Download - Prm Handbook III
-
8/9/2019 Prm Handbook III
1/415
The Professional Risk Managers Handbook
A Comprehensive Guide to Current Theory and Best Practices
___________________________________________________
Edited by Carol Alexander and Elizabeth Sheedy
Introduced by David R. Koenig
Volume III: Risk Management Practices
The Official Handbook for the PRM Certification
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
2/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 2
Copyrighted Materials
Published by PRMIA Publications, Wilmington, DE
No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any
form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise,
except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without
either the prior written permission of the Publisher, or authorization through payment of the
appropriate per-copy fee to the Publisher. Requests for permission should be addressed to
PRMIA Publications, PMB #5527, 2711 Centerville Road, Suite 120, Wilmington, DE, 19808 or
via email to [email protected] .
Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best
efforts in preparing this book, they make no representations or warranties with respect to the
accuracy or completeness of the contents of this book and specifically disclaim any implied
warranties of merchantability of fitness for a particular purpose. No warranty may be created or
extended by sales representatives or written sales materials. The advice and strategies contained
herein may not be suitable for your situation. You should consult with a professional where
appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other
commercial damages, including but not limited to special, incidental, consequential or other
damages.
This book is also available in a !Sealed" digital format and may be purchased as such by members
of the Professional Risk Managers International Association at www.PRMIA.org .
ISBN 0-9766097-0-3 (3 Volume Set)
ISBN 0-9766097-3-8 (Volume III)
PDF created with pdfFactory Pro trial version www.pdffactory.com
mailto:[email protected]://www.prmia.org/http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/http://www.prmia.org/mailto:[email protected]
-
8/9/2019 Prm Handbook III
3/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 3
Contents
Introduction .................................................................................................................................................7
Preface to Volume III: Risk Management Practices .......................................................................9
III.0 Capital Allocation and RAPM..........................................................................................................13
III.0.1 Introduction ..........................................................................................................................13
III.0.2 Economic Capital .................................................................................................................17
III.0.3 Regulatory Capital ................................................................................................................24
III.0.4 Capital Allocation and Risk Contributions.......................................................................31
III.0.5 RAROC and Risk-Adjusted Performance........................................................................35
III.0.6 Summary and Conclusions..................................................................................................39
References................................................................................................................................................40
III.A.1 Market Risk Management........................................... ..................................................................43
III.A.1.1 Introduction......................................................................................................................43
III.A.1.2 Market Risk.......................................................................................................................43
III.A.1.3 Market Risk Management Tasks....................................................................................45
III.A.1.4 The Organisation of Market Risk Management..........................................................47
III.A.1.5 Market Risk Management in Fund Management........................................................49
III.A.1.6 Market Risk Management in Banking...........................................................................57
III.A.1.7 Market Risk Management in Non-financial Firms .....................................................66
III.A.1.8 Summary............................................................................................................................72References................................................................................................................................................73
III.A.2 Introduction to Value at Risk Models.........................................................................................75
III.A.2.1 Introduction......................................................................................................................75
III.A.2.2 Definition of VaR............................................................................................................76
III.A.2.3 Internal Models for Market Risk Capital......................................................................78
III.A.2.4 Analytical VaR Models....................................................................................................79
III.A.2.5 Monte Carlo Simulation VaR.........................................................................................81
III.A.2.6 Historical Simulation VaR..............................................................................................86
III.A.2.7 Mapping Positions to Risk Factors ...............................................................................95
III.A.2.8 Backtesting VaR Models.............................................................................................. 109
III.A.2.9 Why Financial Markets Are Not #Normal......................................................... ....... 111
III.A.2.10 Summary......................................................................................................................... 112
References ........................................... .................................................................. ................................ 113
III.A.3: Advanced Value at Risk Models ......................................................... .................................... 115
III.A.3.1 Introduction................................................................................................................... 115
III.A.3.2 Standard Distributional Assumptions........................................................................ 117
III.A.3.3 Models of Volatility Clustering................................................................................... 121
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
4/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 4
III.A.3.4 Volatility Clustering and VaR...................................................................................... 126
III.A.3.5 Alternative Solutions to Non-normality.................................................................... 134
III.A.3.6 Decomposition of VaR................................................................................................ 142
III.A.3.7 Principal Component Analysis.................................................................................... 149
III.A.3.8 Summary......................................................................................................................... 154
References ........................................... .................................................................. ................................ 155
III.A.4 Stress Testing ......................................................... ...................................................................... 157
III.A.4.1 Introduction................................................................................................................... 157
III.A.4.2 Historical Context......................................................................................................... 158
III.A.4.3 Conceptual Context...................................................................................................... 163
III.A.4.4 Stress Testing in Practice ............................................................................................. 164
III.A.4.5 Approaches to Stress Testing: An Overview............................................................ 166
III.A.4.6 Historical Scenarios ...................................................................................................... 168
III.A.4.7 Hypothetical Scenarios................................................................................................. 174
III.A.4.8 Algorithmic Approaches to Stress Testing ............................................................... 182
III.A.4.9 Extreme-Value Theory as a Stress-Testing Method................................................ 186
III.A.4.10 Summary and Conclusions .......................................................................................... 187
Further Reading.................................................................................. .................................................. 187
References ........................................... .................................................................. ................................ 188
III.B.1 Credit Risk Management................................................. ........................................................... 193
III.B.1.1 Introduction................................................................................................................... 193
III.B.1.2 A Credit To-Do List..................................................................................................... 194
III.B.1.3 Other Tasks.................................................................................................................... 207
III.B.1.4 Conclusions.................................................................................................................... 208
References ........................................... .................................................................. ................................ 209
III.B.2 Foundations of Credit Risk Modelling.......................... ........................................................... 211
III.B.2.1 Introduction................................................................................................................... 211
III.B.2.2 What is Default Risk?................................................................................................... 211
III.B.2.3 Exposure, Default and Recovery Processes ............................................................. 212
III.B.2.4 The Credit Loss Distribution...................................................................................... 213
III.B.2.5 Expected and Unexpected Loss ................................................................................. 215
III.B.2.6 Recovery Rates .............................................................................................................. 218
III.B.2.7 Conclusion ..................................................................................................................... 223
References ........................................... .................................................................. ................................ 223
III.B.3 Credit Exposure....................................... ............................................................ ........................ 225
III.B.3.1 Introduction................................................................................................................... 225
