Basel Committee on Banking Supervision

Experience shows that poor credit quality and deficiencies in credit risk assessment and measurement practices are significant causes of bank failures. Inadequate credit risk policies and procedures can delay the recognition and measurement of increases in credit risk, which in turn affects banks’ capital adequacy and impairs their ability to properly assess and control credit risk exposures.

Banks should apply a disciplined, high-quality approach to the assessment and measurement of expected credit losses (ECL) under the applicable accounting framework.

The following terms are used throughout this chapter and have the meaning given below:

1. Allowances: includes allowances on loans, and allowances or provisions on loan commitments and financial guarantee contracts.
2. Lending exposures: refers to loans, loan commitments and financial guarantee contracts to which an ECL framework applies.
3. Reasonable and supportable information: is information based on relevant facts and sound judgment (see Principle 6 for further guidance).

This chapter sets out principle-based supervisory guidance on how banks should use common elements of the credit risk management process to ensure high-quality and robust assessments and measurements of ECL. While acknowledging differences between ECL accounting frameworks, this guidance provides a consistent interpretation of common elements and their appropriate application. It aims to promote consistency in the assessment and measurement of credit risk, the development of accounting estimates, and the evaluation of capital adequacy. It does not set out any additional requirements regarding the determination of expected loss for regulatory capital purposes.

The scope of this guidance is limited to credit risk practices related to the assessment and measurement of ECL and allowances, rather than broader impairment requirements under the applicable accounting framework. It focuses on lending exposures, for which the Committee expects banks to estimate ECL. While the guidance emphasises the timely recognition of allowances, it also recognises the symmetrical nature of ECL frameworks, where both credit deteriorations and reversals should be considered in the measurement of allowances.

A proportionate implementation of the ECL accounting frameworks can be applied wherein banks are enabled to adopt sound allowance methodologies commensurate with their size, complexity, structure, economic significance, risk profile and broader circumstances but must not compromise its high-quality. Similarly, materiality provisions should consider that individually immaterial exposures may be cumulatively material or could become material in the future. Bank’s approximation methods, adopted on the grounds of materiality or proportionality, should be designed and implemented to avoid bias.

This guidance is equally applicable under all ECL accounting frameworks; it does not contradict applicable accounting standards established by standard setters. For jurisdictions without mandated ECL accounting frameworks, the relevant aspects of it related to sound credit risk practices should be applied, as appropriate, within the context of the applicable accounting framework.

Supervisory guidance for credit risk and accounting for expected credit losses

Principle 1 - A bank’s board of directors1 (or equivalent) and senior management are responsible for ensuring that the bank has appropriate credit risk practices, including an effective system of internal control, to consistently determine adequate allowances in accordance with the bank’s stated policies and procedures, the applicable accounting framework and relevant supervisory guidance.

A bank's board of directors (or equivalent) is responsible for approving and regularly reviewing the bank's credit risk management strategy and significant policies and processes to ensure alignment with the approved risk appetite. The board should also oversee senior management's implementation of sound underwriting practices to mitigate risks to depositors and financial stability.

The board should direct senior management to develop and maintain appropriate processes, which should be systematic and consistently applied, to determine appropriate allowances. The board should also require senior management to report periodically the results of the credit risk assessment and measurement processes, including estimates of its ECL allowances. Senior management should establish, implement and, as necessary, update suitable policies and procedures to communicate the credit risk assessment and measurement process internally to all relevant personnel.

An effective internal control system for credit risk assessment and measurement is essential to enable senior management to carry out its duties. An effective internal control system should include:

1. measures to comply with applicable laws, regulations, internal policies and procedures;
2. measures to provide oversight of the integrity of information used and reasonably ensure that the allowances reflected in the bank’s financial statements and its supervisory reports are prepared in accordance with the applicable accounting framework and relevant supervisory guidance;
3. well defined credit risk assessment and measurement processes that are independent from (while taking appropriate account of) the lending function, which contain:
   1. an effective credit risk rating system that is consistently applied, accurately grades differing credit risk characteristics, identifies changes in credit risk on a timely basis, and prompts appropriate action;
   2. an effective process which ensures that all relevant and reasonable and supportable information, including forward-looking information, is appropriately considered in assessing and measuring ECL. This includes maintaining appropriate reports, details of reviews performed, and identification and descriptions of the roles and responsibilities of the personnel involved;
   3. an assessment policy that ensures ECL measurement occurs not just at the individual lending exposure level but also when necessary to appropriately measure ECL at the collective portfolio level by grouping exposures based on identified shared credit risk characteristics;
   4. an effective model validation process to ensure that the credit risk assessment and measurement models can generate accurate, consistent and unbiased predictive estimates on an ongoing basis. This includes establishing policies and procedures which set out the accountability and reporting structure of the model validation process, internal standards for assessing and approving changes to the models, and reporting of the outcome of the model validation;
   5. clear formal communication and coordination among a bank’s credit risk staff, financial reporting staff, senior management, the board and others who are involved in the credit risk assessment and measurement process for an ECL accounting framework, as applicable (eg evidenced by written policies and procedures, management reports and committee minutes); and
4. an internal audit function2 that independently evaluates the effectiveness of the bank’s credit risk assessment and measurement systems and processes, including the credit risk rating system.

Principle 2 –A bank should adopt, document and adhere to sound methodologies that address policies, procedures and controls for assessing and measuring credit risk on all lending exposures. The measurement of allowances should build upon those robust methodologies and result in the appropriate and timely recognition of expected credit losses in accordance with the applicable accounting framework.

The credit risk assessment and measurement process, underscored by sound credit risk methodologies, provides the relevant information for senior management to make its experienced judgments about the credit risk of lending exposures, and the related estimation of ECL.

Banks should leverage and integrate common processes that are used within a bank to determine if, when and on what terms credit should be granted; monitor credit risk; and measure allowances for both accounting and capital adequacy purposes.

Banks should clearly document the definitions of key terms related to the assessment and measurement of ECL (such as loss and migration rates, loss events and default). Where different terms, information or assumptions are used across functional areas (such as accounting, capital adequacy and credit risk management), the underlying rationale for these differences should be documented and approved by senior management. Information and assumptions used for ECL estimates should be reviewed and updated as required by the applicable accounting framework. Moreover, the rationale for changes in assumptions that affect the measurement of ECL should be well documented.

In accordance with Core Principle 17, banks should have in place adequate processes and systems to appropriately identify, measure, evaluate, monitor, report and mitigate the level of credit risk. During the transition to the relevant new accounting standard, existing processes and systems should be evaluated and, if necessary, modified to collect and analyse relevant information affecting the assessment and measurement of ECL.

