(audited)

The table below shows NWM Group’s carrying values of the principal funding sources based on contractual maturity.

(1)

Funding sources excludes short positions of £9,803 million (2022 - £9,524 million) reflected as trading liabilities on the balance sheet.

(2)

Comprises Central and other bank repos of £3,957 million (2022 - £1,642 million), other financial institution repos of £20,394 million (2022 - £19,354 million) and other corporate repos of

£2,551 million (2022 - £2,744 million).

(3)

Amounts due to holding company and fellow subsidiaries relating to non-financial instruments of £90 million (2022 - £211 million) and Nil intercompany settlement balances (2022 – £26

million) have been excluded from the table.

(4)

Eligible liabilities (as defined in the Banking Act 2009 as amended from time to time) that meet the eligibility criteria set out in the regulations, rules, policies, guidelines, or statements of

Definition

(audited)

Credit risk is the risk that customers, counterparties or issuers fail

to meet a contractual obligation to settle outstanding amounts.

Sources of risk

(audited)

The principal sources of credit risk for NWM Group are lending,

off-balance sheet products, derivatives and securities financing,

debt securities, and settlement risk through trading activities.

Key developments in 2023

(audited)



The credit profile remained stable, but the outlook is uncertain

from inflationary pressure, compounded by higher interest

rates and geopolitical tension. NWM Group has yet to see

signs of financial stress materially affect customers’ ability to

repay.



Expected credit loss (ECL) was stable in 2023, with a small

overall increase driven by business growth.



Current and potential credit exposure increased in the context

of planned business growth. The effect of inflation and higher

interest rates on customers continues to be assessed. Sector

appetite is reviewed regularly and where appropriate adjusted

for those sectors most effected by current economic and

geopolitical conditions.



NWM Group continued to progress embedding climate change

considerations in credit assessment and monitoring, including

scenario analysis to assess the materiality of climate change

risks.

Governance

(audited)

The Credit Risk function provides oversight and challenge of

frontline credit risk management activities. Authority is delegated

to credit risk officers who operate within designated limits set at a

customer level and a portfolio level.

Governance activities include:



Defining and proposing credit risk appetite measures for

Board approval.



Establishing credit risk policy, standards and toolkits which set

out the mandatory limits and parameters required to ensure

that credit risk is managed within risk appetite and which

provide the minimum standards for the identification,

assessment, management, monitoring and reporting of credit

risk.



Oversight of the first line of defence to ensure that credit risk

remains within the appetite set by the Board and that it is

being managed adequately and effectively.



Assessing the adequacy of ECL provisions including approving

key IFRS 9 inputs (such as significant increase in credit risk

(SICR) thresholds) and any necessary in-model and post

model adjustments through provisions and model committees.



Development and approval of credit grading models.



Providing regular reporting on credit risk to the Board.

The Credit Risk Committee provides oversight of the overall credit

risk profile and sector/product/asset class concentrations.

Risk appetite

Credit risk appetite is approved by the Board and is set and

monitored through a risk appetite framework.

The framework has been designed to reflect factors that influence

the ability to operate within risk appetite. Tools such as stress

testing and economic capital are used to measure credit risk

volatility and develop links between the framework and risk

appetite limits.

Risk appetite statements and associated measures are reviewed

at least annually by the Board on the Board Risk Committee’s

recommendation to ensure they remain appropriate and aligned

to strategy.

Operational limits are used to manage concentrations of risk

which may arise across four lenses – single name, sector, country

and product and asset classes.

The framework is supported by a suite of transactional

acceptance standards that set out the risk parameters within

which businesses should operate.

Identification and measurement

Credit stewardship

Risks are identified through relationship management and credit

stewardship of customers and portfolios. Credit stewardship takes

place throughout the customer relationship, beginning with the

initial approval. It includes the application of credit assessment

standards, credit risk mitigation and collateral, ensuring that credit

documentation is complete and appropriate, carrying out regular

portfolio or customer reviews and problem debt identification and

management.