III.B.3.2 Pre-settlement versus Settlement Risk....................................................................... 227
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
5/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 5
III.B.3.3 Exposure Profiles.......................................................................................................... 228
III.B.3.4 Mitigation of Exposures .............................................................................................. 236
References ........................................... .................................................................. ................................ 241
III.B.4 Default and Credit Migration ................................................... ................................................. 243
III.B.4.1 Default Probabilities and Term Structures of Default Rates ................................. 243
III.B.4.2 Credit Ratings ................................................................................................................ 247
III.B.4.3 Agency Ratings. ............................................................................................................. 252
III.B.4.4 Credit Scoring and Internal Rating Models .............................................................. 258
III.B.4.5 Market-Implied Default Probabilities........................................................................ 262
III.B.4.6 Credit Rating and Credit Spreads ............................................................................... 268
III.B.4.7 Summary......................................................................................................................... 270
References ........................................... .................................................................. ................................ 271
III.B.5 Portfolio Models of Credit Loss .............................................. ................................................. 273
III.B.5.1 Introduction................................................................................................................... 273
III.B.5.2 What Actually Drives Credit Risk at the Portfolio Level?...................................... 276
III.B.5.3 Credit Migration Framework ...................................................................................... 279
III.B.5.4 Conditional Transition Probabilities $ CreditPortfolioView .................................. 292
III.B.5.5 The Contingent Claim Approach to Measuring Credit Risk ................................. 294
III.B.5.6 The KMV Approach .................................................................................................... 300
III.B.5.7 The Actuarial Approach............................................................................................... 307
III.B.5.8 Summary and Conclusion............................................................................................ 312
References ........................................... .................................................................. ................................ 312
III.B.6 Credit Risk Capital Calculation.............................................................. .................................... 315
III.B.6.1 Introduction.................................................................................................................. 315
III.B.6.2 Economic Credit Capital Calculation ........................................................................ 316
III.B.6.3 Regulatory Credit Capital: Basel I ............................................................................. 320
III.B.6.4 Regulatory Credit Capital: Basel II............................................................................. 324
III.B.6.5 Basel II: Credit Model Estimation and Validation .................................................. 334
III.B.6.6 Basel II: Securitisation.................................................................................................. 336
III.B.6.7 Advanced Topics on Economic Credit Capital ....................................................... 338
III.B.6.8 Summary and Conclusions .......................................................................................... 340
References ........................................... .................................................................. ................................ 341
III.C.1 The Operational Risk Management Framework............................................. ....................... 343
III.C.1.1 Introduction................................................................................................................... 343
III.C.1.2 Evidence of Operational Failures............................................................................... 345
III.C.1.3 Defining Operational Risk........................................................................................... 347
III.C.1.4 Types of Operational Risk.......................................................................................... 348
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
6/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 6
III.C.1.5 Aims and Scope of Operational Risk Management ................................................ 351
III.C.1.6 Key Components of Operational Risk...................................................................... 354
III.C.1.7 Supervisory Guidance on Operational Risk ............................................................. 357
III.C.1.8 Identifying Operational Risk $ the Risk Catalogue ................................................. 358
III.C.1.9 The Operational Risk Assessment Process............................................................... 359
III.C.1.10 The Operational Risk Control Process...................................................................... 364
III.C.1.11 Some Final Thoughts ................................................................................................... 365
References ........................................... .................................................................. ................................ 366
III.C.2 Operational Risk Process Models........ ......................................................... ............................ 367
III.C.2.1 Introduction................................................................................................................... 367
III.C.2.2 The Overall Process ..................................................................................................... 369
III.C.2.3 Specific Tools ................................................................................................................ 372
III.C.2.4 Advanced Models ......................................................................................................... 374
III.C.2.5 Key Attributes of the ORM Framework................................................................... 378
III.C.2.6 Integrated Economic Capital Model.......................................................................... 381
III.C.2.7 Management Actions.................................................................................................... 384
III.C.2.8 Risk Transfer.................................................................................................................. 386
III.C.2.9 IT Outsourcing.............................................................................................................. 388
References ........................................... .................................................................. ................................ 393
III.C.3 Operational Value-at-Risk................. ............................................................ ............................. 395
III.C.3.1 The #Loss Model Approach............................................................... ......................... 395
III.C.3.2 The Frequency Distribution........................................................................................ 401
III.C.3.3 The Severity Distribution ............................................................................................ 404
III.C.3.4 The Internal Measurement Approach ....................................................................... 407
III.C.3.5 The Loss Distribution Approach............................................................................... 411
III.C.3.6 Aggregating ORC.......................................................................................................... 413
III.C.3.7 Concluding Remarks .................................................................................................... 415
References ........................................... .................................................................. ................................ 416
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
7/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 7
Introduction
If you're reading this, you are seeking to attain a higher standard. Congratulations!
Those who have been a part of financial risk management for the past twenty years, have seen it
change from an on-the-fly profession, with improvisation as a rule, to one with substantially
higher standards, many of which are now documented and expected to be followed. Its no
longer enough to say you know. Now, you and your team need to prove it.
As its title implies, this book is the Handbook for the Professional Risk Manager. It is for those
professionals who seek to demonstrate their skills through certification as a Professional Risk
Manager (PRM) in the field of financial risk management. And it is for those looking simply to
develop their skills through an excellent reference source.
With contributions from nearly 40 leading authors, the Handbook is designed to provide you
with the materials needed to gain the knowledge and understanding of the building blocks of
professional financial risk management. Financial risk management is not about avoiding risk.
Rather, it is about understanding and communicating risk, so that risk can be taken more
confidently and in a better way. Whether your specialism is in insurance, banking, energy, asset
management, weather, or one of myriad other industries, this Handbook is your guide.
In this volume, the current and best practices of Market, Credit and Operational risk
management are described. This is where we take the foundations of Volumes I and II and apply
them to our profession in very specific ways. Here the strategic application of risk management
to capital allocation and risk-adjusted performance measurement takes hold. After studying all of
the materials in the PRM Handbook, you will have read the materials necessary for passage of
Exam III of the PRM Certification program.
Those preparing for the PRM certification will also be preparing for Exam I on Finance Theory,
Financial Instruments and Markets, covered in Volume I of the PRM Handbook, Exam II on the
Mathematical Foundations of Risk Measurement, covered in Volume II of the PRM Handbook
and Exam IV - Case Studies, Standards of Best Practice Conduct and Ethics and PRMIA
Governance. Exam IV is where we study some failed practices, standards for the performance of
the duties of a Professional Risk Manager, and the governance structure of our association, the
Professional Risk Managers International Association. The materials for this exam are freely
available on our web site (see http://www.prmia.org/pdf/Web_based_Resources.htm ) and are
thus outside of the Handbook.
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.prmia.org/pdf/Web_based_Resources.htmhttp://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/http://www.prmia.org/pdf/Web_based_Resources.htm
-
8/9/2019 Prm Handbook III
8/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 8
At the end of your progression through these materials, you will find that you have broadened
your knowledge and skills in ways that you might not have imagined. You will have challenged
yourself as well. And, you will be a better risk manager. It is for this reason that we have created
the Professional Risk Managers Handbook.
