Economic dynamics of defaults in a heterogeneous world

Tsafos, Yannis D. (2019) Economic dynamics of defaults in a heterogeneous world. PhD thesis, University of Glasgow.

Full text available as:
[img]
Preview
PDF
Download (1MB) | Preview

Abstract

This thesis investigates the inter-relationship between incomplete market arrangements, limited commitment in financial contracts and endogenous unsecured consumer bankruptcies in a heterogeneous agent general equilibrium framework. There are three chapters that complete this thesis. The following describes a brief abstract of the three chapters.

Chapter 1: We construct a general equilibrium heterogeneous agent model where a continuum of ex-ante identical agents face uninsurable idiosyncratic shocks. Consumption insurance is restricted to a financial contract in the form of a one period non-contingent bond that is not enforceable. The existence of sequential incomplete markets along with incomplete contracts results in endogenous defaults in equilibrium. A novelty in our framework is the presence of partial defaults and the specification of the stochastic labor endowment shock. We discretize a normal mixture autoregressive process that can capture higher order moments of the earnings distribution. The calibrated model delivers a significant mass of households amounting to 18.9% who hold zero or negative wealth. Furthermore, about 28.7% of the households have a history of bad credit implying that they have limited access to the credit markets. Importantly, these statistics match their data counterparts almost perfectly. Finally, we highlight the fragile assumptions of asymmetric default costs prevalent in the sovereign default literature.

Chapter 2: In the second chapter, we extend our model in two directions. Firstly, we incorporate a risk neutral financial intermediary that takes deposits and offers loans. In this way, we allow a price schedule that is contingent on the loan size, idiosyncratic earnings and employment state. Secondly, we model defaults that resemble the U.S. Chapter 13 bankruptcy law. In particular, defaulted agents face a temporary exclusion from the credit markets along with an earnings loss structured in the form of a wage garnishment. Our results highlight the role of financial intermediaries and risk-neutral pricing on the bankruptcy rates. We also deliver wealth and income distribution moments including the inequality indexes that match household survey data estimates.

Chapter 3: In the third and final chapter, we capture the effects of a “credit crunch” and a “credit easing” shock on the real activity in our heterogeneous agent economic environment. Shocks take the form of a gradual tightening or loosening of the borrowing constraint. Periods of adjustment are set to mimic the Global Financial Crisis episode (2008-09) and the credit boom periods (2000-07). We confirm recent results in the post-crisis literature and find that tight credit generates a recession in the short run. However, we observe that in the long run a credit crunch increases productive capital in turn leading to an output growth. An interesting finding in this chapter is that prolonged periods of credit expansion drives an initial consumption driven output boom accompanied with excessive debt followed by increasing default rates leading to a destruction of productive capital and ending up in a sever recession.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Keywords: Households defaults, heterogeneous agents models, general equilibrium.
Subjects: H Social Sciences > HB Economic Theory
Colleges/Schools: College of Social Sciences > Adam Smith Business School > Economics
Supervisor's Name: Tsoukalas, Professor John
Date of Award: 2019
Depositing User: MR IOANNIS TSAFOS
Unique ID: glathesis:2019-74269
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 28 Jun 2019 08:39
Last Modified: 19 Aug 2019 14:49
Thesis DOI: 10.5525/gla.thesis.74269
URI: http://theses.gla.ac.uk/id/eprint/74269

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year