RBS 2011 Annual Report Download - page 146

144 RBS Group 2011
Risk management: Credit risk continued
Credit risk measurement*
Credit risk models are used throughout the Group to support the
quantitative risk assessment element within the credit approval process,
ongoing credit risk management, monitoring and reporting and portfolio
analytics. Credit risk models used by the Group may be divided into three
categories, as follows.
Probability of default/customer credit grade
These models assess the probability that a customer will fail to make full
and timely repayment of its obligations. The probability of a customer
failing to do so is measured over a one year period through the economic
cycle, although certain retail scorecards use longer periods for business
management purposes.
Wholesale businesses - as part of the credit assessment process, each
counterparty is assigned an internal credit grade derived from a default
probability. There are a number of different credit grading models in use
across the Group, each of which considers risk characteristics particular
to that type of customer. The credit grading models score a combination
of quantitative inputs (for example, recent financial performance) and
qualitative inputs (for example, management performance or sector
outlook).
Retail businesses - each customer account is separately scored using
models based on the most material drivers of default. In general,
scorecards are statistically derived using customer data. Customers are
assigned a score, which in turn is mapped to a probability of default. The
probabilities of default are used to support automated credit decision
making and to group customers into risk pools for regulatory capital
calculations.
Exposure at default
Facility usage models estimate the expected level of utilisation of a credit
facility at the time of a borrower’s default. For revolving and variable draw
down type products which are not fully drawn, the exposure at default
(EAD) will typically be higher than the current utilisation. The
methodologies used in EAD modelling provide an estimate of potential
exposure and recognise that customers may make more use of their
existing credit facilities as they approach default.
Counterparty credit risk exposure measurement models are used for
derivatives and other traded instruments, where the amount of credit risk
exposure may be dependent upon one or more underlying market
variables, such as interest or foreign exchange rates. These models drive
internal credit risk management activities such as limit and excess
management.
Loss given default
These models estimate the economic loss that may be experienced (the
amount that cannot be recovered) by the Group on a credit facility in the
event of default. The Group’s loss given default models take into account
both borrower and facility characteristics for unsecured or partially
unsecured facilities, as well as the quality of any risk mitigation that may
be in place for secured facilities, the cost of collections and a time
discount factor for the delay in cash recovery.
Credit risk assets
In the tables and commentary below, exposure refers to credit risk
assets, which consist of:
xLending - cash and balances at central banks and loans and
advances to banks and customers (including overdraft facilities,
instalment credit and finance leases);
xRate risk management; and
xContingent obligations, primarily letters of credit and guarantees.
Reverse repurchase agreements and issuer risk (primarily debt securities
-refer to pages 175 to 177) are excluded. Where relevant and unless
otherwise stated, the data reflect the effect of credit mitigation
techniques.
Divisional analysis of credit risk assets 2011
£m
2010
£m
2009
£m
UK Retail 111,070 108,302 103,029
UK Corporate 102,468 105,886 110,009
Wealth 20,079 18,875 16,553
Global Transaction Services 34,719 35,462 32,428
Ulster Bank 37,781 40,750 42,042
US Retail & Commercial 56,412 51,699 52,104
Retail & Commercial 362,529 360,974 356,165
Global Banking & Markets 165,616 171,891 205,588
Other 64,518 36,659 3,305
Core 592,663 569,524 565,058
Non-Core 92,710 125,383 158,499
685,373 694,907 723,557
*unaudited
Business review Risk and balance sheet management continued