FINANCIAL RISK MANAGEMENT FOR
ISLAMIC BANKING AND FINANCE
Palgrave Macmillan Finance and Capital Markets Series
For information about other titles in this series please visit the website:
http://www.palgrave.com/business/financeandcapitalmarkets.asp
Financial Risk
Management for
Islamic Banking and
Finance
IOANNIS AKKIZIDIS
AND
SUNIL KUMAR KHANDELWAL
© Dr. Ioannis Akkizidis and Dr. Sunil Kumar Khandelwal 2008
First Foreword © Agil Natt 2008
Second Foreword © Amgad S. Younes 2008
Softcover reprint of the hardcover 1st edition 2008
All rights reserved. No reproduction, copy or transmission of this
publication may be made without written permission.
No paragraph of this publication may be reproduced, copied or transmitted
save with written permission or in accordance with the provisions of the
Copyright, Designs and Patents Act 1988, or under the terms of any licence
permitting limited copying issued by the Copyright Licensing Agency, 90
Tottenham Court Road, London W1T 4LP.
Any person who does any unauthorised act in relation to this publication
may be liable to criminal prosecution and civil claims for damages.
The authors have asserted their rights to be identified as
the authors of this work in accordance with the Copyright,
Designs and Patents Act 1988.
First published in 2008 by
PALGRAVE MACMILLAN
Houndmills, Basingstoke, Hampshire RG21 6XS and
175 Fifth Avenue, New York, N.Y. 10010
Companies and representatives throughout the world.
PALGRAVE MACMILLAN is the global academic imprint of the Palgrave
Macmillan division of St. Martin’s Press, LLC and of Palgrave Macmillan Ltd.
Macmillan® is a registered trademark in the United States, United Kingdom
and other countries. Palgrave is a registered trademark in the European
Union and other countries.
ISBN 978-1-349-36366-7 ISBN 978-0-230-59875-1 (eBook)
DOI 10.1007/978-0-230-59875-1
This book is printed on paper suitable for recycling and made from fully
managed and sustained forest sources. Logging, pulping and manufacturing
processes are expected to conform to the environmental regulations of
the country of origin.
A catalogue record for this book is available from the British Library.
A catalog record for this book is available from the Library of Congress.
10 9 8 7 6 5 4 3 2 1
17 16 15 14 13 12 11 10 09 08
Transferred to Digital Printing in 2011
Other books by Ioannis Akkizidis
1. Guide to Optimal Operational Risk & Basel-II, 2006, Auerbach Publications / Taylor &
Francis, Boca Raton / New York, ISBN: 0849338131.
2. Integrating Market, Credit and Operational Risk: A Complete Guide for Bankers & Risk
Professionals, 2006, Risk Books, London, ISBN: 1904339964.
Apart from its manuscript, the first book also includes a practical operational risk software
tool for designing and evaluating an operational risk management framework based on
advanced measurement approaches.
The second book is a best seller. Its Chapter 3 ‘Operational Risk’ and Chapter 4 ‘Extreme
Value Theory in Risk Management’ are listed from the GARP Association as recommended
material for 2007 FRM Study Guide Readings. For more information please visit:
https://www.garpdigitallibrary.org/display/booktitle.asp?btid=161
This book is dedicated to:
‘My Father’
(Dr. Ioannis Akkizidis)
‘My Larger Family’
(Dr. Sunil Kumar)
CHAPTER
Contents
List of Tables x
List of Figures xii
Foreword by Agil Natt xiv
Foreword by Amgad S. Younes xvi
Preface xviii
About the Authors xxiv
Acknowledgements xxvi
Abbreviations xxviii
1 Principles of Islamic Finance 1
Introduction 1
2 Risk Management Issues in Islamic Financial Contracts 28
Introduction 28
An overview of financial risks 30
Identifying risks in Islamic finance 36
Risk of non-compliance with Shariah rules 41
Main elements used in financial risk analysis 41
Mushãrakah contracts of partnership and financial risks 42
Mudãrabah contracts of partnership 49
Murãbaha contract agreements and financial risk 54
Salam contract agreements and financial risk 58
Istisnã contracts and financial risks 63
Ijãrah contract agreement and financial risk 68
vii
viii CONTENTS
3 Basel II and IFSB for Islamic Financial Risk 80
Introduction 80
The Basel accord 82
Risk management according to Basel II and the Islamic
financial industry 83
The three pillars of Basel II 84
Pillar 2: Supervisory review 96
Pillar 3: Market discipline 98
IFSB principles of risk management 99
4 Credit Risks in Islamic Finance 109
Introduction 109
Credit risk exposure identification 110
Credit risk assessment models 113
Credit risk valuation 118
Credit risk mitigation 133
Credit rating systems 134
Validating the credit rating systems 139
5 Market Risks in Islamic Finance 148
Introduction 148
Identification of market risk factors 149
Rate of return risk 149
Commodity risk in Islamic finance 154
FX rate risks 159
Equity price risks 160
Valuation issues on equity prices and FX rates 160
Quantification of foreign
exchange risk, equity risk, and commodity risk 161
Data referring to market risk factors 161
Sensitivity in market risk 163
Market risk valuation models 164
VaR models for Islamic financial contracts 164
Position and market data 166
Position risk and exposure risk 168
Evaluation methods of market risk 168
Variance–co-variance method 169
Monte Carlo simulation method 173
Historical simulation 174
Back-testing and stress-testing for market risk exposures 176
CONTENTS ix
6 Operational Risk in Islamic Finance 182
Introduction 182
Main elements in operational risk analysis 183
Identification of operational risk 185
Measuring operational risks 197
Loss events 199
Evaluating operational risk based on VaR analysis 203
Elements in the framework of the operational
risk management 206
7 Concluding Remarks 211
Index 217
List of Tables
1.1 Differences between conventional financing and
Islamic financing 3
1.2 Typical balance sheet of an Islamic bank 10
1.3 Differences between Mushãrakah and Mudãrabah 17
1.4 Top 10 Sukûk until February 2007 based on size 25
2.1 Risks in Islamic financial services 39
2.2 Comparative information requirements –
conventional vs Islamic banks 40
2.3 Pattern of risks where financial institutions are
exposed by applying Islamic financial contracts 73
3.1 Alternative approaches to credit, market, and
operational risk 87
3.2 Template for the calculation of the credit risk
exposures as per Pillar 1 under Basel II 91
3.3 Detailed loss event type classification 93
3.4 Beta factors for the different business lines 94
3.5 The IFSB guidelines on risk management 100
4.1 Different cases for estimating the probability of
default based on quantitative criteria 123
4.2 Classification of the levels of credit ratings with
the indication of the creditability and the reliability
as well as the meaning in terms of exposure 137
5.1 Factors that initiate commodity risk in Islamic
financial contracts 158
5.2 Advantages and drawbacks by applying Historic,
Monte Carlo, and Variance–Co-variance Methods 175
x
LIST OF TABLES xi
5.3 The three zone ‘traffic light’ for the overshooting
and the multiplier factor 177
6.1 Examples of types of operational causes, events and,
consequences 188
6.2 Sources of operational risks 190
List of Figures
1.1 Structure of Islamic financial service industry 4
2.1 Major types of risks in the financial industry 32
2.2 Information management in organisations 35
2.3 Risk profiling – conventional vs Islamic banks 38
2.4 Main elements considered in financial risk analysis 42
2.5 Structure of Mushãrakah partnership joint venture
contract agreements 43
2.6 Credit, operational, market, and liquidity risks during
the lifetime of Permanent Mushãrakah contracts 45
2.7 Credit, operational, market, and liquidity risks
during the lifetime of Diminishing Mushãrakah contracts 46
2.8 Ownership pattern and risk in Mushãrakah contracts 48
2.9 Structure of Mudãrabah contract agreements 49
2.10 Risks in Mudãrabah contract agreements during
the investment period 51
2.11 Risks in Mudãrabah contract agreements during
the business profitability and loss period 52
2.12 Murãbaha structure and its main interrelations 55
2.13 Credit, operational, and market risks during the
lifetime of a Murãbaha contract 56
2.14 Relation structure in Salam contract agreements 59
2.15 Operational, credit, market, and liquidity risks during
the lifetime of Salam contracts 61
2.16 Relation structure in Istisnã contract agreements 63
2.17 Operational, credit, market, and liquidity risks during
the lifetime of Istisnã contracts 66
2.18 Ijãrah contract agreements 69
2.19 Credit, operational, and market risks during the
lifetime of Ijãrah contracts 71
xii
LIST OF FIGURES xiii
3.1 The three pillars of Basel II accord 85
4.1 Relations between the Islamic financial counterparties
and contracts and the guaranties and collaterals 110
4.2 Representation of the links between a counterparty
with several contracts and risk coverage by
several guaranties and collaterals 112
4.3 Model layout of the expert system for credit
risk assessment 115
4.4 Distribution of expected and unexpected losses
in credit risk 128
4.5 Periods and samples used for the rating valuation 143
5.1 The four types of market risks that exist in the
different Islamic financial products 150
5.2 Interaction and relation between commodity
risk in Islamic finance with other commodity risk factors 155
5.3 Flow chart referring to the Value-at-Risk approach 165
5.4 Main components considered in VaR models 166
5.5 Normal distribution of the changes caused by the
value of a market analysis portfolio 169
6.1 Operational risk analysis based on operational risks
and losses 183
6.2 Main approaches in operational risk
identification analysis 184
6.3 Sources of operational risks and where they
affect Islamic financial contracts 194
6.4 Interactions among the different causes of
operational risks 197
6.5 The three approaches used to cultivate and
populate the loss event databases 202
6.6 Loss distribution for operational risk 203
Foreword by Agil Natt
I congratulate the authors Messrs Akkizidis and Khandelwal for writing this
book, Financial Risk Management for Islamic Banking and Finance. There
have been many books written on Islamic banking and finance, but few have
been written on risk management aspects, especially within the context of
financial products which are Shariah-compliant. I find the book compre-
hensive in its approach and detailed in information which will render it as a
valuable reference material for practitioners and students alike. The book
starts with an introduction of various products and services that are
currently being marketed and then discusses the various risk management
issues in Islamic financial contracts. It also discusses on the IFSB principles
of risk management and Basel II. The authors have also provided various
methodologies and logical steps involving mathematical models to identify
and measure risks in Islamic finance.
The publication of this book will be greatly welcomed by industry
players and students. Both authors are well versed in this subject and have
done extensive research to write this book, which covers all practical and
theoretical concepts of financial risk management pertinent to the Islamic
financial services industry. The growth in the Islamic financial services
industry has been phenomenal in the last two decades and has formed the
basis of a viable alternative system which in some aspects has rivalled its
conventional counterpart. Today, we see traditional financial capitals of
the West joining the race with the new entrants of the East in laying and
expanding the infrastructure for a robust Islamic financial system. Yet, the
number of knowledgeable Shariah financial practitioners is still trailing
behind. Against this background, institutions like INCEIF have mapped
out their global agenda of developing competent financial practitioners to
serve the global market. The challenge for the future will be to bridge the
gap between the theories propounded by academia and the practical
xiv
FOREWORD BY AGIL NATT xv
activities of the industry and ultimately, to further enhance Islamic
banking and finance into a prime mobiliser of resources for the betterment
of mankind.
I believe this book will be able to spur further thoughts and research with
the objective of adding to this new and exciting body of knowledge.
Agil Natt
President and Chief Executive
INCEIF – The Global University in Islamic Finance
Kuala Lumpur
Foreword
by Amgad S. Younes
The financial industry today has evolved itself to be one of the most
competitive, stringent, diverse, and aggressive environments to operate in.
Successful financial institutions are acquiring, consolidating, and increas-
ing market share, expanding product offerings and offering innovative
structures and services across multi-diverse portfolios. However, the
complexities of risk management threaten to limit growth. The Basel II
accord aggravates the challenges by mandating that all risk-related deci-
sions and processes be fully tracked and analysed to support enterprise risk
management requirements.
Traditionally, risk management was based on risks to financial assets and
people as a function, but risk management has evolved and become less
about managing the present and more about being proactive in tackling
scenarios and achieving business excellence.
With the Basel II accord, being transparent to all conventional financial
institutions, Islamic banks thrive for compliancy in the era of modern and
diversified financial management. With the majority of risks remaining the
same (Credit Risk, Market Risk, Liquidity Risk, Rate of Return Risk,
Operational Risk, etc.), it is crucial for Islamic banks to recognise and
evaluate the overlapping nature and transformation of risks that exist
between and amongst the categories of risks. Adverse changes in the
Islamic banking market, counterparties, or products, as well as changes in
the economic and political environments in which they operate and the
effects of different Shariah rulings, are examples of Business Risks; in this
regard, Islamic banks are expected to view the management of these risks
from a holistic perspective.
This book is a welcome attempt to help Islamic banking and finance
readers to broaden their horizons and enlighten their understanding towards
Risk Management with a new perspective. The objective remains simple:
xvi
FOREWORD BY AMGAD S. YOUNES xvii
‘Risk Management remains to be of crucial importance in today’s modern
conglomerates.’ I do believe that this book will become an invaluable
reference point to all those senior executives, risk managers and practition-
ers, researchers and students.
AMGAD S. YOUNES
Senior Vice-President
Head of Financial Control,
St. Planning and Total Risk Management
Abu Dhabi Islamic Bank
Abu Dhabi, United Arab Emirates
Preface
Nowadays, in European, American and most Westernised markets, financial
institutions such as Credit Swiss, Deutsche Bank, HSBC, the Islamic Bank
of Britain, etc., are offering more products and services for Islamic finance
(it is estimated that only Arab investors have more than US$800 billion on
deposit in overseas banks, much of which is secured in Swiss and European
banks). Moreover, a great number of financial institutions in GCC (Gulf
Co-operation Council) countries and Asia are managing funds of over
US$300 billion and are encouraged from their markets to provide Islamic
financial products. The Asia-Pacific region houses the most populous
Islamic nation on earth, approximately 800 million. As a result of the
growth market of Islamic finance, the demand for risk management for
Islamic financial products and services is becoming a very critical issue for
both Westernised and Islamic financial markets.
In addition to what Westernised (non-Islamic) finance is providing,
Islamic banks are providing specific financial products (contracts) which
make them perform as investors, instead of only as creditors. These
products are based on principles that are driven by Profit and Loss (P&L)
sharing policies. Both the structure of all Islamic financial contracts and
the P&L policies need to fully comply with Shariah (Islamic) laws.
Additionally, Islamic financial institutions and also Westernised institu-
tions that provide Islamic products (via Islamic windows) are on the road
to complying with the Basel II accord. Thus, they are facing the
challenge of developing internal models in order to quantify, supervise,
and manage all three major types of their financial risks; that is, credit,
market, and operational risks. Moreover, they need to provide adequate
solutions and define the associated minimum requirements for the super-
visory regime.
This book covers all risk management issues for all Islamic financial
products and services, including Mushãrakah, Mudãrabah, Murãbaha,
xviii
PREFACE xix
Salam, Ijãrah, and Istisnã, considering all of their unique characteristics
that are underlined by the Shariah principles and law.
Therefore, the book provides an overview of the systematic methodolog-
ical steps for evaluating and managing all types of financial risks for the
Islamic financial products. It describes all types of risks and addresses the
issue of how to identify both qualitatively and quantitatively their signifi-
cance levels. Moreover, the book shows how to define when the risks appear
within the lifetime of the different types of contracts. Measurement
techniques for quantitatively assessing all types of risks are also described
in this book.
Additionally, the book demonstrates the methodologies to model and
evaluate financially, using mathematical approaches, the risks in the con-
tracts referring to Islamic finance. The overall effects of risks including
their impact on the bank’s profit-and-loss and asset and liability on the bal-
ance sheets are also discussed. Hence, all risks that are initiated from the
fluctuations of the balance sheet and the sharing (PLS) and non-sharing
(P&L) profit-and-loss financial products is fully explained in this book.
Risk management for Islamic banking financial products and services is one
of the greatest challenges that many Westernised as well as Islamic banks
are facing and probably will continue to face in the future. Therefore, this
book is a valuable guide for those that are working on both conventional
(non-Islamic) and Islamic finance.
ORGANISATION OF THE BOOK
The book is divided into 7 chapters, consisting of different sections:
Chapter 1: Principles of Islamic Finance
Chapter 1 begins with an overview of Islamic financial principles. The
reasons for the advent of Islamic finance can be traced to social and ethical
dimensions of the financing. The role of Shariah and other classical Islamic
schools of thought in Islamic finance is discussed in this chapter.
Prohibition of Riba, avoidance of Gharar and Haram activities, and
payment of Zakat are important cornerstones of Islamic finance and are
introduced in chapter 1. Additionally, major differences between conven-
tional and Islamic finance are also presented. The chapter further covers sig-
nificant areas of Islamic finance, such as accounting standards, international
risk management framework, and the growth of the industry. Accounting in
Islamic finance is not the same as conventional finance and a glimpse at a
typical Balance Sheet of an Islamic bank reveals some of the differences.
Moreover, the chapter explains the treatment of assets and liabilities as
xx PREFACE
needed in Islamic accounting. One of the fundamental tenets of Islamic
finance is the participation in the business activity and avoidance of a pure
financial debt. Profit and Loss Sharing (PLS) contracts are a result of this.
The relevance of PLS contracts in Islamic finance is explained in this chap-
ter, along with the reasons for their limited use. Six most common types of
Islamic financial contracts are identified as relevant in the current state of
the progress of Islamic finance. Mushãrakah, Mudãrabah, Murãbaha,
Salam, Istisnã, and Ijãrah are the most commonly used Islamic financial
contracts and have been selected for further analysis in this chapter. The
chapter deals with Mushãrakah and Mudãrabah as investment contracts and
explains them in detail with the help of examples drawn from real-life activ-
ities of Islamic banks. Permanent and Diminishing Mushãrakah, along with
differences between Mushãrakah and Mudãrabah, are analysed in this
chapter. Furthermore, the chapter, with the help of supportive examples,
deals with the most popular Islamic contract form, Murãbaha. Further,
Salam and Istisnã as vehicles for forward sale are discussed and examples
are presented to demonstrate their applicability. Being very similar but not
the same, the two are differentiated on the basis of fundamental points.
Ijãrah, the Islamic leasing, is also dealt with in this chapter. Bonds in
Islamic finance, known as Sukûk, are a recent phenomenon. The market for
Sukûk has been growing rapidly, with several large Sukûk being issued
recently. This chapter also presents an analysis of Sukûk.
Chapter 2: Risk Management Issues in
Islamic Financial Contracts
In chapter 2, a brief overview of financial risks is presented, along with a
detailed risk profiling of Islamic financial contracts (Mushãrakah,
Mudãrabah, Murãbaha, Salam, Ijãrah, and Istisnã), for the existence of the
different types of financial risks; that is, credit, market, operational, liquid-
ity. Examples, with detailed graphical timeline analyses of the contract life,
are provided to support the reader’s understanding of the risks inherent
within the different financial contracts. An overview of risks in Islamic
finance is explained, along with a comparison with conventional finance. It
shows that all discussed Islamic financial contracts are exposed to opera-
tional risks. Likewise, credit risks appear in all contracts except in the
Salam and Istisnã contracts. Moreover, it demonstrates that all contracts
that are dealing with commodities that is, Murãbaha, Ijãrah, Salam, and
Istisnã are exposing the financial institutions to commodity price risk; on
the other hand, the Mushãrakah, Mudãrabah contracts, which are partner-
ship agreements, are initiating equity risks. Some of the Islamic financial
contracts are typical due to the changing relationship between the contract-
ing parties over the time period of the contract. Chapter 2 further discusses
PREFACE xxi
in detail, with the help of graphs, the several overlapping risks which can
arise out of single events; for example, in the case of Diminishing
Mushãrakah, an interruption of regular cash flows to the institution can
cause credit and liquidity risks. Furthermore, the chapter also presents risk
management strategies for the Mushãrakah, Mudãrabah, Murãbaha,
Salam, Istisnã, and Ijãrah in relation to credit risk, market risk, operational
risk, and the resulted liquidity risk. This chapter is a useful guide to identi-
fying for all the existing as well as future types of Islamic financial contracts
the different types of risks that may arise within their lifetime. This is
mainly based on the evaluation of the associated financial events linked to
the expected inflow-cash and outflow-cash flows.
Chapter 3: Basel II and IFSB for Islamic
Financial Risk
Chapter 3 provides all necessary information needed for banks that provide
Islamic financial products, to comply with the Basel II accord. Islamic
financial contracts are typified by the changing relationship between the
contracting parties during the lifetime of the contracts. This has a direct
bearing on the risk exposures. This chapter highlights the detailed analysis
of credit, market, and operational risks that has been given by the Basel II
accord. The three mutually-enforcing Pillars are the foundation of Basel II
and this chapter discusses the applicability of these pillars in Islamic finan-
cial industry. It highlights that Basel II is primarily for conventional banks
and thus is not fully relevant to Islamic financial institutions. However, in
line with any other industry guidelines, some of the principles of risk man-
agement as proposed in Basel II are applicable to Islamic financial industry
as well. The basic role of capital cannot be changed and thus the role of risk
management in any institution. Further, the chapter analyses the various
approaches presented under Pillar 1 along with their relevance to the
Islamic financial industry. Pillar 2 and Pillar 3 are supportive in nature and
hence their role in risk management is more at a supervisory level, which
has been presented in this chapter. The chapter also signifies the link
between the role of social responsibility of Islamic finance and market dis-
closure. The chapter further identifies that the IFSB is playing a key role in
the development of standards for risk management in the Islamic financial
industry. The recent release of standards for risk management is the first
systematic attempt at consolidating the efforts for bringing Islamic financial
risk management under one umbrella, and this is discussed. The six major
risks as identified by the IFSB for the Islamic financial industry credit,
market, operational, equity investment, liquidity, and rate of return are
explained in this chapter. Furthermore, the suggested treatment for these
according to the IFSB is also analysed.
xxii PREFACE
Chapter 4: Credit Risks in Islamic Finance
In chapter 4, the risks of credit defaults referring to counterparties,
collaterals, and guarantees are fully covered for all Islamic types of finan-
cial contracts. As discussed in this chapter, financial institutions are exposed
to credit risks that correspond to lending in the Murãbaha, leasing in the
Ijãrah, promises to deliver or to buy in Istisnã and Salam, and investments
failure in the Mushãrakah and Mudãrabah contracts. Furthermore, the
chapter identifies different methods and techniques for developing models
for credit risk that results from Islamic products. The chapter also highlights
credit value-at-risk (VaR) based on expected and unexpected losses. The
Credit VaR estimation is viewed as the economic capital to be held as a
buffer against unexpected losses. The chapter also discusses how financial
institutions are mitigating credit risks by employing collaterals or guarantees
granted by tier counterparties.
Chapter 5: Market Risks in Islamic Finance
Chapter 5 discusses how Islamic financial institutions are exposed to market
risk primarily through four types of risks, which are: rate of return (mark-
up) or benchmark rate risks related to market inflations and ‘interest rates’,
commodity price risks as they typically carry inventory items (predefined
prices), FX rate risks in the same way as conventional banks and equity
price risks mainly in regards to the equity financing through the profit and
loss sharing contract modes. Islamic financial products carry more than one
market risk factor; moreover, even a simple portfolio usually contains dif-
ferent types of contracts. This combination of risk factors and contracts
increases the complexity of collecting and combining the information
needed for market risk analysis. As discussed in this chapter, the sensitivity
to market risk in Islamic contracts reflects the relationship between the
cause from a financial risk factor and its adverse impact to the financial
institution’s earnings from these contracts. In financial analysis, the most
prominent techniques to valuate market risks are the value-at-risk (VaR),
and this analysis is outlined in this chapter.
Chapter 6: Operational Risks in Islamic Finance
In chapter 6, operational risk, one of the major topics in today’s financial
risk management, is discussed. The chapter presents all the key aspects of
operational risk management by giving guidelines on how to identify and
qualitatively map risks in operations within all the business lines as well as
how to transfer their qualitative attributes to quantitative measurement
indicators. It outlines some of the initial main elements of operational risk
CHAPTER TITLE xxiii
analysis, which includes the identification, mapping, assessment, and
measurement and evaluation. It particularly makes reference to Islamic
financial products and especially the compliance with the Shariah
principles and laws. Three different approaches for operational risk identi-
fication analysis are presented: the self-assessment analysis, the quantitative
operational risk indicator approach, and the operational risk loss approach.
The chapter further discusses the identification factors and operational risk
mapping. It particularly goes deeper into the identification of operational
risk causes, events, and consequences. It graphically illustrates the different
sources of operational risks with regard to the different Islamic financial
contracts. It talks about how the loss data collection is a primary approach
for the assessment of operational risks. The issue of IT security is particu-
larly highlighted. It also underlines that in some cases the operational risks
related to IT systems/technologies have a significant impact and can even
lead to credit and market risk.
Chapter 7: Concluding Remarks
Concluding remarks are provided in chapter 7. This chapter brings together
all the issues and ideas discussed throughout the book. A brief summary of
the important conclusions is also presented in this chapter. It discusses
future directions of financial risk management in Islamic Banking and
Finance.
References are given at the end of every chapter for those interested in
strengthening their knowledge beyond the material and scope of this book.
All referenced documents written by the Basel Committee on Banking
Supervision are available free of charge from their website
www.bis.org/bcbs/publ.htm and documents from the IFSB can be found on
http://www.ifsb.org
A list of acronyms is finally given to ease understanding of the terms
used throughout the book.
CHAPTER
About the Authors
Dr Ioannis Akkizidis is a Risk Management Consultant & Analyst at IRIS
Integrated Risk Management AG, Zürich, Switzerland. Ioannis is a main
author of the bestselling book entitled Integrating Market, Credit and
Operational Risk: A Complete Guide for Bankers & Risk Professionals
(2006), where two of its chapters are part of the recommended material for
the GARP’s 2007 FRM Study Guide. He is also the main author of the book
entitled Guide to Optimal Operational Risk & Basel II (2006). His first
degree is in engineering (in Athens, Greece), whereas his postgraduate
masters degree is in applied mathematical analysis and control systems
from the University of Portsmouth, England, UK. He holds a PhD in applied
mathematics and artificial intelligence obtained from the University of
Wales, UK. He has published several scientific papers in international
journals and presented at international conferences and events, giving talks
on the subject of Financial Risk Management including Islamic Finance. He
has work experience worldwide over many years in risk management and
his main interests lie in designing and implementing models as well as
bringing new ideas in the risk and financial management field for financial
institutions and large corporations.
Dr Sunil Kumar Khandelwal is Head of Risk Management – Middle East –
at IRIS integrated risk management ag, Dubai, United Arab Emirates. He is
mainly responsible for providing strategic direction to organisational activities.
He specialises in Basel II, Operational Risk and Islamic Finance. His interests
are also in e-banking, and e-commerce. He holds an MBA in Finance from
The University of Southern Queensland, Australia. He has a PhD in applied
IT (Banking & Finance) from The Indian Institute of Technology, Bombay.
He is providing research support to banks for risk management and Basel II.
He was the Chief Strategist and Architect of the e-services for the Ministry
of Health, UAE. He has delivered several talks, discussions, and speeches
xxiv
ABOUT THE AUTHORS xxv
on banking and finance. He has also chaired numerous sessions in round
tables and meetings on finance. He has several articles to his credit in
journals, newspapers, and magazines. He is an active member of GARP and
ISACA in UAE. He has been actively involved in several co-operative
activities of the financial industry in UAE.
Contacts
The authors will be pleased to have feedback and are open to any discussion
referring to the subject and material presented in this book. Please contact
them via e-mail and/or postal address on:
Dr Ioannis Akkizidis
ioannis.akkizidis@iris.ch
ioannis.akkizidis@irisunified.com
IRIS integrated risk management ag
Bederstrasse 1, CH-8027 Zürich, Switzerland
Dr Sunil Kumar
sunil.kumar@irisunified.com
IRIS integrated risk management ag
Dubai, United Arab Emirates.
Acknowledgements
The authors would like to specially thank all those who gave them support
in different ways for the completion of this book. Ioannis Akkizidis would
like to thank the members of IRIS Integrated Risk Management AG for the
valuable participation in discussions, feedback, and bringing ideas for
improving the quality of this book; special thanks to Mr Jäk Andreas for
encouraging him to write this book and for his valuable comments on its
text. Thanks also to Ms Kathy Cleemput for her useful feedback on the text
of the book and Mr Nikolaj Tsenov for putting the idea in my mind for writ-
ing a book on this topic. Many and special thanks go to Dr Vivianne
Bouchereau for the proof-reading of the manuscript and also for her small
contributions in the material of this book. Her work and contribution played
a key role in the quality of this book. Also, many thanks go to Mr Nikos
Akkizidis for his helpful ideas and discussions in many financial and risk
management aspects; his knowledge and ideas are always more than
valuable.
Sunil Kumar is thankful to Ioannis Akkizidis for taking up and leading
the tasks of this wonderful book. He is thankful to Prof. L. M. Bhole,
Professor Emeritus at The Indian Institute of Technology for going through
the chapters carefully and suggesting improvements. Special thanks to Ms
Sujata for maintaining the time schedule and ascertaining the scheduled
delivery of the contents of the book. Special thanks to Dr. Mohamed Salah
Eldin Mudawi for providing the much-needed library support. Sunil Kumar
is thankful to the large section of bankers in UAE who have regularly pro-
vided inputs needed to strengthen the contents of the book. Finally, thanks
for all those who have supported directly and indirectly in this quest for
knowledge.
The authors would like also to thank Palgrave Macmillan Publications
for giving them the opportunity to publish their work. And last, but not least,
xxvi
ACKNOWLEDGEMENTS xxvii
a great ‘thank you’ goes to all their individual families and friends for their
constant support, encouragement, and patience; special thanks from Ioannis
goes to his mother Ms Kyriaki Akkizidou, his sister Dr. Athena Akkizidou,
as well as to Mr. Vasilios Masmanidis and Mr. Christos Ventiadis for their
initial and endless support. Sunil is thankful to his family for patiently wait-
ing for their share of time.
Abbreviations
Abbreviation Explanation
AAOIFI Accounting and Auditing Organisation
for Islamic Financial Institutions
ADB Asian Development Bank
AMA Advanced Measurement Approach
BIA Basic Indicator Approach
BOT Build Operate Transfer
CAH Current Account Holders
CCR Counterparty Credit Risk
DIB Dubai Islamic Bank
EAD Exposure at Default
IDB Islamic Development Bank
IFSB Islamic Financial Services Board
IIFS Islamic Financial Services
IMA Internal Model Approach
IRB Internal Rating Base
IRR Interest Rate Return
KSA Kingdom of Saudi Arabia
LGD Loss Given Default
LIBOR London Interbank Offered Rate
OIC Organisation of the Islamic Conference
P&L Profit and Loss
PD Probability of Default
PER Profit Equalisation Reserve
PLS Profit and Loss Sharing
RRR Rate or Return Risk
xxviii
ABBREVIATIONS xxix
SA Standardised Approach
Shariah Body of Islamic law
SPV Special Purpose Vehicle
UAE United Arab Emirates
UIAH Unrestricted Investment Account Holder
VaR Value at Risk
CHAPTER 1
Principles of Islamic
Finance
INTRODUCTION
Efficient financial systems are one of the pre-requisites for a developing and
growing economy. Resilient and responsive financial systems enable the
government to implement monetary policies effectively, which is used to
control and manage several macro-economic parameters. Banking is a
major component of the financial system and hence has far-reaching impact
on the overall financial health and stability of an economy. Any disruption
in the banking system has serious implications for the economic conditions
of a country. Banking and financial system in a country connects individual
economic units, thereby creating a national network of financial claims, and
helps in the creation of national financial market. Essentially, banking sys-
tem routes the financial transactions, plays the role of financial intermedi-
ation and helps in the creation of wealth. The intermediation process in
financial market creates a series of interconnected contractual obligations
and relations which alters the values of the variables in the risk equation.
Intermediation helps in rationalising the decision to save and invest by sur-
plus units and to borrow and invest by deficit units. The intermediation is
generally a very long chain involving several financial entities and
processes, each of which is aimed at reducing cost, improving the match
between the needs and availability of funds, creation of tailor-made finan-
cial products based on the market needs, and provision of timely funds.
The last two decades have seen integration of financial markets across
the globe. Due to globalisation and the rapidly growing international trade,
most of the financial entities and markets are connected to each other
through telecommunication network, forming a complex web of global
financial network. These linkages of participants on the global financial
system create a network of interdependencies in terms of financial claims
and settlements. Any disturbance in the settlement of financial claim at any
1
2 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
of the nodes of this web is capable of generating a ripple effect which can
turn a local financial crisis into a global financial catastrophe. The risk of
settlement, also sometimes referred to as Herstatt risk,1 is just one of the
possible damages which these interdependencies can cause. Other than
these, due to the heavy reliance on interest rates, the financial markets are
highly prone to changes in them. Any alteration in the interest rate, either
state governed or market forced, causes impact on several dimensions of
banking and financial sectors. Borrowing and lending, the prime function of
financial intermediaries, is directly related to interest rates.
Conventional banks have been playing the role of financial intermediation
for several decades. Interest differential was the original source of revenue in
primitive models of banking which have been supported by charges, com-
missions and fees as new methods of income generation in modern banking
system. Other than the role of intermediation, banks have been providing
temporary cushions for short-term money requirements in the market as well
as assisting in performing the function of price discovery in the money mar-
ket. Conventional banking is largely based on interest rates, accounting, vari-
ous products and services, risk management activities, as well as long-term
strategies, which are based on interest rates. Islamic scholars have raised
questions about the necessity and validity of the interest in the process of
financial intermediation. In the desire to provide sustainable and justified
distribution of wealth and income, Islamic finance has attempted to find
alternates to the conventional form of financing. Interest has been considered
as a form of exploitation since it is merely a charge on money. Hence, the
prohibition of giving and taking interest among the Muslim population can
be considered as a prime reason for the origin of Islamic banking.2 Interest,
as called Riba (literally meaning ‘extra’), is considered to be Haram (liter-
ally meaning ‘forbidden’) and hence is prohibited. Added to this was the dis-
enchantment of Muslims for investing in economic activities involving
Gharar (literally meaning ‘risk’, ‘uncertainty’, and ‘hazard’). Due to the
strong emphasis in Islamic economics on equitable distribution of wealth in
the society, there has been insistence on sharing the revenue with the less for-
tunate in the form of Zakat3 (literally meaning ‘purification’). Hence, the
Islamic banking is mainly based on the absence of Riba in the transactions,
avoidance of Gharar in contractual terms, payment of Zakat for the needy
and poor, and avoidance of Haram activities. A common thread running
across all these tenets is protection of the poor and weak from exploitation by
the rich and powerful. Islamic finance has a strong root in sustainable soci-
ety with focus on welfare, equality, and justice. Social implications of com-
mercial activities cannot be neglected in Islamic finance since it has a strong
emphasis on a socially responsible form of financing. The activities of
Islamic finance are not purely materialist, although profit is a motive, but it
is supported by strong social responsibilities and accountabilities. The social
PRINCIPLES OF ISLAMIC FINANCE 3
objectives cannot be separated from commercial objectives in Islamic
finance. Some of these differences between conventional financing and
Islamic financing are summarised in Table 1.1.
The formal and structured Islamic banking is of recent origin, although it
was present in unorganised and privately-held activities, the systematic
efforts begun in the mid-seventies. With the opening of the Islamic
Development Bank (IDB) in Saudi Arabia and Dubai Islamic Bank (DIB) in
United Arab Emirates, the formalisation of Islamic financial services began.
IDB is now an international financial institution consisting of members
from 56 countries and DIB is one of the largest Islamic Banks in the world.
This phase of development of Islamic financial sector was marked by set-
ting up of Islamic banks which can offer Riba-free banking. During the late
seventies and early eighties, Islamic banking spread to Egypt, Malaysia,
Table 1.1 Differences between conventional financing and Islamic financing
Conventional finance Islamic finance
Primarily based on Interest rate Interest is prohibited
Facilitate financial activities Facilitate social, economic and
financial activities
Structured and formalised Unstructured and still informal in
many ways
Stress on financial efficiency Stress on social, ethical and financial
efficiency
Restricted moral dimension Strong moral dimension
Highly systematised in terms of risk Standards for risk management,
management, accounting and other accounting and other activities are still
standards developing
Existing set of legislations to deal Legal support still in development
with legal issues with several legal areas under doubt
Highly developed banking and Developing banking and financial
financial product market product market
Existence of conventional Non-existence of significant Islamic
money market money market
Availability of inter-bank funds Non-availability of inter-bank funds
Strong and developed Non-existing secondary market for
secondary market for securities securities
Existence of short-term Non-existence of short-term money
money market market
4 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Jordan, Sudan, Pakistan, and Iran. Faisal Islamic Bank in Egypt, Jordan
Islamic Bank for Finance and Investment in Jordan, Sudanese Islamic Bank
in Sudan and several others in Middle East countries started their oper-
ations. Malaysia systematically adopted Islamic banking with the setting-up
of the Bank Islam Malaysia Berhad. During this phase rapid expansion was
seen, with some of the countries adopting completely Islamic models of
banking, including Sudan and Iran. After the late eighties, the shift was on
developing marketable, consumer-friendly, Shariah-compliant Islamic
financial products. Several innovative products were introduced, with more
and more banks coming up with unique solutions for Islamic financing. At
the same time, the need was felt for standardisation and good governance.
Efforts to develop International Accounting Standards for Islamic financial
services gathered momentum. Organisations such as Islamic Financial
Services Board (IFSB) and Accounting and Auditing Organisation for
Islamic Financial Institutions (AAOIFI) had been working in the direction
of providing standards for the Islamic financial industry. Central Banks in
countries offering Islamic financial services started developing standards
for the supervision of Islamic financial institutions. Islamic insurance
(Takaful) developed as a separate branch of Islamic financial services.
Hence, the framework of Islamic financial services include, among other
things, Islamic banks, other Islamic financial institutions, AAOIFI, IFSB,
Central Banks, and other interest groups (Figure 1.1). During 2007, more
than 250 Islamic banking institutions were providing a wide range of finan-
cial products all over the world. The number is growing rapidly due to the
entry of new institutions and conventional banks into Islamic banking by
opening an Islamic window. The growth is one of the indicators of the success
Accounting Governance
standards standards
Banks Shariah board
Other Central bank Risk
financial supervision management Other interest
groups
Institutions Standards
Figure 1.1 Structure of Islamic financial service industry
PRINCIPLES OF ISLAMIC FINANCE 5
of the Islamic form of banking. Muslims all over the world have been keen
on accepting the Islamic form of financing, largely due to religious reasons.
Equal interest has been shown in Islamic banking by non-Muslims due to
financial reasons.
The growth of sustainable development and the raising concern over
wasteful consumption and exploitative production has also helped in the
development of Islamic banking, which accepts an ethical approach to
financing rather than a pure profit-based financial approach. Money should
not be used to exploit the plight of others is the fundamental rule. Coupled
with this, the other tenets of Islamic banking include avoiding Gharar,
insistence on clear terms in contract, prevention of transactions for specula-
tive activities and linking financing to productivity. Banks do not remain
mute spectators in Islamic financing, but become an active partner and
attempt to participate fully in the sharing of profit and losses with customers
and thus Islamic banking is also known as Profit and Loss Sharing Banking.
In true essence, Islamic banking cannot be conducted without a participa-
tion in profit and loss. The foundation of Islamic banking lies in the partici-
pation and not in simple financial intermediation. While conventional banks
perform the function of transfer of funds from the surplus to the deficit units
by using interest on the funds intermediated as the source of revenue,
Islamic banks perform the same function by way of active participation
between the two units, linking the risks and return in a more formal way.
Thus, we can summarise that Islamic banking and finance wishes to achieve
the objectives of a more equitable society with better distribution of wealth
and income, promote ethical business activities, create a just and fair financial
system where the rich do not benefit at the cost of the poor, and remove uncer-
tainty in business dealings. Due to the emphasis on equity-based financing
over debt-based financing, the risks in Islamic banking differ from conven-
tional banking. The credit risk, market risk and operational risk, albeit present
in Islamic banking, differ in their origin, impact and implications. These will
be explored in detail in the next few chapters of the book.
Although theoretically Islamic banking seems to use the same tenets for
financing as used by conventional banking, they differ greatly in terms of
their application. The method of conducting Islamic banking differs from
bank to bank, from country to country, and sometimes from product to prod-
uct. These differences can be partly attributed to the presence of the Shariah
board, which governs and guides the banks regarding the conduct of Islamic
banking. The Shariah board interprets various products, services and situ-
ations based on Holy Quran and fiqh (Islamic jurisprudence). There are four
classical schools of Islamic thoughts; namely, Hanafi, Maliki, Shafi’i, and
Hanbali. These schools have specific presence in different parts of the
world and hence the Shariah ruling differs which can also be found based
on them. China and Turkey are more influenced by the Hanafi; in a large
6 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
part of Africa Maliki is followed; Indonesia and Malaysia have large fol-
lowers of the Shafi’i school; and Hanbali appears to be followed in Arabia.
These four schools represent most commonly accepted rulings of Islamic
jurisprudence. The interpretation of Shariah scholars can be based on one or
more schools of thought and hence can have impact on the conduct of the
Islamic banking.4 Multiple factors are considered before Shariah provides a
ruling on a particular case. This multiplicity of methods of financing has
been a prime reason for the lack of standardisation of products, processes,
and policies. This did not hamper the growth and development of Islamic
banking, but has resulted in some confusion among the followers of Islamic
banking. Due to the multiplicity of interpretations of situations, the progress
on the front of developing specific legislation for Islamic banking has been
slow. Malaysia, Pakistan and Bahrain have developed specific legislations
dealing with Islamic banking, whereas most of the other countries offering
Islamic banking are using conventional banking legislations with some
modifications for Islamic banking along with Shariah rulings.
The existing literature deals with Islamic banking in isolation and thus
restricts the comprehensive view where Islamic banking is seen as an inte-
gral part of the global financial system. Risk management has been exten-
sively dealt with, but with the focus on conventional banks. There is only a
handful of literature on risk management in Islamic banking and finance.
This chapter begins the systematic attempt to bridge this gap and under-
stand risk management in Islamic banking and finance holistically. The
Balance Sheet of an Islamic Bank is presented as a starting point for under-
standing the grouping of assets and liabilities which are different from the
conventional banks. A typical Balance Sheet of an Islamic Bank is also pre-
sented in order to offer an overview of how reporting is managed in Islamic
financial institutions. Zakat as an obligation has a bearing on the financial
statements and thus is also shown in the Balance Sheets. The role of
AAOIFI has also been highlighted.
The nature of Islamic financial activities cannot be understood without
clear understanding of Profit and Loss Sharing (PLS) principles. This chap-
ter presents an overview of the PLS contracts and their role in Islamic finan-
cial activities. Various rules related to PLS contracts are explained and the
present concerns are also analysed. The reasons for the limited popularity of
the PLS contract are further explored in this chapter.
The chapter presents an analysis of some of the more common Islamic
financial products, including Mushãrakah, Mudãrabah, Murãbaha, Salam,
Istisnã, and Ijãrah. A brief overview of Sukuk is also presented. PLS contracts
are the foundation of Islamic finance and Mushãrakah and Mudãrabah are
the two common forms of contract using PLS in Islamic finance. The bank
and customer join as partners and undertake business activities. Where the
bank contributes only the capital, it is called a Mudãrabah contract. Under
PRINCIPLES OF ISLAMIC FINANCE 7
a Mushãrakah contract, the bank can take part in the management of the joint
business. Several Islamic scholars claim that these two forms of Islamic
finance should preferably be used as compared to Murãbaha. Several rules
related to Mushãrakah and Mudãrabah are enumerated in this chapter. The
analysis is supported with examples from real life.
Murãbaha is the most popular type of Islamic financing contract.
Popularly known as ‘cost-plus’ financing, it has been the instrument of
choice for Islamic banks. Due to limited risk exposures, as compared to
other contract types, Murãbaha has been in use for a long time and has con-
tributed maximally to the total Islamic business (for a detailed analysis of
risks related to Murãbaha, please refer to chapter 2). Murãbaha is most
commonly used for financing consumer goods, real estate, machinery, and
equipment. As per most common Shariah opinions, Murãbaha should not
be used on a long-term basis and should be treated as a ‘stop-gap’ arrange-
ment until other Islamic financial instruments are fully developed.
Islamic finance does not permit forward sale, with the two exceptions of
Salam and Istisnã. These are discussed in detail in the present chapter.
Salam can be used for agricultural commodities, whereas Istisnã can be
used for manufactured (or constructed) goods. Different requirements for
Salam and Istisnã are explained in this chapter. Parallel Salam, where two
simultaneous agreements are used by the banks, is also explained with the
help of examples. A quick comparison that brings out the main differences
between these two types of contract is also presented in this chapter.
Leasing in Islamic finance is managed through Ijãrah, which is dealt
with in this chapter. Ijãrah is subjected to certain rules as provided by
Islamic fiqh, which are explained. Several dimensions of Ijãrah contracts
relating to alternative methodologies for creating a contract are explained in
detail in this chapter. The two main types of Ijãrah contracts – that is, Ijãrah
thumma al bai, and Ijãrah wa Iqtina – are also presented. A risk analysis of
these two is presented in subsequent chapters.
Sukûk is most commonly known as Islamic Bonds. A detailed analysis of
highly dynamic Sukûk is presented in this chapter. Since the market for
Sukûk is going through a rapid transformation, some of the information may
be required to be updated from other sources while using it. There are large
varieties of Sukûk, depending on the type of underlying contracts; but, based
on essential qualifications, Sukûk can be debt-based or equity-based.
Differences between conventional bonds and Sukûk are also presented in
this chapter.
The Islamic banking balance sheet
The extent of liability of the owner provides two alternative models of the
organisations. One can be based on the concept of unlimited liability,
8 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
which is popularly known as ownership form of business, and the other
can be with owners having limited liability. Joint stock companies, which
are based on the concept of limited liability, are a result of rapid industri-
alisation during the early twentieth century. Heavy investments and
involvement of large-scale funds made it necessary to have organisation
with pooled capital. There was also a need for specialised management
which is separate. Joint stock companies with the distinctive features of
limited liability, separation of ownership from management, and legal
entity were a result of industrial revolution. Due to their typical nature and
complicated financial structures, companies need special methods and
standards for maintaining their accounts. Asset valuation and accurate
measurement of income and liabilities is one of the difficulties faced by
the accounting. Availability of alternative methods for doing this, along
with choices of concepts and conventions, has added confusion to the
existing chaotic situation. In the last few decades this problem has been
solved to a large extent by adopting universal concepts, conventions, and
principles. All throughout, the effort has been to monetarise everything.
Focus has been on determining income and valuation. The concept has
been further extended to growth and finally to the wealth maximisation.
Islam has reservations over these issues and offers an alternative view
which includes social, moral, and ethical dimensions to the valuation and
measurement of assets. However, it should not be misunderstood that
Islamic finance is against income and profit. Zakat (obligatory alms) is
calculated on income and profit and thus both are relevant and not Haram
(prohibited). Income in accounting is ex-post calculation and is a historical
concept. The valuation of assets using discounted cash flows is generally
not acceptable in Islamic finance, since, according to Islamic finance,
there is no debt-related time value of money. The concept that ‘a dollar
today is worth more then a dollar tomorrow’ does not find much support
in Islamic finance. This is in line with the prohibition of Riba, as it mainly
represents the time value of money. Islamic finance can use the method
popularly known as ‘current cash equivalent method’. Along with other
reports, due to the calculation of the Zakat, the Income and Expenditure
Account as well as Balance Sheet are also emphasised. Due to the close
integration of all the beneficiaries of a business venture, reporting of true
profits is essential in order to provide justice to them at the time of distri-
bution of profits. Reporting in conventional accounting is limited to finan-
cial information, whereas in Islamic finance there is an emphasis on
reporting non-financial facts, whilst keeping co-operative aspects of eco-
nomic activity in mind rather than competitive aspect. Due to the social
responsibility attached to the economic activity, it is expected that the
reporting should be more elaborate and extensive, also keeping the non-
financial aspects in it. The sustainability of Islamic banking depends on
PRINCIPLES OF ISLAMIC FINANCE 9
the public trust, which is directly related to candid, sufficient, timely and
true and fair information disclosure. Hence, sufficient reporting is of the
utmost importance in Islamic banking.
Accounting aims at presenting the ‘true and fair’ financial view of the
organisation. There has been a series of developments related to account-
ing standards. International standards have evolved and have helped in
standardising the accounting practices in conventional banking. Islamic
banking attempted to align with the success of International Accounting
Standards by creating the AAOIFI standards for Islamic accounting.
Before the introduction of AAOIFI standards, each Islamic bank had its
own accounting standards, practices, and regulations. Differences in these
standards, practices, and regulations have led to inconsistencies, rendered
financial results non-comparable, and created confusion. This was added
to the growing resentment among Islamic institutions towards a blind
adoption of the Western accounting model, which was presumed to be
materialistic and based on consumerism. There has also been concerns
regarding the belief by Islamic and some Western scholars that conven-
tional accounting models do not take into account the ‘other costs’, such
as environment degradations, inequality of income distribution, social
implications, and the degeneration of resources. Islamic accounting ide-
ally wanted to move away from simple monetary measurements, various
conventional accounting concepts and conventions and include non-
pecuniary information so as to provide a better view of the use of resources
as per Islamic jurisprudence. Islamic institutions wanted their own
accounting standards, which are Shariah compliant, leading to the devel-
opment of AAOIFI. The aim of any accounting standard is to record and
present true and fair economic facts. The Islamic accounting differs from
the conventional accounting standard in this respect as the former concen-
trates only on the pecuniary impact of the transactions, where the Islamic
accounting attempts to focus on the pecuniary as well as the ethical and
social impact of the transactions. Out of several controversies related to
Islamic accounting, the most prominent is the treatment of Zakat. Some
scholars believe that Zakat is a cost and hence should be included in the
expenses; whereas some others feel that it is a part of profit distribution
and hence should be treated as appropriation.
Islamic financial institutions will have to prepare similar sets of finan-
cial statements as prepared by conventional banks; statement of income,
statement of cash flows, and a balance sheet. Other than these, they also
have to prepare a statement consisting of transactions related to Zakat
funds. Islamic economists ideally want to have only investment deposits
on the liability side and on the asset side only profit-sharing contracts. But
this is practically not possible due to the many complexities involved. For
all practical purposes, Islamic banks have current account and investment
10 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
deposits on the liabilities side and on the asset side they have several
profit-sharing variants of products such as Mushãrakah, Mudãrabah,
Murãbaha, Istisnã, Salam, and Ijãrah, which are discussed further in this
chapter. The fundamental difference between Islamic banking and conven-
tional banking arises due to the prohibition of Riba. The non-interest-
bearing current accounts are distinguished from the investment deposits.
The asset side of the Islamic bank comprises of funds which are held with
the Central Bank and other financial institutions, property and other
assets, along with Murãbaha, Mudãrabah, Ijãrah, and other financing.
The liability side largely consists of depositor funds. Islamic banks have
to follow certain rules regarding the maintenance of the accounts. First,
each project should be financed and recorded through a special project
account which can specifically record and report the transactions related
to the project. Hence, it is more transaction-based accounting. Second, the
Mudãrabah account should be maintained separately and the settlement
should be on the basis of realised cash and not recorded cash. Similarly,
third, in the case of diminishing Mushãrakah, the net accrued profits
should be considered at the end of the year. Fourth, the profits should be
distributed each year and should not be accumulated. Finally, for
Murãbaha contracts, the profits should be calculated after the termination
of the contract. They generally cannot divide liabilities into different cat-
egories based on liquidity.
At the time of writing this book, not many banks followed the AAOIFI
standards and hence the financial statements of Islamic banks are non-
comparable and are not standardised, although most of them use a similar
format for reporting. A typical balance sheet for an Islamic bank is pre-
sented in Table 1.2.
Table 1.2 Typical balance sheet of an Islamic bank
ASSETS
Cash and cash equivalent
Financing and receivables
– Murãbaha
– Salam
– Istisnã
Investments securities
Mudãrabah investments
Mushãrakah investments
Continued
PRINCIPLES OF ISLAMIC FINANCE 11
Table 1.2 Continued
Investment in other entities
Inventories
Investment in real estate
Assets acquired for leasing
Other investments
Fixed assets
Other assets
Balances with and due from financial institutions
Diminishing Mushãrakah
LIABILITIES
Deposits and other accounts
– Current accounts
– Savings accounts
– Term deposits
– Others
Deposits of other banks
Salam payable
Istisnã payable
Declared but undistributed profits
Zakat and taxes payable
Other accounts payable
NET ASSETS
REPRESENTED BY
Islamic Banking Fund
Reserves
Inappropriate / Un-remitted profit
Surplus / (Deficit) on revaluation of assets
Remuneration to Shariah Advisor / Board
CHARITABLE FUND
Opening Balance
Add: Additional provisions
Less: Utilisation of funds
Closing Balance
12 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Understanding profit and loss contracts
Conventional banking, largely based on interest, is characterised by non-
participation of contracting parties in the profits and losses resulting out of
the fulfilment of contracts. The contracts consist of pure lending with the
onus of making profits solely lying on the borrower and the lender clearly
demarking his liabilities. Further, in the case of profits, the borrower does
not pay anything to the lender which is in proportion to his share and, in the
case of bankruptcy, the shareholders have to suffer, according to which
Islamic finance was the unjust distribution, leading to inequality. Islamic
banking is dominated by Profit and Loss Sharing (PLS). It primarily pro-
motes PLS between the lender and borrower by converting their relation-
ship from borrower and lender to partners. Islamic financial institutions are
facing credit, operational, market, and liquidity risks of a higher degree,
mainly due to PLS.5 It needs transparent contracts as a foundation to the
transaction. The Shariah requires that the contract should be drawn in clear
terms and nothing should be hidden or left unclear. Any uncertainty is not
permitted and is treated as Gharar. Moreover, the contracts should be as per
the Shariah rules and principles. Absence of a universally acceptable legal
framework for Islamic contract has led to several problems. The progress on
designing, developing, and propagating the legal structure has been unsatis-
factory. In countries such as Sudan and Iran that completely follow an
Islamic financing system, there has been a frugal legal system to support,
and in several other countries which are following Islamic banking as a par-
allel system, there is a lack of prudent Islamic legal system in place. There
have also been inconsistencies between the conventional and Islamic legal
interpretations relating to financial transactions.
