BPCE_REGISTRATION_DOCUMENT_2017

FINANCIAL REPORT IFRS Consolidated Financial Statements of Groupe BPCE as at December 31, 2017

LEVEL 2: VALUATION USING OBSERVABLE MARKET INPUTS Level 2 fair value comprises instruments – other than instruments mentioned in Level 1 fair value – measured using a valuation technique incorporating inputs that are either directly observable (prices) or indirectly observable (price derivatives) through to maturity. Thismainlyincludes: Simple instruments Most over-the-counterderivatives, swaps, credit derivatives, forward rate agreements, caps, floors and plain vanilla options are traded in activemarkets, i.e. liquidmarkets inwhich trades occurregularly. These instruments are valued using generally accepted models (discounted cash flow method, Black & Scholes model, interpolation techniques), andon the basisof directly observable inputs. For these instruments,the extent to which models are used and the observability of inputshas been documented. Instruments measuredusing Level 2 inputsalso include: securitiesthat are less liquid than those classifiedas Level ● fair value is determinedbased on external prices put forward by a reasonablenumberof activemarketmakers and which are regularly observable without necessarily being directly executable (prices mainly taken from contribution and consensus databases); where these criteria are not met, the securities are classified as Level 3 fair value; securities not quoted on an active market whose fair value is ● determined based on observable market data (for example, using market data for listed peers or the earningsmultiplemethod based on techniques widely used in the market); Greek sovereign securities, whose fair value is recorded under ● Level 2 given thewide bid-ask price spread on marketprices; shares of UCITS whose NAV is not determined and published on a ● daily basis but is subject to regular reporting or which offer observabledata from recent transactions; debt securitiesdesignatedat fair value, mainly by Natixis, and to a ● lesser extent Crédit Foncier. The methodology used by Natixis to value the “issuer credit risk” componentof issues designatedat fair value is based on the discounting of future cash flows using directly observable inputs such as yield curves and revaluation differences.For each issue, this valuationrepresentsthe product of the notional amount outstanding and its sensitivity, taking into account the existence of calls and the difference between the revaluation spread (based on the BPCE cash reoffer curve at December 31, 2017 as for previous closing dates) and the average issue spread. Changes in own credit risk are generally not material for issueswith an initial maturity of less than one year. Complex instruments Certain hybrid and/or long-maturity financial instruments are measured using a recognized model on the basis of market inputs derived from observable data such as yield curves, implied volatility layers of options, market consensus data or active over-the-counter markets. 1, whose

The main models for determiningthe fair value of these instruments are described below by type of product: equity products: complexproducts are valued using: ● market data, - a payoff, i.e. the formulaof positiveor negativeflows attachedto - the product at maturity, a modelof changesin the underlying asset. - These products can have a single underlying, multiple underlyings or hybrids (fixed income/equity for example). The main models used for equity products are local volatility models, local volatility combined with Hull & White 1 factor (H&W1F), Tskew and Pskew. The local volatilitymodel treats volatility as a function of time and the price of the underlying.Its main propertyis that it considersthe implied volatility of the option (derived from market data) relative to its exercise price. The local volatility hybrid model, paired with the H&W1F, consists of pairing the local volatility model described above with a Hull & White 1 factor type fixed-income model, described below (see fixed-income products). The Tskew model is a valuation model for mono and multi-underlying options. Its principle is to calibrate the distribution of the underlying asset or assets at maturity to standard option prices. The Pskew model is similar to the Tskew model. It is used in particular for simple ratchet equity products such as capped or floored ratchet products; fixed income products: fixed income products generally have ● specific characteristics which justify the choice of model. Underlying risk factors associated with the payoff are taken into account. The main models used to value and manage fixed-incomeproducts are Hull & White models (one-factor and two-factor models or one-factor Hull & White stochastic volatility model), the Hunt Kennedy model and the “smiled” BGM model. The Hull & White models are simple pricingmodels for plain vanilla fixed-incomeproductsand can be calibratedeasily. Productsvalued using these models generallycontaina Bermudan-typecancellation option ( i.e. one that may be exercised at certain dates set at the beginning of the contract). SBGM and Hunt Kennedy models are used to value fixed-income products that are sensitive to volatility smiles ( i.e. implied change in volatilityrelative to the exerciseprice) and to autocorrelation(or correlation between interest rates); foreign exchange products: foreign exchange products generally ● have specific characteristics which justify the choice of model. The main models used to value and manage foreign-exchange productsare local and stochasticvolatilitymodels,as well as hybrid models, which combine modeling of the underlying foreign exchange with two Hull & White 1 factor models to ascertain the fixed-income factors.

5

263

Registration document 2017