Managing Portfolio Credit Risk in Banks

390 Pages·2016·3.078 MB·English

by  Arindam Bandyopadhyay

Managing Portfolio Credit
Risk in Banks
Arindam Bandyopadhyay
4843/24, 2nd Floor, Ansari Road, Daryaganj, Delhi 110002, India
Cambridge Univerisity Press is part of the University of Cambridge.
It furthers the University’s mission by disseminating knowledge in the pursuit of
education, learning and research at the highest international levels of excellence.
www.cambridge.org
Information on this title: www.cambridge.org/9781107146471
© Arindam Bandyopadhyay 2016
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 2016
Printed in India
A catalogue record for this publication is available from the British Library
ISBN 978-1-107-14647-1 Hardback
Cambridge University Press has no responsibility for the persistence or accuracy
of URLs for external or third-party internet websites referred to in this publication,
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.
To my wife Mousumi, whose encouragement
and support made it possible

Contents
Tables, Figures, Charts viii
Preface xv
Acknowledgements xx
Abbreviations xxii
1. Introduction to Credit Risk 1
2. Credit Rating Models 24
3. Approaches for Measuring Probability of Default (PD) 111
4. Exposure at Default (EAD) and Loss Given Default (LGD) 137
5. Validation and Stress Testing of Credit Risk Models 186
6. Portfolio Assessment of Credit Risk: Default Correlation, Asset
Correlation and Loss Estimation 235
7. Economic Capital and RAROC 276
8. Basel II IRB Approach of Measuring Credit Risk Regulatory
Capital 318
Index 355
Tables, Figures, Charts
Tables
Table 1.1: Trends in Quarterly Gross Non-performing Assets of
Indian Banks by Banking Groups (%) 10
Table 2.1: Expert Judgement System vs. Model-driven System
Rating Model 27
Table 2.2: Example of an Internal Rating Template 33
Table 2.3: Use of Two-Dimensional Rating in Credit
Management 37
Table 2.4: Example of Project Finance Expert-based Rating
Model 40
Table 2.5: Risk Criteria of Credit Rating Agencies (CRAs) 47
Table 2.6: Comparison of Predictive Power of New Z-score
vis-à-vis other Z-score Models 59
Table 2.7: Parameters of Indian Logit EPD Models 65
Table 2.8: Example of Statistically Derived Application Scorecard—
Residential Housing Loan 75
Table 2.9: Statistically Derived Risk Weights in Agri-loans 78
Table 2.10: Mapping Scores to PD 80
Table 2.11: Steps in Estimating EDF in MKMV Model 93
Table 2.12: Calibration of Real EDFs with CRISIL Corporate
Rating Grades 95
Table 2.13: Example of Hybrid Corporate Logit Model for Indian
Public Firms 103
Table 3.1: Historical Rating-wise Default Statistics 116
Table 3.2: Average One-year Rating Transition of 572 Indian
Corporate Bonds Rated Externally by CRISIL,
1992–2009 118
Table 3.2a: Indian Corporate Loan Rating Movements in Recent
Years, 2008–15 119
Tables, Figures, Charts ix
Table 3.3: C RISIL’s Published One-year Indian Corporate
Transition Matrix 119
Table 3.4: S&P Global Corporate Transition Matrix in %
(1981–2012) 120
Table 3.5: Annual Industry PDs (%) for Different Loan Grades 121
Table 3.6: One-year Corporate Transition Matrix of a Bank
in India, 2003–09 (%) 123
Table 3.7: Historical Rating-wise Default Statistics 126
Table 3.8: Relationship between Yearly PD and Cumulative PD
(CPD) 127
Table 3.9: Default Rates for Different Horizons 128
Table 3.10: CRISIL’s Indian Corporate Cumulative PDs
(Withdrawal Adjusted) (%) 129
Table 3.11: Estimation of Frequency-based Pooled PD for
Homogeneous Retail Buckets (Personal Loans) –
Illustration 1 131
Table 3.12: Estimation of Long-run Average Pooled Probability of
Default for Homogeneous Retail Pool (Personal Loan) –
Illustration 2 (Exposure-based Method) 133
Table 4.1: UGD per Rating Class on Bank’s Loan Commitments 143
Table 4.2: Facility-wise CCF/UGD (%) Estimates of a Large
Indian PSB 145
Table 4.3: Facility Pool-wise UGD/CCFs of another Large PSB 145
Table 4.4: Rating-wise UGD Estimates of a Large PSB 146
Table 4.5: Illustrative Example for Computing Historical and
Economic LGD 156
Table 4.6: List of Popular Public Studies on Loan LGD 160
Table 4.7: First Round LGD Survey Estimates for Indian Banks:
Commercial and Retail Bank Loans, 1998–2007 161
Table 4.8: LGD (%) Statistics for Defaulted Commercial Loans in
India: Second Round Survey Results 164
Table 4.9: LGD (%) Statistics for Commercial Loans: Secured vs.
Unsecured Loans 165
Table 4.10: Margin-wise LGD (%) Statistics–Secured Commercial
Loans 166
Table 4.11: Collateral-wise Secured Commercial Loan LGD (%) 167
Table 4.12: Historical LGD (%) for Retail Loans: Secured vs.
Unsecured Loans 168
x | Tables, Figures, Charts
Table 4.13: L GD Predictor Models – Multivariate Tobit
Regression Results 174
Table 4.14: Estimation of Long-run Average LGD for a Retail Pool 176
Table 4.15: LGDs: Simple vs. Weighted Average by Default Year
(Corporate Loans) 178
Table 5.1: Validation of CRAs Ratings through Descriptive
Statistics 198
Table 5.2: Group Statistics – Solvent vs. Defaulted Firms 199
Table 5.3: Classification Power of the Model (Within Sample Test) 200
Table 5.4: Validation Report of a Bank’s Internal Rating System
for Commercial Loans, 2003–09 202
Table 5.5: Comparison of Discriminatory Power of Two-rating
Models 206
Table 5.6: Comparing Model Gini Coefficients 207
Table 5.7: Example of KS Test 212
Table 5.8: A Retail-rating Model Calibration 214
Table 5.9: Chi-square Test for Model Comparison 217
Table 5.10: Calibration Test for LGD Rating Model 218
Table 5.10a: Comparing the Discriminatory Power of Models 219
Table 6.1: Portfolio Loss Calculations for Two-asset Example 241
Table 6.2: Assessment Industry Rating Position and Sectoral
Credit Growth 244
Table 6.3: Estimation of Single Default Correlation 249
Table 6.4: Default and Asset Correlation of Indian Banks 250
Table 6.5: Estimation of Rating-wise Default Correlation 253
Table 6.6: Overall IG–NIG Default Correlations (%), 1992–93
to 2012–13 254
Table 6.7: Default Correlation across Rating Grades, 1992–93 to
2008–09 255
Table 6.7a: Global Rating-wise Default Correlations (%) – All
Countries, All Industries, 1981–2002, S&P Credit Pro 255
Table 6.8: Industry Risk Weights 259
Table 6.9: The System-Level Industry Default Correlation
Estimates in India 261
Table 6.10: Descriptive Statistics for Exposure Concentration of
Large Borrowers (of A/Cs > `5 crore exposure) 267
Table 6.11: Estimation of Rating-wise Single Default Correlation 268