Stress testing the UK banking system: Guidance on the 2025 stress test for participants

A1: Overview

A1.1: Introduction

This annex describes the approach that banks are expected to take in the execution of the 2025 stress test with respect to fair valued positions in both trading book and banking book as defined in Section A1.2.^{footnote [10]} More specifically, this annex:

• describes the overall approach that banks should adopt in the execution of the traded risk stress test;
• outlines how the stress scenario should be translated into specific loss numbers^{footnote [11]} and financial and regulatory metrics reported via the templates; and
• defines certain terms and concepts that are used in the templates in the context of the methodology that should be applied.

This annex does not outline the stress scenarios itself, as it is described in the Key elements and the Variable paths for the 2025 stress test and Traded risk scenario for the 2025 stress test.

The traded risk stress-test methodology outlined in this annex expects banks to exercise judgement in the application of the method to their exposures. For example, banks may exercise judgement on the likely time period over which a material, illiquid trading position could be liquidated or hedged under the stress scenario. Banks are expected to explain the judgements that they have made as part of the unstructured data request.

A1.2: Key design features

The Bank’s approach to stress testing traded risk is similar to the approach taken in previous ACSs other than the 2021 exercise, which included some changes due to the different nature of that year’s scenario. The traded risk element of the 2025 stress test incorporates experience of previous historical episodes that is linked to the forward-looking macroeconomic scenario.

The 2025 traded risk scenario is linked to the macroeconomic aspects of the stress test. The market risk factor shocks are broadly aligned to the global and regional impacts of the macroeconomic scenario. Reflecting the Bank’s stress testing framework, the calibration of the shocks takes into account the severity associated with the state of the financial cycle.

The Bank’s approach continues to recognise the importance of market and position liquidity when assessing loss projections under a stress scenario. Banks are expected to apply risk factor shocks that correspond to the likely liquidity of each position under the scenario, and hence the time for which each position is exposed to the scenario.

The Bank’s approach to counterparty credit risk asks banks to identify and default counterparties that are particularly vulnerable to the stress scenario. This approach creates consistency between the counterparty credit risk losses and the macroeconomic stress scenario.

A2: Preliminaries

This section sets out the scope of application and how the different components of the stress test fit together, and outlines several general features of the stress test.

A2.1: Position scope

Broadly, the scope of positions to which the traded risk stress test is applied is: all FVTPL^{footnote [12]} and FVOCI accounted positions. The assets to which the stress is applied can be broken down into several parts as follows:

• All positions that fall within the perimeter of the regulatory trading book.
• All other fair valued items outside the perimeter of the regulatory trading book, including:
  • the FVOCI part of the regulatory banking book, which includes banks’ LABs, and associated hedge positions;
  • the fair value option (FVO) part of the regulatory banking book and associated hedge positions; and
  • other financial assets mandatorily accounted as FVTPL that are not included in the regulatory trading book perimeter, such as underwriting positions and associated hedge positions.

Exceptions to the scope of the traded risk stress are as follows:

• Where a position has a prudential filter that eliminates the impact of changes in its value from capital, then such positions should be omitted in line with the filtering applied in the capital treatment unless explicitly noted otherwise.
• Banking book securitisation positions (per the CRR Chapter 5 definition) and covered bonds are excluded from the traded risk stress test. These are captured as part of the credit stress test but any non-Chapter 5 hedges to these positions should be included. For example CLO hedged with an untranched index CDS would result in the inclusion of losses from the CLO in the credit stress test and the gains from the CDS hedge in the traded risk stress test.
• Securities financing transactions held at amortised cost in the banking book should be included for the purpose of calculating counterparty default losses. This includes all collateral types, even Chapter 5 securities. For clarity, all other types of amortised cost lending are excluded, as they will be captured via the banking book stress test.
• Hedges to amortised cost loans are excluded.

A2.2: Components of the stress test

The traded risk scenario will have an impact on both capital resources (which would be depleted in the event of losses being incurred) and capital requirements (which may increase in response to rises in market volatility and counterparty default risk).

The impact of the traded risk stress test on capital resources is calculated to take into account the separate impacts arising from:

• Market risk losses (described in Section A3) arising in the trading book due to adverse moves in risk factors (market prices and rates) and issuer default.
• Counterparty credit risk default losses (described in Section A4).
• Changes in various valuation adjustments (described in Section A5) such as to the funding valuation adjustment (FVA), and credit valuation adjustment (CVA), which are collectively categorised under the banner of x-value adjustment (XVA) losses.
• Valuation adjustments in addition to XVA, such as bid offer adjustments and regulatory adjustments due to stressed prudent valuation adjustment changes (described in Section A6).
• Other Fair Valued Items losses on FVOCI, FVO and non-trading book FVTPL positions (described in Section A7).
• Revenue and cost changes in the bank’s investment banking business (described in Section A8).

