Attachment 5.5: Minimum Requirements for Risk Quantification Under IRB Approach
1. Introduction
1.1 Terminology
1.1.1 Abbreviations and other terms used in this paper have the following meanings:
• “PD” means the probability of default of a counterparty over one year;
• “LGD” means the loss incurred on a facility upon default of a counterparty relative to the amount outstanding at default;
• “EAD” means the expected gross exposure of a facility upon default of a counterparty;
• “Dilution risk” means the possibility that the amount of a receivable is reduced through cash or non-cash credits to the receivable’s obligor;
• “EL” means the expected loss on a facility arising from the potential default of a counterparty or the dilution risk relative to EAD over one year;
• “IRB Approach” means Internal Ratings-based approach;
• “Foundation IRB Approach” means that, in applying the IRB framework, banks provide their own estimates of PD and use supervisory estimates of LGD and EAD, and, unless otherwise specified by the SAMA, are not required to take into account the effective maturity of credit facilities;
• “Advanced IRB Approach” means that, in applying the IRB framework, banks use their own estimates of PD, LGD and EAD, and are required to take into account the effective maturity of the credit facilities;
• A “borrower grade” means a category of creditworthiness to which borrowers are assigned on the basis of a specified and distinct set of rating criteria, from which estimates of PD are derived. The grade definition includes both a description of the degree of default risk typical for borrowers assigned the grade and the criteria used to distinguish that level of credit risk;
• A “facility grade” means a category of loss severity in the event of default (as measured by LGD or EL) to which transactions are assigned on the basis of a specified and distinct set of rating criteria. The grade definition involves assessing the amount of collateral, and reviewing the term and structure of the transaction (such as the lending purpose, repayment structure and seniority of claims);
• A “rating system” means all of the methods, processes, controls, and data collection and IT systems that support the assessment of credit risk, the assignment of internal risk ratings, and the quantification of default and loss estimates;
• “Seasoning” means an expected change of risk parameters over the life of a credit exposure;
• “VAR” means value-at-risk.
1.2 Application
1.2.1 The requirements set out in this paper are applicable to locally incorporated banks, which use or intend to use the IRB Approach to measure capital changes for credit risk in KSA.
1.2.2 In the case of banks that are branches of foreign banking groups, all or part of their IRB systems may be centrally developed and monitored on a group basis. In applying the requirements of this paper, SAMA will consider the extent to which reliance can be placed on the work done at the group level. Where necessary, SAMA will co-ordinate with the home supervisors of those banking groups regarding the assessment of the comprehensiveness and integrity of the group- wide internal rating systems adopted by their authorized bank in Saudi Arabia. SAMA will also assess whether the relevant systems or models can adequately reflect the specific risk characteristics of the bank’ domestic portfolios.
1.3 Background and Scope
1.3.1 The IRB Approach to the measurement of credit risk for capital adequacy purposes relies on banks’ internally generated inputs to the calculation of capital. To minimize the variation in the way in which the IRB Approach is carried out and to ensure significant comparability across banks, SAMA considers it necessary to establish minimum qualifying criteria concerning the comprehensiveness and integrity of the internal rating systems of banks adopting the IRB Approach. SAMA will employ these criteria for assessing their eligibility to use the IRB Approach.
1.3.2 This paper:
prescribes the minimum requirements relating to risk quantification under the IRB Approach that a bank should comply with at the outset and on an ongoing basis if it were to use the IRB Approach to measure credit risk for capital adequacy purposes; and
Sets out SAMA’s supervisory approach to circumstances where a bank is not in full compliance with the minimum requirements.
1.3.3 The minimum requirements set out herein apply to both the Foundation IRB Approach and the Advanced IRB Approach and to all asset classes1, unless stated otherwise.
1.3.4 The minimum requirements for risk quantification of equity exposures under the PD/LGD Approach are the same as those of the Foundation IRB Approach for corporate exposures, subject to the specifications set out in the Basel II document. The minimum requirements for adopting the internal models approach to calculation of capital charges for equity exposures are set out in section 8 below.
The requirements for internal rating systems described in this paper should be read in conjunction with the “Minimum Requirements for Internal Rating Systems under IRB Approach”.
1 Under the IRB Approach, assets are broadly categorized into five classes: (i) corporate (with specialized lending as a sub-class); (ii) sovereign; (iii) bank; (iv) retail; and (v) equity. Within the corporate and retail asset classes, a distinct treatment for purchased receivables may also apply provided certain conditions are met.
2. Composition of Minimum Requirements
2.1 Overview
2.1.1 The IRB requirements focus on a bank’s ability to rank order and quantify risk in a consistent, reliable and valid manner, and generally fall within the following categories:
(i) Rating system design;
(ii) Rating system operations;
(iii) Corporate governance and oversight;
(iv) Use of internal ratings;
(v) Risk quantification;
(vi) Validation of internal estimates;
(vii) Supervisory LGD and EAD estimates;
(viii) Requirements for recognition of leasing;
(ix) Calculation of capital charges for equity exposures –internal models approach; and
(x) Disclosure requirements.
2.1.2 The minimum requirements under categories (v) to (ix) are detailed in sections 4 to 8 below while those requirements under categories (i) to (iv) and (x) are prescribed in the “Minimum Requirements for Internal Rating Systems under IRB Approach”.
