Back to search

Athens University Economics and Business (AUEB)

Essays on monetary policy in models with frictions

Abstract

dc:description

The scope of the current PhD thesis is to study monetary policy in models with various forms of nominal rigidities and frictions, including price stickiness, wage stickiness, with a focus on bounded rational expectations, and to study optimal monetary policy, when the latter is both conventional and unconventional. This work is divided into three distinct parts, as follows. Chapter 1 studies the impact of bounded rationality in short-term dynamics of macroeconomic variables, in simple, widely used, dynamic stochastic general equilibrium (DSGE) models. Bounded rational expectations are modeled using the learning approach of Evans and Honkapohja (2001) and implemented using three mechanisms, in particular Kalman learning, constant-gain learning and ordinary least squares. Starting with a simple Real-Business cycles model and a New-Keynesian (NK) model, with and without physical capital, we study the stability properties of the learning algorithms examined and whether agent’s initial beliefs affect short-term dynamics. To our knowledge, there exist a few papers highlighting the importance of initial conditions in the learning procedure and how these affect short-term dynamics. Key works in this field are Sargent (1993a), Benveniste et al. (1990),Carceles-Poveda and Giannitsarou (2007), and Ferrero (2007). We find that the presence of capital and the initial conditions affect short-term dynamics, under all learning mechanisms. Moreover, using the NK model we conclude that the distance of initial beliefs from the rational expectations (RE) can lead to changes in the optimized response of the policy-maker to output and inflation gap in the Taylor rule, relatively to the respective one under RE, indicating that optimal monetary policy changes in the presence of learning. In particular, we extend the work of Ferrero (2007) and contribute to the related literature by finding that under learning, the longer the distance of initial beliefs from the rational expectations (RE) solutions, the more aggressive the policy-maker should be to tackle inflation. Furthermore, depending on the relative preference of the policy-maker to output stabilization, the policy-maker should also be more aggressive towards narrowing the output gap, in comparison to RE, in the case of a cost-push shock. Finally, we conclude that under learning, a fast convergence of output is not always desirable, in line with the work of Ferrero (2007).In chapter 2 we investigate the role of learning in a two-country, DSGE model to analyze the differences in the economic cycles between the Core and the Periphery in the Euro Area for the period22000 to 2019. By incorporating learning, we aim to capture the impact of the endogenous evolution of beliefs in the two areas, highlighting behavioral factors beyond the observed discrepancies in the boom and bust cycles of Core and Periphery in this period. Following Milani (2007) we compare the performance between rational expectations (RE) models with New-Keynesian frictions (price, interest rate and wage stickiness) and habits with models augmented with learning using Bayesian techniques. We also implement second moment analysis, following the work of Eusepi and Preston (2007). Our findings suggest that learning plays a significant role in explaining the divergent economic performances of the Core and the Periphery and that mechanical sources of persistence, like habit formation and interest rate stickiness become redundant under learning. Furthermore, we conclude that a monetary authority that aims to minimize welfare losses from output and inflation volatility should react more intense to stabilize inflation under learning. Finally, in chapter 3, we study optimal, Ramsey policies when the fiscal and monetary authorities are treated as different policy entities, in a general equilibrium model. Monetary policy is both “conventional” and “nonconventional”, where the former has to do with the use of a policy interestrate and the latter with balance sheet policies, like purchases of government bonds and transfers to the Treasury. We compare the performance of this model, in terms of welfare, in comparison to the standard case where there is a consolidated fiscal and monetary authority which implements aconventional monetary policy. We find that when the policy maker is allowed to implement balance sheet policies, a higher level of welfare can be achieved and that the optimized response of the policymaker in deviations of inflation and output from their targets does not take corner values as in the standard case.

Degree

thesis:*
Grantor dc:publisher
Athens University Economics and Business (AUEB)
Year dc:date
2023

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Mavrigiannakis, Konstantinos
  • Μαυριγιαννάκης, Κωνσταντίνος

Subjects

dc:subject × 16

Rights

Language dc:language
eng

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/54596

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Mavrigiannakis, Konstantinos; Μαυριγιαννάκης, Κωνσταντίνος. Essays on monetary policy in models with frictions. Athens University Economics and Business (AUEB), 2023. http://hdl.handle.net/10442/hedi/54596