A bank should adopt and adhere to written policies and procedures detailing the credit risk systems and controls used in its credit risk methodologies and the separate roles and responsibilities of the bank’s board and senior management. Robust methodologies for assessing credit risk and measuring allowances should:

1. include a robust process that is designed to equip the bank with the ability to know the level, nature and drivers of credit risk upon initial recognition of the lending exposure to ensure that subsequent changes in credit risk can be identified and quantified;
2. include criteria to duly consider the impact of forward-looking information, including macroeconomic factors. Whether the evaluation of credit risk is conducted on a collective or individual basis, a bank must demonstrate that this consideration has occurred so that the recognition of ECL is not delayed. Such criteria should result in the identification of factors that affect repayment, whether related to borrower incentives, willingness or ability to perform on the contractual obligations, or lending exposure terms and conditions. Economic factors considered (such as unemployment rates or occupancy rates) must be relevant to the assessment and, depending on the circumstances, this may be at the international, national, regional or local level;
3. define the basis for grouping collectively evaluated exposures with shared credit risk characteristics;
4. identify and document the ECL assessment and measurement methods (such as a loss rate method, probability of default (PD) / loss-given-default (LGD) method, or another method) to be applied to each exposure or portfolio;
5. document the reasons why the selected method is appropriate, especially if different ECL measurement methods are applied to different portfolios and types of individual exposures. A bank should be able to explain to its supervisors the rationale for any changes in measurement approach (eg a move from a loss rate method to a PD/LGD method) and the quantitative impacts of such changes;
6. document the inputs, data and assumptions used in the allowance estimation process (such as historical loss rates, PD/LGD estimates and economic forecasts), how the life of an exposure or portfolio is determined (including how expected prepayments and defaults have been considered), the time period over which historical loss experience is evaluated, and any adjustments necessary for the estimation of ECL in accordance with the applicable accounting framework. For example, if current and forecasted economic conditions are different from those that existed during the historical estimation period being used, adjustments that are directionally consistent with those differences should be made. In addition, a bank may have experienced little to no actual losses in the historical period analysed; however, current or forward-looking conditions can differ from conditions during the historical period, and the impact of these changes on ECL should be assessed and measured;
7. include a process for evaluating the appropriateness of significant inputs and assumptions in the ECL assessment and measurement method chosen. The basis for inputs and assumptions used in the estimation process should generally be consistent from period to period. Where inputs and assumptions change, the rationale should be documented;
8. identify the situations that would generally lead to appropriate changes in ECL measurement methods, inputs or assumptions from period to period (eg the bank may state that a loan that had been previously evaluated on a collective basis using a PD/LGD method may be removed and evaluated individually using the discounted cash flow method upon receipt of new, borrower-specific information such as the loss of employment);
9. consider the relevant internal and external factors that may affect ECL estimates, such as the underwriting standards applied to a lending exposure at origination and changes in industry, geographical, economic and political factors;
10. address how ECL estimates are determined (eg historical loss rates or migration analysis as a starting point, adjusted for information on current and expected conditions). A bank should have an unbiased view of the uncertainty and risks in its lending activities when estimating ECL;
11. identify what factors are considered when establishing appropriate historical time periods over which to evaluate historical loss experience. A bank should maintain sufficient historical loss data (ideally over at least one full credit cycle) to provide a meaningful analysis of its credit loss experience for use as a starting point when estimating the level of allowances on a collective or individual basis;
12. determine the extent to which the value of collateral and other credit risk mitigants affects ECL;
13. outline the bank’s policies and procedures on write-offs and recoveries;
14. require that analyses, estimates, reviews and other tasks/processes that are inputs to or outputs from the credit risk assessment and measurement process are performed by competent and well-trained personnel and validated by personnel who are independent of the bank’s lending activities. These inputs to and outputs from these functions must be well documented, and the documentation should include clear explanations supporting the analyses, estimates and reviews;
15. document the methods used to validate models for ECL measurement (eg backtests);
16. ensure that ECL estimates appropriately incorporate forward-looking information, including macroeconomic factors, that has not already been factored into allowances measured on an individual exposure basis. This may require management to use its experienced credit judgment to consider broad trends in the entire lending portfolio, changes in the bank’s business model, macroeconomic factors etc; and
17. require a process to assess the overall adequacy of allowances in accordance with the relevant accounting requirements.

A bank’s credit risk identification process should ensure that factors that impact changes in credit risk and estimates of ECL are properly identified on a regular basis. In addition, consideration of credit risk inherent in new products and activities should be a key part of the risk identification process and the assessment and measurement of ECL.

Consistent with sound model development practices, management should consider relevant facts and circumstances, including forward-looking information, that are likely to cause ECL to differ from historical experience and that may affect credit risk and the full collectability of cash flows.

When assessing the collectability of cash flows, banks should consider factors such as:

1. lending policies, underwriting standards, lending terms, and exceptions to policy;
2. a borrower's sources of recurring income and ability to meet scheduled payments;
3. the borrower's ability to generate sufficient cash flow over the term of the financial instrument;
4. the borrower's leverage level and anticipated changes;
5. unencumbered assets the borrower may pledge as collateral in the market or bilaterally to raise funds and expectations of changes to the value of those assets;
6. reasonably possible one-off events or recurring behaviours affecting repayment ability; and
7. timely evaluations of collateral and factors that may impact its future value.

Where they have the potential to affect the bank's ability to recover amounts due, factors relating to the bank's business model and current and forecasted macroeconomic conditions could be considered, such as:

1. competition and legal and regulatory requirements;
2. trends in credit volume;
3. the overall credit risk profile of lending exposures and any expected changes;
4. credit concentrations by borrower, product, market segment or geography;
5. unencumbered assets the borrower may pledge as collateral in the market or bilaterally in order to raise funds and expectations of changes to the value of those assets;
6. the quality of the credit risk review system and oversight by senior management and the board;
7. other factors that may impact ECL such as, but not limited to, expectations of changes in unemployment rates, gross domestic product, benchmark interest rates, inflation, liquidity conditions or technology; and
8. borrower incentives or willingness to meet their obligations.

Robust methodologies should consider different potential scenarios and should not rely purely on subjective, biased or overly optimistic considerations. A bank should develop and document its process to generate relevant scenarios to be used in the estimation of ECL. In particular:

1. the bank should demonstrate and document how ECL estimates would alter with changes in scenarios, including changes to relevant external conditions that may impact ECL estimates or components of the ECL calculation (such as PD and LGD parameters);
2. the bank should have a documented process for determining the time horizon of the scenarios and, if relevant, how ECL is estimated for exposures whose lives exceed the period covered by the economic forecast(s) used;
3. scenarios may be internally developed or vendor defined. For internally developed scenarios, a bank should have a variety of experts, such as risk experts, economists, business managers and senior management, assist in the selection of scenarios that are relevant to the bank’s credit risk exposure profile. For vendor-defined scenarios, a bank should ensure that the vendor tailors the scenarios to reflect the bank’s business and credit risk exposure profile, as the bank remains responsible for those scenarios;
4. backtesting should be performed to ensure that the most relevant economic factors that affect collectability and credit risk are being considered and incorporated into ECL estimates; and
5. where market indicators of future performance (such as credit default swap spreads) are available, management may consider them to be a valid benchmark against which to check the consistency of its own judgments.

Banks should consider all reasonable and supportable information relevant to the product, borrower, business model, or economic and regulatory environment when developing estimates of ECL. In developing such estimates for financial reporting purposes, a bank should consider the experience and lessons from similar exercises it has conducted for regulatory purposes. Stressed scenarios developed for industry-wide supervisory purposes are not intended to be used directly for accounting purposes. Forward-looking information, including economic forecasts and related credit risk factors used for ECL estimates, should be consistent with inputs to other relevant estimates within the financial statements, budgets, strategic and capital plans, and other information used in managing and reporting on the bank.

Bank management should be able to demonstrate that it understands and is appropriately considering inherent risks when pricing lending exposures. Post-initial recognition increases in credit risk require a bank to reassess ECL and re-measure the amount of the allowance that should be recognised in accordance with the applicable accounting framework. Examples of fact patterns potentially indicative of inadequate estimates of ECL include:

1. the granting of credit to borrowers based on fragile income streams (that could become non-recurrent upon a downturn) or with no documentation or limited verification of borrower income sources;
2. high debt service requirements relative to the borrower’s net available expected cash flows;
3. flexible repayment schedules, including payment vacations, interest-only payments (eg bullet loans) and negative amortisation features;
4. for real estate and other asset-based financing, lending of amounts equal to or exceeding the value of the financed property or otherwise failing to provide an adequate margin of collateral protection;
5. undue increases in restructurings/modifications due to financial difficulties faced by the borrower or other reasons (such as competitive pressures faced by banks);
6. circumvention of the classification and rating requirements, including rescheduling, refinancing or reclassification of lending exposures;
7. undue increases in the volume of credit, especially in relation to the increase in the volume of credit by other lenders in the same market; and
8. increasing volume and severity of delinquent, low-quality and impaired credit.

