Abstract:
We study a simple, microfounded macroeconomic system in which the monetary authority employs a Taylor-type policy rule. We analyze situations in which the self-confirming equilibrium is unique and learnable according to Bullard and Mitra (2002). We explore the prospects for the use of ‘large deviation’ theory in this context, as employed by Sargent (1999) and Cho, Williams, and Sargent (2002). We show that our system can sometimes depart from the self-confirming equilibrium towards a non-equilibrium outcome characterized by persistently low nominal interest rates and persistently low inflation. Thus we generate events that have some of the properties of “liquidity traps” observed in the data, even though the policymaker remains committed to a Taylor-type policy rule which otherwise has desirable stabilization properties.
Keywords:Learning; monetary policy rules; escape dynamics (search for similar items in EconPapers) JEL-codes:E52E32D83D84 (search for similar items in EconPapers) New Economics Papers: this item is included in nep-mac, nep-mon and nep-pol Date: Written Note: An earlier version was presented at the research conference "Expectations, Learning and Monetary Policy" August 2003 sponsored by the Deutsche Bundesbank, the Journal of Economic Dynamics and Control (JEDC) and the Center for Financial Studies (CFS). This paper was originally prepared for a workshop on "Learning and Model Misspecification," in Cleveland, Ohio. We thank the Federal Reserve Bank of Cleveland for sponsoring this event, and John Carlson for organizing it. We also thank discussants Stephanie Schmitt-Grohe, Bob Tetlow, Leopold von Thadden, and seminar participants. View list of referencesView citations in EconPapers
Related works: Working Paper: Escapist policy rules (2003) Journal Article: Escapist policy rules (2005) This item may be available elsewhere in EconPapers: Search for items with the same title.