Our deepest appreciation is extended to Prof. Carol Alexander and Prof. Elizabeth Sheedy, both
of PRMIAs Academic Advisory Council, for their editorial work on this document. The
commitment they have shown to ensuring the highest level of quality and relevance is beyond
description. Our thanks also go to Laura Bianco, President of PRMIA Publications, who has
tirelessly kept the work process moving forward and who has dedicated herself to demanding the
finest quality output. We also thank Richard Leigh, our London-based copyeditor, for his skilful
and timely work.
Finally, we express our thanks to the authors who have shared their insights with us. The
demands for sharing of their expertise are frequent. Yet, they have each taken special time for
this project and have dedicated themselves to making the Handbook and you a success. We are
very proud to bring you such a fine assembly.
Much like PRMIA, the Handbook is a place where the best ideas of the risk profession meet. We
hope that you will take these ideas, put them into practice and certify your knowledge by attaining
the PRM designation. Among our membership are several hundred Chief Risk Officers / Heads
of Risk and tens of thousands of other risk professionals who will note your achievements. They
too know the importance of setting high standards and the trust that capital providers and
stakeholders have put in them. Now they put their trust in you and you can prove your
commitment and distinction to them.
We wish you much success during your studies and for your performance in the PRM exams!
David R. Koenig, Executive Director, Chair, Board of Directors, PRMIA
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
9/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 9
Preface to Volume III: Risk Management Practices
Section III is the ultimate part of The PRM Handbook in both senses of the word. Not only is it
the final section, but it represents the final aims and objectives of the Handbook. Sections I
(Finance Theory, Financial Instruments and Markets) and II (Mathematical Foundations of Risk
Measurement) laid the necessary foundations for this discussion of risk management practices $
the primary concern of most readers. Here some of the foremost practitioners and academics in
the field provide an up-to-date, rigorous and lucid statement of modern risk management.
The practice of risk management is evolving at a rapid pace, especially with the impending arrival
of Basel II. Aside from these regulatory pressures, shareholders and other stakeholders
increasingly demand higher standards of risk management and disclosure of risk. In fact, it
would not be an overstatement to say that risk consciousness is one of the defining features of
modern business. Nowhere is this truer than in the financial services industry. Interest in risk
management is at an unprecedented level as institutions gather data, upgrade their models and
systems, train their staff, review their remuneration systems, adapt their business practices and
scrutinise controls for this new era.
Section III is itself split into three parts which address market risk, credit risk and operational risk
in turn. These three are the main components of risk borne by any organisation, although the
relative importance of the mix varies. For a traditional commercial bank, credit risk has always
been the most significant. It is defined as the risk of default on debt, swap, or other counterparty
instruments. Credit risk may also result from a change in the value of a security, contract or asset
resulting from a change in the counterparty s creditworthiness. In contrast, market risk refers to
changes in the values of securities, contracts or assets resulting from movements in exchange
rates, interest rates, commodity prices, stock prices, etc. Operational risk, the risk of loss
resulting from inadequate or failed internal processes, people and systems or from external
events, is not, strictly speaking, a financial risk. Operational risks are, however, an inevitable
consequence of any business undertaking. For financial institutions and fund managers, credit
and market risks are taken intentionally with the objective of earning returns, while operational
risks are a by-product to be controlled. While the importance of operational risk management is
increasingly accepted, it will probably never have the same status in the finance industry as credit
and market risk which are the chosen areas of competence.
For non-financial firms, the priorities are reversed. The focus should be on the risks associated
with the particular business; the production and marketing of the service or product in which
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
10/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 10
expertise is held. Market and credit risks are usually of secondary importance as they are a by-
product of the main business agenda.
The last line of defence against risk is capital, as it ensures that a firm can continue as a going
concern even if substantial and unexpected losses are incurred. Accordingly, one of the major
themes of Section III is how to determine the appropriate size of this capital buffer. How much
capital is enough to withstand unusual losses in each of the three areas of risk? The measurement
of risk has further important implications for risk management as it is increasingly incorporated
into the performance evaluation process. Since resources are allocated and bonuses paid on the
basis of performance measures, it is essential that they be appropriately adjusted for risk. Only
then will appropriate incentives be created for behaviour that is beneficial for shareholders and
other stakeholders. Chapter III.0 explores this fundamental idea at a general level, since it is
relevant for each of the three risk areas that follow.
Market Risk
Chapter III.A.1 introduces the topic of market risk as it is practised by bankers, fund managers
and corporate treasurers. It explains the four major tasks of risk management (identification,
assessment, monitoring and control/mitigation), thus setting the scene for the quantitative
chapters that follow. These days one of the major tasks of risk managers is to measure risk using
value-at-risk (VaR) models. VaR models for market risk come in many varieties. The more basic
VaR models are the topic of Chapter III.A.2, while the advanced versions are covered in III.A.3
along with some other advanced topics such as risk decomposition. The main challenge for risk
managers is to model the empirical characteristics observed in the market, especially volatility
clustering. The advanced models are generally more successful in this regard, although the basic
versions are easier to implement. Realistically, there will never be a perfect VaR model, which is
one of the reasons why stress tests are a popular tool. They can be considered an ad hoc solution
to the problem of model risk. Chapter III.A.4 explains the need for stress tests and how they
might usefully be constructed.
Credit Risk
Chapter III.B.1 introduces the sphere of credit risk management. Some fundamental tools for
managing credit risk are explained here, including the use of collateral, credit limits and credit
derivatives. Subsequent chapters on credit risk focus primarily on its modelling. Foundations for
modelling are laid in Chapter III.B.2, which explains the three basic components of a credit loss:
the exposure, the default probability and the recovery rate. The product of these three, which
can be defined as random processes, is the credit loss distribution. Chapter III.B.3 takes a more
detailed look at the exposure amount. While relatively simple to define for standard loans,
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
11/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 11
assessment of the exposure amount can present challenges for other credit sensitive instruments
such as derivatives, whose values are a function of market movements. Chapter III.B.4 examines
in detail the default probability and how it can evolve over time. It also discusses the relationship
between credit ratings and credit spreads, and credit scoring models. Chapter III.B.5 tackles one
of the most crucial issues for credit risk modelling: how to model credit risk in a portfolio
context and thereby estimate credit VaR. Since diversification is one of the most important tools
for the management of credit risk, risk measures on a portfolio basis are fundamental. A number
of tools are examined, including the credit migration approach, the contingent claim or structural
approach, and the actuarial approach. Finally, Chapter III.B.6 extends the discussion of credit
VaR models to examine credit risk capital. It compares both economic capital and regulatory
capital for credit risk as defined under the new Basel Accord.
Operational Risk
The framework for managing operational risk is first established in Chapter III.C.1. After
defining operational risk, it explains how it may be identified, assessed and controlled. Chapter
III.C.2 builds on this with a discussion of operational risk process models. By better
understanding business processes we can find the sources of risk and often take steps to re-
engineer these processes for greater efficiency and lower risk. One of the most perplexing issues
for risk managers is to determine appropriate capital buffers for operational risks. Operational
VaR is the subject of Chapter III.C.3, including discussion of loss models, standard functional
forms, both analytical and simulation methods, and the aggregation of operational risk over all
business lines and event types.