A bank’s accounting policies and allowance methodologies should include criteria for:

1. Restructurings/modifications: they can take many forms, including, but not limited to, a renewal or extension of terms, other concessions to the borrower, or a modification of the terms with or without concessions to the borrower. The allowance methodology should deliver a robust assessment and measurement of ECL such that the allowance level continues to reflect the collectability of the substance of the restructured/modified exposure whether or not the original asset is derecognised under the applicable accounting framework. It would not be appropriate to assume that restructurings automatically lead to the conclusion that there has been an immediate decrease in the credit risk on the exposure, and any decrease in the reported allowance level due to improved credit risk should be well supported by strong evidence. Typically, a customer would need to demonstrate consistently satisfactory payment performance over a reasonable period of time before credit risk would be considered to have decreased. Subsequent to a restructuring or modification, a bank may be able to demonstrate that it has increased its likelihood of receiving full payment of outstanding principal and/or interest; however, repayment performance in the form of interest payments alone may not be indicative of whether the collection of loan principal is reasonably assured. In addition, further expected delays in the payment of those cash flows may evidence that credit risk has not improved, and thus the level of ECL should be reassessed carefully. The methodologies should also call upon the lending staff to promptly notify the bank’s accounting function when exposures are restructured or modified to ensure appropriate accounting for the change. For more complex restructurings and modifications, regular communication between the lending staff and the accounting function is warranted.
2. Purchased or Originated Credit-Impaired Lending: The methodology should enable appropriate identification and accounting for purchased or originated credit-impaired lending The cash flow estimates for these lending exposures should be reviewed each reporting period and updated as necessary. Such updates should be properly supported and documented and approved by senior management.

Principle 3 - A bank should have a credit risk rating process in place to appropriately group lending exposures based on shared credit risk characteristics.

Banks should implement comprehensive procedures and information systems to monitor the quality of their lending exposures. These systems should include an effective credit risk rating process that captures the varying level, nature and drivers of credit risk over time, to reasonably ensure that all lending exposures are properly monitored and ECL allowances are appropriately measured.

The credit risk rating process should include an independent review function. While front-line lending staff may have initial responsibility for assigning credit risk grades and ongoing responsibility for updating the credit grade to which an exposure is assigned, this should be subject to the review of an independent review function.

The credit risk grade a bank assigns upon initial recognition of a lending exposure may be based on several criteria, including product type, terms and conditions, collateral type and amount, borrower characteristics and geography or a combination thereof, depending on the bank’s level of sophistication. Existing credit risk grades assigned may subsequently change on either a portfolio or an individual basis due to other relevant factors such as, but not limited to, changes in industry outlook, business growth rates, consumer sentiment and changes in economic forecasts (such as interest rates, unemployment rates and commodity prices) as well as weaknesses in underwriting identified after initial recognition.

The credit risk rating system should capture all lending exposures to allow for an appropriate differentiation of credit risk and grouping of lending exposures within the credit risk rating system, reflect the risk of individual exposures and, when aggregated across all exposures, the level of credit risk in the portfolio. In this context, an effective credit risk rating system will allow a bank to identify both migration of credit risk and significant changes in credit risk.

In describing the elements of its credit risk rating system, a bank should clearly define each credit risk grade and designate the personnel responsible for the design, implementation, operation and performance of the system as well as those responsible for periodic testing and validation (ie the independent review function).

Credit risk grades should be reviewed whenever relevant new information is received or a bank’s expectation of credit risk has changed. Credit risk grades assigned should receive a periodic formal review (eg at least annually or more frequently if required in a jurisdiction) to reasonably ensure that those grades are accurate and up to date. Credit risk grades for individually assessed lending exposures that are higher-risk or credit-impaired should be reviewed more frequently than annually. ECL estimates must be updated on a timely basis to reflect changes in credit risk grades for either groups of exposures or individual exposures.

Banks should group exposures into sufficiently granular portfolios with shared credit risk characteristics to assess changes in credit risk and their impact on ECL estimates. Grouping methodologies should consider factors such as instrument type, product terms and conditions, industry/market segment, geographical location or vintages and should be documented, subject to appropriate review and internal approval.

Lending exposures should be grouped according to shared credit risk characteristics so that changes in the level of credit risk respond to the impact of changing conditions on a common range of credit risk drivers. This includes considering the effect on the group’s credit risk in response to changes in forward-looking information, including macroeconomic factors. The basis of grouping should be reviewed to ensure that exposures within the group remain homogeneous in terms of their response to credit risk drivers.

Exposures must not be grouped in such a way that an increase in the credit risk of particular exposures is masked by the performance of the group as a whole.

Banks should have in place a robust process to ensure appropriate initial grouping of their lending exposures. Subsequently, the grouping of exposures should be re-evaluated, and exposures should be re-segmented if relevant new information is received or a bank’s changed expectations of credit risk suggest that a permanent adjustment is warranted. If a bank is not able to re-segment exposures on a timely basis, a temporary adjustment may be used.

Temporary adjustments to the allowance are adjustments which may be used to account for circumstances when it becomes evident that existing or expected risk factors have not been considered in the credit risk rating and modelling process. The Committee expects that such adjustments would be used only as a temporary solution – for example, in transient circumstances or when there is insufficient time to appropriately incorporate relevant new information into the existing credit risk rating system or to re-segment existing groups of lending exposures, or when lending exposures within a group react to factors or events differently than initially expected.

The use of temporary adjustments requires the application of significant judgment and creates the potential for bias. Temporary adjustments should be directionally consistent with forward-looking forecasts, supported by appropriate documentation, and subject to appropriate governance processes.

Principle 4 – A bank’s aggregate amount of allowances, regardless of whether allowance components are determined on a collective or an individual basis, should be adequate and consistent with the objectives of the applicable accounting framework.

Banks should implement sound and robust credit risk methodologies with the objective that the overall balance of the allowance for ECL is developed in accordance with the applicable accounting framework and adequately reflects ECL within that framework.

A robust assessment of allowances considers relevant factors and expectations at the reporting date that may affect the collectability of remaining cash flows over the life of a group of lending exposures or a single lending exposure. The information that banks consider must go beyond historical and current data to consider relevant forward-looking information including macroeconomic factors that are relevant to the exposure being evaluated (eg retail or wholesale) in accordance with the applicable accounting framework.

Depending on the ability to incorporate forward-looking information into the ECL estimate, a bank may use individual or collective assessment approaches; regardless, the approach should be consistent with the relevant accounting requirements. Together, individual and collective assessments form the basis for the allowance for ECL, and a bank’s use of individual versus collective assessments, if applied appropriately, should not result in materially different allowance measurements.

The ECL estimation technique used should be the most appropriate in the circumstances and typically should be aligned with how the bank manages the credit risk exposure. For example, collective assessment is often used for large groups of homogeneous lending exposures with shared credit risk characteristics, such as retail portfolios. Individual ECL assessments are often conducted for significant exposures, or where credit concerns have been identified at the individual loan level, such as watch list and past-due loans. Regardless of the assessment approach it uses, a bank must ensure this does not result in delayed recognition of ECL. Depending on the level of sophistication of a bank’s credit risk management systems, banks may be challenged to incorporate the impact of forward-looking information, including macroeconomic forecasts, into assessments for individual borrowers, relying instead on collective assessments for a significant portion of their lending exposures, to incorporate forward-looking information.