PLS is one of the fundamental principles of Islamic financing and can
acquire several forms, depending upon the type of contract. The most com-
mon form of it is where a bank provides funds and an entrepreneur provides
time and effort. At the end of a stipulated period, if there are profits, they are
shared in the agreed proportion. If there are losses it is generally the bank
that takes the entire responsibility, unless there is a mismanagement and
negligence on the part of the entrepreneur. Hence, apparently there appears
to be greater risks in profit and loss sharing contracts. Primarily, the rate of
return on any investment is calculated ex-post and thus any pre-determined
fixed rate of return is not permitted. The profit and loss sharing contract
affects the bank balance sheet in several ways. The use of depositors’ funds
is shown as a liability since it represents an unrestricted use of depositors’
funds for financing entrepreneur activity. The provision of funds to an entre-
preneur is shown on the asset side.
Although PLS is the cornerstone of Islamic financing, its use has been
limited and debatable. There are several reflections of PLS used by Islamic
PRINCIPLES OF ISLAMIC FINANCE 13
finance, such as Mushãrakah and Murãbaha. Habib6 observes that it was
expected that Islamic banking would take the form of the two-tier
Mudãrabah model, but was forced to use Murãbaha. Although there is a
strong agreement among Islamic scholars that the most desirable form of
Islamic financing is using Mushãrakah and Mudãrabah, there has been
reluctance from both the parties, the banks and the borrowers, to use PLS.
Institutions are reluctant due to the high degree of uncertainty attached to it,
whereas businessmen have their own reasons for not using PLS. There are
several reasons for the limited use of PLS by Islamic financial institutions.
Desire for secretiveness and lack of transparency, complicated and expen-
sive monitoring, and taxation issues are some of the reasons. Iqbal and
Molyneux7 cite some more reasons, such as the need to keep and reveal
detailed records, limited opportunities to re-invest retained earnings and/or
raise additional funds, and loss of exclusive control over the business. Some
studies, however, hold the moral hazard problem responsible for this.
Another issue is asymmetric information and the nature of banks as short-
term financial institutions.8 Probably the conventional model of information
availability (strong, semi-strong, and weak) does not appear to exist in
Islamic finance, and, due to lack of information, the bank can end up mak-
ing an adverse selection. An entrepreneur having a financially sound project
will shy away from an Islamic bank since it will involve sharing his profits
and thus increasing his cost of funds. On the other hand, projects with lim-
ited financial soundness will come to Islamic banks due to the natural pro-
tection available for the failure in Islamic financing. Hence, Islamic banks
can be left with a choice among weak projects. Similarly, an established
organisation will approach conventional forms of financing since debt-
financing will be a cheaper option. On the other hand, start-up firms will
approach Islamic banks due to hedging against losses. Islamic banks thus
will be left with financing infant firms. The moral hazard problem is at the
core of PLS, and makes it difficult to use. The questionable honesty of the
entrepreneur and his misdirected use of funds can lead banks into difficulty.
Summarising the discussion, for a bank to enter into a PLS agreement
with an entrepreneur, it should have all the necessary information at all the
stages of a contract. Before entering into the contract, the required informa-
tion is generally made available by the entrepreneur as an incentive to obtain
funds. During the life of the enterprise, the problem of under-reporting
may arise, especially when there is no punishment/monitoring/incentive
clause in the contract. It is at this stage where the entrepreneur can misman-
age the funds. Moreover, there are no established models for Islamic banks
to use during contracting which can accurately measure the risk involved. As
discussed before, monitoring is expensive and hence cannot be used in all the
cases by banks. On the other hand, there are incentives for the bank to use
PLS since it provides equity-based funding as compared to debt funding,
14 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
which is expensive. Hence, if a bank can manage information asymmetry,
the PLS model appears to be less expensive.
Islamic financial products and services
Islamic financial products have evolved over two decades. The journey has
been need-based from simple Murãbaha financing to Sukûk. The Islamic
financial products are mostly based on classical Islamic contracts and can
be grouped into three broad types: equity financing, sales financing, and
lease financing. Equity financing mostly takes the form of Mushãrakah and
Mudãrabah. Many of the Islamic financial products are a variation of the
following types of instruments:
Mushãrakah
Mushãrakah is also known as ‘Partnership Financing’ or ‘Joint Venture
Financing’. Islamic Fiqh does not refer to the word Mushãrakah; it is a deri-
vation from the word shirkah, which means sharing. Unequal distribution of
wealth is treated as a sin in Islam and hence any partnership with unjust dis-
tribution of profits or losses is not permitted. It is believed that the interest is
the root cause of unequal distribution of wealth and hence moderation is used
while distributing the profits and losses of the business, where, in the case of
profit, the managing partner should not take a large share and, in the case of
loss, the non-working partners should not be asked to bear all the burden of
loss, which generally is the case in conventional partnerships. To overcome
these issues of conventional partnership, Mushãrakah is used as an alterna-
tive. Mushãrakah is a form of equity financing which refers to a partnership
agreement between the bank and the customer where equity is contributed
jointly and profits and losses are shared on agreed terms; however, it is not just
lending money. The capital can be contributed in cash or in the form of goods
or assets. The profit-sharing ratio can be decided at the time of the agreement
but, in the absence of a loss-sharing ratio, the losses will compulsorily be
shared as per the proportion of the capital. Both the parties have the right to
manage, although one of them may surrender their right in favour of the other.
Mushãrakah is seldom used due to the high degree of uncertainty over the
returns. It is used in cases involving huge investments and for joint-venture
projects. In Mushãrakah, the finance comes from both the parties. In case
only one party finances the whole project, Mudãrabah is used, which is
explained in the next section. Mushãrakah can also take the shape of
Diminishing Mushãrakah, where the entrepreneur keeps purchasing the share
of the financier regularly and thus diminishes the contribution of the financier,
which eventually is completely phased out. A possible use of Diminishing
PRINCIPLES OF ISLAMIC FINANCE 15
Mushãrakah can be for real estate financing. However, Diminishing
Mushãrakah cannot be used easily for financing regular trade transactions.
The general law of partnership applies for the termination of Mushãrakah.
The partners can terminate the Mushãrakah by giving a notice. Death or
insanity of one of the partners also leads to the termination of Mushãrakah.
Example – permanent Mushãrakah A customer X of bank ABC plans to
start a production plant for the manufacturing of ceramic tiles. The bank
enters into a contract with customer X whereby the bank contributes 40%
and customer X contributes 60% of the capital. The bank surrenders its right
to manage in favour of customer X. They agree to share the profits in the
ratio of 20% for the bank and 80% for the customer. In case they make a
profit it will be shared in the ratio of 1:4; but if there is a loss, it will be
shared in the ratio of 2:3 (capital ratio).
Example – diminishing Mushãrakah A customer X of the bank ABC
plans to start a commercial transport service by purchasing a delivery van.
The bank enters into a diminishing Mushãrakah with the customer where it
contributes 90% of the cost of the van and the customer provides the
remaining 10% (generally called down-payment in conventional form). Out
of the profits, the customer keeps 25% and the remaining 75% goes to the
bank. Let us assume the capital contribution to be divided in 20 equal units;
thus, the bank owns 18 units and the customer owns 2. At regular agreed
intervals, say, two months, the customer keeps purchasing one unit from the
bank’s contributed share in units, thus increasing his contribution and
reducing the bank’s. After 36 months, the customer will own all the units of
the bank’s contribution, thus owning the van all by himself. A detailed
analysis of changing risk exposures due to the changing ownership ratio is
shown in the next chapter (Figure 2.8).
Mudãrabah
Mudãrabah is also known as ‘sleeping partnership’. Mudãrabah refers to a
partnership where one of the partners contributes capital (rabb-ul-mal) and
the other partner the management (mudarib). This is also a form of equity
financing and is more popular than Mushãrakah. The investing partner
cannot take part in the management of the firm. The investing partner
can provide the funds with a restriction that they will be invested in a par-
ticular business; it is then called restricted Mudãrabah. On the other hand,
if rabb--ul-mal allows mudarib to invest in any business, it is called
unrestricted Mudãrabah. Many banks use Mudãrabah to mobilise funds
through savings and investment accounts.
16 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Example A customer X approaches the bank ABC to invest in the manu-
facturing of ceramic tiles, but insists that he cannot contribute any capital.
The bank invests all the required funds for the business, whereas the cus-
tomer takes care of the management. The customer and bank are partners in
this joint venture. There are several risk exposures in this form of financing,
which are illustrated in chapter 2.
Note Mushãrakah and Mudãrabah have some common features. Both
should not be confused with simple financing of business. They require par-
ticipation in business either in the form of contribution to the capital or man-
agement or both and prohibit one party benefiting at the cost of the other in
sharing profits and losses. The partners are free to determine the ratio of
profit and loss sharing, subject to the condition that, in the event of losses,
they should be shared as per their capital contributions and profits for the
financing partner (who is not taking part in management) cannot exceed his
share in capital. Both can be securitised, particularly where the investments
are vast. The investment can be divided equally in parts and Mushãrakah /
Mudãrabah certificates can be issued to each contributor like debenture cer-
tificates. Under certain circumstances these certificates can be traded on the
secondary market and hence can provide the much required liquidity for
Islamic banks. Different Shariah experts have different opinions on the
negotiability of these certificates. Some believe that if the investments are
held in cash and assets, it can be traded at a value more than the asset value;
while others believe that it cannot. For practical bankers, Mushãrakah and
Mudãrabah can be used to finance import and export trade. They can also
be used for financing transaction-based working capital needs. In cases
where there are a few assets with business it is easy to use Mushãrakah, but
when the business consists of large number of assets, the profit sharing can
be done on the basis of gross profit rather than net profit and the entrepre-
neur can be compensated by increasing his share in the profit to account for
the indirect costs. The provision of a daily overdraft is also theoretically
possible using these instruments; however, practically they are difficult to
use. However, there are some issues associated with these instruments
which have resulted in making them less popular with bankers. The bank
has a greater risk of loss when it uses these instruments and thus the share-
holders will be reluctant to invest in them. Another problem can arise when
the entrepreneur cheats the bank and does not disclose the true and fair
financial status of the business. The differences between Mushãrakah and
Mudãrabah are summarised in Table 1.3.
Murãbaha
Murãbaha is also referred to commonly as ‘Cost-Plus Financing’ or ‘Mark-
up’. According to Islamic jurisprudence, Murãbaha is a type of sale where
PRINCIPLES OF ISLAMIC FINANCE 17
Table 1.3 Differences between Mushãrakah and Mudãrabah
Feature Mushãrakah Mudãrabah
Type of partners All partners provide capital Two types of partners, one
and contribute to the who contributes the capital
management; hence, there is (rabb-ul-mal), the other who
only one type of partner. takes care of the management
(mudarib).
Rights of All partners have the right Financing partner (rabb-ul-
management of management. Partners mal) cannot take part in
have the right to surrender management. Management
their right of management. exclusively remains with non-
financing partners (mudarib).
Sharing profits All partners share profits in Partners share profits in
and losses agreed proportions and losses agreed proportions and the
in the proportion of their financing partner (rabb-ul-mal)
capital contribution. will share losses to the extent
of his capital.
Liability All partners have unlimited The liability of financing
liability. They will have to partners (rabb-ul-mal) is
bear the extra liability in the limited to the extent of their
proportion of their capital. capital contributions, whereas
managing partners (mudarib)
have unlimited liability.
Ownership of All assets of the partnership The ownership of assets
assets business are jointly owned by remains with financing
all the partners. partners (rabb-ul-mal) and
the managing partners
(mudarib) do not participate
in the ownership of assets
of the business.
the seller discloses the true cost of the products and then adds a mark-up to
sell it at an agreed price. The profit can be based on a percentage or a cer-
tain fixed amount. It also allows spot and future payment; however, there is
no apparent mention about the deferred payment in conventional Islamic
jurisprudence. Since Murãbaha is a contract of sale and not a contract of
loan, it should honour all the requirements of a valid contract of sale. First,
the commodity should be an existing commodity and not one which is going
18 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
to be produced or cultivated or made in future and hence this is the ground
for rejection of commodity futures in Islamic finance. Second, the seller
must be the owner at the time of sale; that is, he must either have physical
or constructive possession at the time of sale. Third, the sale cannot be con-
ditional and the consideration must be certain at the time of entering into the
contract. Fourth, the commodity must have value at the time of sale, and
hence this is the ground for rejection of trading on ‘index future’.
Bai‘Mu’ajjal is the deferred payment Murãbaha. Islamic jurisprudence
allows certain protection to the seller in terms of delayed or unpaid instal-
ments. In case the payment is delayed, the seller can charge some penalty;
however, the amount recovered from the penalty should be used for charita-
ble purposes. In the case of default in instalments, they may include a clause
where remaining instalments can be claimed immediately. There can be
clauses for having some security in the form of mortgage or lien. In bank-
ing, Murãbaha refers to a sale agreement whereby the bank purchases the
goods and sells it to the customer at an agreed profit. The repayment can
either be made in instalments or in a lump sum as decided at the time of the
agreement. Due to the practical difficulties involved in using Mushãrakah
and Mudãrabah, the Shariah experts permitted the use of Murãbaha under
certain circumstances. Shariah experts are of the opinion that Murãbaha
should not be treated as a mode of financing and should not be used on a
long-term basis. Also, Murãbaha should be used where the customer needs
the money for purchasing real commodities and not for any other purposes.
This is the most common mode of financing used by Islamic banks. Almost
all the Islamic banks use this to finance consumer goods, real estate,
machinery, and equipment, and also sometimes for financing Letters of
Credit. Legally, Murãbaha creates a series of relations depending on the
stage of the contract (see chapter 2 for a detailed analysis). It begins with the
relationship of promisor and promisee, which is then converted to buyer and
seller. When the sale between the bank and the customer is finally affected,
it creates a relationship of creditor and debtor.
Example A customer X of bank ABC wishes to purchase a car for
$25,000 from a dealer. He approaches the bank with details regarding the
make, model, and the details of the dealer. The bank enters into a prelimi-
nary contract, where customer X promises to buy the same car from the
bank ABC for an agreed price of $32,000, payable in monthly instalments.
The bank then purchases the car from the dealer at the agreed cost and adds
its profit margin (cost-plus financing), depending on several parameters,
including (but not limited to) total value, total time of financing, repayment
terms, credit standing of the customer, and down payment made by the cus-
tomer. Another contract is then made between the bank and the customer,
where the bank sells the car to the customer and the customer agrees to
PRINCIPLES OF ISLAMIC FINANCE 19
make the payment in the agreed number of instalments. At this stage, the
relationship between the bank and the customer changes to creditor and
debtor. Similar contracts can be made for furniture, electronics, personal
computers, machinery, equipment, and ships. The effect of these changing
relationships on risks in the contract is illustrated in chapter 2.
Salam
Salam is also known as ‘forward sale’. Salam was originally allowed to
meet the needs of small farmers who needed money during the harvesting
period to meet expenses for harvesting as well as to maintain their family.
Since borrowing on interest was not permitted, they were allowed to use
forward sales. It was also used originally for the import and export trade in
the Arab region. As discussed in the previous paragraphs, according to
Islamic jurisprudence the essential conditions of a valid sale include that the
commodity should be ready at the time of the sale. However, Salam and
Istisnã (which is explained in the next section), are the two exceptions
which are permitted according to Islamic fiqh. Hence, Salam is a contract of
forward sale, whereby a commodity is sold on a future date, for which the
complete price is paid on the spot. In this case, the price is paid in cash and
the delivery is deferred. The buyer is called the rabb-us-salam, and the
seller is called the muslam ilaih. Salam is permitted under certain condi-
tions: 1) The price should be paid in full and on the spot. 2) It can be used
for specific products and it can be used for commodities which can be
clearly explained in quantity and quality. 3) The time and place of deliver-
ing the commodity should be specific. 4) Salam cannot be used for barter
transactions. Regarding the time of delivery, opinions are divided. Some
Islamic scholars feel that the time of delivery should not be less than one
month, whereas other scholars feel there is no minimum period. Traditional
Islamic jurists have maintained that the price of Salam should be lower than
the spot price, whereas contemporary Islamic scholars believe that the price
can be decided as per the free will of buyer and seller. The unique feature of
Salam is the confirmed delivery. The seller should deliver the goods on the
specified date. Salam does not include gold, silver, and currencies. Bankers
can use Salam to finance the agricultural sector. It can also be used for
financing commercial and industrial activities, operational costs, and capi-
tal goods. There can be several applications of Salam, depending on the cir-
cumstances. It should be noted that Salam exposes banks to market risk,
especially fluctuations in the commodity prices (a detailed risk analysis of
Salam can be found in chapter 2 and market risk in chapter 5). Banks are not
very happy to take the delivery of the commodity which is a necessary con-
dition in the Salam. To avoid this, modern bankers are using the Parallel
Salam, where a bank enters into two simultaneous agreements for the same
20 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
future date, one as a buyer and the other as a seller. Parallel Salam will be
discussed in the following paragraph with the help of an example. Parallel
Salam takes care of commodity price fluctuations to a certain extent, but
still requires managing the risks from the non-delivery of the commodity on
the due date.
Example – parallel Salam A customer X of bank ABC wishes to arrange
for a forward sale of his crops. The bank enters into a Salam contract (first
contract) where it provides the required finance and in return agrees to pur-
chase the crop from customer X after a specified time. The situation prima-
rily benefits both parties. Customer X is relieved of the burden of repayment
of the amount borrowed as he does not have to repay the same as cash, and
secondly he has sold a certain part (or all) of the produce, thereby reducing
the uncertainty of the sale in the future. The bank can benefit by simultane-
ously entering into another Salam contract (Parallel Salam) to sell the crop
to a third party, thus minimising its risk. The possible price fluctuation in
the market price of the crop (market risk) can be marginalised by entering
into parallel Salam. The bank is thus left with facing the risk of not getting
the right quality or of non-delivery of the crop on time. To shield itself
against this, the bank can include necessary compensation clauses in the
first contract. A detailed analysis of risks and their management for a Salam
contract is presented in chapter 2.
Istisnã
Istisnã is also known as ‘progressive financing’. It is also called purchase
by order. As discussed in the previous paragraph, the two exceptions to
the sale contracts in Islamic finance are Salam and Istisnã. Istisnã is the
second method of sale where the commodities can be sold before they are
ready. Istisnã is applied for goods made-to-order. It is an agreement
aimed at forward contracts in manufactured goods. It allows for on-the-
spot payment and deferred delivery or a deferred payment and a deferred
delivery. There is a provision of deferred payment in instalments. This
flexible combination makes it a popular tool for Islamic banks. In most
of the cases the payment is made periodically as per the progress in man-
ufacturing or construction. Banks use Istisnã to finance construction
activities as well as manufacturing activities. Several banks have suc-
cessfully integrated Istisnã for real-estate financing. Banks can use this
to finance capital-intensive goods by entering into two contracts, one
with the customer who wishes to have the goods and the other with the
manufacturer to produce the goods. The government can use Istisnã for
developing infrastructure by using the Build-Operate-Transfer (BOT)
basis. For example, a highway can be constructed on BOT basis by enter-
ing into Istisnã with a construction company. In this case, the price of
PRINCIPLES OF ISLAMIC FINANCE 21
Istisnã can be the right of the construction company to maintain the high-
way and collect tolls for a fixed period of time.
Example A customer X wants to purchase a ship. Bank ABC agrees to
provide the financing using Istisnã, where is enters into a contract with the
ship manufacturer to manufacture the specified ship and agrees to provide
the finance in instalments, depending on the progress in manufacturing. The
bank also enters into another contract simultaneously with customer Y to
sell the specified ship at an agreed price. Inherent risks in this type of trans-
action are discussed in detail in chapter 2.
Note It might appear that Salam and Istisnã are both sales contracts where
the commodity is not ready and hence are similar; however, it should be
noted that there are some important differences between Salam and Istisnã.
These differences are:
1. Salam can be used for almost all types of commodities, excluding gold,
silver, and currencies; however, Istisnã can only be used for commodi-
ties which require manufacturing.
2. Salam price is always paid in advance; whereas the price of Istisnã can
be paid either in advance or in the future.
3. Salam price is always paid in full; whereas the price for Istisnã can be
paid in instalments.
4. Salam contracts cannot be cancelled unilaterally.
5. The delivery date is fixed in Salam; whereas in Istisnã the delivery date
need not be fixed.
6. The material and manufacturer can be fixed by the buyer in Istisnã;
whereas in Salam this cannot be fixed.
Ijãrah
Ijãrah is also commonly known as leasing. There are two connotations in
Islamic jurisprudence for Ijãrah: one is for hiring services and another is for
assets and properties. In the first case, the service can be hired and the reward
can be paid for labour. In the second, Ijãrah refers to the sale of a definite
usufruct in return for a definite reward. Since the first type of Ijãrah is not
used in Islamic banking, the discussion will be restricted to the second con-
notation. For a bank, the second connotation is more important and applica-
ble. Banks can use Ijãrah for leasing equipment, machinery, vehicles, and
other assets. The lessor in this case is called ‘mu’jir’, and the lessee is called
22 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
‘musta’jir’. Ijãrah is subject to certain rules as provided by Islamic fiqh:
1. The subject of the lease must have a useful value.
2. The ownership of the asset should remain with the lessor and only the
right to use the asset should be transferred to the lessee.
3. The liabilities pertaining to the ownership remain with the lessor;
whereas the liabilities pertaining to the use are with the lessee.
4. The period of lease must be pre-defined.
5. A jointly-owned property can be leased and the lease rental must be
shared in the proportion of their share in the property.
6. The consideration for the lease (rental) should be determined in the
beginning.
7. The lease period will begin subject to delivery of the leased asset and not
the date of commencement of use of the asset by the lessee.
Ijãrah can be used for a future date, subject to the condition that the pay-
ment of rent will begin only after the delivery of the leased asset. The num-
ber of contracts required in Ijãrah can be limited to one; unlike Murãbaha,
where more than two contracts are needed. For long-term leases the bank
can use a variable rental. This rental can be based on a fixed percentage
increase after a specific period. Another alternative is to break down the
lease period into smaller periods and, after the expiry of the first period,
both the parties can mutually decide the rental for the next period. A third
alternative, which has been proposed by several Islamic scholars, is tying up
the rental with some benchmark rate, such as inflation rate, or tax rate, or
bank rate. For example, the rental can be benchmarked to LIBOR and thus
can be revised periodically. There are several criticisms to such benchmark-
ing, especially to the interest rates, and thus it is highly debatable. Some
scholars consider interest rate benchmarking as similar to lending on inter-
est. To overcome this argument, moderators have suggested that a ceiling on
the highest and lowest rental can be fixed, which removes the Gharar and
thus ensures both the parties clear liabilities. In the case of delayed rental
payments, the lessee can be charged some penalty which should be used by
the lessor for charitable purposes. Insurance remains the responsibility of
the lessor. Transfer of the leased asset after the leasing period can be han-
dled in two ways – Ijãrah ‘thumma al-bai’ (leasing and subsequently pur-
chase) where the hirer uses the goods on payment of lease fees and
purchases the goods from the lessor at an agreed price; and Ijãrah ‘wa
Iqtina’ (lease to purchase), where the lessee is committed to purchase the
PRINCIPLES OF ISLAMIC FINANCE 23
goods at the end of the lease period and the rentals are considered to be a part
of the purchase price. Islamic scholars are divided on this issue. According to
Shariah experts, having two contracts – one for lease and the other for buy-
ing, where one is dependent on the fulfilment of the other – is tantamount to
violation of Islamic principle and hence cannot be permitted. Ijãrah con-
tracts can be securitised and hence can have secondary markets. This is pri-
marily because the lessor remains the owner throughout the period of the
lease. In the case of very high value Ijãrah, the bank can issue Ijãrah certifi-
cates, where each certificate can represent a fixed part of ownership and the
rental can thus be proportionately divided among the certificate holders.
Since these certificates would represent definite ownership, they can be
traded and made negotiable. Hence, Ijãrah certificates can be used as a tool
to manage the liquidity problem, which is a critical issue for Islamic banks.
Example An airline approaches ABC bank to finance the purchase of an
Airbus aircraft costing approximately $80 billion US. The bank arranges to
lease the aircraft to the airline using Ijãrah and securitises it as Ijãrah cer-
tificates. The risks in this transaction are discussed in detail in chapter 2.
Sukûk
Sukûk is also commonly known as Islamic Bonds. Sukûk are asset-backed,
stable-income (fixed and floating), Shariah-compliant trust certificates. A
primary requirement for Sukûk is the existence of the asset on the balance
sheet. Scholars point out that the assets should not represent other Islamic
debt contracts. The AAOIFI §17 describes the Sukûk as, ‘certificates of equal
value representing, after closing subscription, receipt of the value of the cer-
tificates and putting it to use as planned, common title to shares and rights in
tangible assets, usufructs and services, or equity of a given project or equity
of a special investment activity’. Sukûk claims are not on the returns but are
on the ownership. The holder of Sukûk holds ownership rights and responsi-
bilities in relation to the underlying asset. Before undertaking any further
discussions on Sukûk, it is important to note basic differences between con-
ventional bonds and Sukûk. A bond is essentially a contractual debt obliga-
tion, whereas Sukûk is not. Sukûk holders are entitled to share the profits
generated from underlying assets as well as share sale proceeds of these
assets on liquidation. Bond holders do not share the losses, whereas Sukûk
holders are under obligation to also share the losses. When the Sukûk repre-
sents a debt, it is not negotiable; unlike bonds, which are generally nego-
tiable. A strong secondary market exists for bonds, but, at the time of writing
this book, there is no visible organised secondary market for Sukûk.
Globally, Sukûk is used more for sovereign issues rather than
corporate issues; however, this trend is now changing, with more and
24 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
more corporates opting for Sukûk. The Sukûk can be medium- to long-
term and can have fixed or variable rates of return. They are also rated by
international rating agencies like their counterpart bonds. And like bonds
they provide regular income streams and can enjoy capital appreciation.
The model of Sukûk is similar to the conventional bond issue, where a
Special Purpose Vehicle (SPV) is set up to acquire assets and issue finan-
cial claims. There are several types of Sukûk issued according to the pref-
erences of the issuer and investor. AAOIFI has issued a standard for
fourteen types of Sukûk, grouped as tradeable and non-tradeable. The
most common Sukûk at the time of writing this book are: Mushãrakah
Sukûk, Ijãrah Sukûk, Istisnã Sukûk, Murãbaha Sukûk, and Mudãrabah
Sukûk. Nearly half of the Sukûk issued globally during 2004–2006
belonged to the category Mushãrakah Sukûk. In terms of value, the
largest Sukûk until April 2007 was by the Ports, Customs & Free Zone
Corporation, Dubai, for $2.8 billion in January 2006. A large number of
Sukûk have been concentrated in the Middle East, with UAE and Bahrain
accounting for the largest shares. As for maturity, the longest tenure
Sukûk of 14 years has been issued by KSA MBS 1 International Sukûk for
$18 million in July 2006. But most of the Sukûk have less than a 5-year
maturity. A handful of them have been rated by international rating agen-
cies. The market for Sukûk has been expanding rapidly and is now
becoming popular with corporates. Table 1.4 presents a summary of the
top 10 of recent Sukûk issues based on size.
Other than the instruments discussed above, there are several variants
of different products/features available in the market used by banks and
financial institutions as per circumstances. Some of the popular ones are
briefly described below:
Hibah: It is a form of gift used to repay account holders in an Islamic bank.
Current accounts and savings account holders in an Islamic bank do not get
any interest; however, at the end of the year the bank can give some hibah
as a part of its compensation to the account holders.
Musawamah: It is very similar to Murãbaha, except that the seller need not
disclose his cost of goods. All other terms are similar to the Murãbaha.
Qard Hassan: It is a loan on a goodwill basis which is totally free of any
extra cost. The debtor is required to return only the principal borrowed
amount, although he may return anything extra, whatever he feels appropri-
ate. Thus, this is a true Riba-free loan.
Takaful: It is an Islamic insurance. It is one of the oldest concepts in Islamic
finance. It is based on pooling the risks together in order to enjoy what is
called the law of large numbers.
PRINCIPLES OF ISLAMIC FINANCE 25
Table 1.4 Top 10 Sukûk until February 2007 based on size
Issue Issue size
Issuer date Country (US$ Mill) Tenure
PCFC Sukûk 6 Jan United Arab 3,500 2 Years
(Corporate) Emirates
Dubai Global Sukûk 4 Nov United Arab 1,000 5 Years
FZCO Emirates
ADIB Sukûk Company 6 Dec United Arab 800 5 Years
Emirates
SABIC Sukûk 6 Jul Kingdom of 800
Saudi Arabia
DIB Sukûk Company 7 Feb United Arab 750 5 Years
Emirates
Government of Qatar 3 Sep Qatar 700 7 Years
(Govt)
DAAR International 7 Jan Kingdom of 600 3 Years
Sukûk Company Saudi Arabia
Pakistan International 5 Jan Pakistan 600
Sukûk Co. Ltd. (Govt)
Malaysia Global Sukûk 2 Jul Malaysia 600 5 Years
(Govt)
Wings FZCO 5 Jun United Arab 550
(Corporate) Emirates
Source: Data sourced from Liquidity Management Centre BSC (c), Bahrain. Latest data can be
accessed at http://www.lmcbahrain.com/
Wadiah: In this case the bank works as the trustee for funds of customers.
The bank does not guarantee any interest but can give some hibah, which
can compensate the customers.
Wakalah: It is a relationship between principal and agent. When a person
appoints someone else as his agent, he uses wakalah.
SUMMARY
The network of financial obligations running through the chain of financial
intermediaries creates numerous risk exposures. The focus of conventional
26 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
financial institutions has been on the monetarisation of activities of finance
with the aim of improving global standards and having better stability.
Islamic financial scholars have tended to disagree with the conventional
finance due to the inherent social role of finance in Islam. The four main
reasons for the evolution of Islamic finance are: prohibition of Riba, avoid-
ance of Gharar, prohibition of Haram activities in business, and payment of
the Zakat. The focus of Islamic finance has been on welfare, equality, and
justice. Several alternative methods of financing have been developed,
keeping in mind these objectives. Islamic Development Bank (IDB) and
Dubai Islamic Bank (DIB) are among the first few Islamic financial institu-
tions which were set up during the seventies. Several countries have adopted
Islamic finance as the core financial activity, including Sudan, Iran, and
Pakistan. In other countries there have been a parallel run of conventional
and Islamic financial activities, either in the form of full Islamic institutions
or through an Islamic window operated by conventional banks.
Although the basic concepts of accounting remain the same for conven-
tional as well as Islamic banking, there are differences in terms of grouping
of the assets and liabilities. The inclusion of Zakat in the accounting and the
treatment of it has been a debatable issue and they are looked at from vari-
ous view-points. Non-recognition of interest (Riba) in Islamic banking
makes the difference in accounting, since the interest income and expenses
are not accounted for. Due to the recent origin of structured and systematic
Islamic financial activities, the international standards for accounting have
been developed recently by AAOIFI. These standards are under develop-
ment and are not mandatory and hence not widely used.
Profit without participating in the actual business activities is not consid-
ered to be ethical and hence Islamic finance has a strong reliance on the PLS
as a mode of financing. The relationship between lender and borrower
should be of partners and not just creditor and debtor. There has always
been strong insistence from Islamic financial experts on PLS to provide
finance. The use of PLS has been rather limited and debatable. High disclo-
sure requirements, difficulties in monitoring the performance of the
financed projects, as well as limited opportunities for re-investment of
retained earnings, are some of the reasons for the non-popularity of the PLS.
PLS contracts appear to be riskier for banks and thus have been avoided.
There are a large variety of Islamic financial products. A large number of
them have been created by banks in order to meet the local and situational
needs. The more common ones are: Mushãrakah, Mudãrabah, Murãbaha,
Salam, Istisnã, and Ijãrah. Sukûk, the Islamic bond, is gaining popularity
with the corporates as well as governments. Mushãrakah and Mudãrabah
are the two common forms of PLS contracts in Islamic banking. Both
involve partnership of financing institution and client. In both types of con-
tract, the bank needs to provide the capital. In Mushãrakah, the bank may
PRINCIPLES OF ISLAMIC FINANCE 27
not participate in the management; whereas in Mudãrabah it does not have
the right to participate in the management of the business. This basic differ-
ence between the two has implications on the risk exposures of the bank.
Salam and Istisnã are the two exceptions to the forward sale, which is not
permitted in Islamic finance. Salam was originally used for financing the
needs of the farmers. In the case of Salam, the price for the commodity is
paid fully on the spot, but the delivery is scheduled for a future date. There
is no provision for deferred payment in Salam. For manufactured goods and
projects under construction, Istisnã is used, where the payment can be made
in trenches to the contracting party. Under both contracts, the delivery is
scheduled for a future date. Istisnã can be used by the government to build
infrastructure on a BOT basis. Ijãrah is leasing in Islamic finance. There are
two types of Ijãrah, one where the lessee commits to purchase the asset at
the end of the leasing period (known as Ijãrah wa Iqtina) and the other
where he has an option to purchase (known as Ijãrah thumma al bai). Ijãrah
contracts can be securitised and have a secondary market. In the case of
high-value Ijãrah, the bank can issue Ijãrah certificates which can be traded
on the secondary market since they represent claims over an asset. Sukûk is
a very important tool for liquidity management in Islamic finance. It can
either be issued based on debts, in which case they are non-negotiable and
must be held until maturity, or on equity, in which case they are negotiable
and need not be held until maturity. The global Sukûk market is expanding
rapidly, with several large Sukûk issues in the recent past.
NOTES
1. Bankhaus Herstatt in 1974, closed for the business after its banking licencse was withdrawn before
settling its claims to counterparties. This type of settlement risk, in which one party in a foreign
exchange trade pays out the currency it sold, but does not receive the currency it bought, is some-
times called Herstatt risk (Source: Herstatt Risk reference http://riskinstitute.ch/134710.htm).
2. Islamic finance is constantly evolving and has several dimensions and interpretations. We have
attempted to represent the most widely accepted views for all the discussion, with all due respect to
alternativee views presented by different schools of thoughts and different Islamic scholars.
3. Every Muslim must pay a portion of their wealth to weaker sections of the society if his annual
wealth exceeds a minimum level. Zakat is one of the five Pillars of Islam and is obligatory.
4. Almost every ruling of Shariah can be linked to one or more schools of thought, but discussing them
is beyond the scope of this book, and hence the references have been generic and not specific.
5. I. Akkizidis and K. Sunil, ‘Risky Business’, Islamic Business and Finance, (2007), pp. 32–34.
6. A. Habib, ‘Incentive-compatible profit-sharing contracts: a theoretical treatment’, Islamic Banking
and Finance, (Ed) by Munawar Iqbal and David T. Llewellyn, Edward Elgar, UK (2002).
7. M. Iqbal and P. Molyneux, Thirty Years of Islamic Banking: History, Performance, and Prospects
(Basingstoke: Palgrave Macmillan, 2005).
8. M. Abalkhail and J. R. Presley, ‘How informal risk capital investors manage asymmetric information
in profit/loss-sharing contracts’, Islamic Banking and Finance, (Ed) by Munawar Iqbal and David T.
Llewellyn, Edward Elgar, UK.
CHAPTER 2
Risk Management
Issues in Islamic
Financial Contracts
INTRODUCTION
Risks are part and parcel of financial intermediation. The survival and
success of a financial organisation depends on the efficiency with which
it can manage its risks. Risk management is one of the critical factors in
providing better returns to the shareholders. It is also a necessity for
stability of the overall financial system. This chapter is an attempt to
understand various risks associated with Islamic finance in general and
Islamic banking products in particular. Increasing complexity and con-
vergence of financial activities has resulted in multiplicity of risks. The
three most common forms of risks – credit, market and operational,
which occupy the maximum attention of the financial community – are
explained in brief in this chapter. The Islamic financial industry has a dif-
ferent orientation towards risks. The risks are more aligned on the basis
of contract types as a result of the special structuring of the contracts in
Islamic banking. Profit and loss sharing is the nature of some of the
Islamic financial contracts, along with the changing relationships of par-
ties during the lifetime of the contract. These expose them to specific
types of risks. These risks are specific to the contract types. This chapter
thus presents and explains the different types of risks arising from the
Mushãrakah, Mudãrabah, Murãbaha, Salam, Istisnã, and Ijãrah Islamic
financial products, elaborated with the help of graphs and examples.
Furthermore, it highlights how financial institutions that provide such
Islamic financial contracts are exposed to these corresponding underly-
ing risks, as well as how they can manage them.
28
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 29
The risks that are initiated by the Mushãrakah partnership, which are
based on capital and business management contribution between the finan-
cial institution and an agent, are discussed in this chapter. Such joint-venture
contracts that combine two forms of contracts – the Permanent Mushãrakah
and the Diminishing Mushãrakah, according to the participation in the
equity, i.e. permanent or limited – are dealt with extensively. The different
exposures of the financial institutions to credit, operational, market, and liq-
uidity risks, whether the financial institution is employing the Permanent or
Diminishing Mushãrakah contracts, is also presented in this part of the
chapter. Due to some differences in the structuring of the Mushãrakah and
Mudãrabah contracts, an analysis of a Mudãrabah agreement along with
the differences between the two in terms of risk exposures is also presented
in this chapter.
The main structure of the Murãbaha agreement is explained which is
based on the interrelations between the three parties: the financial institu-
tion, the buyer (client), and the seller. Consequently, the financial institution
is exposed to operational risk in the Murãbaha contract because of the
inability of the buyer to keep its promise to purchase the commodity.
Furthermore, credit risk initiated from the buyer defaulting to pay the fixed
repayments together with the consequences of liquidity risk is also presented
and discussed. Finally, the exposure to market risks due to commodity price
fluctuation before and after the financial institution is reselling the commod-
ity as well as due to benchmark or mark-up risks are also highlighted.
This chapter also illustrates various aspects referring to Salam contracts,
where the financial institution makes advance payment, but the delivery
from the seller is set at a future date. The different types of risks, including
credit, operational, market – with regards to commodity price fluctuation
and mark-up risk – and liquidity that financial institutions face by providing
Salam contracts are shown. Different approaches on how to manage the
above-mentioned risks are also featured.
The structure and risk issues of the Istisnã financial contract that involves
the financial institution, the producer (manufacturer), and the buyer/user of
a commodity are also presented in this chapter. The main differences
between the Istisnã and Salam contracts are introduced. Furthermore, this
part of the chapter discusses the main risks initiated by the Istisnã contracts.
These risks, which include operational, liquidity, and reputational, are
explained accordingly, together with their resulting losses on financial and
business institutions. Mark-up risk is particularly discussed, as it can be the
main factor for additional losses due to market price fluctuations of the
commodities and constructed assets that are exchanged in the Istisnã con-
tracts. Strategies and techniques on how to minimise the risks arising from
the Istisnã financial contracts are proposed.
30 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
The leasing financial contract called Ijãrah is also presented in this chap-
ter. The different types of risks involved in Ijãrah contracts are featured.
Thus, the exposure of financial institutions to credit risk due to the lessee’s
defaults is presented. Operational risks due to catastrophic events as well as
any damages to the assets on the return (maturity) date are related to the fact
that the financial institution continues to own the assets in Ijãrah contracts.
Furthermore, the exposure of the financial institutions to market risk, which
is initiated from the price fluctuation of the commodities in Ijãrah contracts,
is illustrated. Finally, there is a discussion concerning additional credit and
market risks that arise when the payment is set to the maturity date.
Sukûk, the financial vehicle for Islamic institutions to raise funds, will be
discussed in this chapter. Due to the availability of a large variety of Sukûk
in the market, the discussion on risk management is beyond the scope of this
book and hence has not been dealt with.
In addition, graphical representations showing all cash flows in and cash
flows out of the Islamic financial contracts together with their associated
risks and losses during their lifetime are illustrated. Note that, in the figures
showing the different Islamic contracts, the directions of the arrows (up and
down) represent the cash-out and cash-in or delivery-in, respectively.
Meanwhile, the triple line arrows facing up represent the associated losses
that are covered by the institution. Note, finally, that the analysis is always
performed from the point of view of the financial institution.
As nowadays, the structure of the Islamic contracts is updated and
enhanced with more elements and conditions; institutions that are providing
Islamic financial contracts should be able to identity the types of risks that
they face by employing such contracts. The aim of this chapter is to show
generic means to break down the Islamic financial products in a way that
can be analysed in terms of the risks that may arise during their lifetime.
Institutions may provide more advanced or updated contracts, based on
those that are presented in this chapter (Mushãrakah, Mudãrabah,
Murãbaha, Salam, Ijãrah, and Istisnã), and thus they must be able to iden-
tify the corresponding financial risks. Guides to these identifications are
presented in this chapter.
AN OVERVIEW OF FINANCIAL RISKS
Financial intermediation creates a series of financial inter-linkages, which
results in financial obligations. Individual economic units are interconnected
to each other through the money market intermediaries. There can be disrup-
tions at this level which can create uncertainty and hence the risk. These indi-
vidual entities are linked to each other via local money market. Local money
market contributes its own share of fluctuations and disturbances. Due to the
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 31
global nature of trade, local money markets are connected to each other. The
micro- and macro-economic factors on the global level further add to the
uncertainty. Since all the financial entities are directly or indirectly inter-
woven, interlinked, and interrelated, they create a complicated maze of
uncertainties which makes up the mass of the financial risk. Something
which can be predicted accurately generally does not involve risk; it is
unpredictability which is the risk creator. The higher the degree of uncer-
tainty, the higher is the risk. Risks should not really exist in a market where
there is perfect information symmetry. But this is wishful thinking.
Information asymmetry is an integral and unavoidable part of the market.
Asymmetry leads to uncertainty, which is one of the causes of risk. Events
which are certain are generally out of the purview of risk. Multiplicity of
outcomes with an uncertain chance of occurring results in risks. If a coin
has both the sides marked as ‘heads’ it poses no risk as the result is certain.
The choice of two alternatives, ‘heads’ and ‘tails’, with no certainty of the
happening or non-happening of a specific event, result in a risk element.
The risks cannot be eliminated, they can be managed. The element of risk
also brings opportunities, and, to gain from these opportunities, the risks are
required to be managed efficiently. Banks, due to their typical nature, carry
special risks. For a bank, some of the risks can turn into losses and may even
cause liquidation problems. A risk is in many cases hidden before it is
visible as a loss. Risk and return are usually correlated. A bank with a con-
servative approach may not utilise the funds fully and thus have a higher
cost of capital, whereas a bank with huge risk appetite can over-lend,
thereby increasing the chances of a failure. At the same time, pricing the
loans is also largely based on risk. A risky loan which is under-priced may
prove to be a drag on the profitability, whereas a sound loan which is over-
priced may shy away good customers, leading to loss of business.
At the level of the enterprise, the risks can be grouped into financial,
business, and operational risks. Financial risks will generally include credit,
market, and liquidity risk. Business risk is a combination of management
risk and strategic risk. The operational risk can arise due to people,
processes, systems, as well as several other factors. Some of the other rele-
vant risks for the financial industry can be commodity risk relating to the
changes in the prices of concerned commodities, country risk relating to the
changes in the country rating, equity market risk relating to changes in the
value of equity investments, reputational risk relating to the adverse events
which can affect the organisational reputation, legal risk relating to issues
related to legislations, political risk relating to adverse changes in political
situations, concentration risk relating to grouping of several unfavourable
risks together, regulatory risk relating to non-compliance to regulatory
requirements, and systemic risk related to interconnected unfavourable
events across the industry. For the purpose of this book, the three major
32 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Types of risks
Financial Business Operational
Market Managerial People
Credit Strategic Process
Liquidity Systems
Figure 2.1 Major types of risks in the financial industry
risks – credit risk, market risk, and operational risk – will be considered in
the discussion with reference to other risks, e.g. liquidity risk, wherever
necessary. Figure 2.1 presents the more common categories of risks.
With the fast-changing landscape of financial services in banking and
financial sectors in the last three decades, there is a shift of focus on the
development of Islamic banking. The original issue in the seventies of
developing an interest-free financial system is no more of prime importance
for Islamic bankers. The core issue is currently to develop an Islamic finan-
cial system which does not suffer from the weaknesses of the conventional
system. Thus, the focus has suddenly shifted to liquidity management, risk
management, and common standards in Islamic finance.
Credit risk
Credit risk is simply defined as the potential that a borrower or counterparty
will fail to meet its obligation in accordance with agreed terms.1 It is the risk
of failure of the counterparty to honour their commitment, and is also
referred to as default risk. This arises from the inability of the counterparty
to service the debt on agreed terms. It can also arise when the solvency or
the credit rating of the counterparty changes adversely. In Islamic banking
there is a limited availability of credit ratings defined from external agen-
cies. This book therefore is more emphasizing in the direct credit risk from
the counterparty rather on the risk of changing in the credit ratings. Credit
risk is one of the earliest recognised risks in the financial industry. During
the early eighties, a large part of a bank’s profit came from lending busi-
nesses; hence, the focus was primarily on credit risk. Credit risk exists for
complete portfolio as well as individual accounts. It is important for banks
to recognise such risks since a few large counterparty failures can lead to
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 33
serious consequences. Credit risk cannot be accurately calculated before the
event since the likelihood of default is highly uncertain and thus is difficult
to predict accurately. Although there are developments in the calculation of
credit risks, the major difficulty remains with the availability of the data.
Several standard measures have been developed over the years to calculate
the credit-worthiness of the project and the client. Rating agencies have
played a significant role in standardising the understanding of credit risk,
although they have failed to predict major failures. Rating agencies such as
Moody’s, Standard & Poor’s, and Fitch have developed elaborate techniques
to rate projects and parties. Ratings are based on several factors and rating by
one or more of these rating agencies is generally available to large issues or
issuers. Due to the absence of rating facilities for small customers and proj-
ects, banks have their own in-house-developed rating standards, which are
used to rate their customers and the projects. Ratings are a relative measure
of risk and hence not largely useful for accurately measuring the probability
of default. The first challenge is modelling the probability of default directly
with credit risk models and the second is capturing portfolio effect in credit
risk measurement.2 In the case of credit risk, controls are required at credit
portfolio levels. The risk categories for credit risk include different levels and
segments of credit such that the maximum total loss in case of credit risks is
limited to the total exposure of the credit volume. In some cases credit risk can
be related to other risks, such as market risk and operational risk. In transac-
tions involving cross-border trade, credit risks cover country risks as well as
political risks. Credit risk can also acquire the dimension of concentration risk
when counterparties having similar risk profiles group together.
Market risk
Market risk is the risk of losses in on- and off-balance-sheet positions aris-
ing from movements in market prices, interest rates, FX rates, and equity
values3 where these are the main four market risk factors. Market risk was
recognised in the late eighties after the increasing importance of stock mar-
kets when banks started investing heavily into securities. The exposures
grew and thus the discomfort of possible losses from changes in underlying
prices. Since banks deal in several currencies across different markets in the
world, they are exposed to the fluctuations in national and international
interest rates. Banks are also exposed to changes in foreign exchange posi-
tions, which can apply to transactions as well as assets which are denomi-
nated in foreign currencies. Foreign exchange rate risk can also arise from the
unfavourable revaluation of the trading currencies. Investment in equities by
banks is exposed to changes in the equity prices. Finally, commodity price
changes, where the bank has some interests, can affect the position adversely.
Market risk is difficult to measure due to diversified portfolios, since it will
34 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
consist of several markets, currencies, indexes, and instruments. The larger
the diversification of the portfolio, the more difficult it is to accurately estimate
market risks due to the correlation between different risks. There are several
ways to measure and manage market risk. Most of the banks have limits and
triggers for foreign exchange transactions. Factor sensitivities and VaR can
be used for marked-to-market trading. The market is full of complex deriva-
tive products for hedging the positions to manage the risk. Stress testing and
other financial models are used for better understanding of market risk. The
maximum loss method has been developed to estimate risk for non-linear
portfolios. In the case of market risk, the controls are required at the trading
desk and portfolio level. The risk categories for market risk include differ-
ent interest rate categories, various foreign exchange markets, and different
market commodities. The maximum total loss in case of market risk is
limited to the market value of the security.
Operational risk
Operational risk is defined as the risk of loss resulting from inadequate or
failed internal processes, people, and systems or from external events. This
definition includes legal risk, but excludes strategic and reputational risk.4
Operational risk has been recently recognised and has been gaining promi-
nence among risk-related research. The three major components of opera-
tional risk are people, processes, technology, or some other external events
(usually catastrophic). People’s risks include human errors, lack of expert-
ise, compliance, and fraud. Process risks include risks related to different
aspects of running a business, which may include regular business
processes, risks related to new products and services, inadequate/insufficient
control, etc. Failures related to systems are included in technology risks.
Operational risk is rather difficult to measure and manage. Often, these risks
only become apparent once a problem arises.5 The recent focus of risk
experts has been operational risk. The wide range of activities included in
operational risk makes it difficult to apply a standard model to all organisa-
tions and hence there is a lack of universally accepted standard models.
While the industry is far from converging on a set of standard models, the
banks have developed or are developing models relying on a similar set of
risk factors. Those factors include internal audit ratings or internal control
self-assessments, operational risk indicators such as volume, turnover, or rate
of errors, loss experience, and income volatility.6 The quantification of
various loss events related to a wide range of activities is a difficult task,
which is currently under exploration by the risk community. A daunting task
is to classify the loss events in operational risk and several such classifications
are available. Basel II guidelines (see chapter 3 for details) has published a
scheme which can be taken as a foundation to develop it further based on the
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 35
individual needs of the organisation. The most ideal and simplified ways of
controlling operational risk can be internal audits and internal controls. In the
case of operational risk, the controls are required at business process levels.
The risk categories for operational risk include different business lines and
loss event types (for more details on loss event types see chapters 3 and 6).
Operational risk is becoming a critical factor in risk management since
the maximum total loss in case of operational risk can extend to the bank’s
liquidation value.
The role of information in risk management
It would be impossible to monitor, measure, manage, and mitigate the risk
without sufficient, timely, and accurate information. The monitoring of the
cash flows and the calculation of credit risk, market risk, and operational
risk heavily depends on the appropriate information systems and availabil-
ity of information. Information collection, processing, and preserving plays
an important role in all stages of risk management (Figure 2.2). Information
is generated at all stages of activities in a financial institution and collecting
it at the point of origin will ensure accuracy and reliability. Credit risk is
heavily dependent on information since the assessment of credit-worthiness
of project or client depends on the information available for processing. In
market risk, liquidation analysis, and ALM (Asset Liability Management)
the information referring to financial contracts is playing a key role in the
estimation of the current and future cash flows. Operational risk also relies
Collection at
Compilation
the point of
origin
Information cycle
Monitoring, in organisation
verifications,
and updating Processing
Preserving
Figure 2.2 Information management in organisations
36 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
heavily on collection of information in regards to the performance of
process, systems, and people within the institution. Loss data bases are used
in evaluating all types of risks. Stress and back-testing are fully dependent
on information data and systems. Capital allocation required by an institu-
tion is related to its risk exposures, and since assessing risk exposures is
related to information availability, efficient allocation of capital depends on
information management. Proper information will help the bank in unpacking
risk components and allocating capital efficiently.
Another dimension of risks can be the interaction and mutation of risks.
Some of the risks can work as an initiator for other risks. Sometimes the
risks combine with each other, thus a new risk emerges. For example, the
risk on investments consists of credit risk as well as market risk. A change
in the value of the investment is a market risk, whereas downgrading of the
investment by rating agency will involve credit risk. Similarly, the inability
to manage market risk can be considered as an operational risk rather than a
pure market risk. To what extent this needs to be allocated using market risk
methodology and operational risk methodology is complicated to deter-
mine. The inability to monitor and manage the different types of risks can
be considered as a failure of people and hence in some form can contribute
to operational risks. This is a grey area of risk management and additional
research is required. While allocating capital to manage risks, this merging
of risks can cause duplicate allocations and thus finally increase the capital
allocation. The management of overlapping of capital allocations across
different risks is an area requiring further probing.
IDENTIFYING RISKS IN ISLAMIC FINANCE
A common perception about Islamic banking is that it is safer since it is not
based on interest rates. Also, there is an argument that several of its products
are using a mark-up arrangement, and hence they carry less risk. Both these
points of view are not entirely correct and are understatements. Although
Islamic banks are not directly affected by fluctuations in interest rates, they
are indirectly affected through lease, mark-up, and deferred sale. As discussed
in the previous chapter (chapter 1), Islamic financing relies more on equity
financing rather than debt financing and hence inherently is riskier. There are
several reasons why Islamic banking can be riskier than conventional bank-
ing. Sundararajan and Errico7 point out several of them, including the specific
nature of risk faced by Islamic banks and the virtually unlimited number of
ways to finance a project using either PLS or non-PLS contracts. Along with
this, the Shariah rulings can create a large variety of contracts. PLS contracts
fundamentally increase the risk as they are difficult to monitor. The lack of
standardisation due to the availability of a large number of ways to finance is
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 37
another factor that renders Islamic banking riskier. There is no legal caveat to
control the relationship with the entrepreneur, especially when financed
through Mudãrabah, although the bank can exercise some control over the
working of the enterprise if financed through Mushãrakah. Added to this,
scarcity of hedging instruments, underdeveloped inter-bank money markets,
and a market for government securities which are Shariah-compliant, make
the Islamic banking more vulnerable to unfavourable events. Finally, liquidity
management in the absence of a short-term money market for Islamic finance
adds to the difficulty. Although the liquidity risk faced by banks at present
seems to be low, there are several reasons which may lead to liquidity prob-
lems in the future.8 First, they rely largely on a current account for liquidity.
Second, there is a restriction on the sale of debts. Third, the market for short-
term Islamic instruments is not developed. Fourth, the lender of last-resort
facilities is not available. Another dimension to the high risk is due to the
nature of PLS contracts, since they are based on the profitability of the proj-
ect rather than the credit-worthiness of the borrower.9
The three major risk categories as discussed above – credit risk, market
risk, and operational risk – are all present in Islamic financing in differing
degrees, as compared to conventional financing (Figure 2.3). A quick
comparison of different risks in Islamic finance with conventional finance
reveals that credit risk, commodity risk, liquidity risk, market risk, legal risk,
and regulatory risks are higher in Islamic financing. These risks in Islamic
finance exist with different intensities and have several dimensions (Table 2.1).
Since the bulk of Islamic financing goes to trade finance and mark-up
arrangements, it is susceptible to the adverse fluctuations of the commodity
prices. Also, Islamic banks do not have access to cheap credit and hence have
to manage their own liquidity. In its efforts to manage the liquidity, banks may
adopt an over-cautious approach by following a conservative route to invest-
ments and funds management which will undermine the efficient allocation of
available funds to the profitable channels. On the other hand, in the case of ill-
managing the liquidity, there can be a banking crisis, which may even result
in sudden cash flight and liquidation. There are no universally accepted stan-
dards for reporting for Islamic finance, except AAOIFI (the Accounting &
Auditing Organisation for Islamic Financial Institutions), which is still under
refinement and is not mandatory and hence still not used by several institu-
tions. This renders the reports by Islamic financial institutions non-comparable.
