Stress Testing and other Risk Management Tools - FRM Part 1

Stress testing is a risk management tool that involves analyzing the impacts of the extreme scenarios that are unlikely but feasible. LimitedTimeOffer:Save10%onall2021and2022PremiumStudyPackageswithpromocode:BLOG10  SelectyourPremiumPackage » part-1valuation-and-risk-management StressTesting 20Sep2019 Aftercompletingthisreading,youshouldbeableto: Describetherationalefortheuseofstresstestingasariskmanagementtool. Explainkeyconsiderationsandchallengesrelatedtostresstesting,includingchoiceofscenarios,regulatoryspecifications,modelbuilding,andreversestresstesting. Describetherelationshipbetweenstresstestingandotherriskmeasures,particularlyinenterprise-widestresstesting. DescribestressedVaRandstressedES,includingtheiradvantagesanddisadvantages,andcomparetheprocessofdeterminingstressedVaRandEStothatoftraditionalVaRandES. Describetheresponsibilitiesoftheboardofdirectors,seniormanagement,andtheinternalauditfunctioninstresstestinggovernance. Describetheroleofpoliciesandprocedures,validation,andindependentreviewinstresstestinggovernance. DescribetheBaselstresstestingprinciplesforbanksregardingtheimplementationofstresstesting. Stresstestingisariskmanagementtoolthatinvolvesanalyzingtheimpactsoftheextremescenariosthatareunlikelybutfeasible.Themainquestionforfinancialinstitutionsiswhethertheyhaveadequatecapitalandliquidassetstosurvivestressfultimes.Stresstestingisdoneforregulatorypurposesorforinternalriskmanagementbyfinancialinstitutions.StresstestingcanbecombinedwithmeasurementoftherisksuchastheValue-at-Risk(VaR)andtheExpectedShortfall(ES)togiveadetailedpictureoftherisksfacingafinancialinstitution. Thischapterdealswiththeinternallygeneratedstresstestingscenarios,regulatoryrequirementsofstresstesting,governanceissuesofstresstesting,andtheBaselstresstestingprinciples. RationalefortheUseofStressTestingasaRiskManagementTool Stresstestingservestowarnafirm’smanagementofpotentialadverseeventsarisingfromthefirm’sriskexposureandgoesfurthertogiveestimatesoftheamountofcapitalneededtoabsorblossesthatmayresultfromsuchevents. Stresstestshelptoavoidanyformofcomplacencythatmaycreepinafteranextendedperiodofstabilityandprofitability.Itservestoremindmanagementthatlossescouldstilloccur,andadequateplanshavetobeputinplaceinreadinessforeveryeventuality.Thisway,afirmisabletoavoidissueslikeunderpricingofproducts,somethingthatcouldprovefinanciallyfatal. Stresstestingisakeyriskmanagementtoolduringperiodsofexpansionwhenafirmintroducesnewproductsintothemarket.Theremaybeverylimitedlossdataornoneatall,forsuchproducts,andhypotheticalstresstestinghelpstocomeupwithreliablelossestimates. Underpillar1ofBaselII,stresstestingisarequirementofallbanksusingtheInternalModelsApproach(IMA)tomodelmarketriskandtheinternalratings-basedapproachtomodelcreditrisk.Thesebankshavetoemploystresstestingtodeterminethelevelofcapitaltheyarerequiredtohave. Stresstestingsupplementsotherriskmanagementtools,helpingbankstomitigaterisksthroughmeasuressuchashedgingandinsurance.Byitself,stresstestingcannotaddressallriskmanagementweaknesses,norcanitprovideaone-stopsolution. ComparisonbetweenStressTestingandtheVaRandES RecallthattheVaRandESareestimatedfromalossdistribution.VaRenablesafinancialinstitutiontoconcludewithX%likelihoodthatthelosseswillnotexceedtheVaRlevelduringtimeT.Ontheotherhand,ESenablesthefinancialinstitutionstoconcludewhetherthelossesexceedtheVaRlevelduringagiventimeTandhencetheexpectedlosswillbetheESamount. VaRandESarebackward-looking.Thatis,theyassumethatthefutureandthepastarethesame.ThisisactuallyonedisadvantageofVaRandES.Ontheotherhand,stresstestingisforward-looking.Itasksthequestion,“whatif?”