Tài liệu Risk mgt in banking

  • Số trang: 813 |
  • Loại file: PDF |
  • Lượt xem: 101 |
  • Lượt tải: 0

Đã đăng 59174 tài liệu

Mô tả:

RISK MANAGEMENT IN BANKING RISK MANAGEMENT IN BANKING Joël Bessis Copyright  2002 by John Wiley & Sons Ltd, Baffins Lane, Chichester, West Sussex, PO19 1UD, England National 01243 779777 International (+44) 1243 779777 e-mail (for orders and customer service enquiries): cs-books@wiley.co.uk Visit our Home Page on http://www.wileyeurope.com or http://www.wiley.com All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise, except under the terms of the Copyright, Designs and Patents Act 1988 or under the terms of a licence issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London, UK W1P 9HE, without the permission in writing of the publisher. Joël Bessis has asserted his right under the Copyright, Designs and Patents Act 1988, to be identified as the author of this work. Other Wiley Editorial Offices John Wiley & Sons, Inc., 605 Third Avenue, New York, NY 10158-0012, USA WILEY-VCH Verlag GmbH, Pappelallee 3, D-69469 Weinheim, Germany John Wiley & Sons (Australia) Ltd, 33 Park Road, Milton, Queensland 4064, Australia John Wiley & Sons (Asia) Pte Ltd, 2 Clementi Loop #02-01, Jin Xing Distripark, Singapore 129809 John Wiley & Sons (Canada) Ltd, 22 Worcester Road, Rexdale, Ontario M9W 1L1, Canada Library of Congress Cataloguing-in-Publication Data Bessis, Joël. [Gestion des risques et gestion actif-passif des banques. English] Risk management in banking/Joël Bessis.—2nd ed. p. cm. Includes bibliographical references and index. ISBN 0-471-49977-3 (cloth)—ISBN 0-471-89336-6 (pbk.) 1. Bank management. 2. Risk management. 3. Asset-liability management. I. Title. HG1615 .B45713 2001 332.1′ 068′ 1—dc21 2001045562 British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN 0-471-49977-3 (cloth) ISBN 0-471-89336-6 (paper) Typeset in 10/12pt Times Roman by Laserwords Private Limited, Chennai, India. Printed and bound in Great Britain by TJ International Ltd, Padstow, England. This book is printed on acid-free paper responsibly manufactured from sustainable forestation, for which at least two trees are planted for each one used for paper production. Contents Introduction ix SECTION 1 Banking Risks 1 1 2 Banking Business Lines Banking Risks SECTION 2 Risk Regulations 3 Banking Regulations SECTION 3 Risk Management Processes 4 5 Risk Management Processes Risk Management Organization SECTION 4 Risk Models 6 7 8 9 Risk Measures VaR and Capital Valuation Risk Model Building Blocks 3 11 23 25 51 53 67 75 77 87 98 113 SECTION 5 Asset–Liability Management 129 10 11 12 13 14 131 136 151 164 180 ALM Overview Liquidity Gaps The Term Structure of Interest Rates Interest Rate Gaps Hedging and Derivatives vi CONTENTS SECTION 6 Asset–Liability Management Models 191 15 16 17 18 19 193 201 210 224 233 Overview of ALM Models Hedging Issues ALM Simulations ALM and Business Risk ALM ‘Risk and Return’ Reporting and Policy SECTION 7 Options and Convexity Risk in Banking 245 20 Implicit Options Risk 21 The Value of Implicit Options 247 254 SECTION 8 Mark-to-Market Management in Banking 269 22 23 24 25 271 280 289 300 Market Value and NPV of the Balance Sheet NPV and Interest Rate Risk NPV and Convexity Risks NPV Distribution and VaR SECTION 9 Funds Transfer Pricing 309 26 FTP Systems 27 Economic Transfer Prices 311 325 SECTION 10 Portfolio Analysis: Correlations 337 28 Correlations and Portfolio Effects 339 SECTION 11 Market Risk 357 29 30 31 32 359 363 384 396 Market Risk Building Blocks Standalone Market Risk Modelling Correlations and Multi-factor Models for Market Risk Portfolio Market Risk SECTION 12 Credit Risk Models 417 33 Overview of Credit Risk Models 419 SECTION 13 Credit Risk: ‘Standalone Risk’ 433 34 35 36 37 38 435 443 451 459 479 Credit Risk Drivers Rating Systems Credit Risk: Historical Data Statistical and Econometric Models of Credit Risk The Option Approach to Defaults and Migrations CONTENTS 39 40 41 42 43 Credit Risk Exposure From Guarantees to Structures Modelling Recoveries Credit Risk Valuation and Credit Spreads Standalone Credit Risk Distributions vii 495 508 521 538 554 SECTION 14 Credit Risk: ‘Portfolio Risk’ 563 44 45 46 47 48 49 50 565 580 586 595 608 622 627 Modelling Credit Risk Correlations Generating Loss Distributions: Overview Portfolio Loss Distributions: Example Analytical Loss Distributions Loss Distributions: Monte Carlo Simulations Loss Distribution and Transition Matrices Capital and Credit Risk VaR SECTION 15 Capital Allocation 637 51 52 639 655 Capital Allocation and Risk Contributions Marginal Risk Contributions SECTION 16 Risk-adjusted Performance 667 53 54 669 679 Risk-adjusted Performance Risk-adjusted Performance Implementation SECTION 17 Portfolio and Capital Management (Credit Risk) 689 55 56 57 58 59 60 691 701 714 