The impact of the traded risk stress test on capital requirements is calculated as the sum of the separate impacts from:

• Market risk and CVA RWAs (described in Section A9).
• Counterparty credit risk RWAs (described in Section A9).

The overall impact on a bank’s capital ratios will reflect the impact of the traded risk stress test on both capital resources and capital requirements.

A2.3: Effective date

The stress test should be applied to banks’ fair value positions as of a specified effective date. The effective date for running the stress test is different for different components of the traded risk stress test (and hence for the corresponding templates), as indicated in Table A1.A.

An effective date of 28 Feb 2025 was chosen for market risk, counterparty credit risk and XVA exposures because banks typically reduce their traded positions at year end. Using this chosen date is more likely to provide a representative view of banks’ traded risk positions.

A2.4: Reporting currency

For traded risk positions that would generate profit and loss (P&L) under the stress scenario in currencies other than banks’ reporting currency, such P&L should be translated into the bank’s reporting currency via FX spot rates that are consistent with the stress scenario and the liquidation horizon of the positions that generate the P&L, which determines the time at which the foreign currency P&L is generated and therefore the rate at which it should be converted into the reporting currency.

A2.5: Loss allocation and relationship to management actions

The stress-test horizon is five years and, in line with this, banks should model the stress impact on the fair value positions that are outside of the regulatory trading book, the impact on PVA for positions held in the banking book and the impact on investment banking revenues and costs for each year of the stress scenario. Further details on this are provided in the relevant sections of this annex.

In relation to market risk, counterparty credit defaults, XVA movements and PVA movements on trading book positions, banks should assume that all losses are incurred in the first year of the stress. This is because losses on trading activities would typically be concentrated in the early part of a stress scenario, since market prices tend to reflect worsening conditions relatively quickly.

The allocation of losses over the five years of the stress scenario is summarised in Table A1.B.

Consistent with the overall stress-test results only being collected at an annual frequency, traded risk projections are also annual. However, the intrayear distribution may impact the timing of any assumed management actions, and as a point of reference banks should equally distribute the full-year losses across the four quarters and take this as a floor to possible actions. Banks should then motivate their actions by reference to the liquidity horizon of the positions, and the evolution of the underlying market as represented in the scenario, subject to this floor.

For example: in a real period of market stress, liquid market risk losses may manifest in only a short interval of a few days but structural liquid and illiquid losses will arise over several quarters; uncollateralised counterparty losses are subject to one-year shocks because it is expected these defaults will not occur immediately but only on a lag in quarter four; losses on bond holdings in the Liquid Asset Buffer may occur during the scenario but the extent of bond sales will be motivated by the information available up to the point of sale and not with foresight of future interest rate movements. As a result, some losses incurred in the first year of the stress event may be weighted towards the latter end of the first year of the stress rather than being equally distributed across the quarters. The timing of any management actions that are necessitated by these losses are therefore expected to be late in the first year. An action should not be motivated by an allocation of losses to quarter one that is larger than would occur under an equal-quarters loss allocation. This applies to both business-as-usual and strategic management actions.

Section 11 provides guidance on the difference between strategic and business-as-usual management actions. Traded risk projections should only include business-as-usual management actions and these should be motivated by precise policies and procedures that support the business-as-usual actions eg to stay within limits, to meet enforced limit reductions under stress or in response to activated stop-loss triggers. Traded risk strategic management actions should be recorded alongside banks’ other strategic management actions.

A3: Market risk stress

A3.1: Position types

Banks’ trading books comprise trading positions of varying liquidity. As was apparent in the global financial crisis, the most illiquid positions can inflict the greatest damage to banks’ P&L and capital resources. For this reason, banks are expected to clearly identify illiquid positions and distinguish. For the purpose of the traded risk stress test, banks are requested to classify trading book positions into three categories:

1. Liquid positions are defined to be those which would take two weeks or less to liquidate or hedge under the stress scenario.
2. Illiquid positions are defined to be those that would take more than two weeks to liquidate or hedge under the stress scenario. This longer liquidation period may arise due to the bespoke features outlined in Section A3.6.
3. Structural liquid positions are a further designated position type that is intended to capture positions which, although possibly reduced or neutralised when an adverse stress scenario has its initial impact, may need to be subsequently reopened in order to preserve a bank’s ability to provide financial products in a particular market, for example market‑making positions. By virtue of reopening such a position, a bank exposes itself to further losses associated with further adverse market movements.

Stresses applied to Structural liquid and Illiquid positions are incremental to the stress applied to Liquid positions.

A3.2: Assessment of position liquidity

Banks are expected to make their own assessments of the liquidity horizons of their positions. More specifically, banks should judge how quickly they would be able to exit positions in view of likely market trade volumes under the stress scenario. The Bank will assess banks’ judgements regarding the liquidity of their traded positions.

