Barclays 2009 Annual Report Download - page 303

www.barclays.com/annualreport09 Barclays PLC Annual Report 2009 301
50 Fair value of financial instruments continued
Unrecognised gains as a result of the use of valuation models using unobservable inputs
The amount that has yet to be recognised in income that relates to the difference between the transaction price (the fair value at initial recognition) and the
amount that would have arisen had valuation models using unobservable inputs been used on initial recognition, less amounts subsequently recognised,
was as follows:
Year ended 31st December 2009 2008
£m £m
Opening balance 128 154
Additions 39 77
Amortisation and releases (68) (103)
Closing balance 99 128
Sensitivity analysis of valuations using unobservable inputs
At 31st December 2009 Fair value Favourable changes Unfavourable change
Total Total Profit Profit
assets liabilities and loss Equity and loss Equity
Product type £m £m £m £m £m £m
Commercial real estate loans 7,170 – 429 – (437) –
Asset backed products 5,840 (2,334) 175 4 (175) (4)
Other credit products 2,020 (2,827) 171 – (152) –
Derivative exposure to Monoline insurers 2,027 – 336 – (532) –
Non-asset backed debt instruments 3,127 (3,202) 145 2 (141) (2)
Equity products 1,536 (1,922) 28 15 (28) (15)
Private equity 1,978 – 267 73 (339) (95)
Funds and fund-linked products 1,241 – 100 – (100) –
FX products 761 (379) 33 – (33) –
Interest rate products 2,357 (1,775) 78 – (78) –
Commodity products 748 (581) 36 – (36) –
Other 1,451 – 52 – (52) –
Total 30,256 (13,020) 1,850 94 (2,103) (116)
As part of risk management processes, an analysis is performed on unobservable parameters to generate a range of reasonably possible alternative
valuations.
The effect of stressing the unobservable assumptions to a range of reasonably possible alternatives would be to increase the fair values by up to £1.9bn
(2008: £2.4bn) or to decrease the fair values by up to £2.2bn (2008: £3bn) with substantially all the potential effect impacting profit or loss rather than
equity. The metric has not been offset for the effect of hedging.
The stresses applied take account of the nature of valuation techniques used, as well as the availability and reliability of observable proxy and historic data.
In all cases, an assessment is made to determine the suitability of available data. The sensitivity methodologies are based on a range, standard deviation
or spread data of a reliable reference source or a scenario based on alternative market views. The level of shift or scenarios applied is considered for each
product and varied according to the quality of the data and variability of underlying market. The approach adopted in determining these sensitivities has
evolved during the year, in the context of changing market conditions.
Commercial real estate loans
The unobservable inputs include but are not limited to market quoted origination spreads, internal credit ratings and subordination of the loans. The sensitivity
is determined by applying a +/- 6% shift for each underlying position based on the largest upward and downward price movement of observable published
indices of a similar nature in the preceding 12-month period.
Asset backed products
For non-agency RMBS and non-investment grade ABS, sensitivity is calculated on the price movement on observable ABX.HE indices. For asset backed
credit derivatives, sensitivity is calculated on price movements within the ABX.HE.AAA indices. Sensitivity is based on the average of the largest price
movement upward and the largest price movement downward in the preceding 12-month period.