Banking (Capital)(Amendment) Rules 2023 - Supplementary Guidance on the Revised Credit Risk Framework

Banking Policy Department
Our Ref: B1/15C
B9/75C
S4/3C
13 December 2024
The Chief Executive
All Authorized Institutions
Dear Sir/Madam,
Banking (Capital)(Amendment) Rules 2023 - Supplementary Guidance on the Revised Credit Risk Framework I am writing to inform you that, following consultation with the two industry Associations, the HKMA is issuing the above guidance today. The purpose of the guidance is to facilitate consistent application of the revised credit risk framework introduced by the Banking (Capital) (Amendment) Rules 2023 (BCAR) that take effect on 1 January 2025. The contents of the guidance (mostly in the form of Q&A) include—

• Chapter I - Introduction
• Chapter II - Use of ECAI ratings for the purposes of the Banking (Capital)
  Rules (BCR)
• Chapter III - STC approach
• Chapter IV - IRB approach
• Chapter V - Counterparty credit risk
• Chapter VI - Securitization
  Chapter II contains guidance on the administrative arrangement for ECAI
  nomination under the revised credit risk framework. Chapters III and IV1 are largely a compilation of the HKMA’s responses to the comments received from the industry during previous consultations on the BCAR, Chapter IV also 1 Chapter IV is currently under revision, primarily to accommodate the review of the SPM module CA￾G-4 on “Validating Risk Rating Systems under the IRB Approach”. Please refer to the HKMA’s letters to the HKAB and the DTC Association dated 29 November 2024, available on the HKMA’s public website (https://www.hkma.gov.hk/eng/regulatory-resources/consultations/) and the private website (http://www.stet.iclnet.hk/index.htm).

2 superseding the “IRB Approach” section of the “Questions and Answers on Banking (Capital) Rules”, which was last updated on 31 December 2014 2 .
Chapter V is an updated version of the guidance on counterparty credit risk capital
standard issued by the HKMA on 3 June 20213 , reflecting the revised capital requirements introduced by the BCAR (e.g. reclassification of CVA risk from counterparty credit risk to market risk). Chapter VI is essentially the implementation guidance on securitization framework issued on 26 March 20184 , with updates to the subsection references to section 230 of the BCR. Chapters III, IV and V also incorporate clarifications made by the HKMA in light of a few enquires received from individual authorized institutions, and certain Basel requirements that have not been included in the BCAR (e.g. the eligibility criteria for a covered bond to receive preferential capital treatments). We envisage that the issuance of supplementary guidance will be a continuing process. Updates and additions will be made as appropriate to address common interpretative issues, or to incorporate further guidance issued by the Basel Committee. Authorized institutions are also welcome to submit any questions to their usual supervisory contacts at the HKMA should there be areas where they wish to seek further clarification. On-line access to the guidance is available in the webpage for “Banking Legislation, Policies and Standards Implementation / Capital” on the HKMA’s public website (https://www.hkma.gov.hk) and the Supervisory Communication Website (https://www.stet.iclnet.hk). Yours faithfully, Cho-hoi Hui Acting Executive Director (Banking Policy) Encl cc: The Chairperson, The Hong Kong Association of Banks The Chairperson, The DTC Association FSTB (Attn: Mr Justin To) 2 https://www.hkma.gov.hk/media/eng/doc/key-information/guidelines-and-circular/2014/20141231e1.pdf 3 https://www.hkma.gov.hk/media/eng/doc/key-information/guidelines-and-circular/2021/20210603e1a1.pdf 4 https://www.hkma.gov.hk/media/eng/doc/key-information/guidelines-and-circular/2018/20180326e1.pdf

---

Chapter I Introduction Page 1 of 1
Questions and Answers on
Banking (Capital) Rules in respect of
Credit Risk Framework
I. INTRODUCTION

1. These questions and answers (“Q&As”) relate to the Banking (Capital) Rules
   (“the BCR”) that are subsidiary legislation made by the Monetary Authority (“MA”) under section 97C of the Banking Ordinance (Cap. 155) and have been subject to negative vetting by the Legislative Council.
2. These Q&As, written in simple and non-legal language, are designed to assist
   authorized institutions (“AIs”) in understanding the approach being taken by the MA to the implementation of certain sections of the BCR. They are not intended to provide a comprehensive summary of the BCR (or of the aspects of the BCR to which they refer). The answers given are general in their scope and do not take into account the particular circumstances of any individual AI. Certain sections of the BCR contain exceptions or qualifications which, although not covered in these Q&As, may still apply to particular AIs. In the case of any discrepancy between these Q&As and the BCR, the BCR prevail. As such, reading these Q&As is no substitute for reading the BCR themselves.
3. These Q&As should not be regarded as, or be considered a substitute for
   obtaining, legal or other professional advice. AIs should consider obtaining legal and other professional advice before taking any action on any of the matters covered by the Q&As, particularly if they have any doubt as to how any aspect of the BCR might apply to them.
4. The matters covered by these Q&As include information and issues arising
   during the consultation and rule-making processes undertaken in respect of the BCR; the MA’s responses to some common enquiries raised by AIs; ongoing developments in respect of the Basel Framework; and relevant explanatory guidance provided by the Basel Committee on Banking Supervision (“BCBS”) relating to the Basel Framework. The MA will keep under review the practical implementation of the BCR by AIs and changes to the Basel Framework as promulgated by the BCBS from time to time, and will revise these Q&As as appropriate in the circumstances.
5. Unless stated to the contrary (or as otherwise made clear by the context), the
   terms and acronyms used in the Q&As have the same meaning as in the BCR (as in force from 1 January 2025), and the parts, divisions, subdivisions, sections, schedules, tables or formulae referred to are references to those in the BCR.

---

Chapter II ECAI ratings Page 1 of 3
II. Use of ECAI ratings for the purposes of the BCR

1. Nomination of ECAIs under §4C of the BCR
   For the purpose of using ECAI ratings to derive risk-weights for exposures under the STC approach or for securitization exposures under the SEC-ERBA, an AI should nominate the ECAIs it intends to use for each ECAI ratings based portfolio or specific exposures within certain ECAI ratings based portfolios in accordance with the requirements set out in §4C. AIs that use the STC approach and/or SEC-ERBA only for the purposes of output floor calculations are also required to nominate ECAIs in accordance with §4C. The recognition of the HKMA given to a particular ECAI may only cover ECAI ratings with respect to a particular market segment or geographical region (such an ECAI is referred to as Type B ECAI in the BCR). In such case, AIs should not nominate such ECAI for ECAI ratings based portfolio(s) or exposures that are outside the scope of the HKMA’s recognition. The scope of recognition of each Type B ECAI will be published on the HKMA’s website 1 upon commencement of
   Part 3 of the Banking (Capital)(Amendment) Rules 2023 (“BCAR”).
   In selecting ECAIs to be nominated, AIs should pay special attention to the criterion of “reasonable coverage” specified in §4C(3). AIs should ensure that their nominations are made in such a way to ensure that significant exposures within their ECAI ratings based portfolios are covered by at least one of the ECAIs. Note: Part V of the ECAI policy paper, which sets out guidelines applicable to AIs, will be superseded by this Chapter.
2. Treatment of nominations made before commencement date of §4C
   As provided for by §4E, any nomination made before the commencement date of §4C for securitization exposures remains effective on and after that date. No action is required from AIs. AIs are required to make new nomination(s) under §4C for the following ECAI ratings based portfolios— (a) multilateral development bank exposures; (b) unspecified multilateral body exposures; (c) eligible covered bond exposures; (d) qualifying non-bank financial institution (“QNBFI”) exposures; (e) general corporate exposures;

1 https://www.hkma.gov.hk/eng/key-functions/banking/banking-legislation-policies-and-standards￾implementation/capital/

Chapter II ECAI ratings Page 2 of 3
(f) specialized lending exposures.
Existing nominations for securities firm exposures and corporate exposures cannot be carried forward because the classification criteria has been changed. Some exposures that are currently classified as securities firm exposures will be reclassified as corporate exposures and some exposures (e.g. insurance firms) that are currently classified as corporate exposures will be reclassified as QNBFI exposures. AIs are required to reassess whether the ECAIs nominated for, say corporate exposures, have reasonable coverage for exposures to securities firms (see §4C(3)). There is no specific requirement on the format of the written notice to be given to the MA under §4C(5) for the new nominations in respect of the portfolios listed above. AIs only need to declare in the notice that the requirements in §4C(3) or (4), as the case may be, are complied with. Also, AIs are not required to submit the assessment of “reasonable coverage” to the MA. However, the assessment should be properly documented. For administrative convenience, AIs may submit the written notice to their usual supervisory contact at the HKMA before the commencement date of §4C, but the effective date of the written notice and the new nominations must be specified as that commencement date.
3. Use of ECAI ratings issued by nominated ECAIs
(a) An AI should use the ECAI ratings of its nominated ECAIs within the ECAI ratings based portfolios consistently for both risk-weighting and risk management purposes, and, as required by §4D(1), seek the consent of the HKMA on any subsequent changes to the nomination (e.g. the addition or removal of a nominated ECAI in respect of an ECAI ratings based portfolio or any change in the scope of application of the ECAI ratings of a nominated ECAI). The above requirements aim to ensure that AIs use the ECAI ratings of their nominated ECAIs consistently and to avoid any possible cherry-picking of ECAI ratings provided by different ECAIs or any arbitrary change of the use of ECAIs. (b) As a general rule, AIs should use solicited ratings from ECAIs. Unsolicited ratings may be used only when the issuer and the debt obligations of the issuer do not have any solicited rating. (c) Under §54F(1)(a), if an AI intends to use an ECAI rating issued by a nominated ECAI to risk-weight an exposure, the ECAI rating must take into account and reflect the entire amount of the exposure with regard to all payments owed to the AI. For example, if the AI is owed both principal and interest, the ECAI rating must fully take into account and reflect the credit risk associated with repayment of both principal and interest. (d) ECAI ratings for one entity within a group cannot be used to risk-weight other entities within the same group.