Asset quality

All credit grades map to an asset quality (AQ) scale, used for

financial reporting. This AQ scale is based on Basel probability of

defaults. Performing loans are defined as AQ1-AQ9 (where the

probability of default (PD) is less than 100%) and defaulted non-

performing loans as AQ10 or Stage 3 under IFRS 9 (where the PD

is 100%). Loans are defined as defaulted when the payment

status becomes 90 days past due, or earlier if there is clear

evidence that the borrower is unlikely to repay, for example

bankruptcy or insolvency.

Counterparty credit risk

Counterparty credit risk arises from the obligations of customers

under derivative and securities financing transactions. It is a

material part of NWM Group’s credit risk.

NWM Group mitigates counterparty credit risk through

collateralisation and netting agreements, which allow amounts

owed by NWM Group to a counterparty to be netted against

amounts the counterparty owes NWM Group.

Mitigation

Mitigation techniques, as set out in the appropriate credit risk

toolkits and transactional acceptance standards, are used in the

management of credit portfolios across NWM Group. These

techniques mitigate credit concentrations in relation to an

individual customer, a borrower group or a collection of related

borrowers. Where possible, customer credit balances are netted

against obligations. Mitigation tools can include structuring a

security interest in a physical or financial asset, the use of credit

derivatives including credit default swaps, credit-linked debt

instruments and securitisation structures, and the use of

guarantees and similar instruments (for example, credit insurance)

from related and third parties.

Credit risk continued

Assessment and monitoring

Customers, including corporates, banks and other financial

institutions are managed on an individual basis. Customers are

aggregated as a single risk when sufficiently interconnected to the

extent that a failure of one could lead to the failure of another.

A credit assessment is carried out before credit facilities are made

available to customers. The assessment process is dependent on

the complexity of the transaction. Credit approvals are subject to

environmental, social and governance risk policies which restrict

exposure to certain highly carbon intensive industries as well as

those with potentially heightened reputational impacts. Customer

specific climate risk commentary is now mandatory.

For lower risk transactions below specific thresholds, credit

decisions can be approved through a combination of fully

automated or relationship manager self-sanctioning within the

business. This process is facilitated through an auto-decision

making system, which utilises scorecards, strategies and policy

rules.

For all other transactions, credit is only granted to customers

following joint approval by an approver from the business and the

credit risk function, or by two credit officers. The joint business

and credit approvers act within a delegated approval authority

under the Wholesale Credit Authorities framework policy. The

level of delegated authority held by approvers is dependent on

their experience and expertise with only a small number of senior

executives holding the highest approval authority.

Transactional acceptance standards provide detailed transactional

lending and risk acceptance metrics and structuring guidance. As

such, these standards provide a mechanism to manage risk

appetite at the customer/transaction level and are supplementary

to the established credit risk appetite.

Credit grades (PD) and loss given default (LGD) are reviewed and

if appropriate reapproved annually. The review process assesses

borrower performance, including reconfirmation or adjustment of

risk parameter estimates; the adequacy of security; compliance

with terms and conditions; and refinancing risk.

Problem debt management

Early problem identification

Each sector has defined early warning indicators to identify

customers experiencing financial difficulty, and to increase

monitoring if needed. Early warning indicators may be internal,

such as a customer’s bank account activity, or external, such as a

publicly listed customer’s share price. If early warning indicators

show a customer is experiencing potential or actual difficulty, or if

relationship managers or credit officers identify other signs of

financial difficulty, they may decide to classify the customer within

the Risk of Credit Loss framework. Broader macro-economic

trends including commodity prices, foreign exchange rates and

consumer and government spend are also tracked, helping inform

decisions on sector risk appetite. Customer level early warning

indicators are regularly reviewed to ensure alignment with

prevailing economic conditions, ensuring both the volume and

focus of alerts is aligned to the point-in-time risk within each

sector.