. Whilestresstestinglargelydoesnotinvolveprobabilities,VaR,andESmodelsarefoundedonprobabilitytheory.Forexample,a99.9%VaRcanbeviewedasa1-in-1,000event. Thebackward-lookingESandVaRconsiderawiderangeofscenariosthatarepotentiallygoodorbadtotheorganization.However,stresstestingconsidersarelativelysmallnumberofscenariosthatareallbadfortheorganization. Specifically,forthemarketrisk,VaR/ESanalysisoftentakesashortperiodoftime,suchasaday,whilestresstestingtakesrelativelylongperiods,suchasadecade. Theprimaryobjectiveofstresstestingistocapturetheenterpriseviewoftherisksimpactingafinancialinstitution.ThescenariosusedinthestresstestingareoftendefinedbasedonthemacroeconomicvariablessuchastheunemploymentratesandGDPgrowthrates.Theeffectofthesevariablesshouldbeconsideredinallpartsofaninstitutionwhileconsideringinteractionsbetweendiverseareasofaninstitution. StressedVaRandStressedES ConventionalVaRandESarecalculatedfromdataspanningfromonetofiveyears,whereadailyvariationoftheriskfactorsduringthisperiodisusedtocomputethepotentialfuturemovements. However,inthecaseofthestressedVaRandstressedES,thedataisobtainedfromspecificallystressedperiods(12-monthstressedperiodoncurrentportfoliosaccordingtoBaselrules).Inotherwords,stressedVaRandstressedESgeneratesconditionaldistributionsandconditionalriskmeasures.Assuch,theyareconditionedtoarecurrenceofagivenstressedperiodandthuscanbetakenasahistoricalstresstesting. ThoughstressedVaRandstressedESmightbeobjectivelysimilar,theyaredifferent.TypicallythetimehorizonforthestressedVaR/ESisshort(onetotendays),whileforthestresstesting,itconsidersrelativelylongerperiods. Forinstance,assumethatastressedperiodistheyear2007.ThestressedVaRwouldconcludethatiftherewasarepeatof2007,thenthereisanX%likelihoodthatlossesoveraperiodofTdayswillnotsurpassthestressedVaRlevel.Ontheotherhand,stressedESwouldconcludethatifthelossesoverTdaysdonotexceedthestressedVaRlevel,thentheexpectedlossisthestressedES. However,stresstestingwouldaskthequestions“ifthefollowingyear(2008)isthesameasin2007,willthefinancialinstitutionsurvive?”Alternatively,whatiftheconditionsofthenextyeararetwiceasadverseasthatof2007,willthefinancialinstitutionsurvive?Therefore,stresstestingdoesnotconsidertheoccurrenceoftheworstdaysof2008butrathertheimpactofthewholeyear. ThereisalsoadifferencebetweenconventionalVaRandthestressedVaR.ConventionalVaRcanbeback-testedwhilestressedVaRcannot.Thatis,ifwecancomputeone-dayVaRwith95%confidence,wecangobackanddeterminehoweffectiveitwouldhaveworkedinthepast.Wearenotabletoback-testthestressedVaRoutputanditsresultsbecauseitonlyconsiderstheadverseconditionswhicharegenerallyinfrequent. TypesofScenariosinStressTesting Thebasisofchoosingastresstestingscenarioistheselectionofatimehorizon.Thetimehorizonshouldbelongenoughtoaccommodatethefullanalysisoftheimpactsofscenarios.Longtimehorizonsarerequiredinsomesituations.One-daytoone-weekscenarioscanbeconsidered,butthreemonthstotwo-yearscenariosaretypicallypreferred. Theregulatorsrecommendsomescenarios,butinthissection,wewilldiscussinternallychosenscenarios.Theyincludeusinghistoricalscenarios,stressingkeyvariables,anddevelopingadhocscenariosthatcapturethecurrentconditionsofthebusiness. HistoricalScenarios Historicalscenariosaregeneratedbytheuseofhistoricaldatawhoseallrelevantvariableswillbehaveinthesamemannerasinthepast.Forinstance,variablessuchasinterestratesandcreditratespreadsareknowntorepeatpastchanges.Assuch,actualchangesinthestressedperiodwillbeassumedtorepeatthemselveswhileproportionalvariationswillbeassumedforothers.Agoodexampleofahistoricalscenarioisthe2007-2008UShousingrecession,whichaffectedalotoffinancialinstitutions. Insomecases,amoderatelyadversescenarioismadeworsebymultiplyingvariationsofallriskfactorsbyacertainamount.Forinstance,wecouldmultiplywhathappenedintheloss-makingone-monthperiodandincreasethefrequencyofmovementofallrelevantriskmovementsbyten.Asaresult,thescenariobecomesmoreseveretofinancialinstitutions.However,thisapproachassumeslinearrelationshipsbetweenthemovementsinriskfactors,whichisnotalwaysthecaseduetocorrelationsbetweentheriskfactors. Otherhistoricalscenariosarebasedonone-dayorone-weekoccurrencesofallmarketriskfactors.Sucheventsincludeterroristattacks(suchas9/11terroristattacks)andone-daymassivemovementofinterestrates(suchasonApril10,1992,whenten-yearbondyieldschangedby8.7standarddeviations). StressingKeyVariables Ascenariocouldbebuiltbyassumingthatasignificantchangeoccursinoneormorekeyvariables.Suchchangesinclude: A2%declineintheGDP A25%decreaseinequityprices A100%increaseinallvolatilities A4%increaseintheunemploymentrate A200-basispointincreaseinallinterestrates Someothersignificantvariationscouldoccurinfactorssuchasmoneyexchangerates,pricesofcommodities,anddefaultrates. Inthecaseofthemarketrisk,smallchangesinmeasuredusingtheGreekletters(suchasdeltaandgamma).TheGreekletterscannotbeusedinstresstestingbecausethechangesareusuallylarge.Moreover,Greeksareusedtomeasureriskfromaunitmarketvariableoverashortperiodoftime,whilestresstestingincorporatestheinteractionofthedifferentmarketvariablesoveralongperiodoftime. AdHocStressTests