A3.3: Calibration of risk factor shocks

The risk factor shocks that comprise the traded risk scenario are included in Variable paths for the 2025 stress test and in the ‘Traded risk shocks’ tab of the traded risk scenario for the 2025 stress test. The Bank is specifying a core set of risk factor shocks that are intended to induce an overall shock to the entire set of in‑scope positions. The Bank has specified a number of key risk factor shocks in each material geography and market to provide a secure foundation for the elaboration of the stress scenario in terms of all risk factors that would drive banks’ P&L. Moreover, risk factor shocks are specified for a range of different liquidity horizons.

However, the risk factor shocks provided by the Bank do not include all risk factors to which banks are exposed, and so banks are expected to identify other risk factors that would contribute to their P&L under the stress scenario and to calibrate shocks for these risk factors. These risk factors should be identified based on banks’ understanding of the material risk factors that would be expected to drive P&L under the stress scenario. Further, these additional risk factor shocks should be calibrated with reference to the risk factor shocks and scenario narrative that have been provided by the Bank. If this proves insufficient, banks should gauge the severity of shocks applied to these factors with reference to large moves observed since 2005. Banks should include commentary on how shocks were determined for risk factors that have a material impact on stressed projections in the unstructured data submissions.

Whether market risk factor shocks are provided by the Bank or identified and calibrated by banks themselves, banks should apply the shocks appropriate to the liquidity of each position. The Bank will assess the appropriateness of the shocks that banks apply to their traded positions.

When applying risk factor shocks to any part of their portfolios, banks should consider whether the resulting losses are realistic. Where the profit or loss is material and unrealistic banks should highlight this and provide a realistic assessment of stress results (eg where the size of a position under stress would exceed limits and necessarily be reduced or hedged).

The remaining parts of this section describe the approach that banks are expected to take in the calculation of loss per position type in greater detail.

A3.4: Liquids stress

Having identified all the risk factors that drive the P&L of liquid portfolios, banks should apply the risk shock (whether supplied by the Bank or calculated by the banks themselves) appropriate to the liquidity of each risk factor and thereby obtain the total loss generated by liquid portfolios under the stress scenario. This is to be reported in the ‘Liquids’ column of the ‘Totals’ tab in the ‘Market Risk Stressed P&L’ template. The total loss should be disaggregated and reported at the level of granularity specified in the template.

A3.5: Structural liquids stress

Structural liquids positions may suffer a loss at the onset of a stressed market environment. This is likely to cause a bank to reduce its inventory in the associated products. However, for the franchise reasons noted in Section A3.1, such positions may be reopened and thereby expose the bank to further losses associated with adverse market moves later in the stress scenario. The approach banks are expected to take is detailed as follows:

• Banks should identify desks or position types that are significant for strategic reasons, eg they require a minimum level of inventory in order to maintain a credible market‑making franchise. For example, this could be a bond or swaps market‑making desk whose relative standing in the market (as indicated by rankings or otherwise) needs to be preserved.
• For each such desk or position type, and the risk factors they are exposed to, banks should identify the risk factor that typically has the greatest market risk and identify a typical level of exposure to it. This may coincide with the value as of the effective date or be a representative trailing average calculated as of the effective date.
• As the exposure will be present throughout year one, albeit potentially run down and replenished throughout on a rolling basis, it is reasonable to consider that a longer liquidity horizon, and as a result a larger shock should be applied to this position. This is because even though the position could in principle be liquidated faster, the size of the position is not discretionary because of its strategic importance for the overall franchise. Therefore, the loss should be calculated by following a two‑stage procedure as follows.
• In stage one the loss should be estimated by applying the risk factor one‑year shock to the typical structural liquid exposure and adding together the losses from each of the structural liquids identified. The one‑year shock should not be downscaled to account for the proportion of the shock already suffered in the Liquids stress (eg if the risk factor has a one‑day liquidity horizon and the one‑day shock is 20%, while the one‑year shock is 30%, the Structural liquids shock to be applied is 30% and not 30% minus 20%). The rationale for this is that the overall size of the one‑year shock is used as a proxy measure to capture the effect of multiple repeat losses and also to account for any significant deviations in exposure away from the typical level.
• In stage two, banks should assess whether there are any material artefacts in the loss that make it unrealistic. For example, material gains that would not occur in a real stress and are a by‑product of using a point in time stress approach. When identifying such artefacts banks should consider, but not be limited to, the following:
  • significant differences between the inventory size on the effective date and the typical size;
  • changes to the P&L if the one‑year shock were to be realised over the period of several days, rather than instantaneously;
  • the cost of re‑establishing positions at (increasingly) stressed levels over the course of a year; and
  • whether option positions would be re‑established at current strikes as the stress progresses.

As an example, if a firm expects to be persistently carrying a certain amount of short‑dated variance swap or option risk with an average expiry of three months then the application of a one‑year shock with no offsetting adjustments would not be realistic. The bank should consider the instances where it would have to rollover the three‑month position and the fact that the purchase price may be increasing, and use this to adjust the one‑year shock results.

