Bank balance sheet

The balance sheet is a snapshot in time of the bank's financial strength. Note that the capital amount must, at all times, be more than sufficient to absorb customer loan losses (or losses incurred for other reasons, such as the result of proprietary trading losses or ineffective hedging resulting in unexpected losses). However, the reality of government regulation, which every bank in the world is subject to, means that banks don't need sufficient capital to absorb only losses. They also need sufficient capital to absorb losses and still, after such absorption, be able to demonstrate capital levels that are above the regulatory minimum. Banking is all about confidence. If customers have confidence in the bank, they will continue to place their funds on deposit there. If they do not, they will not. A bank that dips below its regulatory minimum capital ratios, even if it has still absorbed all its losses to date, will not be able to maintain that confidence.

In other words, good capital management, and indeed considered by some practitioners as best‐practice capital management, is all about managing capital on a going concern basis. The Board of a bank takes responsibility to articulate its risk appetite such that this requirement is met. It is generally considered bad practice to manage capital on a gone concern basis. The available capital, the actual amount that can be used to absorb losses, is the surplus above the minimum required by the regulator. Capital that meets the regulator's requirement is not, in truth, available to absorb losses on a going concern basis. It is essential that bank Boards and executive directors manage the institution on this basis.

Invariably banks may fail. In such circumstances the capital sources must be adequate to recover the bank or, if this is not possible, resolve the bank without undue stress to the financial system. This, however, is more feasible in theory than in practice. If a bank fails, one can assume, to a reasonable and safe extent, that its capital base will be insufficient to recover it, and for large banks to resolve it either.

Sources of capital

Core equity Tier 1 capital consists in principle of share capital, share premium, and retained earnings attributed as regulatory capital. Certain deductions apply.

The features that make a long‐dated liability eligible as Additional Tier 1 (AT1) and Tier 2 (T2) are as follows:

AT1:

• Perpetual; not callable prior to year 5;
• Non‐cumulative, discretionary distributions;
• Deeply subordinated;
• Conversion to equity or principal write‐down at a trigger of CET1 < 5.875% (or higher);
• Conversion or write‐down at the point of non‐viability, where the point of non‐viability is determined at the discretion of the SARB prior to the failure of the bank.

T2:

• Minimum maturity of 5 years (with capital credit amortising 20% per year 5 years prior to maturity);
• Subordinated;
• Conversion or write‐down at the point of non‐viability.

Given the requirements of Basel III, at a strategic level the main ingredients of capital planning are essentially given: that is, the attachment points for maximum distributable amounts can be set quite easily. The most transparent way to illustrate this is to assume that one is setting up a bank from scratch, as shown in Table 15.1, with the theoretical illustration at Figure 15.4 and 15.6.

The Pillar 2A add‐on is stipulated by the national regulator.

In other jurisdictions, as in the UK, the PRA has stated that the Pillar IIA add‐on of up to 2% must comprise at least 56% CET1 and the remainder of a combination of AT1 and T2. For a new bank, the minimum capital requirement would allow the requirements under Basel III to be met, as well as to allow for sufficient capital should a countercyclical buffer requirement be implemented by the regulator SARB.

Hypothetical example: vanilla commercial bank

To illustrate the considerations involved from a first‐principles basis, an assumed example of a UK commercial bank that is being set up from scratch, with an inherited portfolio, is considered. What would be the primary factors driving the capital planning process? The balance sheet RWA breakdown is shown in Figure 15.5.

The main factors that (say) a ratings agency review would consider include: asset mix (for example, whether there is a concentration in assets such as commercial real estate (CRE)); advanced – vs foundation – IRB being applied; Pillar II impacts; stress buffers required; and any capital release opportunities. With a total balance sheet RWA of just over GBP 13 billion, the bank is below 1% of UK GDP in size, so the countercyclical, globally systemically important institutions (G‐SIFIs), and ring‐fence buffers do not apply. Hence, the capital considerations that must be accounted for are shown below.

