Credit Ratings, Credit Default Swaps and Credit Correlation
Author(s)
Evans, Leonard Andrew
Type
Thesis
Abstract
This thesis looks at the statistical interaction of credit ratings and Credit Default Swap (CDS)
spreads. Both have been implicated as major contributors to the financial crises of 2007-present.
The body of work contained herein looks to further our understanding of their relationship and in
doing so, I make three empirical contributions to the fields of credit risk and financial economics.
Firstly, in Chapter 2, I uncover a striking empirical artifact contained within CDS correlation
dynamics. Namely, that there is a well-defined credit rating structure embedded in them. Although
much of the extant literature treats credit derivatives and equity as contingent claims on the same
underlying firm value, by contrast, no rating-based structure exists in equity correlations.
In Chapter 3, I demonstrate that rating-based correlation dynamics in CDS markets are not
fully consistent with the traditional framework of financial economics in which a security’s price
merely reflects its fundamental value. I show that the trading behaviour of market participants in
relation to CDS indices, the constituents of which are based on the discrete and somewhat arbitrary
labeling of issuers as either investment-grade or high-yield, drives a distortion in single-name CDS
co-movement. My results can be interpreted as the first evidence of a significant departure from
traditional views of market efficiency in a $30 trillion segment of global derivatives markets.
Finally, in Chapter 4, I go on to explore the complete time-series and cross-sectional interaction
of the credit rating process on CDS spreads. In doing so, I identify that prior to the crisis, credit
rating agencies played a much greater role in the price discovery process of corporate credit risk.
As such, there has been a significant loss of information in credit ratings. This result can be
explained via a loss of confidence in rating agencies due to a spill-over effect of reputational
damage from their role in the collapse of the $3tn structured credit derivatives market. The use
of ex post hyper-inflated AAA ratings on CDOs and RMBS, and the subsequent fall-out from
doing so, has altered how credit market participants react to the information contained in corporate
credit ratings. These results are particularly relevant in light of impending regulatory reform under
the Dodd-Frank act of 2010.
spreads. Both have been implicated as major contributors to the financial crises of 2007-present.
The body of work contained herein looks to further our understanding of their relationship and in
doing so, I make three empirical contributions to the fields of credit risk and financial economics.
Firstly, in Chapter 2, I uncover a striking empirical artifact contained within CDS correlation
dynamics. Namely, that there is a well-defined credit rating structure embedded in them. Although
much of the extant literature treats credit derivatives and equity as contingent claims on the same
underlying firm value, by contrast, no rating-based structure exists in equity correlations.
In Chapter 3, I demonstrate that rating-based correlation dynamics in CDS markets are not
fully consistent with the traditional framework of financial economics in which a security’s price
merely reflects its fundamental value. I show that the trading behaviour of market participants in
relation to CDS indices, the constituents of which are based on the discrete and somewhat arbitrary
labeling of issuers as either investment-grade or high-yield, drives a distortion in single-name CDS
co-movement. My results can be interpreted as the first evidence of a significant departure from
traditional views of market efficiency in a $30 trillion segment of global derivatives markets.
Finally, in Chapter 4, I go on to explore the complete time-series and cross-sectional interaction
of the credit rating process on CDS spreads. In doing so, I identify that prior to the crisis, credit
rating agencies played a much greater role in the price discovery process of corporate credit risk.
As such, there has been a significant loss of information in credit ratings. This result can be
explained via a loss of confidence in rating agencies due to a spill-over effect of reputational
damage from their role in the collapse of the $3tn structured credit derivatives market. The use
of ex post hyper-inflated AAA ratings on CDOs and RMBS, and the subsequent fall-out from
doing so, has altered how credit market participants react to the information contained in corporate
credit ratings. These results are particularly relevant in light of impending regulatory reform under
the Dodd-Frank act of 2010.
Date Issued
2012-04
Date Awarded
2012-07
Copyright Statement
Attribution NoDerivatives 4.0 International Licence (CC BY-ND)
Advisor
El-Jahel, Lina
Cathcart, Lara
Publisher Department
Imperial College Business School
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)