721 733 744 Portfolio Reporting (1) Portfolio Reporting (2) Portfolio Applications Credit Derivatives: Definitions Applications of Credit Derivatives Securitization and Capital Management Bibliography 762 Index 781 Introduction Risk management in banking designates the entire set of risk management processes and models allowing banks to implement risk-based policies and practices. They cover all techniques and management tools required for measuring, monitoring and controlling risks. The spectrum of models and processes extends to all risks: credit risk, market risk, interest rate risk, liquidity risk and operational risk, to mention only major areas. Broadly speaking, risk designates any uncertainty that might trigger losses. Risk-based policies and practices have a common goal: enhancing the risk–return profile of the bank portfolio. The innovation in this area is the gradual extension of new quantified risk measures to all categories of risks, providing new views on risks, in addition to qualitative indicators of risks. Current risks are tomorrow’s potential losses. Still, they are not as visible as tangible revenues and costs are. Risk measurement is a conceptual and a practical challenge, which probably explains why risk management suffered from a lack of credible measures. The recent period has seen the emergence of a number of models and of ‘risk management tools’ for quantifying and monitoring risks. Such tools enhance considerably the views on risks and provide the ability to control them. This book essentially presents the risk management ‘toolbox’, focusing on the underlying concepts and models, plus their practical implementation. The move towards risk-based practices accelerated in recent years and now extends to the entire banking industry. The basic underlying reasons are: banks have major incentives to move rapidly in that direction; regulations developed guidelines for risk measurement and for defining risk-based capital (equity); the risk management ‘toolbox’ of models enriched considerably, for all types of risks, providing tools making risk measures instrumental and their integration into bank processes feasible. THE RATIONALE FOR RISK-BASED PRACTICES Why are visibility and sensitivity to risks so important for bank management? Certainly because banks are ‘risk machines’: they take risks, they transform them, and they embed x INTRODUCTION them in banking products and services. Risk-based practices designate those practices using quantified risk measures. Their scope evidently extends to risk-taking decisions, under an ‘ex ante’ perspective, and risk monitoring, under an ‘ex post’ perspective, once risk decisions are made. There are powerful motives to implement risk-based practices: to provide a balanced view of risk and return from a management point of view; to develop competitive advantages, to comply with increasingly stringent regulations. A representative example of ‘new’ risk-based practices is the implementation of riskadjusted performance measures. In the financial universe, risk and return are two sides of the same coin. It is easy to lend and to obtain attractive revenues from risky borrowers. The price to pay is a risk that is higher than the prudent bank’s risk. The prudent bank limits risks and, therefore, both future losses and expected revenues, by restricting business volume and screening out risky borrowers. The prudent bank avoids losses but it might suffer from lower market share and lower revenues. However, after a while, the risk-taker might find out that higher losses materialize, and obtain an ex post performance lower than the prudent lender performance. Who performs best? Unless assigning some measure of risk to income, it is impossible to compare policies driven by different risk appetites. Comparing performances without risk adjustment is like comparing apples and oranges. The rationale of risk adjustment is in making comparable different performances attached to different risk levels, and in general making comparable the risk–return profiles of transactions and portfolios. Under a competitive perspective, screening borrowers and differentiating the prices accordingly, given the borrowers’ standing and their contributions to the bank’s portfolio risk–return profile, are key issues. Not doing so results in adverse economics for banks. Banks who do not differentiate risks lend to borrowers rejected by banks who better screen and differentiate risks. By overpricing good risks, they discourage good borrowers. By underpricing risks to risky customers, they attract them. By discouraging the relatively good ones and attracting the relatively bad ones, the less advanced banks face the risk of becoming riskier and poorer than banks adopting sound risk-based practices at an earlier stage. Those banking institutions that actively manage