OperationalRiskChapter 20RiskManagementandFinanciallnstitutions3e,Chapter20,CopyrightJohnC.Hull2012
Risk Management and Financial Institutions 3e, Chapter 20, Copyright © John C. Hull 2012 Operational Risk Chapter 20 1
Definition of Operational Risk“Operational risk is the risk of lossresulting from inadequate or failed internalprocesses, people, and systems, or fromexternaleventsBaselCommitteeJan20012RiskManagementandFinancialInstitutions3e,Chapter20,CopyrightJohnC.Hull2012
Risk Management and Financial Institutions 3e, Chapter 20, Copyright © John C. Hull 2012 Definition of Operational Risk “Operational risk is the risk of loss resulting from inadequate or failed internal processes, people, and systems, or from external events” Basel Committee Jan 2001 2
WhatItIncludes The definition includes people riskstechnology and processing risks, physicalrisks, legal risks, etcThe definition excludes reputation risk andstrategic risk3RiskManagementandFinancialInstitutions3e,Chapter20,CopyrightJohnC.Hull2012
Risk Management and Financial Institutions 3e, Chapter 20, Copyright © John C. Hull 2012 What It Includes ⚫ The definition includes people risks, technology and processing risks, physical risks, legal risks, etc ⚫ The definition excludes reputation risk and strategic risk 3
Regulatory Capital (page 431) In Basel Il there is a capital charge forOperationalRiskThree alternatives:Basic Indicator (15% of annual gross income) Standardized (different percentage for eachbusiness line)Advanced Measurement Approach (AMA)RiskManagementandFinancialInstitutions3e,Chapter20,CopyrightJohnC.Hull20124
Risk Management and Financial Institutions 3e, Chapter 20, Copyright © John C. Hull 2012 Regulatory Capital (page 431) ⚫ In Basel II there is a capital charge for Operational Risk ⚫ Three alternatives: ⚫ Basic Indicator (15% of annual gross income) ⚫ Standardized (different percentage for each business line) ⚫ Advanced Measurement Approach (AMA) 4
CategorizationofBusinessLinesCorporate financeTrading and salesRetail bankingCommercial banking Payment and settlementAgency servicesAsset managementRetail brokerage5RiskManagementandFinancialInstitutions3e,Chapter20,CopyrightJohnC.Hull2012
Risk Management and Financial Institutions 3e, Chapter 20, Copyright © John C. Hull 2012 Categorization of Business Lines ⚫ Corporate finance ⚫ Trading and sales ⚫ Retail banking ⚫ Commercial banking ⚫ Payment and settlement ⚫ Agency services ⚫ Asset management ⚫ Retail brokerage 5
Categorization of risksInternal fraudExternal fraudEmployment practices and workplace safetyClients, products and business practicesDamage to physical assetsBusiness disruption and system failuresExecution, delivery and process management6RiskManagementandFinancialInstitutions3e,Chapter20,CopyrightJohnC.Hull2012
Risk Management and Financial Institutions 3e, Chapter 20, Copyright © John C. Hull 2012 Categorization of risks ⚫ Internal fraud ⚫ External fraud ⚫ Employment practices and workplace safety ⚫ Clients, products and business practices ⚫ Damage to physical assets ⚫ Business disruption and system failures ⚫ Execution, delivery and process management 6
The Task Under AMA Banks need to estimate their exposure toeach combination of type of risk andbusiness lineldeally this will lead to 7 X 8=56 VaRmeasures that can be combined into anoverall VaR measure7RiskManagementandFinancialInstitutions3e,Chapter20,CopyrightJohnC.Hull2012
Risk Management and Financial Institutions 3e, Chapter 20, Copyright © John C. Hull 2012 The Task Under AMA ⚫ Banks need to estimate their exposure to each combination of type of risk and business line ⚫ Ideally this will lead to 7×8=56 VaR measures that can be combined into an overall VaR measure 7
Loss Severity vs LossFrequency (page 434)Lossfrequencyshouldbeestimatedfromthebanksown data as far as possible. One possibility is toassume aPoisson distribution so that weneedonlyestimate anaverage lossfrequency.Probability ofnevents in time T is thene-ar (aT)"n!Loss severity can be based on internal and externalhistorical data. (One possibility is to assume alognormal distribution so that we need only estimatethe mean and SD of losses)RiskManagementandFinancialInstitutions3e,Chapter20,CopyrightJohnC.Hull20128
Risk Management and Financial Institutions 3e, Chapter 20, Copyright © John C. Hull 2012 Loss Severity vs Loss Frequency (page 434) ⚫ Loss frequency should be estimated from the banks own data as far as possible. One possibility is to assume a Poisson distribution so that we need only estimate an average loss frequency. Probability of n events in time T is then ⚫ Loss severity can be based on internal and external historical data. (One possibility is to assume a lognormal distribution so that we need only estimate the mean and SD of losses) ! ( ) n T e n −T 8
Using Monte Carlo to combine theDistributionsS(Figure20.2)3010034bLoss severity (SM)Lossfrequency0.250.20.15nqeqo0.1P0.05268041012Loss (SM)RiskManagementandFinancialInstitutions3e,Chapter20,CopyrightJohnC.Hull20129
Using Monte Carlo to combine the Distributions (Figure 20.2) Risk Management and Financial Institutions 3e, Chapter 20, Copyright © John C. Hull 2012 9
Monte Carlo SimulationTrial Sample from frequency distribution todetermine the number of loss events (=n) Sample n times from the loss severitydistribution to determine the loss severityfor each loss event Sum loss severities to determine total loss10RiskManagementandFinancialInstitutions3e,Chapter20,CopyrightJohnC.Hull2012
Risk Management and Financial Institutions 3e, Chapter 20, Copyright © John C. Hull 2012 Monte Carlo Simulation Trial ⚫ Sample from frequency distribution to determine the number of loss events (=n) ⚫ Sample n times from the loss severity distribution to determine the loss severity for each loss event ⚫ Sum loss severities to determine total loss 10