Elizabeth Sheedy, Member of PRMIA! s Academic Advisory Council and co-editor of The PRM Handbook.
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
12/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 12
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
13/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 13
III.0 Capital Allocation and RAPM
Andrew Aziz and Dan Rosen1
III.0.1 Introduction
We introduce in this chapter the definitions and key concepts regarding capital, focusing on the
important role that capital plays in financial institutions. Readers should already be familiar with
Chapter I.A.5, which has presented the basic principles behind the capital structure of the firm.
There it was argued that the actual capital $ the physical capital that a firm holds $ should be
distinguished from the optimal level of capital. The optimal capital depends on many things,
including capital targets that are associated with a desired level of #capital adequacy to cover the
potential for losses made by the firm. Capital adequacy is assessed using internal models (for
economic capital) and for banks a certain level of capital is imposed by external standards ( regulatory
capital). Capital adequacy is a measure of a firms ability to remain a going concern. Thus,
targeted levels of capital are direct functions of the riskiness of the business activities or, from a
balance sheet perspective, the riskiness of the assets.
In this chapter you will learn:
the role of capital in financial institutions and the different types of capital;
the key concepts and objectives behind regulatory capital, as well as the main calculations
principles in the Basel I Accord and the current Basel II Accord;
the definition and mechanics of economic capital as well as the methods to calculate it;
the use of economic capital as a management tool for risk aggregation, risk-adjusted
performance measurement and optimal decision making through capital allocation.
This introductory section presents the definition of capital and its role in financial institutions.
We make the distinction between the various types of capital: book capital, economic capital and
regulatory capital. We discuss briefly the use of economic capital as a management tool for risk
aggregation, decision making and performance measurement.
III.0.1.1 Role of Capital in Financial Institution
Banks generate revenue by taking on exposure to their customers and by earning appropriate
returns to compensate for the risk of this exposure. In general, if a bank takes on more risk, it
can expect to earn a greater return. The trade-off, however, is that the same bank will, in general,
increase the possibility of facing losses to the extent that it defaults on its debt obligations and is
forced out of business. Banks that are managed well will attempt to maximise their returns only
1 Algorithmics Inc.
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
14/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 14
through risk taking that is prudent and well informed. The primary role of the risk management
function in a bank is to ensure that the total risk taken across the enterprise is no greater than the
bank s ability to absorb worst-case losses within some specified confidence interval.
In its pure form, capital represents the difference between the market value of a bank s assets and
the market value of its liabilities. Because capital can be viewed as a buffer against insolvency,
capital adequacy is a measure of a bank s ability to remain a going concern under adverse conditions.
In contrast to a typical corporation, the key role of capital in a financial institution such as a bank
is not primarily one of providing a source of funding for the organisation. Banks usually have
ready access to funding through their deposit-taking activities, which can be increased fairly
fluidly. Instead, the primary role of capital in a bank, apart from the transfer of ownership, is to
act as a buffer to:
absorb large unexpected losses;
protect depositors and other claim holders;
provide enough confidence to external investors and rating agencies on the financial
health and viability of the firm.
A firms credit rating can be seen as a measure of its capital adequacy and is generally linked to a
specific probability that the firm will enter into default over some period of time. If we make the
assumption that liabilities are riskless, then the credit rating of a firm becomes a function of the
overall riskiness of its assets and the amount of capital that the bank holds. Firms which hold
more capital are able to take on riskier assets than firms of similar credit rating which hold less
capital.
Typically, the sources of risk within the assets of a firm are classified as follows:
credit risk $ losses associated with the default (or credit downgrade) of an obligor (a
counterparty, borrower or debt issuer);
market risk $ losses associated with changes in market values;
operational risk $ losses associated with operating failures.
Capital represents an ideal metric for aggregating risks across both different asset classes and
across different risk types.
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
15/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 15
III.0.1.2 Types of Capital
We can broadly classify capital into three types:2
Economic capital (EC) $ an estimate of the level of capital that a firm requires to operate its
business with a desired target solvency level. Sometimes this is also referred to as risk
capital .3
Regulatory capital (RC) $ the capital that a bank is required to hold by regulators in order
to operate; this is an accounting measure defined by the regulatory authorities to act as a
proxy for economic capital.
Book capital (BC) $ the actual physical capital held. While in its strictest definition this
should be simply equity capital , more generally this might also include other assets like
liquid debt or hybrid instruments
In practice, many firms hold book capital in excess of the required economic and even regulatory
capital. This reflects both historical and practical business reasons (for example, given their size,
they might be too slow to invest it effectively), as well as their more conservative view on the
applicability of the models. The combined forces of deregulation and the increased market
volatility in the late 1970s motivated many banks to aggressively grow market share and to
acquire increasingly riskier assets on their balance sheets. This emphasis on growth precipitated a
decline of capital levels throughout the 1980s that led to fears of increasing instability in the
international banking system. These concerns motivated the push for the creation of international
capital adequacy standards such as those ultimately established by the Basel Committee on
Banking Supervision (BCBS). The imposition of the Basel I Accord in 1988 proved to be
successful in its objective of increasing worldwide capital levels to desired levels by 1993 and,
ultimately, to reduce the overall riskiness of the international banking system.
In general, EC is meant to reflect the true #fair market value differential between assets and
liabilities, and thus it is limited by the ability to mark to market a balance sheet in a manner that is
indisputable for all key constituencies $ the financial institution, the regulators and the investors
themselves. As such, the determination of EC has traditionally been highly institution-specific.
The foremost objective of regulations, however, is to define an unarguable standard for capital
comparison that creates a level playing field across all financial institutions. Thus, regulatory
capital has traditionally been defined with respect to accounting book value measures rather than
2 This is a general classification, and there are various alternative definitions of capital and terminology used to describethem.3 Some authors have used alternative definitions: for example, Matten (2000, pp. 222 $ 223) defines economic capital as
risk capital plus goodwill; Perold (2001) defines risk capital in terms of insurance (explained in Section III.0.2.6).
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
16/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 16
to market value measures (notwithstanding the fact that, in some cases, accounting practice
allows balance sheet items to be reported on a market value basis).
To capture the discrepancy between fair values and market values, regulatory capital measures
incorporate ad hoc approaches to normalise asset book values to reflect differences in risk. As
such, it is recognised that regulatory capital calculations tend to contain a number of
inconsistencies, which have led regulators to set prescribed levels on a conservative basis. In
some cases, these inconsistencies have led to the notion of regulatory arbitrage , whereby investment
is determined not on the basis of risk $ reward optimisation but on the basis of regulatory capital $
reward optimisation.
