Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/220189 
Year of Publication: 
2015
Series/Report no.: 
Discussion Paper No. 100
Publisher: 
Institute for Applied Economic Research (ipea), Brasília
Abstract: 
Based on three versions of a small macroeconomic model for Brazil, this paper presents empirical evidence on the effects of parameter uncertainty on monetary policy rules and on the robustness of optimal and simple rules over different model specifications. By comparing the optimal policy rule under parameter uncertainty with the rule calculated under purely additive uncertainty, we find that parameter uncertainty should make policymakers react less aggressively to the economy's state variables, as suggested by Brainard's 'conservatism principle', although this effect seems to be relatively small. We then informally investigate each rule's robustness by analyzing the performance of policy rules derived from each model under each one of the alternative models. We find that optimal rules derived from each model perform very poorly under alternative models, whereas a simple Taylor rule is relatively robust. We also find that even within a specific model, the Taylor rule may perform better than the optimal rule under particularly unfavorable realizations from the policymaker's loss distribution function.
Document Type: 
Working Paper

Files in This Item:
File
Size
270.08 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.