their risks have a competitive advantage. They take risks more consciously, they anticipate adverse changes, they protect themselves from unexpected events and they gain the expertise to price risks. The competitors who lack such abilities may gain business in the short-term. Nevertheless, they will lose ground with time, when those risks materialize into losses. Under a management perspective, without a balanced view of expected return and risk, banks have a ‘myopic’ view of the consequences of their business policies in terms of future losses, because it is easier to measure income than to capture the underlying risks. Even though risks remain a critical factor to all banks, they suffer from the limitations of traditional risk indicators. The underlying major issue is to assign a value to risks in order to make them commensurable with income and fully address the risk–return trade-off. Regulation guidelines and requirements have become more stringent on the development of risk measures. This single motive suffices for developing quantified risk-based practices. However, it is not the only incentive for structuring the risk management tools and processes. The above motivations inspired some banks who became pioneers in this field many years before the regulations set up guidelines that led the entire industry towards more ‘risk-sensitive’ practices. Both motivations and regulations make risk measurement a core building block of valuable risk-based practices. However, both face the same highly challenging risk measuring issue. INTRODUCTION xi RISK QUANTIFICATION IS A MAJOR CHALLENGE Since risks are so important in banking, it is surprising that risk quantification remained limited until recently. Quantitative finance addresses extensively risk in the capital markets. However, the extension to the various risks of financial institutions remained a challenge for multiple reasons. Risks are less tangible and visible than income. Academic models provided foundations for risk modelling, but did not provide instrumental tools helping decision-makers. Indeed, a large fraction of this book addresses the gap between conceptual models and banking risk management issues. Moreover, the regulators’ focus on risks is still relatively recent. It dates from the early stages of the reregulation phase, when the Cooke ratio imposed a charge in terms of capital for any credit risk exposure. Risk-based practices suffered from real challenges: simple solutions do not help; risk measures require models; models not instrumental; quantitative finance aimed at financial markets more than at financial institutions. For such reasons, the prerequisites for making instrumental risk quantifications remained out of reach. Visibility on Losses is Not Visibility on Risks Risks remain intangible and invisible until they materialize into losses. Simple solutions do not really help to capture risks. For instance, a credit risk exposure from a loan is not the risk. The risk depends on the likelihood of losses and the magnitude of recoveries in addition to the size of the amount at risk. Observing and recording losses and their frequencies could help. Unfortunately, loss histories are insufficient. It is not simple to link observable losses and earning declines with specific sources of risks. Tracking credit losses does not tell whether they result from inadequate limits, underestimating credit risk, inadequate guarantees, or excessive risk concentration. Recording the fluctuations of the interest income is easy, but tracing back such changes to interest rates is less obvious. Without links to instrumental risk controls, earning and loss histories are of limited interest because they do not help in taking forward looking corrective actions. Visibility on losses is not visibility on risks. Tracking Risks for Management Purposes Requires Models Tracking risks for management purposes requires models for better capturing risks and relating them to instrumental controls. Intuitively, the only way to quantify invisible risks is to model them. Moreover, multiple risk indicators are not substitutes for quantified measures. Surveillance of risk typically includes such various items as exposure size, watch lists for credit risk, or value changes triggered by market movements for market instruments. These indicators capture the multiple dimensions of risk, but they do not add them up into a quantified measure. Finally, missing links between future losses from current risks and risk drivers, which are instrumental for controlling risk, make it unfeasible to timely monitor risks. The contribution of models addresses such issues. They provide quantified measures of risk or, in other words, they value the risk of banks. Moreover, they do so in a way that allows tracing back risks to management controls over risk exposures of financial