I am usually
suspicious of attempts to "integrate" market, credit and operational risk. They
so often describe the individual risks but leave it up to the reader to figure
out the integration business. At least this book tries.
It starts off with
three "siloed" chapters covering market, credit and operational risk. It then
has a chapter on extreme value theory. There is another assessing
strengths and weaknesses of Basel II. It is only in the last chapter that the
topic of integration is taken up.
That chapter doesn't build on the earlier chapters.
Rather, it makes a fresh start. It proposes that a bank identify integrated key
risk indicators (IKRIs). These are, I suppose, observable quantities that appear
correlated with at least two of the three risk types—market, credit or
operational risk. By monitoring these, the institution can get a sense of its
"integrated" risk. At least, that is the idea. The authors provide some
beautiful two- and three-dimensional graphics of hypothetical outputs from an IKRI-based system.
A lot of information isn't provided. How do you
determine if an IKRI is "correlated" with, say, operational risk? How do you
know if your set of IKRIs capture all, or at least most of, your integrated
risk? How do you calibrate the model? What data requirements are there, and what
challenges does data gathering and cleaning pose? What are the analytics behind
all the beautiful graphical outputs?
Contents
Introduction
1. Market Risk
2. Credit Risk
3. Operational Risk in
Banking Organizations
4.
Characteristics, Strengths and Weaknesses of Basel II
5. Integrated Risk Management Framework
Conclusion
It would have been nice if the authors had given a
list of useful IKRIs. They don't. I spent some time on the book but didn't come
across a single example of an IKRI. Everything is abstract. Given all the
missing details, I think the authors' ideas are still at a "pie in the sky"
stage.
A shortcoming of the book is a large number of
factual errors. For example, the authors state that linear and Monte Carlo VaR
measures are based on the theory of CAPM, that Harry Markowitz published CAPM in
1952, and that the VaR measures make the same restrictive assumptions as CAPM
(markets are frictionless; investors have homogenous beliefs; there is a risk
free rate at which all investors can borrow or lend, etc.). All of this is
wrong.
We are a long way from ever integrating market,
credit and operational risk in any sort of meaningful way. Some, including
myself, don't believe it will ever happen. I think this book is useful for
reminding us of the challenges such a venture would face, and how far we have to
go. [December 1, 2006]
.gif)