2.1.3 The overarching principle behind the requirements is that an IRB-compliant rating system should provide for a meaningful assessment of borrower and transaction characteristics, a meaningful differentiation of credit risk, and reasonably accurate and consistent quantitative estimates of risk. Banks using the IRB Approach would need to be able to measure the key statistical drivers of credit risk. They should have in place a process that enables them to collect, store and utilize loss statistics over time in a reliable manner.
2.1.4 The internal ratings and risk estimates generated by the rating system should form an integral part of the bank’s daily credit risk measurement and management process.
2.1.5 Generally, all banks adopting the IRB Approach should produce their own estimates of PD1 and should adhere to the overall requirements for rating system design, operations, controls, corporate governance, use of internal ratings, recognition of leasing, calculation of capital charges for equity exposures, as well as the requirements for estimation and validation of PD measures. Banks wishing to use their own estimates of LGD and EAD should also meet the additional minimum requirements for these risk factors. See the “Minimum Requirements for Internal Rating Systems under IRB Approach” for the requirements relating to the overall architecture of internal rating systems.
1 Banks are not required to produce their own estimates of PD for certain equity exposures and certain exposures that fall within the specialized lending sub-class (see the “Risk-weighting Framework for IRB Approach” for details).
3. Compliance with Minimum Requirements
3.1 Ongoing Compliance
3.1.1 To be eligible for the IRB Approach, a bank should demonstrate to SAMA that it meets all minimum requirements at the outset and on an ongoing basis.
Furthermore, the bank’s overall credit risk management practices should be consistent with the guidelines and sound practices issued by SAMA.
3.2 Supervisory Approach to Non-Compliance
3.2.1 Where a bank adopting the IRB Approach is not in full compliance with the minimum requirements, bank should produce a plan for a timely return to compliance and seek approval from SAMA. Alternatively, the bank should demonstrate to SAMA that the effect of such noncompliance is immaterial in terms of the risk posed to the bank.
3.2.2 Failure to demonstrate immateriality or to produce and satisfactorily implement an acceptable plan will lead the SAMA to reconsider the bank eligibility for the IRB Approach. During the period of non-compliance, SAMA will consider the need for the bank to hold additional capital under the supervisory review process, or to take other appropriate supervisory action (such as reducing its credit exposures), depending on the circumstances of each case.
4. Risk Quantification
4.1 Overall Requirements for Estimation
General
4.1.1 This section addresses the broad standards for a bank’s own estimates of PD, LGD, and EAD. Except for certain equity and specialized lending exposures, all banks using the IRB Approach should estimate a PD for each internal borrower grade for corporate, sovereign and bank exposures or for each pool in the case of retail exposures.
4.1.2 PD estimates should be a long run average of one-year default rates for borrowers in the grade, with the exception of retail exposures (see paragraphs 4.4.10 to 4.4.12). Requirements specific to PD estimation are provided in subsection 4.4.
4.1.3 Banks on the Advanced IRB Approach should estimate an appropriate LGD (as defined in paragraph 4.5.1) for each of their facilities (or retail pools). Requirements specific to LGD estimation are set out in subsection 4.5. They should also estimate an appropriate long run default weighted average EAD for each of their facilities (as defined in paragraphs 4.6.1 and 4.6.2). Requirements specific to EAD estimation are set out in subsection 4.6.
4.1.4 Banks that are on the Foundation IRB Approach or do not meet the requirements for their own estimation of EAD or LGD for corporate, sovereign and bank exposures should use the supervisory estimates of these parameters.
4.1.5 The quantification process, including the role and scope of expert judgment, should be fully documented. It should cover all stages of the estimation process including data collection, estimation, mapping and application. Adequate documentation would promote consistency and allow third parties to review and replicate the entire process.
4.1.6 Periodic updates to the quantitative process should be conducted to ensure that new data and analytical techniques and evolving industry practices are incorporated into the process.
PD/LGD/EAD estimation
4.1.7 Estimates of PD, LGD and EAD measured by the quantification process should be updated at least annually or whenever it is considered necessary (e.g. when new data and other information have become available or methods for estimation have changed). The updating process should be documented in banks’ internal policies. Particular attention should be given to new business lines or portfolios in which the mix of obligors is believed to have changed substantially.
4.1.8 Estimates should be grounded in historical experience and empirical evidence, and not based purely on subjective or judgmental considerations. They should incorporate all relevant, material and available data, information and methods. Any changes in lending practice or the process for pursuing recoveries over the data observation period should be taken into account.
4.1.9 Banks may utilize internal data and data from external sources (including pooled data) in there own estimation. Where such data are used, banks should demonstrate that their estimates are representative of long run experience.
4.1.10 The population of exposures represented in the data used for estimation, and the lending standards in use when the data were generated, and other relevant characteristics should be closely matched to or at least comparable with those of a bank’s exposures and standards. The bank should also demonstrate that economic or market conditions underlying the data are relevant to current and foreseeable conditions.
For estimates of LGD and EAD, banks should take into account paragraphs 4.5.1 to 4.5.2 and 4.6.3 to 4.6.9 respectively. The number of exposures in the sample, and the data period used for quantification should be sufficient to provide a bank with confidence in the accuracy and robustness of its estimates. The estimation technique should perform well in out-of-sample tests.
4.1.11 SAMA may allow some flexibility in the application of required standards for data that are collected prior to a bank adoption of the IRB Approach. However, in such cases the bank should demonstrate to the SAMA that appropriate adjustments have been made to achieve broad equivalence with the data without such flexibility. Data collected beyond the date of adoption1 should conform to the minimum standards unless otherwise stated.
4.1.12A Date of adoption is the data a bank starts to accumulate data. For applying IRB approaches.
Conservatism
4.1.13 Judgmental adjustments may form a part of the quantification process, but should not be biased toward lower estimates of risk. Consistent signs of judgmental decisions that lower parameter estimates materially may be evidence of bias. The reasoning and empirical support for any adjustments, as well as the mechanics of the calculation, should be documented. Banks should conduct sensitivity analysis to demonstrate that the adjustment procedure is not biased toward reducing capital requirements. The analysis should consider the impact of any judgmental adjustments on estimates and risk weights, and should be fully documented.
4.1.14 Estimates of PD, LGD and EAD should incorporate a degree of conservatism that is appropriate for the overall robustness of the quantification process. In general, such estimates are likely to involve unpredictable errors. In order to avoid undue optimism, banks should add to their estimates a margin of conservatism that is related to the likely range of errors. Where methods and data are less satisfactory and the likely range of errors is larger, the margin of conservatism should be larger.
4.1.15 There should be an appropriate degree of conservatism to adequately account for all uncertainties and weaknesses relating to risk quantification. Improvements in the quantification process (e.g. use of better data and estimation techniques) may reduce the appropriate degree of conservatism over time.
4.1.16 Estimates of PD, LGD, EAD or other parameters should be presented with statistical indicators that facilitate an assessment of the appropriate degree of conservatism.
Review and validation
4.1.17 Banks should subject all aspects of the quantification process, including design and implementation, to an appropriate degree of independent review and validation. An independent review is an assessment conducted by persons not accountable for the work being reviewed. The reviewers may either be internal or external parties.
4.1.18 The review serves as a check on the quantification process to ensure that it is sound and works as intended; it should be broad-based, and should include all of the elements of the quantification process that lead to the ultimate estimates of PD, LGD and EAD. The review should cover the full scope of validation, including:
• an evaluation of the integrity of data inputs;
• an analysis of the internal logic and consistency of the process;
• a comparison with relevant benchmarks; and
• appropriate back-testing based on actual outcomes.
• Detailed requirements for ongoing validation and back testing of estimates are set out in section 5.
1 Date of adoption is the date a bank start to accumulate data on a prospective basis in conformance with SAMA’s minimum qualitative and quantitative requirements.
4.2 Definition of Default for Different Asset Classes
General definition of default.
4.2.1 A default is considered to have occurred with regard to a particular obligor when either or both of the two following events have taken place:
• A bank considers that the obligor is unlikely to pay in full its credit obligations to the bank (or the banking group1 of which it is a part), without recourse by the bank to actions such as realizing security (if held);
• The obligor is past due for more than 90 days2 on any material portion of its credit obligations to the bank (or the banking group of which it is a part). Past due credit obligations are regarded as material if they represent 5% or more of the obligor’s outstanding credit obligations. Banks may however set a lower threshold or choose not to apply the threshold based on their individual circumstances. Overdrafts will be considered as past due once the customer has breached an advised limit or been advised of a limit smaller than the current outstanding balance (see also paragraph 4.2.7). The criteria for determining overdue assets are set out in SAMA’s circular BCS # 312 of 19.1.2004 entitled “SAMA’s Rules Concerning Loan Classifications, Provisioning and Credit Review”.
4.2.2 The elements to be taken as indicators of unlikeliness to pay include:
A Bank puts the credit obligation on non-accrual status.
• The criteria for putting an obligation on non-accrual status and those for restoring the “accrual” status are set out in SAMA’s circular # 312 of 19.1.2004 entitled “SAMA circular on loan classification, provisioning and credit review”.
• A bank makes a charge-off or account-specific provision resulting from a significant perceived decline in asset quality subsequent to the bank taking on the exposure3;
• A bank sells the credit obligation at a material credit related economic loss;
• A bank gives consent to a distressed restructuring/rescheduling of the credit obligation where this is likely to result in a diminished financial obligation caused by the material forgiveness, or postponement, of principal, interest or, where relevant, fees.4 The criteria for determining rescheduled assets and those for uplifting the “rescheduled” status are set out SAMA’s circular # 3125.
• A bank has filed for the obligor’s bankruptcy or a similar order in respect of the obligor’s credit obligation to the bank;
• The obligor has sought or has been placed in bankruptcy or similar protection where this would avoid or delay repayment of the credit obligation to the bank.
4.2.3 For retail exposures, the definition of default can be applied at the level of a particular facility, rather than at the level of the obligor. As such, default by a customer on one obligation does not require a bank to treat all other obligations of the customer to the bank (or its banking group) as defaulted.
4.2.4 Banks should record actual defaults on IRB asset classes using the reference definition mentioned above. They should also use the reference definition for their estimation of PDs, and, where relevant, LGDs and EADs. In arriving at these estimations, banks may use external data available to them that are not itself consistent with that definition, subject to the requirements set out in paragraphs 4.4.3 to 4.4.7.
4.2.5 In such cases, however, bank should demonstrate to the SAMA that appropriate adjustments to the data have been made to achieve broad equivalence with the reference definition. The same condition would apply to any internal data used up to the time when a bank adopts the IRB Approach. Larger discrepancies require larger adjustments for the sake of conservatism. Internal data (including those pooled by bank) used in such estimates beyond the date of adoption of the IRB Approach should be consistent with the reference definition.
4.2.6 If a bank considers that the status of a previously defaulted exposure is such that the trigger of the reference definition no longer applies, the bank should rate the borrower and estimate LGD as it would for a non-defaulted facility. Should the reference definition be subsequently triggered, a second default would be deemed to have occurred.
Treatment of overdrafts
4.2.7 Overdraft facilities authorized by a bank to a customer should be subject to a formal credit limit and brought to the knowledge of the customer. Any breach of this limit should be monitored. If the account were not brought under the limit after 90 days, it would be considered as defaulted. Temporary or non-authorized overdrafts will be associated with a zero limit for IRB purposes. Thus, the days past due commence once any credit is granted to the customer concerned. If such credit were not repaid within 90 days, the exposure would be regarded as in default. Banks should have in place rigorous internal policies for assessing the credit-worthiness of customers who are offered overdraft accounts.
Re-ageing
4.2.8 Re-ageing is a process by which the delinquency status of loans, the terms of which have not been changed, is adjusted based on subsequent good performance, even though not all arrears under the original repayment schedule have been paid off.
The bank must have clearly articulated and documented policies in respect of the counting of days past due, in particular in respect of the re-ageing of the facilities and the granting of extensions, deferrals, renewals and rewrites to existing accounts. At a minimum, the re-ageing policy must include: (a) approval authorities and reporting requirements; (b) minimum age of a facility before it is eligible for re-ageing; (c) delinquency levels of facilities that are eligible for re-ageing; (d) maximum number of re-ageings per facility; and (e) a reassessment of the borrower‘s capacity to repay. These policies must be applied consistently over time, and must support the ‘use test’ (i.e. if a bank treats a re-aged exposure in a similar fashion to other delinquent exposures more than the past-due cut off point, this exposure must be recorded as in default for IRB purposes). Some supervisors may choose to establish more specific requirements on re-ageing for banks in their jurisdiction.
(Refer para 458, International Convergence of Capital Measurement and Capital Standards – June 2006).
1 The banking group covers all entities within the group that are subject to the capital adequacy regime in Saudi Arabia.
2 In the event that a branch owned by a foreign banking group wants to use a different default trigger set by its home supervisor for particular exposures (e.g. 180 days for exposures to retail or public sector entities), the bank will need to satisfy SAMA that such a difference in the definition of default will not result in any material impact on the default and loss estimates generated. Where necessary, if the relevant models are centrally developed and validated at the home country, the views of the home supervisor will be sought.
3 Specific provisions on equity exposures set aside for price risk do not necessarily signal default.
4 Including, in the case of equity holdings assessed under a PD/LGD approach, such distressed restructuring of the equity itself.
5 Also see “Rescheduled Loans”, SAMA circular # 312 of 19.1.2004, which provides guidance on the definition of “rescheduled loans”.4.3 Definition of Loss for All Asset Classes
4.3.1 The definition of loss used in estimating LGD is economic loss. When measuring economic loss, all relevant factors should be taken into account. This should include material discount effects and material direct and indirect costs associated with collecting on the exposure.
4.3.2 Banks should not simply measure the loss recorded in accounting records. They should be able to compare accounting and economic losses (some Banks may also adopt the concept of economic loss in their accounting records). Banks’ own workout and collection expertise significantly influences their recovery rates, and should be reflected in their LGD estimates. However, adjustments to estimates for such expertise should be conservative until a bank has maintained sufficient internal empirical evidence to manifest the impact of its expertise.
4.4 Requirements Specific to PD Estimation
Data observation period
4.4.1 Irrespective of whether a bank is using external, internal, or pooled data sources, or a combination of the three, for its PD estimation, the length of the underlying historical observation period used must be at least five years for at least one source. If the available observation period spans a longer period for any source, and this data is relevant and material, this longer period must be used. (Refer para 463, International Convergence of Capital Measurement and Capital Standards – June 2006)
4.4.1A Irrespective of whether banks are using external, internal, pooled data sources, or a combination of the three, for their estimation of loss characteristics, the length of the underlying historical observation period used must be at least five years.
If the available observation spans a longer period for any source, and these data are relevant, this longer period must be used. A bank need not give equal importance to historic data if it can convince its supervisor that more recent data are a better predictor of loss rates.
(Refer para 466, International Convergence of Capital Measurement and Capital Standards – June 2006)
4.4.2 The SAMA applies the transitional requirement of a minimum of two years of data at the time of adopting the Foundation IRB Approach for corporate, sovereign, and bank exposures or the IRB Approach for retail exposures.
Corporate, sovereign, and bank exposures
4.4.3 Bank should use information and techniques that take appropriate account of the long run experience when estimating the average PD for each rating grade. For example, banks may use one or more of the three specific techniques set out below (i.e. internal default experience, mapping to external data, and statistical default models),
4.4.4 Banks may have a primary technique and use others as a point of comparison and potential adjustment. SAMA will not be satisfied by mechanical application of a technique without supporting analysis. Banks should recognize the importance of judgmental considerations in combining results of techniques and in making adjustments for limitations of techniques and information.
4.4.5 Banks may use data on internal default experience for the estimation of PD. They should demonstrate in their analysis that the estimates are reflective of actual default experience and of any differences in the rating system that generated the data and the current rating system. Where only limited data are available, or where underwriting standards or rating systems have changed, Banks should add a greater margin of conservatism in their estimate of PD. The use of pooled data across banks may also be recognized. A bank should demonstrate that the internal rating systems and criteria of other bank in the pool are comparable with its own.
4.4.6 Banks may associate or map their internal grades to the scale used by an external credit assessment institution (“ECAI”) and then attribute the default rate observed for the ECAI’s grades to the bank’s grades. Mappings should be based on a robust comparison of internal rating criteria to the criteria used by the ECAI and on a comparison of the internal and external ratings of any common borrowers. Biases or inconsistencies in the mapping approach or underlying data should be avoided.
4.4.7 The ECAI’s criteria underlying the data used for quantification should be oriented to the risk of the borrower and not reflect transaction characteristics. A bank’s analysis should include a comparison of the default definitions used, subject to the requirements in subsection 4.2 above. The bank should document the basis for the mapping.
4.4.8 Banks that aggregate the PD of individual portfolio obligors when calculating PD estimates for internal grades should have a clear policy governing the aggregation process. A mean of PD estimates for individual borrowers in a given grade should be used. A bank would only be allowed to calculate this estimate differently if it can demonstrate that the alternative method provides a better estimate of the long run average PD. To obtain this evidence, the bank should at least compare the results of both methods.
4.4.9 Banks’ use of default probability models for estimating PD should meet the standards specified in subsection 4.6 of the “Minimum Requirements for Internal Rating Systems under IRB Approach”.
Retail exposures
4.4.10 Given the bank specific basis of assigning exposures to pools, banks should regard internal data as the primary source of information for estimating loss characteristics. Banks are permitted to use external data or statistical models for quantification provided a strong link can be demonstrated between:(i) the bank’s process of assigning exposures to a pool and the process used by the external data source; and (ii) the bank’s internal risk profile and the composition of the external data. In all cases banks should use all relevant and material data sources as points of comparison.
4.4.11 One method for deriving long run average estimates of PD and default-weighted average loss rates given default (as defined in 4.5.1) for retail would be based on an estimate of the expected long run loss rate. A bank may (i) use an appropriate PD estimate to infer the long run default-weighted average loss given default; or (ii) use a long run default-weighted average loss rate given default to infer the appropriate PD. In either case, it is important to recognize that the LGD used for the IRB capital calculation cannot be less than the long run default-weighted average loss rate given default and should be consistent with the concept defined in paragraph 4.5.1.
4.4.12 Seasoning can be quite material for some long-term retail exposures characterized by seasoning effects that peak several years after origination. Banks should anticipate the implications of rapid exposure growth and take steps to ensure that their estimation techniques are accurate, and that their current capital level and earnings and funding prospects are adequate to cover their future capital needs.
4.4.13 In order to avoid gyrations in their required capital positions arising from short-term PD horizons, banks are also encouraged to adjust PD estimates upward for anticipated seasoning effects, provided such adjustments are applied in a consistent fashion over time.
4.4.14 If a bank does not take seasoning effects into account and its own estimates of PD are considered to be too low, SAMA may require banks to use higher values of PD for the calculation of capital charges. PD’s will be considered too low if validation tests, stress tests, back testing indicates lack of predictability,
4.5 Requirements Specific to Own-LGD Estimates
4.5.1 Banks should estimate an LGD for each facility that aims to reflect economic downturn conditions where necessary to capture the relevant risks. This LGD cannot be less than the long run default-weighted average loss rate given default calculated based on the average economic loss of all observed defaults within the data source for that type of facility. In addition, a bank should take into account the potential for the LGD of the facility to be higher than the default-weighted average during a period when credit losses are substantially higher than average.
In all cases, both the borrower and all recognized guarantors must be assigned a borrower rating at the outset and on an ongoing basis. A bank must follow all minimum requirements for assigning borrower ratings set out in this document, including the regular monitoring of the guarantor’s condition and ability and willingness to honor its obligations.
Consistent with the requirements in paragraphs 430 and 431, International Convergence of Capital Measurement and Capital Standards – June 2006, a bank must retain all relevant information on the borrower absent the guarantee and the guarantor. In the case of retail guarantees, these requirements also apply to the assignment of an exposure to a pool, and the estimation of PD.
(Refer para 481, International Convergence of Capital Measurement and Capital Standards – June 2006)
4.5.2 For certain types of exposures, loss severities may not exhibit such cyclical variability and LGD estimates may not differ materially (or possibly at all) from the long run defaulted-weighted average. However, for other exposures, this cyclical variability in loss severities may be important and bank will need to incorporate it into their LGD estimates. For this purpose, banks may use averages of loss severities observed during periods of high credit losses, forecasts based on appropriately conservative assumptions, or other similar methods. Appropriate estimates of LGD during periods of high credit losses might be formed using either internal and/or external data. SAMA will continue to monitor and encourage the development of appropriate approaches to this issue.
4.5.3 In its analysis, a bank should consider the extent of any dependence between the risk of the borrower and that of the collateral or collateral provider. Cases where there is a significant degree of dependence should be addressed in a conservative manner. Any currency mismatch between the underlying obligation and the collateral should also be considered and treated conservatively in the bank’s assessment of LGD.
4.5.4 LGD estimates should be grounded in historical recovery rates and, when applicable, should not solely be based on the estimated market value of collateral. This requirement recognizes the potential inability of banks to gain both control of their collateral and liquidate it expeditiously. To the extent, that LGD estimates take into account the existence of collateral, bank should establish internal requirements for collateral management, operational procedures, legal certainty and risk management process that are generally consistent with those required for the Standardized Approach for calculating credit risk capital changes.
4.5.5 Recognizing the principle that realized losses can at times systematically exceed expected levels, the LGD assigned to a defaulted asset should reflect the possibility that the bank would have to recognize additional, unexpected losses during the recovery period. For each defaulted asset, the bank should also construct its best estimate of the expected loss on that asset based on current economic circumstances and facility status. The amount, if any, by which the LGD on a defaulted asset exceeds the bank’s best estimate of expected loss on the asset represents the capital requirement for that asset, and should be set by the bank on a risk-sensitive basis. Instances where the best estimate of expected loss on a defaulted asset is less than the sum of specific provisions and partial charge- offs on that asset will attract supervisory scrutiny and should be justified by the bank.
4.5.6 Estimation of LGD may involve mapping facility-specific data elements in a bank’s portfolio to the factors in reference data sets used by ECAIs. The mapping process should be based on a robust comparison of available common elements in the reference data and the bank’s portfolio. The bank should also have a policy describing how it combines multiple sets of reference data. Biases or inconsistencies in the mapping approach or underlying data should be avoided.
4.5.7 Banks that aggregate LGD estimates for facility grades from individual exposures should have a clear policy governing the aggregation process. In general, simple averaging is preferred. This requirement is however irrelevant for bank that choose to assign LGD estimates directly to individual exposures rather than grades, because aggregation is not required in that case.
4.5.8 For corporate, sovereign, and bank exposures, estimates of LGD should be based on a minimum data observation period that should ideally cover at least one complete economic cycle but should in any case be no shorter than a period of seven years for at least one source. If the available observation period spans a longer period for any source, and the data are relevant, this longer period should be used.
4.5.9 For retail exposures, the minimum data observation period for LGD estimates is five years. The less data a bank has, the more conservative it should be in its estimation. A bank need not give equal importance to historical data if it can demonstrate to SAMA that more recent data are a better predictor of loss rates.
4.6 Requirements Specific to Own-EAD Estimates
4.6.1 EAD for an on-balance sheet or off-balance sheet item is defined as the expected gross exposure of the facility upon default of the obligor. For on-balance sheet items, banks should estimate EAD at no less than the current drawn amount, subject to recognizing the effects of on balance sheet netting as specified in the Foundation IRB Approach (see the ”Risk-Weighting Framework for IRB Approach”). The minimum requirements for the recognition of netting are the same as those under the Foundation IRB Approach.
4.6.2 The additional minimum requirements for internal estimation of EAD under the Advanced IRB Approach, therefore, focus on the estimation of EAD for off- balance sheet items (excluding derivatives). Banks using the Advanced IRB Approach should have established procedures in place for the estimation of EAD for off balance sheet items. These should specify the estimates of EAD to be used for each facility type. Banks’ estimates of EAD should reflect the possibility of additional drawings by the borrower up to and after the time a default event is triggered. Where estimates of EAD differ by facility type, the delineation of these facilities should be clear and unambiguous.
4.6.3 Banks using the Advanced IRB Approach should assign an estimate of EAD for each facility. It should be an estimate of the long run default-weighted average EAD for similar facilities and borrowers over a sufficiently long period of time, but with a margin of conservatism appropriate to the likely range of errors in the estimate.
4.6.4 If a positive correlation can reasonably be expected between the default frequency and the magnitude of EAD, the EAD estimate should incorporate a larger margin of conservatism. Moreover, for exposures for which EAD estimates are volatile over the economic cycle, banks should use EAD estimates that are appropriate for an economic downturn, if these are more conservative than the long run average.
4.6.5 For banks that have been able to develop their own EAD models, this could be achieved by considering the cyclical nature, if any, of the drivers of such models. Other banks may have sufficient internal data to examine the impact of previous recessions. However, some banks may only have the option of making conservative use of external data.
4.6.6 The criteria by which estimates of EAD are derived should be plausible and intuitive, and represent what banks believe to be the material drivers of EAD. The choices should be supported by banks’ credible internal analysis. Banks should be able to provide a breakdown of their EAD experience by the factors they see as the drivers of EAD. Banks should use all relevant and material information in their derivation of EAD estimates. Across facility types, banks should review their estimates of EAD when material new information comes to light and at least on an annual basis.
4.6.7 Due consideration must be paid by the bank to its specific policies and strategies adopted in respect of account monitoring and payment processing. The bank must also consider its ability and willingness to prevent further drawings in circumstances short of payment default, such as covenant violations or other technical default events. Banks must also have adequate systems and procedures in place to monitor facility amounts, current outstandings against committed lines and changes in outstandings per borrower and per grade. The bank must be able to monitor outstanding balances on a daily basis.
477(i). For transactions that expose banks to counterparty credit risk, estimates of EAD must fulfill the requirements set forth in Annex 4 of this Framework.
(Refer para 477, International Convergence of Capital Measurement and Capital Standards – June 2006)
4.6.8 For corporate, sovereign, and bank exposures, estimates of EAD should be based on a time period that should ideally cover a complete economic cycle but should in any case be no shorter than a period of seven years. If the available observation period spans a longer period for any source, and the data are relevant, this longer period should be used. EAD estimates should be calculated using a default-weighted average and not a time weighted average.
4.6.9 For retail exposures, the minimum data observation period for EAD estimates is five years. The less data a bank, the more conservative it should be in its estimation. A bank need not give equal importance to historical data if it can demonstrate to SAMA that more recent data are a better predictor of draw-downs.
4.6.10 SAMA applies the transitional requirement of a minimum of two years of data at the time of adopting the IRB Approach for retail exposures to banks that can implement such an approach during the period from 1 January 2007 to 31 December 2009. This requirement will increase by one year for each of the three years after year-end 2009.
5. Validation of Internal Estimates
5.1 General Requirements
5.1.1 Validation is an integral part of a bank’s rating system architecture to provide reasonable assurances about its rating system. Banks adopting the IRB Approach should have a robust system in place to validate the accuracy and consistency of their rating systems, processes and the estimation of all relevant risk components. They should demonstrate to SAMA that their internal validation process enables them to assess the performance of internal rating and risk estimation systems consistently and meaningfully.
5.1.2 The validation process should include review of rating system developments (see subsection 5.2), ongoing analysis (see subsection 5.3), and comparison of predicted estimates to actual outcomes (i.e. back-testing, as described paragraphs 5.1.3 and 5.1.4 and subsection 5.4).
5.1.3 Banks should regularly compare realized default rates with estimated PDs for each grade and be able to demonstrate that the realized default rates are within the expected range for that grade. The actual long run average default rate for each rating grade should not be significantly greater than the PD assigned to that grade. The methods and data used in such comparisons by banks should be clearly documented. This analysis and documentation should be updated at least annually.
5.1.4 Similarly, banks using the Advanced IRB Approach should complete such analysis for their estimates of LGD and EAD. Such comparisons should make use of historical data that are over as long a period as possible. The actual loss rates experienced on defaulted facilities should not be significantly greater than the LGD estimates assigned to those facilities.
5.1.5 Banks should also use other quantitative validation tools and comparisons with relevant external data sources. The analysis should be based on data that are appropriate to the portfolio, are updated regularly, and cover a relevant observation period. Banks’ internal assessments of the performance of their own rating systems should be based on long data histories, covering a range of economic conditions, and ideally one or more complete business cycles.
5.1.6 Banks should have in place a process for vetting data inputs, including the assessment of accuracy, completeness and appropriateness of the data specific to the assignment of an approved rating. Detailed documentation of exceptions to data input parameters should be maintained and reviewed as part of the process cycle of validation.
5.1.7 The process cycle of validation should also include: ongoing periodic monitoring of rating system performance, including evaluation and rigorous statistical testing of the dynamic stability of the models used and their key coefficients; identifying and documenting individual fixed relationships in the rating system or model that are no longer appropriate; and a rigorous change control process, which stipulates the procedures that should be followed prior to making changes in the rating system or model in response to validation outcomes.
5.1.8 Bank should demonstrate that quantitative testing and other validation methods do not vary systematically with the economic cycle1 which incorporate the general impact of economic downturn and upswings of the subject economy. Changes in methods and data (both data sources and periods covered) should be clearly documented.
5.1.9 Some differences across individual grades between observed outcomes and the estimates can be expected.
However, if systematic differences suggest a bias toward lowering regulatory capital requirements, the integrity of the rating system (of either the PD or LGD dimensions or of both) becomes in doubt.
5.1.10 Bank should have well-articulated internal standards for situations where deviations in realised PDs, LGDs and EADs from expectations become significant enough to call the validity of the estimates into question. These standards should take account of business cycles and similar systematic variability in default experiences. Where realised values continue to be higher than expected values, banks should revise estimates upward to reflect their default and loss experience.
1 Economic cycle refer to ensuring that validation of internal estimates incorporate the general impact of economic downturn and upswings of the subject economy.
5.2 Review of Rating System Developments
5.2.1 The first analytical support for the validity of a bank’s rating system is review of rating system developments, in particular analyzing its design and construction. The aim of the review is to assess whether the rating system could be expected to work reasonably if it is implemented as designed. Such review should be revisited whenever the bank makes a change to its rating system. As the rating system is likely to change over time as the bank learns about the effectiveness of the system, the review is likely to be an ongoing part of the process. The particular steps taken in the review depends on the type of rating system.
5.2.2 Regarding a model-based rating system, the review of rating system developments should include information on the logic that supports the model and an analysis of the statistical model-building techniques. The review should also include empirical evidence on how well the ratings might have worked in the past, as such models are chosen to maximize the fit to outcomes in the development sample. In addition, statistical models should be supported by evidence that they work well outside the development sample. Use of out-of-time and out-of-sample performance tests is a good model-building practice to ensure that the model is not merely a statistical quirk of the particular data set used to build the model. Where a bank uses scoring systems for assigning credit ratings, it should demonstrate that those systems have adequate discriminating power.
5.2.3 Regarding an expert judgment-based rating system, the review of rating system developments requires asking two groups of raters how they would rate credits based on the rating definitions, processes and criteria for assigning exposures to grades within the rating system (see sections 4 and 5 of the “Minimum Requirements for Internal Rating Systems under IRB Approach” on requirements for rating criteria and processes). These two sets of rating results could then be compared to determine whether the ratings were consistent. Conducting such tests would help identify any factors, which may lead to different or inconsistent ratings. While some differences and inconsistencies may arise from the exercise of judgment, those findings should be considered for the development of the rating system.
5.2.4 Where an expert judgment-based rating system which employs quantitative guidelines or model results as inputs, the review of the rating system that features guidance values of financial ratios or scores of a scoring model might include a description of the logic and evidence relating the values of the ratios or scores to past default and loss outcomes.
5.3 Ongoing Analysis
5.3.1 The second analytical support for the validity of a bank’s rating system is the ongoing analysis intended to confirm that the rating system is implemented and continues to perform as intended. Such analysis involves process verification and benchmarking.
Process verification
5.3.2 Specific verification activities depend on the rating approach. If a model is used for rating, verification requires reviewers who are independent of the model development to evaluate the soundness of the model, including the theory, assumptions and mathematical/empirical basis. In addition, the evaluation should include the assessment of the compliance with the requirements set out in subsection 4.6 of the “Minimum Requirements for Internal Rating Systems under IRB Approach” on use of models.
5.3.3 If expert judgment is used for rating, verification requires other individual reviewers to evaluate whether the rater has followed rating policy. The minimum requirements for verification of ratings assigned by individuals are:
• a transparent rating process;
• a database with information used by the rater; and
• documentation of how the decisions were made.
5.3.4 Rating process verification also includes override monitoring. The requirements for overrides are set out in subsection 5.3 of the “Minimum Requirements for Internal Rating Systems under IRB Approach”. A reporting system capturing data on reasons for overrides could facilitate learning about whether overrides improve accuracy.
Benchmarking
5.3.5 Benchmarking is a set of activities that uses alternative tools to draw inferences about the correctness of ratings before outcomes are actually known. Benchmarking of a rating system demonstrates whether another rater or rating method attaches the same rating to a particular obligor or facility. At a minimum, banks should establish a process in which a representative sample of its internal ratings is compared to third-party ratings (e.g. independent internal raters, external rating agencies, models, or other market data sources) of the same credits. Regardless of the rating approach, the benchmark can either be a judgment-based or a model based rating. Examples of such benchmarking include: rating reviewers completely re-rate a sample of credits rated by individuals in a judgment-based system; an internally developed model is used to rate credits rated earlier in a judgment-based system; individuals rate a sample of credits rated by a model; internal ratings are compared against results from external agencies or external models.
Banks can also consider benchmarking which includes activities designed to draw broader inferences about whether the rating system – as opposed to individual ratings – is working as expected. Bank can look for consistency in ranking or consistency in the values of rating characteristics for similarly rated credits. Examples of such benchmarking activities include:
analyzing the characteristics of obligors that have received common ratings; monitoring changes in the distribution of ratings over time;
calculating a transition matrix from changes in ratings in a bank portfolio and comparing it to historical transition matrices from publicly available ratings or external data pools.
5.3.6 If benchmarking evidence suggests a pattern of rating differences, it should lead the bank to investigate the source of the differences. Thus, the benchmarking process illustrates the possibility of feedback from ongoing validation to model development.
5.4 Back-Testing
5.4.1 Back-testing is the comparison of predictions with actual outcomes. It is the empirical test of the accuracy and calibration of the estimates, i.e. PDs, LGDs and EADs, associated with borrower and facility ratings, respectively.
5.4.2 At a minimum, banks should:
• develop their own statistical tests to back-test their rating systems;
• establish internal tolerance limits for differences between expected and actual outcomes; and
• have a policy that requires remedial actions be taken when policy tolerances are exceeded.
5.4.3 However, the data to perform comprehensive back testing would not be available in the early stages of implementing an IRB rating system. Therefore, banks should rely more heavily on review of rating system developments, process verification, and benchmarking to assure themselves and other interested parties that there rating systems are likely to be accurate. Validation in its early stages should also depend on a bank’s management exercising informed judgment about the likelihood of the rating system working — not simply on empirical tests.
5.4.4 Where banks rely on supervisory, rather than internal, estimates of risk parameters, they are encouraged to compare realised LGDs and EADs to those set by the SAMA. The information on realised LGDs and EADs should form part of a bank’s assessment of economic capital.