This argues for an indicative capital structure of 15%. However, as this is a fairly small portfolio, the individual capital guidance received from the regulator will more likely impose a higher requirement. This would be a working assumption taking into account:

• Regulator feedback (for example, the requirement for the amount of total loss‐absorbing capacity);
• Potential impact of stress tests on the capital plan. Such stresses are then accounted for to a certain level of confidence using economic capital principles;
• Peer‐group analysis;
• Market expectations;
• Rating agency feedback.

Assuming the 15% capital base is accepted as a minimum requirement, there are two main scenarios that present themselves. The first is Scenario 1, the all‐equity scenario, while Scenario 2 describes an equity plus other capital structure. The latter is shown in Figure 15.7.

A capital base as shown in Figure 15.7 would enable the bank to meet both the standing buffer requirements and the macro‐prudential one. Note that the PRA has stipulated that CET1 must be used to fill Pillar IIA requirements prior to the combined capital buffer. The residual 3.0% of CET1 would be available to meet any remaining capital or buffer requirements, including the Pillar IIA requirements.

What factors should guide the capital planning process with respect to the structure of the capital base? The all‐equity scenario presents the following features:

• The most robust form from a regulatory perspective;
• It may result in an incremental (one‐notch) rating benefit;
• As the most expensive structure, it is least efficient from a shareholder return perspective;
• Compared to the equity plus other capital format, there is a substantial decrease in RoC.

The equity plus other capital liability structure has the following features:

• It is a more efficient capital structure;
• Ultimately, it remains subject to national regulatory approval;
• The rating agency position towards this structure is neutral to negative;
• The T2 issuance would of course be subject to investor demand for this paper;
• The structure allows for further gearing/leverage of the capital base. The level of leverage is restricted under Basel III and in particular in South Africa as discussed below.

Table 15.2 is a summary of the requirements for capital instruments to be AT1 and T2 eligible. In its capital planning, a bank may consider the following rationale for issuance of one or both of these instruments:

• Free up common equity to count towards buffer requirements (capital buffers are additive to minimum total capital requirements);
• Pillar II or bail‐in capital requirements;
• Non‐dilutive (absent conversion);
• Tax deductible;
• Able to be sold to a fixed income investor base;
• Acts as a cushion for senior debt investors;
• Supports credit rating requirements (S&P RAC eligible).

In undertaking its capital planning, a bank must consider the Pillar II requirements. When conducting its Pillar II review (that is the review of the bank's Internal Capital Adequacy Assessment Process or ICAAP), the local regulator will consider both those risks to banks that are not fully captured under Pillar IIA and those risks a firm may be exposed to in future, for example, regulatory changes (Pillar IIB). The key considerations are:

• Pillar IIA: in addition to the Pillar I requirements of Basel III (in the EU, CRR/CRDIV), certain regulators including the PRA regard Pillar IIA capital as the minimum level of regulatory capital a bank should maintain at all times to cover against risks, with the implication being that CET1 must be used to fill the Pillar IIA requirement prior to being used for the capital conservation buffer (CCB);
• Pillar IIB: Pillar IIB buffers will be required for firms where the regulator (PRA included) deems that the Basel III or CRDIV buffers may not be sufficient to enable a firm to meet its capital requirements under stress. The Pillar IIB buffer will replace the existing (in the UK jurisdiction) capital planning buffer (CPB), and will be set based upon a range of factors, including firm‐specific stress test results.

Risk‐weighted assets and economic capital demand

RWA balances (i.e. the regulatory Pillar I view of risk) and economic capital demand (the internal view of risk) must be reported to Finance, Treasury, and the business lines. The frequency of reporting may vary from inter‐day to monthly depending on the type of risks. RWA forecasts are prepared at month‐end; any inconsistency with the Finance general forecast must be reported to ALCO. The RWA forecast should be in line with the bank's capital allocation process. The impact of any business line transaction, whether asset or liability, that is likely to result in a reduction in capital must be reported immediately to ALCO. The process applied is similar in nature to the actuarial control cycle.