When a bank uses individual assessments, the ECL estimate should always incorporate the expected impact of all reasonable and supportable forward-looking information, including macroeconomic factors, that affects collectability and credit risk. When applying an individual assessment approach, the bank’s documentation (similarly to what is expected when performing a collective assessment) should clearly demonstrate how forward-looking information, including macroeconomic factors, has been reflected in the individual ECL assessment.

In instances where a bank’s individual assessments of exposures do not adequately consider forward-looking information, it is appropriate to group lending exposures with shared credit risk characteristics to estimate the impact of forward-looking information, including macroeconomic factors. This process allows identification of relationships between forward-looking information and ECL estimates that may not be apparent at the individual exposure level. Conversely, when banks determine that all reasonable and supportable forward-looking information has been incorporated in the individual assessment of ECL, an additional forward-looking assessment should not be conducted on a collective basis if that could result in double-counting.

As set out in Principle 3, temporary adjustments may be necessary. If the reason for the adjustment is not expected to be temporary, such as the emergence of a new risk driver that has not previously been incorporated into the bank’s allowance methodology, the methodology should be updated in the near term to incorporate the factor that is expected to have an ongoing impact on the measurement of ECL.

Principle 5 - A bank should have policies and procedures in place to appropriately validate models used to assess and measure expected credit losses.

Models may be used in various aspects of the ECL assessment and measurement process at both the individual transaction and overall portfolio levels, including credit grading, credit risk identification, measurement of ECL allowances for accounting purposes, stress testing and capital allocation. ECL assessment and measurement models (“models”) should consider the impact of changes to borrower and credit risk-related variables such as changes in PDs, LGDs, exposure amounts, collateral values, migration of default probabilities and internal borrower credit risk grades based on historical, current and reasonable and supportable forward-looking information, including macroeconomic factors.

A bank should have robust policies and procedures in place to validate the accuracy and consistency of its model-based rating systems and processes and the estimation of all relevant risk components. Model validation should be conducted when the ECL models are initially developed and when significant changes are made to the models. A bank should regularly (for example, annually) review its ECL models.

A sound model validation framework should include, but not be limited to, the following elements:

1. Clear roles and responsibilities for model validation with adequate independence and competence. Model validation should be performed independently of the model development process and by staff with the necessary experience and expertise. Model validation involves ensuring that the models are suitable for their proposed usage, at the outset and on an ongoing basis. The findings and outcomes of model validation should be reported in a prompt and timely manner to the appropriate level of authority.3
2. An appropriate model validation scope and methodology include a systematic process of evaluating the model’s robustness, consistency and accuracy as well as its continued relevance to the underlying portfolio. An effective model validation process should also enable potential limitations of a model to be identified and addressed on a timely basis. The validation should include reviewing:
   1. Model inputs – The bank should have internally established quality and reliability standards on data (historical, current and forward-looking information) used as model inputs. Data used to estimate ECL allowances should be relevant to the bank’s portfolios, and as far as possible accurate, reliable and complete (ie without exclusions that could bias ECL estimates). Validation should ensure that the data used meet these standards.
   2. Model design – For model design, validation should demonstrate that the underlying theory of the model is conceptually sound, recognised and generally accepted for its intended purpose. From a forward-looking perspective, validation should also assess the extent to which the model, at the overall model and individual risk factor level, can take into consideration changes in the economic or credit environment, as well as changes to portfolio business profile or strategy, without significantly reducing model robustness.
   3. Model output/performance – The bank should have internally established standards for acceptable model performance. Where performance thresholds are significantly breached, remedial actions to the extent of model re-calibration or re-development should be considered.
3. Comprehensive documentation of the model validation framework and process. This includes documenting the validation procedures performed, any changes in validation methodology and tools, the range of data used, validation results and any remedial actions taken where necessary. Banks should ensure that the documentation is regularly reviewed and updated.
4. A review of the model validation process by independent parties (eg internal or external parties) to evaluate the overall effectiveness of the model validation process and the independence of the model validation process from the development process. The findings of the review should be reported in a prompt and timely manner to the appropriate level of authority (eg senior management, audit committee).

Principle 6 – A bank’s use of experienced credit judgment, especially in the robust consideration of reasonable and supportable forward-looking information, including macroeconomic factors, is essential to the assessment and measurement of expected credit losses.

Banks should have the necessary tools to ensure a robust estimate and timely recognition of ECL. Information on historical loss experience or the impact of current conditions may not fully reflect the credit risk in lending exposures. In that context, a bank must use its experienced credit judgment to thoroughly incorporate the expected impact of all reasonable and supportable forward-looking information, including macroeconomic factors, on its estimate of ECL. A bank’s use of its experienced credit judgment must be documented in the bank’s credit risk methodology and subject to appropriate oversight.

Historical information provides a useful basis for the identification of trends and correlations needed to identify the credit risk drivers for lending exposures. However, ECL estimates must not ignore the impact of (forward-looking) events and conditions on those drivers. The estimate should reflect the expected future cash shortfalls resulting from such impact.

Consideration of forward-looking information is essential to the proper implementation of an ECL accounting model and should not be avoided on the basis that a bank considers the cost of incorporating forward-looking information to be excessive or unnecessary or because there is uncertainty in formulating forward-looking scenarios. Additional cost and operational burden may not be introduced where they do not contribute to a high-quality implementation of an ECL accounting framework.

Banks should be able to demonstrate that the forward-looking information factored into the ECL estimation process has a link to the credit risk drivers for particular exposures or portfolios. For a variety of reasons, it may not be possible to demonstrate a strong link in formal statistical terms between certain types of information, or even the information set as a whole, and the credit risk drivers. Particularly in those circumstances, a bank’s experienced credit judgment will be crucial in establishing an appropriate level for the individual or collective allowance. When a forward-looking factor that has been identified as relevant is not incorporated into the individual or collective assessment, temporary adjustments may be necessary.

Macroeconomic forecasts and other relevant information should be applied consistently across portfolios where the credit risk drivers of the portfolios are affected by these forecasts/assumptions in the same way. Furthermore, when developing ECL estimates, a bank should apply its experienced credit judgment to consider its point in the credit cycle, which may differ across the jurisdictions in which it has lending exposures.

Banks should exercise care when determining the level of ECL allowances to be recognised for accounting purposes to ensure that the resulting estimates are appropriate (ie consistent with neutrality and neither understated nor overstated).

Banks use a wide range of information, including forward-looking information, for risk management and capital adequacy purposes. Information from all stages of the credit risk management process should inform ECL estimates.

Principle 7 – A bank should have a sound credit risk assessment and measurement process that provides it with a strong basis for common systems, tools and data to assess credit risk and to account for expected credit losses.

There is commonality in the processes, systems, tools and data used to assess credit risk, measure ECL for accounting purposes and determine expected losses for capital adequacy purposes. The use of common processes, systems, tools and data strengthens, to the maximum extent possible, the consistency of the resulting estimates and minimises disincentives to following sound credit risk practices for all purposes.

Credit risk practices should meet fundamental requirements including having the appropriate tools to identify and assess credit risk. These fundamental requirements are equally necessary for assessing credit risk and fairly representing the bank's financial position for accounting and capital adequacy purposes. These common processes strengthen the reliability and consistency of ECL estimates, increase transparency and, through market discipline, provide incentives to follow sound credit risk practices.