An entrepreneur having a financially sound project will generally not come
to an Islamic bank due to a PLS contract; his preference in this case will be
a conventional bank. Thus, entrepreneurs having projects which are less
profitable will approach Islamic banks to finance. Some of the unique risk
factors relating to the Islamic mode of financing are liquidity-originated mar-
ket risk, transformation of credit risk to market risk and market risk to credit
risk during different stages of the contract, bundling of credit risk and market
38 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
0
2 2 2 2 2 2
Risk intensity
1 1 1 1 1 1 1
Credit risk Market risk Liquidity risk Interest rate Legal risk Currency risk Commodity Regulatory
risk risk risk
Conventional banking Islamic banking
Figure 2.3 Risk profiling – conventional vs Islamic banks
risk, market risk arising from ownership of non-financial assets, and treatment
of defaults.10
Some of the factors affecting credit risk in Islamic banks are:11
Less sophistication in risk management practices
Incentive to unscrupulous clients to default due to leniency towards
default
Lower risk due to short-term financing
No (except in Malaysia) trading book exposure
No access to credit derivatives
According to OIC Fiqh Academy, Murãbaha is binding only to the seller
The usual counterparty risks, such as failure to supply/deliver on time,
failures regarding the quality and quantity
That Istisnã poses dual risk
The default of clients and non-performance of sub-contractors, and
some Fiqh does not allow Ijãrah ending in ownership.
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 39
Table 2.1 Risks in Islamic financial services
Type of risk Coverage
Credit risk Attributed to delayed, deferred, and default in payments by
counterparties. Covers profit-sharing contract (Mudãrabah
and Mushãrakah), receivables and lease (Murãbaha,
Diminishing Mushãrakah, and Ijãrah), and working capital
financing (Salam, Istisnã, and Mudãrabah). Covers different
stages of a contract.
Market risk Attributed to adverse movements in interest rates,
commodity prices, and foreign exchange rates. Covers
commodity risks existing in Murãbaha and Ijãrah contracts.
Equity risk Attributed to adverse changes in market value (and liquidity)
of equity held for investment purposes. Covers all equity
instruments (Mudãrabah and Mushãrakah).
Liquidity Risk Attributed to adverse cash flow in situations arising mainly
out of changing market risk exposures, credit risk exposures,
and operational risk exposures.
Rate of return Attributed to changes in account holders’ expectations of the
risk return on investments. Also related to fluctuations in returns
due to changes in underlying factors of the contract.
Operational risk Attributed to the inadequacy of failed processes, people, and
systems. Also includes risks arising from Shariah
non-compliance.
Legal risk Attributed to the inadequate legal framework, conflict of
conventional and Islamic laws, and conflict between Shariah
rulings and legal decisions.
The role of information in the risk management of Islamic financial insti-
tutions can be more critical compared to conventional financial institutions.
The nature of contracts in Islamic finance involves more integration with
the activities of the entrepreneur. The PLS contracts are heavily biased
towards availability of information for managing the risks. In the case of
Mushãrakah and Mudãrabah contracts, there is a heavy bias towards avail-
ability of information for risk management. Whilst using Istisnã, a bank
needs to maintain a steady flow of information about the contractors.
Similarly, the sustainability of the activities of the Islamic financial institu-
tion depends on the sharing of information with the shareholders and their
approval of these activities through Shariah. Sharing of Shariah rulings
with the shareholders is an integral part of activities. Islamic finance has a
special emphasis on transparency in the conduct of activities, calculation of
40 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Table 2.2 Comparative information requirements – conventional vs
Islamic banks
Nature of information required to be shared
User/source of
information Conventional banks Islamic banks
Shareholder Accounting, significant Accounting, significant
management policies, critical management policies,
decisions. critical decisions, business
activities, details of major
sources of income and
investment, Zakat details.
Clients/customers Credit-worthiness, ratings, Credit-worthiness, ratings,
financial. financial, entrepreneurship
abilities, selected personal
information.
Financed projects Feasibility reports before the Feasibility reports before
agreement for financing. the agreement for financing.
During the financing period,
regular cash flows, income
statements, overall conduct
of business. At the time of
termination of financing
(or partnership) contract,
balance sheet, income
statements.
Central bank Statutory reports Statutory reports, Special
reports needed for Islamic
banking.
Shariah Not applicable Project details, client
details, business lines
details, method of financing,
treatment of defaults.
Other Industry-supportive Industry-supportive
banks/institutions information, a few details information, details of
about line of businesses. ethical and Shariah
approved conduct of
business.
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 41
profits and losses, as well as the nature of activities. Transparency depends
on information. Due to the focus on social responsibilities, information
management acquires a special dimension in Islamic financial institutions,
more than a mere statutory compliance. Hence, special efforts are required
for efficient information management.
RISK OF NON-COMPLIANCE WITH
SHARIAH RULES
Shariah is the body of Islamic law. The term means ‘way’ or ‘path’; it is the
legal framework within which the public and some private aspects of life are
regulated for those living in a legal system based on Muslim principles of
jurisprudence. Shariah deals with all aspects of day-to-day life, including
politics, economics, banking, business law, contract law, family, hygiene,
and social issues.
The risk of non-compliance with Shariah rules is referred to as
Shariah risk. This can be included in operational risk, as non-compliance
can lead to reputational damage, which can trigger an exodus of funds
from Islamic investors, causing failures and systemic risks. There are
several variations possible in Shariah rulings, which give rise to a high
degree of uncertainty. The conventional financial market is reasonably
matured in terms of availability of products for financing. A large num-
ber of highly matured products are available for retail and corporate
financing. A constraint of Islamic finance has been the non-availability of
parallel products as compared to conventional finance. With the growing
market for Islamic finance, innovative products are being launched,
which have created some unique solutions for Islamic financing. Some of
these products may not fully comply with Shariah rules and are subject
to Shariah scrutiny and thus can be considered to be an adaptation based
on current market needs. These products pose a special risk in terms of
being rejected under the scrutiny of Shariah rules. As interpretations can
differ, utmost precaution needs to be taken before entering into any con-
tract. An approval from the Shariah council should be obtained. There
should not be any deviations in the contract once the Shariah council has
approved a particular contract.
MAIN ELEMENTS USED IN FINANCIAL
R I S K A N A LY S I S
There are four main elements (as illustrated in Figure 2.4) that should be
well defined and considered in financial risk analysis for both conventional
42 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Behaviour Behaviour
Contracts
Markets Counterparties
Behaviour
Figure 2.4 Main elements considered in financial risk analysis
and Islamic financial products; these are:
The construction of the financial contracts
The identification of the markets
The identification of the counterparties
The behaviour and interaction of the above two within a time period
(past-current-future), mapped into the financial contracts.
The construction of the financial contracts is mainly based on the cash
flows (ins and outs). The ins and outs cash flows are illustrated in the repre-
sentation of the lifecycle of the contracts, (presented in the following
paragraphs), as arrows with down and up directions respectively. The identi-
fication of the markets and the counterparties is a complex process that is
based on many tangible and non-tangible factors that affect their associated
behaviours. This chapter is mainly focused on the Islamic financial contracts,
where their construction is based on the characteristics of markets, counter-
parties and operations. The aim of this chapter is to show to the reader how
to identify and analyse the different types of risks that arise within the life-
time or type of the Islamic financial contracts. Therefore, the guides illus-
trated in the following paragraphs can be applicable to any new or integrated
types of Islamic products by focusing on the financial events and the inflows
and outflows expected cash.
MUSHÃRAKAH CONTRACTS OF
PA R T N E R S H I P A N D F I N A N C I A L R I S K S
The Mushãrakah contract combines the acts of investment and manage-
ment. The Mushãrakah is a partnership (joint venture) contract where two
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 43
Investment
Capital Capital
P&L
Bank Partner
Figure 2.5 Structure of Mushãrakah partnership joint venture contract
agreements
parties, the financial institution and the business partner, are combining
their capital in an investment to share the profits and losses whereby they
have similar rights and liabilities. The structure of the Mushãrakah contract
is illustrated in Figure 2.5.
Note that in general the term Mushãrakah is used to define partnerships
in respect to the levels of the partners’ authority and obligations and the
types of their contributions, which could be management, practical skills,
goodwill, etc. However, in this book, for the financial Mushãrakah contract,
the partnership that is based on the capital contribution is considered and
discussed. Mushãrakah is a contract arrangement that works as follows:
The financial institution together with other partners provides the necessary
capital for a particular project. The contributions of the partners under this
mode may be equal or unequal percentages of capital for the purpose of
establishing a new income-generating project or to participate in an existing
one. All partners, including the institution, have the right to participate in
the project, i.e. on the management and decision-making level. They could
also waive this right according to the agreement. The profits are to be dis-
tributed according to an agreed ratio, which needs to be decided at the time
of signing the agreement and can be different to the capital ratio. Moreover,
losses are shared in the proportion in which the different partners have
provided the finance for the project.
Types of Mushãrakah contracts
The Mushãrakah contracts may take two forms:
Permanent Mushãrakah contracts
In the case where the Permanent Mushãrakah contract is applied, the finan-
cial institution participates in the equity of the company that is investing and
44 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
receives an annual share of the profits and losses on a pre-rated basis.
During the period where the financial institution will participate in this
investment, profit and loss sharing is intended to continue until the company
is dissolved.
Financial institutions provide Permanent Mushãrakah contracts for
income-generating projects. As aforementioned, financial institutions provide
capital for a project in exchange for ownership and profit sharing in the
proportion agreed upon between the two partners. Moreover, the financial
institution may leave the responsibility of management to the partner and
retain the right of supervision and follow-ups.
Diminishing Mushãrakah contracts
When the financial institution agrees on Diminishing Mushãrakah contracts:
– Its intention is to stay in the partnership for a limited time
– Its share of the equity is progressively reduced and the partner even-
tually becomes the full owner
– The partner generally buys the shares by payments on an instalment
basis
Note, however, that until the financial institution participates in the
equity of a company, it receives a share of the profits on a pre-rated basis.
Risks identification in Mushãrakah contracts
Credit, operational, market, and liquidity are the main risks that finan-
cial institutions are exposed to by participating in both Permanent and
Diminishing Mushãrakah contracts.
Risk exposures when dealing with Permanent Mushãrakah contracts
involve: In Permanent Mushãrakah contracts, the financial institution has a
partnership in the business; thus, any external event or in general any
inadequate activities or failures due to business risks that cause losses will
initiate an exposure to operational risks as illustrated in Figure 2.6, points
1 & 2. Note that the financial institution that engages itself in Mushãrakah
contracts has to participate in sharing both the losses and the profits. The
agreed ratio of profit sharing can be different to the capital ratio, whereas
the loss-sharing ratio must be the same as the capital ratio, as explained in
Permanent Mushãrakah in chapter 1 (profit sharing ratio 1:4, loss sharing
ratio – capital ratio – 2:3). In this situation, a financial institution has an
unfavourable spread of profits and losses.
In the business partnership concerning Permanent Mushãrakah contracts,
the financial institution is also sharing the business profits; however, the
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 45
Market risk P and L
fluctuations
Profit and loss sharing Last equity
payment and/or
sum of payments
Loss
Profit
Profit
Loss
Loss sharing
Major loss
Cycle angle income
Investment principle
Balance sheet
fluctuations
Operational risk Inability to carry on
business
Business risk Credit risk
Liquidity risk
1
2
Figure 2.6 Credit, operational, market, and liquidity risks during the lifetime of
Permanent Mushãrakah contracts
business may default to provide the expected cash due to the above-
mentioned risks and losses. Such cases give birth to credit risk exposure as
shown in Figure 2.6, points 1 & 2.
As a result to the above-mentioned credit inability concerning Permanent
Mushãrakah contracts, the financial institution may face an exposure to
liquidity risks as it may not be able to provide enough cash for its other
investments and activities.
Finally, any major losses on Permanent Mushãrakah contracts may cause
inability for further continuation of the business as illustrated in Figure 2.6,
point 2. Such an event may result in a ‘last equity’ payment that will most prob-
ably have a market price lower than the initial nominal one. In this case, the
financial institution is exposed to market risk as shown in Figure 2.6, point 3.
Risk exposures when dealing with Diminishing Mushãrakah contracts
In Diminishing Mushãrakah contracts, the financial institution is selling its
46 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
equities to the partners by payments on an instalment basis, as illustrated in
Figure 2.7. In the event that the partner is unable to buy the equities on the
pre-fixed price, due to business failures, the financial institution is eventually
exposed to operational risks (see Figure 2.7, point 1).
The above-mentioned default on the expected payments from the part-
ner(s), illustrated in Figure 2.7, point 1, results in a credit risk exposure in
financial institutions that participate in Diminishing Mushãrakah contracts.
Credit and operational risks in Diminishing Mushãrakah contracts result
in losses and variations to the financial institution’s cash expectations.
When this is the case, there is an additional exposure to liquidity risks (see
Figure 2.7, point 1) due to the bank’s inability to provide the associated cash
for any further investments and planned activities.
In Diminishing Mushãrakah contracts, the price of the equity is fixed and
thus any mismatch between the actual market price and the fixed one initi-
ates loss of potential profit. In this case, the financial institution is exposed
to market risk, as illustrated in Figure 2.7, point 3.
Both Permanent and Diminishing Mushãrakah contracts consist of
partnership type of contracts and therefore the danger of transparency risks
should be considered by the financial institution.
Market risk
Share price fluctuations
Market price
Initial investment + profit
Payment instalments
=
Payment inability
End of partnership
Cycle angle selling
Operational risk
shares
Credit risk
Liquidity risk
Figure 2.7 Credit, operational, market, and liquidity risks during the lifetime of
Diminishing Mushãrakah contracts
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 47
Operational risk management in
Mushãrakah contracts
In both Permanent and Diminishing Mushãrakah contracts, operational
risks are mainly initiated due to business risks. Financial institutions that
have rights in management of such business partnerships may participate in
and/or monitor the process of the business and thus minimise the associated
risks. Moreover, insurance policies could be applied for covering major
losses that can be initiated due to external events.
Credit risk management in Mushãrakah contracts
Similarly to operational risk management, financial institutions can
minimise credit risks in Permanent Mushãrakah contracts by being
involved in business management activities and/or monitoring the balance
of the business profits and losses. Moreover, the sale of last equities is a type
of guarantee for minimising the losses from such credit risk exposures.
Finally, financial institutions may minimise credit risks in the Diminishing
Mushãrakah contracts by having the rights to sell their equities to a third
party with the approval of the Shariah committee.
Market risk management in
Mushãrakah contracts
Financial institutions should define a worthy strategy in case of market risk
in Permanent Mushãrakah contracts where the stop loss should be clearly
defined for selling the last equity price. On the other hand, for the market risk
in Diminishing Mushãrakah contracts caused by market fluctuation, static
and dynamic analysis can be applied to estimate the current and future Value
at Risk (VaR) and evaluate the significance of the market risk exposure.
For minimising both credit and market risks, financial institutions that
are participating in Diminishing Mushãrakah contracts should set the
payment for the equity sale to the partner on several preset instalments.
Liquidity risk management in
Mushãrakah contracts
In Mushãrakah contracts, the liquidity risk is a result of the other risks.
Financial institutions may avoid facing such risks by either managing the
source of the risk or by reserving additional capital.
Example – Permanent Mushãrakah By continuing with the example from
chapter 1 and as illustrated in Figure 2.6, points 1 and 2, a bank is
exposed to operational risk through business risk arising out of any external
48 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
events, inadequate activities, or failures. In case there is a sudden fall in the
market for the ceramic tiles, the bank is exposed to operational risk. If the
partner (customer) fails to provide the required profit it may give rise to
credit risks as illustrated in Figure 2.6, points 1 and 2. In the case of major
losses, the continuity of the business may be affected and may expose the
bank to market risk as shown in Figure 2.6, point 3. The bank is already in
an adverse position due to unequal sharing in profits (1:4) and losses (2:3).
Example – Diminishing Mushãrakah Diminishing Mushãrakah has a
different set of risks. By continuing the example from the previous chapter,
a bank enters into a Diminishing Mushãrakah with the customer where it
contributed 80% of the cost of the delivery van and the customer provided
the remaining 20% (4:1) and, as discussed, the capital contribution is
divided in equal units of 20; thus, the bank owns 16 units and the customer
owns 4 (4:1). Every two months the customer keeps purchasing one unit of
ownership from the bank, thus increasing his contribution and reducing the
bank’s ownership. After 33 months, the customer will own all the units of
the bank’s contribution, thus owning the delivery van all by himself. If the
partner fails to acquire the units from the bank, the bank is exposed to oper-
ational risk as shown in Figure 2.7, point 1, which in turn may also expose
the bank to credit risk. The change of ownership pattern poses different
risks during different phases of the contract. Figure 2.8 shows the way in
which changing relationships affect the risk pattern.
Before During financing After 32 months
financing period of (financing
the asset the asset period over)
Ownership Financing
percentage Bank period Bank Customer Customer
1st month 80% 20%
3rd month 75% 25%
5th month 70% 30%
100% … … … 100%
29th month 10% 90%
31th month 5% 95%
33rd month 0% 100%
Figure 2.8 Ownership pattern and risk in Mushãrakah contracts
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 49
Before the Mushãrakah contract is entered into, the risk is related to the
asset price, i.e. the price of the delivery van. During the financing (contract)
period, the three main risks faced by banks are credit risk, fluctuations in the
value of delivery van (market risk), and operational risks, which includes
Shariah risk. Interruptions in the future cash flows from the project result-
ing from sharing profits and sales of the partnership share may result in
liquidity risk. As the ownership pattern is dynamic during the period of
financing, the risk exposure also becomes dynamic.
MUDÃRABAH CONTRACTS OF
PA R T N E R S H I P
The Mudãrabah contract is a type of business venture agreement. In Islamic
finance, Mudãrabah is a special kind of partnership of capital and entrepre-
neur, and is therefore considered as the cornerstone of the Islamic financial
intermediation.
The principal structure of the Mudãrabah financial contract is based on
the following two key points:
The contract is an agreement between two parties in which the first
party provides capital (the financial institution) and the other party (agent)
provides the expertise with the purpose of earning profit which will be
shared at a percentage mutually agreed upon.
Any losses from the implementation and lifetime of the Mudãrabah
contract are fully covered by the provider of the capital (the financial
institution).
Figure 2.9 illustrates the flow of the structure for the Mudãrabah contracts.
Investment
Capital Expertise
Profit
Loss
Bank Agent
Figure 2.9 Structure of Mudãrabah contract agreements
50 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Financial institutions that agree on Mudãrabah contracts count on the
abilities and expertise of the agent to run a profitable business. On the other
hand, the agent commits that he will provide the best of his know-how and
experience in balancing the capital that has been invested by the financial
institution for the particular business or project. As a result, by entering into
a Mudãrabah contract, the two parties complement each other, allowing a
business venture to be financed. Mudãrabah is also used to mobilise funds
through savings and investment accounts.
Risks identification in Mudãrabah contract of
partnerships
Financial institutions that are dealing with Mudãrabah contracts are
exposed to operational, market, and liquidity risks. The analysis of risk
identification in a Mudãrabah contract is split into two periods: a) during
the investment lifetime of the agreement, and b) during the time of Profit
Sharing and Covering Losses, if any, of the contract.
Risk issues during the investment period of the Mudãrabah contract consist of:
As illustrated in Figure 2.10 (points 1 and 2), during the investment
period of the Mudãrabah contract of partnership, the financial institution is
exposed to operational risk. Operational risk may arise due to external
events, including catastrophic ones as well as internal business failures.
Such events cause high disruptions for business development and result in
losses (minor to major) which the financial institution is compelled to cover
fully.
As a result of covering the above-mentioned losses, the financial
institution is exposed to liquidity risk. This is due to the fact that the finan-
cial institution has to cash out a capital above its expected lines and plans
and thus it may be unable to fulfil the other financial obligations (i.e. pro-
viding cash for other Mudãrabah financial contracts).
Major losses may result from the inability of the business partner
(or agent) to carry on the business development and/or the project’s imple-
mentation. In this case, the financial institution is facing an additional future
liquidity as well as credit risk due to the defaults from the partner (agent) for
providing the expected (future) cash flows.
Risk issues during the profit and loss period of the Mudãrabah contract
consist of: After the initial investment period, the Mudãrabah contract is
expected to give back financial returns (profits). However, the Mudãrabah
contract further exposes the financial institutions, who are the financial
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 51
Payment investments and
covering losses
Loss
Major loss
Cycle of investments
Business risk Operational risk
Liquidity risk
Inability to carry on
business’ investment
1 2
Figure 2.10 Risks in Mudãrabah contract agreements during the investment
period
partners to operational, credit, market, and liquidity risks as follows:
As the financial institution in a Mudãrabah contract has a partner-
ship in the actual business that it is investing in, it is inherently exposed to
business and operational risks as illustrated in Figure 2.11, points 1 and 2.
This is due to any external or internal events that may arise and are causing
losses to the actual business. Thus, any inadequate activities or failures
which are beyond the scope of due diligence by the agent during the busi-
ness operational processes and activities that cause losses are expected to be
covered fully by the financial institution.
Major losses in Mudãrabah contracts (Figure 2.11, point 2) may
cause the inability of the financial institution to provide additional funds for
the Mudãrabah investment and, thus, the business may be unable to operate
further. This event will result in a last equity payment to the investment’s
equity shares. In this case, the equity price will most probably have a mar-
ket price lower than the initial nominal that reflects a financial exposure to
market equity risk as shown in Figure 2.11, point 3.
Financial institutions are expecting profits resulting from the
Mudãrabah contract. As a consequence of the above-mentioned losses, the
investors of the Mudãrabah business are unable to provide the expected
52 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Market risk P and L
fluctuations
Level of investment
Profit sharing and covering losses Last equity
payment and/or
sum of payments
Cycle angle income
Investment principle
Major loss
Loss
Balance sheet
fluctuations
Operational risk Inability to carry on
business
Business risk credit risk
Liquidity risk
1 2
Figure 2.11 Risks in Mudãrabah contract agreements during the business
profitability and loss period
profit. Thus, as shown in Figure 2.11, points 1 and 2, the financial institu-
tion is exposed to credit risk due to this default on the expected cash-in from
the business profits.
The above-mentioned inabilities to repay expose the financial insti-
tutions to liquidity risk as it is likely that the institution may be unable to
provide enough cash for its other investments and activities.
In Mudãrabah financial contracts, financial institutions have
negligible management rights on the partnership business. As mentioned
above, these limitations may cause a transparency risk that would result
in losses to the financial institution. Therefore, transparency risk should
be highly considered and controlled by the financial institutions that are
providing Mudãrabah contracts.
Operational risk management in Mudãrabah
contracts
Similar to Mushãrakah, the operational risk in Mudãrabah contracts is
mainly initiated due to business risks. Since the agent (business partner) has
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 53
full management responsibilities, the events that initiate losses related to the
operational risks can barely be managed by the financial institution.
Financial institutions must ensure that the business deals using the
Mudãrabah contracts are driven by experienced and knowledgeable agents
where the projects are managed in such ways as to minimise business risks.
Credit risk management in Mudãrabah contracts
The credit risk exposure in Mudãrabah contracts can be minimised by
monitoring the business performance, if possible, and the balance sheet of
the business profits and losses.
Market risk management in Mudãrabah contracts
By applying equivalent principles as in the Mushãrakah contracts, for
managing market risk, financial institutions should define strategies that
will be implemented for the case of market risk in Mudãrabah contracts
where, for instance, the stop loss for selling the last equity price is defined.
Liquidity risk management in Mudãrabah
contracts
In Mudãrabah contracts, the liquidity risk is initiated by the other types of
risks. Thus, financial institutions may need to provide capital adequacy that
needs to be defined, based either on the regulators’ directives or on internal
estimations (see chapter 3).
Example Customer X approaches the bank ABC to invest in the manufac-
turing of ceramic tiles but insists that he cannot contribute any capital
Mudãrabah. The bank invests all the required funds for the business, whereas
the customer takes care of management. During the investment period, the
bank is exposed to operational risk arising due to external and internal events,
as illustrated in Figure 2.10, points 1 and 2. In case the customer (partner) is
unable to generate the profits, the bank may be exposed to liquidity risk. In
case there are major losses, the bank may not be able to provide additional
funds, thus threatening the closure of the business. The market for the ceramic
tiles will be an important factor affecting the risk position of the bank, which
can affect the bank via operational and market risks.
Note Mushãrakah and Mudãrabah instruments have some common fea-
tures and hence share some common risks. When these instruments are used
for financing import and export trade, they carry foreign exchange risk. The
54 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
transparency poses risk for the bank. If the partner does not disclose true
and fair profits, the bank practically does not have ways and means to
control it.
MURÃBAHA CONTRACT AGREEMENTS
AND FINANCIAL RISK
The Murãbaha contract is one of the most popular contracts of sale used for
purchasing commodities and other products on credit. Most of the financial
institutions that are providing Islamic products are extensively using
Murãbaha as an Islamic mode of financing, and most of their financing
operations are based on Murãbaha. This type of contract is suitable for
financing the different investment activities of customers with regard to the
manufacturing of finished goods, procurement of raw materials, machinery,
and other required plant and equipment purchases.
Murãbaha is a contract arrangement where there are three parties
involved: the ‘financial institution’ the ‘buyer’ (client), and the ‘seller’. The
financial institution acts as an intermediary trader between the buyer
(client) who is the end-user and the seller (Figure 2.12). In other words,
upon receipt of an order and agreement to purchase a certain product from
the buyer, the financial institution will purchase the product from the seller
to fulfil the order. This type of financial contract is based on the sale for an
agreed-upon profit. This means that, in addition to the cost of a commodity,
there is an agreed-upon profit payable at the present time or on a date in the
future in a lump sum (at the maturity day) or by instalments (cycle repay-
ments), as defined in equation (2.1).
n
R 兺 (r p )
i1
i i (2.1)
where: n is the number of payment instalments,
r is the commodity’s price of the repayments,
p is the profit earned by the financial institution.
Under this sale transaction, the financial institution is calculating the
profit margin over the initial cost defined by the seller. The profit margin
that is based on a benchmark or mark-up rates may be defined either as a
fixed sum or based on a percentage of the price of the goods. If a percent-
age is used, the percentage shall never be expressed in terms of time, in
order to avoid confusion that the price is a form of interest (Riba), which
is not allowed. The institution must disclose clearly the cost of acquiring
the asset and hence should be transparent about the cost and profit.
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 55
Bank
profit capital
goods
Buyer Seller
Figure 2.12 Murãbaha structure and its main interrelations
Risks identification in Murãbaha contracts
By applying Murãbaha financial transactions, financial institutions are
exposed to credit, operational, market, and liquidity risk in an integrated
way, as illustrated in Figure 2.13. There are three main areas during the
Murãbaha contract where these risks may arise:
In Murãbaha contracts, a client can make a promise to purchase
from the financial institution. This is to either satisfy the promise or to iden-
tify any losses incurred from breaking the promise without excuse. In this
step, the financial institution is exposed to operational risk, as shown in
Figure 2.13 at point 1.
Stock and availability of the commodity/goods is a basic condition
for signing a Murãbaha agreement. However, the financial institution must
purchase the goods in accordance with the specifications of the client,
thereby taking ownership of the goods before signing the Murãbaha
agreement with the client. Therefore, the financial institution assumes the
operational and market risk of the commodity’s ownership. In other words,
the institution is responsible for damages, defects, and/or spoilage to the
merchandise until such time that it is actually delivered to the buyer.
Moreover, within this time period the financial institution is exposed to any
risk from the commodity’s price fluctuation. Point 1 in Figure 2.13 illus-
trates where these risks may appear before the commodity is even re-sold to
the client/buyer.
In the case of non-binding Murãbaha contracts where the client has
the right to refuse the delivery of the product purchased by the financial
institution, the institution is further exposed to market risk due to commod-
ity (price) fluctuation illustrated at point 1 in Figure 2.13.
56 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
The financial institution delivers to the client the commodity at an
agreed initial date, but may not receive the payment from the client at the
agreed cycle repayment instalments or at the maturity date. Thus, such
default of the client’s inability to repay is exposing the financial institution
to credit risk as shown in point 2 of Figure 2.13. Moreover, at this point, the
financial institution is also exposed to liquidity risk as on the repayment
days it is expecting to receive a cash-in that it may use to cover its other
financial obligations. Note also that in case the client agrees to pay at a later
date, or even after the maturity date, issues of additional liquidity risk may
arise due to unexpected cash-in at the later dates.
The financial institution purchases the commodity on a current date
and has to deliver the capital of the price plus the profit at a future date
based on a benchmark or mark-up rates; however, if the profit does not
cover the actual market rate, the financial institution is exposed to the mark-
up or, in other words, market risk (see point 3 in Figure 2.13).
Different types of risks arise at the same points during the lifetime of
Murãbaha contracts, also related to each other to a certain degree. Thus,
financial institutions that are utilising Murãbaha contracts should consider
their financial risks as integrated ones. For instance, financial institutions
may increase the margin of the profit for covering any commodity price
risk; however, such practice may cause inability of repayments, i.e. default
1 3
Market risk Market risk
Commodity price fluctuations
Commodity price risk
Commodity price risk
Market price
Balance sheet
Price + profit fluctuations
Receiving commodity
Payment instalments P and L fluctuations
Profit {
=
Ownership and promise
Payment inability
Cycle angle payment
Re-sell the commodity
Order for purchasing
Maturity date
In case of default collateral and
Operational
guarantees maybe applied
risk Credit
liquidity risk Future cash flows on delayed
payments
2
Figure 2.13 Credit, operational, and market risks during the lifetime of a
Murãbaha contract
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 57
of payments, that will consequently result in credit and liquidity risk. In
Murãbaha financial contracts, the defined profit should cover the actual
profit, the market influence rate as well as the additional services, costs, and
risk components.
Operational risk management in Murãbaha
contracts
In regards to the operational risk exposure – this is whether the client will
keep his promise of buying the product – the financial institution is permit-
ted to take a collateral security to guarantee the implementation of the
promise or to indemnify any losses that may result.
Credit risk management in
Murãbaha contracts
The financial institution accepts from the client that agrees on Murãbaha
contracts, goods or other assets as collaterals against the credit risk; that is,
the risk of default in repayments. Thus, appropriate documentation of the
debt resulting from Murãbaha contracts should be made by a guarantor, or
a mortgage, or both, like any other debt. Mortgage, guarantee, or cash secu-
rity may be obtained prior to signing the agreement or at the time of signing
the agreement. The analysis of credit risk is provided in chapter 4.
Commodity and mark-up risk management in
Murãbaha contracts
By providing Murãbaha contracts, financial institutions are exposed also to
mark-up risk due to the changes in mark-up rate that are used as a price trading
benchmark. In Murãbaha contracts, the mark-up price is fixed for the duration
of the contract, while the benchmark rate may change. In cases where the pre-
vailing mark-up rate increases beyond the benchmark rate, in the Murãbaha
contract, the financial institution is unable to benefit from the increased rates.
In Murãbaha contracts, the total repayment must be greater than the sum of the
repayments considering the fluctuation μ as defined in equation (2.2).
n
R · 兺 ri (2.2)
i1
Financial institutions must apply benchmarks and mark-up prices that
refer to the commodities that are trading in Murãbaha contracts. Moreover,
different scenarios that are driven by dynamic simulations of the market
behaviours can be used to estimate the future commodity prices.
58 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Liquidity risk management in
Murãbaha contracts
The liquidity risk in Murãbaha contracts is a result of the other risks and
may cause a significant impact and additional losses. Financial institutions,
therefore, should invest in efforts to manage the other types of risks and thus
minimise their exposure to liquidity risk.
Example As discussed in the example in the previous chapter, bank ABC
purchases a car for $25,000 from a dealer to sell it to customer X. As shown
in Figure 2.13, point 1, if the customer does not honour its promise to buy the
car from the bank, it is exposed to operational risk. At the same time, in the
case of non-binding Murãbaha contracts, the customer may refuse the deliv-
ery of the car and thus expose the bank to market risk due to fluctuations in
the price of the car. A default from the customer as illustrated in Figure 2.13,
point 2, may expose the bank to credit risk. In case the bank’s mark-up is not
sufficient, it is exposed to market risk as shown in Figure 2.13, point 3.
SALAM CONTRACT AGREEMENTS
AND FINANCIAL RISK
Salam is a term used to define a sale in which the buyer makes advanced
payment, but the delivery is delayed until some time in the future. Forward
sale is prohibited in Islamic finance, with two exceptions: Salam and
Istisnã. Usually, the seller is an individual or business and the buyer is the
financial institution. The Salam contract serves the interests of three parties:
the financial institution (bank), the seller, and the buyer, as illustrated in
Figure 2.14. The seller receives advanced payment in exchange for the
obligation to deliver the commodity at some later date. Firstly, the seller
benefits from the Salam sale by locking the price of the sold commodity. It
also allows covering its financial needs and business expenses. As a conse-
quence, the financial institution (purchaser) benefits from the Salam
contract by receiving the delivery of the commodity when needed to fulfil
some other agreement, without incurring storage costs. Secondly, a Salam
sale is usually less expensive than a cash sale. Finally, a Salam agreement
allows the purchaser to lock in a price, thus protecting both parties from
price fluctuation. In a Salam agreement there are several options for
delivery available to the financial institution on the due date:
The financial institution (bank) may receive the commodity and resell it
to another party for cash or credit;
The financial institution (bank) may authorise the seller to find another
buyer for the commodity;
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 59
Commodity
Sell for Deliver
profit
Bank
Buyer Seller
Figure 2.14 Relation structure in Salam contract agreements
The financial institution (bank) may direct the seller to deliver the com-
modity directly to a third party with whom the financial institution has
entered into another agreement.
In all the different options, the financial institution plays an intermediate
role between the seller and the buyer, as illustrated in Figure 2.14. The
financial institution gets a profit out of the trading between the seller and the
buyer.
In terms of functionality, the Salam contract is structured in the same
way as the conventional forward contracts; however, it defers in terms of
the payment arrangements. In the Salam contract, the payment is set at the
deal date on an agreed price where the seller promises to deliver at the
specified future date. On the other hand, in the conventional forward con-
tract, the full negotiated price is payable at the contract deal date, as
opposed to the Salam, where the full payment is normally set on the
delivery date.
Risks identification in Salam contracts
By undertaking Salam contracts, financial institutions are mainly exposed
to operational, credit, market, and liquidity risks:
Due to external events, the seller may default on delivering the com-
modity at the delivery date agreed in the Salam contract. In such cases, the
financial institution may have some losses due to its exposure to the above-
mentioned operational and credit type of risks (see Figure 2.15 at points 1
and 4). Note that, according to the IFSB,12 such default is considered as
credit (counterparty) risk.
60 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
In addition to this operational risk, there could also be a mismatch on
the agreed specifications referring to the received commodity. Such failures
could cause further delays to the delivery of the actual commodity after the
selling date (agreed between the bank and the buyer), as illustrated in
Figure 2.15 at point 4, or even termination of the contract. Delayed deliver-
ies for receiving the commodities may result in some reputational and
business risks as well as cause additional expenses and opportunity losses.
Even though the price of the commodity is locked in the Salam
contract in order to be protected from commodity risks, both the bank that
receives the commodity as well as the seller that sells it, at the delivery date
the financial institution may face a market risk due to commodity price
fluctuation and mark-up risk (see Figure 2.15 at point 2).
The Salam financial Islamic product exposes the financial institution to
commodity price volatility during the period between the delivery of the
commodity and the sale of the commodity at prevailing market price. The
higher forward price accommodates the cost of commodity from the trade
date to the delivery date, including financing, insurance, and storage costs.
According to the definitions given by the Islamic Financial Services Board,
institutions that are providing Salam financial products are exposed to
commodity price fluctuations on a long position after entering into such a
contract and while holding the subject matter until it is disposed of or even
beyond the maturity date of the contract, as long as the commodity remains
on the balance sheet of the institution.
Note that any inability to forecast and thus estimate the future price of
the commodity based on benchmarks and mark-ups (at the delivery date)
could expose the financial institution to reinvestment risk that makes it
unable to re-sell the commodity at a profitable price.
On the selling date (see Figure 2.15 at point 4) any default from the
buyer to pay the commodity at the agreed price exposes the financial insti-
tution to the credit risk.
Since the price and time for the end-buyer are defined before the date
of delivery of the commodity in the Salam contract, the financial institution
may face additional market risk due to mark-up risk (see Figure 2.15 at
point 3). Moreover, in cases of delivery defaults (operational risks and mar-
ket risks), there is an additional exposure to liquidity risk, as it expects cash-
flow that may not be received at the future selling time, as illustrated in
Figure 2.15, points 3 and 5.
Operational risk management in Salam contracts
Financial institutions that are providing Salam financial Islamic contracts
might be exposed to operational risks due to a seller’s default of the
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 61
2 3
Market risk Market risk P and L and
Commodity’s current balance sheet
Liquidity risk
principle on fixed price fluctuations
Nominal value
Market price
Received commodity
Mark-up risk
Mark-up risk
Advance payment
Future sell
Fixed payment
Default on delivery
Delay on delivery (after SD)
Time
Default on buy
Contract deal date
Delivery Selling Date
Date (DD) (SD)
Credit risk Operational risk
Market risk
Operational risk Credit risk Liquidity risk
1 4 5
Figure 2.15 Operational, credit, market, and liquidity risks during the lifetime
of Salam contracts
commodity’s delivery or due to the commodity’s specification mismatch-
ing. Financial institutions may minimise such types of operational risks by
asking from the seller guarantees that they are following a quality manage-
ment system or following any ‘standard system’, and/or by asking for refer-
ences on past promises on Salam contracts and/or by collateralising their
losses via insurance policies.
Credit risk management in Salam contracts
By providing Salam financial Islamic contracts, the financial institutions
might also be exposed to credit risk due to a default from the buyer’s side to
purchase the commodity at the agreed price. Financial institutions should
apply approaches for estimating the probability of such defaults and the
result from such credit risks and expected losses. Such approaches are
mainly based on quantitative information from past references that are com-
bined with qualitative criteria related to promises on Salam contracts, as
will be discussed in chapter 4, referring to credit risk management in
Islamic finance.
62 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Market risk management in Salam contracts
By evaluating the future market price, set for the Salam contracts, based on
different market scenarios and strategies, financial institutions can minimise
their exposure to market risk and mark-up risk related to commodity price
fluctuations. Moreover, based on static and dynamic analysis, the VaR
analysis can be applied in evaluating and managing the market risk in Salam
contracts.
Hedging the Salam contracts
Financial institutions that provide Salam financial contracts are hedging
their position by entering into Parallel Salam off-setting contracts.
Financial institutions have several options for delivering the commodity and
thus to hedge the corresponding risks:
They may receive the commodity and re-sell it to counterparties for
cash or credit.
They may authorise the seller to find a third counterparty buyer for the
commodity.
Or they may advise the seller to deliver the commodity directly to a
third counterparty with whom the financial institution has entered into
another agreement.
A financial institution that is using parallel Salam contracts expects from
the first contract a commodity to be delivered from the first counterparty at
the agreed delivery date. With the second contract it is obligated to deliver
the commodity to a second counterparty on a future date, after the delivery
date. The financial institution is therefore exposed to commodity risk from
the first contract; thus, when there is a default of delivery from the first
counterparty, the loss is offset by the future contracts that mature on a future
delivery date.
Example – Parallel Salam A customer X of bank ABC wishes to arrange
for a forward sale of its crop. The bank enters into a Salam contract (first
contract) where it provides the required finance and in return agrees to pur-
chase the crop from customer X after a specified time. At the same time, the
bank enters into another Parallel Salam contract (second contract) with a
buyer (or customer Y) to ensure the sale of the crop. If, due to external
events, the customer defaults on the delivery of the crop, the bank is
exposed to operational and credit risk as shown in Figure 2.15, point 1. If
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 63
the market price of the crop increases substantially during the first phase,
there is an incentive for the seller to default. After the bank accepts the
delivery, in case the price of the crop has increased substantially, the buyer
(in a Parallel Salam contract) may default, thus leaving the bank with stor-
age costs and further credit risk and commodity price risk (see Figure 2.15,
point 4).
ISTISNÃ CONTRACTS AND
FINANCIAL RISKS
Istisnã financial contracts are used to define a production of a commodity or
construction of an asset in which the buyer makes advance payment, but the
delivery of the commodity or the asset is set at the future time. In other
words, Istisnã is a contract that ‘a manufacturer makes something’ and ‘it is
a commodity on liability with the provision of work’. The Istisnã contract is
very appropriate for the construction industry and for building infrastruc-
ture such as roads, dams, real estates, hospitals, etc.
The contract serves the interests of three parties: the financial institution
(bank), the producer (manufacturer), and the buyer/user of the commodity,
as illustrated in Figure 2.16. Financial institutions that provide Istisnã
contracts are buying the commodity or the constructed asset and then are
selling them after receipt for cash on deferred payment.
The Istisnã contract comes into existence when the manufacturer under-
takes the production of the commodity or the construction of the asset for
Commodity
construction
of asset
Sell for Produce +
profit deliver
Bank
Constructor
Buyer/user manufacturer
Figure 2.16 Relation structure in Istisnã contract agreements
64 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
the financial institution/buyer. Apart from the fixed price, the parties in the
Istisnã financial contract agree on the specifications of the asset intended to
be manufactured. Note, therefore, that, if there is a deficiency in fulfilling
the agreed specifications, the party that had placed the order has a right to
withdraw the order.
In Istisnã contracts, the feature related to the mode and timing of the
payment is very flexible as it can be fixed at any specific time, from the
beginning of the contract to the delivery time, or it can even be set at
different payment instalment times. There is also flexibility on the delivery
time. The amount of instalments could also be agreed and scheduled by the
parties in the contract.
Differences between the Istisnã and
Salam contracts
The Istisnã and Salam are similar contracts in terms of the creditability of
time. However, the main differences between these contracts are:
Salam can be used for almost all types of commodities, excluding gold,
silver, and currencies; however, in the Istisnã contract, the underlying
asset is required to be manufactured or constructed.
Salam price is paid in full and in advance, whereas the payment instal-
ments for the commodity in Istisnã contracts are flexible and can be
made at any time if the parties agree.
The Istisnã contract can be cancelled before the manufacturer or con-
structor undertakes the associated project, whereas in a Salam contract
the unilateral cancellation of the contract is forbidden.
Unlike the Salam contract, in the Istisnã contract there is flexibility on
the delivery time.
Risks identification in Istisnã contracts
Financial institutions that are providing Istisnã contracts are exposed to
operational, market, credit, and liquidity risks:
Due to external events during the manufacturing or construction
process, there may be a default from the supplier’s side to provide the
goods at all, or supply the goods on time, and the manufacturer or
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 65
constructor may default on:
carrying out the process of producing the commodity/asset, as
illustrated in Figure 2.17 at point 1.
delivering the agreed commodity/asset on time by delaying the
production process and shifting the delivery date after the selling time
agreed between the financial institution and the buyer, as illustrated in
Figure 2.17, point 6.
In both cases, the financial institution is exposed to operational risk where
the associated losses are initiated due to the influence on the bank’s reputa-
tion and future liquidation problems (liquidity risk) that the financial insti-
tution may face due to its inability to sell and cash in the commodity or the
asset to the buyer at the selling date (see Figure 2.17, at points 4 and 6).
Given the structure of the Istisnã contracts, as illustrated in Figure 2.17
at point 2, on the delivery date, the manufacturer or constructor may default
in reaching the agreed quality of the goods as contractually specified.
Assuming that the manufacturers and/or constructors are counterparties in
relation to the bank, such defaults result in a credit risk exposure.
Moreover, the aforementioned defaults arise mainly due to operational
risks and that implies further delays on the delivery of the commod-
ity/asset (see Figure 2.17, at point 6) or even termination of the contract.
Note, however, that by shifting the delivery date to a future time which
is after the agreed selling date (between the financial institution and the
buyer), it may cause losses to the financial institution due to liquidity
risk as well as business risk (including reputation, administration, legal
fees, etc). The failures mentioned above could result in financial losses
due to additional expenses and opportunity losses.
Credit exposures in the Istisnã contracts are arising at the selling
time where the commodity or the constructed asset is billed to the cus-
tomer (see Figure 2.17, at point 5). According to the IFSB§126, there is
credit (counterparty) risk when the financial institution is not receiving
the selling price of the asset from the customer or project sponsor either
in pre-agreed stages of completion and/or upon full completion of the
manufacturing or construction process.
The Istisnã contracts may also expose the financial institution to
market risks. This is due to the fact that the price of a commodity or a
construction is fixed on the deal date. However, at the delivery date, due
to market price, fluctuations of the commodity may result in a
differentiation from the actual market price (see Figure 2.17, points 3
66 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
and 4). This price differentiation is initiated due to the mark-up risk
where any underestimation of the future price that is based on
benchmarks and mark-ups could result in the financial institution being
unable to sell the commodity at a profitable price.
Similarly to the case discussed in Salam contracts, the Istisnã financial
Islamic contracts expose the financial institutions to the volatility of the
commodity and asset price during the period between the delivery of the
commodity or the constructed asset and the sale of the commodity at
prevailing market prices.
In the Istisnã contract, any default on delivery, mainly due to oper-
ational risks after the selling date (Figure 2.17, at points 4 and 6),
exposes the financial institution to liquidity risk as the institution is
expecting a cash flow that it will not receive at the future selling date.
Moreover, the financial institution is exposed to market risk as the price
of the commodity or the asset at a later date may fluctuate and thus dif-
ferentiate the nominal set price with the market one.
Operational risk management in Istisnã contracts
When selling Istisnã financial Islamic contracts, financial institutions may
be exposed to operational risks due to the default of manufacturing and
3 4
Market risk Market risk P and L and
Commodity’s/asset current Liquidity risk balance sheet
Principle on fixed price fluctuations
Nominal value
Market price
Mark-up risk
Mark-up risk
Receive commodity/asset
Sell commodity/asset
Fixed payment
Advance payment
Variable payments
Default on delivery
Default of manufacturing
Delay on delivery after SD
Time
Default on buy
Contract deal date
Delivery Selling
date date
Credit risk Operational risk
Market risk
Operational risk Operational risk Credit risk Liquidity risk
1 2 5 6
Figure 2.17 Operational, credit, market, and liquidity risks during the lifetime
of Istisnã contracts
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 67
construction companies to deliver or due to the mismatching of the
commodity’s or asset’s specification. Similarly to the risk management
practices that should be applied in the Salam contracts, financial institutions
can minimise such types of operational risks by monitoring the process of
the commodity’s manufacturing or the construction of the asset. Moreover,
financial institutions should receive guarantees from the manufacturers on
whether they are following an appropriate quality management system or
that they are following any ‘standard system’. Financial institutions should
also ask for references concerning past manufacturing or construction
projects. Losses that are initiated from the Istisnã contracts can also be
collateralised via insurance policies.
Credit risk management in Istisnã contracts
Financial institutions might also be exposed to credit risk due to a default
from the buyer’s side to purchase the commodity or the constructed asset
agreed in the Istisnã financial Islamic contract. Banks must be able to
estimate the probability of default and the expected losses resulting from the
Istisnã credit risk based on qualitative criteria as well as quantitative
information data, as discussed in chapter 4.
Market risk management in Istisnã contracts
For commodity and construction assets, a good strategy for minimising
market risks in an Istisnã contract is to sell the commodity and the proper-
ties resulting from the asset’s construction before the delivery date.
Moreover, financial institutions can minimise their exposure to market risk
and mark-up risk by evaluating the future market prices of the manufactured
commodity and constructed assets. By applying static and dynamic
analysis, the VaR analysis can evaluate the significance of the market risk
exposure that the financial institution is undertaking on Istisnã sales.
For minimising both operational and market risks, financial institutions
that provide Istisnã contracts should set the payment instalments to the
manufacturer or constructor of the asset on a variable basis instead of on the
contract deal date.
Liquidity risk management in Istisnã contracts
One characteristic of the Istisnã contract is that the delivery and the selling
date of the manufactured commodity or constructed asset may change. As
aforementioned, this shifting of time may have a great impact on liquidity
risk. Financial institutions that sell Istisnã contracts should be aware of this
risk and reserve adequate capital for covering such liquidation issues.
68 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Example A customer X wants to purchase a ship. Bank ABC agrees to
provide the financing using Istisnã, where it enters into a contract with the
ship manufacturer to manufacture the specified ship and agrees to provide
the finance in instalments, depending on the progress in manufacturing. The
bank also enters into another contract simultaneously with customer X to
sell the specified ship at an agreed price. In case the manufacturer fails to
deliver the ship on time as agreed, the bank is exposed to operational risk as
illustrated in Figure 2.17, points 1 and 6. Failure of the customer to pay the
selling price exposes the bank to credit risk at the time of delivery of the
ship (see Figure 2.17, point 5).
IJÃRAH CONTRACT AGREEMENT AND
FINANCIAL RISK
Ijãrah is a leasing financial contract that gives the right to use a commodity
or service for a specific period of time. Ijãrah or leasing is a contract
whereby the owner of something transfers its usufruct to another person or
body for an agreed period. In general, Ijãrah contracts refer to the lease/rent
of tangible assets such as property and merchandise, but it is also designed
to denote the hiring of professional services for a fee.
Ijãrah is a contract between two parties:
1. the individual or organisation who leases out/rents out the property or
service and is called the lessor (mu jir)
2. the lessees (financial institution – musta jir) who have the right to
enjoy/reap a specific benefit from a specified consideration, rent, wages,
from the lessor.
In Ijãrah contracts, the buyer usually makes advance payment and rents,
leases, hires the goods and/or services from the lessor. The delivery time is
set to a fixed date in the future. The payments can be either fixed or variable;
however, in the former case they are based on benchmarking analysis.
Nowadays, extensions of Ijãrah contracts are implemented with the most
commonly used options for purchasing the asset. One of these is the Ijãrah
wa Iqtina (hire-purchase agreement), which is similar to conventional
lease-purchase agreements. Based on this type of agreement in Ijãrah con-
tracts, an additional contract is applied, which includes a promise by the
financial institution (owner of the asset) to sell the leased asset to the lessee
at the maturity date of the original lease of the Ijãrah contract. The price of
the residual value of the asset is pre-determined. The second type of Ijãrah
contract, known as Ijãrah thumma al-bai, provides an option to the lessee to
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 69
Leasing
commodities
or services
Benefit Deliver
Rent/Buy
Lessee Lessor
(borrower) (bank)
Figure 2.18 Ijãrah contract agreements
purchase the asset at the maturity date of the contract or to return it as orig-
inally agreed in the Ijãrah contract.
Risks identification in Ijãrah contracts
A financial institution as the owner of an asset that leases to the lessor based
on Ijãrah contracts is exposed to credit, operational, market, and liquidity
risks:
Any default of payment due to business risks, market (price varia-
tion) risks, or even due to unscrupulous defaults by the lessee, may generate
additional losses. Thus, such cases of inability to re-pay by the lessee is
exposing the financial institution to credit risks, as illustrated in Figure 2.19,
point 1.
In the case that the renter/lessee exits earlier (before the maturity
date), the financial institution will lose the expected payments on the prede-
fined instalments or at the maturity date. Thus, any early leave exposes the
financial institution to credit risks (Figure 2.19, point 2). Moreover, as a
result of the above-mentioned risk, there is an additional exposure to liq-
uidity risk at the times where cash flows were expected from the instalment
payments or the fixed payments (last payment).
Catastrophic events could also cause damage to the assets (such as
properties) that may initiate major losses. The financial institution (owner)
is therefore exposed to operational risk, as shown in Figure 2.19, point 3.
The title of the ownership of the asset remains with the lessor, thus
damages due to misconduct by the lessee propagate operational losses. The
financial institution, who is the owner of the assets, is exposed to
operational risk.
70 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
The value of the payments of rent in Ijãrah contracts is defined on
cycle payment instalments and is based on benchmark analysis. However,
when such an estimation is unable to fulfil the actual market price, the
financial institution is exposed to market risk (see Figure 2.19, point 4).
In the case where the Ijãrah contract is set for payment at the matu-
rity date, the financial institution may be exposed to market as well as to
credit risks. Losses from the latter risk may appear when and if the
renter/lessee defaults on the obligation for payment (full or partial amount),
as shown in Figure 2.19, point 5. The former risk may arise due to the mar-
ket risk fluctuation that could differ from the actual agreed price (on the
Ijãrah contract), as illustrated in Figure 2.19, point 6.
In general, at the maturity date, when the asset will be returned to
the financial institution, any damages to the assets may cause an associated
loss. In such cases there is an exposure to operational risk once more
(Figure 2.19, point 6).
Delayed repayments due to the above-mentioned defaults may
cause liquidity risk (see Figure 2.19, point 7).
In the Ijãrah wa Iqtina, the financial institution might be exposed to
market (commodity) price risk as the price of selling the asset to the renter
at the maturity date is pre-determined. Due to market fluctuations, the price
at the maturity date may differ from the actual market price.
In the Ijãrah thumma al-bai, the financial institution might be
exposed to market (commodity) price risk as the lessee may refuse to pur-
chase the asset on the termination of the lease; thus, at the maturity date, due
to market fluctuations, the market risk may arise.
When in the Ijãrah contract there is an option for the renter/lessee
to buy at a pre-determined price the asset at the maturity date, the financial
institution is exposed to the market (commodity) price risk in the same way
as discussed in the previous point.
The risks encountered by the Ijãrah contracts are, to a high degree,
similar to the ones that exist in conventional leasing contracts.
Credit risk management in Ijãrah contracts
Guarantees (capital or personal) and collaterals may be applied for min-
imising the financial institution’s exposure to credit risk by revoking the
contract’s agreement or inability of payment from the renter in the cycle
instalments or at the maturity date.
Operational risk management in Ijãrah contracts
For minimising their exposure to operational risks, adequate insurance
against any losses and damages to the asset should be defined.
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 71
6
Market risk
Market price
fluctuation Sell the asset
Buy in advance and lease
Fixed payment 4 7
Market risk
Liquidity risk
Received back
Loss
Variable payment Delayed
instalments PF payments
Payment rent inability
Earlier leave
Default of payment
Damage of the asset
Major catastrophe
Contract deal date
Maturity
date
Balance sheet
P and L
fluctuations
Credit risks Operational risk Credit risks
1 2 3 5
Figure 2.19 Credit, operational, and market risks during the lifetime of Ijãrah
contracts
Market risk management in Ijãrah contracts
In undertaking Ijãrah contracts, the exposure to market risks can be
minimised by simulating and evaluating the future market price based on
different market scenarios and strategies that are mainly driven by potential
yield curves. VaR analysis could also be applied to assess and evaluate the
current and potential degree of market risks.