Chapter II ECAI ratings Page 3 of 3
4. §54F(1)(b) - Implicit government support
According to §54F(1)(b), AIs may continue to use ECAI ratings with assumptions of implicit government support for public banks owned by governments (i.e. even after 5 years from the commencement of §54F). In determining whether a financial institution is a public bank owned by a government, AIs must first verify whether the institution falls within the definition of “bank” in §2(1) before considering the ownership. If an AI has any difficulty in determining whether a bank to which it has a credit exposure is owned by a government, it may consult the HKMA.
5. §54F(1)(c) - Use of ECAI ratings assigned to banks and bank exposures with
assumptions of implicit government support
After the 5-year transitional period provided for under §54F(2) has expired— (a) as specified in §54F(1)(c)(i), in determining the applicable ECAI rating in accordance with §54E (application of ECAI ratings) for the purposes of §54D (mapping ECAI rating to credit quality grade), the ECAI ratings assigned to banks and their exposures mentioned in §54D and §54E should not include those ratings that incorporate assumptions of implicit government support; (b) according to §54F(1)(c)(ii), if an AI’s bank exposure has only one ECAI issue specific rating but such rating incorporates assumptions of implicit government support, the bank exposure is regarded as falling within (as applicable) paragraph (a)(i)(B) or (b)(i)(B) of the definition of “unrated exposure” in §51(1). Similarly, when determining whether a bank falls within paragraph (a)(ii) or (b)(ii) of that definition, any ECAI rating that incorporates assumptions of implicit government support should not be counted; and (c) in cases where ECAI ratings without assumptions of implicit government support are not available, AIs should not continue to use ECAI ratings with such assumptions by making any adjustments to any of such ECAI ratings in an attempt to remove the impact of implicit government support on the rating, regardless of whether such adjustment is based on information on the rating methodologies provided by the ECAI concerned. The HKMA will keep in view ECAIs’ progress in providing ECAI ratings without the assumptions of implicit government support and revisit the related policy near the time of the expiry of the 5-year transitional period.

---

Chapter III STC approach Page 1 of 65
III. STC Approach

1. Due diligence requirements
   Q1. 1Could more guidance be provided on the due diligence requirements set out in §54C(1)? A1. An AI must, instead of relying on external credit ratings, conduct its own credit assessments (at origination and at least annually thereafter) to ensure that it has an adequate understanding of the risk profile and characteristics of the obligors in respect of its exposures. §54C(1) applies to both rated and unrated exposures. The sophistication of the due diligence should be commensurate with the size and complexity of the AI’s activities. The AI must take reasonable and adequate steps to assess the operating and financial performance levels and trends of each obligor through internal credit analysis and/or other analytics outsourced to a third party, as appropriate for the obligor. The AI must be able to access information about the obligor on a regular basis to complete due diligence analyses. For exposures to an obligor belonging to a group of companies, due diligence should, to the extent possible, be performed at the solo entity level to which the AI has an exposure. In evaluating the repayment capacity of the obligor, the AI is expected to take into account the support of the group and the potential for the obligor to be adversely impacted by problems in the group. Also, the AI should be prepared to demonstrate that its due diligence analyses are appropriate if requested by the HKMA. As part of the supervisory review, the HKMA will consider whether the AI has appropriately performed its due diligence analyses, and, where appropriate, take necessary supervisory measures when there are deficiencies in any of the analyses. The above requirements are similar to those set out in the SPM modules on credit risk management, which should already be the standard credit risk management practices of AIs. Hence, AIs are not expected to require to make substantial change to their credit risk management process in order to comply with §54C(1). Q2. Could immaterial exposures, such as utility deposits, rental deposits, etc. where an AI does not maintain a corporate relationship with the counterparty concerned (e.g. utility provider or the landlord), be exempted from the due diligence requirements? A2. For exposures that are advance payments made, or collateral provided, by an AI to a service provider/supplier or landlord that pertain to the AI’s operating expenses, such as utility deposits and rental deposits paid by the AI for its offices, there is no need to perform due diligence as required under §54C(1) of the BCR. Q3. Is due diligence required on exposures such as interbank balances, time deposits and nostros?

1 Reference: Basel Framework CRE20.4, 20.5 and 20.6.

Chapter III STC approach Page 2 of 65
A3. Due diligence is required on interbank balances, time deposits and nostros as they are credit exposures to banks. Q4. Regarding the due diligence requirements related to ECAI ratings set out in §54C(2), what is the supervisory expectation of the HKMA? Specifically, would AIs be expected to perform a line by line comparison of the underlying risk characteristics considered in the AIs’ internal credit assessments with those in the external rating assessments? A4. The objective of the due diligence requirements is to ensure that the risk-weight assigned to an exposure is reflective of the level of credit risk assessed by an AI itself. AIs are not required to assess whether the external credit rating of an exposure is appropriate or not. The HKMA’s expectation is as follows— (a) 2AIs should have in place effective internal policies, processes, systems and controls to ensure that appropriate risk-weights are assigned to obligors; (b) for capital purposes, an AI is required to develop a mapping scheme that maps the AI’s own credit assessment result (e.g. an internal credit quality grade) to an equivalent external credit rating (based on the credit standing such rating represents). The mapping does not need to be a line by line comparison of the underlying risk characteristics considered in the AI’s internal credit assessment and the external rating assessment. There is also no requirement that the mapping must be based on outcomes of credit models; and (c) the AI should compare the outcome of its own internal credit assessment and the risk-weight determined based on the external credit rating assigned to the exposure concerned (or, if there is no such rating, an applicable external credit rating determined under §54E). If an AI uses the outcomes of credit models for the purpose of the comparison, the AI is reminded that the use of credit models must never result in the application of a lower risk-weight than that determined based on the external credit rating (see §54C(3)). The HKMA will take a proportionate approach to considering the appropriateness of an AI’s mapping scheme and expect AIs to be prepared to justify the reasonableness of their mapping mechanism. Q5. Could an AI leverage the internal credit rating assessment results and mapping scheme for external ratings prepared by the AI’s parent bank for the due diligence process discussed in A4? A5. An AI may adopt the mapping scheme between internal credit ratings and external credit ratings developed by the AI’s parent bank if the AI’s internal credit rating system is the same as that of its parent bank. In general, the AI is expected to conduct the internal credit assessment by itself. The AI may leverage the internal credit ratings assessment results prepared by its parent bank only under very limited circumstances, for example, both the parent bank’s exposure and the AI’s exposure are to the same bond.

2 Reference: Basel Framework CRE20.6.

Chapter III STC approach Page 3 of 65
Q6. 3Should AIs assess climate-related financial risks as part of the due diligence analyses with respect to obligors and credit exposures? A6. Climate-related financial risks can impact AIs’ credit exposures through obligors/counterparties. To the extent that the risk profile/creditworthiness of an obligor or counterparty is affected by climate-related financial risks, AIs are expected to give proper consideration, on a best effort basis given the challenges arising from, for example, methodological and data limitations that cannot be fully resolved currently, to the climate-related financial risks as part of the due diligence (in other words, AIs are expected to integrate climate-related financial risks in their own credit risk assessment and, if applicable, when performing due diligence in respect of risk-weights based on external ratings as required by §54C(2)). The HKMA will adopt a pragmatic approach and apply the proportionality principle in line with the implementation approach of SPM GS-1 “Climate Risk Management”.
2. Unspecified multilateral body exposures
Q7. Would the HKMA provide any guidance with respect to the definition of “unspecified multilateral body” in §2(1), particularly on the concepts of “large sovereign membership”, “mandate similar to the mandates of multilateral development banks” and “considerable number of the owners of which are also the owners of multilateral development banks”? A7. The HKMA does not intend to set any thresholds for “large”, “considerable” or “similar” as different cases may result in different judgements based on the own merits of each case. An AI may consult the HKMA if there is a specific case in which the AI is unable to form a view on whether an obligor to which it has an exposure is an unspecified multilateral body.
3. Eligible covered bonds
Q8. 4What are the conditions that must be met in order for a covered bond to be classified as an eligible covered bond? A8. In order to be eligible for the risk-weighting treatments set out in §59D(1), (2) and (3), the following conditions must be met in respect of the covered bonds concerned— (a) the assets securing the payment obligations attached to the covered bonds (“cover pool”) include any one or more of the following assets:

3 Reference: FAQs inserted to Basel Framework CRE20.4, 20.20 and 20.42. 4 Reference: Basel Framework CRE20.34 to CRE20.37.

