BPCE_PILLAR_III_2017

CREDIT RISK Organization of credit risk management

Impairment As indicated above, impairment for credit risk will be equal to 12-month expected credit losses or lifetime expected credit losses, depending on the level of increase in credit risk since initial recognition (Stage 1 or Stage 2 asset). A set of qualitative and quantitative criteria isused to assess the increase in credit risk. A significantincrease in credit risk is measuredindividually,based on reasonable and supportable information, and by comparing the default risk on the financial instrument at the closing date with the default risk on the financial instrument at the date of initial recognition.Any significantincrease in credit risk shall be recognized before the transaction is impaired (Stage3). In order to assess a significant increase in credit risk, the Group applies a process based on rules and criteriawhich apply to all Group entities. For the Individual Customers, Professional Customers and SME loan books, the quantitative criterion is based on the measurement of the change in the 12-month probability of default since initial recognition (probability of default measured as a cycle average). For the Large Corporates, Banks and Specialized Financing loan books, it is based on the change in rating since initial recognition. These quantitative criteria are accompaniedby a set of qualitative criteria, including the existence of a payment more than 30 days past due, the classification of the contract as at-risk, the identification of forbearance exposure or the inclusion of the portfolio on a Watch List. Exposures rated by the Large Corporates, Banks and Specialized Financing software tool are also downgraded to Stage 2 depending on the sector rating and the level of country risk. Financial assets where there is objective evidence of impairmentloss due to an event which represents a counterparty risk and which occurs after their initial recognition will be considered as impaired and will be classified as Stage 3. Identificationcriteria for impaired assets are similar to those under IAS 39 and are aligned with the default criterion. IFRS 9 calls for modifiedcontractualcash flows that are renegotiated or otherwise modified (whether or not as a result of financial hardships), but not subsequentlyderecognized,to be identified. Any profit or loss is recognized as a modificationgain or loss. The gross carrying amount of the financial asset shall be recalculated as the present value of the renegotiatedor modified contractualcash flows that are discountedat the financial asset’s original effective interest rate. The materiality of the modifications is analyzed on a case by case basis. Recognition of loans restructured due to financial hardship will be identical toIAS 39. For Stage 1 and Stage 2 assets, expected credit losses are calculated

used for stress tests. Specific adjustments are made to factor in current conditions and forward-looking macroeconomicprojections: IFRS 9 parametersthereforeaim to provide an accurateestimateof ● losses for accounting provision purposes, whereas prudential parameters are more cautious for regulatory framework purposes. Several of these safety buffers are restated; IFRS 9 parametersmust allow lifetime expected credit losses to be ● estimated, whereas prudential parameters are defined to estimate 12-month expected credit losses. 12-month parameters are thus projected overlong periods; IFRS 9 parametersmust be forward-lookingand reflect forecastsof ● future economic conditions over the estimated period, whereas prudential parameters consist of cycle-average (for PD) or cycle-trough estimates (for LGD and EAD). Prudential parameters are therefore also adjusted to reflect forecasts of future economic conditions. The parametersthus defined allow expectedcredit losses for all rated exposures to be measured, regardless of whether they refer to an entity approvedto use the IRB methodor the standardizedmethodto determine its RWA. Conservative default rules are applied to non-rated exposures. Parameters are adjusted to economic conditions by defining reasonable and supportable economic scenarios, coupled with the probability of occurrence and the calculation of a probable average credit loss. This adjustment calls for the definition of models which link IFRS 9 parameters to a set of economic variables. These models are based on those developed for stress tests. Projections are also based on the budget process. Three economic scenarios (the budget scenario,alongwith optimisticand pessimisticviews of this scenario), coupled with probabilities, are defined over a three-year period to estimate the probable economic loss. The scenarios and weightings are defined using analyses produced by Natixis’ Economic Research department and Management’s expert judgment. Although the majority of the parameters are drawn up by the BPCE and Natixis Risk divisions, other entities including Natixis Financement,BPCE Internationaland certain regional institutionsfor their subsidiaries also contribute to the Group IFRS 9 provisioning system. Moreover, regional institutions are responsible for assessing the consistencyof provisionsdeterminedfor the Group with the local and sector characteristics of their loan books and for defining additional sector provisions if necessary. The mechanism for validating IFRS 9 provisions is fully aligned with the Group’s existing model validation process. Parameters are reviewed by the independent internal model validation unit. This unit’s work is reviewedby the GroupModelingCommittee.Finally,the recommendationsissued by the validation unit are followed up. This process is scheduled to ensure that the main parameters will be reviewed before the first-time application of IFRS 9. In short, the new IFRS 9 provisioningmodel points to an increase in the amount of impairment on loans and securities measured at amortized cost or at fair value through non-recyclableOCI, and on off-balance sheet commitmentsas well as on lease receivables and trade receivables. The calibrationand validation process isongoing and cannotcurrently be disclosed inthe financial statements.

5

as the product of three inputs: probabilityof default(PD); ● loss given default (LGD); ●

exposureat default (EAD) – this dependson contractualcash flows, ● the contract’s effective interest rate and the expected prepayment rate. The Group draws on existing concepts and mechanisms to define these inputs, and in particular on internal models developed to calculate regulatory capital requirements and on projection models

79

Risk Report Pillar III 2017