III.0.1.3 Capital as a Management Tool
Capital can be used as a powerful business management tool, since it provides a consistent metric
to determine:
risk aggregation;
performance measurement;
asset and business allocation.
The objective of risk-adjusted performance measurement (RAPM) is to define a consistent metric
that spans all asset and risk classes, thereby providing an #apples to apples benchmark for
evaluating the performance of alternative business opportunities. RAPM thus becomes an ideal
tool for capital allocation purposes. By allocating the appropriate amount of EC to each asset, net
expected payoffs can then be expressed as returns on capital. Each asset can, therefore, be
assessed on a consistent basis, with returns adjusted appropriately in the context of the amount
of risk taken on. (This is further discussed in Sections III.0.4 and III.0.5).
Risk aggregation generally refers to the development of quantitative risk measures that
incorporate multiple sources of risk. The most common approach is to estimate the EC that is
necessary to absorb potential losses associated with each of the risks. EC can be seen as acommon measure that can be used to summarise and compare the different risks incurred by a
firm, across
different businesses and activities;
different types of risk $ market risk, credit risk and operational risk.
According to a recent study (BCBS, 2003b), the application of risk aggregation and EC methods
is still in the early stages of its evolution. While some firms remain sceptical of the value of
reducing all risks to a single number, many now believe that there is a need for a common metric
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
17/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 17
that allows risk $ return comparisons to be made systematically across business activities whose
mix of risks may be quite different (e.g., insurance versus trading). However, there remains a wide
variation in the manner in which aggregated risk measures such as EC are used for risk
management decision making in practice today.
III.0.2 Economic Capital
Economic capital acts as a buffer that provides protection against all the credit, market,
operational and business risks faced by an institution. EC is set at a confidence level that is less
than 100% (e.g., 99.9%), since it would be too costly to operate at the 100% level. The
confidence interval is chosen as a trade-off between providing high returns on capital for
shareholders and providing protection to the debt holders (and achieving a desired rating) as wellas confidence to other claim holders, such as depositors.
In so far as EC reflects the amount of capital required to maintain a firms target capital rating,
the confidence interval can be defined at a very high quantile of the loss distribution. For
example, to achieve a target S&P credit rating of BB, the probability of default over the next year
for the firm cannot be greater than 3.0%, so the quantile should be set at least at 97%. In
contrast, for the same firm to achieve a target S&P credit rating of BBB, it must lower its
probability of default to be at most 0.5%, corresponding to the 99.5% quantile of the loss
distribution. Given the desire to achieve a BBB rating and to remain solvent 99.5% of the time, a
firm must have enough capital to sustain a #0.5% worst-case loss over a one-year time horizon;
that is, 99.5% of the time the future value of the non-defaulted assets must be at least equal to
the future value of the liabilities. Example III.0.1 below gives a simple outline of how this could
be achieved.
III.0.2.1 Understanding Economic Capital
Denote by At and D t the market values (at time t ) of the assets and liabilities, respectively. The
available capital C t for the current time, t = 0 ,and at the end of one year , t = 1, can be expressedas
C 0 = A0 $ D 0, (III.0.1)
C 1 = A1 $ D 1.
If the nominal returns on the assets and liabilities are equal to r A and r D , respectively, then a
worst-case loss from all sources, l (i.e., when C 1 = 0 for a given confidence interval), would result
in the value of assets at t = 1 just being sufficient to cover the value of debt in t = 1. Then
C 1 = 0 = A0 (1 + r A )(1 l ) D 0(1 + r D ).
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
18/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 18
Thus the maximum amount of debt allowable to sustain solvency under the worst-case scenario
cannot exceed
D 0 = A0 (1 + r A )(1 l )/(1 + r D ). (III.0.2)
Since EC 0 is the minimum amount of capital required to sustain such a loss, it is given by:
EC 0 = A0 (1 [(1 + r A )(1 l )/(1 + r D )]). (III.0.3)
Hence the minimum amount of EC a financial institution must take on in order to avoid
insolvency increases as the level of the worse-case loss l increases.
The expected return on EC over the period from t = 0 to t = 1 is given by
[E( EC 1 ) / EC 0 ] 1.
The expected return on EC reflects the impact of leverage on risk and reward. An increase in
expected returns (to compensate for increased risk) is reflected in the numerator, while the
increase in risk is reflected in the denominator (the current EC).
Equation (III.0.3) is often expressed in terms of value-at-risk notation in the following manner:
EC 0 = A0 $ VaR/( 1+r D ), (III.0.4)
where VaR represents the Ai value associated with the worst-case loss, l , corresponding to the
appropriate ( x% ) confidence interval.
For ease of presentation, consider the case where credit risk is the sole source of business risk to
which the firm is exposed. Returning to equation (III.0.3), under the simplifying assumption that
the spread between the nominal return on the assets and the return on the liabilities is roughly
equal to the expected default loss, u , then,
EC 0 = A0 {1 $ (1 + r D )(1 + u )(1 $ l )/(1 + r D )} (III.0.5)
= A0 {1 $ (1 + u )(1 $ l )}.
By then ignoring second-order effects, equation (III.0.5) simplifies to the following more familiar
expression for economic capital:
0 0 ( ) EC A l u . (III.0.6)
This relationship is illustrated with respect to a default loss distribution in Figure III.0.1.
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
19/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 19
Credit Loss Distribution
Loss
Probability
0
Expected Loss
VaR( %)
Volatility
Unexpected Loss
Figure III.0.1: Credit loss distribution: expected and unexpected losses
Expressions (III.0.4) and (III.0.6) highlight the link between VaR measures and EC. The
simplifying assumption leading to equation (III.0.6) and illustrated in Figure III.0.1 is the
approach commonly taken by practitioners and generally leads to conservative estimates (for a
detailed discussion, see Kupiec, 2002). Thus, in its most common definition, EC is defined to
absorb only unexpected losses (UL) up to a certain confidence level (i.e., A0( l u )). Credit reserves are
traditionally set aside to absorb expected losses (EL) over the period (i.e., A0u ). More precisely,
equation (III.0.4) shows that the VaR measure appropriate for EC should in fact measure losses
relative to the assets initial mark-to-market (MtM) value and not relative to the EL in its end-of-period distribution. Also, the VaR measure should explicitly account for the interest payments on
the funding debt. While the UL approximation has very little effect on market risk, where the
horizon is short (and EL is small) it may have a higher impact in credit risk.
Example III.0.1
Consider a BBB-rated firm (or a firm that has targeted a BBB rating). Suppose the firm has
liabilities consisting of D 0 = $92 million in deposits, with a cost of debt of r D = 5%, which have
been invested in A0 = $100 million of assets (40% at a nominal return of 6.75% and 60% at a
nominal return of 7%.). The weighted average nominal return across the $100 million in total
assets is r A = 6.9%, representing a compounded spread of 1.81%. If the nominal values of the
assets and liabilities are equal to the market values, then the current capital for this firm is
calculated as C 0 = $8 million (the difference between the market value of the assets and the
market value of the liabilities).