Chapter III STC approach Page 4 of 65
(i) claims on, or claims guaranteed by, any one or more of the following entities (as defined in the BCR):
 sovereigns;
 public sector entities;
 multilateral development banks;
 unspecified multilateral bodies;
(ii) claims secured by residential real estate that meet the criteria set out in Basel Framework CRE20.71 (or §65(1) or (3) in the case of loans originated in Hong Kong) and with a loan-to-value ratio of 80% or lower; (iii) claims secured by commercial real estate that meet the criteria set out in Basel Framework CRE20.71 (or §65(1) in the case of loans originated in Hong Kong) and with a loan-to-value ratio of 60% or lower; (iv) claims on, or claims guaranteed by, banks that would qualify for a 30% or lower risk-weight under the STC approach. However, such claims cannot exceed 15% of the aggregate principal amount of the outstanding covered bonds. (b) the nominal value of the assets in the cover pool (“cover assets”) should exceed the aggregate principal amount of the outstanding covered bonds by at least 10% (“10% requirement”). The value of the cover assets for this purpose does not need to be that required by the legislative framework that governs the covered bonds. However, if the legislative framework does not stipulate a 10% requirement, the issuer needs to publicly disclose on a regular basis that the cover pool meets the 10% requirement in practice. In addition to the primary assets 5 , additional collateral may include substitution assets 6 (cash or short-term liquid and secure assets held in substitution of the primary assets to top up the cover pool for management purposes) and derivative contracts entered into for the purposes of hedging the risks arising in the covered bond program concerned; and (c) the investing AI is able to demonstrate to the satisfaction of the MA that— (i) it receives portfolio information at least on:
 the value of the cover pool and outstanding covered bonds;  the geographical distribution and type of the cover assets, and their loan size;

5
“Primary assets” means dominant assets included in a cover pool that determine the nature of the pool. 6 “Substitution assets” means assets in the cover pool that contribute to the coverage requirements (i.e. the extent of overcollateralization), other than the primary assets.

Chapter III STC approach Page 5 of 65
 details in relation to interest rate risk and currency risk;  the maturity structure of the cover assets and the covered bonds; and  the percentage of loans more than 90 days past due; and (ii) the issuer of the covered bonds makes the information referred to in paragraph (c)(i) available to the AI at least semi-annually. The conditions set out in paragraphs (a) and (b) must be satisfied at the inception of the covered bonds and throughout its remaining maturity. Q9. According to A8(a)(ii) and (iii) above, the claims within the cover pool must be regulatory real estate exposures with a loan-to-value ratio not exceeding a certain level. Since the information provided by the covered bond issuer is usually on portfolio level only, investing AIs may not be able to verify whether individual claims meet the requirements in A8(a)(ii) and (iii). Guidance is needed from the HKMA in this respect. A9. The loan-to-value ratio (“LTV ratio”) mentioned in A8(a)(ii) and (iii) is the ratio of a single loan (or in cases where more than one loan is secured by the same property, all those loans in aggregate), rather than the weighted average LTV ratio of the cover pool or the LTV ratio of the cover pool calculated by dividing the total of all outstanding balances of the loans in the cover pool by the total values of the properties securing those loans. AIs may rely on the warranties and representations made by the originator of the loans, and the eligibility criteria that must be met by assets for inclusion in the cover pool, as set out in the relevant covered bond programme documentation. If the eligibility criteria under a covered bond programme do not include a criterion on the maximum LTV ratio of a loan, but there is a maximum LTV ratio under the asset coverage test or for the purpose of calculating the extent of overcollateralisation, AIs may refer to the latter maximum LTV ratio for the purposes of the conditions mentioned in A8(a)(ii) and (iii). Q10. Clarification is sought on whether “loan-to-value ratio” mentioned in A8(a)(ii) and (iii) refers to the original LTV ratio of a loan at bond inception and/or the current LTV ratio at the time of capital computation/reporting. A10. Since A8(a)(ii) and (iii) are on-going requirements, “LTV ratio” refers to the status at bond inception as well as the current status. For the purpose of determining whether the condition in A8(a)(ii) or (iii) is met, “LTV ratio” is the one calculated in the way specified in the relevant law or regulations governing issuance of covered bonds or in the covered bond programme documentation*, and the current LTV ratio is the one disclosed in the most recent disclosure (e.g. investor report) made by the covered bond issuer.

Chapter III STC approach Page 6 of 65

• For example, the LTV ratio could be calculated as the ratio of the loan amount included in the
  cover pool to the current market valuation (CMV) of the property. The loan amount included in the cover pool is not necessarily equal to the current outstanding amount of the loan. The legal requirement or the covered bond programme documentation may provide that if the ratio of current outstanding loan amount to CMV of the property is more than 80%, the loan amount that can be included in the cover pool is 80% of the CMV of the property.

4. Bank exposures
   Risk-weights inferred from short-term ECAI issue specific ratings Q11. Does §59A(2) refer to exposures with an original maturity between 6 and 12 months? A11. The exposure referred to in the chapeau of §59A(2) is— (a) an exposure with an original maturity of not more than 6 months that is not associated with cross-border movement of goods, including any 3 months’ exposure as defined in §51(1); or (b) an exposure with an original maturity between 6 and 12 months, whether associated with cross-border movement of goods or not. Reference exposures referred to in §59A(2)(a)(ii) and (b) are exposures that have a short-term ECAI issue specific rating. There is no need to consider the maturity of a reference exposure for the purpose of §59A(2). Standardized credit risk assessment approach (“SCRA”) Q12. 7 To what extent should climate-related financial risks be taken into consideration when determining whether a bank is eligible for a credit assessment grade of A? A12. Given the challenges arising from, for example, methodological and data limitations that cannot be fully resolved currently, an AI is expected to, on a best effort basis, consider the impact of material climate-related financial risks on the bank’s capacity to meet its financial commitments in a timely manner for the projected life of the AI’s exposures to that bank. Prudent practice by the AI to evaluate the bank’s ability to repay commitments could include incorporating consideration of material climate-related financial risks into the entire credit life cycle, including client due diligence as part of the on-boarding process and ongoing monitoring of clients’ risk profiles. The HKMA will adopt a pragmatic approach and apply the proportionality principle in line with the implementation approach of SPM GS-1 “Climate Risk Management”. Q13. When assessing whether a credit assessment grade of A can be assigned to a bank for an AI’s capital reporting at a quarter end, if it is not feasible to base

7 Reference: FAQ inserted to Basel Framework CRE20.22.

Chapter III STC approach Page 7 of 65
the SCRA assessment on the same quarter-end capital position of the bank (which is not publicly available at the time of the AI’s capital reporting), can the AI base the assessment on last year’s or last quarter’s CET1 capital ratio and leverage ratio of the bank? A13. AIs may use the latest publicly available information to conduct the assessment. Q14. How often should the SCRA assessment of a bank be conducted? A14. Generally speaking, the SCRA assessment consists of two components— (a) assessment of a bank’s capacity to meet its financial commitments in a timely manner for the projected life of the assets or exposures concerned; and (b) the extent the published minimum regulatory requirements are met. Component (a) can be based on the annual credit review conducted by an AI using the latest available published information unless there is any adverse change (e.g. any material adverse changes in the business, financial or economic conditions) that warrants immediate reassessment. Component (b) should be updated when new published information is available. Hence, if a bank publishes new regulatory ratios quarterly, component (b) should be updated quarterly. Q15. If a bank declares in its publicly disclosed financial statements/regulatory reports that it has complied with all minimum regulatory requirements, but the detailed ratios are not disclosed in the statements/reports, is it acceptable to treat the bank as having met the quantitative requirements for grade B? A15. For a commercial bank other than an AI, it must fall within paragraph (b)(ii)(A) or (B) of the definition of “bank” in §2(1) of the BCR, otherwise, the bank must not be classified as “bank” for the purposes of the BCR. If the regulatory framework to which the bank is subject does not include disclosure requirements on the bank’s regulatory ratios (e.g. requirements similar to Template KM1 (Key metrics (at consolidated group level)) set out in DIS20 of the Basel Framework), this would call into question whether such bank falls within the definition of “bank” in §2(1) of the BCR. Q16. Under the SCRA, if the sovereign in which a bank is incorporated has an attributed risk-weight of 100% and the bank is assigned a credit assessment grade of B (i.e. the applicable risk-weight is 75% or 50% under §59C(1), should the risk-weight of 100% be allocated to exposures to the bank to which §59C(3) applies? A16. The applicable risk-weights for unrated exposures to banks are those set out in Table 4A under §59C(1) and §59C(2) of the BCR. The wording of §59C(3) is that “… an unrated exposure to a bank must not be allocated under subsection (1) or (2) a risk-weight that is lower than the attributed risk-weight of the sovereign of the bank’s home jurisdiction if …” (emphasis added). Accordingly, if the attributed risk-weight of the sovereign concerned is 100%, the risk-weight available in