The aligned Risk of Credit Loss and Viability framework

This framework focuses on all Wholesale customers to provide

early identification of credit deterioration, support intelligent risk-

taking, ensure fair and consistent customer outcomes and provide

key insights into Wholesale lending portfolios. Expert judgement is

applied by experienced credit risk officers to classify cases into

categories that reflect progressively deteriorating credit risk to

NWM Group. There are two classifications in the framework that

apply to non-defaulted customers who are in financial stress –

Heightened Monitoring and Risk of Credit Loss. For the purposes

of provisioning, all exposures categorised as Heightened

Monitoring or Risk of Credit Loss are categorised as Stage 2 and

subject to a lifetime loss assessment. The framework also applies

to those customers that have met NWM Group’s default criteria

(AQ10 exposures). Defaulted exposures are categorised as Stage

3 impaired for provisioning purposes.

Customers classified in the Heightened Monitoring category are

those who are still performing but have certain characteristics –

such as trading issues, covenant breaches, material PD

downgrades and past due facilities – that may affect the ability to

meet repayment obligations. Heightened Monitoring customers

require pre-emptive actions to return or maintain their facilities

within risk appetite.

Risk of Credit Loss customers are performing customers that

have met the criteria for Heightened Monitoring and also pose a

risk of credit loss to NWM Group in the next 12 months should

mitigating action not be taken or not be successful.

Once classified as either Heightened Monitoring or Risk of Credit

Loss, a number of mandatory actions are taken in accordance

with policies. Actions include a review of the customer’s credit

grade, facility and security documentation and the valuation of

security. Depending on the severity of the financial difficulty and

the size of the exposure, the customer relationship strategy is

reassessed by credit officers, by specialist credit risk or

relationship management units in the relevant business, or by

Restructuring.

Restructuring

Where customers are categorised as Risk of Credit Loss,

relationships are mainly managed by the Restructuring team in

NatWest Bank Plc as a service to NWM Group. The team protects

NWM Group’s capital by working with corporate and commercial

customers to support their turnaround and recovery strategies

and enable them to return to mainstream banking.

Forbearance

(audited)

Forbearance takes place when a concession is made on the

contractual terms of a loan/debt in response to a customer’s

financial difficulties. The aim of forbearance is to support and

restore the customer to financial health while minimising risk. The

type of forbearance offered is tailored to the customer’s individual

circumstances and may involve covenant waivers, amendments

to margins, payment concessions and loan rescheduling (including

extensions in contractual maturity), capitalisation of arrears, and

debt forgiveness or debt-for-equity swaps.

Provisioning for forbearance

Provisions for forborne loans are assessed in accordance with

normal provisioning policies. The customer’s financial position and

prospects – as well as the likely effect of the forbearance,

including any concessions granted, and revised PD or LGD

gradings – are considered in order to establish whether an

impairment provision increase is required.

Loans granted forbearance are individually credit assessed in

most cases. Performing loans subject to forbearance treatment

are categorised as Stage 2 and subject to a lifetime loss

assessment.

Credit risk continued

Forbearance may result in the value of the outstanding debt

exceeding the present value of the estimated future cash flows.

This difference will lead to a customer being classified as non-

performing.

In the case of non-performing forborne loans, an individual loan

impairment provision assessment generally takes place prior to

forbearance being granted. The amount of the loan impairment

provision may change once the terms of the forbearance are

known, resulting in an additional provision charge or a release of

the provision in the period the forbearance is granted.

The transfer of loans from impaired to performing status follows

assessment by relationship managers and credit. When no further

losses are anticipated and the customer is expected to meet the

loan’s revised terms, any provision is written-off or released and

the balance of the loan can be returned to performing status once

exit criteria, as set out by regulatory guidance, is met.

Credit grading models

Credit grading models is the collective term used to describe all

models, frameworks and methodologies used to calculate PD,

exposure at default (EAD), LGD, maturity and the production of

credit grades.

Credit grading models are designed to provide:



An assessment of customer and transaction characteristics.



A meaningful differentiation of credit risk.



Accurate internal default rate, loss and exposure estimates

that are used in the capital calculation or wider risk

management purposes.

