ING Direct 2009 Annual Report Download - page 286

RISK MEASUREMENT AND REPORTING
ING distinguishes three separate information requirements from senior management related to the Advanced IRB (AIRB) approach for
credit risk:
Reporting on (minimum) regulatory capital requirements;•
Model monitoring reports; and•
Stress testing reports.•
The acceptance, maintenance, measurement, management and reporting of credit risks at all levels of ING Bank is accomplished through
promotion of single, common credit risk data standards and the integration into common credit risk tools that support standardised and
transparent credit risk practices.
THE IRB METHOD IN SHORT
There are four elements which drive the Basel II ‘risk-based approach’ to the determination of the capital base. For each of these elements,
ING has developed a series of statistical, expert and hybrid models based on ING’s historical experience and other market observations.
Probability of Default (PD):• The first is the borrower’s, counterparty’s, or issuer’s (collectively referred to as the ‘obligor’) probability
of default, which measures an obligor’s creditworthiness in terms of likelihood to go into default. The result of this calculation attempts
to measure the senior, unsecured standalone creditworthiness of an organisation without consideration of structural elements of the
underlying transactions, such as collateral, pricing, or maturity;
Exposure at Default (EAD): • The second element is the obligor’s exposure at default. These models are intended to estimate the
outstanding amount or obligation at the moment of default in the future. Since the fact that an obligor will go into default is not
known, and the level of outstandings that may occur on that date is also not known, ING uses a combination of statistical, expert and
hybrid models to estimate the Exposure at Default. With the exception of guarantees and letters of credit, the EAD is always equal to or
higher than the associated credit risk outstandings, under the assumption that obligors tend to absorb liquidity from available credit
resources before financial problems become apparent to the obligor’s creditors;
Loss Given Default (LGD): • The third element is the loss given default. These models are intended to estimate the amount ING will lose
when liquidating collateral pledged in association with a given loan or financial obligation, or alternatively, liquidating the company as a
whole, as part of a workout process. LGD models are based on cover types, estimated recovery rates given orderly liquidation, and (in)
direct cost of liquidation;
Maturity (M): • The fourth element is the time to the maturity of the underlying financial obligation. Basel II caps the maturity element at
five years, despite the fact that many obligations extend longer than five years.
Expected Loss (EL): The expected loss provides a measure of the value of the credit losses that ING may reasonably expect to incur on its
portfolio. ING must hold a reserve (as part of its capital base) to cover the expected losses in its credit portfolio. In its basic form, the
expected loss can be represented as:
EL = PD * EAD * LGD
Unexpected Loss (UL): Additionally, ING must also maintain a capital buffer against unexpected losses in order to protect itself against
credit losses associated with unusual market events outside of the statistical norms.
Basel II uses these same components (expected loss and unexpected loss) conceptually in the determination of the Risk Weighted Assets
(RWA). Like EL, RWA takes PD, EAD, and LGD into account, but also includes variables associated with the type of obligor and its size.
The PD, EAD and LGD models that are used in the calculation of Basel II regulatory capital are the same models that ING uses in the
determination of its internally based economic capital models. Additionally, these models are used for loan pricing and customer
profitability calculations, as well as forming the foundation for loan loss provisioning calculations.
CREDIT RISK MODELS
ING considers a well-balanced and controlled set of rules around model development, maintenance and validation to be an essential
component for professional risk measurement and risk management. In 2006, ING developed and implemented a Credit Risk Model
Governance framework, which consists of a set of extensive guidelines and requirements to which all stakeholders must adhere when
developing, implementing and maintaining PD, LGD and EAD models.
Types of Credit Risk Modelling
Within ING Bank, there are three types of modelling which form the foundation of the PD, EAD and LGD models used throughout
the bank.
ING Group Annual Report 2009
284
Additional Pillar 3 information for ING Bank only (continued)
2.4 Additional information