institutions. Without such links, risk measures would ‘float in the air’, without providing management tools. xii INTRODUCTION Financial Markets versus Financial Institutions The abundance of models in quantitative finance did not address the issues that financial institutions face until recently, except in certain specific areas such as asset portfolio management. They undermined the foundations of risk management, without bridging the gap between models and the needs of financial institutions. Quantitative finance became a huge field that took off long ago, with plenty of pioneering contributions, many of them making their authors Nobel Prize winners. In the market place, quantification is ‘natural’ because of the continuous observation of prices and market parameters (interest rates, equity indexes, etc.). For interest rate risk, modelling the term structure of interest rates is a classical field in market finance. The pioneering work of Sharpe linked stock prices to equity risk in the stock market. The Black–Scholes option model is the foundation for pricing derivative instruments, options and futures, which today are standard instruments for managing risks. The scientific literature also addressed credit risk a long time ago. The major contribution of Robert Merton on modelling default within an option framework, a pillar of current credit risk modelling, dates from 1974. These contributions fostered major innovations, from pricing market instruments and derivatives (options) that serve for investing and hedging risks, to defining benchmarks and guidelines for the portfolios management of market instruments (stocks and bonds). They also helped financial institutions to develop their business through ever-changing product innovations. Innovation made it feasible to customize products for matching investors’ needs with specific risk–return bundles. It also allowed both financial and corporate entities to hedge their risks with derivatives. The need for investors to take exposures and, for those taking exposures, to hedge them provided business for both risk-takers and risk-hedgers. However, these developments fell short of directly addressing the basic prerequisites of a risk management system in financial institutions. Prerequisites for Risk Management in Financial Institutions The basic prerequisites for deploying risk management in banks are: • Risks measuring and valuation. • Tracing risks back to risk drivers under the management control. Jumping to market instruments for managing risks without prior knowledge of exposures to the various risks is evidently meaningless unless we know the magnitude of the risks to keep under control, and what they actually mean in terms of potential value lost. The risk valuation issue is not simple. It is much easier to address in the market universe. However, interest rate risk requires other management models and tools. All banking business lines generate exposures to interest rate risks. However, linking interest income and rates requires modelling the balance sheet behaviour. Since the balance sheet generates both interest revenues and interest costs, they offset each other to a certain extent, depending on matches and mismatches between sizes of assets and liabilities and interest rate references. Capturing the extent of offsetting effects between assets and liabilities also requires dedicated models. INTRODUCTION xiii Credit risk remained a challenge until recently, even though it is the oldest of all banking risks. Bank practices rely on traditional indicators, such as credit risk exposures measured by outstanding balances of loans at risk with borrowers, or amounts at risk, and internal ratings measuring the ‘quality’ of risk. Banking institutions have always monitored credit risk actively, through a number of systems such as limits, delegations, internal ratings and watch lists. Ratings agencies monitor credit risk of public debt issues. However, credit risk assessment remained judgmental, a characteristic of the ‘credit culture’, focusing on ‘fundamentals’: all qualitative variables that drive the credit worthiness of a borrower. The ‘fundamental’ view on credit risk still prevails, and it will obviously remain relevant. Credit risk benefits from diversification effects that limit the size of credit losses of a portfolio. Credit risk focus is more on transactions. When moving to the global portfolio view, we know that a large fraction of the risk of individual transactions is diversified away. A very simple question is: By how much? This question remained unanswered until portfolio models, specifically designed for that purpose, emerged in the nineties. It is easy to understand why. Credit risk is largely invisible. The simultaneous default of two large corporate firms, for whom the likelihood of default is small, is probably