It is not considered necessary at the current time for banks to model the detailed intra‑year profile of risk to combat the artefact problem. However, banks should assess the results for the existence of material artefacts, identify and report them in their submissions, and make approximate adjustments for their effect. The Bank does not expect banks to be generating large gains from structural liquids.

A3.6: Illiquids stress

The loss sustained by each portfolio of illiquid positions should be identified separately and reported in the market risk template. Banks should clearly articulate their approach to the identification of illiquid portfolios. As noted in Section A3.1, a position is designated as illiquid if it is likely to take more than two weeks to liquidate or hedge under the stress scenario. For guidance purposes, examples of illiquid positions are provided as follows:

• Positions that would take longer than two weeks to liquidate or hedge, whether complex or not. This could, for example, include a corporate bond held in large size relative to the amount of the bond in issue.
• Positions for which there are only thin or one‑way hedging markets available, and so the ability to ascribe a liquidity horizon to the position may be compromised.
• Positions that are difficult to value and consequently may have significant non‑modelled characteristics that are not captured in the stressed value such as legal enforceability risk and rating downgrade contingencies.
• Positions for which values may be modelled, but with significant uncertainty.

Banks should articulate their approach when calculating the Illiquids stress‑test loss in sufficient detail to allow the Bank to understand, in respect of each illiquid portfolio:

• the nature of the positions that comprise the portfolio;
• the risk factors that drive portfolio P&L;
• the risk factor shocks utilised (and how they were calibrated to be consistent with the scenario);
• the details of the stress loss calculation applied;
• the loss outcome itself; and
• which trading desk manages the portfolio.

In identifying the risk factors that drive P&L of illiquid portfolios and in calibrating the corresponding risk factor shocks, banks should take due account of:

• The risk factor shocks and scenario narrative published by the Bank.^{footnote [13]}
• The market structure and dynamics for the products that comprise the illiquid positions. Banks are expected to take into account that illiquid product valuations are heavily influenced by other broker‑dealer activity.

Banks should review their results for material artefacts, disclose any that are identified and apply appropriate adjustments.

The Bank does not typically expect banks to generate large gains from illiquids in the stress.

A3.7: Issuer default

The market risk template includes a tab relating to ‘Issuer Default’ losses. Such losses would be associated with those counterparties identified as defaulting in the counterparty credit risk stress described in Section A4.^{footnote [14]} That is, if a counterparty were to default under the counterparty credit risk stress, then any issuer exposure to that name arising in the trading book (from bonds, equities, traded loans, and derivatives where the defaulting counterparty is referenced as an issuer, eg CDSs) should also be assumed to default and be reported in the ‘Market Risk Stressed Profit and Loss’ template.

A4: Counterparty risk default stress

This section discusses counterparty default loss, which comprises two parts: portfolio‑wide default losses across particular cohorts of clients, and additional losses arising from the default of specifically named, large counterparties that are deemed to be vulnerable to default under the stress scenario. Banks should detail the rationale for choosing which counterparties to default under the stress scenario (both in terms of the cohorts and specific names).

A4.1: Definition of vulnerable counterparties

The selection of vulnerable counterparties requires expert judgement regarding the creditworthiness of counterparties, and banks are expected to consider multiple factors in making this determination. For example, banks should consider both the current creditworthiness of counterparties, and how that creditworthiness might deteriorate under the stress scenario. Banks should also consider the nature of the exposure and, in particular, whether it exhibits wrong‑way risk. Therefore, the selection of vulnerable counterparties should not be based solely on simple application of measures such as probabilities of default (PDs) (or external ratings), but should also take into account idiosyncratic credit factors arising from the stress scenario itself.

A4.2: Portfolio default losses

Regarding portfolio losses, banks are expected to:

• Stress significant cohorts as specified in the scenario. The significance of a cohort should be judged in terms of both the materiality and the vulnerability of the exposure under the stress scenario.
• Estimate a cohort default loss that would arise from a portion of this portfolio defaulting at the end of the first year of the stress scenario, and with no further losses beyond the one‑year point. Banks should estimate this cohort default loss as follows:
  • Calculate the stressed exposures of the uncollateralised counterparties in the cohort by applying one‑year market risk factor shocks. For collateralised counterparties, the stressed exposure should equal the current exposure on the effective date.
  • Calculate the stressed expected loss, using market‑implied stressed PD and loss given default (LGD) rather than those used to project impairments in the banking book.
  • Using the stressed PD implied from the cohort’s stressed expected loss, estimate the proportion of pre‑stress CVA that relates to the defaulted portion of the overall cohort and deduct this from the stressed expected loss to arrive at the cohort default loss.

A4.3: Specific name default losses

Banks are also expected to default a number of specifically named, vulnerable counterparties under the stress scenario. Details of the minimum number of counterparties that banks should default will be provided as part of the traded risk scenario. The approach to determining the default loss varies according to whether a bank’s exposures to a counterparty are collateralised or uncollateralised.