Thestresstestingscenarioswehavebeendiscussingaboveareperformedregularly,afterwhichtheresultsareusedtotestthestabilityofthefinancialstructureofafinancialinstitutionincaseofextremeconditions.However,thefinancialinstitutionsneedtodevelopadhocscenariosthatcapturethecurrenteconomicconditions,specificexposuresfacingthefirm,andupdateanalysisofpotentialfutureextremeevents.Thefirmseithergeneratenewscenariosormodifytheexistingscenariosbasedonpreviousdata. AnexampleofaneventthatwillpromptthefirmstodevelopanadhocscenarioisthechangeinthegovernmentpolicyonanimportantaspectthatimpactsthefinancialinstitutionsorchangeinBaselregulationthatrequiresincrementofthecapitalwithinshortperiodsoftime. Theboards,seniormanagement,andeconomicexpertsusetheirknowledgeinmarkets,globalpolitics,andcurrentglobalinstabilitiestocomewithadversescenarios.Theseniormanagementcarriesoutabrain-stormingevent,afterwhichtheyrecommendnecessaryactionstoavoidunabsorbablerisks. UsingtheStressTestingResults Whilestresstesting,itisvitaltoinvolvetheseniormanagementforittobetakenseriouslyandthususedfordecisionmaking.Thestress-testingresultsarenotonlyusedtosatisfythe“whatif”question,butalsotheBoardandmanagementshouldanalyzetheresultsanddecidewhetheracertainclassofriskmitigationisnecessary.StresstestingmakessurethattheseniormanagementandtheBoarddonotbasetheirdecision-makingonwhatismostlikelytohappen,butalsoconsiderotheralternativeslesslikelytohappenthatcouldhaveadramaticresultonthefirm. ModelBuilding Itispossibletoseehowthemajorityoftherelevantriskfactorsbehaveinastressedperiodwhilebuildingascenario,afterwhichtheimpactofthescenarioonthefirmisanalyzedinanalmostdirectmanner.However,scenariosgeneratedbystressingkeyvariablesandadhocscenarioscapturethevariationsofafewkeyriskfactorsoreconomicvariables.Therefore,inordertoexhaustthescenarios,itisnecessarytobuildamodeltodeterminehowthe“leftout”variablesareexpectedtobehaveinastressedmarketcondition.Thevariablesstatedinthecontextofthestresstestingaretermedascorevariables,whiletheremainingvariablesaretermedasperipheralvariables. Onemethodisperforminganalysis,suchasregressionanalysis,torelatetheperipheralvariablestothecorevariables.Notethatthevariablesarebasedonthestressedeconomicconditions.Usingthedataofthepaststressedperiodsismostefficientindeterminingappropriaterelationships. Forexample,incaseofthecreditrisklosses,datafromtheratingagencies,suchasdefaultrates,canbelinkedtoaneconomicvariablesuchasGDPgrowthrate.Afterward,generaldefaultratesexpectedinvariousstressedperiodsaredetermined.Theresultscanbemodified(scaledupordown)todeterminethedefaultratefordifferentloansorfinancialinstitutions.Notethatthesameanalysiscanbedonetotherecoveryratestodeterminelossrates. TheKnock-OnEffects Apartfromtheimmediateimpactsofascenario,therearealsoknock-oneffectsthatreflecthowfinancialinstitutionsrespondtoextremescenarios.Initsresponse,afinancialinstitutioncanmakedecisionsthatcanfurtherworsenalreadyextremeconditions. Forinstance,duringthe2005-2006UShousingpricebubble,bankswereconcernedwiththecreditqualityofotherbanksandwerenotreadytoengageininterbanklending,whichmadefundingcostsforbanksrise. ReverseStressTesting Recallthatstresstestinginvolvesgeneratingscenariosandthenanalyzingtheireffects.Reversestresstesting,asthenamesuggests,takestheoppositedirectionbytryingtoidentifycombinationsofcircumstancesthatmightleadfinancialinstitutionstofail. Byusinghistoricalscenarios,afinancialinstitutionidentifiespastextremeconditions.Then,thebankdeterminesthelevelatwhichthescenariohastobeworsethanthehistoricalobservationtocausethefinancialinstitutiontofail.Forinstance,afinancialinstitutionmightconcludethattwicethe2005-2006UShousingbubblewillmakethefinancialinstitutiontofail.However,thiskindofreversestresstestingisanapproximation.Typically,afinancialinstitutionwillusecomplicatedmodelsthattakeintoconsiderationcorrelationsbetweendifferentvariablestomakethemarketconditionsmorestressed. Findinganappropriatecombinationofriskfactorsthatleadthefinancialinstitutiontofailisachallengingfeat.However,aneffectivemethodistoidentifysomeofthecriticalfactorssuchasGDPgrowthrate,unemploymentrates,andinterestratevariations,thenbuildamodelthatrelatesallotherappropriatevariablestothesekeyvariables.Afterthat,possiblefactorcombinationsthatcanleadtofailurearesearchediteratively. RegulatoryStressTesting