Business line profit

The net profit after direct and indirect costs of each business line must be transferred to the central book at year‐end. This is a direct cash transfer.

Capital allocation

Each business line prepares a business case for capital demand, based on RWAs and economic capital. Each business case contains key metrics, including RoC, net generation of equity, and economic profit. The Balance Sheet Management department sets targets for RWAs and capital usage as part of the budget forecast and allocation process. Capital is allocated to produce optimum return, in line with the strategy and risk appetite of the bank. The strategy and risk appetite will drill down to each business line. Treasury Balance Sheet Management will present capital usage limits at month‐end, for approval by ALCO (see Chapter 10 for template Treasury organisation structure). Treasury Balance Sheet Management will report current and forecast capital usage to ALCO on a weekly basis.

If a forecast exceeds a limit, the business line will submit a request for mitigating action to be taken, or for an increase in limit.

Performance metrics

The Board, having delegated authority to the executive management committee (ExCo), will set performance metrics targets for each business line. This will include RoC and capital usage (RWA) metrics, against which performance of each business is evaluated. Some banks also consider economic profit or profit after regulatory capital cost, depending on which view of risk is more onerous.

Portfolio credit risk management

ALCO is responsible for reviewing and approving the asset pool for credit risk management purposes. This is to ensure that provisions are signed off in a consistent manner, that all transactions are in line with the business strategy and risk profile, and that they follow policy on capital usage and regulatory requirements.

Risk taxonomy and quantification

Before quantifying balance sheet risk exposure, it is important to categorise it, or as some banks term it, compile a taxonomy of risk. Not all banks exhibit all identified risk type exposures. Under Pillar II, all material risk types need to be quantified, and thereby covered with adequate capital provision. Certain types of risk (conduct risk, reputation, etc.) are not straightforward to quantify, so banks will apply a qualitative assessment. Typically, banks and, crucially, certain regulators employ the Value‐at‐Risk (VaR) technique to estimate risk exposure on the balance sheet. For those familiar with VaR, the logic of its methodology is consistent with a trading book, or a pool of assets that are repriced on a regular basis. However, this logic is less consistent in a banking book. Nevertheless, the VaR approach remains the one that is most in use and often demanded by the regulator to assess the VaR to the firm. The actual methods vary from simple risk weights to detailed economic risk assessments. In addition, regulators use peer group comparisons as well and any outliers per business area are interrogated further.

Table 15.3 summarises the Level 1 risks within the risk taxonomy for a vanilla UK commercial bank, with 1st and 2nd line‐of‐defence (LoD) responsibilities indicated.

It is common to see the variance‐covariance (“parametric”), historical simulation, and Monte Carlo simulation approaches used in equal measure among banks. Typically, banks apply a 1‐year holding period and one‐sided confidence levels of 95% or 99%. For market risk, banks often apply a 1‐day and 10‐day holding period. Note that most banks classify interest‐rate risk in the banking book (IRRBB) as a market risk, albeit a non‐traded market risk. Again, it is common to observe VaR as the preferred measurement technique for IRRBB as well; however, the author's recommended approach is to supplement this with traditional modified duration analysis (“gap” analysis) as well.

Stress tests

In the post‐2008 era, all banks, except perhaps the very smallest, specify a large number and variety of stress tests. That said, there is considerable commonality in approach. Generally, the primary focus is on scenario analyses, often based on 5‐year historical worst‐case values or hypothetical scenarios. The latter are sometimes also required to include a “reverse stress test”, which is a hypothetical scenario that would break the bank. In theory, as stress tests allow for the identification of sensitivities to specific risk factors, they should be a worthwhile exercise as they provide value‐added input to the bank's risk management process and subsequent management actions. The danger, as with all exercises subject to considerable regulatory supervision, is that the stress testing process becomes an excessively bureaucratic one and more of a “box‐ticking” exercise. This must be avoided and ALCO is the appropriate forum to review the process and ensure this does not happen.