Credit risk practices should not be static and should be reviewed periodically to ensure that relevant data available throughout a banking organisation are captured and that systems are updated as the bank’s underwriting or business practices change or evolve over time. Moreover, a feedback loop should be established to ensure that information on estimates of ECL, changes in the credit risk and actual losses experienced on loans is shared among credit risk experts, accounting and regulatory reporting staff, and with the loan underwriting personnel.

Common processes, systems, tools and data that are used in assessing credit risk and measuring ECL for accounting purposes and expected losses for capital adequacy purposes could include credit risk rating systems, estimated PDs (subject to appropriate adjustments), past-due status, loan-to-value ratios, historical loss rates, product type, amortisation schedule, down payment requirements, market segment, geographical location, vintage, and collateral type.

ECL allowances may vary across jurisdictions due to differences in accounting standards. However, consistent and sound credit risk practices should be applied to narrow differences in interpretations and practices across jurisdictions.

Principle 8 – A bank’s public disclosures should promote transparency and comparability by providing timely, relevant and decision-useful information.

Banks should disclose, quantitative and qualitative information to communicate to users the main assumptions/inputs used to develop ECL estimates. Additionally, disclosures should highlight policies and definitions that are integral to the estimation of ECL (such as a bank’s basis for grouping lending exposures into portfolios with similar credit risk characteristics and its definition of default, guided by the definition used for regulatory purposes, factors that cause changes in ECL estimates, and the way management’s experienced credit judgment has been incorporated. Disclosure of significant policies should be decision-useful and should describe, in the specific context of the bank, how those policies have been implemented.

Banks should provide qualitative disclosures on how forward-looking information has been incorporated into ECL estimates, particularly for individual assessments.

Banks should disclose how management satisfies itself that lending exposures are appropriately grouped, such that these groups continue to share credit risk characteristics.

Banks should explain, based on qualitative and quantitative information, significant changes to the estimation of ECL from period to period.

Management should regularly review disclosure policies to ensure relevance to the bank's risk profile, product concentrations, industry norms, and market conditions. Disclosures should facilitate peer comparisons and enable users to monitor changes in ECL estimates over time.

Principle 9 – Banking supervisors should be satisfied that the methods employed by a bank to determine accounting allowances lead to an appropriate measurement of expected credit losses in accordance with the applicable accounting framework.

In assessing the methods employed by a bank to estimate allowances, supervisors should be satisfied that the bank is following policies and practices consistent with the ECL measurement principles outlined in this guidance, including:

1. the procedures used by a bank to measure ECL are robust and timely and consider criteria such as updated valuations of credit risk mitigants (and, in particular, collateral), cash flow estimates based on assessments of borrower-specific factors and current and future macroeconomic conditions, together with other relevant forward-looking information that affects the expected collectability of the bank’s lending exposure;
2. the framework and methodology for establishing allowances, whether determined collectively or individually, are robust;
3. aggregate allowances on lending exposures are appropriate in accordance with relevant accounting requirements and in relation to the credit risk exposure in the bank’s portfolio;
4. uncollectability is recognised in the appropriate period through allowances or write-offs; and
5. regardless of the method used to determine ECL, the bank’s internal processes for measuring ECL take account of the credit risk that the bank has taken on and changes in the credit risk of the bank’s lending exposures.

Supervisors may make use of the work performed by internal and external auditors in reviewing a bank’s credit risk assessment and ECL measurement functions.4

Principle 10 – Banking supervisors should consider a bank’s credit risk practices when assessing a bank’s capital adequacy.5

In performing their assessments of a bank’s capital adequacy, supervisors should consider whether management has:

1. maintained effective systems and controls for identifying, measuring, monitoring and controlling the level of credit risk, significant increases in credit risk and asset quality problems in a timely manner;
2. analysed all significant relevant factors that affect credit risk and the collectability of the portfolio; and
3. established an acceptable allowance estimation process that, at a minimum, meets the principles set out in this guidance, including the relevant accounting requirements.

In communicating deficiencies or recommending improvements in a bank’s credit risk practices, supervisors should consider the full range of supervisory measures at their disposal to bring deficiencies to the attention of management and encourage timely correction. The supervisory response, including the extent of the supervisor’s communication with the board, should be commensurate with the severity of the deficiencies, the impact on the bank’s risk level and the bank’s risk profile, as well as the risk-bearing capacity of the bank and management’s responsiveness in addressing concerns. For example, supervisory responses could include the following approaches and measures:

1. communicating concerns routinely or on an ad hoc basis to a bank’s senior management and/or board and evaluating management’s response as to how the bank will address these concerns (remediation plan);
2. factoring into supervisory ratings any concerns about the bank’s credit risk practices (eg factoring this into prudential risk management or capital adequacy ratings); and
3. taking informal or formal supervisory action (which can be of a non-public or public nature) requiring management and the board to remedy the deficiencies within a specified time frame and to provide the supervisor with periodic written progress reports.

When assessing capital adequacy, supervisors should consider how a bank’s accounting and credit risk assessment policies and practices affect the measurement of the bank’s assets, earnings and, therefore, its capital position.

To the extent that credit risk assessment or ECL measurement deficiencies are significant or are not remedied on a timely basis, the supervisor should consider whether such deficiencies should be reflected in supervisory ratings or through a higher capital requirement under Pillar 2 of the Basel Framework.

This section sets out guidelines specific to banks reporting under International Financial Reporting Standards (IFRS). It is limited to providing complementary guidance on ECL requirements in the impairment sections of IFRS 9 that are not common to other ECL accounting frameworks, particularly on (i) the loss allowance at an amount equal to 12-month ECL; (ii) the assessment of significant increases in credit risk; and (iii) the use of practical expedients.

In accordance with the International Accounting Standard Board’s (IASB’s) impairment standard for financial instruments, ”if, at the reporting date, the credit risk on a financial instrument has not increased significantly since initial recognition, an entity shall measure the loss allowance for that financial instrument at an amount equal to 12-month expected credit losses.”6 A bank should always measure ECL for all lending exposures. A nil allowance should be rare7 because ECL estimates are a probability-weighted amount that should always reflect the possibility that a credit loss will occur.8

In accordance with Principle 6 of this guidance, banks should adopt an active approach to assessing and measuring 12-month ECL that enables changes in credit risk to be identified in a timely manner. The 12-month ECL is the expected cash shortfalls over the life of the lending exposure or group of lending exposures, due to loss events that could occur in the next 12 months.9 To assess whether a financial instrument should move to a lifetime expected credit loss (LEL) measure, the change in the risk of a default occurring over the expected life of the financial instrument must be considered.

IFRS 9 does not directly define default, but requires entities to define default in a manner consistent with that used for internal credit risk management. IFRS 9, paragraph B5.5.37, also includes a rebuttable presumption that default does not occur later than 90 days past due. The Committee recommends that the definition of default adopted for accounting purposes should be guided by the definition used for regulatory purposes. The default definitions provided in CRE20.104 to CRE20.105 and CRE36.68 include both:

1. a qualitative criterion by which the bank considers that the obligor is unlikely to pay its credit obligations in full, without recourse by the bank to actions such as realising security (“unlikeliness to pay” events); and
2. an objective indicator where the obligor is past due more than 90 days on any material credit obligation, equivalent to the rebuttable presumption in IFRS 9, paragraph B5.5.37.