US,UK,andEUregulatorsrequirebanksandinsurancecompaniestoperformspecifiedstresstests.IntheUnitedStates,theFederalReserveperformsstresstestsofallthebankswhoseconsolidatedassetsareoverUSD50billion.ThistypeofstresstestistermedasComprehensiveCapitalAnalysisandReview(CCAR).UnderCCAR,thebanksarerequiredtoconsiderfourscenarios: BaselineScenario AdverseScenario Scenario AninternalScenario ThebaselinescenarioisbasedontheaverageprojectionsfromthesurveysoftheeconomicpredictorsbutdoesnotrepresenttheprojectionoftheFederalReserve. Theadverseandtheseverelyadversescenariosdescribehypotheticalsetsofeventswhicharestructuredtotestthestrengthofbankingorganizationsandtheirresilience.Eachoftheabovescenariosconsistsofthe28variables(suchastheunemploymentrate,stockmarketprices,andinterestrates)whichcapturesdomesticandinternationaleconomicactivityaccompaniedbytheBoardexplanationontheoveralleconomicconditionsandvariationsinthescenariosfromthepastyear. Banksarerequiredtosubmitacapitalplan,justificationofthemodelsused,andtheoutcomesoftheirstresstesting.Ifabankfailstostresstestduetoinsufficientcapital,thebankisrequiredtoraisemorecapitalwhilerestrictingthedividendpaymentuntilthecapitalhasbeenraised. BankswithconsolidatedassetsbetweenUSD10millionandUSD50millionareundertheDodd-FankActStressTest(DFAST).ThescenariosintheDFASTaresimilartothoseintheCCAR.However,intheDFAST,banksarenotrequiredtoproduceacapitalplan. Therefore,throughstresstests,regulatorscanconsistentlyevaluatethebankstodeterminetheirabilitytoextremeeconomicconditions.However,theyrecommendthatbanksdeveloptheirscenarios. ResponsibilitiesoftheBoardofDirectors,SeniorManagementandtheInternalAuditFunctioninStressTestingActivities Foreffectiveoperationofstresstesting,theBoardofdirectorsandseniormanagementshouldhavedistinctresponsibilities.What’smore,thereshouldbesomesharedresponsibilities,althoughafewrolescanbesetasideexclusivelyforoneofthetwogroups. ResponsibilitiesoftheBoardofDirectors ThebuckstopswiththeBoard:TheBoardofdirectorsis“ultimately”responsibleforafirm’sstresstests.Evenifboardmembersdonotimmersethemselvesinthetechnicaldetailsofstresstests,theyshouldensurethattheystaysufficientlyknowledgeableaboutstress-testingproceduresandinterpretationofresults. Continuousinvolvement:Boardmembersshouldregularlyreceivesummaryinformationonstresstests,includingresultsfromeveryscenario.Membersshouldthenevaluatetheseresultstoensuretheytakeintoaccountthefirm’sriskappetiteandoverallstrategy. Continuousreview:Boardmembersshouldregularlyreviewstresstestingreportswithaviewtonotjustcritickeyassumptionsbutalsosupplementtheinformationwiththeirviewsthatbetterreflecttheoverallgoalsofthefirm. Integratingstresstestingresultsindecisionmaking:TheBoardshouldmakekeydecisionsoninvestment,capital,andliquiditybasedonstresstestresultsalongwithotherinformation.Whiledoingthis,theBoardshouldproceedwithacertainlevelofcautionincognizanceofthefactthatstresstestsaresubjecttoassumptionsandahostoflimitations. Formulatingstress-testingguidelines:It’stheresponsibilityoftheBoardtocomeupwithguidelinesonstresstesting,suchastherisktolerancelevel(riskappetite). ResponsibilitiesofSeniorManagement Implementationoversight:SeniormanagementhasthemandatetoensurethatstresstestingguidelinesauthorizedbytheBoardareimplementedtotheletter.ThisinvolvesestablishingpoliciesandproceduresthathelptoimplementtheBoard’sguidelines. RegularlyreportingtotheBoard:SeniormanagementshouldkeeptheBoardup-to-dateonallmatterstodowithstresstesting,includingtestdesigns,emergingissues,andcompliancewithstress-testingpolicies. CoordinatingandIntegratingstresstestingacrossthefirm:Membersofseniormanagementareresponsibleforpropagatingwidespreadknowledgeonstresstestsacrossthefirm,makingsurethatalldepartmentsunderstanditsimportance. Identifyinggreyareas:Seniormanagementshouldseektoidentifyinconsistencies,contradictions,andpossiblegapsinstressteststomakeimprovementstothewholeprocess. Ensuringstresstestshaveasufficientrange:InconsultationswiththeBoardofdirectors,seniormanagementhastoensurethatstresstestingactivitiesaresufficientlyseveretogaugethefirm’spreparationforallpossiblescenarios,includinglow-frequencyhigh-impactevents. Usingstressteststoassesstheeffectivenessofriskmitigationstrategies:Stresstestsshouldhelpthemanagementtoassessjusthoweffectiveriskmitigationstrategiesare.Ifsuchstrategiesareeffective,significantlysevereeventswillnotcausesignificantfinancialstrain.Ifthetestspredictsignificantfinancialturmoil,itcouldbethatthehedgingstrategiesadoptedareineffective. Updatingstressteststoreflectemergingrisks:Astimegoes,aninstitutionwillgraduallygainexposuretonewrisks,eitherasaresultofmarket-widetrendsoritsinvestmentactivities.Itistheresponsibilityofseniormanagementtodevelopnewstress-testingtechniquesthatreflecttheinstitution’snewriskprofile. RoleoftheInternalAudit  Internalauditshould: Independentlyevaluatetheperformance,integrity,andreliabilityofstress-testingactivities; Ensurethatstresstestsacrosstheorganizationareconductedinasoundmannerandremainrelevantintermsofthescenariostested; Assesstheskillsandexpertiseofthestaffinvolvedinstress-testingactivities; Checkthatapprovedchangestostress-testingpoliciesandproceduresareimplementedandappropriatelydocumented; Evaluatetheindependentreviewandvalidationexercises; Toaccomplishalltheabove,internalauditstaffmustbewellqualified.Theyshouldbewell-groundedinstress-testingtechniquesandtechnicalexpertisetobeabletodifferentiatebetweenexcellentandinappropriatepractices. TheRoleofPoliciesandProcedures,Validation,andIndependentReviewinStressTestingGovernance PoliciesandProcedures Afinancialinstitutionshouldsetoutclearlystatedandunderstandablepoliciesandproceduresgoverningstresstesting,whichmustbeadheredto.Thepoliciesandproceduresensurethatthestresstestingofpartsofafinancialinstitutionconvergestothesamepoint. Thepoliciesandproceduresshouldbeableto: Explainthepurposeofstresstesting; Describetheproceduresofstresstesting; Statethefrequencyatwhichthestresstestingcanbedone; Describetherolesandresponsibilitiesofthepartiesinvolvedinstresstesting; Provideanexplanationoftheprocedurestobefollowedwhilechoosingthescenarios; Describehowtheindependentreviewsofthestresstestingwillbedone; Givecleardocumentationonstresstestingtothirdparties(e.g.,regulators,externalauditors,andratingagencies); Explainhowtheresultsofthestresstestingwillbeusedandbywhom; Theywereamendedasthestresstestingpracticeschangesasthemarketconditionschange; Accommodatetrackingofthestresstestresultsastheychangethroughtime;and Documenttheactivitiesofmodelsandthesoftwareacquiredfromthevendorsorotherthirdparties. ValidationandIndependentReviewGovernance Thestresstestinggovernancecoverstheindependentreviewprocedures,whichareexpectedtobeunbiasedandprovideassurancetotheboardthatstresstestingiscarriedoutwhilefollowingthefirm’spoliciesandprocedures.Financialinstitutionsusediversemodelsthataresubjecttoindependentreviewtomakesurethattheyservetheintendedpurpose. Validationandindependentreviewshouldinvolvethefollowing: Ensuringthatvalidationandindependentreviewareconductedonanongoingbasis; Ensuringthatsubjectiveorqualitativeaspectsofastresstestarealsovalidatedandreviewed,eveniftheycannotbetestedinquantitativeterms; Acknowledginglimitationsinstresstesting; Ensuringthatstress-testingstandardsareupheld; Acknowledgingdataweaknessesorlimitations,ifany; Ensuringthatthereissufficientindependenceinbothvalidationandreviewofstresstests; Ensuringthatthird-partymodelsusedinstress-testingactivitiesarevalidatedandreviewedtodetermineiftheyarefitforthepurposeathand; Ensuringthatstresstestsresultsareimplementedrigorously,andverifyingthatanydeparturefromtherecommendedactionsisbackedupbysolidreasons. BaselStress-TestingPrinciples TheBaselCommitteeemphasizesthatstresstestingisacrucialaspectbyrequiringthatthemarketriskcalculationsarebasedontheinternalVaRandtheExpectedShortfall(ES)models,whichshouldbeaccompaniedby“rigorousandcomprehensive”stresstesting.Moreover,banksthatusetheinternalratingapproachoftheBaselIItocalculatethecreditriskcapitalshouldperformastresstesttoevaluatethestrengthoftheirassumptions. Influencedbythe2007-2008financialcrisis,theBaselCommitteepublishedtheprinciplesofstress-testingforthebanksandcorrespondingsupervisors.TheoverarchingemphasisoftheBaselcommitteewastheimportanceofstresstestingindeterminingtheamountofcapitalthatwillcushionbanksagainstlossesduetolargeshocks. Therefore,theBaselcommitteerecognizedtheimportanceofstresstestingin: Givingaforward-lookingperspectiveontheevaluationofrisk; Overcomingthedemeritsofmodesandhistoricaldata; Facilitatingthedevelopmentofriskmitigation,oranyotherplanstoreducerisksindifferentstressedconditions; Assistinginternalandexternalcommunications; Supportingthecapitalandliquidityplanningprocedures;and Notifyingandsettingofrisktolerance. WhentheBaselcommitteeconsideredthestresstestsdonebefore2007-2008,theyconcludedthat: ItiscrucialtoinvolvetheBoardandtheseniormanagementinstresstesting.TheBoardandtheseniormanagementshouldbeinvolvedinstresstestingaspectssuchaschoosingscenarios,settingstresstestingobjectives,analysisofthestresstestingresults,determiningthepotentialactions,andstrategicdecisionmaking.Duringthecrisis,banksthathadseniormanagementinterestedindevelopingastresstest,whicheventuallyaffectedtheirdecision-making,performedfairlywell. Theapproachesofthestress-testingdidnotgiveroomfortheaggregationofdifferentexposuresindifferentpartsofabank.Thatis,expertsfromdifferentpartsofthebankdidnotcooperatetoproduceanenterprise-wideriskview. Thescenarioschoseninthestresstestsweretoomoderateandwerebasedonashortperiodoftime.Thepossiblecorrelationsbetweendifferentrisktypes,products,andmarketswereignored.Assuch,thestresstestreliedonthehistoricalscenariosandleftoutrisksfromnewproductsandpositionstakenbythebanks. Someoftheriskswerenotconsideredcomprehensivelyinthechosenscenarios.Forexample,counterpartycreditrisk,risksrelatedtostructuredproducts,andproductawaitingsecuritizationswerepartiallyconsidered.Moreover,theeffectofthestressedscenarioonliquiditywasunderrated. BaselCommitteeStressTestingPrinciples AccordingtotheBaselCommitteeonBankingSupervision’s“StressTestingPrinciples”publishedinDecember2017: Stresstestingframeworksshouldincorporateaneffectivegovernancestructure. Thestresstestingframeworksshouldinvolveagovernancestructurethatisclear,documented,andcomprehensive.Therolesandresponsibilitiesofseniormanagement,oversightbodies,andthoseconcernedwithstresstestingoperationsshouldbeclearlystated. Thestresstestingframeworkshouldincorporateacollaborationofallrequiredstakeholdersandtheappropriatecommunicationtostakeholdersofthestresstestingmethodologies,assumptions,scenarios,andresults. Stresstestingframeworksshouldhaveclearlyarticulatedandformallyadoptedobjectives. ThestresstestingframeworksshouldsatisfytheobjectivesthataredocumentedandapprovedbytheBoardofanorganizationoranyotherseniorgovernance.Theobjectiveshouldbeabletomeettherequirementsandexpectationsoftheframeworkofthebankanditsgeneralgovernancestructure.Thestaffmandatedtocarryoutstresstestingshouldknowthestresstestingframework’sobjectives. Stresstestingframeworksshouldcapturematerialandrelevantrisksandapplysufficientlyseverestresses. Stresstestingshouldreflectthematerialandrelevantriskdeterminedbyarobustriskidentificationprocessandkeyvariableswithineachscenariothatisinternallyconsistent.Anarrativeshouldbedevelopedexplainingascenariothatcapturesrisks,andthoserisksthatareexcludedbythescenarioshouldbedescribedclearlyandwelldocumented. Stresstestingshouldbeutilizedasariskmanagementtoolandtoconveybusinessdecisions. Stresstestingistypicallyaforward-lookingriskmanagementtoolthatpotentiallyhelpsabankinidentifyingandmonitoringrisk.Therefore,stresstestingplaysaroleintheformulationandimplementationofstrategicandpolicyobjectives.Whenusingstresstestingresults,banksandauthoritiesshouldcomprehendcrucialassumptionsandlimitationssuchastherelevanceofthescenario,modelrisks,andriskcoverage.Lastly,stresstestingasariskmanagementtoolshouldbedoneregularlyinaccordancewithawell-developedschedule(exceptadhocstresstests).Thefrequencyofastresstestdependson: Theobjectiveofthestresstestingframework; Thesizeandcomplexityofthefinancialinstitution;and