ICAAP implementation

Before a bank can begin designing its ICAAP, it should first define its relevant target state. The steps involved are those illustrated in Figure 15.10.

What does Figure 15.10 mean in practice? Each step is considered in turn below.

Definition of bank‐specific requirements (target state): In the first step, the bank compiles its strategy based on its goal or vision. It will test this strategy against a list of requirements based on national regulator statements, for example, the EU CRDIV. Next, these requirements have to be specified for the individual bank, thus in the course of a self‐assessment the bank should identify its material risks, which arise from the implementation of its strategy. The ICAAP therefore facilitates assessing the sensitivity and risks in a firm's strategy. The requirements with regard to ICAAP methods should be defined in light of the bank's risk profile, as they need to be fit for purpose but also meet regulatory rigour. Typically, the introduction of new methods begins with relatively simple, robust solutions which are then developed and refined on an ongoing basis. The full list of requirements then represents the target state for ICAAP purposes and defines requirements with regard to methods, procedures, processes, and organisation.

Gap analysis: Once the target state has been defined, the bank should analyse those requirements that are currently not (or not completely) fulfilled. These gaps pertain both to the execution of strategy as well as the capital assessment process. It is the business team's responsibility to address the former and its risk management teams would survey the current state of methods, processes, and organisation in place in the internal risk management system. This might include the regulator's requirements for calculating capital amounts, or activities aimed at fulfilling the minimum standards for the credit origination business. The process is not restricted to risk managers. In terms of the three lines of defence model, the process starts with all business lines, which should analyse their current state and report it centrally to Risk (and ALCO). Gaps in implementation can then be identified by comparing the requirements with the current state. This comparison of target and actual states could be carried out in the course of a workshop attended by representatives from the business lines, with the results documented and reported to all relevant stakeholders. The bank can then assess the significance and consequences of the gaps identified as well as identify the necessary management actions as a result.

Implementation planning: In implementation planning, the first step is to prioritise the required measures identified in the previous process. This way a clear ranking can be defined in order to deploy implementation resources effectively. Measures identified should be combined in individual work packages and coordinated with business lines; this process requires that specific roles and responsibilities be identified and assigned. As part of the process, the bank should set binding deadlines, which reflect resource and capacity available within the firm.

Implementation: Put simply, this stage involves undertaking and delivering on the process measures identified in the previous stage. This includes employing human resources as well as IT capacity as required by the ICAAP. The process‐related aspects and responsibilities within the ICAAP can then be defined and documented. This may involve quantifying and aggregating risks and coverage capital, monitoring limits, or taking measures in the ex‐post control process. The ICAAP is integrated into the bank's strategic and operational control mechanisms (for example, annual budgeting and planning on the basis of risk indicators and coverage capital). Once implementation is completed, the bank should have in place adequate methods, processes, and systems to ensure its risk‐bearing capacity over the long term.

What makes a good ICAAP?

In essence, the ICAAP process and its published output as presented to the national regulator should be a value‐added exercise that reflects the interests of all stakeholders in the long‐term viability of the bank as a going concern. This is not a platitude. However, because of the onerous regulation and reporting process in place in most jurisdictions, the process often boils down to producing something that the regulator signs off on. This is not the best approach to ICAAP production. To the question “What makes a good ICAAP?” the obvious answer is a document that presents the firm's strategy with all material risk exposures, on a forward‐looking basis, and demonstrates that these exposures are covered by adequate capital levels, risk management systems, and management actions to rectify stresses in order to ensure the successful implementation of the strategy.

As a checklist, we should note that the following factors are crucial in the actual implementation and successful presentation of an ICAAP:

• Early detection of gaps in fulfilment:

      A bank should make efforts to detect gaps in the fulfilment of requirements as early as possible so that it can take the appropriate measures in a timely and economical manner. Closing these gaps quickly will improve the bank's internal risk management and thus enhances its ability to ensure its risk‐bearing capacity.