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
20/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 20
Assume that, in a #0.5% worse-case scenario the firm has a potential for a loss of 15% in the
value of total assets. Under this scenario, the firm will become insolvent as the value of the assets
will be A1 = $100 million ? 1.069 ? 0.85 = $90.9 million, while the value of the liabilities will be
D 1 = $92 million ? 1.05 = $96.6 million, giving a capital shortfall of $5.7 million.
From equation (III.0.3), the minimum amount of capital the firm must hold to avoid insolvency
in the worst-case scenario is
EC 0 = A0 (1 [(1 + r A )(1 l )/(1 + r D )]) = 100(1 $ [1.069(1 0.15)/1.05]) = 13.46.
Therefore, for the firm to improve its capital adequacy to the desired level, it must increase its
capital from $8 million to $13.46 million. The shareholders should be 99.5% sure that such an
increase in capital will ensure solvency from t = 0 to t = 1.
III.0.2.2 The Top-Down Approach to Calculating Economic Capital
EC can be seen as a common measure that can be used to summarise and compare the different
risks incurred by a firm, across different businesses and activities, and across different types of
risk: market, credit and operational risk. At the enterprise level, EC can be estimated based on
aggregate information of the firms performance. Such #top-down approaches generally use one
of two types of information: earnings or stock prices.
III.0.2.2.1 Top-Down Earnings Volatility Approach
A top-down approach based on a firms earnings makes the simplifying assumption that the
market value of capital is equal to the value of a perpetual stream of expected earnings. In other
words, by assuming that all expected future earnings of the firm are equal to the next periods
expected earnings, the value of capital can be expressed as
C 0 = Expected earnings / k,
where k represents the required return associated with the riskiness of the earnings. As the
determination of EC is based on the ability to sustain a worst-case loss associated with a given
confidence interval,
EC 0 = EaR / k,
where EaR represents the difference between expected earnings and the earnings under the
worst-case scenario, for a given confidence interval (see Saita, 2003). Often this approach relies
on the additional assumption that earnings are normally distributed, and thus the confidence
interval can be determined as a multiple of the standard deviation.
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
21/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 21
Limitations of the earnings volatility approach include the following:
It requires historical performance data for reliable estimates of the mean and standard
deviation of earnings; few companies have enough data to yield reliable estimates.
It does not link EC directly to the sources of risk.
In general, it does not naturally allow capital to be separated out into its market, credit
and operational risk components, nor across different business lines or activities.
III.0.2.2.2 Top-Down Option-Theoretic Approach
A top-down approach based on the Black $ Scholes $ Merton (BSM) framework (see Chapter
III.B.5) assumes that the market value of capital can be modelled as a call option on the value of
the firms assets where the strike price is the notional value of the debt. If the value of assets at
the end of the period ( t = 1) is greater than the value of the debt, then the value of capital is equalto the difference between the value of the assets and the debt; otherwise (in the case of
insolvency), it is equal to zero. Using this approach assumes we have the following information
available:
the current market value and volatility of the company s net assets;
the time horizon (e.g., the average duration of the firms assets);
the risk-free interest rate (maturity corresponding to the time horizon);
the default threshold (the asset level at which the debt holders demand repayment and
bankruptcy can occur).
The BSM model allows us to estimate the implied probability of insolvency for the firm over the
period from t = 0 to t = 1. The EC can then be determined on the basis of reapplying the BSM
model for a level of debt that ensures, even under the worst-case scenario (at a given confidence
interval), that the firm remains solvent.
An advantage of this approach over the one based on EaR is the availability of stock market data.
However, several simplifications regarding the capital structure and model assumptions must be
made to apply this tool in practice. Similar to the EaR approach, a key limitation is that it does
not allow the separation of capital into different risks such as market, credit and operational risks,
nor does it suggest how to allocate it across different business lines or activities.
III.0.2.3 The Bottom-Up Approach to Calculating Economic Capital
In this approach, EC is estimated by modelling individual transactions and businesses and then
aggregating the risks using advanced statistical portfolio models and stress testing. The bottom-
up approach has now become best practice and, in contrast to the top-down approaches,
provides greater transparency with regard to isolating credit risk, market risk and operational risk
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
22/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 22
capital. Furthermore, it naturally accommodates various methodologies to allocate capital to
individual businesses, activities and transactions.
In a bottom-up approach, the estimation of enterprise EC requires consolidation of risks at two
levels:
First, it computes market risk, credit risk and operational risk at the enterprise level. To
achieve this, a firm might use an internal VaR model for market risk, a credit VaR
methodology for credit risk and a loss-distribution approach for operational risk.
Then, at the second level, the firm must consolidate the capital across these risks.
To estimate total capital, it is currently common practice to add up the credit risk capital, market
risk capital and operational risk capital. This produces a conservative capital measure (basicallyassuming that the risks are perfectly positively correlated). However, today, many firms are now
devoting considerable effort to measuring the correlations between these risks (and hence the
levels of diversification), as well as developing frameworks to measure these risks in a more
integrated way.
III.0.2.4 Stress Testing of Portfolio Losses and Economic Capital
In addition to the statistical approaches inherent in credit portfolio models, practitioners usually
use stress testing as an important part of their EC methodology (see Chapter III.A.4).
Commonly, the stress-testing methodology involves the development of one or several specific
adverse scenarios, which are judged to be extreme (falling beyond the desired confidence level).
Current portfolio losses are then assessed against these specific scenarios. Stress scenarios may be
based on historical experience or management judgment.
The translation of the specific stress scenario losses, and the combination of stress testing and
statistical measures, to develop EC measures is today more an art than a science, largely based on
managements objectives and judgement. In essence, firms make their own decision on the
relative weights of the statistical and stress test results in estimating the amount of EC required tosupport a portfolio. For example, an institution might assign the EC for market risk as 50% times
the 99% VaR plus 50% the loss outcome from some stress scenarios (thus normally being higher
than the actual 99% VaR).
III.0.2.5 Enterprise Capital Practices Aggregation
A large firm, such as a bank, acquires different types of financial risk through various businesses
and activities. Capital is indeed a powerful tool for understanding, comparing and aggregating
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
23/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 23
different types of risk to determine the overall health of the firm and to support better business
decisions.
Such an institution is likely to have separate methodologies to measure market risk, credit risk
and operational risk. In addition, it is likely that the institution has different methodologies to
measure the credit risk of its larger commercial loans or its retail credits. More generally, a firm
may have any number of methodologies for various risks and segments. If each type of risk is
modelled separately, then the amounts of EC estimated for each need to be combined to obtain
an enterprise capital amount. In making the combination, the firm needs to incorporate, either
implicitly or explicitly, various correlation assumptions. The methods of aggregation are as
follows:
Sum of stand-alone capital for each business unit and type of risk. This methodologyessentially assumes perfect correlation across business lines and risk types and does not
allow for diversification from them.