Changesinthemacroeconomicenvironment. Resourcesandorganizationalstructuresshouldbeadequatetomeettheobjectivesofthestresstestingframework. Stresstestingframeworksshouldhaveadequateorganizationalstructuresthatmeettheobjectivesofthestresstest.Thegovernanceprocessesshouldensurethattheresourcesforstresstestingareadequate,suchthattheseresourceshaverelevantskillsetstoimplementtheframework. StresstestsshouldbesupportedbyaccurateandsufficientlygranulardataandrobustITsystems. Stresstestsidentifyrisksandproducereliableresultsifthedatausedisaccurateandcomplete,andavailableatanadequatelygranularlevelandontime.Banksandauthoritiesshouldestablishasounddatainfrastructurewhichiscapableofretrieving,processing,andreportingofinformationusedinstresstests.Thedatainfrastructureshouldbeabletoprovideadequatequalityinformationtosatisfyandobjectivesofthestresstestingframework.Moreover,structuresshouldbeputinplacetocoveranymaterialinformationdeficiencies. Modelsandmethodologiestoassesstheimpactsofscenariosandsensitivitiesshouldbefitforpurpose. Themodelsandmethodologiesutilizedinstresstestingshouldservetheintendedpurpose.Therefore, Thereshouldbeanadequatedefinitionofcoverage,segmentation,andgranularityofthedataandthetypesofrisksbasedontheobjectivesofthestresstestframework.Allisdoneatthemodelingstage; Thecomplexityofthemodelsshouldberelevanttoboththeobjectivesofthestresstestingandtargetportfoliosbeingassessedusingthemodels;and Themodelsandthemethodologiesinastresstestshouldbeadequatelyjustifiedanddocumented. Themodelbuildingshouldbeacollaborativetaskbetweenthedifferentexperts.Assuch,themodelbuildersengagewithstakeholderstogainknowledgeonthetypeofrisksbeingmodeledandunderstandthebusinessgoals,businesscatalysts,riskfactors,andotherbusinessinformationrelevanttotheobjectivesofthestresstestingframework. Stresstestingmodels,results,andframeworksshouldbesubjecttochallengeandregularreview. Periodicreviewandchallengeofstresstestingforthefinancialinstitutionsandtheauthoritiesisimportantinimprovingthereliabilityofthestresstestingresults,understandingofresults’limitations,identifyingtheareasthatneedimprovementandensuringthattheresultsareutilizedinaccordancewiththeobjectivesofthestresstestingframework. Stresstestingpracticesandfindingsshouldbecommunicatedwithinandacrossjurisdictions. Communicatingthestresstestingresultstoappropriateinternalandexternalstakeholdersprovidesessentialperspectivesonrisksthatwouldbeunavailabletoanindividualinstitutionorauthority.Furthermore,disclosureofthestresstestresultsbybanksorauthoritiesimprovesthemarketdisciplineandmotivatestheresilienceofthebankingsectortowardsidentifiedstress. Banksandauthoritieswhochoosetodisclosestresstestingresultsshouldensurethatthemethodofdeliveryshouldmaketheresultsunderstandablewhileincludingthelimitationsandassumptionsonwhichthestresstestisbased.Clearconveyanceofstresstestresultspreventsinappropriateconclusionsontheresilienceofthebankswithdifferentresults. Question1 HardikandSimriticompareandcontraststresstestingwitheconomiccapitalandvalueatriskmeasures.Whichofthefollowingstatementsregardingdifferencesbetweenthetwotypesofriskmeasuresismostaccurate? A.Stressteststendtocalculatelossesfromtheperspectiveofthemarket,whileEC/VaRmethodscomputelossesbasedonanaccountingpointofview B.Whilestresstestsfocusonunconditionalscenarios,EC/VaRmethodsfocusonconditionalscenarios C.Whilestresstestsexaminealongperiod,typicallyspanningseveralyears,ECmodelsfocusonlossesatagivenpointintime,say,thelossinvalueattheendofyear\(t\). D.StressteststendtousecardinalprobabilitieswhileEC/VaRmethodsuseordinalarrangements ThecorrectanswerisC. OptionAisinaccurate:Stressteststendtocalculatelossesfromtheperspectiveofaccounting,whileEC/VaRmethodscomputelossesbasedonamarketpointofview. OptionBisinaccurate:Whilestresstestsfocusonconditionalscenarios,EC/VaRmethodsfocusonunconditionalscenarios. OptionDisalsoinaccurate:Stresstestsdonotfocusonprobabilities.Instead,theyfocusonordinalarrangementslike“severe,”“moresevere,”and“extremelysevere.”EC/VaRmethods,ontheotherhand,focusoncardinalprobabilities.Forinstance,a95%VaRlosscouldbeinterpretedas5-in-100events. Question2 OneoftheapproachesusedtoincorporatestresstestinginVaRinvolvestheuseofstressedinputs.Whichofthefollowingstatementsmostaccuratelyrepresentsagenuinedisadvantageofrelyingonriskmetricsthatincorporatestressedinputs? A.Themetricsareusuallymoreconservative(lessaggressive) B.Themetricsareusuallylessconservative(moreaggressive) C.Thecapitalsetaside,asinformedbytheriskmetrics,islikelytobeinsufficient