For uncollateralised counterparty losses, banks should:

• Estimate stressed current exposure by applying one‑year market risk factor shocks and assuming the default occurs at the end of the one‑year period (and with no additional losses beyond the one‑year point).
• Identify and rank their top exposures by stressed current exposure.
• Identify and default vulnerable counterparties from these rankings according to the minimum numbers set out in the scenario. A bank should default more than the minimum number of counterparties if it deems that more than the minimum number are likely to default under the scenario.
• For calculating default losses, use the severity rate from the banking book analysis to inform their choice of LGD, with appropriate consideration of the specific name being defaulted.

For collateralised counterparty losses, banks should:

• Assume the counterparty does not post any additional margin or honour existing margin calls that are still unpaid.
• Assess the total time to close out all the open positions for each of the counterparties, including allowance for any delays in exercising collateral rights, liquidating illiquid collateral or covering illiquid market risk exposures. As a result, this close out period may not be the same for all counterparties.
• Apply market risk shocks to the exposures and collateral that are appropriate to the close out period identified.
• Calculate stressed current exposure for each counterparty.
• Rank the top exposures as detailed in the traded risk scenario. Banks should rank their counterparties by stressed current exposure (net of stressed collateral).
• Identify and default vulnerable counterparties from these rankings according to the minimum numbers set out in the traded risk scenario.
• Note that banks should use the severity rate from their banking book analysis to inform their choice of LGD, with appropriate consideration of the specific name being defaulted.

Where a counterparty is treated as having defaulted, no additional impact on the market due to the default of that name needs to be modelled, and the pre‑stress CVA should be deducted from the default loss. For all counterparties chosen to default, banks should consider the impact on other templates consistent with guidance in Section A3.7 and Section A7.1.

A5: Stressed XVA

Banks’ fair value positions are subject to various types of valuation adjustments that will be impacted by the scenario, and so the following sections provide guidance to banks on how these adjustments should be modified under the stress scenario.

A5.1: Credit valuation adjustment (CVA)

In their trading activities banks enter into derivative contracts with counterparties. If a derivative contract gives rise to credit exposure for a bank ‒ in other words, the contract has produced or may produce a mark-to-market profit for the bank ‒ then there is a risk that the counterparty will default and fail to pay what is owed under the contract. The CVA measures the negative adjustment to the contract’s value today in order to take account of this risk of default by the counterparty. Under the scenario, credit quality will deteriorate for some counterparties and credit spreads will widen and so the CVA should be modified to reflect this.

CVA should be reported in three traded risk templates, with consistency between the entries:

• The ‘Counterparty Credit Risk Losses’ template should show CVA before and after the application of the risk factor shocks and exclusive and inclusive of all associated hedges (credit and market risk hedges).
• The ‘Stressed XVA projections’ template should report the change in the CVA under the stress both with and without associated hedges.
• The ‘Stressed PVA projections’ template should report the CVA as a related fair value adjustment on the ‘Totals’ and ’Unearned Credit Spreads’ tabs.

Banks are asked to note the following when calculating the CVA stress impact:

• When calculating the adjustment to CVA to reflect the impact of the stress scenario, banks should maintain consistency with the calculation of CVA in their accounts. Specifically, banks should use either market‑implied or actual measures of PD and LGD, in line with their accounting CVA.
• Shocks to the credit spread and funding spread risk factors that drive CVA should be calibrated to a one‑year liquidity horizon for both CVA and the associated credit risk hedges in place at the effective date, regardless of the frequency of hedge‑adjustment used by the CVA hedging desk.
• For collateralised counterparties, banks should assume the counterparty continues to post additional margin.
• Banks should pay particular attention to the more complex CVA risks, such as index/single‑name proxy basis and cross-gamma between market and credit risk factors. Further to this, in specifying the credit‑spread shocks for individual counterparts, banks should conservatively explore how proxy hedges may react differently from the underlying credit and how the maturity of hedges may differ from the underlying exposures.
• Banks should decompose the aggregate CVA loss in their accompanying submissions so that the incremental contributions of these bespoke illiquid CVA risk factor shocks are apparent.
• Banks should provide detailed commentary on the resulting CVA adjustment to support the calculations that they have made.

A5.2: Debit valuation adjustment (DVA)

In symmetry with CVA, which adjusts valuations to account for the risk of counterparty default, the DVA adjusts valuations to reflect variations in a bank’s own credit quality.

The approach that banks are expected to follow in respect of DVA under the stress test requires that any impact of DVA is not recognised in the ultimate bottom line loss reported in traded risk templates. This is because regulatory capital treatment assumes that any DVA benefit cannot be realised and so any impact of DVA is not recognised in the calculation of regulatory capital resources. Nonetheless, because of the complications of how DVA is related to and managed alongside FVA and particularly in circumstances where a bank is hedging its DVA, banks are asked to report DVA gross in the XVA template and show the explicit deduction taken to remove the DVA in the bottom line loss number. Hedges are also separately included.