• Selection of methods:

      The bank should determine the methods and procedures that best suit its needs, as these determine the validity of the ICAAP as well as the required implementation resources. In the course of selecting methods, the bank should not only consider its current risk profile but also anticipate planned developments in individual risk types. If, for example, a decision has already been made that trading will be expanded in the medium term, then it makes sense to introduce more advanced procedures from the outset when designing the ICAAP.

• Master plan and project management:

      The bank should develop a master implementation plan that covers planning, budgeting, and a prioritisation of all ICAAP implementation tasks. There should be one overall, dedicated project manager (most probably someone within Risk or Treasury). This master plan forms the basis for requesting internal and external capacities and may well involve planning resources over a period of several years. For example, implementation might already be well under way for the most important risk types, while measures for other risk types are still being planned. Once it reaches a certain scale, the master plan should be transformed into a detailed project plan, which serves to reduce complexity and create transparency with regard to the current implementation status. It is also important to set binding deadlines and responsibilities on the basis of this plan. A project manager should then monitor and control the performance of individual tasks. Project management should seek to prevent any conflicts of interest between the business lines involved in implementation and to maintain an aggregate and holistic view of the project.

• Communication:

      The need for, and benefits of, the ICAAP have to be clearly communicated to all staff. The fundamental concept of the ICAAP is not something for senior executives only to appreciate, but for all business lines. An example is a newly designed limit allocation system or a change to the organisation chart, more likely to be supported by staff if they are informed about the need for these measures in a transparent and understandable manner. Insufficient communication in implementation projects often results in low levels of identification or even rejection and demotivation. By applying an appropriate communication policy and setting a good example, senior executives should generate the employee acceptance necessary for successful implementation of the ICAAP.

• Know‐how and resources:

      One major objective of the ICAAP is to foster the development of an appropriate internal risk management culture. For this reason, expertise in this area is a key success factor in the implementation of the ICAAP. It is important for the bank to have the necessary resources (employees, systems) at its disposal in the ICAAP implementation process. Resource requirements will depend on the bank's size and risk profile as well as the difference between the current status and the defined requirements.

• Data quality and IT systems:

      Data quality (completeness, availability) is especially important because it determines the reliability and accuracy of calculated results (for example, risk indicators and coverage capital). The process of data quality assurance begins with accurate data capture and goes as far as ensuring data availability in the ICAAP. Especially for risk management, it is necessary and worthwhile to ensure timely automated evaluations due to the large data quantities involved and the sometimes complex calculation algorithms used. In its ICAAP, the bank can rely on existing risk management systems (risk measurement, limit monitoring) if they meet the defined requirements. Historically, maintaining and updating the IT structures of many banks requires copious resources. It is a required investment, however, because the lack of uniform data pools can create considerable difficulties in assessing the true extent of balance sheet risk.

Credit risk indicators

The structure of the credit portfolio provides initial indications of a bank's risk appetite. A large share of loans in a certain asset class (for example, CRE) may point to increased risk. In addition, the presence of complex financing transactions such as specialised lending (project finance, for instance) may also indicate a larger risk appetite. For an approximate initial assessment, in the EU a bank can use the asset classes defined in Directive 2000/12/EC, or the asset classes outlined in the BA200 submission to the SARB, to examine the distribution of its credit portfolio. A bank can use credit assessments (such as credit ratings) to measure the share of borrowers with poor creditworthiness in its portfolio; this provides an indication of default risk. The bank will also consider the portfolio in terms of delinquency and impairment. The amount of available collateral – and thus the unsecured volume – also plays a role in this context. The lower the unsecured volume is, the lower the risk generally is; this relationship is also reflected in future supervisory regulations for calculating capital requirements. In this context, however, the type and quality of collateral are decisive; this can be assessed by asking the following questions:

• To what extent is the retention or liquidation of the collateral legally enforceable?
• How will the value of the collateral develop?
• Is there any correlation between the value of the collateral and the creditworthiness of the debtor?