D.Theriskmetricsprimarilydependonportfoliocompositionandarenotresponsivetoemergingrisksorcurrentmarketconditions. ThecorrectanswerisD. Themostcommondisadvantageofusingstressedriskmetricsisthattheydonotrespondtocurrentissuesinthemarket.Assuch,significantshocksinthemarketcan“catchthefirmunaware”andresultinextensivefinancialturmoil. Question3 SarahWayne,FRM,worksatCapitalBank,basedintheU.S.Thebankownsaportfolioofcorporatebondsandalsohassignificantequitystakesinseveralmedium-sizecompaniesacrosstheUnitedStates.Shewasrecentlyrequestedtoheadariskmanagementdepartmentsubcommitteetaskedwithstresstesting.Theaimistoestablishhowwellpreparedthebankisfordestabilizingevents.Whichofthefollowingscenarioanalysisoptionswouldbethebestforthepurposeathand? A.Hypotheticalscenarioanalysis B.Historicalscenarioanalysis C.Forward-lookinghypotheticalscenarioanalysisandhistoricalscenarioanalysis D.Cannottellbasedonthegiveninformation ThecorrectanswerisC. Scenarioanalysesshouldbedynamicandforward-looking.Thisimpliesthathistoricalscenarioanalysisandforward-lookinghypotheticalscenarioanalysisshouldbecombined.Purehistoricalscenarioscangivevaluableinsightsintoimpactbutcanunderestimatetheconfluenceofeventsthatareyettooccur.What’smore,historicalscenarioanalysesarebackward-lookingandhenceneglectrecentdevelopments(riskexposures)andcurrentvulnerabilitiesofaninstitution.Assuch,scenariodesignshouldtakeintoaccountbothspecificandsystematicchangesinthepresentandnearfuture. Question4 Seniormanagementshouldberesponsibleforwhichofthefollowingtasks? Ensuringthatstresstestingpoliciesandproceduresarefollowedtotheletter Assessingtheskillsandexpertiseofthestaffinvolvedinstress-testingactivities Evaluatingtheindependentreviewandvalidationexercises Makingkeydecisionsoninvestment,capital,andliquiditybasedonstresstestresultsalongwithanyotherinformationavailable. Propagatingwidespreadknowledgeonstresstestsacrossthefirm,andmakingsurethatalldepartmentsunderstanditsimportance A.I,II,andIV B.IandV C.IIIandIV D.Vonly ThecorrectanswerisB. RolesIIandIIIbelongtointernalaudit.RoleIVbelongstotheboardofdirectors. × Share Copy Addedtoclipboard × Featured SwapsPrinciplesforSoundStressTesting–PracticesandSupervisionCountryRisk:Determinants,Measures,andImplicationsViewMore StudywithUs CFA®ExamandFRM®ExamPrepPlatformofferedbyAnalystPrep StudyPlatform LearnwithUs Subscribetoournewsletterandkeepupwiththelatestandgreatesttipsforsuccess OnlineTutoring Ourvideosfeatureprofessionaleducatorspresentingin-depthexplanationsofalltopicsintroducedinthecurriculum. VideoLessons DanielGlyn2021-03-24IhavefinishedmyFRM1thankstoAnalystPrep.AndnowusingAnalystPrepformyFRM2preparation.ProfessorForjanisbrilliant.Hegivessuchgoodexplanationsandanalogies.Andmorethananythingmakeslearningfun.AbigthankyoutoAnalystprepandProfessorForjan.5starsalltheway!michaelwalshe2021-03-18ProfessorJames'videosareexcellentforunderstandingtheunderlyingtheoriesbehindfinancialengineering/financialanalysis.TheAnalystPrepvideoswerebetterthananyoftheothersthatIsearchedthroughonYouTubeforprovidingaclearexplanationofsomeconcepts,suchasPortfoliotheory,CAPM,andArbitragePricingtheory.WatchingtheseclearedupmanyoftheunclaritiesIhadinmyhead.Highlyrecommended.NykaSmith2021-02-18EveryconceptisverywellexplainedbyNilayArun.kudostoyouman!BadrMoubile2021-02-13Veryhelpfull!AgustinOlcese2021-01-27Excellentexplantions,veryclear!JaakJay2021-01-14Awesomecontent,kudostoProf.JamesFrojansindhushreereddy2021-01-07CrispandshortpptofFrmchaptersandgreatexplanationwithexamples.Trustpilotratingscore:4.7of5,basedon61reviews. PreviousPostMeasuresofFinancialRisk NextPostApplyingDuration,Convexity,andDV01 RelatedPosts part-1quantitative-analysis Oct20,2019 RegressionwithaSingleRegressor... Aftercompletingthisreadingyoushouldbeableto:Calculateandinterpretconfidence...ReadMore part-1quantitative-analysis Aug19,2019 RegressionwithMultipleExplanatoryV... Aftercompletingthisreading,youshouldbeableto:Distinguishbetweentherelative...ReadMore part-1quantitative-analysis Mar20,2019 CorrelationsandCopulas Aftercompletingthisreadingyoushouldbeableto:Definecorrelationandcovariance...ReadMore part-1valuation-and-risk-management Sep05,2019 ModelingandHedgingNon-ParallelTerm... Aftercompletingthisreadingyoushouldbeableto:Describetheprincipalcomponents...ReadMore LeaveaCommentCancelreplyYoumustbeloggedintopostacomment. X StartstudyingforFRMorSOAexamsrightaway! Registerforfree