Impairment, provisioning and write-offs

(audited)

In the overall assessment of credit risk, impairment provisioning

and write-offs are used as key indicators of credit quality.

NWM Group’s IFRS 9 provisioning models, which use existing IRB

models as a starting point, incorporate term structures and

forward-looking information. Regulatory conservatism within the

IRB models has been removed as appropriate to comply with the

IFRS 9 requirement for unbiased ECL estimates.

Five key areas may materially influence the measurement of

credit impairment under IFRS 9 – two of these relate to model

build and three relate to model application:

Model build:



The determination of economic indicators that have most

influence on credit loss for each portfolio and the severity of

impact (this leverages existing stress testing models which are

reviewed annually).



The build of term structures to extend the determination of

the risk of loss beyond 12 months that will influence the

impact of lifetime loss for exposures in Stage 2.

Model application:



The assessment of the SICR and the formation of a

framework capable of consistent application.



The determination of asset lifetimes that reflect behavioural

characteristics while also representing management actions

and processes (using historical data and experience).



The choice of forward-looking economic scenarios and their

respective probability weights.

For accounting policy information refer to Accounting policy 2.3

for further details.

IFRS 9 ECL model design principles (audited)

Modelling of ECL for IFRS 9 follows the conventional approach to

divide the estimation of credit losses into its component parts of

PD, LGD and EAD.

To meet IFRS 9 requirements, the PD, LGD and EAD parameters

differ from their Pillar 1 internal ratings based (IRB) counterparts

in the following aspects:



– material regulatory conservatism has been

removed from IFRS 9 parameters to produce unbiased

estimates.



– IFRS 9 parameters reflect actual economic

conditions at the reporting date instead of long-run average

or downturn conditions.



– IFRS 9 PD estimates and, where

appropriate, EAD and LGD estimates reflect forward-looking

economic conditions.



– IFRS 9 PD, LGD and EAD are

provided as multi-period term structures up to exposure

lifetimes instead of over a fixed one-year horizon.

IFRS 9 requires that at each reporting date, an entity shall assess

whether the credit risk on an account has increased significantly

since initial recognition. Part of this assessment requires a

comparison to be made between the current lifetime PD (i.e. the

PD over the remaining lifetime at the reporting date) and the

equivalent lifetime PD as determined at the date of initial

recognition.

For assets originated before IFRS 9 was introduced, comparable

lifetime origination PDs did not exist. These have been

retrospectively created using the relevant model inputs applicable

at initial recognition.

PD estimates

PD models use a point-in-time/through-the-cycle framework to

convert one-year regulatory PDs into point-in-time estimates that

reflect economic conditions at the reporting date. The framework

utilises credit cycle indices (CCIs) for a comprehensive set of

region/industry segments. Further detail on CCIs is provided in the

Economic loss driver’s section.

One-year point-in-time PDs are extended to forward-looking

lifetime PDs using a conditional transition matrix approach and a

set of econometric forecasting models.

LGD estimates

The general approach for the IFRS 9 LGD models is to leverage

corresponding IRB LGD models with bespoke adjustments to

ensure estimates are unbiased and, where relevant, forward-

looking.

Forward-looking economic information is incorporated into LGD

estimates using the existing point-in-time/through-the-cycle

framework. For low-default portfolios, including sovereigns and

banks, loss data is too scarce to substantiate estimates that vary

with economic conditions. Consequently, for these portfolios, LGD

estimates are assumed to be constant throughout the projection

horizon.

EAD estimates

EAD values are projected using product specific credit conversion

factors (CCFs), closely following the product segmentation and

approach of the respective IRB model. However, the CCFs are

estimated over multi-year time horizons and contain no regulatory

conservatism or downturn assumptions.

No explicit forward-looking information is incorporated, on the

basis of analysis showing the temporal variation in CCFs is mainly

attributable to changes in exposure management practices rather

than economic conditions.