A5.3: Funding valuation adjustment (FVA)

The stress scenario will impact a bank’s own cost of funding and FVA, to the extent that funding costs are partly or wholly reflected in the bank’s mark‑to‑market accounting. Banks should ensure that this funding loss is included in the XVA template. To determine the loss, banks should estimate their stressed funding curve in line with the overall narrative and severity of the macroeconomic scenario, and with the price paths provided in the scenario. This stressed funding curve should then be used to determine any fair values that are a function of it, in line with banks’ existing valuation methodologies.

To the extent that there is also a PVA against funding costs (specifically, the Investment and Funding Cost component of PVA), then there may be additional capital erosion due to changes in PVA under the stress scenario. This additional PVA amount should be calculated according to banks’ existing methodologies and reported in the Stressed PVA template. Further details are provided in Section A6.

A6: Stressed prudent valuation adjustment (PVA)

The scope of the traded risk stress test is fair‑valued positions. However, accounting fair value may fall short of what would be considered prudent in the context of regulatory capital resources. For example, when valuation of a security is subject to a large degree of uncertainty ‒ perhaps because liquidity in the market for the security is thin ‒ fair value would require the security to be marked within the range of possible prices for the security, whereas prudence would require the security to be marked at a lower (upper) estimate of price if the position were long (short).

The scope of PVA stress includes all components of PVA as set out in the CRR, namely Market Price Uncertainty Additional Value Adjustment (AVA), Close‑Out Cost Uncertainty AVA, Model Risk AVA, Concentrated Position AVA, Unearned Credit Spreads AVA, Investing and Funding Cost AVA Future Administration Cost AVA, Early Termination AVA and Operational Risk AVA. It also includes the accounting bid/offer reserve stress.

Banks should project each component of PVA consistently with the scenario and where necessary maintain consistency with accounting fair value adjustment projections already reported in other templates eg for CVA, FVA. Projections for accounting fair value adjustments related to components of PVA should also be reported on the PVA template.

For trading book related losses or deductions (ie increases in fair value adjustments or PVA in relation to FVTPL trading book positions), the resulting losses or deductions should be allocated to year one with no recovery assumed in subsequent years.

For banking book related losses or deductions (ie PVA in relation to FVOCI or FVO positions), the resulting losses or deductions should be projected over the scenario horizon in accordance with conditions implied by the scenario.

A6.1: PVA projections under stress

PVA is motivated by the concept that there is often a range of values when estimating the fair value of a position. This valuation uncertainty range may change when market conditions change. Therefore, when projecting PVA, banks should apply this principle and design their methodology to capture the changes in valuation uncertainty in the market as implied from the scenario.

We expect that banks will utilise their existing PVA framework to project future PVA in stress. Therefore, the level of granularity of the analysis will be the same as for PVA that is calculated in the ordinary course of business.

For example, when projecting Market Price Uncertainty and Close‑Out Uncertainty AVAs for interest rate swaps, banks should take into account whether a sharp rise in an interest rate curve may lead to increased valuation uncertainty in the market price and bid‑offer spread for this product.

As another example, when projecting Concentrated Position AVA in stress, banks should incorporate the liquidity horizon assessment described in Section A3 so as to identify any concentrated positions that might arise due to a change in market liquidity under the stress scenario.

A6.2: Fair value adjustment projections under stress

Several accounting fair value adjustments are reported in the PVA template, including the bid offer reserve. This is necessary whenever PVAs rely on accounting fair value adjustments as a starting point. Where such adjustments are also captured in the XVA template over the same projection horizon the reported values should be consistent.

Banks should also utilise their existing fair value adjustments framework as much as possible to project future fair value adjustments in stress. The level of granularity of the analysis, where applicable, should be the same as for fair value adjustments that are calculated in the ordinary course of business.

For the bid‑offer reserve stress, banks should assess the impact on bid‑offer spreads arising from the scenario, applying the level of granularity that they would apply to their own internal analysis and using their own netting method.

For XVA, the detailed changes should be captured in the XVA template but a high‑level summary should also be recorded in the PVA template to allow holistic analysis on Unearned Credit Spread PVA and Investing and Funding Cost PVA. Specifically, the approach for stressing funding costs should be identical to that laid out in Section A5.3 and banks should use the same stressed funding curve.

A7: Other Fair Valued Items (OFVI)

The ‘Other Fair Valued Items projections’ template is intended to capture positions measured at fair value which reside outside of the regulatory trading book. It is intended to be a comprehensive balancing item to capture a wide variety of fair valued items whose impact on capital resources would otherwise not be captured in other traded risk templates.

For the bid‑offer reserve stress, banks should assess the impact on bid‑offer spreads arising from the scenario, applying the level of granularity that they would apply to their own internal analysis and using their own netting method.