In general, there is less exposure to the above-mentioned risks when the
Ijãrah contract is set for repaying the instalments at different time periods
(cycle payments) than when the repayment is fixed to the maturity date.
Liquidity risk management in Ijãrah contracts
The exposure to credit, operational, and market risk by applying Ijãrah
contracts results in a liquidity risk to different degrees. A good practice to
minimise the liquidity risk is to monitor and manage its causes from the
other risks.
Example An airline approaches ABC bank to finance the purchase of an
Airbus aircraft costing approximately $80 billion US. The bank arranges to
lease the aircraft to the airline using an Ijãrah contract. If the airline defaults
on the payment of rent, it exposes the bank to credit risk (see Figure 2.19,
72 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
point 1). In case the airline decides to leave the lease early, the bank is
exposed to credit risk, as shown in Figure 2.19, point 2. In the case of a cat-
astrophic event affecting the aircraft, the bank is exposed to operational
risks, as illustrated in Figure 2.19, point 3. On the maturity date when the
aircraft is returned, the bank is exposed to operational risk from damages to
the assets (see Figure 2.19, point 6).
Sukûk
Sukûk, as used in recent Islamic finance, refers to a certificate issued by a
beneficiary as evidence of the collected funds entitling rights in certain assets.
Shariah approval is needed for the underlying assets and for Sukûk contracts.
These assets are driven by Islamic contracts and accordingly there are various
types of Sukûk, named Murãbaha Sukûk, Ijãrah Sukûk, Salam Sukûk, Istisnã
Sukûk, Mushãrakah Sukûk, and Mudãrabah Sukûk. Thus, in the Sukûk
contract, the risk and the return associated with cash flows generated by
underlining assets belong to the pool and are passed to the Sukûk holders
(investors). As a result, all risks that may appear in a Sukûk contract are related
to the associated risks that arise in its underlining Islamic contracts, as
described in the above paragraphs of this chapter. Note that when a Sukûk is
underlined by a pool of assets there is a distribution of different types of risks;
however, a high degree of these risks are resulting in a high complexity in col-
lecting the associated information and in analysing/evaluating their actual
effect on the performance of the actual Sukûk returns.
The Sukûk structure depends on the creation of a Special Purpose Vehicle
(SPV) and the underlying assets. Sukûks representing debts are not trace-
able on a secondary market, although there are some alternative viewpoints
from some of the Shariah scholars. For example, a Sukûk structured on
Murãbaha, Salam, and Istisnã should be held until maturity. On the other
hand, a Sukûk structured on equity basis can be traded on the secondary
market (Mushãrakah and Mudãrabah). On similar lines, Sukûk can be clas-
sified either as having pre-determined returns or profit and loss sharing. It
should be noted that Sukûk is always an asset-backed security. Sukûk is
issued with varying maturity, such as from a few months to a few years.
Salam Sukûk has been mostly issued for managing short-term liquidity
since its structure resembled that of conventional treasury bills. The Ijãrah
Sukûk has been most popular and can be issued for varying maturity, gener-
ally at least five years. Other than these, Mushãrakah Sukûk and Istisnã
Sukûk are also being tried. Due to the availability of a large variety of Sukûk
and their differing risk structures, it is beyond the scope of this book to dis-
cuss them. Fundamentally, the risks in Sukûk can be traced back to the
underlying contract type, which has been discussed in the above paragraphs
in this chapter.
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 73
SUMMARY
Risks and returns are strongly related in Islamic financial products. The
unique structure of Islamic financial products exposes them to different
types of risks at different stages of the contract. Financial institutions that
are providing Islamic financial products or formulate deals based on Islamic
financial contracts are exposed to financial risks, including credit, opera-
tional, business, market, and liquidity risks. Such risks may result in major
financial losses and excessive disruptions in the financial institutions’
performances. Islamic finance is riskier as compared to conventional
finance. The information management plays a crucial role in risk manage-
ment in Islamic finance. The specific nature of Islamic financial contracts
exposes the Islamic financial institutions to risks which are shifting and
changing over the time of the contract.
Table 2.3 illustrates an area of risks that financial institutions are
exposed to by applying Mushãrakah, Mudãrabah, Murãbaha, Salam,
Ijãrah, and Istisnã Islamic financial contracts. As illustrated in this Table,
all of the Islamic financial contracts are exposed to operational risks.
Likewise, credit risks appear in all contracts, except in the Salam and
Istisnã contracts, where the main drivers of these agreements is a
commodity or an asset delivered to the financial institution some time in
the future. Moreover, all contracts that are dealing with commodities, i.e.
Murãbaha, Ijãrah, Salam, and Istisnã, are exposing the financial institu-
tions to commodity price risk; on the other hand, the Mushãrakah and
Table 2.3 Pattern of risks where financial institutions are exposed by applying
Islamic financial contracts
Types of risks
Market risk
Types of Operational Credit Commodity Equity Liquidity
contracts risk risk risk risk risk
Mushãrakah ✓ ✓ ✓ ✓
Mudãrabah ✓ ✓ ✓ ✓
Murãbaha ✓ ✓ ✓ ✓
Salam ✓ ✓ ✓ ✓
Ijãrah ✓ ✓ ✓ ✓
Istisnã ✓ ✓ ✓ ✓
74 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Mudãrabah contracts, which are partnership agreements, result in equity
risks.
In the usual course of events, most of the above-mentioned risks addi-
tionally expose the institution to liquidity risks, as any default or delays
regarding payments or deliveries from the buyers’, sellers’, or manufactur-
ers’ side, may initiate an elevated interruption on the financial institutions’
expected cash flow. Such interruptions may result in the financial institution
facing difficulties or even the inability to fulfil its financial obligations,
especially the ones that are related to other financial products and services.
Additionally, in the case that Islamic financial products are trading in
foreign countries and currencies, they may also be exposed to foreign
exchange risk. Note finally that the losses resulting from the different types
of risks arising from Murãbaha contracts may cause fluctuations on the
balance sheet and P&L of the financial institution.
In terms of managing the above-mentioned risks, financial institutions
should define a risk-management framework of measuring, monitoring,
assessing, evaluating, controlling, and managing their associated risks. For
instance, guarantees or cash securities can be applied to collateralise the
market and operational risks’ exposures. Scenarios that are driven by
dynamic simulations of the market behaviours can also be employed to
define and evaluate the benchmarks and the mark-up prices.
In the following paragraphs, a short summary and concluding remark are
presented and discussed for each of the examined Islamic financial products
and business agreements.
Mushãrakah
The Mushãrakah contract, a partnership contract between a financial
institution and a partner, involves both sides in management and decision
making where the profits and losses are shared according to the proportion
in which the two parties have provided the finance for the project.
According to the financial institution’s participation towards the equity of
the investment company, the Mushãrakah partnership is distinguished in
two types of contracts: the Permanent Mushãrakah and the Diminishing
Mushãrakah. In the first type, the financial institution is sharing the profit
and loss and therefore all the associated investment risks until the termina-
tion of the business. In the latter type, the institution is sharing the business
investment and responsibilities for a limited pre-defined period. In both
cases, however, the Mushãrakah contract that combines the capital invest-
ment and business management initiates credit, operational, market, and
liquidity risks.
When the financial institution is dealing with the Permanent Mushãrakah
contract, it is exposed to operational risks due to any inadequacies or
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 75
failures concerning business activities or even due to external events that
cause major losses. Any resulting defaults from the business side for
providing the expected cash exposes the financial institution to credit as
well as to liquidity risks. Finally, the last equity payment that results from
any inability of undertaking the business activities is exposing the financial
institution to market equity risks.
The Diminishing Mushãrakah contract exposes the financial institution
to operational, credit, and liquidity risk when it does not receive the equities
due to a failure of the partner’s business. Moreover, the fixed price of the
equity, at the start-up of the Mushãrakah contract, may also cause some
market risks to the institution. This is due to certain mismatching between
the actual market price and the fixed prices, which results in loss of
potential profit.
The operational and credit risks in both Permanent and Diminishing
Mushãrakah contracts may be minimised by the financial institution’s man-
agement rights, decision making, as well as by monitoring the balance of
the business profits and losses. Additionally, insurance policies may be
applied to cover any major losses that are initiated due to external events.
Finally, for assessing and evaluating the market equity risk a VaR approach
should be applied.
Mudãrabah
The Mudãrabah contract is a partnership where the financial institution is
investing in a project and agrees with a partner that runs the business to
share only the profit. However, in this partnership, the financial institution
(who is the capital investor) is the only one that takes the whole responsi-
bility concerning business losses. The Mudãrabah contract is a partnership
that does not entitle the capital investor to take business responsibilities;
however, it is exposing the financial institution to operational, market, and
liquidity risks.
During the investment period of the Mudãrabah agreement, the financial
institution is exposed to operational risk due to internal business failures or
even to external events, including catastrophic ones. As a result of its
obligation to fully cover the business losses, the financial institution is
additionally exposed to liquidity risk. Moreover, any defaults to provide the
expected cash due to the above-mentioned risks and losses initiates credit
risk. During the profit and loss period of the Mudãrabah contract, the
business partner has full responsibility for management. Any external or
internal unexpected events may initiate business and operational risk losses.
Moreover, in case the businesses that are based on Mudãrabah contracts
return only minor profits, the financial institution may also face credit as
well as liquidity risks.
76 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
As mentioned above, in the Mudãrabah partnership the financial institu-
tion is fully responsible for covering all losses (minor to major) and thus its
exposure to risk has more financial impact. Financial institutions that are
dealing with both Mushãrakah and Mudãrabah contracts are exposed to
similar types of risks, but in the Mudãrabah contracts the financial institu-
tion is exposed to a higher degree due its obligation to exclusively cover any
financial losses. Moreover, in Mudãrabah contracts there is a higher level of
transparency risk due to the fact that, in this business venture, the financial
institution has very limited management rights. In general, however, most
of the Islamic financial products and deals are based on goodwill and must
be transparent, so that the financial institution knows its real exposure to
potential losses.
Murãbaha
The Murãbaha contract involves three parties: the financial institution, the
buyer (client), and the seller; which is exposing the financial institution to
all types of risks. Therefore, Murãbaha contracts may initiate an exposure
to operational risks due to a reasonable or unreasonable default of the buyer
to keep its promise for buying the commodity that the financial institution
has purchased on his/her behalf. In the Murãbaha contract, exposure also
arises from credit risks since defaults by the client when they come to repay
the commodity on the maturity day or at agreed periodical times. In addi-
tion, the institution is also exposed to market risk due to commodity price
fluctuation even before it actually re-sells the commodity to the client/
buyer. Note, however, that market risks are mainly initiated from failures to
estimate the actual margin on profit set in Murãbaha contracts than those
based on benchmark or mark-up rates.
Ijãrah
The Islamic financial contract that refers to leasing of commodities or
services is called an Ijãrah contract. This involves two parties: the lessor, an
individual or organisation which leases out/rents out the property or service;
and the lessee, the financial institution that rents from the lessor. By being
involved in Ijãrah contracts, financial institutions are exposed to credit,
operational, market, and liquidity risks. Thus, any defaults on payments due
to business risk, market risk, or even due to the lessee’s unconsciousness, is
exposing the financial institution to credit risks. Any early exit of the
renter/lessee from the contract results in a discontinuation of payments and
thus exposes the financial institution to credit risk. This also results in
liquidity risks. The institution is also exposed to operational risk if
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 77
catastrophic events arise or should catastrophic events arise during the of
Ijãrah contracts. Moreover, as the financial institution is the owner of the
assets, it is exposed to operational risks, e.g. due to any damages to these
assets. Any mismatch between the rent price and the actual market price of
the commodity or service exposes the financial institution to market risk.
Note that the prices in Ijãrah contracts are pre-determined. Finally, financial
institutions are also subjected to additional credit and market risks when an
Ijãrah contract is set for payment at the maturity day.
Salam
The Salam contract is a sale in which the buyer, who is the financial institution,
makes advance payments; however, the delivery from the seller on the
Salam contract is defined some time in the future. The Salam is a type of
forward contract where the three parties – the financial institution, the
seller, and the buyer – receive several benefits, i.e. by fixing the price of the
commodity, storage costs, etc. On the other hand, financial institutions that
are providing Salam contracts are exposed to all main types of risks, including
operational, credit, market, and liquidity risks. Any failures on delivering or
delays on the commodity at the agreed date or on the agreed specifications
may initiate expenses as well as losses for the current business and future
opportunities. As a result, Salam contracts may expose the financial institu-
tions to operational risks in addition to reputational and business risks. In
addition to collateralisation policies, such as insurance, determined for the
Salam contracts, financial institutions may minimise such types of risks by
using references as well as guarantees in regards to the commodities’ qual-
ity standardisation. By providing Salam contracts, the financial institutions
are also exposed to credit risk due to any default from the buyer to pay the
commodity at the agreed price. Moreover, as a result of commodity price
fluctuation and mark-up risks, the financial institution may also be exposed
to market risks. Furthermore, any failure in forecasting the future price of
the commodity at which the institution is planning to resell and is based on
mark-ups and benchmarks may expose the institution to reinvestment risks.
As already mentioned, different market scenarios and strategies as well as
mathematical techniques such as VaR can be applied to evaluate and min-
imise these market and mark-up risks. Finally, a parallel Salam may be
employed as a hedge strategy to financial risks.
Istisnã
Financial deals that are structured for forming a relationship between the
financial institution and a manufacturer of commodities or construction
industry for building infrastructure are driven by the Istisnã financial
78 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
contracts. In Istisnã financial contracts, the institution makes advance
payment, but the delivery of the commodity or the asset is set in the future.
After receiving the commodity or the constructed asset, financial institutions
are selling them for cash on deferred prices. One of the most intriguing
characteristics of the Istisnã contract is that it is one of the most flexible
financial Islamic contracts in terms of the volume and the time of payment
instalments. There are two types of defaults that may arise during the life-
time of the manufacturer or constructor supported by the Istisnã contracts;
these are any inabilities and/or any delays for carrying on with the process
of producing the commodities or constructing the assets. Such risks are
linked mostly to business operations and may result in significant losses as
well as liquidation problems and negative reputational records for the
financial institution. Similar to the Salam contracts, in the Istisnã contracts
any failures to arrive at the agreed quality of the goods may result in losses
due to delays or even termination of the delivery of the commodity or assets.
Such failures could also result in opportunity losses as well as reputational,
administrative, and legal costs. Monitoring the manufacturing and construc-
tion processes may minimise such types of risks. Moreover, an efficient
production quality management system could also act as a guarantee for
avoiding failures and delays. Financial institutions should also look for any
references on previous manufacturing or construction projects by the man-
ufacturer or construction industry. Finally, a collateralisation via insurance
policies may also be applied. The credit (counterparty) risk arises when the
buyer or user of the commodity or asset is unable to follow the agreed
payment obligations set in the Istisnã contract. In addition to the above
risks, financial institutions that are dealing with Istisnã contracts may also
be exposed to market risks as the price of the manufactured commodities
and constructed assets which are driven from the mark-ups may fluctuate to
a value that highly defers from the actual market price. As common practice
to most of the financial products, VaR analysis is an important method to
evaluate the significance of the above-mentioned market risks.
NOTES
1. Basel Committee of Banking Supervision, ‘Principles for the Management of the Credit Risk, Bank
for International Settlement’, Basel, Switzerland (1999), p. 1.
2. J. Bessis, Risk Management in Banking (John Willey & Sons, 2002), p. 15.
3. Bank for International Settlement, Basel Committee of Banking Supervision, ‘The supervisory treat-
ment of market risks’, Basel, Switzerland (1993), Annex 1, p. 45.
4. Bank for International Settlement, Basel Committee of Banking Supervision, ‘International
Convergence of Capital Measurement and Capital Standards’, Basel, Switzerland (2006), p. 144.
5. T. Davies, Modern Risk Challenges (GARP, 2004), Issue 19, pp. 31–33.
6. Bank for International Settlement, Basel Committee on Banking Supervision, ‘Operational Risk
Management’, Basel, Switzerland (1998), p. 2.
RISK ISSUES IN ISLAMIC FINANCIAL CONTRACTS 79
7. V. Sundararajan and L. Errico, ‘Islamic Financial Institutions and Products in the Global Financial
System: Key Issues in Risk Management and Challenges Ahead’, IMF Working Paper
No. WP/02/192, International Monetary Fund (2002).
8. M. U. Chapra and T. Khan, ‘Regulation and Supervision of Islamic Banks’, Occasional Paper
No. 3, Islamic Research & Training Institute, Saudi Arabia (2000).
9. F. Al-Omar and A. H. Mohammed, Islamic Banking, Theory, Practice and Challenges (London:
Oxford University, 1996).
10. T. Khan, ‘Risk Management in Islamic Banking: A Conceptual Framework, distance learning
lecture’, Islamic Research and Training Institute (2004).
11. M. U. Chapra and T. Khan, ‘Regulation and Supervision of Islamic Banks’, Occasional Paper
No. 3, Islamic Research & Training Institute, Saudi Arabia (2000), pp. 52–54.
12. Capital Adequacy Standard for Institutions (Other Than Insurance Institutions) Offering only
Islamic Financial Services, Islamic Financial Services Board (2005), pp. 6–7.
CHAPTER 3
Basel II and IFSB for
Islamic Financial Risk
INTRODUCTION
Financial liberalisation, which was a part of globalisation, was keenly
followed by developing countries in the 1990s. Several restrictions were
eased, paving the way for easy entry of financial institutions. Multinational
financial institutions from developed countries entered developing
economies. Relaxing governance was considered as an essential part of this
liberalisation. Self-regulation was considered to be the motivating factor.
But everything did not work well. There were several instances of malprac-
tice, financial frauds and, finally, some failures. A study by Demirgüç-Kunt
and Detragiache found that there was strong correlation between financial
liberalisation and banking crisis.1 Governing bodies started looking at the
existing set of standards and ways to overcome the issue of balancing
control and freedom. It was realised that the existing standards were not
sufficient and hence revision and additions were needed for bringing in
sound financial risk-management practices. From simple capital provisions
for risky assets to a comprehensive framework for risk management, practice
of risk management has undergone wholesale transformation over the past
several years. To protect financial systems from failure, special regulatory
provisions are created on a regular basis. Each country has its own set of
regulations based on several parameters. The most common among them is
the requirement to hold some minimum capital indexed to the activities of
the bank.
The major cushion of safety and soundness arrives from the capital held
by the bank; thus, capital adequacy regulations are an important part of the
regulatory provisions. The larger the capital, the better the protection
against insolvency can be. But holding large capital leads to increased cost,
since capital is a costly option. A bank which is over-capitalised will have
low return on its capital and will not be able to pay decent dividends on
80
BASEL II & IFSB FOR ISLAMIC FINANCIAL RISK 81
investment to its shareholders. At the same time, under-capitalised banks
are highly prone to the risk of insolvency and can also suffer from
retarded growth. Thus, arriving at an optimal level of capital is in the best
interest of banks and the shareholders. Regulators attempts to monitor
the level of capital maintained by the banks by defining the mandatory
levels of capital known as regulatory capital. Capital adequacy is at the
core of the supervisory activities of the Central Bank all over the world.
It is an important benchmark for the soundness of the financial institu-
tion. Capital adequacy is considered to be a better tool compared to
deposit-reserve requirement to control lending. During the eighties, there
were several Central Banks who accepted a minimum limit of 8 percent
of risk-weighted capital as being safe, which was proposed by the Basel I
accord.
The previous two chapters have dealt with the evolution of Islamic
finance and the major types of risks involved in conventional and Islamic
finance. The present chapter provides a brief review of the Basel II
accord (2006) and hence is largely based on the document issued by the
Basel Committee on Banking Supervision. A brief summary of the
original accord (1988) is presented, bringing out the major limitations of
the original accord. An analysis of the three Pillars of Basel II and their
applicability for Islamic financial institutions are also presented. The
minimum capital requirement under Pillar 1 is examined and its rele-
vance for Islamic financial institutions is discussed. Credit risk as
perceived and presented by Basel II is similar in Islamic financial insti-
tutions, but varies in origin and intensity. The Standardised Approach and
Internal Rating Based Approach are also discussed and their applicability
to Islamic finance is analysed. The three alternative approaches as pro-
posed under Basel II for operational risk, Basic Indicator Approach,
Standardised Approach, and Advanced Measurement Approach, are also
presented in brief in order to understand their applicability to Islamic
finance. A review of different risk weights as allocated under Basel II for
the Standardised Approach in the light of risk exposures for Islamic
financial contracts is also presented in this chapter. The risk exposures of
Islamic financial institutions in the case of market risk are spread to
cover all major types of financial contracts, as discussed in chapter 2. The
Basel II approach to market risk is discussed in the light of these
exposures. The Second Pillar of Basel II, the supervisory review, is
equally important for the Islamic financial institutions and is discussed in
this chapter. Market disclosure requirements as proposed by Basel II are
presented and analysed in the light of Islamic financial disclosure needs.
The Islamic Financial Services Board (IFSB) has issued risk manage-
ment standards for Islamic financial industry, which are discussed in
detail in the last section of the chapter.
82 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
THE BASEL ACCORD
The Bank for International Settlements (BIS) was established on 17th May
1930 and is the oldest international financial organisation. It serves as the
bank for the central banks. It provides a platform for consultative co-
operation among the central banks. The role of BIS has undergone change
as per the needs of the international financial sector. BIS now also acts as an
institution for collection, compilation, and dissemination of economic and
financial statistics. It actively promotes global financial stability. It also per-
forms the traditional banking function for the central bank community (gold
and foreign exchange transactions). It has several committees working on
different aspects of international financial stability. The Basel Committee
for Banking Supervision (BCBS, henceforth referred to as the Committee)
was formed at the end of 1974 by the Governors of G-10 countries. The
Committee issued a series of documents beginning from 1975 on banking
supervision.
The original accord was the first major international capital adequacy
standard released by BIS with the focus on credit risk.2 A minimum capital
of 8 percent of risk-weighted asset was prescribed by the Committee.
A simplified risk-weighting structure was given. For example, 0 percent for
cash, 20 percent for claims on multilateral development banks, 50 percent
for residential mortgages, and 100 percent for loans to private sector. The
original accord served the purpose of bringing capital adequacy to the
centre of the banking supervision. However, it was soon proved insufficient
and rendered obsolete with rapid changes in the financial and banking
sector. Several deficiencies in the existing accord were found and thus
amendments were introduced in 1996 with additional focus on market risk.3
It covered the four major risk categories of market risk: (a) interest rate-
related instruments, (b) equities, (c) foreign exchange risk, (d) commodities
risk. Technology-driven banks were more prone to operational risk and the
amended accord failed to take this type of risk into account. Moreover, there
were large-scale changes in the banking industry all over the world.
Globalisation and liberalisation introduced new elements of risks in bank-
ing. Changing supervisory practices also brought about changes in the risk
exposures. Several limitations in the existing standards were realised. First,
it prescribed a single rate for capital adequacy, irrespective of the degree of
risk within the category (‘one size fits all’ approach). Second, there was no
incentive for quality credit; hence, there was no distinction between ‘good’
and ‘bad’ credit in terms of provision of capital. Third, it failed to ade-
quately recognise operational risk. Finally, risk in newer financial instru-
ments such as securitisation of assets and derivatives was not addressed.
Several rounds of consultations took place at BIS with Central Banks,
bankers, academicians, researchers, and industry experts. Hence, after
BASEL II & IFSB FOR ISLAMIC FINANCIAL RISK 83
many amendments and taking into account all these factors, a comprehen-
sive version of the Basel II revised framework was documented in 2006.4
The contribution of the Committee has been noteworthy in promoting bet-
ter supervision. Persistent effort has not only brought changes in the way
risk was perceived but also forced banks to adopt best standards, promoting
better financial health for banks and financial institutions. The accord has
been voluntarily adopted by more than one hundred countries. The new
Basel II accord is now considered as a benchmark for financial risk
management.
RISK MANAGEMENT ACCORDING TO
BASEL II AND THE ISLAMIC
FINANCIAL INDUSTRY
One of the objectives of Basel II is to improve the risk management
practices of internationally active banks, thus promoting international
financial soundness and stability. There has been substantial improvement
in Basel II as compared to the original accord in relation to recognition of
risk. As seen in the previous paragraphs, the new accord recognises the
importance of operational risk as well as providing a better understanding
of market and credit risks. Basel II marks a shift from transaction-based
supervision to risk-based supervision. It provides a wide range of risk
weights which can be applied to calculate the required capital. It also
provides a range of menus to choose from for the approach towards risk
measurement. Basel II has covered a wide range of risks which were not
previously included in the original accord, such as country risk, legal risk,
liquidity risk, and reputational risk. The emphasis has been on efficient
capital management. The importance of developing internal capabilities for
risk assessment has been recognised and Basel II has an undertone of devel-
oping an organisational culture of risk management rather than a simple
framework. Basel II has improved response to risk management from four
different perspectives. First, it provides improved risk measurement
methodologies by incorporating the latest innovations in financial engineer-
ing. Second, it provides improved risk monitoring by providing a wide
range of indicators which can be used by the regulators. Third, it attempts to
bring to the forefront the importance of reporting by including the third
Pillar, which is market disclosure. Finally, it bridges the gap in risk
management by introducing the role of supervision at the centre of risk
management in Pillar 2. The delicately balanced three Pillars of Basel II are
reinforcing on each other and provide the required support for developing a
sound, stable, and dynamic risk-management structure.
84 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
The Islamic financial industry is a part of the global financial industry
and hence cannot remain unconcerned and unaffected. The capital adequacy
is going to remain the core issue for risk management, whether it is a con-
ventional bank or an Islamic bank. The overall impact of a failure in either
a conventional bank or an Islamic bank will be felt by the entire financial
market and thus risk management needs to be integrated. The common con-
cept of having sufficient capital cannot be refuted in Islamic finance. Thus,
application of Basel II is a matter of adoption of the standard to the needs of
the Islamic financial industry. The three pillars of Basel II are discussed in
the following paragraphs, along with their relevance and applicability to
Islamic finance.
THE THREE PILLARS OF BASEL II
The Basel II accord is based on three mutually enforcing ‘Pillars’. The
strong relevance of supervisory review and market discipline has been
introduced along with the fundamental minimum capital requirement as
three Pillars (Figure 3.1). The strength of the Basel II lies in the mutual
support which is provided by each of the three Pillars. By including the
supervisory review, the Committee has brought risk management into the
realms of supervisors, who were in the past not directly related to risk
management. It also re-emphasised the role of supervisors in overall risk
management by providing them with an upper hand in the decision relating
to the validity of the minimum capital calculation. Market discipline helped
in bringing out the importance of transparency within the range of risk-
management activities. The three Pillars together provide a flexible, risk-
sensitive capital management framework, which can change the way risk is
perceived by financial organisations.
The role of minimum capital requirement is equally important in the
Islamic financial industry, although the calculation of the capital charge
may differ, depending on the risk exposures. The role of supervisors is more
critical due to the evolving nature of the Islamic financial industry. Strong
regulatory support in the form of monitoring and assistance is needed for
the Islamic financial industry. The social responsibility is of utmost impor-
tance in Islamic finance. Activities which are conducted by the Islamic
financial institution should not be Haram and, when profits are generated,
the Zakat should be paid. Along with this, there is a greater emphasis on the
transparency and thus the third Pillar of Basel II has more relevance for the
Islamic financial industry.
When estimating the minimum capital requirements, there are two types
of capital that can be calculated by financial institutions: economic capital
and regulatory capital. Economic capital is the amount of capital estimated
BASEL II & IFSB FOR ISLAMIC FINANCIAL RISK 85
Pillar 1
Minimum capital Pillar 2 Pillar 3
requirements Supervisory review Market discipline
Calculation of capital Supervisors can review and The details of risk
charge is connected to revise the charge if they management are
credit risk, market risk, feel that the calculated required to be made
and operational risk, charge does not public by the bank,
signifying higher adequately reflect and improving the
charge for higher risk cover the risk. sharing of information
in each category. within the industry.
Figure 3.1 The three pillars of Basel II accord
by the bank’s management that is required to cover and protect the
shareholders from potential economic losses (unexpected negative changes
in economic value). Regulatory capital is the amount of capital that the reg-
ulators require the institutions to hold as a safety against risks. The primary
purpose of economic capital is to support transaction-level decision making,
whereas the main intention of regulatory capital is to assess the bank’s
financial capability to manage its intrinsic risks. As opposed to regulatory
capital, which is set by the regulators, economic capital is the internal
estimate of capital required to be maintained. It is the required capital for an
unregulated institution, which may or may not be the same as the regulatory
capital required for that same institution. If a bank is constrained by regula-
tory capital, it should use the regulatory capital for its assessment; and if the
bank is constrained by economic capital, it should use the economic capital.
Basel II is attempting to align economic and regulatory capital more closely
to reduce the scope for regulatory arbitrage. Setting a higher limit for eco-
nomic capital provides some room for leverage for bankers. The Basel II
stresses on a complete risk-management framework which can tie regula-
tory capital with the economic capital.
Islamic financial institutions have a more difficult task balancing
between the economic capital and statutory capital. Non-availability of liq-
uidity support from the money market, non-existence of inter-bank money
market for Islamic banks, and absence of a secondary market created more
pressure on the capital requirement of the Islamic financial institutions.
This, coupled with the commitment for current account and investment
account holders, exposes the Islamic financial institutions to larger capital
requirements.
Pillar 1: Minimum capital requirement
The first Pillar of Basel II deals with the minimum capital requirement. This
capital is the regulatory capital. The calculation of the minimum capital
86 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
is presented with the help of the capital adequacy ratio (CAR), which is
defined as a ratio with the numerator representing the amount of available
capital and the denominator a measure of the risk faced by the bank. Basel
II defines the CAR as the ratio of the bank’s capital (Tier I and Tier II) to
its risk-weighted assets and it should not be lower than 8 percent. As com-
pared to Basel I, there is no change in the numerator and the value of the
ratio (8 percent). The change is in the denominator, by assigning the risk
weights to different business activities faced by the bank. Symbolically,
CAR is,
Tier I capital Tier II capital
CAR 8% (3.1)
Risk weighted assets
The Tier I capital must be at least 50 percent of the total capital and it
includes paid-up share capital and disclosed reserves. Tier II capital, which
can be a maximum 100 percent of Tier I, includes undisclosed reserves,
asset revaluation reserves, general provisions and general loan-loss
reserves, hybrid capital instruments, and subordinated term debt. The treat-
ment to market risk from trading activities remains unchanged and there are
substantive changes in the treatment of credit risk and introduction for
explicit treatment of operational risk. There are several alternative
approaches proposed by Pillar II for the treatment of the three financial risks
(Table 3.1).
Having discussed the different aspects of credit risk, market risk, opera-
tional risk, and liquidity risk in chapter 2, it is evident that the risk exposures
of the Islamic financial institution differ from the conventional financial
institution. The risks in Islamic financial institutions are more contract-
centric, rather than conventional product-centric. Even though the lines of
business can be grouped into retail, corporate, and other, the risk exposures
among them are not similar, as in conventional banks. The capital as it is
classified in conventional banking as Tier I and Tier II cannot be used in
Islamic banking since the capital in Islamic banking consists of different
Islamic product groups.
Credit risk
The Basel II approach to credit risk is very exhaustive. Several options with
detailed guidelines are provided. The Committee suggests choosing
between two prime methodologies: Standardised Approach (SA) and
Internal Rating Based (IRB) Approach. The first is an ‘out-of-the-box’ kind
of solution which can be directly applied, whereas the second requires
explicit approval of the bank’s supervisors.
BASEL II & IFSB FOR ISLAMIC FINANCIAL RISK 87
Table 3.1 Alternative approaches to credit, market, and operational risk
Type of risk Proposed approaches Specific approaches
Credit risk Standardised
Approach
Internal Rating Based Foundation Approach
Approach Advanced Approach
Securitisation Standardised Approach
Approach Internal Rating Based
Approach
Market risk Standardised
Measurement
Approach
Internal Models
Approach
Operational risk Basic Indicator
Approach
Standardised Alternative Standardised
Approach Approach
Advanced Internal Measurement
Measurement Approach
Approach Loss Distribution Approach
Scorecard Approach
Standardised Approach
The Standardised Approach (SA) is based on observable characteristics of
the exposure. It provides fixed-risk weights for each supervisory category.
Banks may use external ratings if approved by the national supervisors. For
loans considered past-due, the recommended risk weight is 150 percent,
unless already provided for. In the absence of any external rating, a risk
weight of 100 percent is recommended. A new element of ‘credit mitigants’
consisting of an extended range of collaterals has been added to the Basel II
accord. It also expands the range of external guarantors who meet certain
minimum criteria of external credit ratings. The Retail segment has been
modified to the extent that the risk weights for household mortgages are
reduced and some loans for small- and medium-sized enterprises can be
included in the retail segment.
Internal Rating Based Approach
There are two approaches which have been added in the new accord: they
are foundation and advanced IRB. Banking-book exposures are to be
88 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
classified as (a) corporate, (b) sovereign, (c) bank, (d) retail, and (e) equity.
The four quantitative inputs needed for IRB calculations for exposure to
corporate, sovereign, and bank entities are: Probability of default (PD),
which measures the likelihood that the borrower will default over a given-
time horizon; Loss given default (LGD), which measures the proportion
which will be lost from exposure in case the default occurs; Exposure at
default (EAD), which for a loan commitment measures the amount of the
facility that is likely to be drawn if a default occurs; and Maturity (M),
which measures the remaining economic maturity of the exposure.5 For
retail exposures there is a single advanced IRB approach using the three
inputs, PD, LGD, and EAD. Specialised lending is dealt with separately.
Specialised lending has five sub-classes: (a) project finance, (b) object
finance, (c) commodity finance, (d) income-producing real estate, and
(e) high volatility commercial real estate. Under retail, the three sub-classes
of exposures are: (1) exposures secured by residential properties, (2) quali-
fying revolving retail exposures, and (3) all other retail exposures. Banks
working with IRB need to treat the equity exposures separately. The two
approaches have some fundamental differences. Foundation IRB is based
more on values set by the Committee, whereas the Advanced IRB largely
uses estimates drawn by the bank.
Securitisation framework
According to the Committee, exposures arising from traditional and
synthetic securitisation of similar structures that contain features common
to both should apply a securitisation framework. The capital treatment of
securitisation exposure must be determined on the basis of economic
substance and not on legal substance. Banks must hold capital against all
securitisation exposures. There are two approaches provided by the
Committee for a securitisation framework: SA and IRB. The use of IRB is
subject to the approval of the bank’s supervisors.
Islamic financial institutions’ exposure to credit risk is different and
unstructured as compared to conventional financial institutions. Credit risk
arises at different time intervals in a contract, and thus is highly dynamic
and difficult to accurately estimate. Credit risk is more concentrated in
contracts which are large in number. Both the approaches, standardised
and IRB, as proposed by Basel II, cannot be applied to Islamic banks with-
out several modifications. The applicability, even after modifications, does
not appear to be clear. The highly distributed credit risk in Islamic banking
across different products and time-lines warrants the need for an alterna-
tive or an amended approach to the Basel II methodology. A more detailed
analysis of the estimation of credit risks in Islamic finance is presented in
chapter 4.
BASEL II & IFSB FOR ISLAMIC FINANCIAL RISK 89
Market risk
The Committee described detailed methods for the calculations of capital
charges for (1) interest rate risk, (2) equity position risk, (3) foreign
exchange risk, and (4) commodities risk. It provides several alternatives for
assessing capital charge for options. The capital charge for interest rate is
applied to the current trading book items. The capital charge for foreign
exchange risk may exclude structured foreign exchange positions. The com-
mittee has prescribed two alternative models to measure market risk: the
Standardised Measurement Approach and the Internal Model Approach.
Standardised Measurement Approach
The Standardised Measurement Approach deals with interest rate risk,
equity position risk, foreign exchange risk, commodities risk. It also
provides guidelines for the treatment of options. The minimum capital
requirement for interest rate and equity exposures is based on two different
charges calculated for ‘specific risk’ and ‘general risk’. Regarding the
foreign exchange risk exposures, there are two processes which are
prescribed by the Committee. The first measures the exposures in a single
currency and the second measures the risk inherent in long and short
positions in different currencies. Commodity risk covers all the commodi-
ties except gold, which is covered under foreign exchange risk. The
Committee addresses the complexity involved in the calculation of com-
modity risk. The commodity risk includes basis risk, interest rate risk, and
forward gap risk. There are two approaches prescribed under the calculation
of commodity risk exposures: a simplified approach and a maturity ladder
approach with a time-band of seven steps (0 to 1 month, 1 to 3 months, 3 to
6 months, 6 to 12 months, 1 to 2 years, 2 to 3 years, and over 3 years). In
both cases, the methodology adopted for measuring commodity risk should
encompass directional risk, forward gap, and interest rate risk as well as
basis risk.
Internal Model Approach
The use of the Internal Model Approach (IMA) is subject to the approval of
the bank’s supervisors. There are several qualitative criteria prescribed by
the Committee for using IMA. These include: a regular back-testing
programme, on-going validation of internal risk models, a routine and
rigorous programme of stress testing, and an independent review of risk
management systems. For interest rate, the risk measurement system should
model the yield curve using one of the generally accepted approaches and it
must incorporate separate risk factors to capture spread risk. For foreign
90 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
exchange, the risk measurement system should incorporate risk factors
corresponding to the individual foreign exchange currencies. A more
detailed approach for different industry sectors can be used for equity
risk. For commodity prices, risk factors should correspond to each com-
modity market. Financial institutions using IMA must compute ‘Value at
Risk’ on a daily basis using the 99th percentile with one-tailed confidence
interval.
Classifying the risk exposures according to the Pillar 1 requirements is
thus the first step towards ascertaining the capital adequacy. A template has
been provided in Table 3.2 for the calculation of the total risk exposures
under the three risks – credit, operational, and market – as described by
Pillar 1. The template provides all alternatives in one place and thus can be
used for all approaches; however, it is only suggestive and should be used
after suitable adaptation.
Market risk exposures are mainly concentrated in commodity prices and
foreign exchange in Islamic banking. Interest rate risk is not present in
Islamic financial institutions directly, although it affects the position of the
institution through mark-up rates. Equity price risk in Islamic financial
institutions is through Mushãrakah and Mudãrabah financing. The Basel II
recommendations of SMA and IMA do not fit directly for Islamic banking.
Commodity risk is clearly visible in almost all types of contracts in Islamic
banking, such as Mushãrakah, Mudãrabah, Murãbaha, Salam, Istisnã, and
Ijãrah, as discussed in detail in the previous chapter. Due to the absence of
regular hedging tools such as derivatives, the exposure of Islamic banks for
foreign exchange is larger and more difficult to manage. A detailed analysis
of market risk management in Islamic financial institutions is presented in
chapter 5.
Operational risk
One of the major changes in the Basel II accord has been the recognition of
operational risk as an important category of risk. With increasing reliance
on technology, operational risk has acquired new dimensions. Operational
risk is another area where a new approach has been developed under Basel II.
Although a new field, reasonable clarity exists as far as the application of
operational risk is concerned in the Basel II accord. Operational risk under
Basel II includes legal risk, but excludes strategic risk and reputational risk.
There is loss event-type classification provided by the Committee which
can be used as a base to develop the bank’s own classification based on the
organisational needs (Table 3.3). The Committee stresses the need for
adequate data collection to manage operational risk and provides three
approaches for calculating operational risk capital charge with increasing
BASEL II & IFSB FOR ISLAMIC FINANCIAL RISK 91
Table 3.2 Template for the calculation of the credit risk
exposures as per Pillar 1 under Basel II
Credit risk
Standardised Approach
– Corporate
– Banks
– Sovereign
– Commercial real estate exposures
– Residential real estate exposures
– Other real estate exposures
– Qualifying revolving retail exposures
– Other retail exposures
Total standardised approach
Foundation IRB
– Corporate, sovereign and bank
– Equity portfolios
– Residential mortgage
– Qualifying revolving retail exposures
– Other retail exposures
Total foundation IRB approach
Advanced IRB
– Corporate, sovereign, and bank
– Equity portfolios
– Residential mortgage
– Qualifying revolving retail exposures
– Other retail exposures
– Total Advanced IRB approach
Securitisation exposures
Total of credit risk exposure
Operational risk
Operational risk under
– Basic Indicator Approach
– Standardised Approach
– Advanced Measurement Approach (AMA)
– Total of operational risk exposure
Continued
92 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Table 3.2 Continued
Market risk
Market risk under
– Standardised Approach
– Interest rate risk
– Equity position risk
– Foreign exchange risk
– Commodity risk
Total of Standardised Approach
– Internal Models Approach
Total of market risk exposure
sophistication and risk sensitivity. The Basic Indicator Approach and
Standardised Approach are directed towards banks with less significant
operational risk exposures. The Advanced Measurement Approach is
recommended for banks with a developed risk management framework.
Basic Indicator Approach
The Basic Indicator Approach (BIA) is a simple approach provided by the
Committee for banks with non-significant risk exposures. This approach
provides for a standards capital charge of average annual gross income over
the past three years multiplied by a factor of 0.15 (known as alpha), which
is given by
KBA [兺(GI1…n ␣)] n (3.2)
where:
KB/A the capital charge under the Basic Indicator Approach
GI annual gross income, where positive, over the previous three
years
n number of the previous three years for which gross income is
positive
␣ 15%, which is set by the Committee, relating to the industry-
wide level of required capital to the industry-wide level of the
indicator.
There is no criterion prescribed under Basel II for using BIA except
that banks that use BIA are encouraged to comply with the Committee’s
guidelines on Operational Risk as provided in Sound Practices for the
Management and Supervision of Operational Risk.6 It is recommended
BASEL II & IFSB FOR ISLAMIC FINANCIAL RISK 93
Table 3.3 Detailed loss event type classification
Event type Category
Internal fraud Unauthorised activity
Theft and fraud
External fraud Theft and fraud
System security
Employment practices and Employee relations
workplace safety Safe environment
Diversity and discrimination
Clients, products and business Suitability, disclosure, and fiduciary
practices Improper business or market
practices
Product flaws
Selection, sponsorship, and
exposure
Advisory activity
Damage to physical assets Disaster and other events
Business disruption and system Systems
failure
Execution, delivery, and process Transaction capture, execution, and
management maintenance
Monitoring and reporting
Customer intake and documentation
Customer/Client account
management
Trade counterparties
Vendors and suppliers
Source: BCBS, International Convergence of Capital Measurement and Capital Standards,
Basel Committee on Banking Supervision, Basel, June 2006, Annexure 9, pp. 305–307
that banks using BIA should attempt to move on to SA or AMA in the
future.
Standardised Approach
The Standardised Approach (SA) is superior to BIA in terms of risk
sensitivity. It provides specific risk weights for each business lines. Eight
business lines are prescribed in the Basel II document. A factor beta ()
representing a proxy for the industry-wide relationship between the
94 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
operational risk loss experience and gross income for the specific business
line is prescribed. The total capital charge is given by:
KTSA {兺 years 13 max [ 兺 (GI 18  18 ), 0]} / 3 (3.3)
where:
KTSA the capital charge under the Standardised Approach
GI1–8 annual gross income in a given year, as defined in the Basic
Indicator Approach, for each of the eight business lines
1–8 a fixed percentage, set by the Committee, relating to the level of
required capital to the level of the gross income for each of the
eight business lines. The values are described in Table 3.4.
The main differences between the two approaches are the risk weight
assigned for the business. In BIA, a common risk weight of 15 percent is
allocated for all the business activities. More risk-sensitive values ranging
from 12 percent to 18 percent are used in SA, depending on the business
lines. Hence, banks have an incentive to move on to SA.
Advanced Measurement Approach
Under the Advanced Measurement Approach (AMA), the risk measure is
generated by the bank’s internal operational risk measurement system using
quantitative and qualitative criteria prescribed by the Committee. The use of
AMA is subject to approval of the bank’s supervisors. Under AMA, a bank
Table 3.4 Beta factors for the different business lines
Business lines  factor
Corporate finance (1) 18%
Trading and sales (2) 18%
Retail banking (3) 12%
Commercial banking (4) 15%
Payment and settlement (5) 18%
Agency services (6) 15%
Asset management (7) 12%
Retail brokerage (8) 12%
Source: BIS, International Convergence of Capital Measurement
and Capital Standards, A Revised Framework, BCBS, Basel, June
2006, p. 147
BASEL II & IFSB FOR ISLAMIC FINANCIAL RISK 95
is allowed to use risk mitigation to an extent (maximum 20 percent) of
the total operational risk capital charge. Some of the criteria for using
AMA are:
The bank must have an independent risk management function.
The bank’s internal operational risk measurement system must be
closely integrated into its day-to-day risk management processes.
There must be regular reporting of operational risk exposures and loss
experiences.
The bank’s operational risk management system must be well
documented.
Internal and/or external auditors must perform regular reviews of the
operational risk management processes and measurement systems.
If the minimum regulatory capital requirement is to be based on
unexpected losses alone, then the bank must demonstrate that it has
measured and accounted for its expected losses through internal
business practices.
The bank’s risk measurement system must be sufficiently ‘granular’ to
capture the tail of the loss estimates.
The bank needs to have a credible, transparent, well-documented,
and verifiable approach for weighting the fundamental elements,
including: use of internal data, relevant external data, and scenario
analysis.
Operational risk exposures appear to be higher in Islamic banks. The BIA
approach as indicated by Basel II does not appear to be a case of perfect fit
for Islamic banking. The 15 percent provision for operational risk of the
average of three years gross income needs to be examined thoroughly
before any implementation. The different risk weights as proposed in the
SA under Basel II are not entirely applicable to Islamic banking. The
allocation of 18 percent risk weight for business lines such as corporate
finance, trading and sales, and payment and settlements may not represent
the true picture of the risk exposures of Islamic banks as trading and sales in
Islamic banking may include some Murãbaha transactions and some
exposures from financing of large accounts through Istisnã. The 15 percent
96 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
risk weight allocated to commercial banking and agency services needs to
be further examined before applying to Islamic banks. Retail banking, asset
management, and retail brokerage has been allocated 12 percent risk weight
in SA, which does not fully apply to Islamic banking. As discussed in
chapter 2, Murãbaha financing is most commonly used for retail funding,
such as car financing, etc. The risk exposures differ greatly during different
stages of the contract and hence a blanket weight of 12 percent does not
represent a true risk exposure. Istisnã, which is generally used to finance
large accounts, appears to carry high risk from the analysis presented in
chapter 2, Figure 2.15. The risk weight of 18 percent, if classified as
corporate, or 12 percent, if classified as retail, does not appear to map the
exposure completely. Operational risk management is further discussed in
chapter 6.
PILLAR 2: SUPERVISORY REVIEW
The second Pillar of Basel II goes beyond the importance of supervisory
review and establishes it at the core of risk management activity. It is
believed that, as risk profiles change due to increasing complexities of the
business, banks may take greater risks by hiding behind the existing risk
framework, which may not be adequate to manage the new exposures. The
role of supervisors is thus important in reviewing the constantly changing
risk positions. The supervisory review process is to ensure that banks
develop and use better risk management techniques. The Committee cau-
tions that increased capital should not be taken as the only option for
addressing increased risks confronting the bank. It advises the use of other
means to address risks, such as: strengthening risk management, applying
internal limits, strengthening the level of provisions and reserves, and
improving internal controls. Capital should not be treated as a substitute
for inadequate control or risk management processes. It focuses on three
main areas: credit concentration risk, Interest Rate Risk (IRR) in banking
books, and business cycle risks. The four key principles of supervisory
review are:4
1. Banks should have a process for assessing their overall capital
adequacy in relation to their risk profile and a strategy for maintaining their
capital levels.
2. Supervisors should review and evaluate the banks’ internal capital
adequacy assessments and strategies, as well as their ability to monitor and
ensure their compliance with regulatory capital ratios. Supervisors should
take appropriate supervisory action if they are not satisfied with the result of
this process.
BASEL II & IFSB FOR ISLAMIC FINANCIAL RISK 97
3. Supervisors should expect the banks to operate above the minimum
regulatory capital ratios and should have the ability to require the banks to
hold capital in excess of the minimum.
4. Supervisors should seek to intervene at an early stage to prevent cap-
ital from falling below the minimum levels required to support the risk char-
acteristics of a particular bank and should require rapid remedial action if
capital is not maintained or restored.
Regarding the interest rate risk in the banking book, the Committee held
the opinion that it is most appropriate at present to treat it under Pillar 2 of
the framework. For credit risk, the bank must ensure that it has sufficient
capital to meet the Pillar 1 requirements. Supervisors must review the credit
risk stress test. When the banks use credit risk mitigation techniques, they
give rise to residual risks which include legal risks, documentation risks,
liquidity risks, etc. The supervisors should be concerned particularly about
the residual risks when credit risk mitigation techniques are used.
Identifying risk concentrations as the single most important cause of major
problems in banks, and since they are not covered by Pillar 1, the
Committee places the responsibility of reviewing them onto the supervisors.
Risk concentration is any single exposure or group of exposures which can
produce large losses. Since lending is the most common activity of the
banks, credit risk concentrations can be a threat to the stability of the bank.
A Counterparty Credit Risk (CCR) should be in place including identifica-
tion, measurement, management, approval, and internal reporting of CCR.
Regarding operational risks, the Committee suggests that supervisors
should verify if the capital requirement under Pillar 1 is consistent and com-
parable with the industry standards. The Committee recommends enhanced
communication and co-operation among supervisors of different countries.
Some of the recommendations of Basel II in Pillar 2 can be applied to
Islamic financial institutions. Some of the specific recommendations which
can be applied to Islamic financial institutions are:
Strengthening risk management system
Applying internal limits
Strengthening the level of provisions and reserves
Improving internal controls
Focus on credit concentration risk
Focus on business cycle risks
98 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
A few of the other recommendations, although very relevant for
conventional banks, do not hold grounds for Islamic banks in the absence of
supporting risk management models. Liquidity risk, which is classified as
residual risk, is one of the most important risks in Islamic banks.
Conventional banks have a strong inter-bank facility, strong overnight
borrowing facility from the Central Bank, and a matured and liquid second-
ary market. Liquidity management is of paramount importance in Islamic
banks due to the absence of regular inter-bank money market, overnight
credit lines from the Central Bank, and secondary market. Liquidity risk
management is thus at the core of risk management in the Islamic bank.
Hence, Basel II recommendations on liquidity risk management under
Pillar 2 are generally insufficient for Islamic banks. Credit risk concentra-
tion recommendations are applicable to Islamic banks since they have large
credit exposures.
PILLAR 3: MARKET DISCIPLINE
The main objective of Pillar 3 – marker disclosure, is to compliment the
minimum capital requirement and supervisory review process. Market dis-
closure is considered essential since it can provide effective means for con-
sistent and comparable information. Market discipline can contribute to a
safe and sound banking environment as required by the supervisors.
Supervisors can use moral suasion, reprimands, and financial penalties to
enforce disclosures. Realising the broader coverage of disclosures under
accounting standards, the Committee has narrowed the scope of disclosure
under Pillar 3 to avoid conflicts. The disclosures should be based on the
materiality concept. The Committee did not set specific thresholds for
information disclosure and rely on the users’ test as a benchmark. The
disclosures should be made semi-annually. There are clear guidelines
regarding the capital disclosure. For Tier 1 capital, there should be separate
disclosure relating to each category. In case there are any innovative, com-
plex, or hybrid capital instruments in the capital, they should be summarised
under their main features. There are specific guidelines for disclosure of
credit risk exposures. The major quantitative disclosures regarding credit
risk exposures include: major types of credit exposures, geographic
distribution of exposures, industry type distribution of exposures, residual
contractual maturity breakdown of the entire portfolio, amount of impaired
loans, and the amount of exposures under each approach.
The absence of comparable information is one of the main issues in
Islamic financial reporting. Since AAOIFI is not mandatory, there have been
limited implementations, and the problem of non-comparability remains.
Basel II recommendation regarding consistent and comparable information
BASEL II & IFSB FOR ISLAMIC FINANCIAL RISK 99
is highly applicable to the Islamic financial industry. Due to social commit-
ment attached to Islamic finance, there is a special need for market disclo-
sure. Transparency is considered to be at the core of Islamic financial
contracts and thus should also be reflected in reporting. It is mandatory to
report the investment of funds, lines of business, activities, and sources of
revenue. Along with this, as Basel II recommendations under Pillar 3, risk
management details should also be made public. Basel II could not answer
all the risk management issues for Islamic financial institutions; hence, there
has been a need for alternative and supportive standards on risk management
for Islamic financial institutions. The release of the Islamic Financial
Services Board (IFSB) principles on risk management has been an important
step in the direction of risk management in Islamic financial institutions.
IFSB PRINCIPLES OF RISK MANAGEMENT
As discussed in the previous chapter, the risk exposures of Islamic financial
institutions differ substantially as compared to conventional banks. Due to
the PLS nature of contract as well as the changing relationship between
parties over time, a variety of risk exposures are introduced at different
stages of the transaction. Safety and soundness of Islamic financial institu-
tions and the Islamic financial market in general is thus a common cause of
concern among regulators. Conventional models of risk management and
supervision cannot be applied without modifications to the Islamic institu-
tions. The Basel II standards have been developed with a conventional
bank’s perspective and hence do not apply to the Islamic financial institu-
tions without suitable modifications. With the growing size of Islamic
banking and financial industry all over the world, there have been efforts to
develop prudent supervisory norms. Setting up of the IFSB is a consolida-
tion of these efforts. The IFSB started its operations in March 2003 and is
based in Kuala Lumpur. It serves as an international standard-setting body
of regulators and supervisory agencies that have interest in the soundness
and stability of the Islamic financial services industry. The work of the IFSB
complements that of the Basel Committee on Banking Supervision,
International Organisation of Securities Commission, and International
Association of Insurance Supervisors. At the time of writing this book,
there were 110 members of the IFSB, consisting of regulators, IMF, World
Bank, BIS, IDB, ADB, commercial banks, and other institutions. One of
the objectives of the IFSB is to promote the development of a prudent and
transparent Islamic financial services industry through introducing new, or
adapting existing, international standards consistent with Shariah princi-
ples and recommending these for adoption. In December 2005, the IFSB
released the Guiding Principles of Risk Management and Capital
100 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Adequacy Standard for Institutions (other than Insurance Institutions)7
exclusively for Islamic financial services (IIFS). The following paragraph
is a summary of these guidelines, which is based largely on the original
document issued by the IFSB. Guidelines issued by the IFSB are in the
form of fifteen principles focusing on six major risk areas of Islamic
finance: credit risk, equity investment risk, market risk, liquidity risk, rate
of return risk, and operational risk. A brief summary of these standards is
provided in Table 3.5.