an unobservable event. Still, this is the issue underlying credit risk diversification. Because of the scarcity of data available, the diversification issue for credit risk remained beyond reach until new modelling techniques appeared. Portfolio models, which appeared only in the nineties, turned around the difficulty by modelling the likelihood of modelled defaults, rather than actual defaults. This is where modelling risks contributes. It pushes further away the frontier between measurable risks and invisible–intangible risks and, moreover, it links risks to the sources of uncertainty that generate them. PHASES OF DEVELOPMENT OF THE REGULATORY GUIDELINES Banks have plenty of motives for developing risk-based practices and risk models. In addition, regulators made this development a major priority for the banking industry, because they focus on ‘systemic risk’, the risk of the entire banking industry made up of financial institutions whose fates are intertwined by the density of relationships within the financial system. The risk environment has changed drastically. Banking failures have been numerous in the past. In recent periods their number has tended to decrease in most, although not all, of the Organization for Economic Coordination and Development (OECD) countries, but they became spectacular. Banking failures make risks material and convey the impression that the banking industry is never far away from major problems. Mutual lending–borrowing and trading create strong interdependencies between banks. An individual failure of a large bank might trigger the ‘contagion’ effect, through which other banks suffer unsustainable losses and eventually fail. From an industry perspective, ‘systemic risk’, the risk of a collapse of the entire industry because of dense mutual relations, is always in the background. Regulators have been very active in promoting pre-emptive policies for avoiding individual bank failures and for helping the industry absorb the shock of failures when they happen. To achieve these results, regulators have totally renovated the regulatory framework. They promoted and enforced new guidelines for measuring and controlling the risks of individual players. xiv INTRODUCTION Originally, regulations were traditional conservative rules, requiring ‘prudence’ from each player. The regulatory scheme was passive and tended to differentiate prudent rules for each major banking business line. Differentiated regulations segmented the market and limited competition because some players could do what others could not. Obvious examples of segmentation of the banking industry were commercial versus investment banking, or commercial banks versus savings institutions. Innovation made rules obsolete, because players found ways to bypass them and to compete directly with other segments of the banking industry. Obsolete barriers between the business lines of banks, plus failures, triggered a gradual deregulation wave, allowing players to move from their original business field to the entire spectrum of business lines of the financial industry. The corollary of deregulation is an increased competition between unequally experienced players, and the implication is increased risks. Failures followed, making the need for reregulation obvious. Reregulation gave birth to the current regulatory scheme, still evolving with new guidelines, the latest being the New Basel Accord of January 2001. Under the new regulatory scheme, initiated with the Cooke ratio in 1988, ‘risk-based capital’ or, equivalently, ‘capital adequacy’ is a central concept. The philosophy of ‘capital adequacy’ is that capital should be capable of sustaining the future losses arising from current risks. Such a sound and simple principle is hardly debatable. The philosophy provides an elegant and simple solution to the difficult issue of setting up a ‘pre-emptive’, ‘ex ante’ regulatory policy. By contrast, older regulatory policies focused more on corrective actions, or ‘after-the-fact’ actions, once banks failed. Such corrective actions remain necessary. They were prompt when spectacular failures took place in the financial industry (LTCM, Baring Brothers). Nevertheless, avoiding ‘contagion’ when bank failures occur is not a substitute for pre-emptive actions aimed at avoiding them. The practicality of doing so remains subject to adequate modelling. The trend towards more internal and external assessment on risks and returns emerged and took momentum in several areas. Through successive accords, regulators promoted the building up of information on all inputs necessary for risk quantification. Accounting standards evolved as well. The ‘fair value’ concept gained ground, raising hot debates on what is the ‘right’ value of bank assets and how to accrue earnings in traditional commercial banking activities. It implies