For XVA, the detailed changes should be captured in the XVA template but a high‑level summary should also be recorded in the PVA template to allow holistic analysis on Unearned Credit Spread PVA and Investing and Funding Cost PVA. Specifically, the approach for stressing funding costs should be identical to that laid out in Section A5.3 and banks should use the same stressed funding curve.

A7.1: OFVI projection assumptions

Losses for OFVI positions under the stress scenario should be calculated with respect to each year of the scenario. Banks only need to provide annual losses for the 2025 stress test; there is no need to break down losses during the first year into first month and quarterly losses.

In constructing the stress scenario to be applied to the OFVI positions, banks are expected to refer to:

• the macroeconomic scenario, published in the Key elements of the stress test and Variable paths for the 2025 stress test, which provide full paths for a small number of the market risk factors relevant to OFVI positions; and
• the ‘Traded risk shocks‘ tab of the traded risk scenario for the 2025 stress test, which provides more detailed risk factor shocks for the first year of the scenario, for more of the risk factors relevant to OFVI positions.

Banks are expected to infer from these parts of the Bank’s stress scenario the complete scenario horizon that should be applied to OFVI positions.

For all OFVI positions except the Liquid Asset Buffer, the balance sheet size should be held constant with no ageing or changing of positions. Where banks have in place written procedures requiring the sell down of foreign currency gains or losses from OFVI positions, then banks should follow these procedures in their stress‑test calculation. This is the only type of rehedging permitted in stress testing OFVI positions that are not part of the Liquid Asset Buffer.

Different treatment of Liquid Asset Buffer positions is permitted and should be considered in two stages:

1. At each period end banks should revalue the positions they held as at 31 December 2024, and thereby produce gain or loss projections under the scenario. In calculating the valuations for each period, banks should not age nor change any of the positions. For instance, if a bank holds a ten‑year gilt this position should be revalued each year end as a ten‑year gilt; it should not be revalued in year one of the stress scenario as a nine‑year gilt. This will be reported in the pre‑management action area of the template.
2. The buffer may be adjusted in accordance with justifiable business‑as‑usual management actions. Where an action applies, the bank should report the adjusted gains or losses in the post management action area of the template.

Admissible changes to the buffer under a business‑as‑usual management action must be fully supported by appropriate policies and procedures and evidence of how these are invoked eg with regard to monetisation of the buffer or investment changes due to stop‑loss triggers. Actions meeting the definition of a Strategic Management Action, as set out in Section 11 of this document, must not be included. Unstructured information concerning the business‑as‑usual management action must also be provided in the unstructured data submission, as detailed in the Basis of Preparation.

Note the following points of clarification regarding the treatment of the default risk of OFVI positions:

• The ‘Counterparty Credit Risk Losses’ template only covers derivative and Security Financing Transaction counterparty defaults, and excludes both unsecured lending and issuer defaults on bond and equity holdings. Positions where the loan is designated at fair value under FVO are also excluded. No default losses should therefore be reported in the Counterparty Credit Risk Losses template for OFVI assets. These should instead be reported in the ‘Issuer Default Loss’ tab of the ‘Other Fair Valued Items’.
• However, counterparty default losses on derivative hedges to OFVI items should be reported in the Counterparty Credit Risk Losses template, as this template covers all trading book and banking book derivatives.
• Unlike market risk losses on OFVI positions, which are allocated across the full five years of the stress scenario, default losses for OFVI positions should be allocated to year one of the stress scenario.

For private equity investments in OFVI, banks should as a starting point consider the methodologies used in their current valuation approach, for example their pre‑existing choices of comparable assets (eg listed securities), and any adjustments already taken into account for the difference between the position held and a comparable listed asset. Application of the stress scenario may require approximations such as the use of betas to simplify one or more of the steps in the valuation approach, when applied under the stress scenario. Where these approximations are employed, they should be calibrated to relevant historical reference periods. Banks’ methodology should also consider any impairments under the stress scenario.

A7.2: Additional note for underwriting commitments

Banks should use the ‘non‑trading book positions mandatorily at fair value through profit or loss’ template category to capture any other in scope fair valued items that have not been otherwise captured.

Underwriting commitments in the firm’s pipeline, including those in the process of syndication, should be included in scope. This includes equity, bond, loan and securitisation pipelines that are FVTPL, as well as all FVTPL hedges against these commitments. An example of equity commitment risk would be the underwriting of rights issues. The securitisation pipeline refers to whole loans warehousing, gestation repo, or other pre‑issuance activity where the associated exposure is FVTPL and not subject to amortised cost accounting; if accounted for at amortised cost, then the exposures should be excluded.

In this context, loan commitments refer to conditional agreements to proceed to full loan documentation, where the commitment has a fair value, but is not yet fully documented or funded.

The loan underwriting syndication timeline in particular is often complex and proceeds through various documentary stages that are often completed before the recognition of a credit agreement and the resulting recognition of credit RWA.

