameter Values


page205 Policy Evaluation

page205.1 Analyses of Policy Functions

page205.2 Comparison of the Impulse Response Functions

page21Figure 3: Policy functions; nonZLB vs. ZLB

page22Figure 4: Policy functions; Stochastic Expectations vs. Perfect Foresight

page23Figure 5: Policy functions; Active Policy Regime vs. Passive Policy Regime


page245.3 Monte Carlo Study of the Model

page24Active Regime vs. Passive Regime

page25Figure 6: Impulse response of Aggregate Demand Shock conditional on each

page26Figure 7: Impulse response of Aggregate Demand Shock under each expectations

page28Table 2: Monte Carlo Simulations conditional on Specified Regimes

page29Figure 8: Intuitive Interpretation of Calibration Results; Active Policy vs. Passive Policy

page30Figure 9: Calibrations; Active Policy vs. Passive Policy


page31RegimeSwitching Policy vs. Fixed Policy

page31The Impact of Mitigating the Uncertainty of Demand Shocks to an Economy

page32Table 3: Monte Carlo Simulations under Regime Switching and Fixed Policies

page33Figure 10: Calibrations; Regime Switcthing Policy vs. Fixed Policy

page34Figure 11: 20% Reduction of St.D of Aggregate Demand Shock

page35Table 4: 20% Reduction of Std Dev of Aggregate Demand Shock under Regime Switching and Fixed Policies




page366 Conclusion

page37References