that a loan providing a return not in line with its risk and cost of funding should appear at lower than face value. The last New Basel Accord promotes the ‘three pillars’ foundation of supervision: new capital requirements for credit risk and operational risks; supervisory processes; disclosure of risk information by banks. Together, the three pillars allow external supervisors to audit the quality of the information, a basic condition for assessing the quality and reliability of risk measures in order to gain more autonomy in the assessment of capital requirements. Regulatory requirements for market, credit and operational risk, plus the closer supervision of interest rate risk, pave the way for a comprehensive modelling of banking risks, and a tight integration with risk management processes, leading to bank-wide risk management across all business lines and all major risks. FROM RISK MODELS TO RISK MANAGEMENT Risk models have two major contributions: measuring risks and relating these measures to management controls over risks. Banking risk models address both issues by embedding the specifics of each major risk. As a direct consequence, there is a wide spectrum of INTRODUCTION xv modelling building blocks, differing across and within risks. They share the risk-based capital and the ‘Value at Risk’ (VaR) concepts that are the basic foundations of the new views on risk modelling, risk controlling and risk regulations. Risk management requires an entire set of models and tools for linking risk management (business) issues with financial views on risks and profitability. Together, they make up the risk management toolbox, which provides the necessary inputs that feed and enrich the risk process, to finally close the gap between models and management processes. Risk Models and Risks Managing the banking exposure to interest rate risk and trading interest rate risk are different businesses. Both commercial activities and trading activities use up liquidity that financial institutions need to fund in the market. Risk management, in this case, relates to the structural posture that banks take because of asset and liability mismatches of volumes, maturity and interest rate references. Asset–Liability Management (ALM) is in charge of managing this exposure. ALM models developed gradually until they became standard references for managing the liquidity and interest rate risk of the banking portfolio. For market risk, there is a large overlap between modelling market prices and measuring market risk exposures of financial institutions. This overlap covers most of the needs, except one: modelling the potential losses from trading activities. Market risk models appeared soon after the Basel guidelines started to address the issues of market risk. They appeared sufficiently reliable to allow internal usage by banks, under supervision of regulators, for defining their capital requirements. For credit risk, the foundations exist for deploying instrumental tools fitting banks’ requirements and, potentially, regulators’ requirements. Scarce information on credit events remains a major obstacle. Nevertheless, the need for quantification increased over time, necessitating measuring the size of risk, the likelihood of losses, the magnitude of losses under default and the magnitude of diversification within banks’ portfolios. Modelling the qualitative assessment of risk based on the fundamentals of borrowers has a long track record of statistical research, which rebounds today because of the regulators’ emphasis on extending the credit risk data. Since the early nineties, portfolio models proposed measures of credit risk diversification within portfolios, offering new paths for quantifying risks and defining the capital capable of sustaining the various levels of portfolio losses. Whether the banks should go along this path, however, is no longer a question since the New Basel Accord of January 2001 set up guidelines for credit risk-sensitive measures, therefore preparing the foundations for the full-blown modelling of the credit risk of banks’ portfolios. Other major risks appeared when progressing in the knowledge of risks. Operational risk became a major priority, since January 2001, when the regulatory authorities formally announced the need to charge bank capital against this risk. Capital and VaR It has become impossible to discuss risk models without referring to economic capital and VaR. The ‘capital adequacy’ principle states that the bank’s capital should match risks. Since capital is the most scarce and costly resource, the focus of risk monitoring and xvi INTRODUCTION risk measurement follows. The central role of risk-based capital in regulations is a major incentive to the development of new tools and management techniques. Undoubtedly a most important innovation of recent years