Ad hoc or top-down estimates of cross-business and cross-risk correlation. In order to
allow for some cross-business and cross-risk diversification, a firm might aggregate the
individual stand-alone capital estimates using analytical models and simple cross-business
(asset) correlation estimates.
The enterprise aggregation of capital is still in its infancy and is a topic of much research today.
III.0.2.6 Economic Capital as Insurance for the Value of the Firm
Standard practice is to define EC as a buffer to cover unexpected losses. Thus it is defined in
terms of the tail of the loss distribution, using measures such as VaR. Alternatively, some
economists have used the term #risk capital to define capital in economic terms (Merton and
Perold, 1993; Perold, 2001): risk capital is the smallest amount that can be invested to insure the
value of the firms net assets4 against loss in value relative to a risk-free investment.
For example, under this definition, the risk capital of a long US Treasury bond position is the
value of a put option with strike equal to the forward price of the bond. As pointed out by Perold(2001), in general, the put option accounts for the full distribution of losses, whereas VaR ignores
the magnitude of outcomes conditional on being in the extreme tail of the distribution. When
returns are normally distributed, the value of such a put option is approximately proportional to
the standard deviation of the return on the bond, and thus is approximately proportional to VaR.
4 #Net assets refers to #gross assets minus customer liabilities (swaps, insurance contracts, etc.), valued as if these
liabilities are default-free.
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
24/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 24
III.0.3 Regulatory Capital
This section considers the key concepts and objectives behind regulatory capital, as well as the
main principles used in regulatory capital calculations in the Basel I Accord and the latest
proposals of the current Basel II Accord.
III.0.3.1 Regulatory Capital Principles
In this subsection, we focus mainly on the capital regulation in the banking industry. As defined
in Section III.0.1, regulatory capital refers to the capital that an institution is required to hold by
regulators in order to operate. It is largely an accounting measure defined by the regulatory
authorities to act as a proxy for economic capital. Capital adequacy is generally the single most
important financial measure used by banking supervisors when examining the financial
soundness of an institution.
As also mentioned in Section III.0.1, from an internal bank perspective, capital is designed as a
buffer to absorb large unexpected losses, protect depositors and other claim holders, and provide
enough confidence to external investors and rating agencies on the financial health and viability
of the firm. In contrast, from the external perspective of the regulator, capital adequacy
requirements fulfil two objectives:
Reducing systemic risk: to safeguard the security of the banking system and ensure its
ongoing viability. In a sense, national governments act as guarantors. They have an
interest in ensuring that banks remain capable of meeting their obligations and in
minimising potential systemic effects on the economy. Regulatory capital helps to ensure
that banks bear their share of the burden, otherwise borne by national governments.
Creating a level playing field: to ensure a more even playing field for internationally active
banks, by submitting all banks to (roughly) the same rules.
As we review the key concepts in regulatory capital, it is important to highlight two points:
Regulatory requirements are continuously changing, and it is vital for practitioners to be
familiar with both the latest regulations and the specific requirements in each
jurisdiction.
While the general intention is to make regulatory capital more risk-sensitive and align it
more closely to economic capital, it is important to understand the limitations of using it
directly for managing risk, measuring performance and pricing credits.
The overall objective should be to set up an enterprise risk management framework, which
measures economic capital and reconciles it with regulatory capital.
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
25/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 25
III.0.3.2 The Basel Committee of Banking Supervision and the Basel Accord
A key cornerstone of international banking capital regulation is the Basel Committee on Banking
Supervision, which first introduced the framework for international capital adequacy standards.
This framework has been adopted as the underlying structure of all bank capital adequacy
regulations throughout the G10, as well as many other countries around the world. Today, over
100 countries are expected to implement the latest guidelines set by the BCBS, also referred to as
the Basel II accord).
Summary Chronology of Banking Regulatory Capital
1988 $ The BCBS introduces the framework for international capital adequacy standards.
It is adopted throughout the G10, as well as in over 100 other countries (BCBS, 1988).
Commonly referred to as the Basel I Accord or the BIS I Accord , it was the first step inestablishing a level playing field across member countries for internationally active banks.
The 1988 accord focused mainly on credit risk.
1995 $ An amendment to the initial accord further allows banks to reduce #credit-
equivalent exposures when netting agreements are in place (BCBS, 1995).
1996 $ The 1996 amendment5 extends the capital requirements to include risk-based
capital for the market risk in the trading book (BCBS, 1996).
1999 $ The BCBS issues a proposal for a new capital adequacy framework to replace the
1988 Basel I Accord. This is commonly referred to as the Basel II Accord or BIS II Accord .
The new accord attempts to improve the capital adequacy framework by substantially
increasing the risk sensitivity of the minimum capital requirements, and also
encompassing a supervisory review and market discipline principles. Under the
proposal, banks are required specifically to allocate capital against operational risks for
the first time.
2000 $ 2003 $ The BCBS releases various consultation documents and conducts major
data collection exercises called quantitative impact studies (QIS), intended to gather
information to assess whether it has met its goals.
April 2003 $ The BCBS releases the third consultative paper (CP3) on the new Basel
Accord (BCBS, 2003a).
5 Sometimes referred to as BIS 98 , after its date of implementation.
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
26/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 26
June 2004 $ The final version of the Basel II Accord is published (BCBS, 2004).
2006 $ 2007 $ Currently scheduled implementation of Basel II.
All the papers from the BCBS can be downloaded from www.bis.org .
III.0.3.3 Basel I Regulation
The 1988 accord focused mainly on credit risk, establishing minimum capital standards that
linked capital requirements to the credit exposures of banks. Prior to its implementation in 1992,
bank capital was regulated through simple, ad hoc capital standards. While generally prescriptive,
Basel I left various choices to be made by local regulators, thus resulting in several variations of
the implementation across jurisdictions. The 1996 amendment further extended the capitalrequirements to include risk-based capital for the market risk in the trading book. Basel I does
not cover capital charges for operational risk.
III.0.3.3.1 Minimum Capital Requirements under Basel I
Capital requirements under Basel I are the sum of:
credit risk capital charge, which applies to all positions in the trading and banking books
(including OTC derivatives and balance sheet commitments);
market risk capital charge for the trading book portfolio and off-balance sheet items.
For market risk capital, the accord allows, in addition to a standardised method, the use of
internal VaR models covering both general market risk (or systemic risk) and specific risk.
Specific VaR applies to both equities and bonds. For bonds it covers the risk of defaults,
migration and changes in spreads. The reader is referred to Chapter III.A.2 for the basics of
market risk VaR.