Credit risk continued

Governance and post model adjustments (audited)

The IFRS 9 PD, EAD and LGD models are subject to NWM

Group’s model risk policy that stipulates periodic model

monitoring, periodic re-validation and defines approval procedures

and authorities according to model materiality. Various post model

adjustments were applied where management judged they were

necessary to ensure an adequate level of overall ECL provision.

All post model adjustments were subject to review, challenge and

approval through model or provisioning committees.

Post model adjustments will remain a key focus area of NWM

Group’s ongoing ECL adequacy assessment process. A holistic

framework has been established including reviewing a range of

economic data, external benchmark information and portfolio

performance trends with a particular focus on segments of the

portfolio that are likely to be more susceptible to high inflation,

rising interest rates and supply chain disruption.

ECL post model adjustments

(audited)

The table below shows ECL post model adjustments.

Significant increase in credit risk (SICR)

(audited)

Exposures that are considered significantly credit deteriorated

since initial recognition are classified in Stage 2 and assessed for

lifetime ECL measurement (exposures not considered deteriorated

carry a 12-month ECL). NWM Group has adopted a framework to

identify deterioration based primarily on relative movements in

lifetime PD supported by additional qualitative backstops. The

principles applied are consistent across NWM Group and align to

credit risk management practices, where appropriate.

The framework comprises the following elements:

on

modelled portfolios, the assessment is based on the relative

deterioration in forward-looking lifetime PD and is assessed

monthly. To assess whether credit deterioration has occurred,

the residual lifetime PD at balance sheet date (which PD is

established at date of initial recognition) is compared to the

current PD. If the current lifetime PD exceeds the residual

origination PD by more than a threshold amount, deterioration

is assumed to have occurred and the exposure transferred

into Stage 2 for a lifetime loss assessment. In broad terms, a

doubling of PD would indicate a SICR. However, the PD uplift

must be at least 0.1%.

– the PD assessment is

complemented with the use of qualitative high-risk backstops

to further inform whether significant deterioration in lifetime

risk of default has occurred. The qualitative high-risk backstop

assessment includes the use of the mandatory 30+ days past

due backstop, as prescribed by IFRS 9 guidance, and other

features such as forbearance support and exposures

managed within the Risk of Credit Loss framework.

The criteria are based on a significant amount of empirical

analysis and seek to meet three key objectives:

– the criteria should be effective in

identifying significant credit deterioration and prospective

default population.

– the criteria should not introduce

unnecessary volatility in the Stage 2 population.

– the criteria should produce results which

are intuitive when reported as part of the wider credit

portfolio.

Monitoring the effect on relative PD deterioration when originating

new lending at times of weaker economic outlook (therefore,

higher PDs at initial recognition) is important to ensure SICR

criteria remains effective.

Asset lifetimes

(audited)

The choice of initial recognition and asset duration is another

critical judgement in determining the quantum of lifetime losses

that apply.

The date of initial recognition reflects the date that a transaction

(or account) was first recognised on the balance sheet; the PD

recorded at that time provides the baseline used for subsequent

determination of SICR as detailed above.

For asset duration, the approach applied in line with IFRS 9

requirements is:

Term lending – the contractual maturity date, reduced for

behavioural trends where appropriate (such as, expected

prepayment and amortisation).

Revolving facilities – asset duration is based on annual

customer review schedules and will be set to the next review

date.

Credit risk continued

Economic loss drivers

(audited)

Introduction

The portfolio segmentation and selection of economic loss drivers

for IFRS 9 follows the approach used in stress testing. To enable

robust modelling the forecasting models for each portfolio

segment (defined by product or asset class and where relevant,

industry sector and region) are based on a selected, small number

of economic variables, (typically three to four) that best explain

the temporal variations in portfolio loss rates. The process to

select economic loss drivers involves empirical analysis and expert

judgement.

The most material economic loss drivers for the UK portfolios

include UK gross domestic product (GDP), world GDP, the

unemployment rate, the house price index, and the Bank of

England base rate. Similar metrics are used for other key country

exposures in NWM Group.