A close inspection of the credit portfolio will provide further insights with regard to any existing concentration risks. In order to assess the size structure or granularity of its portfolio, the bank can also assess the size and number of large exposures. The bank should also consider the distribution of exposures among industries (for example, construction business, transport, tourism, and so on) in assessing its concentration risk. If a bank conducts extensive operations overseas (share of foreign assets), it is appropriate to take a close look at the risks associated with those activities as well, such as country and transfer risks. The share of foreign currency loans in a bank's credit portfolio can also point to concentration risks. If the share of foreign currency loans is very high, exchange rate fluctuations can have adverse effects on the credit quality of the borrowers. If the foreign currency loans are serviced using a repayment vehicle that is heavily exposed to market risks, this indicates an additional source of risk that should be monitored accordingly and controlled as necessary.

Market risks in the trading book, foreign exchange risks at the overall bank level

This exposure is calculated using a VaR approach. A bank can determine its sensitivity to foreign exchange fluctuations on the basis of its open foreign exchange positions and (in the broadest sense) open term positions. The influence of foreign exchange fluctuations on the default probability of borrowers with foreign currency loans is also considered.

Interest‐rate risk in the banking book

The results reported in interest‐rate risk statistics (part of regulatory reporting requirements) constitute an essential indicator of the level of interest rate risk in the banking book. The traditional approach here is to apply modified duration analysis such as parallel yield curve shifts (for example, the effects of a 200 basis point interest‐rate shock on the present value of the balance sheet). Of course, if this method demonstrates that material interest‐rate risks exist in the banking book, regulators usually require more “sophisticated” risk measurement methods to be applied, with output including a precise quantification of risks in terms of their effects on the income statement. Another risk indicator should cover proprietary trading both on‐ and off‐balance‐sheet. In accordance with the proportionality principle in the ICAAP (larger, more sophisticated balance sheets require larger, more sophisticated ICAAPs), the corresponding requirements increase in line with the scale of derivatives trading activities. Even in cases where a bank primarily uses derivatives to hedge other transactions or portfolios, the effectiveness of hedging transactions (the hedge effectiveness) should be examined in order to avoid undesirable side effects. In the case of on‐balance‐sheet proprietary transactions, the need for more precise risk control grows along with the scale and complexity of the positions held, for example, “alternative investments” or structured bonds.

Operational risk indicators

Two important indicators of operational risk are the size and complexity of a bank. As the number of employees, business partners, customers, branches, systems, and processes at a bank increases, its risk potential also tends to rise. Another risk indicator in this category is process intensity, for example, the number of transactions and volumes handled in payments processing, loan processing, securities operations, and proprietary trading. Failures (for example, due to overloaded systems) can bring about severe economic losses in banks with high levels of process intensity.¹ The number of lawsuits filed against a bank can also serve as an indicator of operational risks. A large number of lawsuits suggests that there are substantial sources of risk within the bank, such as inadequate system security or insufficient care in processes and control mechanisms. In cases where business operations (for example, the processing activities mentioned above) are outsourced, the bank cannot automatically assume that operational risks have been eliminated completely. This is because a bank's dependence on an outsourcing service provider means that risks incurred by the latter can have negative repercussions for the bank. Therefore, the content and quality of the service level agreement, as well as the quality and creditworthiness of the outsourcing service provider, can also serve as risk indicators in this context.

The risk indicators may be presented as per the following template (note that more than one bank has been amalgamated to reflect different approaches), as illustrated in Table 15.4.

For Bank A, the risk types mentioned above would have little significance under the proportionality principle. The bank shows a low level of complexity and low risk levels. Besides, Bank A does not have any trading positions. For the purpose of measuring its risks and calculating its internal capital needs, Bank A could calculate its capital requirements using the Standardised Approach, or the Basic Indicator Approach in the case of operational risk.