Chapter III STC approach Page 8 of 65
§59C(1) or (2) that is not lower than 100% is 150%. Hence, the grade B bank exposure should be allocated a risk-weight of 150% instead of 75%.
5. Qualifying non-bank financial institution (“QNBFI”) exposures
Definition of “qualifying non-bank financial institution” in §2(1) Q17. It appears that “financial activities” mentioned in paragraph (c) of the definition of QNBFI is not defined in the BCR for the purposes of the STC approach. Would the HKMA define such term in order to enhance clarity? A17. There is no need to define “financial activities” for the purposes of the definition of QNBFI because each jurisdiction will define what types of financial activity require authorization and supervision. If an AI has an exposure to an entity that carries on certain financial activities in a jurisdiction, the AI is expected to verify whether these financial activities are regulated activities under the law of the jurisdiction. If the financial activities are not regulated activities, the entity is not a QNBFI for the purposes of Part 4 of the BCR. Q18. In jurisdictions where banks and non-bank financial institutions (including insurance firms and securities firms) are regulated by a single regulator, can AIs assume that those non-bank financial institutions are QNBFIs for the purposes of the BCR? A18. AIs should not make such assumption because a single regulator may apply different supervisory requirements to different financial sectors based on their specific characteristics and it is possible that some of the supervisory requirements may not be equivalent to those applied to banks. Q19. Guidance is sought on how AIs could determine whether the condition set out in paragraph (c)(i) of the definition of QNBFI is met. For example, whether AIs are expected to keep track of the announcements made by the relevant banking supervisory authorities. A19. We do not expect that banking supervisory authorities will make an announcement on QNBFIs, but the capital requirements issued by the authorities 8 will usually clarify which types of non-bank financial institution in their jurisdictions can be risk￾weighted as banks under their capital adequacy requirements. AIs may refer to the capital adequacy requirements of the jurisdictions concerned and the public registers of regulated entities maintained by regulators (e.g. Monetary Authority of Singapore 9 ) for the purposes of paragraph (c) of the definition of QNBFI. Q20. How to identify whether an entity is a “licensed corporation” under the Securities and Futures Ordinance (“SFO”) or an “authorized insurer” under the Insurance Ordinance?

8 For example, definition of “institution” in Article 4 point 1 of Regulation (EU) No 575/2013, Part VII, Division 3, Sub-division 1, paragraph 7.3.1(e) of the MAS Notice 637, and paragraph 56 of Chapter 4 of the Capital Adequacy Requirements 2023 issued by the Office of the Superintendent of Financial Institutions, Canada). 9 https://eservices.mas.gov.sg/fid

Chapter III STC approach Page 9 of 65
A20. The Securities and Futures Commission and the Insurance Authority have maintained public registers that enable the public to verify whether an entity is a licensed corporation or authorized insurer. Q21. Are entities with temporary licences granted under §117 and §118 of the SFO considered as licensed corporations? A21. Licensed corporation is defined in section 1 of Part 1 of Schedule 1 to the SFO, which means a corporation that is granted a licence under §116 or §117 of the SFO. Hence, entities with temporary licences granted under §117 of the SFO are also “licensed corporations”. Q22. If at least one subsidiary of a holding company (without any licence to carry on financial activities) is a licensed corporation (other than one that is licensed for Type 10 regulated activity) or bank, can the holding company be classified as QNBFI? A22. A holding company that does not meet the requirements specified in paragraph (a), (b) or (c) of the definition of QNBFI in §2(1) is not a QNBFI. Q23. Although currently all entities licensed for Type 10 regulated activity are not licensed for other regulated activities, if in case there is an entity that is licensed for more than one regulated activity (including Type 10 regulated activity), can such entity be classified as QNBFI? A23. It is unlikely that an entity will be licensed for Type 10 regulated activity as well as other types of regulated activity as this will usually give rise to the concern of conflict of interest which is one of the issues that triggered enhancements in the regulation of credit rating agencies after the 2008 global financial crisis. Q24. Will Stored Value Facility (“SVF”) licensees that are under the HKMA’s supervision be regarded as QNBFIs under the STC approach? A24. SVF licensees that are not banks will not be regarded as QNBFIs under the STC approach because they are not subject to regulatory capital and liquidity requirements that are similar to those imposed on AIs. Treatment of unrated QNBFI exposures Q25. Given the SCRA set out in §59B and §59C are designed for exposures to banks, guidance is sought on how the SCRA applies to QNBFIs, in particular, when the regulatory requirements imposed on QNBFIs by relevant regulators may not be easily translated into capital adequacy and countercyclical buffer requirements for banks. A25. The credit assessment under the SCRA consists of two major components:

Chapter III STC approach Page 10 of 65
(a) a credit opinion on the capacity of a QNBFI to meet its financial commitments; and (b) a need to ascertain whether the QNBFI meets or exceeds the minimum regulatory requirements (and buffers if applicable) imposed by its home supervisor. Component (a) is a generic requirement that is applicable to any type of counterparty. For component (b), since financial institutions incorporated outside Hong Kong may be classified as QNBFIs only if they are subject to a regulatory and supervisory framework that is equivalent to that for banks, the minimum regulatory requirements applicable to QNBFIs in a jurisdiction (e.g. minimum capital ratios) and the actual regulatory ratios maintained by individual QNBFIs are expected to be publicly available so that AIs can conduct the assessment in the same manner as in the case of banks. There is no need to translate the ratios into capital adequacy and countercyclical buffer requirements for banks. For example, to qualify for a credit assessment grade of A under the SCRA, the QNBFI concerned must meet or exceed the published minimum regulatory requirements and buffers established by its national supervisor. Hence, an AI only needs to identify what the minimum regulatory requirements and buffers established by the QNBFI’s supervisor are and determine whether the actual ratios maintained by the QNBFI meet or exceed those requirements/buffers. If there is no sufficient information for determining whether the QNBFI is qualified for a credit assessment grade of A or B, the QNBFI must be assigned a credit assessment grade of C. In the case of QNBFIs in Hong Kong that are supervised by the Insurance Authority or the Securities and Futures Commission, a risk-weight of 75% (and 50% for short￾term exposures (essentially exposures with an original maturity of 3 months or less)) would apply (see §60(4)). In other words, there is no need to assign any credit assessment grade to these QNBFIs. If an AI is uncertain about whether a financial institution is eligible for being classified as a QNBFI, the AI must treat the financial institution as a corporate for capital purposes. Q26. If for a QNBFI, the CET1 capital ratio and leverage ratio are not available or not applicable, what benchmarks can be used to determine whether the 30% risk-weight under §59C(2) can be allocated to an exposure to the QNBFI? A26. If the regulatory requirements applicable to a QNBFI do not include a CET1 capital ratio requirement and/or a leverage ratio requirement similar to those applicable to banks, §59C(2) is not applicable to the QNBFI (i.e. the 30% risk-weight is not one of the risk-weights that may be assigned to the QNBFI under the SCRA).

