气候风险管理良好实践

Thisannex summarises the key observations from the thematic examinations andconsultative sessionsrecently conducted by the HKMA,highlighting good practicesidentifiedamongthe participating AIs1tostrengthenclimate risk management frameworks The good practicesspan the risk management cycle,intertwined andcomplementary to oneanother.Thesegood practices demonstrate the following trends observed amongparticipating AIs when managing theirclimate risks: (i) advancing toward a morequantitative-oriented risk management framework; (ii)bridging data gaps to further As international practices and regulatory standards on climate risk management evolve,AIsareencouraged to draw upon these good practices,considering their individual (I)Advancing toward amore quantitative-oriented climate risk management Key observations Allparticipating AIs have established and enhanced their climate risk managementframeworks.These frameworksalign withthe AIs’climate strategiesand reflectthescale,nature,and complexity of their business and operationsin Hong Kong.Theytypicallyincludededicatedpolicies forclimate risk management, whichare integrated into theirexisting risk management frameworksand policies.Climate risksare incorporated into the Good practices 1.Establishingarobustclimate risk limit structuregoverned byaquantitative Mostparticipating AIs have built upon their qualitative climate RAS to establishquantitativemetricsandlimitswithin their climate RAS. These metrics and limits are designed togoverntheirbank-wide climate risk limitstructureeffectively.Commonmetricsand limitsinclude the concentration of credit exposures to high-emitting sectorsinthelending and investment portfolios, the amount of credit exposures with propertycollateral vulnerable to physical risks,andthe size and growth of the green and sustainable UsingRAS limits and metricsas a guide, many participatingAIs also establishappropriatekey risk indicators (KRIs)and key performance indicators (KPIs)toinformtheirbusiness-as-usual (BAU) risk monitoring and managementprocesses.Forexample, some AIs haveimplementedKRIsto trackemissionstrendsof theirtop high-emitting clients,therebyenhancingcounterparty-level monitoring.In addition, some AIs haveemployed KRIsto 2.Underpinning the understanding of climate risk impacts with regular and Some participating AIs conduct regular assessments to refresh their understanding ofhowclimate risksmay affecttraditional risk types. These assessmentsevaluatethe transmissionchannels, materiality, and time horizons of potential impacts on each traditional risk type.Theyconsider risk drivers, including physical risk, transition risk, and other nature-relatedrisks such as biodiversity risk, as well as factorslikepollution, and changes in land use.Inaddition, some AIsincorporatetheir risk mitigation measuresto evaluate theresidual -Multi-dimensional concentration analysisbroken down by sector andgeographyisconducted by some AIs in order togain a deeperunderstanding ofhowclimate risks -Scenario analysis and stress testingareusedby many participating AIs to quantifyand assess the materiality of climate risk impacts. For transition risk, AIsevaluatethe changesin the repayment ability of counterpartiesinhigh-emitting sectors underclimate-focused stressed scenarios. Some AIs also analysesecond-order impactsonand from the counterparties’ upstream and downstream business sectors.Forphysical risk, AIsassessthe impact of physical risk events on their property-related Some participating AIs and/or their banking groups have developed integratedclimate risk stress tests as a BAU tool,and introducedclimate shocksinto the stresstesting programmesfor traditional risk types. For instance,to estimate the impacton liquidity risk, some AIsassumeincreased deposit runoff and usage of committedundrawn credit facilitiesduringclimate events.A few AIs have alsointegrated -Climate risk materiality assessments focused on local operations in Hong Kongareperformed by some participating AIs in addition tothoseconducted by their bankinggroups.Furthermore, some participating AIsactivelycontribute togroup-levelclimate risk materiality assessmentsby participating in discussion, providing local 3.Exploring suitable Fintech solutions with the objective of enhancing the SomeparticipatingAIs havebeen actively exploringFintech solutionstoimprovetheeffectiveness and efficiency ofclimate risk management.Examples of such solutions -UsingGenerative Artificial Intelligence(A.I.)to complete climate or ESGquestionnaires.Some participating AIshavedeployed GenerativeA.I.toassiststaffincompletingtheclimateorESG questionnairesused to inform counterparty-levelclimate risk assessments.The Generative A.I. tool analyses input documents,such -Using HKSAR Government APIs to assess the physical risk of collateral.OneparticipatingAIassesses the physical risk associated with property collateral byleveragingtheHKSAR Government’s APIs (e.g.theLand Parcel and Public UtilityNumber Search API