Table 3.5 The IFSB guidelines on risk management
Risk Principle Guideline
General Principle 1.0 IIFS shall have in place a comprehensive risk
requirement management and reporting process
Credit risk Principle 2.1 IIFS shall have in place a strategy for
financing, recognising the potential credit
exposures at various stages of the agreement
Principle 2.2 IIFS shall carry out due diligence review
Principle 2.3 IIFS shall have in place an appropriate
methodology for measuring and reporting
the credit risk exposures
Principle 2.4 IIFS shall have in place Shariah-compliant
credit risk mitigating techniques
Equity Principle 3.1 IIFS shall have in place appropriate strategies,
investment risk management, and reporting processes in
risk respect to the risk characteristics of equity
instruments
Principle 3.2 IIFS shall ensure that their valuation
methodology are appropriate and consistent
Principle 3.3 IIFS shall define and establish the exit
strategies in respect of their equity
investment activities
Market risk Principle 4.1 IIFS shall have in place appropriate framework
for market risk management
Liquidity Principle 5.1 IIFS shall have in place a liquidity
risk management framework
Principle 5.2 IIFS shall assume liquidity risk commensurate
with their ability to have sufficient recourse to
Shariah-compliant funds
Continued
BASEL II & IFSB FOR ISLAMIC FINANCIAL RISK 101
Table 3.5 Continued
Risk Principle Guideline
Rate of Principle 6.1 IIFS shall establish a comprehensive risk
return risk management and reporting process to assess
the potential impact of market factors
affecting rate of return on assets
Principle 6.2 IIFS shall have in place an appropriate
framework for managing displaced
commercial risk
Operational risk Principle 7.1 IIFS shall have in place adequate systems
and controls
Principle 7.2 IIFS shall have in place appropriate
mechanisms to safeguard the interests of all
fund providers
Note: IIFS – Institutions (other than Insurance Institutions) offering only Islamic Financial
Services
Source: Guiding principles of risk management, Islamic Financial Services Board,
December 2005, Kuala Lumpur
IFSB principles for credit risk
The IFSB principles are based on specific products and are applicable to
profit-sharing assets (Mushãrakah and Mudãrabah). Due to their exposure
to capital impairment, the IFSB suggests rigorous risk evaluation. The
credit risk associated with securitisation and investment activities should
also be taken into account while ascertaining total credit risk exposures. The
IIFS can have credit risk exposures relating to receivables and lease
(Murãbaha, Diminishing Mushãrakah, and Ijãrah) and working capital
financing (Salam, Istisnã, and Mudãrabah). The IFSB focuses on default,
downgrading, and concentration risks relating to credit risk. Since the
IIFS can assume the role of financiers, suppliers and partners, they face
risk relating to deferred payment and bad delivery of assets. For example,
the risk can be related to delayed delivery of the subject-matter of Salam
or Parallel Istisnã. Due to the changing nature of the relationship at dif-
ferent contractual stages, the commencement stage of credit risk may be
different for different products. Hence, the IFSB advises to assess credit
risk separately for each Islamic financial instrument. It addresses the inte-
grated risks, where other risks give rise to credit risk. The IFSB recognises
the high cost of default due to prohibition of penalty for default in some
jurisdictions.
102 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
The Board of Directors should define and set the overall level of risk
appetite, risk diversification, and asset allocation strategies, geographical
spread, seasons, currency, and tenor. For example, in the case of Salam
contracts, the IIFS may enter into the contract during a season when the
products are most likely to be delivered and sold at maturity. Non-binding
promise and legal enforcement may give rise to operational risk. Hence,
the IIFS should ensure that (a) the expected rate of return on a transaction
is commensurate with the risk incurred; and (b) excessive credit risk and
risk concentration are managed effectively. Referring to the due diligence
process, the IFSB recommends that counterparty reviews must be under-
taken, especially where significant investment risks are present in partici-
pation. The review may include evaluation of business purpose,
operational capabilities, enforcement and economic substance and realistic
forecast of estimated future cash flows. Risk mitigation structure should be
used wherever possible. The IFSB recommends taking up help from appro-
priate experts, including Shariah advisors and other industry experts for
evaluation of new projects. For financing involving several related agree-
ments, the IIFS should ensure that all components of the financial structure
are contractually independent. The IIFS should establish limits on the
degree of reliance and the enforceability of collaterals and guarantees and
should formally agree with the counterparty regarding the utilisation of
collaterals. The delicate issue of remedial action is approached with caution
by the IFSB. There are two types of suggested measures: administrative
and financial. The former includes: negotiations, debt-rescheduling, debt-
restructuring, using debt-collection agency, legal action, and making a
claim under Shariah-compliant insurance. The latter includes: imposing
penalties and establishing the enforceability of collateral or third-party
guarantees. For leased assets under Ijãrah, Shariah-compliant insurance
coverage is advised.
Equity investment risk
This part of the chapter deals with risks inherent in the equity instruments
which are held for investment purposes, particularly instruments based on
Mushãrakah and Mudãrabah. On a cautious note, although both these
instruments contribute substantially to IIFS’s earnings, they include market,
liquidity, credit, and other risks. Since both are profit-sharing financing,
they are exposed to capital impairment risk. IIFS should agree with
Mudarib and/or Mushãrakah partners on the appropriate valuation method
for profits before entering into the contract. The IFSB provides a note of
caution on overstatements or understatements of partnership earning. IIFS
should have exit strategies for investments. Realising the difficulty of having
liquidity for profit distribution, IIFS should agree with partners on methods
of treatment of retained profits.
BASEL II & IFSB FOR ISLAMIC FINANCIAL RISK 103
Market risk
Market risk exposure may occur at certain times throughout the contract
period. Market risk exists in the case of tradeable, marketable, or leaseable
assets and off-balance sheet individual portfolios. This risk is related to
current and future fluctuations of market values of assets. In operating
Ijãrah, the lessor is exposed to market risk on the residual value on prema-
ture termination of the lease. Similarly, in Salam, the IIFS is exposed to
commodity price fluctuations on a long position. In the case of Parallel
Salam, failure of the counterparty may expose the IIFS to commodity price
risk, since it will have to be procured in spot market to honour the Parallel
Salam contract. The IFSB suggests using Shariah-compliant methods to
hedge the foreign exchange exposures. It also suggests establishing a
sound and comprehensive market risk management process and informa-
tion system. For valuation of assets with no direct market prices, the IIFS
should have its own valuation of market risk positions. The IIFS should
clearly define their risk appetite which is adequately supported by the
related capital.
Liquidity risk
For the IIFS, there are two major types of fund providers: current account
holders (CAH) and unrestricted investment account holders (UIAH). The
UIAH are more prone to market information regarding rate of return, finan-
cial conditions, and Shariah rulings relating to IIFS. Any adverse event can
trigger an exodus of these account holders, causing a serious liquidity crisis.
Hence, a detailed analysis of UIAH with details of their expectations and
incentives should form part of net funding requirements. IIFS should have
liquidity management policies covering:8
strategy for managing liquidity
framework for developing and implementing sound processes for meas-
uring and monitoring liquidity
monitoring and reporting liquidity exposures on a periodic basis
adequate funding capacity, including willingness and ability of share-
holders to provide additional capital if needed
access to liquidity through fixed asset realisation and options such as
sale and lease-back
liquidity crisis management
104 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
The quantitative factors addressed in the risk management policy will
include: diversity and sources of funds, concentration of funding base,
reliance on marketable assets, standby lines of external funding. The
qualitative factors include: particular skills of treasury management,
quality of MIS, IIFS’s reputation in market. For measuring and monitor-
ing liquidity, the IFSB has suggested constructing maturity ladders with
appropriate time bands. Three types of cash flows are identified: known
cash flows (as in the case of Murãbaha, Ijãrah, and Diminishing
Mushãrakah), conditional but predictable cash flows (like Salam and
Istisnã), and conditional and unpredictable cash flows (like certain
categories of Mushãrakah – where investment is for an open-ended
period). IIFS should assess maximum amounts of cumulative liquidity
mismatches.
Rate of return risk
Rate of return risk (RRR) is generally associated with overall balance sheet
exposures where mismatches arises between assets and balances from fund
providers.9 Relating to the importance of the RRR, it is argued that it is a
strategic risk since one of the primary responsibilities of the IIFS is to
manage investment account holderd expectations, and hence should form a
part of the balance sheet risk management. Since RRR emanates from the
balance sheet positions, the IIFS should have competent staff to undertake
analysis of risk exposures arising from consolidated balance sheet activi-
ties. IIFS is exposed to increasing long-term fixed rates in the market
which can affect balance sheet positions. While calculating the rate of
return, IIFS can use a gapping method for allocating positions into time
bands. IIFS should use appropriate cash flow forecasting techniques based
on behavioural maturity and underlying assumptions and parameters. The
IFSB suggests using balance sheet techniques to minimise exposures
using: determining and varying future profit ratios, developing new
Shariah-compliant instruments, and issuing securitisation tranches of
Shariah-permissible assets. Competitive pressure may bring in displaced
commercial risks, where the IIFS may waive their rights to the part of
profit (Mudãrabah) in favour of investment account holders. IIFS should
have clear and well-defined policies and procedures which should have the
approval of the IIFS’s board of directors. A Profit Equalisation Reserve
(PER) can be formed out of gross income of IIFS which can be used to
maintain a desired level of return over a period of time. This should have
formal approval by the IIFS’s board of directors. Similarly, an Investment
Risk Reserve (IRR) can be created out of the income of investment account
holders.
BASEL II & IFSB FOR ISLAMIC FINANCIAL RISK 105
Operational risk
Other than the risk arising from failed or inadequate internal processes, peo-
ple, and systems, the IIFS should include the risk of Shariah non-compliance
and failure of fiduciary responsibilities in operational risk management. The
risk of diminishing reputation and limiting business opportunities is also
included under the operational risk. Shariah compliance is extremely
important to the IIFS and hence a failure to comply with such principles
may result in a transaction being cancelled and income being considered as
illegitimate. Shariah compliance should be considered at the time of accept-
ing deposits and investment funds, while providing finance and conducting
investment activities for their customers. A Shariah compliance review
should form part of existing internal or external audits. A historical database
can be created consisting of cases of non-Shariah-compliant incomes,
which can be used as a guideline for the future. The IFSB advises on limit-
ing the risk transmission between current and investment accounts. There
should be timely and adequate disclosure of material information to invest-
ment account holders and to the market. To manage a fall in the rate of
return, the IIFS can set up reserves with detailed guidelines regarding trans-
fers in and out of the reserves, thresholds for specific reserves, and closure
of the reserves.
Other than these specific principles, the IFSB emphasises the role of
the supervisory authority. Some of the recommendations regarding the
role of supervisory authorities include: maintaining a detailed descrip-
tion of each instrument, developing information-sharing procedures,
establishing criteria and procedures for dealing with non-wilful default-
ers, and developing adequate risk management framework, policies, and
procedures.
SUMMARY
Islamic financial contracts are typified by the changing relationship
between the contracting parties during the lifetime of the contract. This has
a direct bearing on the risk exposures. Soundness and safety for banks
depend to a great extent on the capital held by them. Since capital is expen-
sive to hold, there is a constant attempt to find the optimal mix of capital for
business and regulatory requirement. The original Basel accord was the
first-ever systematic attempt at a global level to provide a framework for
capital adequacy. A minimum charge of 8 percent of risk-weighted capital
was proposed in the original accord, which was voluntarily accepted by
many countries. Due to the rapid changes on the financial canvas, the origi-
nal accord proved to be insufficient to cover increasing intricacies. The new
106 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Basel II accord revolutionised the concept of risk management. With its
detailed analysis of credit, market, and operational risk, the Basel II accord
brought operational risk into the realms of risk management. The three
mutually enforcing Pillars are the foundation of Basel II and were discussed
in this chapter.
The first Pillar concentrates on the calculation of the minimum capital
charge using several optional methodologies. It provides three approaches
for credit risk: SA, IRB, and Securitisation frameworks. Similarly, for mar-
ket risk, SMA and IMA are proposed. There are three approaches for oper-
ational risk provided under Pillar 1: BIA, SA, and AMA. All the different
approaches were discussed in this chapter. These approaches are not largely
applicable to Islamic financial industry due to several reasons: first, due to
the absence of rating for Islamic finance; second, due to the changing nature
of the relationships during the lifetime of the contract; third, due to difficul-
ties in estimating PDs, LGDs, and EADs for Islamic finance. On the front
of market risk, Basel II has provided two possible alternatives. Market risk
concentrations are very high in Islamic finance. Almost all the products are
exposed to market risk in Islamic finance, including: Mushãrakah,
Mudãrabah, Murãbaha, Salam, Istisnã, and Ijãrah, through commodity
prices.
The calculation of operational risk exposures under Basel II uses three
alternatives. Under BIA, the standard capital charge of 15 percent is used
for the average gross profit of the last three years. SA goes a step further and
allows for bifurcation of business lines and thus the risk weights also. The
allocation of the different weights for Islamic financial activities is under
review and is still not clear.
As discussed, Supervisory Review, the second Pillar of Basel II, has
limited applicability for Islamic finance. Some of the common recom-
mendations can be applicable to Islamic financial institutions. However,
the liquidity risk, which is treated as a residual risk under Basel II, is one
of the core risks in Islamic banking. Non-existence of inter-bank money
market, daily overdrafts from the Central Banks, and secondary market
makes the liquidity risk exposures very huge for the Islamic financial
institutions. Credit concentration risk can also sometimes be very large in
Islamic banking and Basel II recommendations can be applied in order to
manage it.
The social responsibility of Islamic financial activities is of utmost
importance and hence there is a great emphasis on disclosure. The third
Pillar of Basel II on market disclosure is largely applicable to Islamic finan-
cial institutions. Disclosures related to business activities, profits and losses,
use of funds, investments, etc., are all necessary in order to maintain the
confidence of customers in Islamic financial institutions. Risk management
practices in Islamic financial institutions are evolving and thus it is more
BASEL II & IFSB FOR ISLAMIC FINANCIAL RISK 107
important to share this information in order to develop best practices for the
industry.
The IFSB is playing a key role in the development of standards for risk
management in the Islamic financial industry. Their recent release of
standards for risk management is the first attempt at consolidating the
efforts for bringing Islamic financial risk management under one
umbrella. As shown in this chapter, the IFSB standards attempt to provide
a guideline for risk management. The IFSB has recognised six major
types of risks in the Islamic financial industry: credit risk, equity invest-
ment risk, market risk, liquidity risk, rate of return risk, and operational
risk. The IFSB has released 15 principles covering these risks. It recog-
nises the credit risk exposures at different stages of the contract. It focuses
on both types of credit risk, exposures relating to receivables and lease as
well as working capital financing. The exposures for equity investment
risk are mainly through Mushãrakah and Mudãrabah. The IFSB suggests
identifying methods for the treatment of retained profits to avoid difficul-
ties of liquidity arising out of profit distributions. According to the IFSB,
market risk exists for tradeable, marketable, and off-balance sheet indi-
vidual portfolio. Banks may use Shariah-compliant hedging tools for
managing foreign exchange exposures. The volatility related to UIAH
poses the liquidity risk for Islamic financial institutions. Some of the
suggested ways to manage liquidity includes diversification of sources of
funds, concentration of funding base, and a standby line of external
funding. Rate of return risk is generally on balance sheet exposure.
Islamic financial institutions should use appropriate cash flow forecasting
techniques. The IFSB suggest the creation of a profit equalisation reserve
and investment risk reserve.
Basel II is primarily for conventional banks and thus does not offer a
great help to Islamic financial institutions. However, in line with any other
industry guidelines, some of the principles of risk management as proposed
in Basel II are applicable to the Islamic financial industry. The fundamental
role of capital cannot be changed and thus the role of risk management in
any institution.
NOTES
Disclaimer: Note that neither the Basel Committee nor the BIS have endorsed the summary and are not
liable for any misinterpretation that may or may not occur during or due to the summarisation as appear-
ing in the text. Please note that the original text of the Basel II guidelines is available free on the BIS
website at www.bis.org
1. Demirgüç-Kunt, Asli and Detragiache, Enrica, 1997, ‘The Determinants of Banking Crises –
Evidence from Developing and Developed Countries’, IMF Working Paper Series No. 97/106.
2. BIS, ‘International Convergence of Capital Measurement and Capital Standards’, Basel Committee
on Banking Supervision, Basel, July 1988.
108 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
3. BIS, ‘Amendment to the Capital Accord to Incorporate Market Risks’, Basel Committee on Banking
Supervision, Basel, January 1996.
4. BIS, ‘International Convergence of Capital Measurement and Capital Standards’, Basel Committee
on Banking Supervision, Basel, June 2006.
5. BIS, ‘Consultative Document, Overview of The New Basel Capital Accord’, BCBS, Basel, 31st July
2003, p. 5.
6. BIS, ‘Sound Practices for the Management and Supervision of Operational Risk’, February 2003.
7. IFSB, ‘Capital Adequacy Standard for Institutions (Other Than Insurance Institutions) Offering only
Islamic Financial Services’, Islamic Financial Services Board, December 2005.
8. IFSB, Guiding Principles of Risk Management, December 2005, p. 20.
9. IFSB, Guiding Principles of Risk Management, December 2005, p. 23.
CHAPTER 4
Credit Risks in Islamic
Finance
INTRODUCTION
Credit risk appears when a financial institution is expecting a payment that has
been contractually agreed between the institution and the counterparty and the
obligors are unable – or in other words defaults – to fulfill their obligations.
Credit risk also initiated also when there is a change or underestimation in the
rating of the counterparty. Financial institutions that provide Islamic financial
products are also exposed to credit risk because of the emphasis on lending in
the Murãbaha, leasing in the Ijãrah, promising to deliver or to buy in Istisnã
and Salam, and investing on business performance in the Mushãrakah and
Mudãrabah contracts. Financial problems related to either the individual
counterparties (i.e. health problems) or to more general economic situations
(i.e. market recession) may be some of the reasons for the obligors to default.
Islamic contracts are very much driven by agreements that have unique
and individual characteristics. Such characteristics are more qualitative in
nature and are not always well defined and recorded in comparison to the
more conventional contracts. For instance, during the lifetime of Islamic
contracts there could be cases of unclear definition of whether and how the
credit risk exposure is covered from guaranties and collaterals. Another
important aspect that should be noted is that, due to relatively recent imple-
mentation of Islamic financial contracts, there is limited information about
the qualitative as well as quantitative criteria used to set the contracts. In
addition, the above institutions are facing challenges of introducing new or
modified structures of the existing contracts and portfolios. Note, finally,
that the increasing number of bankruptcies globally and the competitive
spreads in loans and the risk-adjusted credit policies are some of the reasons
that renders credit risk assessment more important than ever before.
Factors such as globalisation of financial markets, economic conditions of
the country, and individual cultural characteristics in different regions and
109
110 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
markets increase the complexity of credit risk management. In contrast to the
conventional contracts, the majority of credit risk assessments for the Islamic
financial products are based on subjective analysis. Banks that are operating
in Islamic regions or providing Islamic contracts are in many cases motivated
by information stemming from different obligors, and the result may be sub-
jective and based on experts to approve (or not) the contracts, e.g. a Murãbaha
contract. Traditionally, financial institutions were placing great emphasis on
the relationship with their customers; however, nowadays institutions must
also be able to predict and then consider the behaviour of their customers in
terms of their payment obligations. Having a great number of borrowers with
different financial statuses and from various market sectors, financial institu-
tions have to apply techniques and methodologies for implementing sophisti-
cated models in order to upgrade their credit risk management systems and
approve or reject in an objective way the applications for Islamic contracts.
Therefore, a wide variety of credit models have been built in order to measure
credit risk and provide reliable estimates for credit risk parameters such as the
probability of default (PD) and loss given default (LGD).
This chapter consists of three sections that deal with qualitative- and
quantitative-based credit risk assessment models. Furthermore, it gives a
description of credit risk parameters and credit rating systems and their
validation.
C R E D I T R I S K E X P O S U R E I D E N T I F I C AT I O N
In credit risk exposure analysis, a key factor is the identification of the rela-
tions between the counterparties, the Islamic financial contracts and the
Risk Risk
Guaranties,
coverage coverage
collaterals,
etc
Islamic
financial Counterparties
contracts
Agreement
Figure 4.1 Relations between the Islamic financial counterparties and contracts
and the guaranties and collaterals
CREDIT RISKS IN ISLAMIC FINANCE 111
guaranties and collaterals used to cover a percentage of the potential losses
in the case of defaults. More analytically, each counterparty is linked with
one or more contracts; moreover, it may be linked with other counterparties
defined as guarantees and collaterals. In reality, within the institution there
are many counterparties that may be linked to several contracts. In addition,
many Islamic financial contracts are linked to a great number of counter-
parties and many different types of guaranties and collaterals are used to
cover in parallel the exposure of several contracts and counterparties.
Bearing in mind that these three parties are linked together with different
degrees, such as in regards to the contracts’ lifecycle or in regards to the
coverage from guaranties and collaterals, the problem of identifying and
measuring the credit risk exposure is a complicated process.
Counterparties
In the structure of the Islamic financial products, counterparties are all the
parties that are involved in the Islamic contract agreements and partner-
ships. Thus, in the Murãbaha and the Salam contracts, alongside the bank,
both the seller and buyer are considered as the counterparties. Accordingly,
in the Ijãrah contracts the renters/lessees and in the Istisnã contracts the
buyer, user contractor, or manufacturer are the counterparties accordingly.
Finally, in the Mushãrakah and the Mudãrabah partnership agreements, the
business partners and agents are the corresponding counterparties.
As mentioned previously, what really matters in the credit risk exposure
analysis is to have a system in place that links all counterparties with the
involved contracts. Also, there is a need to show the links with guaranties
and collaterals that will be used to cover the risks in case of defaults.
For instance, a counterparty may agree with an institution on several
Murãbaha and Istisnã contracts, and at the same time may also have a
Mushãrakah partnership business agreement, as illustrated in Figure 4.2.
On the other hand, as can be seen from the same Figure 4.2 several coun-
terparties can also be linked to the same contract; for instance, counterpar-
ties A and D participate by 5 and 95 percent accordingly on a Mushãrakah
contract; whereas in the Istisnã contract there are three different counter-
parties: A, C, and D. Furthermore, different collateral and guaranties may
cover several counterparties and contracts directly or via other counterpar-
ties. For instance, the Istisnã contract is directly collateralised by 30 percent
from the collateral C; also indirectly by the counterparties A and C that are
collateralised from collateral/guaranty A (25 percent) and B (5 percent)
respectively.
Another important issue is that collaterals or guaranties may be linked
with other collaterals or guaranties that have different ratings; such groups
must be well defined and linked according to the degree of dependence and
rating. Any group of the same contracts or a link between them must also be
112 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Islamic Counterparties Risk coverage
Market
financial
contracts
20%
Counterparty B
Risk coverage
Guaranty A 100%
collateral A Murãbaha Guaranty B
25% 100% 5%
Counterparty C Collateral B
72%
18%
8%
Counterparty A Istisnã Collateral C
30%
42%
20% Collateral D
5%
Mushãrakah Counterparty D
95% 15%
Strategies
Figure 4.2 Representation of the links between a counterparty with several
contracts and risk coverage by several guaranties and collaterals
identified. Note that market conditions and the institution’s strategies
influence the behaviour of the counterparties as well as the performance of
the Islamic products. In a case where the counterparty is unable to fulfil its
payment or delivery obligations related to one or more contracts, the insti-
tution is exposed via several other contracts that are linked to the same
counterparty. Bear in mind that in many such cases, a default in one contract
may cause a consequence of defaults to the others. Of course, there are
cases where the loss in case of default is relatively small due to collaterals
or guaranties that are applied. Collaterals or guaranties, however, must be
rated at a level that would enable then to cover the potential losses initiated
from the defaults of counterparties. Summarising the above, in counterparty
analysis the following points should be considered:
The market conditions and the institution’s strategies that may
influence the counterparty’s behaviour
The type and the volume of the contracts where the counterparty is
linked, defining also the degree of participation
CREDIT RISKS IN ISLAMIC FINANCE 113
The links between the contracts that refer to the same counterparty
The rates of the guaranties and collaterals and their inter-links, as well
as the links to the contracts and counterparties. Note that the counter-
parties must be covered by guaranties and collaterals that are rated with
a higher grade.
In Islamic finance where there may be special identification in regards to
the ratings for the guaranties and collaterals, financial institutions may have
only certain indications of the actual risk coverage to their contracts’
exposure. However, such indications must be defined quantitatively and be
part of the risk management system. The assessment of the counterparties is
also related to qualitative criteria such as the external or internal rating, the
industry, or the geographical location.
CREDIT RISK ASSESSMENT MODELS
The definition and implementation of the methodologies used for modelling
credit risks is an important issue to be considered by the risk analysts. In this
section, different methods and techniques for developing models for credit
risks that are arising from Islamic products are described. Moreover, the
assumptions in this modelling are also highlighted. The corresponding
advantages and disadvantages of the models are also discussed to evaluate
their use at the implementation phase.
Qualitative methods – expert systems
Financial institutions are assessing the creditability of the borrower that
agrees on Islamic financial contracts based on qualitative methods. Such
methods define systems that are based on the judgement of experts who are
involved in the credit-approval process. The resulting expert systems
combine the analysis of the credit-worthiness of the obligor with the
practical experience and observations of the experts who apply the analysis.
Therefore, the quality of empirical models of expert systems is very much
dependent on how accurately they depict the subjective experience of
financial experts. Most commonly, the factors that are considered in the
assessment process of the credit-worthiness analysis of the obligor are
determined empirically; moreover, their influence and weight in the overall
assessments are based on subjective experiences.
The design of the expert systems is driven by rules that combine the
criteria used for the expert’s judgements. The aim of such systems is to
design and then combine expert rules by taking into consideration experiences
related to the behaviour of the client. Moreover, the rules must consider any
114 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
information related to the type of contracts, the market conditions, as well
as any other factors that influence the client’s behaviour.
More analytically, the general practice is that the experts, based on their
experiences, are predefining several credit-worthiness characteristics that
are playing an essential role in the future behaviour of the obligor. Based on
a predefined rating scale, the degree of credit-worthiness is rated accord-
ingly. Each rating is driven by the individual qualitative characteristics of
the obligor that refer to his/her expected behaviour. For instance, qualitative
criteria that are related to family, educational, and professional background
may be used to evaluate and mark the degree of rating correspondingly,
according to the experts’ subjective opinion concerning the specific obligor.
The above-mentioned ratings may also be determined by other factors such
as the actual type of contracts that the client has agreed on, the conditions
referring to the future market conditions (for instance, economic recession),
and to market risk factors. Note that credit officers assign ratings that are
based on financial ratios, opinions on quality management, and other data.
The aforementioned ratings are then linked to a risk value that is defined
qualitatively, i.e. ‘low’, ‘below average’, ‘medium’, ‘above average’, and
‘high’. These risk values may correspond to a scale of quantitatively crisp
grades. Finally, the individual factors’ grades are combined to produce an
overall assessment grade. Summarising the above, the credit risk assessment
based on qualitative criteria involves the following steps:
1. The experts are rating the obligor based on predefined qualitative credit-
worthiness characteristics together with some additional factors that
may influence the client’s behaviour
2. The links for the ratings are defined in a qualitative manner that is
determined by the experts
3. The quantitative base risk grade drives the level of risks
4. The individual grades are aggregated to generate an overall assessment.
Note that the aggregation process may use expert system’s aggregation
rules, which cannot be influenced by the credit analysts.
Financial institutions should rate qualitatively the credit-worthiness of
the client based on as much information as possible and well-defined
subjective criteria that may be driven by objective factors.
Figure 4.3 illustrates the expert process as discussed above.
As mentioned above, based on qualitative criteria, the expert has to
identify the expected behaviour of the client under some circumstances that
may influence their ability to keep their promise in regards to the Islamic
CREDIT RISKS IN ISLAMIC FINANCE 115
Ratings Risk values Grades
Very good Low 1
characteristics
Assessment
Qualitative
Good Below average 2
Satisfactory Average 3
Sufficient Above average 4
Insufficient High 5
Figure 4.3 Model layout of the expert system for credit risk assessment
financial contract as agreed. Thus, by applying the Murãbaha contract, the
‘experts’ of financial institutions are facing the challenge of how to identify
the criteria that will evaluate whether the client – or, in other words, the
‘buyer’ – of the goods will comply with the agreed payment obligations that
are set as instalments on fixed time buckets.
In Ijãrah leasing financial contracts, the financial institution that plays
the role of the lessees should define rules and criteria that are related to
future behaviour of the counterparty (lessor) that may expose the institution
to credit risk, such as covering cases of early leave based, for instance, on
the type of the asset or service that is being rented (i.e. car, house, business
property, or service, etc.) and its dependency on the external (market, busi-
ness, and operational) factors.
In the Mushãrakah and Mudãrabah partnership type of contracts, the qual-
itative criteria that financial institutions may define and apply to assess the
credit risk exposure are more subjective and rather complex. The default on
expected cash flows is mostly related to the actual resulting business profit,
where the financial institution may be directly or indirectly responsible.
It is important, finally, to highlight that, by providing Islamic contracts,
financial institutions need to justify whether the client’s guaranties, beyond
the actual goods, commodity, or physical assets contained within the con-
tract, will be able to cover any case of a default from the buyer’s side.
Quantitative methods
The use of quantitative methods for assessing credit risks that financial
institutions are exposed to by providing Islamic financial products are based
mainly on quantitative statistical-based models. There are mainly two types
of information data that financial institutions can consider for modelling the
credit risks that they are exposed to:
the data that refer to the past and current behaviour of the counterparty
(i.e. lender, partner, etc.) and
the data that defines the losses of the associated risks.
116 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
By building quantitative models, risk analysts may have to make several
assumptions and therefore there is a level of uncertainty that can influence
the results of the model. For instance, there may be an uncertainty for the
factors that might be difficult to predict, or an uncertainty for the correct-
ness of the estimation of parameters, and thus uncertainty of how close the
model is to reality. Wrong assumptions, or wrong explanations of the
assumptions, may result in an incorrect development of a model that may
also lead to wrong decisions.
In the construction of the quantitative methods, risk analysts should
follow processes with certain steps that need to be concrete and adequately
validated so as to avoid a ‘black-box’ nature of the model. The main steps
are presented as follows:
Identifying the availability and accessibility of historical data, data
clearance, unification, and selection to be used for credit financial risk
analysis.
Due to several systems installed within different departments and bank-
ing business sectors, the data availability and accessibility may become a
complex process. Nowadays, financial institutions tend to use data-
warehouses where, in a single location, all financial data can be stored
and shared by various analytical departments. The IT system engineers
must be able to perform some cleansing of the data – processes that
enhance data quality such as converting all dates to a common date for-
mat and using one unique code per counterparty. Additionally, there
must be a clear definition and identification process in regards to the
meaning of the financial data that are stored in data-warehouses in a
logical and unified way. Consider, for instance, the case where,
although each transaction system works with the concept of notional
value, the name of this field varies across transaction systems. One sys-
tem might call it ‘notional value’, another ‘nominal value’, and yet
another ‘current principle’ or ‘balance’. Such information is not unified
even though they are stored into a common warehouse.
Moreover, the accuracy, efficiency, and appropriateness of data are
very important issues, so that the model is built not only upon a suffi-
cient number of data but also upon real and representative data of the
population. The selection and aggregation of the available information
data should be set in a system that extracts, transforms, and loads them
to a risk management system tool.
In credit risk analysis, it is very crucial that the model is built upon a
sufficient amount of information referring to the defaults of the oblig-
ors. The percentage of such failures to obligations should be at least
25 percent of the sample. This is because in reality the financial institution
usually has low-defaulted portfolios and a big number of non-defaulted
CREDIT RISKS IN ISLAMIC FINANCE 117
clients. The remaining 75 percent of the sample is the percentage of the
clients that succeed in their payment obligations.
Simulating data used for the credit financial risk analysis.
Based on mathematical approaches, on scenarios, or even on their com-
bination, information data can be generated to be used for credit finan-
cial risk analysis. Using mathematical techniques, one approach is to
‘shock’ the existing historical data and generate slightly different
behaviour in regards to the creditability of the obligator. On the other
hand, using some scenarios of future behaviours, data can be generated
accordingly. Note, however, that by applying simulation techniques the
resulted data are dependent on the selected mathematical methodolo-
gies as well as the assumptions made for the scenarios.
Determination of model methodology.
The model that should be used for assessing credit risks could be based
upon empirical methodologies as mentioned above or based on quanti-
tative statistical methodologies. Alternatively, a combination of the two
methodologies may form hybrid models, as is mentioned in the follow-
ing paragraph. It is important to know that by applying models that are
based fully or partly on quantitative information, the functional rela-
tionship of the model should be well defined.
Assessment of the parameters of the model.
The parameters that are used in the construction of the model should be
scientifically established to ensure a good predictive power of the
model mapping, performance, and stability.
Qualitative and quantitative validation.
It is important that the performance of the model and the stability of the
parameters are validated on an ongoing basis.
Conclusions.
By applying quantitative models, the resulting conclusions need to be
very carefully extracted, based upon the appropriateness of the model,
the application of the parameter assumptions, and the proper choice of
the confidence levels. The model should also be assessed by human
judgement. This is for the risk analysts to ensure that it performs well
and has good predictive power (forward-looking model). Note that the
human judgement should take into consideration all other relevant
information that the model is unable to consider.
Financial institutions that are providing Islamic financial products are
facing the challenge of developing and applying quantitative-based models.
118 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Banking information systems that are performing in Islamic markets are
usually dealing with many problems in regards to the data clearance and
unification. Moreover, cultural issues play an important role in keeping
information data referring to borrowers’ behaviour and/or to the resulting
losses due to credit risks. Added to this, the legal issues in collecting infor-
mation related to borrowers are critical in several markets offering Islamic
financial products.
Hybrid models
The combination of empirical-based models with the quantitative-based mod-
els result in the hybrid form of credit-assessment models. In practice, by using
only empirical or only quantitative models, financial institutions may have
poor and inadequate forward-looking results. Quantitative-based models, on
one hand, have the great advantage of evaluating in an objective way the meth-
ods and results related to the prediction of the obligors credit-worthiness. At
the same time, quantitative models are unable to take into consideration all the
credit-worthiness factors, and so very important information on the borrowers’
characteristics would be lost in individual cases without the empirical help of
credit experts. The objectivity of quantitative systems in the elaboration of
quantitative data combined with the superiority of empirical systems in the
incorporation and analysis of qualitative criteria leads to a better classification
of risks. The above-mentioned combined model could employ advance fuzzy
logic systems that use both the qualitative empirical systems expressed as lin-
guistic rules and the results from the quantitative analysis. Nonlinear functions
are used to express mathematically the qualitative linguistic terms of the rules
for applying such model integration.
C R E D I T R I S K VA L U AT I O N
The main parameters that should be considered in the valuation of credit risks
are the expected and unexpected losses. The calculation of the expected and
unexpected losses requires the calculation of the probability of default (PD),
the loss given default (LGD), and the exposure at default (EAD). Their defi-
nitions and estimation is one of the main challenges that financial institutions
are facing nowadays. Moreover, defining the occurrence of default should
also be clearly set by the institution. The modelling of credit risk is entirely
based on the estimation of the probability of default and the loss given default.
Defaults
In general, from the institution’s point of view, default occurs when there is
a loss that is initiated from the counterparty’s inability to comply with its
CREDIT RISKS IN ISLAMIC FINANCE 119
obligations. However, as there is no standard definition of what constitutes
a default there are different definitions that may be used for different
purposes. Specifically, for the typical Islamic financial contracts default
occurs either due to payment inability such as in the Murãbaha and Ijãrah
contracts, or due to delays and/or failures of deliveries referring to a
commodity or constructed asset at the agreed date, i.e. in the Salam and the
Istisnã Islamic financial contracts. In the case of the Mushãrakah partner-
ship financial contract, the responsibilities for the defaults that refer to the
inabilities of payments or delays on deliveries are shared, to a certain
degree, between the institution and the business partner(s). Finally, in the
Mudãrabah partnership agreement, the occurrence of default is relatively
unclear as it refers to the delays or inabilities to repay in regards to the
expectations from the business venture counterparty.
What is defined as inability of repayments or delays in repayments may
be subjective to the institution that is providing such contracts. Time is
usually set to a fixed amount of days and may differ from one region/
market to the other and correspondingly from the different regulatory
bodies (for Basel II, the days to default for credit risk is set to 90). It is
important to highlight, however, that the definition and the measurement
criteria referring to the different types of defaults should be clearly
explained and defended to the regulators in regards to their use in the
evaluation of the institution’s exposure.
Probability of default (PD)
The Probability of Default (PD) for the Islamic financial products is the
likelihood that the counterparty will be unable to comply with his/her pay-
ment and/or delivery obligations.
Specifically:
For the Murãbaha contracts, it is the likelihood that the buyer of the
goods (counterparty) will be unable to re-pay the instalments that he/she is
obligated to.
For the Ijãrah contracts, it is the probability that the lessee (coun-
terparty) will be unable to re-pay at instalment points or at the end of the
contracts (according to what have been agreed). Moreover, it could be the
probability of any early leave, from the lessee side, before the contract’s
maturity date.
For the Salam contracts, it is the probability from the seller’s (coun-
terparty) side to default on delivering the commodity at the delivery date. On
the other hand, it is also the default probability from the buyer (counterparty)
to buy the commodity at the agreed price.
Similarly to the Salam contracts, for the Istisnã contracts, from the
manufacturers and/or constructors (counterparties) side, it is the likelihood
120 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
of default on delivering the commodity or constructed asset at the delivery
date. Moreover, for the financial institution, it is the probability of not
receiving the agreed selling price from the buyer or user (counterparty).
In the Permanent Mushãrakah contracts of partnership where the
business partner(s) has the role of the counterparty, it is the likelihood for
the business to default in providing the expected cash. However, in the
Diminishing Mushãrakah contracts, it is the probability that the partner(s)
(counterparty) default on buying the equities at the agreed pre-fixed price
using the instalment basis.
In the Mudãrabah contracts of partnership agreement and during
the investment period, it is the probability that the business venture (coun-
terparty) defaults in carrying on the business development and/or the pro-
ject’s implementation. Moreover, during the profit and loss period it is the
likelihood that the business venture defaults in providing the expected
profit.
The estimation of the PD for the Islamic financial products is one of the
most challenging issues for financial institutions. The main steps that are
commonly used for this purpose are:
Analysing the credit risk aspects of the counterparty;
Mapping the counterparty to an internal risk grade which has an associ-
ated PD; and
Calculate the PD.
In general for the calculation of the PD, the credit history of the counter-
party and the nature of the investment are taken into account. There are
several approaches that financial institutions could implement for estimat-
ing the counterparty’s probability of default. Most of them are based on
qualitative and/or quantitative models as discussed in the previous section.
Past information data referring to the counterparties can be used to evaluate
the corresponding behaviour. Such information for the PD calculation may
come either by extracting them from the existing historical databases, or by
observing the current data, or by simulating future market conditions and
counterparty behaviours that are driven by predefined scenarios.
Probability of Default is the most straightforward parameter to estimate
using data available from the credit institutions. However, by using readily
available data, estimating PDs is very difficult, because it is assumed that
the credit institution will keep reliable records about the behaviours of the
credit assets and to efficiently estimate some other parameters necessary for
its estimation. Thus, many banks are using external ratings agencies such as
CREDIT RISKS IN ISLAMIC FINANCE 121
Egan Jones, Fitch, Moody’s Investors Service, or Standard and Poors for
estimating the PDs.
Estimating the probability of default
The estimation of the probability of default is rather a difficult task, but not
as much as the LGD. The main challenge for estimating the PD is to define
the criteria set to form the quantitative as well as the qualitative definition of
default. In most cases the two types of criteria are complementary to each
other in the sense that quantitative criteria should be assessed after using
qualitative ones. For instance, the criteria referring to the ‘unlikeliness-to-
pay’ clause in the Basel II document (see Basel Committee on Banking
Supervision, 20051) should be first assessed qualitatively and then quantita-
tively. Note, finally, that the definition of default should be based on the
nature of the credit products under study.
Quantitative criteria
Statistically based, the probability of default (PD) for a specified class of
Islamic financial contracts, e.g. the one based on Murãbaha financial
contracts, is defined as the number of such contracts that were defaulted
during a certain period of time. Thus, the annual default rate or probabil-
ity of default is the percentage rate of the contracts that were defaulted in
relation to the total portfolio of contracts during the period of one year,
while the biannual default rate is the same measure for a period of two
years. Mathematically expressed, the probability of default for
‘Contracts’ within the period t (i.e. one year) is represented as in the
following equation (4.1):
DCt
PDt 100 (4.1)
TC
where PDt is the annual percentage probability of default, DCt is the num-
ber of defaulted contracts during the period t under examination, and TC is
the total number of contracts of the examined class.
For the different types of financial Islamic contracts and their underlying
assets, the definition of their defaults on, for example, payment, delivery,
expected profit, is based on criteria that are derived either from the financial
institution or from the regulators. Therefore, for the normal Murãbaha
financial contracts, default is the inability or insufficiency of the debtors to
pay a substantial portion of their lend capital for a predefined set period
(such as three months or 90 days).
122 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Quantitative criteria should be set therefore in regards to the parameters
such as ‘portion’, which may refer to the loan or renter payments in the
Murãbaha and Ijãrah contracts respectively, to the sales of the commodities
or assets in the Salam and Istisnã contracts, and to the profits in the
Mushãrakah and Mudãrabah contracts of partnerships. Accordingly, the
parameter named ‘period’ could be referred to the delays in payments or
deliveries as defined in the construction of the Islamic contracts described
in chapter 2. Moreover, the qualitative criteria could also be combined to
define the default driven by their inter-relation. For instance, in regards to
the two aforementioned criteria, when the amount of payment instalments
in the Murãbaha contract or deliveries in the Salam and Istisnã contracts is
high, then the buffer period for the obligor to default is increasing accordingly.
The ‘materiality’ is another point that must be clarified in order to
provide a concrete definition of the probability of default. The ‘materiality’
notion coincides with the meaning of ‘substantiality’. The most common
implied definition that banks consider as ‘material’ is the fact that the debtor
is unable or unwilling to pay the instalments (including or not the profit), to
buy or to deliver the asset as agreed for more than μ consecutive months
(usually, μ is set as three or six months). This is the level after which the
‘road’ to default is irreversible for a certain confidence level.
All approximations assume that the single measure of PD is estimated by
averaging a number of annual estimations of PDs realised during different
past years. The averaging should be extended to a sufficient number of years
in order to cover a full economic cycle. Nonetheless, the estimation of PD
should contain a forward-looking element in order to protect the financial
institution from anticipated future loses arising from credit risk.
Based on the scheduled payments, repayments, or deliveries, the estima-
tions of PD for μ number of past due months can be defined by replacing the
n
DCt in (4.1) with 兺 di. Thus, the probability of default is found as in the fol-
i1
lowing equation (4.2):
冤冢 冣 冏
n
兺d
冢TPDAI M 冣
冥
i
PDt i1
100 di 1, i
(4.2)
TC CAI
i AIF i
where n is the number of loans or underlying assets within the credit
portfolio, di is a parameter that equals to zero (0) if the loan or underlying
asset is ‘active’ or equals to one (1) if it is ‘in default’, TPDAI is the total
past due amount of instalments or delivery for the agreement i, CAIi is the
contractual amount of instalments or payments that had to be paid since the
CREDIT RISKS IN ISLAMIC FINANCE 123
first past due amount had appeared, M is the number of months for the
period t (that is, 12 for an annual period), AIFi is the frequency of instal-
ments or deliveries (if more than one) within a period t that is usually set as
a year.
Case Study A simple example concerning fixed monthly repayment
instalments of a Murãbaha contract is illustrated by the following Table 4.1
of default events.
The Probability of default for every year is estimated by applying equa-
tion (4.1). Thus, the PD for the first composition of the Murãbaha contract
portfolio is:
DC2006
PD2006 100 60 100 0.69%
TC 8600
The PD for the second period and the composition of the portfolio is esti-
mated by, first, subtracting the already defaulted loans from the total num-
ber of loans. The remaining number of outstanding, ‘healthy’, loans is 9,000
minus the 40 defaulted ones, resulting in 8,960 loans; assuming, however,
that none of the defaulted loans during the last period have returned into
‘healthy’ ones. Thus, the contracts that have remained defaulted should be
subtracted from the total portfolio. As a result, the PD is given by:
DC2007 80
PD2007 100 100 0.86%
TC (930060)
Table 4.1 Different cases for estimating the probability of default
based on quantitative criteria
Period for Sum of the Number of
measuring examined defaulted Probability
the defaults loan class loans of default
31 Dec 2005– 8,600 60 0.69%
1 Jan 2007
31 Dec 2006– 9,300 80 0.86%
1 Jan 2008
31 Dec 2007– 10,500 90 0.91%
1 Jan 2009
124 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Based on the above consideration, in the third period and during the com-
position of the portfolio, PD is also given by:
DC2008 90
PD2008 100 100 0.91%
TC (100006080)
Finally, if the three-year period from 2006 to 2008 is conceived as a full
economic cycle period, the PD, applied in the forthcoming calculations for
the estimation of credit risk, is the average value of the above-mentioned
separate annual PDs. Hence, the PD is given by:
DC2006 DC2007 DC2008
PD2006 0.82%
3
The above-mentioned calculation is rather a statistical one and this is
based on actual past information of default cases.
Qualitative criteria
In many cases the estimation of the probability of default needs more than
the aforementioned quantitative definitions to effectively describe the
credit-standing of the debtor. Thus, the ‘unlikeliness-to-pay’ is a qualitative
criterion that refers to cases where the debtor, albeit that it has not been
exceeding the threshold of μ months, is observed to be unlikely to pay its
obligations to the bank. Note, moreover, that the qualitative criterion
‘unlikeliness-to-pay’, as well as the quantitative set period μ after which the
contract (i.e. Murãbaha) is considered as default, are both unclear in the
estimation of the probability of default.
Using qualitative criteria, most of the times they are driven by the
quantitative ones. Therefore, the ‘materiality’ referring to the past payment
instalments or defaults of a debtor (or type of debtor) is the quantitative
element that is considered in the behaviour analysis of the same contract or
similar ones that have been defaulted. In other words, and going back to the
historical events, the risk manager should separate all default events according
to the quantitative definition of default and then establish the aforemen-
tioned level as the ‘materiality’ that the risk manager is pursuing. Bear in
mind, however, that the past behaviour may give only a part of the truth
about the future behaviour of the obligator. Such analysis is very much
dependent on the qualitative criteria that the risk analyst has to set accord-
ing to the availability of the corresponding quantitative information that will
support the analysis. For instance, the past payment instalments of a debtor
under study that refers to a Murãbaha short-term loan may give a minor
indication for the future rental payments in the long-term Ijãrah contract.
CREDIT RISKS IN ISLAMIC FINANCE 125
Failure to choose the right selection of qualitative and quantitative criteria
will most probably give a wrong profile in terms of the debtors’ creditability.
Exposure at default (EAD)
Exposure at default (EAD) is the estimation of the institutions’ exposure in
the event of, and at the time of, counterparty defaults. Based on the Basel
Credit Risk Model, the potential exposure, in currency, is measured for the
period of 1 year or until the maturity date, whichever comes first. Under
Basel II, a bank must provide an estimate of the exposure for each transaction,
commonly referred to as Exposure at Default (EAD), in a bank’s internal
system. All these loss estimates should seek to fully capture the risks of an
underlying exposure.
Under the foundation approach proposed by Basel II, the EAD is
calculated by taking into account the underlying asset, forward valuation,
facility type, and commitment details. It is important to note that guaran-
tees, collaterals, or securities are not considered (i.e. the estimation ignores
Credit Risk Mitigation Techniques with the exception of on-balance-sheet
netting, where the effect of netting is included in Exposure At Default). For
on-balance-sheet transactions, EAD is identical to the nominal amount of
exposure. If the advanced approach is applied, banks are allowed and
obliged to determine the appropriate EAD to be considered for each
exposure. However, in this case, the bank will need to demonstrate to its
supervisor that it is able to meet additional minimum requirements pertinent
to the integrity and reliability of these estimates. All estimates of EAD
should be calculated for any specific provisions where a bank may have
raised against an exposure.
Loss Given Default (LGD)
Loss Given Default (LGD) is a weighting of the loss when a default occurs.
After realising that a default has taken place, financial institutions that have
granted a credit to an obligor try to recover the outstanding amount of the
defaulted event. The LGD is derived within a credit risk model by taking
into account any collateral or security that covers the corresponding loss
due to the default. If a credit institution fully recovers the outstanding
amount (exposure at default), the resulting loss would be zero. However, the
procedure of the loss coverage may take a long time, in many cases several
years, and thus there are various elements that should be taken into account
within this period.
In the case where the procedure is long-term, all the cash flows are
shifted to an unknown future date, but not necessarily to the same date. The
payments of the agreed amount or the deliveries that are received at a future
126 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
date are not worth the same (most of the time, less) than if they were
received or delivered according to the original schedule. The financial insti-
tution may also face liquidation problems, i.e. from default of payments in
Murãbaha contracts, as well as inability to fulfil its own obligations, i.e.
defaulting to deliver in Salam and Istisnã contracts. In addition to the actual
cost resulting from the default, the procedure for collecting the outstanding
amount assumes some administrative costs. Such additional costs may
include operational costs and legal costs, such as cost incurred when suing
the obligator and the costs of lawyers.
The loss given default (LGD) is usually defined as the ratio of losses to
exposure at default (EAD). There are broadly three ways of measuring
LGD:
Market LGD, which is based on the observed prices from the
market soon after the time where the actual default event occurs. This is the
market price of the goods in the Murãbaha contract, the market rental in the
Ijãrah contract, the commodity or asset price in the Salam and Istisnã
contracts.
Workout LGD, which is based on the estimated cash flows, result
from the contracts that default considering the timing of the event. This is
the expected cash flow from the payments in the Murãbaha contract, from
the renter in the Ijãrah contract, from the sales of the commodities or assets
in the Salam and Istisnã contracts, and from the profits in the Mushãrakah
and Mudãrabah contracts of partnerships. Note that this method is more
complicated than the directly observed market LGD.
Implied Market LGD, which is an entirely different approach to
obtain an estimate of LGD. It considers the ‘credit spreads’ of the (much
larger universe of discourse) non-defaulted, but however risky, cases in
Islamic contracts that may result in defaulted events.
An important parameter to estimate in the identification and assessment
analysis of the LGD is the estimation of the recovery rate (RR). When the
EAD is expressed in absolute terms – that is, in its nominal value at the time
of default – the recovery rate is expressed in percentage terms, as shown in
the following equation (4.3):
兺冤 冥
T i i i
i1 (1)
RR 100 (4.3)
EAD
where i is the variable attributed to the recovered part of the asset paid in
cash by the obligor at time interval i, i is the variable attributed to the
CREDIT RISKS IN ISLAMIC FINANCE 127
recovered part of the potential collateral related to the asset at time interval i,
i is the variable attributed to all administrative costs realised at time interval
i, and is the variation of market value in percentage. Note that, as it is very
difficult to assign costs or receivables at a specific point in time, financial
institutions should be permitted to assign them using time intervals (yearly or
shorter time buckets). In this case, the cash inflows and cash outflows are
assumed to be realised at the end of the time interval. Note, finally, that the
accuracy of recovery rate increases when the time interval decreases.
The LGD is related to the recovery rate as defined in the following
equation (4.4):
LGD ⫽ 1 – RR (4.4)
As mentioned above, the LGD parameter increases its value in relation to
the collateral asset. Thus, the LGD parameter relies heavily on the nature of
the asset. If the asset is not backed by a collateral asset, the LGD will be
high; if the asset is backed by a collateral asset, the LGD will be lower.
The calculation of the LGD is one of the most challenging tasks for
financial institutions. Some of the main reasons that make such estimation
so difficult are:
the availability of quality data for defining the default, assessing the
recovery from the collaterals, and estimating the cost resulting from the
defaults;
the different priorities of payments in relation to other payments the
obligor is past due;
the legal regime for collections before or after the obligor declares itself
bankrupt may differ from country to country;
the legal regime for collections before or after the obligor declares itself
bankrupt may differ from time to time throughout the collection period;
the uncertainty of the duration of cash payments made by the obligor.
the continuous changes (variations) in the market values of the assets
that are directly applied in equation (4.3).
Credit value-at-risk based on expected and
unexpected losses
In market risk, the actual returns on assets are used in order to construct the
returns’ distribution of a specific product, or the returns’ distribution of a
128 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
portfolio of products. This distribution is the basis for market VaR estimation,
driven by the notion of maximum loss for a certain confidence level and a
pre-specified holding period. Similarly to market risk, the estimation of the
credit VaR is based on the distribution of the actual losses observed in a
credit portfolio considering also certain confidence level and a pre-specified
loss-holding period. Note that, in market risk, the assumption that usually
underlies the distribution of the portfolio is that it follows multivariate
normal distribution. On the other hand, credit risk portfolios are likely to
follow distributions that are close to lognormal.
There are two types of losses, notably expected and unexpected, that are
considered in the estimation of Credit VaR. These losses are placed in
the map of loss distribution as illustrated in Figure 4.4. In principle, the
expected loss is related to the loss that a financial institution may incur
during its ordinary business. Therefore, financial institutions are able to
continue their operations by simultaneously having a certain level of losses.
In the Basel II framework, this level of losses is related to the level of
provisions that a bank should set aside in order to cover these losses. On the
other hand, unexpected loss is the loss that a financial institution may face
under extreme circumstances that lie beyond the level of expected losses.