Chapter III STC approach Page 11 of 65
6. General corporate exposures
Risk-weights inferred from short-term issue specific ratings Q27. Clarification is sought on whether the following understanding of the requirements set out in §61(3) and (4) is correct:
(a) If—
(i) no issue specific rating is assigned to an exposure to a corporate that is not a small business, but the corporate has an issuer rating or there is a similar debt security issued by the same corporate with a short-term rating; and (ii) according to Table 5 in §61(1), the issuer rating or the short-term rating of the similar debt security would map to 150% risk-weight, AIs should assign 150% risk-weight to the exposure. (b) Exposures with an original maturity of one year or less and without issue specific rating should not be risk-weighted below 100%, even if the obligor’s issuer rating or a short-term rating of a reference exposure would map to a risk-weight lower than 100%. A27. §61(3) and §61(4) apply to an AI’s general corporate exposures to a corporate only if one of the corporate’s debt obligations (including those that are held by other banks/investors) has a short-term issue specific rating. (a) According to §61(3), the AI should— (i) map the short-term issue specific rating of the reference exposure (i.e. the exposure referred to in §61(3)(a)(ii)) to the corresponding ST ECAI rating mapping table to determine the risk-weight that would be applicable based on that short-term issue specific rating and in accordance with §61(1) (see §61(3)(b)); and (ii) if the risk-weight determined under (i) above is 150%, allocate a risk￾weight of 150% to all general corporate exposures to the same corporate that do not have issue specific ratings (i.e. the exposure referred to in §61(3)(a)(i)), regardless of whether the corporate has an issuer rating or not, and whether any of the corporate’s debt obligations has a long-term issue specific rating or not. (b) If an AI’s general corporate exposure to a corporate (“subject exposure”) has an original maturity of one year or less and the subject exposure does not have an issue specific rating, according to §61(4), the AI should— (i) map the short-term issue specific rating of the reference exposure (i.e. the exposure referred to in §61(4)(a)(ii)) to the corresponding ST ECAI rating mapping table to determine the risk-weight that would be

Chapter III STC approach Page 12 of 65
applicable based on that short-term issue specific rating and in accordance with §61(1) (see §61(4)(b)); (ii) do either of the following—  if the corporate has an issuer rating or another debt obligation of the corporate has a long-term issue specific rating (i.e. one of the situations mentioned in §61(4)(a)(i))—map the issuer rating or long￾term issue specific rating, as the case may be, to the corresponding LT ECAI rating mapping table to determine the risk-weight that would be applicable based on that rating and in accordance with §61(1); or  if the subject exposure is an unrated exposure as defined in §51(1) (i.e. another situation mentioned in §61(4)(a)(i))—determine the risk￾weight in accordance with §61(2); and (iii) if the risk-weight determined under (i) above is 50% or 100% and the risk-weight determined under (ii) above is lower than 100% (e.g. 75%), the AI must allocate 100% risk-weight, instead of 75% risk-weight, to the subject exposure. Definition of “small business” in §51(1) Q28. What is the supervisory expectation on the frequency of the review of a corporate’s annual sales for the purpose of determining whether the corporate is, or remains, eligible for being classified as “small business”? A28. AIs are expected to determine the annual sales of a borrower based on the latest information they obtain in their usual credit risk management process (e.g. annual credit review). The annual sales in a financial year should be ascertained through the audited financial report. However, if audited financial reports are not available or not up￾to-date, other recent information for that year (e.g. management accounts or internal management reports prepared by the borrower, transaction/payment data) that are considered by AIs as reliable may be used. Q29. For retail banking business, term loans and instalment loans may be granted to companies which are not subject to annual credit review. Annual sales information is only obtained in the underwriting process. Clarification is sought on which of the following approaches would be acceptable for the purpose of determining whether a borrower is “small business”:
(a) Option 1: Continue to classify the borrower as “small business” over the loan tenor based on the annual sales information obtained at underwriting (which may not be the latest), unless there is information

Chapter III STC approach Page 13 of 65
that contradicts this classification coming to the AI’s attention obtained in the credit risk management process. (b) Option 2: The borrower is only eligible for being classified as “small business” for the first and second years due to acceptable range of “latest information”, and should then be reclassified as “corporate other than small business” in the absence of more recent figures. (c) Option 3: If AIs foresee that more recent annual sales cannot be obtained regularly throughout the loan tenor, the borrower can be classified as “corporate other than small business” even it is qualified for “small business” at loan origination. A29. For cases where the only facility granted to the borrower is a term loan or instalment loan that is not subject to annual credit review, Option 2 and Option 3 are acceptable as they can be accommodated by the relevant requirements under the BCR (see the definition of “small business” in §51(1) and the treatment under §65D(3) of the BCR). Q30. Clarification is sought on whether a SPV/BVI company/shell company without operating activities, i.e. no annual sales value, can be deemed as small business for risk-weighted amount calculation. A30. It is supposed that the question refers to— (a) an entity that is not part of a corporate group with operating activities and non￾zero annual sales (e.g. the entity is neither the holding company nor a subsidiary of that group); or (b) an entity that is not subject to consolidation under relevant accounting standards. The definition of “small business” in §51(1) of the BCR only captures corporates with annual sales. Hence, an entity without operating activities and therefore without annual sales does not fall within that definition. Q31. The audited financial report of a company as of 31 December, Year 1 is usually published in late February of Year 2. As such, an AI has scheduled the annual credit review of the company in March of each year. Since the CAR return for the position as of 31 March has to be submitted in April, but by that time the credit review of the company is not yet completed and therefore the annual sales in the AI’s relevant system has not yet been updated (whether based on in-house or audited numbers), is there any flexibility for the AI to use the annual sales of Year 1 to determine whether the company is a small business for the CAR return as of 31 March, Year 3 (although Year 2 is literally “the most recent financial year” specified in the definition of “small business” in §51(1))?

Chapter III STC approach Page 14 of 65
A31. As mentioned in A28, the annual sales of a company in a financial year should be ascertained through the audited financial report or, where updated audited financial report is not yet available, other recent and reliable information for that year. In the absence of more recent figures, using the annual sales reported in the financial report for Year 1 to calculate the AI’s capital ratios as of 31 March, Year 3 is still acceptable but using any sales figures in financial years earlier than Year 1 will not be considered as reasonable in general. An AI has the responsibility to obtain the most updated information available from the company for credit assessment. Q32. In paragraph (b) of the definition of “small business” in §51(1), does “consolidated group” refer to the consolidated group on financial reporting basis? A32. Yes, the understanding is correct. Q33. Is the requirement on timeliness of financial report figures mentioned in A31 also the HKMA’s general expectation for asset classes other than “small business”? A33. Under the STC approach, only the definition of “small business” requires AIs to obtain financial information on the most recent financial year of the borrower for capital adequacy purposes. There is no similar requirement for other asset classes.
7. Specialized lending exposures
Features of specialized lending
Q34. When determining whether a transaction falls within paragraph (b)(i) of the definition of “specialized lending” in §51(1), what is the supervisory expectation on how to confirm the primary repayment source in a case where the obligor is a non-SPV? A34. AIs are expected to possess the necessary expertise in determining the primary repayment source for a loan or credit facility as this is one of the major parts of a credit assessment for credit decision making. Hence, AIs can make use of their usual credit assessments for classification purpose in relation to specialized lending. The distinct feature of specialized lending is that the borrower, among other things, has little or no independent capacity to repay the exposure concerned. In other words, the cash flow generated by the asset(s) being financed is the sole or almost exclusive source of repayment of the exposure. Object finance Q35. What types of transactions are generally regarded as object finance? A35. Object finance is a transaction for financing the acquisition of physical asset(s) (e.g. satellites, ships, aircrafts, railcars and fleets), and the primary source of repayment

Chapter III STC approach Page 15 of 65
of the loan concerned is the cash flows generated by the asset(s) being financed. These cash flows could be from rental or lease contracts entered into with one or more third parties. If the borrower has the ability to repay the loan without undue reliance on the cash flows generated by the asset(s) being financed and the lending AI has recourse to the borrower for repayment of the loan, the loan should be treated as a corporate exposure rather than as specialized lending. Q36. If an AI owns a ship or aircraft that is directly leased to a shipping company or airline without a lessee-owned SPV and the AI has recourse to the shipping company or airline for payment, may the AI classify such operating lease as specialized lending - object finance? A36. According to the definition of “specialized lending” in §51(1), one of the characteristics of specialized lending is that the primary source of repayment of the exposure is the income generated by the asset(s) being financed by the lender, either in legal form or economic substance. The AI may classify the operating lease as specialized lending if the lease possesses the above characteristic and the other characteristics specified in the definition (also see Q35 above). Q37. Would an exposure (other than an operating lease) to a non-SPV with indication that the primary repayment is based on the asset(s) being financed be classified as specialized lending – object finance? A37. In order for an exposure to be classified as specialized lending – object finance, the exposure must possess the required characteristics set out in the definitions of “specialized lending” and “object finance” in §51(1), either in legal form or economic substance (also see Q34, Q35 and Q36). Hence, an AI should determine whether the exposure falls within the above definitions based on its assessment of the particularities of the obligor and the exposure. To avoid doubt, the obligor in respect of object finance does not need to be an SPV and the form of financing does not need to be operating lease. AIs are expected to put in place systems and controls to ensure that their classification is reliable and consistent over time. Commodities finance Q38. 10What are the typical features of transactions that are eligible for being classified as commodities finance? A38. An eligible transaction is self-liquidating in nature (i.e. the commodity financed by the loan concerned will be sold to repay the loan) and the structure of the transaction is designed to compensate for the weak credit quality of the borrower because the

10 Reference: Page 6 of “Working paper on the internal ratings-based approach to specialised lending exposures” issued by the BCBS in October 2001.