In the Basel Framework, the “unlikeliness to pay” criterion of the debtor permits identification of default before the exposure becomes delinquent with the 90-days-past-due criterion acting as a backstop. The list of elements provided in the Basel Framework as indications of unlikeliness to pay should be implemented in a way that ensures a timely detection of “unlikeliness to pay” events that precipitate eventual cash shortfalls.

IFRS 9 requires a bank to identify significant increases in credit risk since initial recognition for all financial instruments, including those measured at 12-month ECL. IFRS 9 includes the option of making assumptions about low credit risk exposures, the application of which is addressed in PAP20.124 to PAP20.127 below. The measurement of an amount equal to 12-month ECL must be updated each reporting period, and any changes to this amount are to be recorded and monitored through the allowance account.

If a bank originates high-credit-risk exposures10 and their allowances are initially measured at 12-month ECL, the bank should monitor these exposures closely for significant increases in credit risk to ensure a timely movement of the exposure to LEL measurement. That is because high-risk exposures are likely to exhibit greater volatility and to more readily experience a rapid decline in credit risk. If a bank has a policy that allows it to extend credit for high-risk lending exposures, the rationale for extending these exposures and the associated governance process should be well documented, sound underwriting practices adhere to and commensurately robust credit risk management practices implemented.

An amount equal to 12-month ECL measurement may be determined on an individual or collective basis. A robust implementation of the IFRS 9 ECL requirements, considering the migration of credit risk, should allow increases in credit risk to be reflected in increased allowances well before exposures move, either individually or collectively, to LEL measurement.

Even if an increase in credit risk is not judged to be significant, a bank must adjust its estimate of 12-month ECL to adequately reflect changes in credit risk that have taken place.

Where a collective assessment is performed, exposures within that group must adhere to the requirements set out in Principle 3. Where information becomes available to management indicating that further or different segmentation within a group of lending exposures is required, the group should be split into subgroups and the measurement of the amount equal to 12-month ECL should be updated separately for each subgroup11 or, in the case of transient circumstances, a temporary adjustment should be applied.

Assessment of significant increases in credit risk

“The objective of the impairment requirements is to recognise LEL for all financial instruments for which there have been significant increases in credit risk since initial recognition – whether assessed on an individual or collective basis – considering all reasonable and supportable information, including that which is forward-looking.”12

If banks assess that the credit risk of an exposure increased significantly since its initial recognition, the lending exposure should be subject to LEL measurement.13 While the creditworthiness of the counterparty, and thus the ECL anticipated upon initial recognition, is taken into account in the pricing of credit at that time, a post-origination increase in credit risk may not be fully compensated by the interest rate charged.14

The IFRS 9 approach to impairment assessment and measurement is demanding in its requirements for data, analysis and use of experienced credit judgment, particularly regarding whether an exposure has suffered a significant increase in credit risk and the measurement of required 12-month ECL and LEL. Strong governance, systems and controls must be placed around these processes. Banks should have systems that are capable of handling and systematically assessing the large amounts of information that will be required to judge whether particular lending exposures or groups of lending exposures exhibit a significant increase in credit risk, and to measure LEL where that is the case. A consistent approach across entities within a consolidated group should be ensured. For example, processes should be in place to ensure that forecasts of economic conditions in different jurisdictions and economic sectors are reviewed and approved by an entity’s senior management, and that the process, controls and economic assumptions around developing forecasts and linking these to expectations of credit loss are consistent across the entity (ie at the jurisdictional and the group level). The need for consistency should not be interpreted as a requirement that the practice be identical across a group. On the contrary, within a consistent framework there may be differences across jurisdictions and products, depending for instance on the availability of data. These differences should be well documented and justified.

The IFRS 9 objective stated above means that the timely determination of whether there has been a “significant” increase in credit risk after the initial recognition of a lending exposure is crucial. Banks must have processes in place that enable them to determine this on a timely and holistic basis so that an individual exposure, or a group of exposures with similar credit risk characteristics, is transferred to LEL measurement as soon as credit risk has increased significantly, in accordance with the IFRS 9 impairment accounting requirements.

As noted in the IFRS 9 Application Guidance, the range of information that will need to be considered in making this determination is wide. In broad terms, it will include information on macroeconomic conditions, and the economic sector and geographical region relevant to a particular borrower or a group of borrowers with shared credit risk characteristics, in addition to borrower-specific strategic, operational and other characteristics. A critical feature is the required consideration of all reasonable and supportable forward-looking information in addition to information about current conditions and historical data.15

To recognise allowances on a timely basis in line with the IFRS 9 requirements, banks will need to:

1. assemble data and forward projections for the key drivers of credit risk in their portfolios; and
2. be able to quantify the credit risk in each of their exposures or portfolios based on these data and projections. This will both enable management to judge whether there has been a significant increase in credit risk and form a key input to the measurement of ECL and allowances.

It is important that banks’ analyses take into account that the determinants of credit losses very often begin to deteriorate a considerable time (months or, in some cases, years) before any objective evidence of delinquency appears in the lending exposures affected. Delinquency data are generally backward-looking, and they will seldom on their own be appropriate in the implementation of an ECL approach by banks.

For example, within retail portfolios adverse developments in macroeconomic factors and borrower attributes will generally lead to an increase in the level of credit risk long before this manifests itself in lagging information such as delinquency. Thus, to meet the objective of IFRS 9 in a robust manner, banks should consider the linkages between macroeconomic factors and borrower attributes to the level of credit risk in a portfolio based on reasonable and supportable information. To that end, banks should start with a detailed analysis of historical patterns and current trends, which would allow for identification of the most relevant credit risk drivers. Experienced credit judgment should facilitate the incorporation of current and forecasted conditions likely to affect those risk drivers, the expected cash shortfalls and therefore loss expectations.

Analyses of this kind should be performed not only in the context of portfolios of individually small credits, such as credit card exposures, but also for large, individually managed exposures. For example, for a large commercial property loan, banks should take account of the considerable sensitivity of the commercial property market in many jurisdictions to the general macroeconomic environment and consider using information such as levels of interest rates or vacancy rates to determine whether there has been a significant increase in credit risk.

Banks must have a clear policy including well developed criteria on what constitutes a “significant” increase in credit risk for different types of lending exposures. Such criteria and the reasons why these approaches and definitions are considered appropriate should be disclosed in accordance with IFRS 7, paragraph 35F. IFRS 9, paragraph 5.5.9, requires that, when making the assessment of significant increases in credit risk, “an entity shall use the change in the risk of default occurring over the expected life of the financial instrument instead of the change in the amount of expected credit losses”. In other words, this assessment is made in terms of the risk of a default occurring and not expected credit loss (ie before consideration of the effects of credit risk mitigants such as collateral or guarantees).

In developing their approach to determining a significant increase in credit risk, banks should consider each of the 16 classes of indicators in IFRS 9 (insofar as they are relevant to the financial instrument being assessed) as set out in paragraphs B5.5.17 (a)–(p) and, in addition, to consider whether there is further information that should be considered. Such indicators (in both IFRS 9 and this guidance) should not be viewed as a “checklist”. Some will be more relevant than others to assessing whether a particular type of exposure exhibits a significant increase in credit risk. At the same time, banks should take particular care to avoid the risk of a significant increase in credit risk not being acknowledged promptly when it is, in fact, present. Banks should not restrict significant increases in credit risk to situations when a financial instrument is anticipated to move to the third stage. Rather, debtors may exhibit a significant increase in credit risk without evidence that the related exposures are likely to become impaired. The fact that credit risk has increased significantly does not necessarily mean that default is probable – merely that it is more likely than at initial recognition. This point is underlined by the symmetry of the IFRS 9 model: it is possible for exposures to move to LEL but subsequently be moved back to 12-month ECL if the threshold of a significant increase in credit risk is no longer met.