Economic scenarios

At 31 December 2023, the range of anticipated future economic

conditions was defined by a set of four internally developed

scenarios and their respective probabilities. In addition to the base

case, they comprised upside, downside and extreme downside

scenarios. The scenarios primarily reflected the current risks

faced by the economy, particularly in relation to the path of

inflation and interest rates.

For 2023, the four scenarios were deemed appropriate in

capturing the uncertainty in economic forecasts and the non-

linearity in outcomes under different scenarios. These four

scenarios were developed to provide sufficient coverage across

potential rises in unemployment, inflation, asset price declines and

the degree of permanent damage to the economy, around which

there remains pronounced levels of uncertainty.

Upside

– This scenario assumes robust growth as inflation falls

sharply and rates are lowered more quickly than expected.

Consumer spending is supported by savings built up since COVID-

19 and further helped by fiscal support and strong business

investment. The labour market remains resilient, with the

unemployment rate falling. The housing market slows down

compared to the previous year but remains robust.

Compared to 31 December 2022, the upside scenario remains

similarly configured, exploring a more benign set of economic

outcomes, including a stronger performing stock market, real

estate prices, and supported by a stronger global growth

backdrop, relative to the base case view. Reflecting recent

outturn data, inflation falls back quicker and the labour market is

tighter than previously assumed.

Base case

– High inflation and tight monetary policy leads to

muted economic growth. However, continued disinflation allows

an easing cycle to start in 2024. The unemployment rate rises

modestly but there are no wide-spread job losses. Inflation

moderates and falls to a target level of 2% by early 2025. The

housing market experiences modest nominal price decline but the

extent of the decline is lower than experienced during prior

stresses. Housing market activities remain weak but gains pace

gradually as interest rates fall and real income recovers.

Since 31 December 2022, the economic outlook has improved as

energy prices fell sharply and the labour market remained

resilient. The near-term inflation outlook remains elevated and

upside risks remain but they have reduced since last year. Rates

increased to levels higher than expected previously and are

expected to remain higher for longer. Economic growth is still

expected to be muted in the near-term. The base case now

assumes muted growth in 2023 as opposed to a mild recession

assumed previously. The unemployment rate still rises but the

peak is marginally lower and is underpinned by a resilient labour

market The peak to trough house price correction remains

broadly similar to the previous assumption but the timing of the

fall is more spread out.

Downside

– Inflation resurges as energy prices rise and core

inflation remains persistently high. The economy experiences a

recession as consumer confidence weakens due to a fall in real

income. Interest rates are raised higher than the base case and

remain elevated for longer. High rates are assumed to have a

more significant impact on the labour market. Unemployment is

higher than the base case scenario while house prices experience

declines comparable to previous episodes of stress.

Compared to 31 December 2022, the downside scenario explores

risks associated with ongoing price pressures and significantly

higher interest rates across the period. This contrasts with last

year’s scenario, which assumed lower rates than the base case

view. Partly as a result, UK economic activity and labour market

are slightly weaker. Nominal asset prices, while experiencing

declines comparable with past downturns, perform slightly better

than previously assumed.

Extreme downside

– This scenario assumes a classical recession

with loss of consumer confidence leading to a deep economic

recession. This results in widespread job losses with the

unemployment rate rising above the levels seen during the 2008

financial crisis. Rates are cut sharply in response, leading to some

support to the recovery. House prices lose approximately a third

of their value.

Compared to 31 December 2022, the extreme downside again

captures an extreme set of economic outcomes, with very sharp

falls in asset prices and a marked deterioration in the labour

market. The key difference is the assumed path for interest rates.

Unlike at 31 December 2022, when recessionary risks were

explored in the context of a stubbornly high inflation environment,

both inflation and interest rates are now assumed to follow a

significantly lower trajectory – consistent with recession driven by

material weakness in domestic demand.

Credit risk continued

Main macroeconomic variables

The main macroeconomic variables for each of the four scenarios used for expected credit loss (ECL) modelling are set out in the main

macroeconomic variables table below.