Chapter III STC approach Page 16 of 65
borrower has no independent capacity to repay the loan. For example, a bank extends a short-term documentary credit to a small independent trading company that acts as an intermediary between producers of commodities and their customers. Each commodity shipment handled by the trading company is financed and secured separately. Credit is extended by the bank upon delivery of the commodity (e.g. corns) to the trading company, who has already contracted for the resale of the commodity. A trust-worthy third party controls the shipment of the commodity, and the bank controls payment by the customer. If the borrower has cash flows generated from other business activities that can be used to repay the loan and the lending AI has recourse to the borrower for the repayment, the loan should be treated as a corporate exposure rather than as commodities finance. Project finance Q39. 11What are the typical features of transactions that are eligible for being classified as project finance? A39. Project finance is usually for financing large, complex and expensive installations, such as power plants, chemical processing plants, mines, transportation infrastructure, and telecommunication infrastructure. This type of financing may take the form of financing of the construction of a new installation, or refinancing of an existing installation (with or without improvements). The lender is usually paid solely or almost exclusively out of the cash flows generated by the contracts for the installation’s output, such as the electricity sold by a power plant, and the borrower is usually an SPV established solely for developing, owning and operating the installation. As such, the repayment of the loan concerned depends primarily on the project’s cash flows and the value of the project’s assets. In contrast, if the loan depends primarily on a well-established and credit-worthy corporate for repayment (e.g. a loan to finance the installation of transatlantic fibre optic cable where the borrower is an established telecommunications firm with diversified revenue stream, i.e. the repayment ability of the borrower does not depend primarily on the performance of the new installation being financed), the loan should be treated as a corporate exposure rather than as specialized lending - project finance. Loans that are for financing the development or acquisition of real estates (e.g. industrial buildings to let or for sale) should be classified as real estate exposures rather than as specialized lending - project finance.

11 Reference: Page 3 of “Working paper on the internal ratings-based approach to specialised lending exposures” issued by the BCBS in October 2001.

Chapter III STC approach Page 17 of 65
Q40. If an AI granted a loan to a corporate to construct a tunnel primarily based on a guarantee given by the holding company of the corporate rather than the toll income to be generated from that tunnel, is the loan a project finance? A40. Such loan would not fall within the definition of “specialized lending” if— (a) the AI is of a view that the tunnel’s cash flows alone may not be sufficient to repay the loan; and (b) the AI’s credit decision is based primarily on the guarantor’s obligations under the guarantee and its capacity to repay the loan in the event of default of the borrower. High quality project finance (a) Climate-related financial risk Q41. 12To what extent does the classification as high quality project finance require consideration of climate-related financial risks? A41. Changes in environmental policy, technological progress or investor sentiment can leave projects exposed to transition risks. At the same time, projects may be exposed to physical risks depending on their type and location. When assessing the ability of a project finance entity to meet its financial commitments in a timely manner, AIs are expected to, on a best effort basis given the challenges arising from, for example, methodological and data limitations that cannot be fully resolved currently— (a) consider the extent to which climate-related financial risks may have an adverse impact on the ability of a project finance entity to meet its financial commitments in a timely manner; and (b) in view of uncertainty of the materiality and timing of the impact of climate￾related financial risks, evaluate on an ongoing basis the impact of climate￾related financial risks as the capacity to evaluate climate-related financial risk data improves. The HKMA will adopt a pragmatic approach and apply the proportionality principle in line with the implementation approach of SPM GS-1 “Climate Risk Management”. Q42. Project finance might be relevant to financing transition to a lower carbon intensive infrastructure or other similar green project. Would the HKMA consider providing supplementary and indicative reference to the relevant sector that might or would be considered eligible for being treated as high quality project finance, with reference in particular to the classification scheme

12 Reference: FAQ inserted into Basel Framework CRE20.52.

Chapter III STC approach Page 18 of 65
introduced with the HKMA “Hong Kong Taxonomy for Sustainable Finance” published in May 2024? A42. For the purposes of capital calculation, “high quality” is in terms of the project entity’s financial capability to meet financial commitments and the degree of protection to creditors provided by the terms and conditions of the project documents and loan documents. Since transition to a lower carbon intensive infrastructure (or the level of “greenness”) does not guarantee financial sustainability of the project concerned, green-related projects, regardless of whether they are related to or fall within the economic activities encompassed in the Hong Kong Taxonomy, should also be assessed against the same criteria as set out in §62(3). (b) §62(3)(d) - revenues Q43. Does §62(3)(d) only apply to cash flows that are availability-based or subject to a rate-of-return regulation or take-or-pay contract and that are used to amortize 100% of the term debt? A43. There is no such requirement, but it is understood that the cash flows are the primary source of repayment of the project finance. Q44. For contracts that contain both availability-based/rate-of-return regulation related/take-or-pay cash flows (“concerned cash flows”) and non-contracted cash flows, is there any materiality qualifier on the percentage of debt that should be repaid by the concerned cash flows in order to fulfil §62(3)(d)? A44. As the types of facility granted under a project finance may be in various forms and there may not be a fixed ratio between revenues and repayment amounts during the whole term of the project finance, it does not seem feasible to set any hard limit. The principle is that the primary source of repayment must be revenues that are availability-based or subject to a rate-of-return regulation or take-or-pay contract. The underlying rationale of §62(3)(d) is that there should be a stable income source that enables the project finance entity to meet its financial commitments in a timely manner, even when there are adverse changes in the economic cycle and business conditions. (c) §62(3)(e) - main counterparty Q45. If a project has multiple off-takers and the revenues from them in aggregate contributes the majority of the project’s total revenue, could §62(3)(e) be regarded as being met on a combined basis? A45. In §62(3)(e), “one main counterparty” means a single legal entity instead of a group of counterparties considered on a collective basis.

Chapter III STC approach Page 19 of 65
Q46. Does “main counterparty” refer to the off-taker that contributes the highest percentage of revenue in a project even though the amount contributed is less than 50%? A46. The underlying yardstick is that the revenue of the project finance entity generated from the main counterparty should enable the project finance entity to meet its financial commitments in a timely manner, even when there are adverse changes in the economic cycle and business conditions. Q47. Clarification is sought on whether the following scenarios can be considered as fulfilling §62(3)(e) (in other words, whether all projects with more than one off￾taker would be ineligible for being classified as high quality project finance, irrespective of their underlying risk profile)— (a) a project with multiple off-takers where the total revenues from these off￾takers account for a majority of the project’s revenue; (b) there are 3 off-takers that contribute 40%, 30% and 30% of the project’s total revenues respectively, and the off-taker with 40% contribution is an investment group. A47. (a) There is no requirement in §62(3) that there must be only one counterparty. Hence, projects with more than one off-taker will not be disqualified automatically. For a project with more than one off-taker, §62(3)(e) reflects the requirements in CRE20.52(4) of the Basel Framework that— (i) the project finance entity’s revenue should depend on one main counterparty (in other words, the other off-takers are not the major source of revenue of the project); and (ii) this main counterparty should be a central government, PSE or a corporate, that has a risk-weight of 80% or lower. (b) In such scenario, AIs should consider whether the project finance entity is still able to meet its financial commitments in a timely manner if any one or more of the off-takers default. Hence, the decision is not just based on the percentage of revenue contributed by each off-taker. (d) §62(3)(g) - protection against losses resulting from termination of project Q48. A project can be terminated due to a variety of reasons including force majeure, inefficiency of the project finance entity to generate revenue, etc. Also, the pay-out of the counterparties or off-takers is based on the reason of project termination and they usually do not provide blanket security that covers all scenarios of project termination. Would more details be provided on the project termination scenarios being considered for §62(3)(g)? A48. §62(3)(g) is equivalent to CRE20.52(6) of the Basel Framework, which is one of the typical features of public-private partnership availability-based projects that,

Chapter III STC approach Page 20 of 65
according to ECAIs’ default studies, exhibit lower risk profile. Termination in this case refers to a termination during the operational phase triggered by an event (other than borrower default) defined in the project documentation (“relevant termination event”). Since no particular termination event is specified in CRE20.52(6), it is reasonable to conclude that the contractual provisions of a high quality project should provide for termination compensation that cover all relevant termination events. Q49. Clarification is sought on the application of §62(3)(g) to projects like the following:
(a) a power project where all power can only be sold to a government-owned company, and it is not possible to find a replacement off-taker; and (b) a power project where all power is sold to a single private company. In case that company terminates the off-take contract with the project company, the project company can still find an alternative buyer for its output or sell the power in the merchant electricity market. There is usually a 6-month grace period (which can be extended further with lenders’ consent) for the project company to find an alternative buyer. During the grace period, lenders can rely on the sales in the merchant electricity market or the Debt Service Reserve Account for payment of the loan instalments due. A49. §62(3)(g) requires that there must be a person that will protect the creditors from losses due to a termination of the project. Such person does not need to be the main counterparty. It can also be a counterparty which similarly complies with the eligibility criteria for the main counterparty. In (a), the requirement of §62(3)(g) is met if the government (provided that its risk￾weight is 50% or lower) or the government-owned company (provided that it is a PSE or a corporate with a risk-weight of 80% or lower) will protect the creditors from the losses resulting from a termination of the power project. In (b), the termination of the off-take contract with the private company is not necessarily equal to the termination of the project since the project will continue when a replacement off-taker is found. If it is possible that the termination of off￾take contract will lead to a termination of the project, the key consideration is whether contractually the private company or another high quality party will protect the creditors from losses due to the termination of the project. Since a replacement off-taker is not guaranteed, this will not be regarded as termination protection for the purpose of §62(3)(g). Q50. It appears that §62(3)(g) contradicts the definition of “specialized lending” in §51(1) that requires the primary source of repayment to be the income generated from the project. If the main counterparty undertakes to recuperate all the losses incurred by the creditors due to project termination, the underlying credit risk is changed from the project’s revenue generating