The Expected Credit Loss (ECL) is calculated based on the PD, LGD,
and EAD as shown in equation (4.5):
ECL PD LGD EAD (4.5)
Taking into account that the EAD depends on the culture and specific
policy of the individual financial institution, in this analysis it is assumed to
UCL ECL
Probability
Confidence
level
Losses
Figure 4.4 Distribution of expected and unexpected losses in credit risk
CREDIT RISKS IN ISLAMIC FINANCE 129
be one unit currency. Based on the above assumption, the estimation of the
expected credit loss defined in equation (4.5) transforms into equation (4.6):
ECL PD LGD (4.6)
The Unexpected Credit Loss (UCL), on the other hand, is based on full
parametric approaches and thus can be represented as in the following
equation (4.7):
UCL credit m (4.7)
where m is the appropriate quantile of the credit loss distribution, in order
for the model to reach the credit VaR confidence level given by equation
(4.8):
creditEL
VaR(•, • •)
m credit UL (4.8)
credit
where VaR(•, • •)
credit is the measure of Credit VaR for • confidence level and ••
holding period, and credit is the volatility of the individual asset expressed
in terms of the volatility of PD and LGD as defined in the following
equation (4.9):
credit 兹PD 2LGD LGD2 2PD (4.9)
Using the distribution illustrated in Figure 4.4, the ECL and UCL are
distributed over a target horizon. The credit VaR is then measured as the
deviation away from ECL as defined in equation (4.10):
credit ECLUCL
VaR(•, • •)
(4.10)
Based on the above equation (4.10), the difference between credit VaR
and expected loss defines the unexpected credit loss as in equation (4.11):
UCL ECLVaR(•, • •)
credit (4.11)
The credit VaR refers to the unitary one that is the VaR used to quantify
the risk of a unique credit element, such as a single Murãbaha contract. In
reality, however, there are multiple assets in a credit portfolio. The analysis
of correlations between the portfolio’s assets is particularly useful for the
estimation of the LGD of a credit portfolio. Moreover, in the multiple assets
portfolio, the correlations between the defaults that have occurred on its dif-
ferent assets are used in the estimation of the portfolio’s PD. Note that the
130 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
former type of correlation is named ‘assets correlation’ and the latter
‘default correlation’. The analysis of the assets correlation is performed in a
similar way to the returns in market risks. On the other hand, the analysis of
the default correlation is based on the cases and the degree where a default
in one asset may or may not cause another asset to become defaulted. In
turn, using the default correlations, the standard deviation p, credit of the
credit portfolio is estimated as illustrated in equation (4.12):
冪兺 兺 UL UL
n n
p, credit ij i j (4.12)
i1 j1
where ij is the correlation coefficient for assets i and j and it is given by
equation (4.13):
P(Di Dj)PDi PDj
ij (4.13)
兹PDi (1PDi)兹PDj (1PDj)
where P(Di Dj) is the correlation between the defaults on assets i and j.
In the case of parametric models and based on equation (4.7) and the
standard deviation (p, credit) defined in equation (4.12), the corresponding
UCL of the credit portfolio is given by the equation (4.14):
UCLp p, credit m (4.14)
Note that the assumption in the above definition of the UCL is that the
portfolio follows the same probability distribution as the individual asset.
The unexpected loss of the portfolio of credit assets consists of the sum of
the unexpected losses of the individual assets a that the portfolio comprises
of, as defined in equation (4.15):
a
UCL 兺 UCL
i1
i (4.15)
Economic capital and Credit VaR
The Credit VaR estimation should be viewed as the economic capital to be
held as a buffer against the UCL. This is because the estimation of the eco-
nomic capital considers the risk arising from the unexpected nature of the
losses as distinct from expected losses (as illustrated in Figure 4.4), which
are considered as part of doing business and are covered by reserves and
incomes. Thus, in credit risk, economic capital is required as a cushion for
an institution’s risk of unexpected credit default losses, because the actual
level of credit losses suffered in any one period could be significantly higher
than the expected level. Note that economic capital covers all unexpected
CREDIT RISKS IN ISLAMIC FINANCE 131
events except the catastrophic ones, for which it is not possible to hold
capital.
In economic capital estimation, there is no common agreement on the
definitions referring to key parameters such as confidence level and time
horizon that should be well defined and set. Most of the time during the
estimation of the economic capital, credit institutions choose a higher
confidence level; for instance, more than 99 or 99.9 percent that is usually
fixed by the regulatory authorities. Moreover, the time horizon in economic
capital calculation may also vary. Financial institutions that provide Islamic
products should apply a time horizon that reflects the contracts that may
construct the institution’s accounts of their portfolios. Thus, when portfolios
of Mushãrakah and Mudãrabah contracts are considered, the time horizon
may need to be long (years) in the estimation process of the corresponding
economic capital. On the other hand, the confidence level referring to
accounts or portfolios for the Murãbaha contracts should be set on a high
degree to align with the expected value of credit risk.
Loss data and Credit VaR estimation
The suitability of the methodologies used for producing efficient estima-
tions of the Credit VaR is related to the loss data distribution and thus should
be assessed within the context of the corresponding availability of historical
loss data. Thus, financial institutions may keep records in regards to the
losses resulting from the counterparty defaults of re-payments in Murãbaha
and Ijãrah contracts. Moreover, in the Salam and Istisnã contracts, the
losses initiated from the delays of the sellers, manufacturers, and/or
constructors. Also, the losses resulted from the corresponding payment
inability from the buyers or users. Financial institutions should also know
the history of their losses concerning their Mushãrakah and Mudãrabah
partnership agreements.
Within financial institutions, the above-mentioned data may only be fully or
partly available, or in some cases are even missing completely. Therefore, as
will be discussed in the following paragraphs, different techniques should be
employed, depending on the availability or non-availability of the loss data.
Fully available historical loss data
In the case where financial institutions have full availability of historical
loss data, there are two methodologies that could be applied for estimating
the Credit VaR. These are the historical simulation method as well as the
variance–co-variance method applied to historical losses.
When the availability of historical loss data refers to a long time period,
for the entire distribution of losses, the credit VaR can be estimated either
132 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
over the entire observation period, named ‘inter-temporal credit VaR’, or
based on an individual period, e.g. year, named ‘year-in-time credit VaR’
estimation. In the former case, data need to be collected in large quantities
for each year, attributing each loss to the respective credit asset. The
sequence of annual credit losses comprises time series of credit losses. In
the second case, where only one year is considered, the credit loss observed
should be attributed to each asset or each asset class. Note that asset classes
exhibit similar characteristics, behaving similarly in the marketplace, and
are subject to the same laws and regulations. Optimally, a completed and
constructing credit loss distribution should include the loss that has
appeared during the past year (365 days) for each day and each asset class.
The loss for each asset should be accounted for in the year in which it
appears. Consequently, there would be daily observations for annual losses
for every year and every asset class.
Partly available historical loss data
Most financial institutions do not have adequate historical loss data that can
be used for the construction of the entire probability distribution over the
entire period of time. Thus, in many cases the historical data referring to
existing past losses are only partly available, which means that the adequacy
of the historical database is unable to fully aid the credit risk analysis for
estimating the Credit VaR. Therefore, when historical loss data are missing,
the most appropriate methodology to model credit losses should be based
on both available historical data and Monte Carlo simulation.
For instance, if for each year there is availability of only scattered and
scarce data of asset losses then, based on these data, the Monte Carlo
simulation should be applied. This method defines a number of the possible
simulated distributions of the loss data for each single year. To complete the
construction of the credit loss evolution, Monte Carlo simulation should
be applied on each asset or asset loss for multiple years.
Lack of historical data
In order to estimate the loss distribution, financial institutions that lack
historical loss data should apply Monte Carlo simulations. Based on the
values of the assets that construct the credit portfolio, the loss distribution
can be inferred. By applying Monte Carlo simulations, these asset values
are shocked and thus generate new distribution of losses. However, it is
important to note that, while using Monte Carlo simulations, an assumption
is made about the distribution from which random numbers are drawn. This
assumption should be based on the anticipated shape of the distribution of
credit losses.
CREDIT RISKS IN ISLAMIC FINANCE 133
C R E D I T R I S K M I T I G AT I O N
The exposure at default (EAD) can be reduced using several mechanisms
that are also recognised by Basel II. As already discussed earlier in this
chapter, institutions are exposed to credit risk that is initiated from Islamic
financial products and correspond to lending in the Murãbaha, leasing in
the Ijãrah, promises to deliver or to buy in Istisnã and Salam, and invest-
ments failure in the Mushãrakah and Mudãrabah contracts. Moreover,
institutions are exposed to additional credit risk via the guarantees, referring
to the above Islamic financial products, granted by institutions, organisa-
tions, or even individuals. Institutions, however, are mitigating credit risk
exposures by employing collaterals or guarantees granted by tier counter-
parties. A Guarantee is a contractual agreement with a third party which has
a lower risk weight than the original counterparty of the exposure and will
take over the claim in case of default of the original counterparty. The
fraction of the exposure which is covered (by the guarantee amount) has a
lower risk weight and contributes to a lower capital charge for the institu-
tion. Collateral, on the other hand, is an asset used partly or totally to secure
the obligation of the counterparty by refunding the losses in the event that a
default occurs.
For financial institutions using collaterals, the mitigation of credit risks is
expressed in terms of reducing the EAD or LGD whenever they apply the
Basel II standardised or IBR foundation approach, respectively. On the
other hand, using guarantees (for the covered part of the exposure), the
reduction is expressed in terms of risk weights. Taking these mechanisms
into consideration, financial institutions are allowed to reduce the capital
charge reserve associated to different credit risk activities. Note that the
quality (credit rate) of the guarantor should be better than the quality of the
direct counterparty that guarantees its exposure.
When collaterals and guaranties are applied, the financial institution
should consider specific rules concerning:
1. The eligibility that looks at whether the operational conditions are
met, and thus the institution may or may not be allowed to use specific
pieces of collateral. Such conditions are defined by the regulators of cen-
tral banks, IBSB,2 Basel II, etc. The same selection mechanisms could
apply in a similar way for guarantees. One should pay attention to the case
where the counterparty, which exposure is to be covered by a guarantee,
belongs to the same group as the one granting the coverage. There could be
some accrued risk that both counterparties could experience default at the
same time.
2. The amount of collaterals and guaranties that cover all the exposures
from all contracts that are attached to the counterparty. The exposures are
134 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
summed up and the collateral amount is used to reduce the sum of the whole
exposure.
3. The allocation of the coverage. The way that the coverage amounts
from collateral and guarantees are allocated (proportionally or based on the
actual risks) to reduce the risk associated to different claims.
C R E D I T R AT I N G S Y S T E M S
Credit rating is indicating the credit-worthiness and the reliability of the
counterparties (i.e. individuals and corporations, etc.) in regards to its finan-
cial obligations. There are different types of credit rating systems according
to the rating dimensions, the rating assignment techniques, and the histori-
cal data used to produce the ratings. The degree of credit rating is mainly
based on an assessment that combines both quantitative and qualitative crite-
ria, as were discussed earlier in this chapter. Such criteria are referring to the
behaviour (mainly historical) of the counterparty’s in regards to their bor-
rowing and repayments, as well as the availability of assets and extent of lia-
bilities. Regardless of the type and the methodology used for the rating
system, it should be well validated in order to give confidence in its suit-
ability for the specific financial institution and in order that it depicts the
pragmatic credit-worthiness of the obligors. Therefore, validation is the
internal assessment of the credit institutions’ rating system and it should
involve regular reviews of the rating system’s quality and suitability on the
basis of ongoing operations.
Rating systems therefore consist of several credit rating grades that rank
the credit-worthiness of the obligors. At the same time, the credit ratings are
also indicating the relative risk on the institution’s credit exposure. In most
Westernised econonomies, ratings are available from other independent
firms, i.e. Standard & Poor’s, Moody’s, and Fitch. However, many financial
institutions, including the ones that are providing Islamic financial prod-
ucts, may hire their own credit analysts and thus prepare and use their credit
ratings for internal use.
In its most widely known form, a rating system might be either expressed
as letters (A, B, C, etc.) or as whole numbers – for instance, from 1 to 10;
the former type of ranging expression is mainly used by the public agencies,
whereas the latter is usually expressed by a bank’s internal ratings. Table 4.2
provides a classification of the credit rating degrees from AAA to D. The
larger the number of grades, the better, because it means better discrimina-
tion of credit risk and, at the same time, avoids granularity. Note that, for
most rating bodies, the counterparties that are rated between AAA to BBB
are within the investment zone, whereas lower grade, i.e. BB to D, are
within the speculative zone.
CREDIT RISKS IN ISLAMIC FINANCE 135
Financial institutions may classify credit risks depending on obligor char-
acteristics; in this case, the rating system is assumed to be one-dimensional.
On the other hand, institutions may assess the rating based on the charac-
teristics of the transactions and thus evaluate the loss severities; in such a
case the rating system is two-dimensional. Obligor and loss severity ratings
must be calibrated to values of the defaults and the loss given defaults
(LGD), respectively. Therefore, financial institutions may assign obligor
ratings that are associated with the PDs. Accordingly, the loss severity rat-
ing is associated with the values of the LGD, or is directly assigned to the
LGD values for each transaction (facility).
There are two different types of techniques for assessing the credit rating
systems, namely:
point-in-time and through-the-cycle rating; and
qualitative and quantitative rating.
They are both explained analytically in the following paragraph.
Point-in-time and through-the-cycle
Based on the point-in-time rating systems, the main driver of the assessment
is the information at the time of the rating. Thus, the obligors are assigned
on certain ratings according to the frequency of ‘default’ cases concerned
within a particular time or relatively short time period, i.e. a year. Thus, the
point-in-time ratings changes from one period to the other as borrowers’
circumstances change, moreover including the changes initiated from the
current or short time market economic conditions. Applying, on the other
hand, the through-the-cycle rating systems, the ratings take into considera-
tion the obligors’ attitude through all the phases of the cycle, as well as dis-
tressed circumstances that might influence the obligors’ credit-worthiness
in the long term. Similarly to the point-in-time, the through-the-cycle rat-
ings system should be updated from period-to-period (i.e. year-to-year) due
to the corresponding changes in a borrower’s conditions. However, since the
through-the-cycle rating systems take into consideration downturns in the
economy and consider stress circumstances, period-to-period ratings are
less influenced by the changes in the economic environment.
Obviously, the specific short periods for the point-in-time and the cycle
periods for the through-the-cycle systems that should be considered for
assessing the ratings defers for the different types of the Islamic financial
products linked to the counterparties. Thus, in the Murãbaha and Ijãrah
contracts which are dealing with the normal repayments, the period and
cycles under consideration are usually smaller that in the Salam and the
136 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Istisnã contracts that mainly refer to manufacturing and/or construction
projects.
In general, the point-in-time rating systems are more suitable for estimat-
ing the economic capital as it is taking into account the defaults for a par-
ticular period of time. It is also suitable for identifying the counterparty’s
credit behaviour of a loan as well as for calculating provisions. On the other
hand, for long-run contracts and investment purposes, the decision-making
should be based on through-the-cycle rating systems analysis. Rating agen-
cies are commonly known to use through-the-cycle rating approaches.
Qualitative and quantitative rating techniques
In rating, financial institutions and agencies may use methodologies that are
based on qualitative judgemental criteria or quantitative information data or
a combination of the two. Although there are different techniques and
approaches for applying them, the best choice depends on the case under
consideration and the availability of the information and the risk philosophy
of each financial institution.
The qualitative rating systems demonstrate the results of an empirical
rating model. Empirical models are based on experts’ judgement and
experience. Their combination with quantitative information, as discussed
later, is most appropriate. The quantitative rating systems, however, are for-
mulated based on mathematical rating models that are capable of using and
combining the available quantitative information. Such credit risk rating
classification models utilise techniques based on two types of structure:
parametric; and
non-parametric
The first category is based on more analytical mathematical techniques
where the different model parameters and their relations need to be well
defined. Moreover, any assumptions made should also be defined and
considered in the model’s results accuracy. For rating purposes, statistical
mathematical models are mainly used; however, such models are very
dependent on the data quality and availability as well as the techniques used
to analyse the statistical results. Poor data and analysis may provide inefficient
rating results.
Non-parametric-based models are used to estimate and forecast the main
components of credit risk assessment such as the PD and LGD. Such
models are usually based on neural networks, expert systems, fuzzy logic
systems, and multi-criteria analysis. Using neural networks, for instance, a
large number of quantitative information may be used and transferred to
CREDIT RISKS IN ISLAMIC FINANCE 137
artificial intelligence knowledge; expert and logic management systems are
capable of combining qualitative criteria with quantitative information data.
By applying linguistic rules, the rating systems are designed similarly to the
way the human ‘experts’ think; increasing, therefore, their robustness to a
high degree. The linguistic rule-based expressions are simple mathematical
equations and thus they allow fast and uncomplicated design of the pro-
posed rating grading methodology, where previous knowledge on default
cases may not be fully available. Moreover, using such techniques, rapid
prototyping of the rating strategies can be implemented, because their
design is only based on the expert knowledge and thus an immediate com-
mencement of the implementation can be made. As a result, such techniques
minimise the cost of design and implementation. Finally, and more impor-
tantly, a significant benefit in applying such an approach is the transparency
of the rating process based on linguistic rules that are easy to understand
and modify. However, such techniques are mainly dependent on the
designer’s expert knowledge; thus, a knowledgeable designer will produce
more efficient rating rules.
The structure of most Islamic financial contracts has a high degree of ‘qual-
itative’ flexibility, in terms of the definition of default as well as in the assess-
ment of grading the counterparties. As a result, a mixed based approach where
both qualitative criteria and quantitative information are combined should be
most appropriate for applying rating systems to such products.
Table 4.2 Classification of the levels of credit ratings with the indication of the
creditability and the reliability as well as the meaning in terms of exposure
Rating Creditability Exposure Meaning
Degree Reliability
AAA Highest Lowest The creditability of the counterparty
is rated to the highest (best) degree –
the counterparty is extremely
reliable with regard to financial
and/or delivery obligations and
thus is exposing the institution
to a lower degree.
AA Very high Very low The counterparty’s creditability is
rated to a very high (very good)
degree – the reliability of the
counterparty in regard to its financial
and/or delivery obligations is still very
high and thus the exposure of the
institution is still to a very low degree.
Continued
138 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Table 4.2 Continued
Rating Creditability Exposure Meaning
Degree Reliability
A High Low The creditability and the reliability of
the counterparty to financial and/or
delivery obligations are still rated high,
whereas the exposure of the
institution to credit risk is to a low
degree.
BBB Medium Medium The degree of both creditability and
reliability of the counterparties may
be susceptible to economic
conditions, the credit risk exposure
are at a medium degree.
BB Caution Caution Caution for the creditability and
reliability of the counterparties as
well as for the institution’s exposure
is necessary. However, the
counterparties rate have the best
sub-investment credit quality.
B Partially Partially The creditability and reliability of the
acceptable acceptable counterparties with this rate are
partially acceptable. It is vulnerable to
changes in economic conditions.
CCC Low High Such rate implies that the
counterparties are currently vulnerable
to non-payment and/or non-deliveries,
whereas their behaviour is highly
dependent on favourable economic
conditions. The degree of exposure in
credit risk is high.
CC Very low Very high The counterparties with this rate are
highly vulnerable to a payment and
delivery default. Their creditability is
very minor. Consequently, the
institution is exposed to a high
degree.
Continued
CREDIT RISKS IN ISLAMIC FINANCE 139
Table 4.2 Continued
Rating Creditability Exposure Meaning
Degree Reliability
C Significantly Significantly Counterparties with such rates are
low high close to or already bankrupted;
however, for payment and/or delivery it
may continue with its obligations. The
institution is exposed to a significant
degree.
D Dangerous Alerted Payment default on some financial
and/or delivery obligations have
actually occurred from the counter-
parties that are rated with this rating
grade. Institution is on alert position
due to such defaults.
VA L I D AT I N G T H E C R E D I T R AT I N G S Y S T E M S
The credit rating system that the institution is using should be validated at
pre-defined intervals as well as whenever there are new or modified
parameters due to market and contract conditions. For this validation, the
qualitative criteria must firstly be examined and then the quantitative analy-
sis should be used in the systems’ rating process.
Qualitative validation
The validation of the Credit Rating Systems that is based on qualitative cri-
teria should consider:
the design and implementation of the model for the rating system;
the quality of the data used as input to the rating system; and
the internal use of the rating system in the credit institution’s credit
approval process (testing the use of the rating system).
Financial institutions must give a great emphasis on the validation of the
qualitative criteria. Insufficient qualitative validation could provide the
institution or the regulators with a wrong valuation of the appropriateness of
the rating system. As the qualitative criteria are rather subjective, their
140 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
validation is a rather difficult process and their validation, however, is very
critical as it can influence the future exposure of the institution.
Design and implementation of the rating
system’s model
The institution must provide a full documentation for the design and
implementation process for the precise model used for the rating system so
that it can easily be valuated. Therefore, the documentation should be trans-
parent, consistent, and complete. More analytically, such document should
at least include:
the selection of the criteria chosen for the rating system that is
implemented;
the type of the rating system model;
the description of the rating method;
the dataset used as an input to the rating system;
the validation process and possible results during the model development;
the mathematical functions used to implement the model;
the definitions of the above functions as well as of the non-parametric
components of the model;
the definitions of the obligor’s ‘default’ as well as of the ‘loss’; and
the calibration of the model output to the default probabilities.
The data integrity issue
Data integrity plays a major role in the quality and trustworthiness of the
results of the rating systems. The data collection builds the historical
database referring to the past default cases, in payments or deliveries, and
generally all the information used to identify the counterparty’s behaviour.
This is essential for the development of a rating system that is built on
adequate and reliable history and that leads to accurate estimates of the
parameters of credit risk – such as the PD, LGD, and EAD. Any definitions
together with the assumptions made for the use of these data as input infor-
mation to the rating models should also be well defined.
CREDIT RISKS IN ISLAMIC FINANCE 141
Financial institutions that provide conventional and/or Islamic financial
products should follow some basic requirements for the data definition,
collection, and usage:
the rating history of the obligors and guarantors;
the time where the data have been used for assigning the ratings;
the key data and the associated data management methodology used to
derive the rating;
the data that identify the obligors and exposures resulting out of their
default status;
the dates and circumstances of such defaults;
data on the PDs associated with rating grades.
In the event that the financial institutions use their own estimates of the
LGDs, then they should be able to collect and store complete historical
information data on the ratings such as the LGD estimates associated with
each rating scale, the dates the estimates were done, the methodology used
to derive the facility ratings, and realised LGDs associated with each
defaulted exposure.
Testing the use of the rating system
Financial institutions must be able to test and provide an approval process
for the internal use of their rating systems. The consistent development of
the rating system should be followed by a credit approval process that
includes, and are not limited to, the setting of the strategic limit, the pricing
of the profits (payments), and the definitions of the delivery for the different
Islamic products and partnership contracts, and in general the institution’s
policy concerning credit risk management.
Quantitative validation
In the quantitative validation, what should be examined is the ability of the
rating system to discriminate, based on quantitative analysis, between
solvent and distressed obligors and to categorise them according to the
appropriate classes and seniorities. Moreover, the accuracy and robustness
of the rating results should be valuated as they play a significant role in
terms of the rating system’s efficiency and reliability. Good robustness is a
142 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
guarantee of the stability of the system that will need to be sensible on small
modifications of the input data without, however, applying any changes in
the rating model. Note that the rating systems are updated from time to time
and thus should be capable of being modified when there are changes in
conditions referring to market, counterparties, strategies, etc.
Based on the quantitative analysis, the validation of the Credit Rating
Systems, in regards to the accuracy of their estimates, should consider:
the functionality (parametric and non-parametric) and the supported
methodologies;
the back-testing;
the benchmarking; and
the stress-testing.
Functionality test
The functionality and the supported methodologies applied in the rating
system are mainly tested based on statistical tools, including the CAP and
ROC curves, the Gini co-efficient,3 and the Kolmogorov–Smirnov test.
Such tools are comparing the usage of the parametric and non-parametric
components of the system with its ability to discriminate between solvent
obligors and the ones with a higher probability of default; moreover, they
measure the integrity of a rating system, regarding the allocation of risk
between the obligors.
One of the most common techniques for the statistical test is to define
good benchmarks where the performance of the model will be compared in
terms of its discriminatory power and results, to some confidence level.
Unfortunately, it is difficult to find or define realistically the perfect bench-
mark as every structure of the financial institution and every synthesis of
their contracts’ portfolios varies. Several statistical tests should be
employed by the experts that analyse the results of the system’s functional-
ity. An efficient number of tests (according to the complexity of the model)
will give the ability of explaining the results accordingly and in relation to
each other. As mentioned above, the discriminatory power, the stability, and
the robustness of rating systems should be on a satisfactory level in regards
to the following basic validations:
the in-sample validation, where two-thirds (66.66 percent) of the
sample is the ‘training sample’ and is used for modelling buildings or, in
CREDIT RISKS IN ISLAMIC FINANCE 143
other words, identifying the parametric and non-parametric components
based on the observed data, and fitting them to the credit risk evaluation;
the out-of-sample validation, where the remaining one-third (33.33
percent) of the development in-sample is used as the ‘test sample’ of the rat-
ing system’s performance;
the out-of-time validation, where the validation is divided in two
periods: the observation period (usually a year of data collection from
accounts booked during this period) and the performance period (usually
12–18 months after the observation end date) of the accounts observed.
Then, the rating system’s analysts should test the behaviour of the model
with a sample from a more recent period, which is usually the period after
the performance date.
Figure 4.5 illustrates the periods and samples used for the rating valuation.
Back-testing
By using back-testing, the model of the rating system is compared with the
realised historical data of the credit institution. Note, however, that the data
used for the back-testing should not only be available but also reliable for
the purpose of the particular testing. Normally, the frequency of such testing
is at minimum once every year. Financial institutions should then check
whether the realised default rates are between (or beyond) the estimated PD
range of each rating class. In the case where the rating system is built in-
house, back-test should also be performed for the LGD parameters. In cases
where the estimated parameters differ significantly from the institution’s
history, the causes of the problem should be detected and the appropriate
adjustments should be made to the model. It is important to highlight that
such comparisons in the back-testing process, together with any model’s
modification, should be clearly documented. Note, finally, that it is often
alleged that, using the back-testing approach, the resulting parameters of
credit risk (PD, LGD) may not be very accurate. This allegation concerns
mostly low-default portfolios. In general, financial institutions should
In-sample Out-of-sample Out-of-time
Past Future
Observation Performance Recent
period period period
Figure 4.5 Periods and samples used for the rating valuation
144 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
perform statistical tests using back-testing and define (as a policy) the
allowable deviations between realised and estimated data.
Benchmarking
By applying the benchmarking approach, there is a comparison between the
estimations of the rating system’s model with the results of external assess-
ment institutions and prestigious and broadly known models. This method
is based on qualitative comparative analysis where the two types of results,
actual and benchmarked, need to be aligned using the same underlying
measurements. Such an approach is very useful, especially when there is a
lack of efficient and reliable past data and the back-testing is unable to give
credible results. Data referring to Islamic financial products may raise such
issues and thus benchmarking may be extremely beneficial for the financial
institutions to apply and develop a system of Islamic products’ benchmarking
that will be interpreted in relation to the results from back-testing
Stress-testing
The stress-testing scenarios are employed to ensure the forward-looking
ability of a model. In other words, it tests the ability of the model from nor-
mal to extreme conditions. Therefore, the scenarios take into consideration
the phases of a down-turn in the economic cycle that may affect the pay-
ment or delivery capabilities of the counterparties of the Islamic contracts
(i.e. Murãbaha, Ijãrah, Salam, and the Istisnã); and consequently may
influence the business performances that are related to the Islamic financial
partnership contracts (i.e. Mushãrakah and Mudãrabah).
Stress-testing exercises must be a part of the institution’s internal
ongoing risk management processes as one of the main tools for evaluating
the model of the rating system, and measurement of the influence on the
capital adequacy. Risk analyst must design and perform stress-testing
scenarios that are representative of the credit institution’s history and
experience, by also taking into consideration the economic and business
conditions that are referring to Islamic financial product and contracts.
Stress-testing scenarios should be performed on a regular basis, at least
once every year, considering, in addition to the above, the assessment of the
capital adequacy as well as any strategies of the financial institution in terms
of their product, portfolios, and overall financial and business policies.
All the assumptions made for risk scenarios must be well defined and
considered in the testing evaluation process. Thus, based on Basel II regula-
tions, the institution must hold a capital amounting to at least 8 percent of
their risk-weighted assets in order to cover their credit risks. Another crucial
parameter is the probability of default by the individual borrowers. Analysts
CREDIT RISKS IN ISLAMIC FINANCE 145
may increase the probability of default by 30 to 60 percent during the
different scenario designs and thus to increase stress-expected loss that are
corresponding to the different downgrades in the rating scale of the Islamic
products’ portfolio.
Risk analysts may also select the appropriate scenarios that refer to the
rating models that can be based on statistical methods. The risk parameters
are estimated from historical data that may also be used in the identification
process of the rating model, and then scenarios representing the extreme
events of the probability distribution are selected. Finally, another scenario
option is the model-based analysis, especially in the case where macro-
stress tests are applied. This method is based on an econometric model, in
which the interrelationships of the relevant risk factors can be shown.
Stability and robustness of a rating system
Both the stability and robustness of the rating system being used by the
financial institution should be tested according to the recent profile of the
obligors. Any changes in terms of payments and delivery obligations in the
contracts’ agreements or even on the institution’s strategies should be well
integrated with the model. Thus, the model should be providing efficient
ratings adaptable to small changes; on the other hand, it should be easily
modified when major changes occur. The stability of a rating system should
be tested according to the phases of the economic and business cycle.
Realistically talking in most cases, historical data from a full economic and
business cycle is not available. This implies that the financial institutions are
unable to measure the consequences of a downturn of the above-mentioned
cycles. This is the reason why it is highly recommended that when models are
built, micro-economic and macro-economic variables should be considered,
which could be stressed in order to give results in the case of an economic
recession. The stress scenarios of the macro-economic variables could con-
cern the GDP growth, the unemployment rate, the Consumer Price Index, etc.
SUMMARY
The estimation of the expected loss is relatively easier for simple and
homogenous contracts than for the complex heterogeneous ones. Since the
Islamic financial contracts are more complex than the conventional ones,
financial institutions that provide such types of contracts are facing addi-
tional challenges. As discussed in this chapter, financial institutions are
exposed to credit risk that correspond to lending in the Murãbaha, leasing
in the Ijãrah, promises to deliver or to buy in Istisnã and Salam, and invest-
ments failure in the Mushãrakah and Mudãrabah contracts.
146 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
This chapter has given an insight into credit risk exposure identification.
In credit risk exposure analysis, a key factor is the identification of the rela-
tions between the counterparties, the Islamic financial contracts, and the
guaranties and collaterals used to cover a percentage of the potential losses
in case of defaults. Islamic financial contracts are linked to counterparties
and different types of guaranties and collaterals are used to cover in parallel
the exposure of several contracts and counterparties.
Furthermore, the chapter identified different methods and techniques for
developing models for credit risk that results from Islamic products. The
corresponding advantages and disadvantages of the models were also dis-
cussed to evaluate their use in the implementation phase. The models that
were considered fell under two groups: qualitative and quantitative. Models
under the qualitative group are namely expert systems driven by rules that
combine the criteria used by expert’s judgements. These non-parametric
qualitative rating systems are flexible and free from statistical bias as they
are not based on certain statistical hypotheses. The ones falling under the
quantitative group are based mainly on quantitative statistical-based models.
The aforementioned validation of the quantitative analysis is mainly subject
to statistical random fluctuations and thus there is a certain degree of
freedom when interpreting the results. This is mainly the reason why such
validation is less significant and critical than the qualitative ones. Most
quantitative rating systems are driven by mathematical techniques that are
transparent in their supported methodologies and rating results. However,
they could also be driven by qualitative parameters and so can be influenced
by subjective characteristics. Frequently, the two aspects of such validation
complement each other and are used to form hybrid models. The hybrid
models bring together the advantage of evaluating in an objective way the
methods and results related to the prediction of the obligors’ credit-worthiness
as used in the quantitative models and at the same time take into considera-
tion all the credit-worthiness factors and important information on the
borrowers’ characteristics used in the qualitative models.
The main parameters that should be considered in the valuation of credit
risks are the expected and unexpected losses. This chapter discusses in
depth, using mathematical functions, the calculation of the expected and
unexpected losses, which requires the computation of the probability of
default (PD), the loss given default (LGD), and the exposure at default
(EAD). The entire modelling of credit risk is based on the estimation of the
probability of default and the loss given default. Their definitions and esti-
mation is one of the main challenges that financial institutions are facing
today. This section of the chapter also highlighted credit value-at-risk (VaR)
based on expected and unexpected losses. The estimation of the credit VaR
is based on the distribution of the actual losses observed in a credit portfolio
considering also certain confidence level and a pre-specified loss holding
CREDIT RISKS IN ISLAMIC FINANCE 147
period. The Credit VaR estimation is viewed as the economic capital to be
held as a buffer against unexpected losses. The suitability of the method-
ologies used for producing efficient estimations of the Credit VaR is related
to the loss data distribution and thus should be assessed within the context
of the corresponding availability of historical loss data, as discussed. As
explained, financial institutions that lack historical loss data should apply
Monte Carlo simulations in order to estimate the loss distribution.
The chapter also talked about how financial institutions are mitigating
credit risks by employing collaterals or guarantees granted by tier counter-
parties. For financial institutions using collaterals, the mitigation of credit
risks is expressed in terms of reducing the EAD or LGD. When a guarantee
is used, which is a contractual agreement with a third party, it has a lower
risk weight than the original counterparty of the exposure and will take over
the claim in case there is a default of the original counterparty.
Furthermore, the chapter discussed the credit rating system that the
institution should use. This system must be validated at pre-defined
intervals as well as whenever there are new or modified parameters due to
market and contract conditions. For this validation, the qualitative criteria
must firstly be examined and then the quantitative analysis should be used
in the systems’ rating process. For the qualitative validation analysis, the
design and implementation of the rating system’s model, the data integrity,
and the testing of the usage of the rating system must all be considered. On
the other hand, in the quantitative validation analysis such tests as the
functionality test, back-testing, benchmarking, stress-testing, should be
undertaken. Finally, the stability and robustness of the chosen rating system
must be determined. The model should be able to provide efficient ratings
when small changes arise, but be easily modified when major changes
occur.
NOTES
1. Basel Committee on Banking Supervision, ‘International Convergence of Capital Measurements and
Capital Standards, A Revised Framework’ (2005).
2. Islamic Financial Services Board, Guiding Principles Of Risk Management For Institutions Offering
Only Islamic Financial Services (December 2005).
3. G. Corrado (1912), Variabilità e mutabilità, Reprinted in Memorie di metodologica statistica (Eds.
E. Pizetti and T. Salvemini). (Rome: Libreria Eredi Virgilio Veschi, 1955).
CHAPTER 5
Market Risks in Islamic
Finance
INTRODUCTION
The notion of market risk management and hedging is of recent origin in
Islamic financial markets. Importers and exporters, for example, would
often adopt a ‘do nothing until you need to’ approach to hedging against
adverse exchange rate fluctuations. This was partly due to a culture that, for
many years, encouraged the view that risk was something to be accepted,
rather than to be reduced. It was also an attitude promoted by the fixed
exchange rate to the US dollar ($) which was powerful in most Middle
Eastern countries, particularly because the most important export for most
of these economies – oil – was priced in dollars, where nowadays it is also
exchanged in Euros (€).
According to the guiding principles (provided by the Islamic Financial
Services Board) of risk management for institutions offering Islamic finan-
cial services, market risk is defined as the risk of losses in on- and off-
balance-sheet positions arising from adverse price movements in market
prices, i.e. fluctuations in values in tradeable, marketable, or leaseable
assets and in off-balance-sheet individual portfolios. Additionally, market
risks are risks that are related to the current and future volatility of market
values from specific assets, e.g. the commodity price of Salam or Istisnã
assets, the market value of Ijãrah agreements, the market value of
Murãbaha assets purchased to be delivered over a specific period, and of
foreign exchange rates. Furthermore, market risk appears on trading posi-
tions in Sukûk products. Market risk in the Islamic financial markets inher-
ently exists within the lifetime of the Islamic contracts and, as such, its
increased severity attracted market risk participants to be involved with its
identification, modeling, and control.
In this chapter, the main concepts related to market risks are discussed,
including identification of market risk factors, rate of return risks,
148
MARKET RISKS IN ISLAMIC FINANCE 149
commodity risk in Islamic finance, FX rate risks, and equity price risks. It
also discusses valuation issues on equity prices and FX rates, the quantifi-
cation of foreign exchange risk, equity risk and commodity risk, the data
that are referring to market risk factors, and the sensitivity analysis in
market risk. Market risk valuation models are presented in regards to
identification and implementation of VaR models for Islamic financial
contracts. Evaluation methods of market risk are also presented, including
variance–co-variance method, Monte Carlo simulation method, and
Historical simulation. Finally, the use of back-testing and stress-testing for
market risk exposures are also discussed.
I D E N T I F I C AT I O N O F M A R K E T
R I S K FA C T O R S
By providing Islamic financial products, banks are facing four major cate-
gories of market risk factors that influence the value of the assets held over the
lifetime of the contracts that build up the accounts of financial institutions.
The four types of risks that expose financial institutions to market risk are:
1. rate of return (mark-up) or benchmark rate risks related to market infla-
tion and interest rate;
2. commodity price risks because, unlike conventional banks, they typi-
cally carry inventory items (predefined prices);
3. FX rate risks in the same way as conventional banks;
4. equity price risks, mainly in regards to the equity financing through the
PLS modes.
Figure 5.1 illustrates the four risk factors that exist in the different
Islamic financial products.
R AT E O F R E T U R N R I S K
Interest rate is a significant element used in Westernized conventional finan-
cial products to valuate, align, and harmonise their current and future mar-
ket prices with the corresponding market movements. On the other hand,
the use of interest rate is forbidden in the structure of the Islamic financial
product. Changes in the market interest rate, however, result in some risks
regarding the earnings of Islamic financial institutions. Therefore, for
150 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Market risk
Rate of return risk Commodity risk FX rate risk Equity risk
Murãbaha Murãbaha Murãbaha Mudãrabah
Ijãrah Salam Ijãrah Mushãrakah
Salam Istisnã Salam
Istisnã Istisnã
Mudãrabah
Mushãrakah
Figure 5.1 The four types of market risks that exist in the different Islamic
financial products
considering the risk initiated from the fluctuation of the market, in the
construction of Islamic contracts, financial institutions are using an alterna-
tive rate called rate of return, which is based on different benchmarks.
In the Murãbaha financial contracts, the considered rate of return should
be aligned with the repayment instalments that include the price of the com-
modity together with the institutions’ profit. In the case that the rate of
return, which is fixed in the Murãbaha contract, appears to be different from
the actual market rate, then there may be a loss in the benefits that should be
earned from this contract. Moreover, in the Ijãrah leasing financial contract,
the rate of return payments defined by the institution’s benchmarks should
also correspond to the actual market prices. A failure in this case will cause
the institution loss in opportunity that could be gained out of the Ijãrah leas-
ing. Finally, by providing Salam and Istisnã contracts, financial institutions
should define the price of the commodity in respect to the future delivery
date, based on the estimated benchmark rates. In case the actual market
price differs from the rate of return value, the institution may face difficul-
ties to resell and reinvest the delivered commodity and receive the expected
profit from the Salam and Istisnã agreements.
Islamic financial contracts that have fixed income assets driven by rate of
returns cannot be adjusted in regards to changes in the benchmark market
rates. As a result, financial institutions that provide such contracts are facing
risks arising from the movements in the market interest and inflation rates.
Note that financial institutions are using benchmark rates, for the absence of
any Islamic index, based in most cases on the London Inter-Bank Offered
Rate (LIBOR). This is a widely used benchmark for short-term interest
rates. Such index is used to align the market risk associated to Islamic
MARKET RISKS IN ISLAMIC FINANCE 151
financial product with the movements in LIBOR rates. Nowadays, in all
markets, new benchmark rates are also defined mainly based on polls of
indexes. On the other hand, different types of market yield curves are
mainly used to model the predictive market rates.
Market yield curves
The market yield curve provides information about the market future
expectations illustrated on a graphical representation of the yields for a
range terms to maturity. For instance, yields for maturity from one day
(overnight) to thirty years terms can be illustrated by a yield curve. The
yield curves are considered to be a predictor of future economic activities
and may provide signals of pending changes in economic fundamentals.
However, there is no single universal yield curve that maps and describes
the future economic changes.
Types of yield curves
In Islamic financial products the yield curve attempts to represent the future
behaviour of the markets for the performing contracts. As the markets are
reacting differently to economic situations, there are many yield curves for
representing them within the future set time periods. There are mainly four
types of yield curves that can be used to define such market behaviours and
are driving the market parameters of the Islamic financial contracts:
The Normal yield curve, where the curve rises as the maturity
lengthens, is increasing positively. This curve reflects an expectation for the
economy to grow smoothly in the future. Note, however, that the degree and
the rate of the growth vary according to the market conditions, driven by a
great number of internal and external parameters.
The Steep yield curve, where the curve rises steeply, represents the
behaviour of an economy that is expected to improve quickly in the future.
Such types of curves may be applied at the beginning of an economic
expansion.
The Flat and Humped yield curve, where the former one maps an
uncertainty in the economy, e.g. high volatility with constant mean; more-
over, in the latter curve, the short-term and long-term yields are equal and
the mid-term yield is higher.
The Inverted yield curve that occurs when long-term yields fall
below short-term yields. This curve maps rather abnormal and bad
economic situations in the future. Inverted yield curves imply that the mar-
ket believes inflation will remain low due to situations like recession in the
economic or economic sector. In general, an inverted curve indicates a
worsening economic situation in the future.
152 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Careful analysis on the design of market yield curves will provide fair
contracts from both the bank and lender, where the former will earn enough
profit; whereas the latter will benefit from those contracts providing addi-
tional benefits to the market. Yield curves can be applied to define the rate
of returns that should be considered when the Islamic financial products are
offered on the market.
Applying yield curves
One of the very important factors in determining a yield curve is the
currency in which it is denominated. Moreover, the economic situation of
the countries and companies using each currency is a primary factor that
should be considered in determining the yield curve. Thus, financial institu-
tions and investors that are dealing with Islamic financial contracts and
agreements may combine different types of yield curves by considering the
expected economic and market conditions. For instance, in the recession
times the Islamic financial contracts should map their parameters related to
market risks by using the Inverted yield curve. However, the market prices
for constructed assets such as buildings that are being invested in, by the
Istisnã financial contracts, are rising rapidly on such extreme market
conditions and thus should be mapped by the Steep yield curve.
The direction and the shape of the yield curve is widely analysed and
monitored by the market participants. Yield curves, i.e. in Murãbaha and
Ijãrah Islamic financial contracts, are usually upward sloping asymptoti-
cally with a positive slope, as lenders/investors demand, form longer
lending terms, higher rates from borrowers; the longer the maturity, the
higher the yield, with diminishing marginal growth. There are few explana-
tions for this phenomenon: it may be that the market is anticipating a rise in
the risk-free rate; if investors hold off investing now, they may receive a
better rate in the future, since the chance of default from the borrower
increases in terms of maturity, so the lenders demand to be compensated
accordingly.
The yield curves are affected by the expected inflation rate. In contracts
like Murãbaha, the expected inflation rate is at least considered in the profit
added to the repayments. Therefore, in the event that the future expected
inflation rate tends to be higher, investors demand greater premiums for
longer terms to compensate for this uncertainty. Therefore, the longer the
term, the higher the profit resulting from the upward-sloping yield curve.
However, note that nowadays the demand for shorter-term contracts is
increasing substantially; as a result, the long-term yields fall below short-
term yields. This is actually an inversion of the yield curve and slopes
downward. As a result of this abnormal and contradictory situation, in the
MARKET RISKS IN ISLAMIC FINANCE 153
long term the bank may settle for lower yields if, however, the bank has
strong indications that the economy will slow or even decline in the future.
The high cost of short-term contracts diverts from gains that would
otherwise be obtained through investments and business expansions. This
makes the economy vulnerable and slows it down, moving towards a
recession status.
Valuation based on rate of return risks
The valuation formula of the Islamic financial products that are based on
rate of returns and benchmarks is expressed as follows:
n
CF(t)
P 兺 (5.1)
t1
(
1m
y
)
t
where P is the price and/or profits of the financial instruments such as
Murãbaha, Ijãrah, CF(t) is the cash flow at time point t, m is instalment
payments (times) per year, y is yield to maturity, n is total number of
periods.
Keeping all other components of the above valuation constant, the level
of cash flows affects the agreement for the price of the contract according to
a non-linear positive relationship. On the other hand, the higher the yield to
maturity is, the lower the resulting price or profit and vice versa.
The above Islamic financial products exhibit a convex relationship
between yields mentioned above and the price of the commodities or asset
for large changes in market interest rates. However, this relationship can-
not be fully captured by the measured duration. The change in the price
and profits of the above-mentioned instruments in a percentage value,
which is driven directly or indirectly by the rate of returns and bench-
marks – or, in other words, by the market interest rate changes – can be
modelled and valuate-based on the Taylor expansion series as defined in
equation (5.2):
dP 1 dP dy 1 1 d 2P (dy)2 1 error
P P dy 2 P dy2 P
where P is the price or profits of the Islamic contracts, and y is the yield to
maturity.
Convexity introduces a non-linear element (optionally) in the market price
and profit valuation of the above-mentioned contracts. Mathematically
expressed, it yields the second derivative of the contract’s price or profit in
154 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
respect to interest rate changes. Analytically, it is given by:
t(t 1)
m CF(t)
n
d2P
兺 (5.3)
dy2 t1
( 1m )
y t2
where C(t) is the time period at time point t, m is the number of profit
payments that is set per year.
The impact of duration and convexity is given by the first and the second
term of equation (5.3), respectively. The effects from theta, rho and lambda
(vega) can also be incorporated in the error term. In the case that all ‘Greek’
effects are included, the VaR is based on the full-valuation method. It is
clear that a higher degree of non-linearity indicates a higher degree of risk
(see Group of Thirty,1 1993). However, estimation of gamma is a very
difficult task to carry out analytically. Thus, it is preferable to employ non-
parametric or parametric models.
COMMODITY RISK IN ISLAMIC FINANCE
Commodities are physical assets with unique attributes. Commodity risk is
the risk arising from movements in commodity prices. The fluctuation of
the commodity prices is a significant source of market risk. In Islamic
finance, contracts that deal with purchasing commodities and/or their
production are exposed to commodity price risk.
Commodity risk occurs when there is potential for changes in the price of
a commodity that is planned to be purchased or sold, such as in Murãbaha
Islamic financial contracts. In Murãbaha contracts, a seller agrees with
the purchaser to provide a specific commodity. The bank is financing the
contract on a certain profit added to the initial commodity’s price. The
difference between the agreed and the future market price of the commodity
is the actual exposure of the corresponding risk that banks take. Furthermore,
when the banks are providing the Salam and the Istisnã Islamic financial
contracts, they are also exposed to commodity risk. When the Salam contract
is applied, the financial institution (who is the buyer) makes advance pay-
ments on a negotiated price for a commodity. However, the delivery of the
commodity is at a specific time in the future, where its price may differ from
the set one. Similarly, when the Istisnã contract is used, the price of the com-
modity is paid in advance at the time of the contract and the commodity of
sale is manufactured and delivered later. A detailed description on how and
when the commodity risk is exposed in Murãbaha, Salam, and Istisnã
Islamic financial contracts is presented in chapter 2.
MARKET RISKS IN ISLAMIC FINANCE 155
Influence
rate risk
Future
Quantity
delivery
risk
risk
Commodity
risk
Invert Price
price risk risk
FX Cost
rate risk risk
Figure 5.2 Interaction and relation between commodity risk in Islamic finance
with other commodity risk factors
There are several drivers that influence commodity risks, which are
needed to be identified for the different types of contracts. Islamic financial
contracts that are exposed to such risks are similar to forward contracts and
they have to be hedged accordingly.
Drivers of the commodity risk
The behaviour of commodities in Islamic finance are driven both directly
and indirectly by many different risk factors, including price risk, cost risk,
market’s influence rate risk, FX rate risk, quantity and time risk, future
delivery risk, and invert price risk, as described below. Figure 5.2 illustrates
the interaction and relation between commodity risks with other risk
factors.
Financial institutions should consider the drivers of the commodity risk
referring to Islamic financial contracts:
Price risk
Commodity price risk affects consumers and end-users such as manufac-
turers, governments, processors, and wholesalers. When a commodity
changes, the cost of its purchase also changes; modifying, however, the
profit, such as in Salam and Istisnã contracts. Price risk also affects
commodity producers, as their variation affects directly the production
and the business income. Thus, since the prices of the commodity influ-
ence the production and business revenues, they also influence directly
156 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
the decision making about them. If the price goes below a threshold, pro-
ducers may reduce existing production or limit new production. On the
other hand, if the price of the commodity rises, the associated production
may become attractive and increase in its level. As a result, the decision
making concerning the production affects the supply and demand, which
in turn is affected by the changes of the commodity price. The factor of
supplying the commodities is a function of production and product deliv-
ery. Finally, the value of commodities is also affected by attributes such
as physical quality and location. In general, there are several factors that
affect the price of the commodities, including:
– The expected level of inflation
– The exchange rates
– The general economic conditions
– The cost of production
– The ability to deliver on time
– The availability of substitutes
– The weather conditions, especially for agricultural commodities and
energy
– The political stability.
Price risk is generally the greatest risk affecting the livelihood of com-
modity producers and should be managed accordingly.
Cost risk
The cost of manufacturing the commodity is an additional input risk
that should also be considered in Islamic contracts. For instance, the
manufacture of the commodities or the construction of assets referring
to Istisnã contract are directly or indirectly affected by the cost of raw
materials and other production costs. Thus, if the producer of the
commodity is unable to cover such manufacturing costs, a default on
delivery may arise with downstream consequences on other risks, e.g.
credit and operational. Financial institutions that agree on such
contracts must consider and evaluate the risk of cost by effectively
defining the price risks of the surrounding products and services needed
to produce the commodity.
MARKET RISKS IN ISLAMIC FINANCE 157
Market’s influence rate risk
Market’s influence rate is directly influenced by the commodity price
where they need to be harmonized in proportion to this rate. Thus,
changes in the influence rate must be adequately considered in the
Murãbaha financial contracts. This is because one of the main parame-
ters for estimating the profit that financial institutions are making out of
this contract is the price of the commodity at the maturity date of the
repayments. Any underestimation of this price is exposing the financial
institution to commodity price risk. Furthermore, commodities that are
purchased in advance, such as in the case of the Salam and the Istisnã
Islamic financial contracts, expose the bank to market influence rate
risks that correspond to the commodity price fluctuations.
FX rate risk
When the commodity price, set, in Islamic contracts by the financial
institution (e.g. a buyer) and the other party (e.g. a seller), on the
domestic currency and the commodity is traded in another currency, the
exchange rate will be a component of the total price of the commodity.
Therefore, financial institutions are directly or indirectly exposed to
commodity risks that are initiated from their trading in multi-currency
FX rates. Thus, in the Istisnã and Salam contracts, the bank may agree
to purchase a commodity from the domestic market and make an agree-
ment on selling it to a foreign market; in this case, the trade uses more
than one currency and thus is exposed to FX rate risk.
Quantity and time risk (through the demand for purchasing the
commodity)
Quantity and time risk is related to the capability of the manufacturer to
produce the planned amount of commodities at a specific period. Istisnã
contracts may expose the bank to a high degree to this type of risks. On
the other hand, the market demand for the commodity on different time
periods may influence the quantity that should be produced. For
instance, if a manufacturer or a farmer expects a high demand for a
commodity and thus plans the season accordingly, there is a risk that the
demand from the market is less than has been produced or vice versa.
The commodity’s market demand could be dependent on uncontrollable
factors. However, both the manufacturer or farmer and the financial
institution will suffer losses by being unable to sell the expected
amounts of products.
Future delivery risk
In the Salam and Istisnã financial contracts, there is a delivery of a com-
modity at a future time. In Salam agreements, the financial institution is
158 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
expecting to receive the commodity at a set future time. The sale in the
Salam contract should be less expensive than a cash sale. The Salam
contract agreement allows the purchaser to lock in a price, thus protect-
ing the purchaser from price fluctuation. However, the price of the
commodity at the delivery time may be different from the market price
and thus the financial institution is exposed to commodity price risk.
Moreover, in the Istisnã sale, the price is paid in advance at the time
of the contract and the object of sale is manufactured and delivered
later. However, any failure from the seller, who has already received the
advanced payment, to deliver the commodity at the set time may cause
commodity risk with significant losses. Delay of the delivery may result
in additional charges and in general commodity price fluctuation.
Invert price risk
There are occasions when the market is following specific commodity
price structures. This may appear when the demand for cash, by the
seller, or near-term delivery, by the purchaser, of a commodity exceeds
supply, or even when there are supply problems. Then, an invert or
backwardation market may result. Therefore, the market influences
directly the price of the commodities for immediate available supplies,
resulting in prices for near-term delivery to rise above prices for longer-
term deliveries. Such types of commodity risks affect both the Salam
and the Istisnã contract agreements. The period where the backwarda-
tion market is applied affects the deals of the above contracts. When the
market moves from backwardation to the normal price structure, both
the seller and purchaser may suffer significant losses.
Table 5.1 illustrates a pattern of the factors that initiate commodity price
risk in the Murãbaha, Salam, and Istisnã Islamic financial contracts.
In Islamic finance, the financial contracts that carry commodity risks,
such as Murãbaha, Salam, and Istisnã, differ from the other contracts in
Table 5.1 Factors that initiate commodity risk in Islamic financial contracts
Factors initiating commodity risks
Quantity Future Invert
Types of Influence FX rate and time delivery price
contracts Price risk Cost risk rate risk risk risk risk risk
Murãbaha ✓
Salam ✓ ✓ ✓ ✓ ✓
Istisnã ✓ ✓ ✓ ✓ ✓ ✓ ✓
MARKET RISKS IN ISLAMIC FINANCE 159
several significant ways, primarily due to the fact that most of them have the
potential to involve physical commodity deliveries. Note that commodities
involve issues that affect their prices and trading activities, such as quality,
transportation, delivery location, shortages, spooling, and storability.
Moreover, market demand and the availability of the commodity’s substitutes
could be important factors in the emergence of commodity risks. Substitutes
may become attractive when the commodity price fluctuates, becomes
expensive or, more importantly, when there are delivery difficulties. The
effect of commodity may not only be temporary but in many cases may
become permanent.
By providing Islamic products that are related to commodities, i.e.
Murãbaha, Salam, and Istisnã, financial institutions need to have a risk
tolerance of managing their potential commodity risks. Institutions and
scholars are required to understand the actual nature of the commodity risk
that they are exposed to as well as the available products that could be used
to support in developing risk-management strategies. Scholars and financial
risk managers should also consider how to manage commodity risk through
good decision- and strategy-making.
