Impairment Calculation

The iFRS-VBox is designed to perform estimation of impairment of financial assets and provisions for off-balance sheet credit exposures using individual as well as portfolio (collective) methods. Furthermore the system is designed to guarantee a smooth migration from IAS 36/39 (incurred loss model) to IFRS 9 (expected loss model).

Impairment overview

Identification and Measurement of Individual Impairment

Individual impairment is calculated based on the estimated future recovery cash flows (both from repayment of loan and/or proceeds from collateral realization). For significant assets, the estimation of the recovery cash flows is done on a case-by-case basis (RCF Approach). For non-significant assets which nevertheless need to be impaired specifically, the estimation is based using a Lump Sum Approach.

The rate used in the discounting of the cash flows is the original EIR. The impairment is the difference between the carrying amount of the financial asset and recoverable amount (discounted recovery cash-flows).

Exposures that are individually assessed for impairment may be identified automatically using various parameters like ‘materiality threshold’, the customer or product type, the past due duration, or can be manually flagged by the user through GUI. The system automatically calculates impairment provision as well as income adjustments on impaired assets.

Unwinding

For IFRS purposes, the interest income on individually impaired loans is calculated using the same original EIR but the (lower) recovery cash flows (Unwinding). This implies that the bank needs separately the interest according to the repayment schedule and the accounting interest computed based on the IFRS method. The unwinding calculation is based on the last available and valid provision calculation. The system assures the calculation of provisions and unwinding on the single contract level and automatically handles for the events related to repayments of exposure or/and change of expected recoveries.

Identification and Measurement of Collective Impairment

All the remaining loans in the portfolio not included in the individual assessment are considered for collective impairment. This is also the case of loans individually impaired but which resulted in a nil provision.

Collective impairment is calculated by dividing the loan portfolio into groups of loans constituting homogeneous portfolios considering their credit risk profile. The calculation is based on portfolio definitions and historical parameters using a range of available statistical analysis.

Credit risk parameters (PD, LGD, CCF)

iFRS-VBox applies in the calculation of collective impairment parameters like:

  • Probability of Default (PD).
  • Loss Given Default (LGD).
  • Credit Conversion Factor (CCF) for off-balance sheet exposures.
  • Available collateral information (if not already included in the LGD).
  • Loss Identification Period (LIP, if not already included in PD, for IAS 39 only).

The bank has several options with respect to the determination of the above credit risk parameters:

  • Calculation of credit risk parameters within the ETL (interface) part of iFRS-VBox by means of special scripting for statistical analysis and/or
  • Imported from external sources, like an existing Basel II/III application and/or
  • Imputed manually on the configuration level or on the specific exposure level (e.g. for handling of special cases).

The calculation of the credit risk parameters like PD or LGD highly customizable an depends on the specific business model of the bank and on the type, quality and quantity of the available data. Best-practice approaches implemented by our team in other projects for PD estimation include:

  • For consumer loans and commitments: roll-rate model on the past-due days.
  • For business and commercial loans: credit scoring models (e.g. Z-score and/or probit/logit regression of balance sheet information).
  • For notes and securities: rating migration analysis.

Best-practice approaches implemented by our team in other projects for LGD estimation include:

  • Analysis of the payment and recovery history of already defaulted loans.
  • Evaluation of available collateral.
  • Discounted cash flow analysis of defaulted debt.
  • Market prices of defaulted assets.