in terms of the modelling ‘toolbox’ is the VaR concept for assessing capital requirements. The VaR concept is a foundation of risk-based capital or, equivalently, ‘economic capital’. The VaR methodology aims at valuing potential losses resulting from current risks and relies on simple facts and principles. VaR recognizes that the loss over a portfolio of transactions could extend to the entire portfolio, but this is an event that has a zero probability given the effective portfolio diversification of banks. Therefore, measuring potential losses requires some rule for defining their magnitude for a diversified portfolio. VaR is the upper bound of losses that should not be exceeded in more than a small fraction of all future outcomes. Management and regulators define benchmarks for this small preset fraction, called the ‘confidence level’, measuring the appetite for risk of banks. Economic capital is VaRbased and crystallizes the quantified present value of potential future losses for making sure that banks have enough capital to sustain worst-case losses. Such risk valuation potentially extends to all main risks. Regulators made the concept instrumental for VaR-based market risk models in 1996. Moreover, even though the New Accord of 2001 falls short of allowing usage of credit models for measuring credit risk capital, it ensures the development of reliable inputs for such models. The Risk Management Toolbox Risk-based practices require the deployment of multiple tools, or models, to meet the specifications of risk management within financial institutions. Risk models value risks and link them to their drivers and to the business universe. By performing these tasks, risk models contribute directly to risk processes. The goal of risk management is to enhance the risk–return profiles of transactions, of business lines’ portfolios of transactions and of the entire bank’s portfolio. Risk models provide these risk–return profiles. The risk management toolbox also addresses other major specifications. Since two risks of 1 add up to less than 2, unlike income and costs, we do not know how to divide a global risk into risk allocations for individual transactions, product families, market segments and business lines, unless we have some dedicated tools for performing this function. The Funds Transfer Pricing (FTP) system allocates income and the capital allocation system allocates risks. These tools provide a double link: • The top-down/bottom-up link for risks and income. • The transversal business-to-financial sphere linkage. Without such links, between the financial and the business spheres and between global risks and individual transaction profiles, there would be no way to move back and forth from a business perspective to a financial perspective and along the chain from individual transactions to the entire bank’s global portfolio. Risk Management Processes Risk management processes are evolving with the gradual emergence of new risk measures. Innovations relate to: INTRODUCTION xvii • The recognition of the need for quantification to develop risk-based practices and meet risk-based capital requirements. • The willingness of bankers to adopt a more proactive view on risks. • The gradual development of regulator guidelines for imposing risk-based techniques, enhanced disclosures on risks and ensuring a sounder and safer level playing field for the financial system. • The emergence of new techniques of managing risks (credit derivatives, new securitizations that off-load credit risk from the banks’ balance sheets) serving to reshape the risk–return profile of banks. • The emergence of new organizational processes for better integrating these advances, such as loan portfolio management. Without risk models, such innovations would remain limited. By valuing risks, models contribute to a more balanced view of income and risks and to a better control of risk drivers, upstream, before they materialize into losses. By linking the business and the risk views, the risk management ‘toolbox’ makes models instrumental for management. By feeding risk processes with adequate risk–return measures, they contribute to enriching them and leveraging them to new levels. Figure 1 shows how models contribute to the ‘vertical’ top-down and bottom-up processes, and how they contribute as well to the ‘horizontal’ links between the risk and return views of the business dimensions (transactions, markets and products, business lines). Risk Models & Tools Credit Risk FTP Capital Allocation Market Risk ALM Others ... Risk Processes Global Policy Business Policy Risk−Return Policy Reporting Risk & Capital Earnings Business Lines FIGURE 1 Comprehensive and consistent set of models for bank-wide risk management xviii INTRODUCTION THE STRUCTURE OF THE BOOK The structure of the book divides each topic into single