F X R AT E R I S K S
FX rate risk is a form of risk that arises from the change in price of one
currency against another. Most of the Islamic financial contracts that are
provided by financial institutions may be exposed to foreign exchange
fluctuations arising from general FX spot rate changes in cross-border
transactions, operations, and the resultant foreign currency receivables and
payables.2
FX rate risk may arise when the financial institution buys the
commodities in the Murãbaha contract using foreign currency. Moreover,
in the agreement of a Ijãrah contract, the deal of leasing between the
financial institutions (lessees) and the lessor may involve commodity or
service from foreign markets that requires trading on FX currencies and
therefore financial institutions are exposed to FX rate risk. Moreover,
institutions are also exposed to such risks when the commodities in Bai-
Salam contracts are related to the foreign currency, either before the
delivery date or at the time of reselling them. Additionally, Istisnã
financial agreements concerning construction of assets using capital
investment or operating in foreign regions may also cause such exposure.
Finally, in the Mushãrakah and Mudãrabah financial contracts of partner-
ship, when the investment of the financial institution refers to commodities,
assets, or business operations across the national borders, they may face
FX rate risk too.
160 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
EQUITY PRICE RISKS
Equity price risk in regards to the Islamic financial contract agreements is
the risk that the financial institution is exposed to by its investments’ depre-
ciation as a result of business or market dynamics that are causing losses.
The Mushãrakah and Mudãrabah Islamic financial contracts may result in
equity price risks. This is mainly because one of the main characteristics of
both the joint venture Mushãrakah Islamic financial contracts and
Mudãrabah Islamic business venture agreements is the sharing, between the
financial institution and the partner, of profit and loss that is driven by the
share in the investment’s equity.
Based on the main definition of the permanent Mushãrakah partnership
contracts, the participation of both parties in the investment’s profit and loss
is corresponding to the share of its equity stake. Any changes of the equity
price may modify the balance of the equities between the two parties. Such
modifications may cause either losses or additional gains to the financial
institution according to its equity participation and the business profit and
loss balance sheet. Moreover, any major losses and defaults on the expected
profits may result in selling the equities at the market price. This price is
usually less than the nominal one. In this case, the financial institution is
exposed to a major equity price risk.
When the partnership between the financial institution and the partner is
based on diminishing Mushãrakah contracts, there is an agreement for the
partner to buy all equity shares on a pre-defined fixed price. Any mismatch
between the fixed equity prices with the market (potential) prices is expos-
ing the institution to equity price risk. Moreover, any inability from the
partnership side to buy the equity or carrying on with the business may
influence the equity price.
In the Mudãrabah contract, any inability of business continuation may
result in a last equity payment and force the equity price to a level lower
than the nominal one and thus exposing the financial institution to equity
price risks.
VA L U AT I O N I S S U E S O N E Q U I T Y P R I C E S
A N D F X R AT E S
The valuation models for financial products referring to equity prices and
FX rates has a high degree of complexity due to the lack of being able to fol-
low the prices in the marketplace, and so they do not exhibit gaps in their
returns. Moreover, based on valuation models, not all inputs are known or
cannot be somehow estimated, such as the dividends for equities. The main
drawback of equity models is that it is difficult to quantify the level of
MARKET RISKS IN ISLAMIC FINANCE 161
dividends on a perpetual basis, even on a profit and loss sharing model
applied for Islamic finance. There are also several micro-economic and
macro-economic factors that cannot be easily modelled and drive the prices
of FX rates. Furthermore, both risk factors, namely equities and FX rates,
are liquid-traded financial products and their distributions of returns do not
exhibit gaps. Thus, instead of using a model for their valuation, it is prefer-
able for the risk manager to use their prices directly from the market.
Q U A N T I F I C AT I O N O F F O R E I G N
EXCHANGE RISK, EQUITY RISK,
AND COMMODITY RISK
The quantification analysis and valuation of rate of return risk and bench-
marks must consider a high number of market factors that are modelled by
future yield curves. These curves (or factors) have high complexity in their
identification as well as a high degree of interrelation. On the other hand, for
FX risks, every percentage of change in FX rates affects the price of the posi-
tion with the same percentage. Thus, their evaluation is rather simply applied
for generic cases where most of the Islamic financial contracts are classified.
The equity price may be fixed in Diminishing Mushãrakah contracts; how-
ever, any last equity price and exit of the partnership is based on the market
price of the equity. Equities appear to have quite smooth distributions of
returns and they can efficiently substitute for any valuation models.
Nevertheless, the image is different when market indexes are used, instead of
the data that exactly represent the movement of the equity under examination.
In this case, market risk exists by holding the equity, which is not represented
by the movement of the index. This risk is known as specific risk of the equity.
For all Islamic financial contracts, the quantification of the commodity
risk which is arising from movements of the commodity’s prices should be
based on these prices. Note that the behaviour of the commodity price risk
can be very similar to that of equities, especially in the case of specific risks.
For instance, the use of an energy index cannot fully explain the movement
of oil because there are specific factors that drive the price of oil. Similarly,
the analysis of the specific risk for equities coincides with the respective
analysis for commodities.
D ATA R E F E R R I N G T O M A R K E T
R I S K FA C T O R S
Islamic financial products carry more than one market risk factor; more-
over, even a simple portfolio usually contains different types of contracts.
162 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
This combination of risk factors and contracts increases the complexity of
collecting and combining the information needed for market risk analysis.
The different risk factors referring to commodity prices, equity prices, and
FX rates may replicate movements in the cash products. Most of the move-
ments can be easily picked up from organised exchanges; however, several
risk factors cannot be directly observed in the market place and should be
calculated by using other sources of information (market interest and
benchmark rates). In cash products, such as in the Murãbaha contracts, this
difficulty may arise from the fact that the yield to maturity is driven by a
variable function where the payments that produce the cash flows follow
different time points from those implied by market benchmarks.
Market data such as prices referring to Islamic products and agreements
are exchanged directly or indirectly at regular points in time. In the Islamic
financial contracts, the different points in time that are mainly considered
for trading market data used during market risk analysis are the ones that
refer to the contracts’ agreement date and the delivery dates. These prices
correspond to trading actions such as buy, sell, rent, etc. Moreover, all
market data underlying the yield curves and benchmarks are mainly used
for the rate of return risk analysis in the Murãbaha, Ijãrah, Salam, and
Istisnã financial contracts. In the agreements of Salam, Istisnã, and par-
tially for the Murãbaha financial contracts, the prices and the cost of the
commodities or services, the influence rate and the FX rate, the quantity
and time, and the invert process are the main market data that should be
considered for their market risk analysis. In the financial contracts based
on profit and loss, notably the Mushãrakah and Mudãrabah contracts, the
equity prices and their possible trading should be extracted for their market
risk analysis. Finally, the data corresponding to the currencies and FX rates
are used for analysing the associated risk for all types of Islamic financial
contracts.
In general, the more complex the synthesis of the portfolio or balance
sheet accounts constructed by Islamic products, the higher the number of
risk factors involved and the higher the amount of data needed from the
associated databases. It is important to highlight that the trading processes
during the lifetime of Islamic financial products is less dynamic than
conventional financial products. The frequency of data collection referring
to the above-mentioned time points is less than in conventional financial
products and thus the collection of information data, from the database
systems, is rather a straightforward process. In the selection of market data,
financial institutions have multiple accesses to different systems and types
of databases. However, any inability to extract information data that refers
to market risk factors could make the market risk valuation and management
process weak without providing and adding any value to the management of
the bank’s portfolios or the accounts on its balance sheet.
MARKET RISKS IN ISLAMIC FINANCE 163
There are many problems arising during this data extracting process,
where some of them are: misplaced digits or a misplaced decimal point,
any repeated datum, and any missing datum or a datum that is equal to
zero that has been inputted by mistake. Financial institutions are mainly
using ETL (Extract, Transfer, and Load) systems that are transferring the
information from the core database/warehouse to the risk management
system. Such process must be efficiently done to map all financial con-
tracts within the associated accounts. Errors in information data may
result in an under- or over-estimation of the volatility and thus will affect
the measurement of the Value at Risk (VaR). There are different tech-
niques and methodologies that financial institutions could employ to min-
imise such risks and loss of information data. For instance, in order to fill
up the gap arising from a missing datum, interpolation techniques can
usually be applied.
SENSITIVITY IN MARKET RISK
Sensitivity is a significant factor that plays a critical role in the market risk
analysis. Sensitivity to market risk in Islamic products reflects the relationship
between the cause from a financial risk factor and its adverse impact to the
financial institution’s earnings from these contracts. The causes are initiated
from the volatility in risk factors arising in Islamic financial contracts, i.e.
rate of returns, foreign exchange rates, commodity prices, or equity prices,
and the impact usually refers to the changes in price or return of the
contracts.
In market risk analysis, the volatility of the risk factors is transformed
into market risk by either linear or non-linear means that is encompassed
into a unique measure, called market sensitivity. Thus, every change in risk
factors has an equal quantitative percentage effect on the total position
invested in the product (sensitivity of one). Note, however, that the rate of
return or interest rate risk measurement is a more complicated task. Its sen-
sitivity, in respect to market volatility, is a non-linear function and is usually
a difficult one in terms of the computational power needed.
Based on the above, the market risk measurement is established by the inte-
gration between the Sensitivity and Volatility as illustrated in equation (5.4):
Market risk = Sensitivity Volatility (5.4)
The degree of significance referring to the sensitivity factor becomes
higher when the characteristics of the contracts involved in the quantification
of market risk are non-linear.
164 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
M A R K E T R I S K VA L U AT I O N M O D E L S
As already mentioned, market risks referring to Islamic financial contracts
(or agreements) arise from the movements in the level or volatility of the
market prices underlying the products and services that refer to these con-
tracts. There are two types of analysis for valuating the market risks. The
first is called ‘static’, simply because no change in positions related to
financial contract is taken into account; however, the only element that
might vary is the market conditions. The second is called ‘dynamic’, which
is a forward-looking analysis based on defined market price scenarios, busi-
ness strategies, and customer behaviours. The market positions are moving
over time and the prices are always influenced by the customers’ behaviour.
This behaviour should be considered in the future business strategies that
should be related to market prices. In other words, market prices are driven
by the rules of the market and such rules have to be considered in financial
risk analysis. Moreover, the business strategies need to be defined accord-
ing to the customer behaviour and future market price conditions. Market
changes arising over a certain time frame and thus effective changes over
time should be taken into account. In order to do that, assumptions on future
cash flows and their reinvestment need to be made. Dynamic analysis is a
strong tool for planning the future business in regards to what and how the
bank should provide its Islamic financial contracts so that they can be
profitable to the institution and beneficial to its clients. Both static and
dynamic simulation plays a key role in risk management and institutions
must be able to perform both types of analysis. One of the main elements of
such evaluation analysis is the VaR that is applied in both types of analysis,
used as a measurement indicator for the market risk referring to Islamic
financial contracts.
VaR MODELS FOR ISLAMIC FINANCIAL
CONTRACTS
Value-at-Risk is one of the most well-known methodologies to quantify and
valuate market risk in a systematic fashion. Based on Islamic financial con-
tracts that contain market risks, financial institutions attempt to measure
losses on a constant portfolio over a specified period. This period for con-
ventional finance contracts usually refers to the next day or next ten days;
whereas, for the Islamic contracts, this may be different (usually it is longer
as the Islamic financial products are less dynamic in terms of the market
risk factors; in addition, due to the luck of available market risk loss data).
The measurements of losses are calculated with regards to whether this time
limit is exceeded on a given fraction of time (typically, a probability level of
MARKET RISKS IN ISLAMIC FINANCE 165
1 percent). Except for the simplifying assumption of constant portfolio syn-
thesis during the time horizon, VaR models possess some latent, but equally
crucial, weaknesses, arising from the fact that they are tailor-made models.
Because of this, the data inputs should be carefully assessed before the
appropriate model is applied.
In VaR model analysis, both the position data and the market data are
used. Figure 5.3 shows the process flow of market and position data for
implementing the VaR approach that finally results in the risk evaluation
analysis reports. Note that the definition and the nature of the data are the
key factors that drive the decision about which model is going to be used.
Moreover, the position risk and risk factors in the construction of the
Islamic contract-based portfolios should be identified.
Having both position and market data, every VaR mechanism, implicitly
or explicitly, is driven by the selection or estimation of the components
illustrated in Figure 5.4 and includes the assumptions for the data distribu-
tion, the window length of the data used for parameter estimates, the confi-
dence level, the holding period that defines the time horizon for holding the
investment, and the individual volatilities and co-movements between or
among risk factors defined by using return time series. Note that in a case
where the frequency of the data used is very high, VaR models assume zero
mean return; however, the inclusion of the mean of returns in lower-
frequency data is suggested when VaR measures are estimated.
Market risk factors Portfolio exposure
commodities, FX, equity, Positions of the institution
benchmark, yield curve (Islamic financial products)
movements
Value at risk
mechanisms
Risk evaluation
reports
Figure 5.3 Flow chart referring to the Value-at-Risk approach
166 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Distribution
assumptions
Volatilities and Window
co-movements length
Value at risk
Holding Confidence
period level
Figure 5.4 Main components considered in VaR models
P O S I T I O N A N D M A R K E T D ATA
For the implementation of the VaR model, there are two classes of data that
should be carefully used and assessed: position data and market data. It is
impossible to estimate the VaR without position data and the necessary mar-
ket data. The former type of data (position) is defined within the systems
that map the performance of contracts (or underlining contracts) in the insti-
tution’s accounts or portfolio; whereas the source of the latter type of data
(market) is usually done by using either interpolation techniques (described
briefly in the following paragraphs) or extensive simulation methods as dis-
cussed later in this chapter.
Interpolation techniques
Interpolation is a technique for constructing new data points based on a dis-
crete set of known data points. There are different interpolation techniques,
including linear interpolation, the exponential and cubic spine interpolation
that can be applied according to the data availability, and the level of
approximation is needed as described in the following sections.3
Linear interpolation
There are some cases where the information data of the Islamic financial
product, such as prices and rates, are not observed directly in the market
MARKET RISKS IN ISLAMIC FINANCE 167
place or are partially missing. Therefore, it is essential to find methods for
constructing those values from the most relative values observed. The sim-
plest approximation of the estimation of the aforementioned information
data is the linear interpolation. This approximation is based on the data
points that are located in between the actual data which are assumed to be
on a straight line. Thus, the estimation of a rate r* for a Sukûk bond corre-
sponding to time distance is located between the payment rates rn⫹1 and rn
that are corresponding to time distances n⫹1 and n. The rate r* is therefore
estimated by the following equation (5.5):
rn1rn
r* rn ( n) (5.5)
n1 n
Note that the time distance belongs to [ n⫹1 and n].
The advantage of the linear interpolation is its simplicity and the fact
that, in many cases, it can produce good approximations of reality. On the
other hand, if the yield curve is seen as a whole, the method produces sharp
changes between time points under consideration. Using polynomial inter-
polation where it is assumed that a yield curve exhibits smoother alterations
in its shape can be used to overcome the above-mentioned drawback.
Exponential interpolation
Another natural candidate for producing missing data is the exponential
interpolation. This approximation is used whenever the shape of the yield
curve is considered to be of a similar shape. The exponential interpolation is
expressed as in the following equation (5.6).
r* rne(ln((rn1rn) (n1n)))*n e(ln (rn1 (n1 n)))*n (5.6)
Note that this approximation is largely and successfully applied with the
factor that estimates the profit in the Islamic products. Thus, this technique
is suitable for interpolating between discount factors corresponding to dif-
ferent periods in time that corresponds to the contract negotiations such as
in Salam and Istisnã, where there is market price consideration on delivery
date and the future selling date.
Cubic spline interpolation
One of the most widely used techniques in financial engineering is cubic
spline interpolation.4 It is the most suitable technique for modelling yield
curves, zero curves, and forward curves. Note, however, that it may produce
biased results if some data are more concentrated in relation to some others.
168 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
POSITION RISK AND EXPOSURE RISK
Nowadays, various financial institutions offer a wide spectrum of Islamic
financial products. For such products, the VaR estimation is based on infra-
structure that combines the market data and position data in order to
achieve continuous portfolio valuation. Note that position risk or exposure
risk is the risk attributed to a specific position due to the exposure in one or
more risk factors. In the VaR analysis, there is a strong assumption that the
synthesis of the portfolio, i.e. the one that contains, for instance, Islamic
financial contracts, remains unchanged throughout the holding period.
Thus, in the construction of the portfolio, risk managers must be aware that,
during their VaR analysis, the portfolio should remain steady.
Under the above assumption, the set of market risk factors remains the
only source of risk. Thus, the market factors that drive the value of com-
modities in the Murãbaha, Salam, and Istisnã Islamic financial contracts
are commodity prices. The main factors that drive the FX in all Islamic
financial products are the FX rates. Similarly, the factors that drive the
equities in the Mushãrakah and Mudãrabah agreements are the equity
prices. Moreover, the factors that drive the rate of returns in the Murãbaha,
Ijãrah, Salam, and Istisnã financial contracts are the yield curves. Finally,
the factors that drive the profits in the Sukûk bonds are the ones that are
related to the risk factors of the underlying contracts as well as their
corresponding volatility of the underlying asset prices.
E VA L U AT I O N M E T H O D S O F M A R K E T R I S K
The evaluation of a portfolio is based on the same approaches invented dur-
ing the 90s where a professional team from JP Morgan Bank started to
engage themselves with the bank’s ‘morning report’. This report was pro-
duced early in the morning by the risk management department and was
produced to inform the top management about the perceived market risk for
the next day’s positions. The market risk was translated in terms of how
much money the bank had the possibility of losing by assuming that it
retained the previous day’s portfolio. In order to evaluate the overall market
risk that was assumed to be undertaken by the bank, risk professionals
employed by JP Morgan developed three alternative quantitative methods:
The Variance–Co-variance (VC) – parametric – approach
The Monte Carlo (MC) approach
The Historical-Simulation (HS) approach.
MARKET RISKS IN ISLAMIC FINANCE 169
VaR
Figure 5.5 Normal distribution of the changes caused by the value of a market
analysis portfolio
These three approaches can be used to quantify not only the Westernized
type but also the Islamic financial products by deriving the distribution of the
changes caused in the value of a portfolio at the end of the holding period. In
most cases, in market risk analysis this distribution appears to be nearly sym-
metric and is often approximated as normal (see Figure 5.5), which may
allow for analytical solutions to be developed. The distribution may also be
estimated using Historical-Simulation returns. Finally, a Monte Carlo simu-
lation can be used to create a distribution based on the assumption of certain
stochastic processes for the underlying variables. The data that are consid-
ered in this distribution and are used in VaR models are the ones within the
boundaries set by the confidence level as illustrated in Figure 5.5.
As will be explained later, the first and the second methods (VC and MC)
are based on parametric estimation, while the third one (HC) is based on
non-parametric estimation. Moreover, the number of different product types
or, in other words, the number of different data types, has a direct effect on
the difficulty of VaR estimation; that is to say, the higher the number of
assets, the higher the time needed for VaR estimation under all three tradi-
tional approaches. More specifically, under the VC approach, an increase in
the number of positions taken has a direct exponential effect on the number
of calculations needed. Under the MC approach, a similar increase in the
number of positions causes more significant calculation difficulties.
Instead, the magnitude of the effect is lower under the HS approach,
although there is also an increased storage need.
VA R I A N C E – C O -VA R I A N C E M E T H O D
The variance–co-variance method is applied when there is the assumption
that the risk factors follow a multivariate normal distribution and that they
170 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
exhibit serial independence. Thus, this method is suitable for Mushãrakah
and Mudãrabah Islamic financial contracts where there are cash earnings/
profits. The method could also be applied to the Murãbaha and Ijãrah
contracts, where the market value may vary from the actual payment
instalments.
The key parameter for VaR estimation under the variance–co-variance
method is the measurement of volatility. Note that, in order for the volatil-
ity to be estimated, it is essential to determine the time series of the prices
or returns (profits) for each of the contracts that are under consideration, i.e.
commodity prices referring to Mushãrakah contracts, returns on payments
on the Ijãrah contracts, and returns referring to the profits or shares earned
by the Mushãrakah and Mudãrabah agreements. The above-mentioned
returns implicitly drive the risk factors and are presented by using either
logarithmic or arithmetic scales, although the former produce smoother
measures of volatility. The concept of volatility, according to the variance–co-
variance method, is approximated by the variance or standard deviation of
returns. The estimation of the standard deviation can be made by using
either equally-weighted historical data or unevenly-weighted historical data.
The difference of the latter in relation to the former technique lies in the fact
that each of the historical market data receives a different weight when the
volatility is estimated.
Equally-weighted historical volatility
The estimation of the volatility it of the risk factor i, which is based on
equally-weighted historical data of returns Ris in the time interval between
tT and t1, is estimated as in the following equation (5.7):
冪
t1
it 1
兺 (Risit)2
T1 stT
(5.7)
where
it 1 (Ri1 Ri2 … RiT) (5.8)
T
Note that T is the time window and t is the point in time. Nowadays, the
‘effective’ time window T is usually set by the regulatory agents, even
though the original concept was related to the average life of weights
assigned to the observations5 (see Jorion, 2002). According to the docu-
ments of the Bank for International Settlements6 (BIS), the ‘effective’ time
window for the conventional contracts should be no less than one year or, by
interpreting the statement, 250 observations (T = 250). However, in the
MARKET RISKS IN ISLAMIC FINANCE 171
estimation of the equally-weighted volatility ˆ it referring to Islamic finan-
cial contracts, to create an ‘effective’ time window more observations may
need to be taken into consideration, as these types of contracts are less
dynamic in measurements than the conventional ones. Financial analysts
usually exclude the mean from the volatility estimation formula as, in most
financial time series, the true mean is close to zero and any attempt to esti-
mate it will produce statistically insignificant results. Thus, the mean metric
can be excluded from the above estimation without significant changes in
the volatility estimation.
Weighted schemes
By implementing the weighted schemes7 approach, the estimation of the
volatility it of the risk factor i is based on the assumption that the current
volatility depends on the past squared residuals and on short-term and
long-term volatility. The following equation 0 defines the generic model of
the weighted schemes approach, widely known as the GARCH(s,s)
specification:
冪
t1 t1
it ␥VL 兺
stT
as (Risit)2 兺 
stT
s
2
is (5.9)
where,
t1 t1
␥ 兺a
stT
s 兺
stT
s 1 (5.10)
VL is the long-term variance, ␥ is the weight assigned to VL, ␣s and s are at
point in time s, the positive amount of weight given to residuals and volatil-
ity, respectively.
The GARCH(1,1) model defined in equation (5.11) has an economic
interpretation in the sense that it depends on the most recent financial
variability – shocks – by defining the squared residual and the prevailing
circumstances, by defining the most recent estimates of the variance rate.
This model calculates relatively good short-run predictions, incorporating
the impact of the long-run variance level as in equation (5.11).
ˆ t 兹␥VL a (Rt1)2 2t1 (5.11)
where ␣  ␥ 1.
Note, finally, that the GARCH(1,1) model recognises that, over time, the
variance tends to get pulled back to a long-run average level of variance;8 on
the other hand, it eliminates the serial correlation from the squared residuals.
172 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
An alternative way of weighted schemes is the exponentially-weighted
moving average model (EWMA), also known as integrated GARCH
(IGARCH), according to which ␣ 1,  , and ␥ 0. The volatil-
ity it estimated based on the exponentially-weighted moving average
model is defined as formulated in equation (5.12).
冪
t1
it (1) 兺
stT
ts1
(Ris)2 (5.12)
where is a parameter called ‘decay factor’ and ranges between zero and
one. This smoothing parameter places less weight on observations as the
time distance from the present becomes larger. Moreover, the smaller the
value of , the less weight is assigned to past observations and the higher
the ratio according to which the weights are lowered.
Variance–co-variance matrix and VaR
One of the main elements in constructing the variance–co-variance method
is the estimation of the variance–co-variance matrix. The elements of this
matrix classify the possible co-movements among the spectrum of the uni-
tary VaR estimates as they are addressed by the pair-wise correlations for-
mulated in equation (5.13):
2ijt
ijt (5.13)
it jt
t1
2ijt 1
兺 (Risit) (Rjsjt)
T1 stT
(5.14)
where ijt is the correlation between asset i and j at time t, and ijt is the
co-variance between asset i and j at time t. Note that the estimation of
co-variances according to the RiskMetrics approach, presented above, is
given by:
t1
2ijt (1) 兺
stT
(Risit)(Rjsjt)
ts1
(5.15)
When the portfolio includes more than one position, the computation of
VaR requires an estimation of the entire variance–co-variance, or correla-
tion matrix. Thus, the estimation of the portfolio’s VaR needs more compli-
cated approaches than simply adding the unitary VaRs. The assets in a
portfolio move in independent directions, mapped by the correlation matrix.
Based on this matrix together with the unitary VaRs, the portfolio’s VaR is
MARKET RISKS IN ISLAMIC FINANCE 173
estimated as defined in equation (5.16):
VaRP 兹xᠬTCxᠬ (5.16)
冤 冥
1,1 1,2 … 1,n
2,2 … 2,n
C 2,1 (5.17)
… … … …
n,1 n,2 … n,n
冤 冥
VaR1
VaR2
xᠬ (5.18)
…
VaRn
where the matrix C in equation (5.17) contains all the co-movements among
the volatilities as estimated in equation (5.13) and the elements in vector xᠬ
in equation (5.18) as defined by the unitary VaRs.
M O N T E C A R LO S I M U L AT I O N M E T H O D
The Monte Carlo simulation is applied when, on a portfolio, the risk factors
have a high degree of non-linearity. In this case, the probability distribution
of the changes in the portfolio value is rather unknown. Even by making
assumptions about the distribution of the risk factors, it is hard to deduce the
P&L distribution analytically. Based on different mathematical approaches,
the changes in risk factors (rate of returns, price volatility, shares, and FX
rates) can be simulated. For instance, implementing Geometric Brownian
Motion (GBM), the method produces numerous changes in the value of the
portfolio with the help of a random generator procedure.
The Monte Carlo simulation is by far the most powerful method to
compute VaR and is able to account for a wide range of exposures and risks,
including non-linear price risk, volatility risk, and even model risk. The
Monte Carlo method is a flexible enough method to incorporate time varia-
tion in volatility, fat tails, as well as extreme scenarios. Moreover, this
method is capable of performing simulations in regards to passage of time
which is used to create structural changes in the portfolio. On the other hand,
in Monte Carlo simulations, the main drawback is the computational time. If
the valuation of assets on the target data is also involved in the simulation,
the method requires an additional simulation, which implies an increased
computational time. As a result, this method is very ‘expensive’ to
174 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
implement in terms of system infrastructure and intellectual development.
Another weakness of this method is the ‘model risk’, meaning that it relies
on the model of stochastic processes which adds underlying risk factors.
Finally, the Monte Carlo method requires the use of the variance–co-
variance matrix, as well as having some kind of dependency on historical
data. Detail analysis on how to formulate the matrices and the steps to
estimate the VaR by applying Monte Carlo simulation can be found in
Kalyvas & Akkizidis.3
H I S T O R I C A L S I M U L AT I O N
The historical simulation method provides a straightforward implementa-
tion of full valuation. The underlying methodology consists of going back
in time and applying current weights to time series of historical asset returns
(equation (5.19)) involved in the computation of VaR. By doing this, a his-
tory of a hypothetical portfolio of Islamic contracts is reconstructed using
the current position.
N
R 兺w
i1
i,t Ri,k (5.19)
where the weights wt are kept at their current values and the returns recon-
struct the history of a hypothetical portfolio using the current position. The
assumption applying this method is that the present portfolio construction is
frozen at all points in time in the past. According to the historical simulation
method, m different scenarios concerning probable future price movements
can be applied. Those scenarios are taken from past experiences by utilising
all past multivariate returns included in the used time window. The full set
of hypothetical prices is used to define the new portfolio value and thus it
creates the portfolio distribution of returns which, together with a predeter-
mined confidence level, is used to estimate the VaR. In historical simulation
analysis, VaR is the worst loss observed, after ignoring n negative returns
that correspond to the confidence level.
The main characteristics of the historical simulation method are that it is
the most cost-effective and the least time-consuming approach in terms of
computational needs. Moreover, this method is relatively simple to implement
as long as there is availability of (high frequency) historical data represented
mainly in time series. Additionally, using historical simulation, the time hori-
zon defined in VaR estimation is rather an easy choice. Furthermore, this
method relies on actual measurement returns and therefore it considers non-
linearities and non-normal distributions. The full valuation is simply obtained
based on the actual historical data and not on any valuation models as applied,
for instance, in the Monte Carlo method. Finally, and more importantly, the
MARKET RISKS IN ISLAMIC FINANCE 175
historical simulation is capable of considering fat tails because it refers to the
true distribution of returns without assuming any specific distribution.
On the other hand, in the implementation of the historical simulation
method there is a strong assumption that there is sufficient history of price
(or profits) changes. This is in many cases very hard to obtain from the
Islamic financial contracts. For instance, to obtain the corresponding 1,000
simulations, it may require a few years of collecting continuous historical
data. Thus, such a method is probably the most costly in terms of storage
requirements. Note, however, that some Islamic financial contracts have
short lifetimes and histories or may not have the associated history that is
needed on record. Another strong assumption in this method is that the past
represents somehow the immediate future to an acceptable level. Moreover,
the method gives the same weight to all observations in the window,
including old data points. Finally, this method becomes rather difficult to
implement for large and complicated structured portfolios.
A short description of the advantages and drawbacks that financial
organisations may face by applying Historic, Monte Carlo, and Variance–
Co-variance Methods is presented in Table 5.2:
Table 5.2 Advantages and drawbacks by applying Historic, Monte Carlo, and
Variance–Co-variance Methods
Methodology Advantages Disadvantages
Variance – Easy to implement Requires variance–
Co-variance Computationally fast co-variance matrix
Easily amenable to analysis Dependent on
historical data
Can account for
non-linear risks (with
difficulty)
Monte Carlo Can account for complex Computationally complex
non-linear risk structure Requires variance–
co-variance matrix
Dependent on historical
data
Historic Easy to calculate Requires large database
Based on actual results Historical results may not
Non-parametric be appropriate
Can account for Lack of drill-down ability
non-linear risks
176 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
B A C K-T E S T I N G A N D S T R E S S -T E S T I N G F O R
MARKET RISK EXPOSURES
The different methodologies for applying the VaR models to portfolios that
contain Islamic financial products should be tested by the financial institutions
to evaluate their efficiency and applicability. The test referring to the
efficiency should be made by the institution on its specific portfolios. This
is due to the fact that financial institutions may possess different financial
products with different combinations and thus may follow different invest-
ment policies. Moreover, when structural changes are taking place in the
nature of the portfolios, financial institutions must also test the efficiency of
the VaR model by periodically performing back-tests. The back-test is
applied to check whether the VaR predictions correspond to observed
market changes. Moreover, institutions that provide Islamic financial prod-
ucts should also check their vulnerability in extreme circumstances that
conventional VaR measures fail to predict. As these extreme market events
may be catastrophic for the institution, their consideration and analysis have
significant degree. Therefore, financial institutions should conduct regular
stress-tests along with the results produced by VaR models. The following
section presents the theoretical and, occasionally, the practical aspects of
back-testing and stress-testing applied by the financial institutions.
Back-testing
In back-testing there are two perspectives that should be considered to
evaluate the efficiency of the models under consideration. The first one is
about testing if the criteria referring to the confidence levels are fully set;
the second one is about testing other models that are candidates for
introduction as internal models in the financial institution. Note that, in the
former perspective, the confidence level together with the ratio of the
number of times that the model underestimates the actual portfolio change
to the total number of predictions made are the two characteristics of the
model that should be used in order to evaluate its performance. In the event
that the above ratio exceeds the confidence level set by the VaR model, the
model is considered to be insufficient to represent reality. Moreover, when
the phenomenon appears on an occasional basis, the number of failures may
sporadically deviate from the confidence level mentioned above. Note that
the Basel Committee introduced a matrix that assigns the number of fail-
ures, or in other words ‘exceptions’, to a multiplication factor, as illustrated
in Table 5.3.
As shown in this table, the exceptions (or overshooting) are distributed
within three ‘traffic light’ zones, where the multiplication factor increases
accordingly. When the number of exceptions falls in the green and yellow
MARKET RISKS IN ISLAMIC FINANCE 177
Table 5.3 The three zone ‘traffic light’ for the
overshooting and the multiplier factor
Zone Overshooting Multiplier
Green 1 3.00
2 3.00
3 3.00
4 3.00
Yellow 5 3.40
6 3.50
7 3.65
8 3.75
9 3.85
Red 10 4.00
zones, this indicates that the internal model is appropriate and fully or par-
tially acceptable; whereas, when the number of exceptions falls in the red
zone, it indicates that the internal model is either inappropriate or needs
major corrections.
Stress-testing
Financial institutions are using ‘Stress-testing’ for assessing the event and
consequences of risks that may appear under extreme circumstances. It is
actually used to fill the gap in the concept of VaR methodology. Note that
the VaR is used to provide a probabilistic prediction on losses that are likely
to happen for a pre-specified holding period and confidence level. Thus, it
is difficult to ensure that, by using VaR, extreme cases are fully covered.
Moreover, as the VaR methodologies are based on predefined models, they
inherently carry a model error risk and thus their resulting evaluation of risk
may fail to a certain degree. Supervisory authorities have introduced stress-
testing as a complementary, but obligatory, test to VaR estimations, in order
to assess the vulnerability of financial institutions to exceptionally
unexpected, but plausible, usually non-probabilistic, financial events.
In modern risk management analysis, both financial institutions and
regulatory entities need to evaluate specific aspects of extreme risk events
arising from either individual business lines of the organisation or individual
178 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
position types of trading and/or banking portfolios. Thus, extreme com-
modity price risks that appear in the Murãbaha, Ijãrah, Salam, and Istisnã
financial products, as well as in the equity price for the Mushãrakah and
Mudãrabah agreements, need to be tested and evaluated by the institutions.
In general, there are two different types of ‘stress-testing’ that financial
institutions and supervisory regulators accept:
scenario-based
sensitivity-based.
Scenario-based tests assume that, for all categories of risk factors, their
changes happen simultaneously. This assumption reflects the consensus of
risk professionals that the aforementioned spectrum of events will happen
instantaneously in the near future. Scenarios can be chosen from either a
historical or hypothetical event or a combination of both.
The aforementioned prices of the Islamic products can be evaluated
based on the price shift and price volatility shift analysis. Such analysis is
driven by stress scenarios that are defined at the level of the basic risk factor
categories (rate of return and benchmark rates, FX rates, equity markets,
commodities) or the level of individual risk factors. Note that the term ‘price
volatility’ is used to describe the price fluctuations, i.e. of a commodity.
Thus, volatility is measured by the day-to-day percentage difference in the
price of the commodity. Despite the level of prices, the degree of variation
defines a volatile market. Since price is a function of supply and demand, it
follows that volatility is a result of the underlying supply and demand
characteristics of the market. Therefore, high levels of volatility reflect
extraordinary characteristics of supply and/or demand.
In the case of price shifts, the stress scenarios may be defined as additive
or multiplicative shifts. Moreover, they may be defined by means of volatil-
ity functions of the risk factors. The scope in the analysis of price shift and
price volatility shift is to estimate the value changes of Islamic financial
contracts based on stress scenarios, which may be defined for changes in
prices and volatilities. Moreover, the scope of such analysis is also to fulfil
the regulatory requirements with respect to the calculation of market risk
based on the stress scenario-matrix approach.
Financial institutions employ the sensitivity-based tests to evaluate the
impact of price changes on the value of a portfolio. Because financial insti-
tutions exhibit a variety of portfolios that could contain several Islamic
financial products, the tests are made for both directional changes; that is,
for negative and positive changes. In addition to the bidirectional changes in
market risk factors appearing in Islamic products, financial institutions are
also employing stress-tests that are referring to the links between the related
MARKET RISKS IN ISLAMIC FINANCE 179
positions of a portfolio. These positions are usually taken according to a
sensitivity factor and/or a correlation estimator. Thus, financial institutions
should be able to identify what kind of change in each sensitivity factor
and/or each correlation estimator would adversely change the value of
the portfolio; moreover, they should also set the exact test for individual
estimations.
It is somewhat unavoidable to distinguish between scenario-based tests
and sensitivity-based tests, in the sense that some scenario-based tests
implicitly assume some scenarios for sensitivity or correlation estimators.
The most apparent paradigm is the case of the simultaneous adverse change
on all market risk factors. In practice, this technique assumes a unique unity
correlation for all bilateral product co-movements.
SUMMARY
Islamic financial institutions are exposed to market risk primarily through
four types of risks, which are rate of return or benchmark rate risks related
to market inflation and interest rate, commodity price risks as they typically
carry inventory items (predefined prices), FX rate risks in the same way as
conventional banks and equity price risks, mainly in regards to the equity
financing through the profit- and loss-sharing contract modes. These types
of risks were discussed extensively in this chapter and are summarised
below. Yield curves are considered to be a predictor of future economic
activities and may provide signals of pending changes in economic funda-
mentals. There are mainly four types of yield curves that can be used to
define such market behaviours and drive the market parameters of the
Islamic financial contracts: the Normal yield curve, the Steep yield curve,
the Flat and Humped yield curve, the Inverted yield curve. Each one is
useful for different purposes and for different types of Islamic financial
contracts. The use of these curves is explained in this chapter.
The fluctuation of the commodity prices is a significant source of
market risk which creates commodity risks, arising from movements in
commodity prices. In Islamic finance, contracts that are dealing with
purchasing of commodities and/or their production are exposed to com-
modity price risk. The behaviour of commodities in Islamic finance are
driven both directly and indirectly by many different risk factors, including
price risk, cost risk, market’s influence rate risk, FX rate risk, quantity and
time risk, future delivery risk, and invert price risk. By providing Islamic
products that are related to commodities, i.e. Murãbaha, Salam, and
Istisnã, financial institutions need to be able to manage these potential
commodity risks. These commodity risks were explained extensively in the
course of this chapter.
180 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
One of the main characteristics of both the joint venture of the
Mushãrakah and Mudãrabah contracts is the sharing of profit and loss that
is driven by the share in the investment’s equity. This type of venture can
result in equity price risks. The quantification and management of equity
risk should be based on available market prices.
Islamic financial products carry more than one market risk factor; more-
over, even a simple portfolio usually contains different types of contracts.
This combination of risk factors and contracts increases the complexity of
collecting and combining the information needed for market risk analysis.
Market data underlying the yield curves and benchmarks are mainly used
for the rate of return risk analysis in the Murãbaha, Ijãrah, Salam, and
Istisnã financial contracts. In general, the more complex the synthesis of the
portfolio or balance sheet accounts constructed by Islamic products, the
higher the number of risk factors involved and the higher the amount of data
needed from the associated databases. In the selection of market data,
financial institutions have multiple access to different systems and types of
databases. Any inability to extract information data that refers to market risk
factors could make the market risk valuation and management process weak
without providing and adding any value to the management of the bank’s
portfolios or the accounts on its balance sheet.
In market risk analysis, the volatility of the risk factors is transformed
into market risk by either linear or non-linear means that is encompassed
into a unique measure, called market sensitivity. Sensitivity is a significant
factor that plays a critical role in market risk analysis. Sensitivity to market
risk in Islamic products reflects the relationship between the cause from a
financial risk factor and its adverse impact on the financial institution’s
earnings from these contracts.
The evaluation of market risk is translated in terms of how much money
the bank had the possibility of losing by assuming that it retained the previ-
ous day’s portfolio. In financial analysis, the most prominent techniques to
valuate the market risks are the value-at-risk (VaR). In order to evaluate the
overall VaR, three alternative quantitative methods have been developed,
which are the Variance–Co-variance (VC), the Historical-Simulation (HS),
and the Monte Carlo (MC) approaches. These three approaches can be used
to quantify the Islamic financial products by deriving the distribution of the
changes caused in the value of a portfolio at the end of the holding period.
For the implementation of the Value-at-Risk (VaR) model, there are two
classes of data that should be carefully used and assessed: position data and
market data. It is impossible to estimate the VaR without position data and
the necessary market data. The latter are defined by using interpolation
techniques, including linear interpolation, the exponential and cubic spine
interpolation that can be applied according to the data availability and the
level of approximation. Position risk or exposure risk is the risk attributed to
MARKET RISKS IN ISLAMIC FINANCE 181
a specific position due to the exposure in one or more risk factors. In
constructing the portfolio, risk managers must be aware that during their
VaR analysis the portfolio should remain steady.
The different methodologies for applying VaR models to portfolios that
contain Islamic financial products should be tested by the financial institu-
tions to evaluate their efficiency and applicability. There are two known
ways of conducting such tests. The back-test is applied to determine
whether the VaR predictions correspond to observed market changes;
whereas the stress-test is used for assessing the event and consequences of
risks that may appear under extreme circumstances. Supervisory authorities
have introduced stress-testing as a complementary, but obligatory, test to
VaR estimations, in order to assess the vulnerability of financial institutions
to exceptionally unexpected, but plausible, financial events.
Islamic financial institutions should have in place an appropriate
framework for market risk management (including reporting) in respect of
all assets held, including those that do not have a ready market and/or are
exposed to high price volatility. Adopting and applying the Basel II
principles is the way forward, as described in chapter 3.
NOTES
1. Group of Thirty, ‘Special Report on Global Derivatives – Derivatives: Practices & Principles’ (1993).
2. Islamic Financial Services Board, Guiding Principles Of Risk Management For Institutions Offering
Only Islamic Financial Services (December 2005).
3. L. Kalyvas, I. Akkizidis, I. Zourka, and V. Bouchereau, Integrating Market, Credit and Operational
Risk: A Complete Guide for Bankers and Risk Professionals (London, Risk Books, 2006).
4. G. A. Holton, Value at Risk: Theory and Practice (London: Academic Press, 2003).
5. P. Jorion, ‘Fallacies about the effects of market risk management systems’, Financial Stability Review
(2002), pp. 115–127.
6. www.bis.org
7. T. Bollerslev, ‘Generalized Autoregressive Conditional Heteroskedasticity’, Journal of Econometrics
(1986) 31, pp. 307–327.
8. J. Hull, Options, Futures and Other Derivatives (Prentice Hall, 2003).
CHAPTER 6
Operational Risk in
Islamic Finance
INTRODUCTION
According to the guidelines of the Islamic Financial Services Board
(IFSB),1 financial institutions are exposed to operational risks when losses
occur due to failures in their internal controls involving processes, people,
and systems. In addition, institutions should also incorporate possible
causes of losses resulting from Shariah non-compliance and the failure in
their fiduciary responsibilities. A special characteristic in applying Islamic
financial contracts is the strong engagement between the institution and the
counterparties. In addition, when applying partnership agreements (i.e. in
Mushãrakah and Mudãrabah), both sides may share profits and losses.
There is, therefore, an involvement from all parties (banks, buyers, renters,
business partners, etc.) in the cause of the operational risk.
This chapter presents all the main elements of operational risk analysis.
It shows the three different approaches for operational risk identification
analysis, including the self-assessment analysis, the quantitative operational
risk indicators, and the operational risk losses. The latter two approaches are
extensively discussed as they are mostly used in quantitative operational
risk assessment and identification analysis. The definition of the identifica-
tion factors and the design of the operational risk mapping are also
discussed. The detection of causes, events, and consequences and their
interactions are presented, illustrating cases of different types of risks that
may appear within this chain. A particular emphasis in operational risks
arising from the employees and/or IT systems is also given. Two techniques for
measuring operational risks via the key risk indicators and the actual losses are
also discussed. Specifically, for the former technique, the identification and
data accessing issues are highlighted; whereas, for the latter technique, the
internal operational risk loss event data, the external operational risk loss
event data, and the losses based on scenario-simulation analysis are
182
OPERATIONAL RISK IN ISLAMIC FINANCE 183
covered. Finally, the key issues in the evaluation and management of
operational risks are also examined in this chapter.
M A I N E L E M E N T S I N O P E R AT I O N A L R I S K
A N A LY S I S
The main drivers in market risk analysis are the commodity price risk, the
equity price risk, the FX rate risk, and the mark-up risk market risk factors.
Any change in these factors has a computed impact on the value of the
bank’s trading portfolio. On the other hand, the actual losses (their proba-
bility and coverage) resulting from the defaults of the counterparties are
considered as the main drivers in credit risk evaluation analysis. Operational
risk analysis, however, can be based on risk factors that affect the institu-
tion’s business performances and objectives; it can also be based on the
actual operational risk losses (Figure 6.1). Identification and measurement
of the actual risks and their resulting losses is a process that requires a
special framework and effort from the institution.
There are three different approaches (see Figure 6.2) for operational risk
identification analysis:
1. Self assessment analysis
2. Quantitative operational risk indicators
3. Operational risk losses.
The self-assessment approach is mainly based on internal assessment
that involves people’s contribution (i.e. the employees of the institution) for
reporting any possible risks and resulting losses within the actual business
Losses
Financial Risks in
institution operations
Figure 6.1 Operational risk analysis based on operational risks and losses
184 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Operational risk
Self Risk Losses
assessment indicators
what would what can be what already
happen? measured? happened?
Figure 6.2 Main approaches in operational risk identification analysis
operations. Such an approach uses interviews and questionnaires to identify
what could happen in regards to operational risks. A main drawback of this
approach is that the collected information may be biased in nature.
Therefore, although nowadays some financial institutions are assessing and
managing their operational risks via self-assessment approaches, they are
based on subjective information that may minimise their real applicability
and usage. In this book, we will focus on the second and third approaches,
as they are based on more objective quantitative approaches. The quantita-
tive operational risk analysis is driven by risk indicators as well as by the
actual resulting losses. The latter approach is based on historical informa-
tion, whereas the former is based on the actual risk measurements within the
operations. In regards to the advance measurement approach (see chapter 3),
operational risk indicators could be used in the scorecard analysis, whereas
the operational risk losses may be used in internal measurement
approaches.
Financial institutions should apply the appropriate approach according to
their needs and abilities to implement them. Some key points that should be
considered are:
Data availability and integrity: approaches based on risk indicators
and losses are data-intensive. The quantitative parameters of the indicators
must be clearly measured, mainly via an ongoing data collection process.
Moreover, the data referring to losses must be well defined and updated
according to the appearance of the loss events. Although in the analysis of
the Islamic financial products there may be a lack of information referring
to what operational risk is and the resulting losses, financial institutions
should be able to have enough data in order to implement quantitative-based
approaches.
Commitment to implement an operational risk framework and
develop a risk management culture: This requires that the institution will
invest in technologies to support the risk management system. Moreover,
the institution should develop an operational risk management culture as it
OPERATIONAL RISK IN ISLAMIC FINANCE 185
involves nearly every employee in the institution. It should also ensure that
it has an up-to-date knowledge of the risk analysis and management
process; also, an appropriate budget is provided accordingly.
Size of the institution: the size of the institution is usually an indi-
cator of the investment in more advanced and quantitative risk management
systems, analysis, and process. On the other hand, smaller institutions are
more flexible in the implementation process and have less complexity in
their products, services, and operational systems.
Level of complexity: Institutions with complex Islamic financial
products, services, and operations have a higher degree of exposure to more
extreme events (tail risks). Quantitative approaches will provide more
insight into the true risk profile.
Degree of sophistication: Institutions with a high degree of sophis-
tication in their process and systems need to use management approaches
that are more risk-sensitive.
The first step in the analysis of the operational risk is the identification
phase, which entails the mapping of the business lines and operational
processes, the mapping of the operational risks, and the capturing of the
strategic business objectives. A good approach for mapping the institution’s
business lines is based on those identified by Basel II.2 The mapping of the
operations within the business lines, together with the operational risks
associated to the causes, events, and consequences, is also an important step
in the risk identification and evaluation process. The measurement phase
includes the identification of qualitative and the definition of the quantita-
tive parameters. These are used to define and measure the risk indicators
and the resulted operational risk losses. Based on this risk measurement the
evaluation process is the next step of the analysis.
I D E N T I F I C AT I O N O F O P E R AT I O N A L R I S K
Identification factors and operational risk mapping
A main part in designing an effective operational risk management system
is the identification of both internal and external operational risks. In
Islamic finance, such identification should also refer to the operational risks
for the insufficient compliance with Shariah rules and principles. Financial
institutions should identify and assess the operational risk inherently in all
products, activities, processes, and systems. Moreover, based on Basel II
and IFSB directives, further states that risk identification is essential for the
consequent development of a practical operational risk monitoring and con-
trol system. However, the key factors that negatively affect the financial
186 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
institution in terms of reaching their business objectives should be identified
first.
Effective risk identification considers both internal and external factors
that could negatively affect the process of reaching the financial institution’s
objectives. Some important internal factors are:
the structures of the institution’s accounts;
the corresponding contracts;
the nature of the institution’s activities;
the quality of the institution’s human resources;
the organisational changes and employee turnover.
Moreover, some external factors are:
the changes in the industry; and
the technological advances.
It should be standard practice for a financial institution’s management to
implement policies and procedures to manage risks arising from their
operational activities. The institution should maintain written policies and
procedures that identify the risk tolerances approved by the board of
directors and should clearly define lines of authority and responsibility for
managing the risks. The institution’s employees should be fully aware of all
policies and procedures that relate to their specific duties.
The above factors should also be considered in the process for
mapping the business operations and the risks that influence them. The
mapping of the operations’ processes is used to define the key business
operations, the various business units, the organisational functions, and
process flows as well as their direct or indirect links to business targets
and objectives. Note that operations used in the Islamic financial con-
tracts must also be linked to Shariah compliance. For instance, the com-
modities, assets, or constructions agreed in the Istisnã and Salam
contracts should always be linked to the Shariah principles. Moreover,
the operations that refer to the process of producing and delivering prod-
ucts and services should be well defined and monitored in regards to risk
of not complying with Shariah principles. In addition, when the financial
institution agree on a partnership type of agreement, such as the
Mushãrakah and Mudãrabah Islamic financial contracts, additional
OPERATIONAL RISK IN ISLAMIC FINANCE 187
mappings of the operational processes that are linked to these contracts
should also be designed.
The operational process mapping exercise is used to identify key operations
and design a roadmap of the combined key operations by defining inputs
and outputs and linkage between them. In the risk-mapping process, all
possible risks that might affect the operational processes are identified and
linked to the operations process map. The operational risk mapping is used
as the basis to identify the types of operational risks’ causes and their
existence in Islamic financial contracts.
Identification of causes, events, and consequences
Having performed the operational risk mapping, financial institutions
should be able to identify what are the causes of the risks, what are the
events, and what are the downstream effects and consequences. It is some-
times difficult, however, to identify the differences between causes, events,
and consequences. In general, operational risk analysts and managers
should have in their minds that:
what is characterised as a ‘cause’ should result in one or more events;
what is characterised as an ‘event’ should have at least one cause and it
must result in one or more consequences; and finally
what is characterised as a ‘consequence’ must result from one or more
events and may result in new cause(s).
Note that any changes from the initial definition must be documented,
and all those that are involved in the operational risk management process
and systems must be informed.
Table 6.1 illustrates certain types of causes, events, and consequences.
Operational risk causes, events, and consequences are usually associated
with internal control weaknesses or lack of compliance with existing internal
procedures as well as with the Shariah principles. They are found in all areas
of an institution and are mainly caused by the combined actions of people,
technological systems, processes, and some unpredictable external events.
People are the area of greatest variability and, as a result, the sources of
the majority of operational risks. It is recommended that the organisation
look for root causes as opposed to effect. When a risk event is formulating,
the causes or originating source of it must be identified as well as what
consequences it will have and the resulting effect it will have on other risks.
The resulting consequences if the risk is to be ‘accepted’, ‘avoided’, or ‘mit-
igated’ must also be understood. It is important that this categorisation relies
188 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Table 6.1 Examples of types of operational causes, events and, consequences
Cause types Event types Consequence types
Deception of Individual’s Internal fraud. Regulatory and
Behaviour. Compliance.
External Fraud.
Organisational and Legal Liability.
Employment Practices
Corporate Behaviour.
and Workplace Safety.
Faults due to Information Loss/Damage to Assets
Business Disruption.
Technology.
Systems Failures. Third Party Losses and
External Political and Damages to Assets (i.e.
Damage to Physical
Financial Uncertainties. in Ijãrah Contracts).
assets.
Inefficient Agreements
Clients, Products, and Loss of Reputation.
with the Counter-
Business Practices.
parties / Partners due to
Restitution.
inefficient operational Execution, Delivery, and
evaluation of processes. Process Management.
Loss of Resources.
Non Financial External Default of keeping the
Uncertainties. Promise to Buy the Loss of Opportunities.
Commodity (i.e. in
Mismatching Specification
Murãbaha contract). Loss of Market Share.
in Commodities, Assets.
Defaults of the
Uncertainties in Commodity’s Delivery Exposure to Market and
Manufacturing and (i.e. in Salam and Istisnã Credit Risks.
Construction Process. Contracts).
Losses from Covering
External Partnership Failures on Deliveries by
Business Failures
Business Risk. the Partnership
(Mushãrakah and
Obligations (i.e. in
Unclear definitions in Mudãrabah business
Mushãrakah and
Business activities for the agreements).
Mudãrabah Contracts).
Partnership Agreements
that may be against the Default in following the Non-compliance with
Shariah principles. principles of Shariah. Shariah principles.
on a root cause analysis; that is, causes of operational risk loss events, which
are captured in the loss event database.
Identifying root causes can help to identify additional, related risks. By
linking causation to relevant business activities, through correlation analysis,3
this structure is intended to be used as a tool with which to act upon opera-
tional risks. This provides management with an effective operational risk
management framework. The structure also lends itself to quantification of
operational risks by drawing on data sources relevant for modelling.
Realistically, some operational risks must be accepted. How much is
accepted, or not accepted, mainly depends on the operational risk impact
and internal policies of the organisation. Operational risks with a high
degree of impact should not be accepted, even if their probability is low.
OPERATIONAL RISK IN ISLAMIC FINANCE 189
The decision to accept operational risk is affected by many inputs and
policies. When a manager decides to accept operational risks, the
decision should be co-ordinated whenever practical with the affected per-
sonnel and organisations, and then documented so that in the future
everyone will know and understand the elements of the decision and why
it was made.
Types of operational risk causes
Mapping the operational risk during the identification process allows
financial institutions to define and measure the risks within the business
and better understand their operational risk loss profile. Each financial
institution has its own, individual and unique operational settings. Thus, to
be able to manage operational risk might require tailoring its definition to
the institution’s specific settings. In operational risk identification analysis,
all major business disruptions that result in operational risk losses initiated
from People, Systems and Technology, Policies, Processes and Delivery
Failures, Transactions, and/or Internal and External Events should be
considered.
1. People – Humans are one of the main sources of operational risks and
play a major role in Islamic financial contracts (see Table 6.2).
2. Transactions – Failures in financial transactions are illustrated in
Table 6.2.