In terms of its total assets and number of employees, Bank B is comparable to Bank A, but Bank B's transactions show a markedly higher risk level. In addition, concentration risks exist with regard to size classes (for example, several relatively large loans to medium‐sized businesses), borrowers in the same industry, and foreign currency loans. In this bank, methods that go beyond the Standardised Approach should be employed and/or adequate qualitative measures (monitoring/reporting) should be set. Furthermore, Bank B should pay attention to concentration risks, for example, by adhering to suitable individual borrower limits based on creditworthiness or by implementing minimum standards for foreign currency loans. In this example, using more advanced systems may be appropriate in other areas, such as interest rate risk in the banking book.

Bank C shows high credit exposures to SMEs and has also granted a number of relatively large loans. This results in a certain degree of concentration risk. In addition, the bank is exposed to relatively high interest‐rate risks. In fact, Bank D has large exposures to almost all risk types. The bank's size and structure can be described (such as a VaR model) for interest‐rate risk at Banks C and D; Bank D should also use a more sophisticated model for market risk. Due to the higher risk level and the existing complexity with regard to credit risk, the bank may use other risk‐sensitive techniques based on the internal ratings‐based (IRB) approach or a credit portfolio model.

The individual institutions in this example have to define the scale and type of risk management system that is appropriate to their activities, with due attention applicable to the regulator's requirements. The choice of suitable risk measurement procedures to determine risks and internal capital needs plays a key role in this context. Moreover, the proportionality concept also has effects on process and organisational design: banks that demonstrate a high level of complexity or a large risk appetite have to fulfil more comprehensive requirements.

An example of the presentation of the risk indicators is given in Table 15.5.

Capital treatment of material risks

In our example illustration, the MIRA material risks are summarised below. These are used to determine whether the risk is either “managed and capitalised” or else just “managed”. In this case, the MIRA has identified that the material add‐on risks are for credit risk concentration, operational risk, and non‐traded market risk and these are then discussed in more detail in Table 15.8.

Pillar II add‐on

From the above, the bank in question has to calculate its Pillar II add‐on for credit risk concentration and non‐traded interest‐rate risk. The relevant points here are:

• Single name concentration (SNC):

  One methodology to address this is the Pythkin approach, which is aligned with the theoretical (Vašíček) approach underlying the regulatory capital formula. The alternative Gordy approach, which is sometimes advocated, is also used although it is based on a different underlying model. In our example the add‐on for SNC is highest for large exposures with high PD.

• Sector concentrations (SRC):

  This is an add‐on for excess concentration in certain sectors such as CRE.

• Non‐traded market risk (NTMR) – method of assessment:

  The bank does not incur traded market risk. Non‐traded market risk arises from:

  • Repricing risk: arises from timing differences in the maturity (for fixed‐rate) and repricing (for floating‐rate) of assets and liabilities. There repricing mismatches can expose net interest income and economic value to fluctuations in interest rates.
  • Yield curve risk: arises when yield curve shifts differently between the short end and the long end and this can have adverse effects on a bank's economic value where repricing mismatches exist. The Liquid Asset Buffer (LAB) of £3.5bn is expected to be held in gilts for an average duration of 4 to 7 years, and this would be subject to fluctuating gilts prices if unhedged or to swap spread risk if hedged with swaps. For the purpose of an interim capital assessment, NTMR is measured using a VaR model based on a 99% confidence interval over a 1‐year holding period. In addition, a parallel shift of +/–100 and +/–200 basis point parallel shifts are also calculated and reported to ALCO. Non‐traded market risk (NTMR) – estimates of exposure: The risk exposure on the estimated LAB is a PV01 of circa £820k for an average 4‐year duration. The VaR estimated on a hedged or unhedged basis (using a 99% confidence interval for a 1‐year holding period) is between £50 and £80m. Thus, the Pillar II add‐on will be £80mm.

Operational risk

In normal circumstances, to calculate a robust Pillar II operational risk capital charge, a bank would use a variety of data including internal loss event data, external loss event data, key risk indicators, and the output of scenario analyses. For a new bank, there may be insufficient internal loss event data specific to its operations to be able to undertake any meaningful modelling capability. As a consequence, such a bank (as in our case study here) would develop a comprehensive programme for scenario analysis, which it would use until such time as 3 years' worth of loss event data was accumulated.