While it is neither possible nor desirable for universally applicable criteria to be developed, consideration should be given to conditions (a)–(f) below in assessing a significant increase in credit risk:

1. a discretionary decision by management such that, were an existing loan newly originated at the reporting date, the element of the price of the loan that reflects the credit risk of the exposure would be significantly higher than it was when the loan was originated because of an increase in the credit risk of the specific borrower or class of borrowers since inception;
2. a decision by management to strengthen collateral and/or covenant requirements for new exposures that are like exposures already advanced because of changes in the credit risk of those exposures since initial recognition;
3. a downgrade of a borrower by a recognised credit rating agency, or within a bank’s internal credit rating system;
4. for performing credits subject to individual monitoring and review, an internal credit assessment summary credit-quality indicator that is weaker than upon initial recognition;
5. deterioration of relevant determinants of credit risk (eg future cash flows) for an individual obligor (or pool of obligors); and
6. expectation of forbearance or restructuring due to financial difficulties.

Most of the factors listed above are related to a bank’s credit risk management practices. While implementation of IFRS 9 should reflect such practices where possible, in some cases that would not be appropriate. For example, in a case where a bank manages most exposures in the same way regardless of credit risk – with the exception only of particularly strong or weak credits – the way an exposure is managed is unlikely to be a sound indicator of whether there has been a significant increase in credit risk.

In addition, the assessment of whether there has been a significant increase in credit risk for a lending exposure should take account of the more general factors below:

1. deterioration of the macroeconomic outlook relevant to a particular borrower or group of borrowers. Macroeconomic assessments must be sufficiently rich to include factors relevant to sovereign, corporate, household and other types of borrowers. Furthermore, they must address any relevant regional differences in economic performance within a jurisdiction. See Principle 6 in the main section of this guidance for additional considerations regarding guidance on the consideration of forward-looking information, including macroeconomic factors; and
2. deterioration of prospects for the sector or industries within which a borrower operates.

Accurate identification of drivers of credit risk, and reliable demonstration of the linkages between those drivers and the level of credit risk, are both critical, as a seemingly small change in a qualitative characteristic of a loan can potentially be a leading indicator of large increase in the risk of a default occurring. Furthermore, IFRS 9, paragraph 5.5.9, states that the significance of a change in credit risk since initial recognition depends on the risk of a default occurring at initial recognition. In this regard, where a bank uses changes in probability of default (PD) as a means of identifying changes in the risk of a default occurring, the significance of a given change in PD can be expressed in a ratio (or the rate of fluctuation) proportionate to the PD at initial recognition (ie a change in the PD divided by the PD at initial recognition). However, the change in PD itself (ie PD at measurement date minus PD at initial recognition) should also be taken into consideration.

It is necessary to look beyond how many “grades” a rating downgrade entails because the change in PD for a one-grade movement may not be linear (for example, the default probability over five years of an exposure rated BB is around three times that of one rated BBB, based on current data and analyses applicable to certain jurisdictions). Furthermore, because the significance of a one-grade movement would depend on the granularity of a bank’s rating system – and hence the “width” of each grade – an appropriate initial segmentation is important to ensure that a significant increase in credit risk for an individual exposure or group of exposures is not masked within a segment. As such, a bank should ensure that credit risk rating systems include enough grades to appropriately distinguish credit risk. A bank should also be mindful of the fact that a significant increase in credit risk could occur prior to a movement in a credit grade.

There are some circumstances in which an adverse movement in the factors listed in PAP20.100 to PAP20.103 might not be indicative of a significant increase in credit risk. For example, it may be the case that the default probability of an exposure rated AA is low, and not much greater than one rated AAA. However, very few bank loans are of such apparently low credit risk – and, as noted in PAP20.104, the sensitivity of default probability to rating grades may increase strongly as rating quality declines.

There could also be circumstances in which some factors move in an adverse direction but may be counterbalanced by improvement in others.16 Nonetheless, in view of the importance of detecting whether there has been a significant increase in credit risk, banks must put in place governance processes capable of reliably validating any judgment that negative factors are counterbalanced by positive ones.

Thorough consideration and full weight must be given to discretionary decisions by a bank’s management which point to a change in credit risk. For example, if because of concerns about credit risk a decision is made to intensify the monitoring of a borrower or class of borrowers, it is unlikely that such action would have been taken by the decision-maker had the increase in credit risk not been perceived as significant.

Sometimes a bank will assess that there has been significant increase in credit risk for some, but not all, of its exposures to a counterparty. While it is possible for this to be the case – for example, because of differences in the timing of when lending was provided – particular care should be taken in this situation to ensure that all exposures are identified where there has been a significant increase in credit risk.

IFRS 9, paragraph B5.5.1, states that, in order to meet the objective of recognising LEL for significant increases in credit risk since initial recognition, it may be necessary for the assessment to be performed on a collective basis by considering information that is indicative of significant increases in credit risk in a group or subgroup of financial instruments even if evidence of such significant increases in credit risk at the individual instrument level is not yet available. Accordingly, in instances where it is apparent that some exposures in a group have experienced a significant increase in credit risk, a subset or a proportion of the group should be transferred to LEL measurement of ECL even though it is not possible to identify this on an individual exposure basis (see IFRS 9, Illustrative Example 5).

Consistent with paragraph B5.5.6 of IFRS 9 and paragraph IE39 of the Implementation Guidance for IFRS 9, if it is not possible based on shared credit risk characteristics to identify a particular subgroup of borrowers for which credit risk has increased significantly, an appropriate proportion of the overall group should be subject to LEL measurement.

“Significant” should not be equated with statistical significance, meaning that the assessment approach should not be based solely on quantitative analysis. For portfolios which have many individually small credits, and a rich set of relevant historical data, it may be possible to identify “significant” increases in credit risk in part by utilising formal statistical techniques. However, for other exposures, that may not be feasible.

“Significant” should also not be judged in terms of the extent of impact on a bank’s primary financial statements. Even where an increase in credit risk defined in terms of probability of default is unlikely to affect the allowance made – for example, because the exposure is more than fully collateralised – identification and disclosure of such increases are likely to be important to users seeking to understand trends in the intrinsic credit risk of a bank’s loans.

The IASB ECL model is a relative model: the assessment of significant increases in credit risk is based on comparing credit risk on exposures at the reporting date relative to credit risk upon initial recognition. IFRS 9, paragraph BC5.161, and Illustrative Example 6 suggest that banks can set a maximum credit risk for particular portfolios upon initial recognition that would lead to that portfolio moving to LEL measurement when credit risk increases beyond that maximum level. This is an example of the application of the principle in the Standard, whereby changes in the risk of default need to be assessed relative to that upon initial recognition, rather than an exception to that principle. This simplification is only relevant when exposures are segmented on a sufficiently granular basis such that a bank can demonstrate that the analysis is consistent with the principles of IFRS 9. Specifically, it would be necessary to demonstrate that a significant increase in credit risk had not occurred for items in the portfolio before the maximum credit grade was reached.

Banks should develop ways of rigorously reviewing the quality of their approach to assessing whether credit risk has increased significantly. This could involve some form of analysis of the treatment of exposures through time. Management should consider whether there are additional factors that should be considered in the assessment of significant increases in credit risk which would improve the quality of their approach.

Banks should be alert to any possibility of bias being introduced that would prevent the objectives of the Standard from being met. For this reason and to implement IFRS 9 in a robust manner, practical expedients (see below) should have limited use by banks, as these have the potential to introduce significant bias. For example, as noted below, use of a 30-days-past-due criterion introduces bias leading to a move to LEL later than the objective of the Standard requires.

In cases where banks believe that their approach to implementation is likely to have introduced bias, they should correct their assessment for identified bias and thus ensure that the objective of the Standard is met (see IFRS 9, paragraphs B5.5.1–B5.5.6).

IFRS 9, in paragraphs 5.5.12 and B5.5.25–B5.5.27, sets out the requirements for the assessment of significant increases in credit risk for lending exposures whose contractual cash flows have been renegotiated or modified. For modifications that do not result in de-recognition in accordance with IFRS 9, a bank must assess whether credit risk has increased significantly by comparing (a) the risk of a default occurring at the reporting date based on the modified contractual terms with (b) the risk of default occurring upon initial recognition based on the original, unmodified contractual terms.

Modifications or renegotiations can mask increases in credit risk, resulting in ECL being underestimated, and delaying the transfer to LEL for obligors whose credit risk has significantly deteriorated, or can inappropriately result in a move from LEL measurement back to 12-month ECL measurement.

When determining whether there is a significant increase in credit risk for a modified lending exposure, a bank should demonstrate whether such modifications or renegotiations have improved or restored the ability of the bank to collect interest and principal payments compared with the situation upon initial recognition. In developing ECL estimates, a bank should also consider whether the modification or renegotiation has improved or restored the ability of the bank to collect interest and principal payments as compared with the situation prior to modification. Consideration should also be given to the substance of modified contractual cash flows as well as the implications of the modifications for the future credit risk of the exposure (taking into consideration the obligor’s credit risk). Factors to consider include;

1. whether the modification or renegotiation of the contractual terms and resulting cash flows is economically beneficial to the obligor, compared with the original, unmodified contractual terms, and how the modification economically affects the obligor’s ability to repay the debt;
2. whether factors can be identified that support a bank’s assessment of the obligor’s ability to repay the debt, including circumstances leading up to the modification, and future prospects of the obligor as a result of the modifications, considering current conditions, macroeconomic forecasts, and prospects for the sector/industry within which the obligor operates, the obligor’s business model, and the obligor’s business (management) plan that outlines the obligor’s expectations of its future performance, financial resilience and cash flows; and
3. whether the obligor’s business plan is feasible, realisable and consistent with the repayment schedule of interest and principal under the modified contractual terms of the lending exposure.

Exposures transferred to LEL that are subsequently renegotiated or modified, and not de- recognised, should not move back to 12-month ECL measurement unless there is sufficient evidence that the credit risk over the life of the exposure has not increased significantly compared with that upon initial recognition. For example, where a bank grants various concessions such as interest rate reductions or postponements of principal repayments to obligors in financial difficulty, the lending exposure may exhibit characteristics of a lower credit risk even though in reality the obligor may continue to experience financial difficulty with no realistic prospects of making scheduled repayments over the remaining term of the exposure. IFRS 9 notes that evidence that the criteria for the recognition of LEL are no longer met could include a history of up-to-date and timely payment performance against the modified contractual terms. Typically, a customer would need to demonstrate consistently good payment behaviour over a period before the credit risk is considered to have decreased. For example, a history of missed or incomplete payments would not typically be erased by simply making one payment on time following a modification of the contractual terms.

Use of practical expedients

IFRS 9 includes several practical expedients, intended to ease the implementation burden for a wide range of companies. Internationally active banks’ use of the practical expedients will be limited, because the cost of obtaining relevant information is not likely to involve “undue cost or effort”.

In instances where the practical expedients below are applied, banks should clearly document justifications for their use. They will be subject to increased scrutiny by supervisors to determine appropriateness.

The information set

IFRS 9 states that “an entity shall consider the best reasonable and supportable information that is available, without undue cost and effort” and that “an entity need not undertake an exhaustive search for information”.17 Banks should not read these statements restrictively. Since the objective of the IFRS 9 model is to deliver fundamental improvements in the measurement of credit losses, banks should develop systems and processes that use all reasonable and supportable information that is relevant to the group or individual exposure, as needed to achieve a high- quality, robust and consistent implementation of the approach. This will potentially require costly upfront investments in new systems and processes but t the long-term benefit of a high-quality implementation far outweighs the associated costs, which should therefore not be considered undue. Nevertheless, additional cost and operational burden need not to be introduced where they do not contribute to a high-quality implementation of IFRS 9.

“Low credit risk” exemption18

For “low credit risk” exposures, entities have the option not to assess whether credit risk has increased significantly since initial recognition. However, banks should conduct timely assessment of significant increases in credit risk for all lending exposures. The use of this exemption by banks should be limited and its use for the purpose of omitting the timely assessment and tracking of credit risk would reflect a low-quality implementation of the ECL model and IFRS 9.

To achieve a high-quality implementation of IFRS 9, any use of the low-credit-risk exemption must be accompanied by clear evidence that credit risk as of the reporting date is sufficiently low that a significant increase in credit risk since initial recognition could not have occurred.

According to IFRS 9, paragraph B5.5.22, the credit risk on a financial instrument is considered low if:

1. the financial instrument has a low risk of default;
2. the borrower has a strong capacity to meet its contractual cash flow obligations in the near term; and
3. adverse changes in economic and business conditions in the longer term may, but will not necessarily, reduce the ability of the borrower to fulfil its contractual cash flow obligations.

IFRS 9, paragraph B.5.5.23, cites as an example for low credit risk an instrument with an external “investment grade” rating. However, lending exposures that have an “investment grade” rating from a credit rating agency cannot automatically be considered low credit risk. Banks should rely primarily on their own credit risk assessments to evaluate the credit risk of a lending exposure, and not solely or mechanistically on external ratings. Nevertheless, optimistic internal credit ratings, as compared with external ratings, would require additional analysis and justification by management.

More-than-30-days-past-due rebuttable presumption

Delinquency is a lagging indicator of significant increases in credit risk. Banks should have credit risk assessment and management processes in place to ensure that credit risk increases are detected well ahead of exposures becoming past due or delinquent. As noted in PAP20.95 and PAP20.115, a bank should not use the more-than-30-days-past-due rebuttable presumption as a primary indicator of transfer to LEL, while recognising that appropriate use of this rebuttable presumption as a backstop measure would not be precluded in accordance with IFRS 9 alongside other, earlier indicators for assessing significant increase in credit risk.

Any assertion that the more-than-30-days-past-due presumption is rebutted because there has not been a significant increase in credit risk will be accompanied by a thorough analysis clearly evidencing that 30 days past due is not correlated with a significant increase in credit risk.19 Such analysis should consider both current and reasonable and supportable forward-looking information that may cause future cash shortfalls to differ from historical experience.

In this regard, a bank should use relevant forward-looking information that is reasonable and supportable, to analyse whether there is any substantive relationship between such information and credit risk drivers. A bank should not use the 30-days-past-due rebuttable presumption unless it has demonstrated that the forward-looking information had no substantive relationship with the credit risk driver or such information is not available without undue cost or effort.20

In the limited instances where past-due information is the best criterion available to a bank to determine when exposures should move to the LEL category, banks should pay particular attention to their measurement of 12-month ECL allowance to ensure that ECL are appropriately captured in accordance with the measurement objective of IFRS 9. Moreover, banks should recognise that significant reliance on backward-looking information will introduce bias into the implementation of an ECL model and that banks should pay particular attention to ensuring that the objectives of the IFRS 9 impairment requirements (ie to reflect ECL that meet the stated measurement objectives and to capture all significant increases in credit risk) are met.