The regulatory charge for banks using internal market risk models is given by
[ ]8
MR SR Trigger Market Risk Capital M VaR M SpecificVaR , (III.0.7)
where VaR and Specific VaR denote, respectively, the 99% market VaR and specific VaR over a
10-day horizon, and M MR and M SR are multipliers designed to adjust the capital to cover for
modelling errors and reward the quality of the models. The first one ranges between 3 and 4, and
the second one between 4 and 5. Finally, the Trigger is related to quality of controls in the bank.
Currently it is set to 8 in North America and between 8 and 25 in the UK.
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.bis.org/http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/http://www.bis.org/
-
8/9/2019 Prm Handbook III
27/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 27
The methodology for credit capital is simple. Minimum capital requirements are obtained by
multiplying the sum of all the risk-weighted assets by the capital adequacy ratio of 8% (also
referred to as the Cook ratio ):
8%kk
Capital RWA . (III.0.8)
Thus the calculation of credit regulatory requirements has three steps: converting exposures to
credit equivalent assets; computing loan equivalents for off-balance sheet and OTC portfolios;
and applying the capital adequacy ratio. This is described in greater detail in Chapter III.B.6.
A great strength of Basel I is the simplicity of the framework. This has allowed it to be
implemented in countries with different banking and accounting practices. Thus, it has been quite
successful in achieving its two general objectives (to safeguard the stability of the banking system
and to ensure an level playing field internationally).
Its simplicity also has been its major weakness, as the accord does not effectively align regulatory
capital requirements closely with an institutions risk. For example, some criticisms on the credit
risk capital include the lack of proper differentiation for credit quality and maturity, insufficient
incentives for credit mitigations techniques, and lack of recognition of portfolio effects (these are
discussed briefly in Chapter III.B.6).
III.0.3.3.2 Regulatory Arbitrage under Basel I
The lack of differentiation in the accord, together with the financial engineering advances in
credit risk over the last decade, have lead to the development of a regulatory capital arbitrage
industry. This refers to the process by which regulatory capital is reduced through instruments
such as credit derivatives or securitisation, without an equivalent reduction of the actual risk
being taken. Through regulatory arbitrage instruments, for example, banks typically transfer low-
risk exposures from their banking book to their trading book, or simply place them outside the
regulated banking system.
III.0.3.3.3 Meeting Capital Adequacy Requirements
Available regulatory credit capital is divided into two categories:
Tier 1 capital: essentially shareholder funds $ equity $ and retained earnings.
Tier 2 capital: long-term subordinated debt, other qualifying hybrid instruments and
reserves (such as loan loss reserves).
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
28/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 28
From a regulatory perspective, Tier 1 capital must cover at least 50% of the total capital; that is,
Tier 2 cannot exceed Tier 1 capital. In addition, the subordinated debt included in Tier 2 cannot
exceed 50% of the Tier 1 capital.6
Capital adequacy is generally expressed as a ratio. For example, an 8% capital ratio means that the
total Tier 1 and Tier 2 capital is 8% of the risk-weighted assets (RWA). A 6% Tier 1 capital ratio
refers to Tier 1 capital being 6% of the RWA.
III.0.3.4 Basel II Accord Latest Proposals
The final version of the Basel II Accord was published in June 2004 (BCBS, 2004).7 Its
implementation will take effect between the end of 2006 and the end of 2007.
In this subsection we present a brief summary of the basic principles of Basel II; the key
formulae for minimum capital requirements for credit risk are given in Chapter III.B.6. For
greater detail, the reader is referred to the BCBS papers.
Basel II attempts to improve capital adequacy framework along two important dimensions:
First, the development of a capital regulation that encompasses not only minimum
capital requirements, but also supervisory review and market discipline.
Second, a substantial increase in the risk sensitivity of the minimum capital requirements.
The new accord intends to foster a strong emphasis on risk management and to encourage
ongoing improvements in banks risk assessment capabilities. This is to be accomplished by
closely aligning banks capital requirements with prevailing modern risk management practices,
and by ensuring that this emphasis on risk makes its way into supervisory practices and into
market discipline through enhanced risk- and capital-related disclosures.
The Basel II Accord consists of three pillars: minimum capital requirements , supervisory review , and
market discipline . We briefly summarise these below and then present the key principles behind thecomputation of minimum capital requirements.
III.0.3.4.1 Pillar 1 - Minimum Capital Requirements
Minimum capital requirements consist of three components:
1. definition of capital (no major changes from the 1988 accord);
6 A Tier 3 capital was introduced with the market risk requirements. Short-term subordinated debt can be used to meetmarket risk requirements as well, but not credit risk.
7 A small number of open issues are still to be resolved during 2004.
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III
29/415
The PRM Handbook Volume III
Copyright ? 2004 The Authors and the Professional Risk Manager ! s International Association 29
2. definition of RWA;
3. minimum ratio of capital/RWA (remains 8%).
Basel II proposes to modify the definition of risk-weighted assets in two areas:
substantive changes to the treatment of credit risk relative to the Basel I Accord;
the introduction of an explicit treatment of operational risk that will result in a measure
of operational risk being included in the denominator of a bank s capital ratio.
Basel II moves away from a one-size-fits-all approach to the measurement of risk, through the
introduction of three distinct options for the calculation of credit risk and three others for
operational risk. These approaches present increasing complexity and risk-sensitivity. Banks and
supervisors can thus select the approaches that are most appropriate to the stage of developmentof banks operations and of the financial market infrastructure. Chapter III.B.6 briefly reviews the
three credit risk approaches. The operational risk approaches can be found in Chapter III.C.3.
III.0.3.4.2 Pillar 2 - Supervisory Review
The second pillar is based on a series of guiding principles, which point to the need for banks to
assess their capital adequacy positions relative to their overall risks, and for supervisors to review
and take appropriate actions in response to those assessments. Banks under internal ratings-based
credit models will be required to demonstrate that they use the outputs of those models not only
for minimum capital requirements but also to manage their business. The inclusion of
supervisory review provides benefits through its emphasis on strong risk assessment capabilities
by banks and supervisors alike. Important new components of Pillar II also include the treatment
of stress testing, concentration risk and the residual risks arising from the use of collateral,
guarantees and credit derivatives as well as specific securitisation exposures (these are discussed
further in Chapter III.B.6).
III.0.3.4.3 Pillar 3 - Market Discipline
Also referred to as public disclosure , the third pillar aims to encourage safe and sound bankingpractices through effective market disclosures of capital levels and risk exposures. This will help
market participants assess better a bank s ability to remain solvent.
III.0.3.5 A Simple Derivation of Regulatory Capital
Recognising that the simple difference between the value of assets and the value of liabilities in
accounting value terms is not a good indicator of the true difference in market value terms has
led regulators to make appropriate adjustments in the calculation of regulatory capital. In this
section, we follow a similar approach to Section III.0.2.1 to understand regulatory capital.
PDF created with pdfFactory Pro trial version www.pdffactory.com
http://www.pdffactory.com/http://www.pdffactory.com/http://www.pdffactory.com/
-
8/9/2019 Prm Handbook III