Chapter III STC approach Page 21 of 65
capability to the credit worthiness of the main counterparty. Clarification is sought in this regard. A50. Paragraph (b)(i) of the definition of “specialized lending” in relation to the primary source of repayment is a feature that helps AIs to determine whether an exposure should be classified as specialized lending for risk-weighting purposes. The risk￾weights for specialized lending are intended to reflect a series of factors including the protection available to creditors when the asset/assets fail to generate sufficient cash flows for loan repayments. Termination compensation can be considered as a form of risk mitigant that partly contributes to the lower risk-weight of 80%. (e) §62(3)(h) - assets pledged Q51. In §62(3)(h)— (a) does “all assets and contracts necessary” refer to all key assets indispensable for operating the project? (b) does “have been pledged (or otherwise provided as security)” refer to having appropriate restrictions in place to ensure that the assets and contracts cannot be pledged to other creditors? A51. (a) The full requirement set out in §62(3)(h) is “all assets and contracts necessary to operate the project” (emphasis added). The word “necessary” has its usual dictionary meaning. AIs should conduct the assessment based on this requirement. Also see Q52. (b) The wording “have been pledged (or otherwise provided as security)” does not mean a negative pledge clause or other measure with a similar effect. The requirement essentially means that the lending parties (including the AI concerned) must have a legally enforceable security interest in the assets and contracts concerned, created through, for example, first legal charge or assignment. Q52. In certain projects, providing the project assets to the lenders as collateral is not prohibited by law, but it is not practically possible to do so due to the nature of the assets (e.g. road, prisons, hospitals, etc.). Would these projects be regarded as failing to meet the condition set out in §62(3)(h)? A52. Assets in this context do not need to be physical assets (e.g. the road in a toll road project). If it is a standard practice in a particular type of project that the physical assets will not be pledged to the creditors, the HKMA would not consider it unreasonable to regard the requirement in §62(3)(h) as being met if the assets and contracts that have been pledged to the creditors are already sufficient in providing the creditors with step-in rights and the ability to protect key contracts/assets and enforce the terms of key project agreements (e.g. the concession agreement).

Chapter III STC approach Page 22 of 65
8. Retail exposures
Regulatory retail exposure
Q53. Should “other real estate exposure” secured by residential properties be excluded when calculating the aggregate exposure for the purpose of determining whether §64(2)(d) and (e) are met? A53. In calculating the aggregate exposure, all forms of exposures (including real estate exposures secured by non-residential properties but excluding real estate exposures secured by residential properties) must be aggregated. Real estate exposures secured by residential properties mentioned in §64(3)(a) mean exposures falling with the definition of “real estate exposure” in §51(1) where the immovable property concerned is a residential property. In other words, all real estate exposures secured by residential properties (regardless of whether they are regulatory real estate exposures) are excluded from the calculation of the aggregate exposure. Transactor Q54. Should “over limit” commitment from credit cards within tolerance level be regarded as overdraft for the purposes of the definition of “transactor” in §2(1)? A54. Since the “over limit” commitment is an integral part of a credit card facility granted to a customer, it is not regarded as an overdraft for the purpose of determining whether the customer is a transactor. Q55. For the purposes of the definition of “transactor” in §2(1), are redrawable loans and loans on cards (e.g. balance transfer, instalment loan on card) revolving facilities? A55. (a) Instalment loans where the repaid principals can be redrawn are treated as non￾revolving loans instead of revolving facilities. (b) Loans on cards will be regarded as outstanding balances under revolving facilities if all the following conditions are met— (i) the loan is granted under the borrower’s credit card limit; (ii) if the loan is an instalment loan, each monthly instalment amount is billed to the borrower’s credit card account monthly. An instalment amount is treated in the same way as a purchase transaction and included as part of the credit card monthly statement balance; (iii) if the loan is not an instalment loan, the whole loan amount is billed to the borrower’s credit card account and treated in the same way as a

Chapter III STC approach Page 23 of 65
purchase transaction. Any outstanding balance of the loan will be included as part of the credit card monthly statement balance; (iv) in the cases mentioned in subparagraphs (ii) and (iii) above, the borrower has an option to pay only a minimum payment due; and (v) such minimum payment due is calculated by using the same method as that for retail purchases. Q56. As regards the condition set out in A55(b)(v), clarification is sought on the meaning of “same method as that for retail purchases” and how that paragraph should be applied for the following products— (a) cash advance instalment plan under existing credit card limit; (b) merchant instalment plan or spending instalment plan under existing credit card limit; (c) additional line of credit granted under the total credit card limit; and (d) balance transfer or balance consolidation. A56. The “Minimum Payment” set out in the monthly credit card statement is usually calculated as the higher of— (a) a fixed amount (e.g. HKD230); and (b) the sum of— (i) a fixed percentage (e.g. 1%) of the total outstanding balance of the transactions (other than transactions that are subject to subparagraph (ii) below) shown in the statement; and (ii) the full amounts of other outstanding payments such as fees and charges. In A55(b)(v), “same method” means that—  in the case of cash advance instalment plan or merchant or spending instalment plan, or balance transfer or balance consolidation that is an instalment loan, the whole monthly instalment is included as part of the total outstanding balance used in the calculation referred to in paragraph (b)(i) above; and  in the case of additional line of credit, or balance transfer or balance consolidation that is not an instalment loan, the whole outstanding balance under such facility is included as part of the total outstanding balance used in the calculation referred to in paragraph (b)(i) above. (See Annex 1 for an illustrative example) If the whole monthly instalment of an instalment loan, or the whole outstanding balance of a line of credit or balance transfer/consolidation loan, is subject to

Chapter III STC approach Page 24 of 65
paragraph (b)(ii) rather than paragraph (b)(i), the line of credit or loan must not be regarded as revolving facilities. Q57. According to paragraph (a)(i) of the definition of “transactor” in §2(1), an obligor should have repaid in full the balance due under the facility concerned at each scheduled repayment date for the previous 12 months. If, under a credit card facility granted by an AI to a customer, there is one late payment record during the previous 12 months purely due to unpaid annual fee, and such annual fee (and the resultant late charge, if any) was subsequently waived by the AI, may the AI treat the obligor as “transactor”? A57. If— (a) the customer only refused to pay the annual fee (and any resultant late charge) but repaid the balances of other billed transactions on or before the due date; (b) the AI subsequently waived the annual fee and any resultant late charge; and (c) the AI is not aware of any evidence suggesting an increase in credit risk of the customer, it is acceptable to regard the customer as having repaid in full the balance due for the purpose of identifying transactors under the BCR. Similarly, if the annual fee was waived by the AI on or before the due date and the customer only paid an amount that was equal to the difference between the statement balance and the annual fee, the AI may regard the customer as having repaid in full the balance due. The same principle also applies to other bank charges or fees payable by customers for other types of revolving facility.
9. Unhedged credit exposure
Non-revolving loans versus revolving facilities Q58. Is the currency mismatch multiplier applicable to exposures under revolving facilities (e.g. revolving loans, overdrafts, loan on cards) given that such exposures may have features similar to pre-designated principal and interest repayment schedules? A58. Subject to any further guidance or clarifications from the Basel Committee, all revolving facilities are excluded. Please refer to the definition of “revolving” in §2(1), and Q59, Q60 and Q61 below. Q59. For the purposes of §51A(1) (meaning of unhedged credit exposure), are redrawable loans and loans on cards revolving facilities?