3. Systems and Technology – Failures referring to systems and technology
that are initiated by internal and external events such as those listed in
Table 6.2 may also be sources of the institution’s operational risk losses.
4. Process and Delivery Failures – Such disruption may refer to Process
execution and Delivery and are present in most Islamic financial con-
tracts, as shown in Table 6.2.
5. Internal and External Events – These are events that cause losses to
the financial institutions due to external events, political uncertainties,
natural disasters, and the actual implementation of the Islamic
contracts.
6. Policies – Policies may refer to incomplete/missing legal documenta-
tions which affect the compliance to the Shariah principles.
Furthermore, it includes unapproved access given to client accounts, or
even to employment practices and workplace safety.
190 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Table 6.2 Sources of operational risks
Sources of
operational Some key risks
risks Types of risks indicators
People Unauthorised usage of internal Staff turnover rates
control, corporate governance
Staff training and
breakdown, mistakes, and
experience levels
incompetence in operational
processes, incomplete or miss- Number of people that
checking the creditability of the have unauthorised
counterparty, authorisation and access to third party
approvals given to contracts that accounts
are not Shariah-compliant.
Number of approvals
Internal fraud: that are not compliant
Intentional misreporting of to Shariah rules and
repayments and deliveries from principles
the counterparty and business
partner, employee theft and
smuggling, insider and outsider
trading, cheque kiting, bribes, etc.
External fraud:
Robbery, Forgery, Damage from
computer hacking, etc.
People’s risks in reference to
Islamic contracts:
Failure to keep the promise of
buying the commodity, i.e. in the
Murãbaha contract.
In the Ijãrah contract, the
ownership of the asset remains
with the lessor and thus any
misconducts by the lessee
(during the lifetime of the
contract or at the maturity day)
are initiating losses accordingly.
In the Mushãrakah contract, the
inability to buy the equities, due
to business failures or the partner.
Continued
OPERATIONAL RISK IN ISLAMIC FINANCE 191
Table 6.2 Continued
Sources of
operational Some key risks
risks Types of risks indicators
Transactions Transactions: Transaction and trades
Clients, Products and Business volume
Transactions, Data entry errors,
Income instability
Product complexity, Document/
contract error, Privacy Breaches, Number of entry errors
Misuse of confidential customer
Frequency and/or
information, Improper trading
severity of transactions
activities on the bank’s account,
errors and omissions
Money laundering, producing and
sale of unauthorised products
that is against the
Shariah principles.
Technological System failures caused by internal Number of instances of
Systems and external events: network or systems
downtime
Internal:
Programming errors, IT crash Maintenance fail over
caused by new applications, rate,
loss of Information data,
Systems’ failure rate
incompatibility
with existing systems, internal Data integrity after
telecommunication failures, failure outage
of system to meet business
System failure retrieval
requirements, etc.
time
External:
Backup failure rate
Utility outages such as power cut,
business disruption caused by Number of failed
system failures, telecommunication operations
problems, information risks and
losses, external security
breaches, etc.
Specification Mismatching
Specification mismatching in
commodities’ or asset’s
productions that were
Continued
192 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Table 6.2 Continued
Sources of
operational Some key risks
risks Types of risks indicators
agreed in the Salam and
Istisnã contracts.
Processes Process execution: Number of contracts
and Delivery Delivery and process management, non-compliant with
Failures product and service complexity, organisation’s policy
management failures, security
failures, incomplete/missing legal Settlement failure rate
documentation, unapproved access
given to client accounts, delivery Accounting losses’ rate
failure and vendor disputes,
payment/settlement delivery risk, Number of issues raised
employment practices and workplace by regulators/auditors
safety, workers’ compensation
claims, violation of employee health Payment failure rate
and safety rules, organised labour
Number of defaults
activities (strikes, etc.),
discrimination claims and general
Number of Islamic
liability.
contracts that are not
Delivery on contracts’ agreement: Shariah compliant
Defaults on the commodity’s
delivery in Salam contracts and
defaults in manufacturing and
construction in Istisnã contracts,
whereas in Mushãrakah and
Mudãrabah contracts the
inabilities of the partner to
further operate the business.
External events External events: Amount of Physical losses
Utility outages, power cut, (financial, people,
telecommunication problem. systems)
Political uncertainties: Loss data rate
War (global effect), damage to
physical assets, fires,viruses/
Continued
OPERATIONAL RISK IN ISLAMIC FINANCE 193
Table 6.2 Continued
Sources of
operational Some key risks
risks Types of risks indicators
mass diseases, terrorism,
vandalism, etc.
Natural disasters:
Earthquakes, Fires, Viruses,
Terrorism, Vandalism, Floods,
Hurricanes, Volcanoes, Damage
To Physical Assets, etc.
Breach of environmental
management, bankruptcy of
supplier, transportation failures, etc.
Events to contracts:
In the Ijãrah Islamic contract,
catastrophic events may cause
damages to the assets (such as
properties) which consequently
result in major losses.
In Mushãrakah and Mudãrabah
contracts, where the financial
institution has a partnership in
the business, any external event
or in general any inadequate
activities or failures due to
business risks may cause
operational risks losses.
In the Mudãrabah contract, high
disruption on business
development that results loss is
fully covered by the
financial institution.
194 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Operational risk
Systems and Policies/ Process and delivery Internal and
People
technology transactions failures external events
Murãbaha Salam Shariah Law Istisnã Ijãrah
Ijãrah Istisnã Mudãrabah Mudãrabah
Mushãrakah Mushãrakah Mushãrakah
Figure 6.3 Sources of operational risks and where they affect Islamic financial
contracts
Figure 6.3 illustrates the different sources of operational risks and where
they affect Islamic financial contracts.
The greatest losses among all operational risks are the ones that are initi-
ated from the employees’ activities and the failures, inefficient, or inappro-
priate use of IT systems/technology. Financial institutions therefore should
pay particular attention to these areas and be able to identify and manage
such sources of operational risks.
Employees’ risks in financial institutions
The risk of a loss intentionally or unintentionally caused by an employee,
such as employee error and employee misdeeds, or involving employees
such as in the area of employment disputes, is the risk class that covers
internal organisational problems, fraud, and losses. Unfortunately, the
largest amount of losses results from intentional activities such as fraud and
unauthorised trading. It is people, not businesses or systems, who commit
fraud or mistakes and such risks increase dramatically every day. Fraud
invades every area of businesses and is committed when a motive coincides
with an opportunity.
Financial organisations should establish and maintain appropriate sys-
tems and controls for the management of operational risks that may arise
from employees. Moreover, they should ensure that all employees are aware
of their operational risk management responsibilities, including by estab-
lishing and maintaining:
A selection of employees that respect and follow the Shariah principles
A separation of the employees’ duties
An internal supervision of the employees’ performances
OPERATIONAL RISK IN ISLAMIC FINANCE 195
A monitoring of the employees’ behaviour
Well-established policies that are complying with the Shariah princi-
ples and are well known by all employees.
Training processes to direct the employees in the process of the risk
management.
Well-defined employment termination policies and procedures.
Using qualitative criteria the identification and assessment of the risks
that result from people is defined and measured via parameters that describe
how people’s risks affect operational performances, which are directly or
indirectly linked to business targets and objectives. The organisation’s
process map highlights the links between operational processes that can
identify and measure operational risks, performances, and business
objectives. The identification process defines the key operational risks and
performance indicators that measure quantitatively both risks and perform-
ances, respectively. Therefore, when it is feasible, the qualitative ‘measure-
ments’ should be transferred to quantitative ones.
IT systems/technology risks
Nowadays, information technology plays a primary role in financial institu-
tions, bringing, however, the potential to transform operational risks from
manual processing errors to system failure risks. The growth of the elec-
tronic banking, electronic payments, and transactions has resulted in the
risk of electronic frauds and system security issues that are not yet fully
understood. The pressure from all the quarters to invest continuously and
dramatically in modern processes and technologies is enormous. Internet-
related technologies enable much higher and more sophisticated levels of
co-ordination, globality, efficiency, and flexibility. However, they open the
door for chaos and operational risks if they are not consistent, structured,
co-ordinated, and stable over time. Systems that are related to IT failures are
significantly increasing in number and severity as most of the business
transactions and processes are fully dependent on IT, and thus failures in IT
are more likely to impact the business, and that impact is more likely to be
severe.
IT security issue
The growing complexity of the IT environment – that is, the underlying
backbone to modern financial institutions, and the potential for substantial
196 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
monetary losses – has increased the importance of IT security. As financial
institutions continuously improve their technological capabilities, the com-
plexity of maintaining a secure IT environment will certainly increase
accordingly. At the same time, attacks on IT systems are increasing and
exposure to security breaches has also increased during the past few years.
As a result, financial institutions should establish and maintain appropriate
systems and controls to manage their information security risks via: appro-
priately trained staff and users, integrity in regards to the accuracy and
completeness of information, setting back-up systems, ensuring that there is
isolation of networks, using firewalls, ensuring confidentiality of the infor-
mation, defining non-denial and accountability where the person or system
that processed the information cannot deny their actions, updating the
virus/worm protections, using data encryption, applying continuous vulner-
ability assessments, using intrusion detection systems and external moni-
toring systems.
Effect of IT operational risk on credit risk
There are cases where the IT systems/technology risks have a significant
impact on credit risk events. Thus, credit-related operational risk may occur
through the following:
Processes: when failures in collateral management occur via, for
instance, an inappropriate or missing documentation, incorrect assign-
ment to trading portfolio, inaccurate valuation.
People: if there is a violation of competence and evaluation rules in
credit-approval processes.
Systems: when there is unavailability of system support for credit-
approval processes.
External events: credit fraud in case, for instance, the credit being
granted is based on forged balance sheets.
Interactions between the causes of
operational risks
Between the individual operational risks types, mentioned above, there is a
certain amount of overlap or interaction (Figure 6.4). The intertwined rela-
tionship among them can be very complex. This interactive relationship
must somehow be understood. From an internal and external perspective,
people and systems and technology interact to produce a successful process
OPERATIONAL RISK IN ISLAMIC FINANCE 197
Internal and
external events People
Operational
risk
Systems
Process and and
delivery failures technology
Transactions
Figure 6.4 Interactions among the different causes of operational risks
as well as comply with the obligations underlined in the different types of
Islamic financial contracts. When an operation referring to internal
processes or to external activities via the business partnership contracts such
as Mushãrakah and Mudãrabah fails, an incident occurs that may result in
losses. Thus, the operational risk management system must be analysed and
the inputs and interaction among the elements must be thoroughly
reassessed. The organisation is often the controlling factor in operational
success or failure.
M E A S U R I N G O P E R AT I O N A L R I S K S
During the identification analysis there are cases where the definitions of
operational risks are initially described qualitatively; however, a process to
transfer these definitions to quantitative values is desirable to convert all
these operational risks into measurable ones. These quantitative values are
defined as Key Risk Indicators (KRIs) and should be established for
operational risk analysis to ensure the escalation of significant risk issues to
appropriate management levels. Operational risks affect to certain degrees,
directly or indirectly, some of the operational business performances. Thus,
these performances are required to be defined, and their variations should
be measured quantitatively.
Identifying operational key risk indicators
The identification and measurement of operational risks via KRIs is becom-
ing an important tool in the framework of operational risk management
198 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
systems. Financial institutions include risk indicators into their measure-
ment approach to clearly identify, measure, and alert when, for instance, the
level of an operational risk is above a set threshold. Using KRIs is one of the
most regular ways to link the quantitative data with the actual values of risk
causes, events, and their consequences. KRIs are mathematical functions
that describe the operational discrepancies of specific operations within
specific business lines. The main challenge in identifying KRIs is to quan-
tify the qualified risks’ parameters by determining their mathematical func-
tions. One of the most common and efficient ways to define the KRIs is by
measuring the variation (differentiation ) and volatility of the operational
systems’, processes’, and people’s performances KRIs (OP)sy, pr, pe and
the deterioration of the business targets when an operational risk occurs.
The variation/volatility in operational performances within the business
lines can be characterised as a risk event or a cause to a new event or even
to a consequence.
Some key points that financial institutions should consider during the
design and implementation phases of the KRIs’ approach are listed as
follows:
In most cases the parameters of the KRIs are linked to the informa-
tion that is frequently being updated. The frequency of updates describes
the number of times a risk of a given size occurs within a given time period
or a given organisational unit. All this information must be accessible and
easily collected.
Information may be sensible in regards to confidentiality.
Therefore, the institutions should identify and ensure that such data can be
used in operational risks analysis.
A simple but very important rule is to keep the identification and
design of the KRI functions simple and transparent by not including large
numbers of parameters. It is preferable to have a larger number of simple
and easy-to-understand KRIs than a smaller number of complex functions.
It is inefficient to have duplicated KRIs where the same measure-
ments of operational risk use slightly different functions.
The correlation between the indicators should be identified and
analysed to evaluate the degree of their significance. High correlation
(negative or positive) indicates high degree of significance.
A complete index that describes and explains all parameters used in
the KRI functions must be in place so that future users and operational risk
analysts can understand where data is coming from.
Akkizidis and Bouchereau (2006)3 proposed techniques for constructing
KRIs for operational risks as well as on how to estimate their significance
values.
OPERATIONAL RISK IN ISLAMIC FINANCE 199
Accessing information used for KRIs
As mentioned above, the access to information data and metrics referring to
the key risk indicators should be available at any requested time. In opera-
tional risk management systems, an important issue is to set the measure-
ments on different types of frequency for collecting such information data.
Thus, there are three ways of accessing them:
On-going access: Information referring to operational risks may
need to be monitored constantly in real-time or near-real-time or at least
many times within a short period (i.e. each day). KRIs with high degree of
significance may need constant monitoring to give an insight into the oper-
ational risks using the parameters that are associated with them.
Periodic access: Operational risks that frequently resulting signifi-
cant losses should be monitored on a periodical basis using KRIs.
Moreover, periodical monitoring should also be applied to external opera-
tional risks that refer to the partnership Islamic financial contracts.
Access as on demand: Any major operational risk occurrence must
be tracked and recorded in terms of the time of appearance and its value.
Additionally, financial institutions should be able to access and record any
particular operational risk measurements when they are requested.
Note, finally, that, based only on past data, we may not always have a
good indication for the future occurrences in regards to risks in operations.
This is the main reason why accessing the measurement and analysing the
most recent information is very useful. KRIs can be used as predictive
indicators of arising risks, risk events, and potential losses.
LO S S E V E N T S
A fundamental principle in Islamic finance is the profit and loss sharing,
where both institutions and counterparties are committed to it. As illustrated
in chapter 2, operational losses arise by applying all types of Islamic finan-
cial contracts, including the Murãbaha, Ijãrah, Salam, Istisnã. Operational
losses also appear in the Mushãrakah and Mudãrabah contracts, where the
institution has a close business relation with the counterparties that are
called partners. In such contract agreements, the institution can be exposed
to a great degree of operational risk as it has the full responsibility for
covering the entire amount of associated losses. The collection, measure-
ment, and evaluation of all these losses are the key issues in the area of
operational risk management and in the development of regulatory capital
requirements.
200 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
As mentioned earlier in this chapter, the loss data collection is an ele-
mentary approach for the assessment of operational risks. The resulting loss
event database should include all these losses that are initiated from the
causes of risks, risk events, and downstream effects. Apart from information
on the gross loss amount, the institution should collect information about
the date of the event and any recoveries. A methodological approach for
identifying and collecting this information is to link the losses with the
internal operational risks. However, in cases where there is a lack (or miss-
ing) of sufficient information, it can be supplemented by external loss event
data, assuming that the provided data are relevant to the institution.
Financial institutions and supervisors are nowadays recognising the sharing
of loss data, based, however, on consistent definitions and metrics. When
there is a lack of both internal and external data, scenario analysis could also
be used as a vital tool.
Internal operational risk loss event data
The institution may implement an internal operational risk loss event
database when it has collected loss information for several years (at least
three). The loss data is obtained across the institution’s business lines,
events, product types, and geographic locations. On the other hand, they can
be obtained via the relevant business activities, such as manufacturing,
construction or services, products, events, and locations that refer to the
contracts with their counterparties (i.e. Ijãrah contract) and business part-
ners (i.e. Mushãrakah and Mudãrabah agreements). The institutions must
make sure that operational risk loss event information associated to their
direct internal business as well as to their external businesses that influence
their contracts should be reported uniformly and accurately. This can be
achieved by having a system that has the ability to automatically track the
losses or where different people are responsible to report them.
The reporting and loss collection process should be driven by the finan-
cial institution’s policies that identifies when an operational risk loss
becomes a loss event and it must be added to the loss event database.
Policies and procedures should be communicated to the relevant staff to
ensure that there is a satisfactory understanding of operational risk and the
data-capture requirements under the AMA operational risk management
framework. In addition, the institution must set up appropriate operational
risk data thresholds. These thresholds can be used as a loss indication and
even in the prevention and corrective actions processes.
External operational risk loss event data
In the case where there are limitations to obtain the above-mentioned
internal operational risk loss event data, financial institutions could use an
OPERATIONAL RISK IN ISLAMIC FINANCE 201
external data. Such data can then be used for defining the institution’s level
of operational risk exposure as long as the data are relevant. This means
losses that are coming from similar institutions and providing similar types
of contracts (especially in regards to the counterparties and business part-
ners) in the same or similar markets, with similar types of operational risks.
Note that operations and people’s behaviours vary in different geographical
areas and markets. Even institutions that are exclusively providing Islamic
products or mixed products (Islamic and Westernised) may have different
processes in operations and thus in their associated risks. Moreover, Islamic
products that are referring to clients in Europe are exposed to a different
degree and type of operational risk than the ones referring to the Middle
East and Far East regions. External data enables institutions to comprehend
industry experience and, in turn, present a way for assessing the adequacy
of its internal data.
External loss data can be obtained via any reasonable manner, including
external databases or direct links to other institutions with similar business
processes and activities. The financial institution may use data gained
through membership of industry groups and others use data obtained from
vendor databases or public sources. It is important to note, however, that the
institution should have policies and procedures for the use of external loss
data in the operational risk framework. This is to make sure that they get
what they should and thus feel confident that the information used and
reported is relevant and relatively accurate. Institutions using the advance
measurement approach must use sophisticated data management practices
to produce realistic and dependable operational risk estimates for calculating
their capital requirements. Finally, note that external data may also be used
as a driver to build up scenario analysis; additionally, they can be used as a
benchmark for the existing (but limited) data or to the overall operational
risk exposure results.
Losses based on scenario simulation analysis
The scenario simulation analysis is based on systems that integrate past
experiences and professional expertise of the designer. Past experiences on
failures and operational risks with the resulting losses are useful in risk
management analysis, but in many cases not enough to give a realistic
estimation of the exposure to future losses. Financial institutions, and
especially the ones that provide Islamic financial contracts, are moving fast
in their Islamic financial product implementation and so are the technology
and operational processes that support them. The know-how and the
processes are rapidly updating until Islamic financial systems will become
mature and the exposure of the institution will be stabilised accordingly.
Scenario analysis is very useful for institutions that provide Islamic finan-
cial contracts where the risk and loss analysis based only on past data may
202 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Internal
losses
External
losses Loss events
database
Loss distribution
Scenario
analysis
Figure 6.5 The three approaches used to cultivate and populate the loss event
databases
give a false warning of the present and future operational risks. Moreover,
scenarios are very useful when used to evaluate high-severity events.
Business managers and risk management experts could have a good
indication of the future losses and thus they can build scenarios that define
such cases. However, institutions must have policies and procedures that
describe scenario analysis and identify its role in the operational risk frame-
work. These policies should refer to all fundamental information for the
scenario analysis, including details of the structure of the scenario, the time
and the reasons why they have been created, the updates and the correspon-
ding extents, the losses that are being looking for, etc.
The distribution of the losses is defined based on the three approaches
used to evaluate and manage the operational risks, as shown in Figure 6.5.
Operational risk profile based on loss distribution
The operational risk profile is based on the distribution of the above-
mentioned operational losses within the space of the axes that define the risk
probability, impact, and exposure values. The first dimension of this distri-
bution is the likelihood or probability of operational risks occurring and lead-
ing to an undesirable event. The second one is about the resulting impact of
operational risks to the institution before and after undertaking risk control
actions, and the third dimension refers to the overall exposure areas of the
operations to all types of operational risks. Figure 6.6 illustrates the distribu-
tion of the operational risk losses within the probability and impact space.
The space that is defined by the first two dimensions (probability and
impact) can be divided into four subspaces: (a) Low impact and low
probability, (b) Low impact and high probability, (c) High impact and high
probability, and (d) High impact and low probability. The first two cases
indicate trivial operational risks in the institution’s and/or its partnership busi-
nesses (referring to Islamic contracts) and a low degree of the corresponding
OPERATIONAL RISK IN ISLAMIC FINANCE 203
Probability
Impact
Figure 6.6 Loss distribution for operational risk
implications. The third case is out of the operational risk pragmatic state of
existence, which means such risks exist in the normal course of doing business.
The fourth case is where the probability of the operational risks appearances is
‘low’, whereas their impact is ‘high’ and must receive significant attention
during the operational risk profiling analysis. All financial institutions at
unexpected times have experienced such cases where the impacts of business
disruptions and losses are very significant. Note, finally, that the distribution
starts from the positive impact due to the fact that in some cases operational
risk events may even have a positive impact on the institution.
The degree of operational risk exposure that defines the third dimen-
sion of the risk profiling axes can be based on the loss distribution within
the probability and impact axes. The operational risk exposure can be
defined based on the methodology called clustering operational risk profile
(ClORiP)2 that is used to identify clusters within the loss distribution.
According to this approach, the co-ordinates defined by the probability,
impact, and significance degrees create groups of loss position within the
axes. These groups of losses that have different density within the axes,
called clusters, are analysed and evaluated based on a methodology called
operational risk mountain surface.2
The operational risk profiling, as well as the measurements of the key
risks within the operations, are the two main elements for operational risk
evaluation. These are performed by applying operational VaR analysis as
discussed in the following sections.
E VA L U AT I N G O P E R AT I O N A L R I S K
B A S E D O N VA R A N A LY S I S
There are mainly two ways for evaluating and managing operational risks in
operational risk analysis: one is by considering the risks and the operations
204 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
that refer to the different business lines of the institution; and the other is by
analysing the actual operational risk losses, i.e. internal, external, and those
that are initiated from the partnership in Islamic-based agreements. The lat-
ter is a top-down approach, whereas the former is a bottom-up approach.
The main advantage in the top-down approach is its simplicity in terms of
loss data considerations. Top-down approaches in operational risk measure-
ment frameworks are appropriate for the determination of overall economic
capital levels for financial institutions. However, it is a rather straightfor-
ward way to determine a capital amount for increasing operational losses
that may not be covered by insurance. Bottom-up approaches, on the other
hand, engage in the mapping of workflows where failures may occur. It
analyses the operational risks from the point of view of individual business
activities within the business lines. The operational risks are modelled
based on risk parameters that construct the KRIs used in the risk quantifica-
tion process. These approaches can make use of operational VaR, as
discussed later.
In terms of management using the top-down approach, financial institu-
tions may be able to drill-down in operational procedures that are linked to
these losses, but it is difficult to discover the related operational risks that
are causing them and are exposing the particular area(s) of the institution.
Thus, such an approach is mostly unable to inform management about par-
ticular weaknesses in the operational processes and therefore they cannot
exhibit fine-tuning capabilities for the implementation of operational risk
controls. The operational risk losses are difficult to be incorporated and
characterise dynamic changes in the operational risk environment that
might affect the operational loss distribution over time. In contrast, the bot-
tom-up approach is more analytical and sophisticated, as it is based on the
analysis of the actual operational risk events. Such approaches are more
transparent and forward looking, providing the ability to the institutions for
drilling-down into the business processes, systems, and people’s actions to
explore their performances and risk status. Thus, bottom-up solutions
involve many parties within the institution from the risk analysts and busi-
ness-line managers to the internal risk auditors and regulators.
By using bottom-up approaches, however, a great number of information
data that refer to operational risks are required in order to perform the
evaluation and management analysis. Such analysis may prove difficult and
time-consuming, as it should take into account the interdependencies/
correlations across business lines and operations. It may, therefore, give too
much in-depth and sometimes irrelevant details about the institution’s
operational structures and related risks. Bottom-up models are more accu-
rate and targeted to the measurement of specific operational risk problems,
but at the same time are more complicated and difficult to characterise and
analyse than top-down models.
OPERATIONAL RISK IN ISLAMIC FINANCE 205
Both approaches may also be employed in parallel to increase the degree
of confidence in terms of evaluating and managing operational risks. Thus,
the institution may try to detect sources of operational risk and identify the
resulting exposure within the business lines; on the other hand, it may use
the past, current, and future (simulated) losses to evaluate the position in
terms of financial capital. Note that ideally the capital requirements defined
for both approaches should converge to the same estimation values.
Operational VaR analysis
Using the above analysis (top-down and bottom-up) and the measuring of
operational risks as well as the resulting losses, the evaluation of operational
risk based on VaR analysis can be implemented. As already presented in the
market risk chapter of this book, the market VaR analysis is based on the
changes in risk factors (rate of return rates, price volatility, shares, and FX
rates), whereas the credit VaR, presented in the credit risk chapter, is driven
by the credit risk losses. Operational VaR analysis is based on the actual
operational risks within the business lines or, in other words, the measure-
ment of risk factors using KRIs. On the other hand, the estimation of the
operational VaR can also be based on the actual operational risk losses as
distributed in the probability and impact risk profile space. As in the other
types of risks, there are three methods for estimating the operational VaR.
These are Historic, Variance–Co-variance, and Monte Carlo methods. The
non-parametric Historic method is the easiest for calculating VaR, as it is
based on actual results; however, it requires a large database and is difficult
to drill-down in the results. Similarly to the historical method, the
Variance–Co-variance also uses historical data. This method is based on the
Variance–Co-variance matrix and the calculation complexity is low with the
capability of drilling-down the calculation process and the results. The
above two methods cannot really consider any non-linear behaviour. Such
consideration can be fully supported by applying the Monte Carlo method,
but it requires a high degree of computational process as it is based on algo-
rithmic steps and loops. This method allows shocking or stressing the risks
that affect the operations within the business lines or even the actual
operations that are at risk to any required limits and thus evaluating their
VaR up to its extreme points. It is the most recommended method to be
implemented in the financial industry as the random values used to shock
the underlying business operations represent, in a way, the reality of
unexpected operational risks and the consequence of losses. Detailed
information about operational VaR analysis can be found in Kalyvas &
Akkizidis, 2006.4
Based on the evaluation results, risk analysts are able to construct the
framework of risk management from defining and constructing the risk
206 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
policies and thus to correct the weaknesses of the institution in regards to
operational risk and the corresponding exposure.
ELEMENTS IN THE FRAMEWORK OF THE
O P E R AT I O N A L R I S K M A N A G E M E N T
All financial institutions, including also those providing Islamic financial
products, should define a framework for their operational risk management.
The main principles are common in both Westernised and Islamic banking,
with the latter having an additional dimension to consider. This is the man-
agement of operational risk so that the institution’s processes and people’s
activities comply with the Shariah principles. This dimension should be
fully integrated with the whole operational risk management framework.
Financial institutions should first plan, schedule, and define the frame-
work of the operational risk actions and policies. The main elements for
constructing this framework are:
Planning the activities for reducing the operational risk probability,
impact, and exposure. Defining a business impact analysis.
Scheduling the every-day actions and policies with regards to the
involvement of people in the operational risk management process and
in decision making, such as on whether and how to accept, avoid,
transfer, or mitigate the operational risks and their resulted losses.
Define the scenario driven by the strategies and policies of the institu-
tion which are mainly covering extreme events of operational risk
emergence.
Continuity or contingency planning and recoveries to ensure that the
business is running (fully or even partially) when extreme events occur
and that the business is also able to recover within a pre-defined set
time.
Defining the operational risk reporting that includes the results of the
risk analyses referring to both the evaluation and management faces.
Institutions should also define internal control actions driven by the
above-mentioned elements together with the Shariah principles.
A detailed description of the guides to operational risk management is
beyond the scope of this book; however, for more explanations, refer to the
list of references.
OPERATIONAL RISK IN ISLAMIC FINANCE 207
SUMMARY
This chapter has presented some of the initial main elements of operational
risk analysis, which includes the identification, mapping, assessment, and
measurement and evaluation. It particularly made references to Islamic
financial products and particularly the compliance with the Shariah
principles. It firstly shows that operational risk identification can be based
on risk factors that affect the institution’s business performances and objec-
tives and also on the actual operational risk losses. Three different
approaches for operational risk identification analysis were presented.
Firstly, the self-assessment analysis uses interviews and questionnaires to
identify what could happen internally in regards to operational risks.
Secondly, the quantitative operational risk indicator approach defines and
measures key risk indicators. And thirdly, the operational risk loss approach
is based on historical information.
Financial institutions should apply the appropriate approach according to
their needs and abilities to implement them. Some key points that should be
considered were discussed and are availability and integrity of data, the
commitment of staff to implement an operational risk framework and
develop a risk management culture, the size of the institution, the level of
complexity and the degree of sophistication of the institution.
The chapter further discussed the identification factors and operational
risk mapping. A main part in designing an effective operational risk man-
agement system is the identification of both internal and external opera-
tional risks. In Islamic finance, such identification should also refer to the
operational risks for the insufficient compliance with the Shariah
principles. It showed that the mapping of the operations’ processes is a
useful tool to define the key business operations, the various business units,
the organisational functions and links to business targets and objectives. It
particularly dwelled into the identification of operational risk causes,
events, and consequences. After performing the operational risk mapping,
financial institutions should be able to identify what are the causes of their
operational risks, what are the events, and what are the downstream effects
and consequences. Operational risk causes, events, and consequences are
usually associated with internal control weaknesses or lack of compliance
with existing internal procedures, as well as with the Shariah principles.
The chapter also talked about interactions between the causes of
Operational Risks (i.e. between People, Systems and Technology,
Transactions, Process and Delivery, Internal and External events) and
operational risk as an initiator of market and credit risks.
It graphically illustrated the different sources of operational risks with
regards to the different Islamic financial contracts and identified that the
greatest losses among all operational risks are the ones that are initiated
208 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
from the employees’ activities and failures and inefficient or inappropriate
use of IT systems and technology. The chapter paid particular attention to
employees’ risks in financial institutions and IT systems and technological-
related risks. It highlighted that the largest amount of losses results from
intentional activities such as fraud and unauthorised trading caused by
employees. Furthermore, the growth of electronic banking, electronic
payments, and transactions has resulted in the risk of electronic frauds and
system security issues. Although internet-related technologies have given
much higher and more sophisticated levels of co-ordination, efficiency, and
flexibility, they have at the same time increased the level of operational risks
when they are not consistent, well-structured, and co-ordinated. The issue
of IT security was particularly discussed because, if financial institutions
continuously improve their technological capabilities, the complexity of
maintaining a secure IT environment will certainly increase. It also high-
lighted that in some cases the operational risks related to IT systems/
technology have a significant impact and can even lead to credit and market
risk. This led on to talk about how operational risks have a complex
interaction among each other and also with other types of risks.
The chapter further highlighted that during the identification analysis
there are cases where the definitions of operational risks are initially
described qualitatively, but a process to transfer these definitions to quanti-
tative values is desirable to convert all these operational risks into measura-
ble ones. These quantitative values are defined as Key Risk Indicators
(KRIs). Financial institutions include key risk indicators into their measure-
ment approach to clearly identify, measure, and alert when thresholds are
reached. It was shown that the main challenge in identifying KRIs is to
quantify the qualified risk parameters by determining their mathematical
functions and also gave some key points that financial institutions should
consider when using the KRI approach. It was noted that the availability of
data as well as accessing this data when it is essential are some of the
biggest challenges of this approach.
It was also discussed that a fundamental principle in Islamic finance is
the profit and loss sharing, where both institutions and counterparties are
committed. Operational losses appear in the Mushãrakah and Mudãrabah
contracts, where the institution has a close business relation with the coun-
terparties that are called partners. In such contract agreements, the institu-
tion can be exposed to a great degree of operational risk as it has the full
responsibility for covering the entire amount of associated losses.
It talked about how the loss data collection is a primary approach for the
assessment of operational risks. It further talked about how the resulting
loss event database should include all the losses that are initiated from the
causes of risks, risk events, and downstream effects. The institution may
implement an internal operational risk loss event database when it has
OPERATIONAL RISK IN ISLAMIC FINANCE 209
collected loss information for a sufficient amount of years. If the internal
data is insufficient or inadequate, it can be substituted with external data,
but the data should come from the same type of institution, with trades from
similar markets. External loss data can be obtained in any practical way,
including using external databases or direct links to other institutions with
similar business processes and activities.
The chapter then described how losses can be estimated based on
Scenario Simulation Analysis. The scenario simulation analysis is based on
systems that integrate past experiences and the professional expertise of the
designer. This analysis is very useful for institutions that provide Islamic
financial contracts where the risk and loss analysis based only on past data
may not give appropriate indications of present and future operational risks.
The distribution and estimation of operational losses is defined based on
internal and external losses and scenario analysis to develop and populate
the loss event database. The graphical representation of the loss distribution
is based on the probability and impact of operational risks and their resulting
losses.
In addition, the chapter described how operational risk evaluation is
performed based on the Value-at-Risk (VaR) analysis. Operational risk
profiling, as well as the measurements of the key risks within the opera-
tions, are the two main elements used for operational risk evaluation.
The chapter also dealt with the top-down approach as opposed to the
bottom-up approach for measuring operational risks. Top-down approaches
are appropriate for the determination of overall economic capital levels for
financial institutions. Bottom-up approaches, on the other hand, engage in
the mapping of workflows where failures may occur and get to the root
causes of operational risks.
For the measuring of operational risks, as well as the resulting losses, the
operational VaR analysis was discussed as a very useful tool. Operational
VaR analysis is based on the actual operational risks within the business
lines or, in other words, the measurement of risk factors using KRIs. It was
shown that there are three methods for estimating the operational VaR.
These are historic, variance–co-variance, and Monte Carlo methods. Each
method is briefly explained, with their benefits and shortcomings.
The chapter further explains that all financial institutions, including also
the ones providing Islamic financial products, should define a framework
for their operational risk management. It adds that there is an additional
element to consider for institutions providing Islamic financial products,
which is the compliance with the Shariah principles. It finally lists some of
the main elements for constructing an efficient operational risk manage-
ment framework. Throughout the chapter, there are references to various
useful methods and techniques for assessing and evaluating operation risks
and their associated losses.
210 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
NOTES
1. Islamic Financial Services Board, Guiding Principles Of Risk Management For Institutions Offering
Only Islamic Financial Services (December 2005).
2. Basel Committee on Banking Supervision, ‘International Convergence of Capital Measurements and
Capital Standards, A Revised Framework’ (2005).
3. Akkizidis & Bouchereau, Guide to Optimal Operational Risk and Basel II (New York: Auerbach
Publications, Taylor & Francis Group, 2005). ISBN: 0849338131.
4. Recommended study for GARP exams, FRM Study Guide, 2007, Chapter 5 / Operational Risk, from
the book by L. Kalyvas, I. Akkizidis, I. Zourka, and V. Bouchereau, Integrating Market, Credit and
Operational Risk: A Complete Guide for Bankers and Risk Professionals (London, Risk Books,
2006).
CHAPTER 7
Concluding
Remarks
Wealth maximisation has been an important goal of the financial institu-
tions for many decades. The quest for generating better revenues is ongoing.
Conventional financial institutions have been playing the role of financial
intermediation for several years. However, the finer details of this role have
changed dramatically. From simple borrowing and lending institutions, they
have emerged as giants who provide a wide range of financial services and
have grown to a such a proportion that the stability of entire financial mar-
kets depends on the stability of these financial institutions. For efficient
financial intermediation, several factors are critical for the organisation.
First, the availability of a basic infrastructure which includes local payment
networks, local money markets, a deep and wide financial market, and an
adequate legal framework. Second, it requires products, people, systems,
and processes. Third, organisation requires an integration of all its activities
from a risk perspective. Finally, it needs a sound economy with vigilant
supervision. The management of risks in financial institutions can be
decisive for the efficient performance of the entire economy. The Islamic
financial institutions re-emerged in the last three decades. They were infor-
mally existent in history. The structured Islamic financial industry started
with the setting-up of IDB and DIB. The prime reasons for this were: dis-
enchantment of the Muslim population with the conventional form of
financing, a quest for a religious form of financing, the rapid economic
growth of countries with significantly higher Muslim populations, changing
political, legal, and economical situations all over the world, and improved
wealth among the Muslims. Social dimension has an important position in
Islamic finance. Equitable distribution of wealth and income and protection
of poor are fundamental to Islamic finance. Thus, prohibition of Riba,
avoidance of Gharar, payment of Zakat, and avoidance of Haram activities
are the cornerstones of Islamic finance.
211
212 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
Keeping the social objectives in mind, the Islamic financial industry is
attempting to evolve and develop. There have been serious efforts towards
developing international accounting standards by the AAOIFI and an
international risk management framework by the IFSB. The Islamic account-
ing is more or less based on conventional accounting. However, there are
some fundamental differences in terms of representation of assets and liabil-
ities, treatment of income and losses, as well as payment of Zakat. Islamic
finance has always placed importance on participation in financed business
activities. Based on this tenet, Mushãrakah and Mudãrabah were developed,
which are two types of partnership contracts in Islamic financing. Islamic
scholars have always insisted on using Mushãrakah and Mudãrabah, but this
did not happen, due to several reasons. Critics of modern Islamic finance
argue that using Murãbaha, which is the most popular type of Islamic con-
tract in recent times, is not the long-term solution and should be used spar-
ingly. Forward sale is prohibited in Islamic finance, with the exception of
Salam and Istisnã. Both these Islamic financial contracts can be used for for-
ward trade under specific circumstances. Leasing is recognised and is per-
mitted through Ijãrah in Islamic finance. The use of all the above-mentioned
products is always subject to specific conditions as instructed by Shariah.
The Shariah interpretations vary and may create confusion for those who fail
to look at the Islamic finance holistically. Extreme caution is thus needed
before making any comments on any Shariah ruling.
Financial contracts in Islamic finance are archetypal. They are structured
in a specific format and have special relationships between the contracting
parties, which sometimes changes during the different stages of the con-
tract. Moreover, the risks in Islamic finance are not distributed as conven-
tional finance. The origin, intensity, and the spread of risks are unique for
Islamic financial institutions. They are mainly concerned with the manage-
ment of credit, operational, market, and liquidity risk, in addition to several
others which are specific to the products, places, and processes. A combi-
nation of these two facts produces a treacherous situation for risk management.
On one hand, the relationship is dynamic and on the other hand the risk pro-
files are also dynamic. Hence, risk management in Islamic financial institu-
tions is far more of a serious issue when compared to conventional financial
institutions. Investment contracts, Mushãrakah and Mudãrabah, exposes
the institution to all four major risks; that is, credit, operational, market, and
liquidity at different stages of the contract. Murãbaha, the most commonly
used Islamic financial product, faces mainly credit, operational, and market
risk during different time period of the contract. Commodity risk is highly
prominent in Salam and Istisnã contracts, although they are also exposed to
credit, operational, market, and liquidity risks. Ijãrah is considered to have
larger risk exposures, primarily because of the long-term nature. It is
exposed to all major risks at different stages of the contract.
CONCLUDING REMARKS 213
The fundamental principle of having sufficient capital cushion for risk
management remains the same for Islamic banks. Basel II guidelines,
although primarily aimed at conventional banks, can be applied to Islamic
banks. Several approaches provided under Pillar 1 can be applied to Islamic
banks with the necessary modifications and adaptations. Pillars 2 and 3 are
equally relevant for Islamic banks. The role of supervisors cannot be under-
mined in Islamic banking and market disclosure is in the larger interest of
the financial industry, which is not specific to conventional banking alone.
Hence, the common misnomer of Basel II being completely irrelevant for
Islamic banks is not true. However, serious and sustained efforts are needed
to find the applicability which is specific to countries and markets. The
contribution of the IFSB is commendable in terms of risk management for
Islamic financial institutions. The risk management principles as released
by them are gladly received by the community and efforts are ongoing to
apply them to banks. However, much is needed to be done. The risk
management for the Islamic financial industry needs more effort.
Credit risk in Islamic financial institutions is corresponding to invest-
ment failures through Mushãrakah and Mudãrabah, lending through
Murãbaha, forward sale complications through Salam and Istisnã, and leas-
ing through Ijãrah. The book goes through methodological steps for esti-
mations of credit risk using several options. The understanding of
interrelationships between the counterparties, the Islamic financial con-
tracts, and the guaranties and collaterals used for coverage is critical for
credit risk exposure identification. There are various methods and tech-
niques for the development of models for credit risk. Expert systems based
on rules developed by the expert’s judgement fall under the category of
qualitative models, whereas quantitative models are based on statistical
approaches. A hybrid model attempts to combine the best of both. The two
types of losses that institutions are interesting to identify are expected and
unexpected ones. The book explains the mathematical functions needed to
compute PD, LGD, and EAD. Credit risk analysis will remain incomplete
without a proper treatment for credit VaR. Estimation of credit VaR depends
on the loss data distribution and which many times is not available. The
alternative approach in the absence of loss data distribution is the Monte
Carlo Simulation. The role of collaterals and guaranties is important in
mitigation of credit risk. A reduction should be allowed in EAD and LGD
when guaranties are used. Ratings also play an important role in the credit
risk management. Several critical issues related to internal rating models are
discussed in the book.
Market risk exposure in Islamic financial institutions is initiated through
rate of return or benchmark rate risk, commodity price risk, FX risk, and
equity price risk. The book explains the four main types of yield curves –
the Normal yield curve, the Steep yield curve, the Flat and Humped yield
214 RISK MANAGEMENT FOR ISLAMIC BANKING AND FINANCE
curve, and the Inverted yield curve – that are extensively used in market risk
analysis. Commodity price risk is very high in Murãbaha, Salam, and
Istisnã. The behaviour of commodities underlying these contracts is driven
by several factors. For Mushãrakah and Mudãrabah contracts, the quantifi-
cation and management of equity risk should be based on available market
prices. A more complex synthesis of portfolio or balance sheet accounts
represents a greater need for data related to market risk. Market sensitivity
represents the volatility of the risk factors which is transformed into market
risk. VaR is the most prominent technique used in the estimation of market
risk. The book explains the three alternative quantitative methods for evalu-
ating market VaR: the Variance–Co-variance, the Historical Simulation, and
the Monte Carlo approaches. Such approaches are mainly driven by the
position and market data referring to market risk factors. The Back-Testing
and Stress-Testing are important to assess the efficiency and applicability of
market VaR methodology.
The most difficult to measure and manage among all the risks nowadays
is the operational risk. Probably very few may disagree with this. The two
ways to connect to operational risk identification are factors affecting the
institution’s business performance and actual operational losses. Digging
deep into the organisation and conducting a self-assessment analysis can
provide important information regarding operational risk. Another approach
to the information can be the quantitative operational risk indicator
approach, which defines and measures key risk indicators. Alternatively,
operational risk loss approach can be used, which is based on historical
information. Management or operational risk goes beyond the mathematical
models and capital adequacy. A cultural change in the organisation regard-
ing the operational risk is needed in order to develop sound operational risk
management practices. Developing a learning organisation is the way for-
ward for better risk management. Shariah risk is a part of operational risk
and should be estimated. Mapping the operational risk is an essential part of
designing an effective operational risk management framework. Mapping
helps the organisation in identification of causes of operational risk. Many
causes of operational risk interact with each other and create a complicated
maze of interconnected risks. People, process and delivery, systems and
technology, transactions, internal and external events, etc., are all interre-
lated and affect each other. The highest amount of losses is caused by frauds
and misappropriations by internal staff members. IT as a business enabler
has played a dual role; it has helped business for better co-ordination,
efficiency, and flexibility, but at the same time heightened the risk expo-
sures due to weak IT security. The most difficult part in operational risk
management is quantification of qualitative values. The KRIs are functions
where their parameters are having quantitative values and thus are applica-
ble for measuring quantitatively the operational risks. Some of the Islamic
CONCLUDING REMARKS 215
financial contracts carry larger operational risk exposures. In the
Mushãrakah and Mudãrabah, for example, the partnership contracts exhibit
larger operational risk exposures due to the typical structure of such
contracts. The loss data collection exercise is at the beginning of the devel-
opment of an operational risk management framework. This data needs to
be collected internally as well as externally. Where past data is not indicative
of present or future operational risk, Scenario Simulation Analysis is a use-
ful tool. The analysis integrates past experiences and the professional
expertise of the designer. Operational VaR analysis can also be used for esti-
mations of operational risk. Calculation of operational VaR using the three
alternatives – the Variance–Co-variance, the Historical, and Monte Carlo –
can be used suitably by institutions aiming to estimate their operational risk.
The Islamic financial industry is at the crossroad. The right direction can
provide the required impetus for the sustained growth in the long run. The
only alternative is through better risk management practices and systems.
Integrated risk management is needed now and for the future. Furthermore,
implementing enterprise-wide risk management is going to be one of the
major steps towards building a risk-sensitive culture in the organisation. The
focus is on risks as risks leads to misdirection, miscalculation, and misrep-
resentation. Investment in better risk management systems and tools solidi-
fies the position of the institution, and thus should be viewed as a strategic
opportunity rather than a situational necessity. Risk management is not a
project – it is a process – an ongoing process.
Index
AAOIFI, 4, 6, 9, 10, 23, 24, 26, 27, 98, 212 functionality test, 142–143
stability and robustness, 145
back-testing, 176–177 stress-testing scenarios, 144–145
Basel II, 34, 133, 80–108, 119, 121, 125, 128, testing of usage, 141
133, 185, 213 credit rating validation, 139–145
behaviour, 113–114, 117, 118, 134, 140, 164 qualitative validation, 139–140
benchmark, 150 quantitative validation, 131–145
credit risk, 32–33, 39, 57, 86–89, 100, 101, 102
causes, 187, 188, 189 assessment, 113–118
consequences, 187 exposure, 110–111, 134
commodity risk, 154–159 internal rating based approach, 87–88
cost risk, 156 modeling, 113–118
future delivery risk, 157–158 operational risk and IT, 196
invert price risk, 158 securitization framework, 88
market’s influence rate risk, 157 standardized approach, 86–87
price risk, 155 valuation, 118–132
quantification, 161 see also credit risk management
quantity and time risk, 157 credit risk management, 109–147
see also market risk management expert systems, 113–115
counterparties, 32, 111–113 hybrid models, 118
see also credit risk ijãrah, 70, 109, 111, 115, 119, 122, 126, 131,
collaterals, 111–113, 133 133, 135, 144, 145
confidence level, 131 istisnã, 67, 109, 111, 112, 119, 122, 126, 131,
correlation coefficient, 130 136, 144, 145
credit rating, 33, 109–111, 113–115, 134–145 mudãrabah, 53, 109, 111, 115, 119, 120, 122,
agencies 121, 126, 131, 133, 144, 145
classification, 137–139 murãbaha, 57, 109, 110, 111, 112, 115, 119,
non-parametric, 136–137 121, 122, 123, 124, 126, 131, 133, 144,
parametric, 136 145
point-in-time, 135–136 mushãrakah, 47, 109, 111, 112, 115, 119, 120,
qualitative rating, 136–139 122, 126, 131, 133, 144, 145
quantitative rating, 136–139 qualitative assessment, 113–115
through-the-cycle, 135–136 quantitative assessment, 115–118
validation, 139–145 salam, 61, 109, 111, 119, 122, 126, 131, 144
credit rating systems, 134–139 credit risk mitigation, 133–134
back-testing, 143–144 credit VaR, 127–130
benchmarking, 144 see also loss data
217
218 INDEX
data, 116–117, 120 information systems, 35–36
availability and integrity, 184 interest rate risk,
integrity, 140–141 see rate of return risk
loss, see loss data see commodity risk
market risk factors, 161–163 interpolation techniques, 166–167
position and market, 166–167 cubic spline, 167
simulating, 117 exponential, 167
defaults, 118–119 linear, 166–167
istisnã, 7, 20–21, 29, 63–68, 77, 78
EAD see exposure at default structure, 63–64
ECL, see expected credit loss see also risk identification, operational risk,
economic capital, 84–85, 130 market risk, credit risk, liquidity risk
employees’ risks, 194–195 management
equity price risk, 160 IT security issues, 195–196
quantification, 161 ijãrah, 7, 21–23, 68–72, 212, 213
valuation issues, 160–161 structure, 68–69
see also market risk see also risk identification, operational risk,
evaluation methods, 168–175 market risk, credit risk, liquidity risk
see also Variance–Co-Variance approach management
see also Monte Carlo (MC) approach
see also Historical-Simulation (HS) approach key risk indicators, 197–199
events, 187 KRI, see key risk indicators
expected credit loss, 128–129
expected losses, 127–130 liquidity risk management
expert systems, see credit risk ijãrah, 71
exposure istisnã, 67
see credit risk mudãrabah, 53
exposure at default, 118, 125, 133 murãbaha,
exposure risk, see position risk mushãrakah, 47
loss data, 183, 184
financial intermediation, 1, 2, 5, 28, 30 based on scenario simulation, 201
financial risk analysis, 41–42 credit VaR estimation, 131
financial risks overview, 30 external, 200
foreign exchange risk, 157, 159 fully available, 131–132
quantification, 161 internal, 200
valuation issues, 160–161 lack of, 132
see also market risk partly available, 132
see also commodity risk loss distribution, 202–203
fuzzy logic, 118, 136 loss events, 199–203
FX see foreign exchange risk see also, loss data
loss given default, 118, 125–127, 133, 135
gharar, 2, 5, 12, 22, 26, 211 implied market LGD, 126
guaranties, 111–113, 133 market LGD, 126
measuring,126
haram, 2, 8, 26, 84, 212 workout LGD, 126
hedging salam, 62 LGD, see loss given default
historical simulation approach, 174–175
HS, see historical simulation market data, see data
market risk, 33–34, 39, 89–92, 103, 148–181
IFSB, 4, 59, 65, 80–108, 182, 185, 212, 213, exposures, 176
credit risk, 101–102 factors, 149
equity investment risk, 102 identification, 149
liquidity risk, 103–104, internal model approach, 89–92
market risk, 103 see also valuation
operational risk , 105 standardized measurement approach, 89
rate of return risk, 104 market risk management
INDEX 219
ijãrah, 71, 67, 148, 150, 152, 153, 159, 162, PD, see probability of default
167, 168, 170, 178, 179, 180 position risk, 168
istisnã, 67, 150, 152, 153, 154, 155, 156, 157, probability of default, 118, 135
158, 159, 162, 167, 168, 170, 178, 179, estimation, 121–125
180 profit and loss sharing, 6, 12, 13, 26, 37
mudãrabah, 53, 150, 159, 160, 162, 168, 170, PLS, see profit and loss sharing
178, 180
murãbaha, 57, 148, 150, 152, 153, 154, 157, qualitative criteria, 124–125, 134
158, 159, 162, 168, 170, 178, 179, 180 qualitative validation, see credit rating
mushãrakah, 47, 150, 159, 160, 161, 162, 168, validation
170, 178, 180 quantitative criteria, 121–124, 134
salam, 62, 53, 148, 150, 154, 155, 157, 158, quantitative validation, see credit rating
159, 162, 167, 168, 170, 178, 180 validation
see also commodity risk
monte carlo simulations, 132, 173–174, 175 rate of return risk, 149–154
MC see monte carlo simulations rating see credit rating
mudãrabah, 13, 15–16, 17, 49–54, 75, 76, 120 recovery rate, 126–127
structure, 49–50 regulatory capital, 84–85
see also risk identification, operational risk, riba, 2, 3, 8, 10, 24, 26, 54, 212
market risk, credit risk, liquidity risk risk identification, 36–41
management ijãrah, 69–70
murãbaha, 7, 10, 16–18, 29, 54–58, 74, 76, 111, istisnã, 64–66
119, 121, 122, 123, 124, 126 mudãrabah, 50–52
structure, 54–55 murãbaha, 55–57
see also risk identification, operational risk, mushãrakah diminishing, 45–46
market risk, credit risk, liquidity risk mushãrakah permanent, 44–45
management salam, 59–60
mushãrakah, 14–15, 16, 17, 29, 42–49, 74–75 risk management
diminishing, 44 see also credit risk management
permanent, 43–44, see also market risk management
structure, 42–43 see also operational risk management
see also risk identification, operational risk, risk mitigation, see credit risk mitigation
market risk, credit risk, liquidity risk
salam, 7, 19–20, 21, 58–63, 64, 77, 119, 126, 131
operational risk, 34–35, 39, 87, 90–96, 101, parallel, 62–63
104–105, 182–210 structure, 58–59
advanced measurement approach, 94 see also risk identification, operational risk,
basic indicator approach, 92–94 market risk, credit risk, liquidity risk
elements, 183–185, 206 management
identification 185–197 scenario, 178, 201
interactions, 196–197 securitization framework, 88
measuring, 197–199 self assessment analysis, 183
profile, 202 sensitivity in market risk, 163, 178
standardized approach, 93 standard deviation, 130
operational risk indicators, 183 stress-testing, 177–179
operational risk losses, 183 shariah,
operational risk management compliance, 4, 9, 23, 37, 39, 41, 100, 103, 104,
ijãrah, 70, 188, 190, 193, 194, 199, 200 105, 107, 182, 186, 190, 192
istisnã, 66, 186, 188, 192, 194, 199 principles, 7, 12, 99, 185, 186, 187, 188, 189,
mudãrabah, 52–53, 182, 186, 188, 192, 193, 191, 206, 207, 209
194, 197, 199, 200, 208 others, 5, 6, 12, 16, 18, 23, 27, 36, 39, 40, 102,
murãbaha, 57, 188, 190, 194, 199 212, 214
mushãrakah, 47, 182, 186, 188, 190, 192, 193, sukuk, 23–25, 72
194, 197, 199, 200, 209
salam, 60, 186, 188, 192, 194, 199 takaful, 4, 24
operational VaR, 203–206 technology risks, 195
220 INDEX
UCL, see unexpected credit loss valuation,
unexpected credit loss, 129 based on rate of return risks, 153–154
unexpected losses, 127–130 models for market risk, 164
see credit risk
validating credit rating, see credit rating Variance–Co-Variance approach, 169–173, 175
validation VC, see Variance–Co-Variance approach
value at risk, 164–166
see credit VaR yield curves, 151–153
see historical simulation applying, 152
see monte carlo simulations types of, 151
see operational VaR
see variance-co-variance zakat, 2, 6, 8, 9, 26, 27, 211, 212