In the meantime, while the scenario analysis approach is being developed, the example bank here adopts a scaling approach based on the Group operational risk economic capital model, which considers Group internal loss event experience, its own programme of scenario analyses outputs, and relevant external loss event data.

Stress testing approaches

Table 15.9 is a slide that would be presented at the start of the ICAAP deck to the regulator during the SREP, with initial capital assessment shown at Table 15.10 and projection at Table 15.11.

The stress testing framework should emphasise integrated stress testing for the aggregate balance sheet. Remember that the ICAAP is designed to be a forward‐looking process, so that the anchor point for start of testing would be today and then built around scenarios over the next 1, 3, and 5 years.

For the case study here on the UK commercial bank, the top‐level scenario planning looked like this:

• A recession for the first 2 years, triggered by contagion from a re‐intensification of Eurozone sovereign concerns; this scenario is relevant for the UK bank as it is a UK‐based recession although the trigger is outside the UK. The scenario addresses the key vulnerabilities in the bank's portfolio, being UK focused affecting CRE and mortgage exposures with the key economic parameters below:
  • UK real GDP falls at an annualised rate of 5.7% at the trough of the recession;
  • Unemployment peaks at the highest to 11.4%, affecting household consumption and business consumption;
  • House prices fall by more than 20% over the first two scenarios and remain volatile thereafter;
  • M4 lending and holdings contract but at different rates;
  • Interest rates (3‐month Libor) increase in the first 3 years and then fall for the next 2 years.
• The contraction is followed by a slow recovery, with growth eventually returning close to trend towards the end of the scenario.

Of course, banks would design their stress test relevant to their operating environment. A South African bank would concentrate on the economic impact of recession in South Africa.

The scenario is then subject to drilldown to see the impact on financials:

• Balance sheet under stress:
  • Asset balances decline under stress due to debt flow, fall in new lending, and in consumer/household spending as reflected in the relevant economic parameters;
  • Additional assumptions are being taken with regard to CRE decline as shown in scenario 3.
• Operating and funding costs:
  • Balance sheet decline will cause a corresponding fall in income in the first 2 years of the stress and then picks up the following 3 years;
  • Net interest margin declines compared to base case and remains low in the first few years of the recovery;
  • Costs will follow the base balance sheet forecast and any cost cutting will be taken as a management action;
  • Funding costs would reflect the scenario interest rates but would increase if the funding gap increases.
• Impairments:
  • Impairments are equal to EL and taken into the P&L;
  • The scenario has a negative effect on unemployment, house prices, and CRE index and this has a significant impact on impairments during the stress period as creditworthiness deteriorates. Consequently, PDs and expected losses increase causing impairments to more than double during the stress period;
  • Similarly, an economic recovery would result in lower impairments.
• RWA credit:
  • The effects of a decline in house prices and property values during stress would decrease the value of the security of the loan and hence increase LGDs. This and the increase in PDs would cause a sharp increase in RWAs and this is offset by debt flow (defaults) and contraction of the balance sheet.
• Operational risk losses and RWAs:
  • It is assumed that certain operational risk losses (for example, external fraud) crystallise under stress;
  • Operational risk RWAs are recalculated according to the Standardised Approach and would decline in line with income.
• Capital plans:
  • Capital deductions remain although there will be no excess provisions over EL as provisions are equal to or greater than EL;
  • The assets in the Liquid Asset Buffer (LAB), also termed Available For Sale (AFS), are revalued and profits/losses taken against capital and not P&L in line with accounting rules;
  • The combined effects of a deterioration in P&L and an increase in RWAs causes CT1 and total capital ratios to decline compared to base. This will then be assessed against regulatory minima.

The above summary is presented as a template as to how a bank could set the stress testing framework for impact on financials.