Banks should rely on their internal risk management definitions to determine the moment when they consider themselves to be ‘on risk’, which may be synonymous with the recognition of an accounting fair value for the commitment or the existence of a signed legal agreement (at least signed by the bank), and is also likely to be before the recognition of any RWA. Banks do not have to include unsigned or soft commitments unless they believe there is a necessary franchise reason to honour these commitments.

When projecting the loss for underwriting positions, banks should follow the same principles outlined in A7.1 to construct shocks to valuation inputs such as credit spreads and equity prices, taking account of any contractual mitigants such as flex and fees. Each commitment should be assessed individually to take into account its size and idiosyncratic risk particularly where the commitment amount is large. The balance sheet for the positions should be held constant. For banks that have fair value hedges to their commitment positions, these positions should be stressed separately in accordance with the scenario and should not a priori be assumed to be fully effective unless the scenario allows for this.

A8: Revenue and cost projections

Banks should provide stress scenario revenue and cost projections for the IFRS operating segments that include investment banking activities such as trading and capital markets activity, and also for non‑core segments if relevant. Investment banking activity is defined as one or more of the following items:

• Markets: cash and derivatives trading activity including for example products such as FX, Rates, Credit, Equities, Commodities and Prime Finance.
• Capital Markets: activity such as Advisory, Debt Capital Markets, Equity Capital Markets, and Syndicate desks.
• Banking book activity that is readily identifiable inside the bank as supporting Markets or Capital Markets activity, and which is internally managed alongside it with this exclusive aim eg a dedicated relationship lending book for large corporate or institutional clients. If there is no such clear segregation then this activity can be omitted from reporting in this template.

The traded risk templates capture separate income statement information at a more granular level than these segments, focusing investment banking activities in isolation. This allows review of the underlying, bottom‑up assumptions that have been used to build the stressed projections. Banks are expected to present both the aggregated projections at operating segment level and more granular views at business line level and to assign direct and indirect costs at a level that is consistent with their business-as-usual processes.

The income and expense projections should reflect the plausible execution of a bank’s business plan under the stress scenario and be consistent with the stress assumptions made for RWAs.

Banks should assess the impact of the scenario on trading and capital markets activities separately with an appropriate level of granularity. Different business lines are likely to perform differently in the stress scenario, and banks should document the key assumptions that drive performance.

Banks should prudently project revenues that could realistically be earned during the stress. The bid/offer widening assumptions to calculate the bid/offer stress in in Section A6.2 do not apply. Banks should also justify the use of any caps or floors in their approach eg in maintaining certain revenues flat at year zero levels with no decreases below this level. Banks should not assume reduced competition in the investment banking sector as a consequence of the stress scenario.

A9: Risk‑weighted assets projections

Banks should submit more granular information on their starting traded risk^{footnote [15]} RWAs (ie as at the effective date defined in Section A2.3) and projected traded risk RWAs under the stress scenario for each year‑end date over the time horizon via the following two structured data templates:

• Market Risk and CVA RWA; and
• Counterparty Credit Risk RWA.
• This information is used to supplement the projected traded risk RWAs provided in the capital projections template.

The Market Risk and CVA RWA template captures starting and projected components of capital requirements for both market risk and CVA risk, while the Counterparty Credit Risk RWA template captures a breakdown of starting and projected capital requirements for counterparty default risk by counterparty group and exposure type. Other traded risk related components of RWA (such as settlement risk and large exposures) are not captured in the traded risk templates, but are captured in the capital projections template.

A9.1: General guidance

The starting values as at the effective date should reflect reported year‑end values corresponding to the prescribed time period of the stress test. In addition:

• For the stress scenario, banks should reflect a plausible execution of a bank’s business plan under the stress scenario if the business plan remains feasible. Otherwise, the projections should reflect a plausible variation to the bank’s business plan, where these variations are clearly identified and where they have been appropriately assessed for inclusion against the BAU management action criteria in Section 11.
• For both the stress scenario, banks should be consistent with balance sheet, income and expense growth assumptions. Specifically, an increase in projected balance sheet size as a result of increased trading business is expected to result in an increase in projected traded risk RWAs. Similarly, a bank’s plans to increase traded risk appetite should be reflected in an increase in projected traded risk RWAs.
• It is expected that traded risk RWAs submitted in the ‘Market Risk and CVA RWA’ and ‘Counterparty Credit Risk RWA’ templates are projected using a continuation of hedging practices documented and in place in year zero. Additional hedging in response to scenario shocks should be assessed against the management action criteria and only included in projections where it is a business‑as‑usual action supported by appropriate policies and procedures that existed at year zero.
• Changes in market variables such as foreign exchange rates that have a material impact on market risk, CVA risk or counterparty credit risk RWAs must be taken into account when calculating projected traded risk RWAs.

A9.2: Specific guidance

Further details of the methodology that banks are expected to apply in the production of RWA projections under the stress scenario are provided in Table A1.C: