9 Financial instruments - valuation
Valuation of financial instruments carried at fair value
Control environment
Common valuation policies, procedures, frameworks and models apply across the RBS Group. Therefore, for the most part, discussions on these aspects below reflect those in the RBS Group as relevant for businesses in the Group.
The Group’s control environment for the determination of the fair value of financial instruments includes formalised protocols for the review and validation of fair values independent of the businesses entering into the transactions. There are specific controls to ensure consistent pricing policies and procedures, incorporating disciplined price verification. The Group ensures that appropriate attention is given to bespoke transactions, structured products, illiquid products and other instruments which are difficult to price.
Independent price verification (IPV)
IPV is a key element of the control environment. Valuations are first performed by the business which entered into the transaction. Such valuations may be directly from available prices, or may be derived using a model and variable model inputs. These valuations are reviewed, and if necessary amended, by a team independent of those trading the financial instruments, in the light of available pricing evidence.
IPV differences are classified according to the quality of independent market observables into IPV quality bands linked to the fair value hierarchy principles, as laid out in IFRS 13 ‘Fair Value Measurement’. These differences are classified into fair value levels 1, 2 and 3 (with the valuation uncertainty risk increasing as the levels rise from 1 to 3) and then further classified into high, medium, low and indicative depending on the quality of the independent data available to validate the prices. Valuations are revised if they are outside agreed thresholds.
Governance framework
IPV takes place at least each month end date, for all fair value positions. The IPV control includes formalised reporting and escalation of any valuation differences in breach of established thresholds. The Pricing Unit determines IPV policy, monitors adherence to that policy and performs additional independent reviews of highly subjective valuation issues.
The Modelled Product Review Committee sets the policy for model documentation, testing and review, and prioritises models with significant exposure being reviewed by the RBS Group Pricing Model Risk team. The Valuation Committee is made up of valuation specialists and senior business representatives from various functions and oversee pricing, reserving and valuations issues as relevant to businesses within the RBS Group. This committee meets monthly to review and ratify any methodology changes. The Executive Valuation Committee meets quarterly to address key material and subjective valuation issues, to review items escalated by the Valuation Committees and to discuss other relevant matters including prudential valuation.
Valuation hierarchy
Initial classification of a financial instrument is carried out by the Product Control team following the principles in IFRS 13. They base their judgment on information gathered during the IPV process for instruments which include the sourcing of independent prices and model inputs. The quality and completeness of the information gathered in the IPV process gives an indication as to the liquidity and valuation uncertainty of an instrument.
These initial classifications are reviewed and challenged by the Pricing Unit and are also subject to senior management review. Particular attention is paid to instruments crossing from one level to another, new instrument classes or products, instruments that are generating significant profit and loss and instruments where valuation uncertainty is high.
Valuation techniques
The Group derives fair value of its instruments differently depending on whether the instrument is a non-modelled or a modelled product.
Non-modelled products
Non-modelled products are valued directly from a price input typically on a position by position basis and include cash, equities and most debt securities.
Modelled products
Modelled products valued using a pricing model range in complexity from comparatively vanilla products such as interest rate swaps and options (e.g. interest rate caps and floors) through to more complex derivatives. The valuation of modelled products requires an appropriate model and inputs into this model. Sometimes models are also used to derive inputs (e.g. to construct volatility surfaces). The Group uses a number of modelling methodologies.
Inputs to valuation models
Values between and beyond available data points are obtained by interpolation and extrapolation. When utilising valuation techniques, the fair value can be significantly affected by the choice of valuation model and by underlying assumptions concerning factors such as the amounts and timing of cash flows, discount rates and credit risk. The principal inputs to these valuation techniques are as follows:
· |
Bond prices - quoted prices are generally available for government bonds, certain corporate securities and some mortgage-related products. |
· |
Credit spreads - where available, these are derived from prices of credit default swaps or other credit based instruments, such as debt securities. For others, credit spreads are obtained from pricing services. For counterparty credit spreads, adjustments are made to market prices (or parameters) when the creditworthiness of the counterparty differs from that of the assumed counterparty in the market price (or parameters). |
Inputs to valuation models
· |
Foreign currency exchange rates - there are observable prices both for spot and forward contracts and futures in the world’s major currencies. |
· |
Equity and equity index prices - quoted prices are generally readily available for equity shares listed on the world’s major stock exchanges and for major indices on such shares. |
· |
Commodity prices - many commodities are actively traded in spot and forward contracts and futures on exchanges in London, New York and other commercial centers. |
· |
Price volatilities and correlations – volatility is a measure of the tendency of a price to change with time. Correlation measures the degree which two or more prices or other variables are observed to move together. |
· |
Prepayment rates - the fair value of a financial instrument that can be prepaid by the issuer or borrower differs from that of an instrument that cannot be prepaid. In valuing prepayable instruments that are not quoted in active markets, the Group considers the value of the prepayment option. |
· |
Recovery rates/loss given default - these are used as an input to valuation models and reserves for asset-backed securities and other credit products as an indicator of severity of losses on default. Recovery rates are primarily sourced from market data providers or inferred from observable credit spreads. |
Consensus pricing
The Group uses consensus prices for the IPV of some instruments. The consensus service encompasses the equity, interest rate, currency, commodity, credit, property, fund and bond markets, providing comprehensive matrices of vanilla prices and a wide selection of exotic products. NatWest Markets contribute to consensus pricing services where there is a significant interest either from a positional point of view or to test models for future business use. Data sourced from consensus pricing services are used for a combination of control processes including direct price testing, evidence of observability and model testing. In practice this means that the Group submits prices for all material positions for which a service is available. Data from consensus services are subject to the same level of quality review as other inputs used for IPV process.
In order to determine a reliable fair value, where appropriate, management applies valuation adjustments to the pricing information gathered from the above sources. The sources of independent data are reviewed for quality and are applied in the IPV processes using a formalised input quality hierarchy. These adjustments reflect the Group’s assessment of factors that market participants would consider in setting a price.
Furthermore, on an ongoing basis, the Group assesses the appropriateness of any model used. To the extent that the price determined by internal models does not represent the fair value of the instrument, for instance in highly stressed market conditions, the Group makes adjustments to the model valuation to calibrate to other available pricing sources.
Where unobservable inputs are used, the Group may determine a range of possible valuations derived from differing stress scenarios to determine the sensitivity associated with the valuation. When establishing the fair value of a financial instrument using a valuation technique, the Group considers adjustments to the modelled price which market participants would make when pricing that instrument. Such adjustments include the credit quality of the counterparty and adjustments to compensate for model limitations.
Valuation reserves
When valuing financial instruments in the trading book, adjustments are made to mid-market valuations to cover bid-offer spread, liquidity and credit risk.
Credit valuation adjustments (CVA)
CVA represents an estimate of the adjustment to fair value that a market participant would make to incorporate the counterparty credit risk inherent in derivative exposures. CVA is actively managed by a credit and market risk hedging process, and therefore movements in CVA are partially offset by trading revenue on the hedges. The CVA reserve at 31 December 2017 was £11 million (2016 - £26 million).
The CVA is calculated on a portfolio basis reflecting an estimate of the amount a third party would charge to assume the credit risk.
Where a positive exposure exists to a counterparty that is considered to be close to default, the CVA is calculated by applying expected losses to the current level of exposure. Otherwise, expected losses are applied to estimated potential future positive exposures which are modelled to reflect the volatility of the market factors which drive the exposures and the correlation between those factors.
Expected losses are determined from market implied probabilities of default and internally assessed recovery levels. The probability of default is calculated with reference to observable credit spreads and observable recovery levels. For counterparties where observable data do not exist, the probability of default is determined from the credit spreads and recovery levels of similarly rated entities.
Collateral held under a credit support agreement is factored into the CVA calculation. In such cases where the Group holds collateral against counterparty exposures, CVA is held to the extent that residual risk remains.
Valuation hierarchy
The following tables show financial instruments carried at fair value on the balance sheet by valuation hierarchy for continuing operations - level 1, level 2 and level 3.
|
|
2017 |
|
2016 |
||||||||||||
|
|
Level 1 |
|
Level 2 |
|
Level 3 |
|
Total |
|
Level 1 |
|
Level 2 |
|
Level 3 |
|
Total |
|
|
£bn |
|
£bn |
|
£bn |
|
£bn |
|
£bn |
|
£bn |
|
£bn |
|
£bn |
Assets |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Loans and advances |
|
— |
|
0.2 |
|
— |
|
0.2 |
|
— |
|
19.6 |
|
0.4 |
|
20.0 |
Debt securities |
|
0.4 |
|
0.2 |
|
— |
|
0.6 |
|
4.2 |
|
0.3 |
|
— |
|
4.5 |
- of which AFS |
|
0.4 |
|
0.2 |
|
— |
|
0.6 |
|
0.3 |
|
0.1 |
|
— |
|
0.4 |
Equity shares |
|
— |
|
— |
|
— |
|
— |
|
— |
|
0.1 |
|
— |
|
0.1 |
Derivatives |
|
— |
|
2.3 |
|
— |
|
2.3 |
|
— |
|
3.9 |
|
— |
|
3.9 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
0.4 |
|
2.7 |
|
— |
|
3.1 |
|
4.2 |
|
23.9 |
|
0.4 |
|
28.5 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Proportion |
|
12.9% |
|
87.1% |
|
— |
|
100% |
|
14.7% |
|
83.9% |
|
1.4% |
|
100% |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Liabilities |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Deposits |
|
— |
|
0.8 |
|
— |
|
0.8 |
|
— |
|
15.7 |
|
— |
|
15.7 |
Short positions |
|
— |
|
— |
|
— |
|
— |
|
4.6 |
|
— |
|
— |
|
4.6 |
Derivatives |
|
— |
|
3.2 |
|
— |
|
3.2 |
|
— |
|
4.6 |
|
— |
|
4.6 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
— |
|
4.0 |
|
— |
|
4.0 |
|
4.6 |
|
20.3 |
|
— |
|
24.9 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Proportion |
|
— |
|
100% |
|
— |
|
100% |
|
18.5% |
|
81.5% |
|
— |
|
100% |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Level 3 balances within continuing operations at 31 December 2017 were nil (2016 – level 3 comprised loans and advances of £0.4 billion with sensitivity of +£40 million / -£30 million, valued using a traded price with a range of 85% to 95%). Level 3 balances within disposal groups were immaterial at 31 December 2017.
Notes:
(1) |
Level 1: valued using unadjusted quoted prices in active markets, for identical financial instruments. Examples include G10 government securities, listed equity shares, certain US agency securities.| |
Level 2: valued using techniques based significantly on observable market data. Instruments in this category are valued using:
(a) quoted prices for similar instruments or identical instruments in markets which are not considered to be active; or
(b) valuation techniques where all the inputs that have a significant effect on the valuations are directly or indirectly based on observable market data.
Level 2 instruments included non-G10 government securities, most government agency securities, certain mortgage products, most bank loans, repos and reverse repos, less liquid listed equities, and most OTC derivatives.
Level 3: instruments in this category have been valued using a valuation technique where at least one input which could have a significant effect on the instrument’s valuation, is not based on observable market data. Level 3 instruments primarily include cash instruments which trade infrequently such as certain mortgage loans.
(2) |
Transfers between levels are deemed to have occurred at the beginning of the quarter in which the instruments were transferred. There were no significant transfers between level 1 and level 2. |
(3) |
For an analysis of debt securities (by issuer and measurement classification) and derivatives (by type of contract) refer to Capital and risk management – Credit risk. |
(4) |
The determination of an instrument’s level cannot be made at a global product level as a single product type can be in more than one level. For example, a single name corporate credit default swap could be in Level 2 or Level 3 depending on whether the reference counterparty’s obligations are liquid or illiquid. |
The following table shows the carrying value and fair value of financial instruments carried at amortised cost on the balance sheet.
|
|
Group |
|
Bank |
||||||||||||||||
|
|
Items where |
|
|
|
|
|
|
|
|
|
Items where |
|
|
|
|
|
|
|
|
|
|
fair value |
|
|
|
|
|
|
|
|
|
fair value |
|
|
|
|
|
|
|
|
|
|
approximates |
|
Carrying |
|
Fair |
|
Fair value hierarchy level |
|
approximates |
|
Carrying |
|
Fair |
|
Fair value hierarchy level |
||||
|
|
carrying value |
|
value |
|
value |
|
Level 2 |
|
Level 3 |
|
carrying value |
|
value |
|
value |
|
Level 2 |
|
Level 3 |
2017 |
|
£bn |
|
£bn |
|
£bn |
|
£bn |
|
£bn |
|
£bn |
|
£bn |
|
£bn |
|
£bn |
|
£bn |
Financial assets |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Cash and balances at central banks |
|
35.8 |
|
|
|
|
|
|
|
|
|
34.8 |
|
|
|
|
|
|
|
|
Loans and advances to banks |
|
0.6 |
|
79.1 |
|
79.5 |
|
1.0 |
|
78.5 |
|
0.6 |
|
55.0 |
|
55.5 |
|
1.1 |
|
54.4 |
Loans and advances to customers |
|
|
|
191.9 |
|
191.3 |
|
0.2 |
|
191.1 |
|
|
|
160.7 |
|
160.1 |
|
— |
|
160.1 |
Debt securities |
|
|
|
1.1 |
|
1.1 |
|
— |
|
1.1 |
|
|
|
1.1 |
|
1.1 |
|
— |
|
1.1 |
Settlement balances |
|
— |
|
|
|
|
|
|
|
|
|
— |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Financial liabilities |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Deposits by banks |
|
18.8 |
|
34.4 |
|
35.6 |
|
17.0 |
|
18.6 |
|
4.5 |
|
27.4 |
|
28.5 |
|
17.0 |
|
11.5 |
Customer accounts |
|
206.3 |
|
26.9 |
|
26.9 |
|
4.9 |
|
22.0 |
|
187.2 |
|
13.8 |
|
13.8 |
|
4.9 |
|
8.9 |
Debt securities in issue |
|
|
|
0.4 |
|
0.4 |
|
— |
|
0.4 |
|
|
|
— |
|
— |
|
— |
|
— |
Settlement balances |
|
— |
|
|
|
|
|
|
|
|
|
— |
|
|
|
|
|
|
|
|
Notes in circulation (1) |
|
0.8 |
|
|
|
|
|
|
|
|
|
— |
|
|
|
|
|
|
|
|
Subordinated liabilities |
|
|
|
5.8 |
|
6.0 |
|
5.9 |
|
0.1 |
|
|
|
5.6 |
|
5.9 |
|
5.9 |
|
— |
2016 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Financial assets |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Cash and balances at central banks |
|
2.6 |
|
|
|
|
|
|
|
|
|
1.2 |
|
|
|
|
|
|
|
|
Loans and advances to banks |
|
0.6 |
|
87.6 |
|
88.1 |
|
3.6 |
|
84.5 |
|
0.5 |
|
63.8 |
|
64.2 |
|
3.3 |
|
60.9 |
Loans and advances to customers |
|
|
|
166.0 |
|
166.0 |
|
0.9 |
|
165.1 |
|
|
|
150.0 |
|
150.0 |
|
— |
|
150.0 |
Debt securities |
|
|
|
— |
|
— |
|
— |
|
— |
|
|
|
— |
|
— |
|
— |
|
— |
Settlement balances |
|
1.7 |
|
|
|
|
|
|
|
|
|
0.1 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Financial liabilities |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Deposits by banks |
|
5.1 |
|
11.1 |
|
12.1 |
|
0.9 |
|
11.2 |
|
4.5 |
|
4.2 |
|
5.2 |
|
— |
|
5.2 |
Customer accounts |
|
197.3 |
|
24.9 |
|
24.9 |
|
5.7 |
|
19.2 |
|
180.1 |
|
12.1 |
|
12.1 |
|
5.7 |
|
6.4 |
Debt securities in issue |
|
|
|
0.3 |
|
0.3 |
|
— |
|
0.3 |
|
|
|
— |
|
— |
|
— |
|
— |
Settlement balances |
|
1.8 |
|
|
|
|
|
|
|
|
|
0.1 |
|
|
|
|
|
|
|
|
Notes in circulation (1) |
|
0.8 |
|
|
|
|
|
|
|
|
|
— |
|
|
|
|
|
|
|
|
Subordinated liabilities |
|
|
|
7.3 |
|
7.0 |
|
1.6 |
|
5.4 |
|
|
|
5.9 |
|
5.7 |
|
1.3 |
|
4.4 |
Note:
(1) |
Included in Accruals and other liabilities. |
The fair value is the price that would be received to sell an asset or paid to transfer a liability in an orderly transaction between market participants at the measurement date. Quoted market values are used where available; otherwise, fair values have been estimated based on discounted expected future cash flows and other valuation techniques. These techniques involve uncertainties and require assumptions and judgments covering prepayments, credit risk and discount rates. Furthermore there is a wide range of potential valuation techniques. Changes in these assumptions would significantly affect estimated fair values. The fair values reported would not necessarily be realised in an immediate sale or settlement.
The assumptions and methodologies underlying the calculation of fair values of financial instruments at the balance sheet date are as follows:
Short-term financial instruments
For certain short-term financial instruments: cash and balances at central banks, items in the course of collection from other banks, settlement balances, items in the course of transmission to other banks, customer demand deposits and notes in circulation, carrying value is a reasonable approximation of fair value.
Loans and advances to banks and customers
In estimating the fair value of loans and advances to banks and customers measured at amortised cost, the Group’s loans are segregated into appropriate portfolios reflecting the characteristics of the constituent loans. Two principal methods are used to estimate fair value:
(a) |
Contractual cash flows are discounted using a market discount rate that incorporates the current spread for the borrower or where this is not observable, the spread for borrowers of a similar credit standing. This method is used for portfolios where counterparties have external ratings: institutional and corporate lending in NatWest Markets. |
(b) |
Expected cash flows (unadjusted for credit losses) are discounted at the current offer rate for the same or similar products. This approach is adopted for lending portfolios in UK PBB, Commercial Banking (SME loans) and Private Banking in order to reflect the homogeneous nature of these portfolios. |
For certain portfolios where there are very few or no recent transactions a bespoke approach is used based on available market data.
Debt securities
The majority of debt securities are valued using quoted prices in active markets, or using quoted prices for similar assets in active markets. Fair values of the rest are determined using discounted cash flow valuation techniques.
Deposits by banks and customer accounts
Fair values of deposits are estimated using discounted cash flow valuation techniques.
Debt securities in issue and subordinated liabilities
Fair values are determined using quoted prices for similar liabilities where available or by reference to valuation techniques, adjusting for own credit spreads where appropriate.