modular pieces addressing the various issues above. The first section develops general issues, focusing on risks, risk measuring and risk management processes. The next major section addresses sequentially ALM, market risk and credit risk. Book Outline The structure of the book is in 17 sections, each divided into several chapters. This structure provides a very distinct division across topics, each chapter dedicated to a major single topic. The benefit is that it is possible to move across chapters without necessarily following a sequential process throughout the book. The drawback is that only a few chapters provide an overview of interrelated topics. These chapters provide a synthesis of subsequent specific topics, allowing the reader to get both a summary and an overview of a series of interrelated topics. Section Section Section Section Section Section Section Section Section Section Section Section Section Section Section Section Section 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. Banking Risks Risk Regulations Risk Management Processes Risk Models Asset–Liability Management Asset–Liability Management Models Options and Convexity Risk in Banking Mark-to-Market Management in Banking Funds Transfer Pricing Portfolio Analysis: Correlations Market Risk Credit Risk Models Credit Risk: ‘Standalone Risk’ Credit Risk: ‘Portfolio Risk’ Capital Allocation Risk-adjusted Performance Portfolio and Capital Management (Credit Risk) Building Block Structure The structure of the book follows from several choices. The starting point is a wide array of risk models and tools that complement each other and use sometimes similar, sometimes different techniques for achieving the same goals. This raises major structuring issues for ensuring a consistent coverage of the risk management toolbox. The book relies on a building block structure shared by models; some of them extending across many blocks, some belonging to one major block. The structuring by building blocks of models and tools remedies the drawbacks of a sequential presentation of industry models; these bundle modelling techniques within each building block according to the model designers’ assembling choices. By contrast, INTRODUCTION xix a building block structure lists issues separately, and allows us to discuss explicitly the various modelling options. To facilitate the understanding of vendors’ models, some chapters provide an overview of all existing models, while detailed presentations provide an overview of all techniques applying to a single basic building block. Moreover, since model differentiation across risks is strong, there is a need to organize the structure by nature of risk. The sections of the book dealing directly with risk modelling cross-tabulate the main risks with the main building blocks of models. The main risks are interest rate risk, market risk and credit risk. The basic structure, within each risk, addresses four major modules as shown in Figure 2. FIGURE 2 I Risk drivers and transaction risk (standalone) II Portfolio risk III Top-down & bottom-up tools IV Risk & return measures The building block structure of risk models The basic blocks, I and II, are dedicated by source of risk. The two other blocks, III and IV, are transversal to all risks. The structure of the book follows from these principles. Focus The focus is on risk management issues for financial institutions rather than risk modelling applied to financial markets. There is an abundant literature on financial markets and financial derivatives. A basic understanding of what derivatives achieve in terms of hedging or structuring transactions is a prerequisite. However, sections on instruments are limited to the essentials of what hedging instruments are and their applications. Readers can obtain details on derivatives and pricing from other sources in the abundant literature. We found that textbooks rarely address risk management in banking and in financial institutions in a comprehensive manner. Some focus on the technical aspects. Others focus on pure implementation issues to the detriment of technical substance. In other cases, the scope is unbalanced, with plenty of details on some risks (market risk notably) and fewer on others. We have tried to maintain a balance across the main risks without sacrificing scope. The text focuses on the essential concepts underlying the risk analytics of existing models. It does detail the analytics without attempting to provide a comprehensive coverage of each existing model. This results in a more balanced view of all techniques for modelling banking risk. In addition, model vendors’ documentation is available directly from sites dedicated to risk management modelling. There is simply no need to replicate such documents. When developing the analytics, we considered that providing a
- Xem thêm -