Chapter III STC approach Page 25 of 65
A59. Please see Q55 and Q56.
Q60. If a borrower is allowed to draw down a fixed rate loan under an uncommitted revolving facility, and the borrower can choose when to repay and the amount to be repaid, also, any unpaid portion of the loan at maturity will be charged at overnight interest rate as an overdraft and the AI concerned has the right to recall the loan at any time, is such loan part of a revolving facility? A60. Any amount drawn under a revolving facility will not be regarded as a separate non￾revolving loan for the purpose of the definition of “unhedged credit exposure”, unless the drawn amount is in economic substance a term loan that meets the description of paragraph (b) of the definition of “unhedged credit exposure” in §51A(1). Q61. In the case of retail exposures, are there any intended differences between “instalment loans” and “non-revolving loans with pre-specified schedules of repayment”? A61. Non-revolving loans, according to paragraph (b) of the definition of “unhedged credit exposure” in §51A(1), are loans— (i) with a pre-specified schedule of repayments of principal and interest (which does not need to be regular monthly repayments) and predefined repayment amounts such that the whole loan amount will be repaid within a fixed repayment period, including such a loan that allows re￾borrowing of the repaid principal (whether there is an extension of the fixed repayment period or not); or (ii) that will be repaid within a fixed repayment period with a bullet payment. Since instalment loan also falls within paragraph (b) of the definition of “unhedged credit exposure”, there is no reference to “instalment loan” in §51A. “Instalment loan” is typically considered as a loan the whole loan amount of which must be repaid in regular monthly amounts (each payment is made up of both principal and interest) within a fixed repayment period. The most common example is mortgage loans that finance acquisition of real estates. Q62. Clarification is sought on whether the following margin lending facility is subject to the currency mismatch multiplier:
(a) the margin lending facility is fully collateralized and mark-to-market is conducted on a regular basis; (b) the customer can draw multiple term loans under the facility with tenors between 1 to 12 months subject to the total outstanding amounts not exceeding the approved total credit limit; and

Chapter III STC approach Page 26 of 65
(c) the term loans are allowed to be rolled-over upon maturity or can be repaid fully or partially as per the customer’s instruction at any point of time. A62. Subject to any further guidance or clarifications from the Basel Committee, a margin lending facility is not subject to the currency mismatch multiplier if the facility is a revolving facility. The loan drawn under the margin lending facility is not subject to the currency mismatch multiplier if the loan does not fall within paragraph (b) of the definition of “unhedged credit exposure” in §51A(1). Q63. Would private bank and wealth-based lending be exempted from the currency mismatch multiplier? A63. There is no presumption that private bank and wealth-based/Lombard lending refer to revolving loans/facilities only. Hence, any private bank or wealth￾based/Lombard lending that falls within paragraph (b) of the definition of “unhedged credit exposure” in §51A(1) should be subject to the currency mismatch multiplier. Q64. Are facilities with repayment methods different from traditional instalment loans considered in-scope for the currency mismatch multiplier? For example— (a) a secured facility with interest-only repayment, i.e. the obligor repays interest on a monthly basis but is not allowed to redraw the approved limit of the facility; and (b) a secured facility with yearly repayment of principal and monthly repayment of interest. A64. According to paragraph (b) of the definition of “unhedged credit exposure” in §51A(1)— (a) subject to any guidance that may be issued by the Basel Committee from time to time, the facility with interest-only repayment is not subject to the multiplier unless there is a fixed repayment date on which the obligor is required to repay the whole outstanding loan balance (i.e. a bullet payment); and (b) the facility with yearly repayment of principal and monthly repayment of interest is subject to the multiplier. Source of income Q65. Does “the currency of the obligor’s source of income” in paragraph (c)(i) of the definition of “unhedged credit exposure” in §51A(1) refer to the currency at the time of reporting? A65. In the case of credit facilities that are not subject to periodic credit reviews but have regular scheduled repayments of principal and interest (such as regulatory

Chapter III STC approach Page 27 of 65
residential real estate exposures, including refinancing loans and top-up loans), an AI may continue to rely on the information obtained at loan origination to determine the currency of the obligor’s source of income unless the AI is aware of a change in the currency of the obligor’s source of income. In the case of other types of credit facility that are subject to periodic credit reviews based on updated information provided by obligors, AIs are expected to verify whether there is any change in the currency of an obligor’s source of income as part of a credit review. Also see Q80 for more information. Q66. If an AI does not have sufficient information to verify the currency of an obligor’s income, is the AI allowed to regard the currency of the place of residence of the obligor as the currency of the obligor’s income? A66. As provided for under §51A(4), if an AI does not have readily available or sufficient information to assess whether there is a currency mismatch between the exposure and the obligor’s source of income, either one of the following treatments will apply— (a) for loans granted by the AI before 1 January 2025 (“cut-off date”), the AI may assume that there is no currency mismatch unless the AI has recent information that suggests otherwise; or (b) for loans granted by the AI on or after the cut-off date, the loan must be regarded as having currency mismatch if— (i) the place of residence of the borrower is jurisdiction A; and (ii) the loan is denominated in a currency other than the currency of jurisdiction A. Q67. Clarification is sought on whether the following situations are considered as “does not have readily available or sufficient information” for the purposes of §51A(4):
(a) There is no sufficient information on the currency of the obligor's source of income, and the obligor has not declared that all the sources of income are in HKD. (b) The currency of the obligor's source of income is not captured by the application/system under current design. (c) In the current loan application process, the currency of the obligor's source of income is not required. A67. In situation (a), an AI may rely on §51A(4)(a) or (b), as applicable, if the AI has already made reasonable efforts to obtain the information from the obligor but the information is still unavailable.

Chapter III STC approach Page 28 of 65
In situation (b), AIs have the responsibility to enhance their systems to enable capturing information necessary for compliance with the requirements set out in the BCR. In situation (c), similar to situation (b), AIs are expected to adjust the loan application process to require information on the currency of the obligor’s source of income. In both situations (b) and (c), AIs may, before completion of the enhancement in the system and/or application process for capturing information on borrowers’ currencies of incomes, rely on §51A(4)(a) or (b), as applicable, for the purpose of determining whether there is currency mismatch. AIs are expected to complete the enhancement within a reasonable timeframe (usually no later than 12 months from 1 January 2025). For exposures granted before that date, the AIs may continue to rely on §51A(4)(a) after the enhancement unless there is new information available to the AIs (see Q80). Q68. If an obligor has multiple sources of income in different currencies, which of the currencies should be regarded as the currency of the obligor’s source of income? A68. As required by §51A(2), if an obligor has sources of income in two or more currencies (say HKD, CNY and USD), an AI must regard an exposure to that obligor as falling within paragraph (c) of the definition of “unhedged credit exposure” (regardless of the currency in which the exposure is denominated), i.e. must assume that there is a mismatch between the currency of the exposure and the currency of the obligor’s source of income. Determination of applicability of currency mismatch multiplier (a) Pegged currencies Q69. Can pegged currencies, such as HKD and USD, be considered as the same currency for the purposes of the currency mismatch multiplier? A69. Pegged currencies must be considered as different currencies. (b) Initial assessment Q70. For customers with incomes in multiple currencies, if an AI uses the dominant currency (e.g. the currency that accounts for 90% of the customer’s total income) to compare with its exposure to the customer instead of comparing each currency against the exposure, is this approach acceptable? A70. When considering whether there is a natural hedge for the exposure, the currencies of other sources of income may need to be considered. For example, if the monthly instalment of a personal loan is HKD50,000 and the borrower’s monthly income consists of HKD60,000 and CNY100,000, there is currency mismatch based on the

Chapter III STC approach Page 29 of 65
dominant source of income but there is a natural hedge for the HKD instalment and therefore the currency mismatch multiplier is not applicable to the loan. In practice, the AI may take a shortcut in this case by comparing the HKD instalment amount with the amount of the HKD income of the borrower. This will not be considered as a contravention of the requirement of the BCR as the BCR are not meant to dictate, in practice, what steps, or the order of the steps, taken by AIs to determine whether the currency mismatch multiplier is applicable. Q71. For cases where the credit assessment of an exposure does not depend on the income of the borrower, how should an AI assess whether the currency mismatch multiplier is applicable to the exposure? A71. If an AI does not have sufficient information to verify the currency of a borrower’s income (including cases where the credit assessment does not require income proof from the borrower), the AI should apply the treatment set out in §51A(4)(a) or (b), as applicable, to the exposure concerned. To avoid doubt, collateral provided by and assets belonging to the borrower cannot be regarded as income for the purpose of determining whether there is currency mismatch, but they can be considered as natural hedges (please see Q86 to Q89 for details). For wealth-based lending, an AI (assuming its relevant underwriting practice is prudent) may regard any acceptable financial assets (see Q87) denominated in the same currency as the exposure concerned as a natural hedge to determine whether the currency mismatch multiplier should be applied. In the case of collateralized lending (including Lombard lending) where information on the borrower’s income is unavailable and the collateral is not recognized collateral, the collateral may be regarded as a natural hedge only if the collateral is an acceptable financial asset (see Q87) denominated in the same currency as the exposure concerned. Q72. If, for innovative retail products (e.g. the New Personal-Lending Portfolio), customers are not required to provide any income proof and AIs will estimate customers’ incomes by using models, how should the AIs determine whether the currency mismatch multiplier is applicable? A72. Subject to any guidance that may be issued by the Basel Committee from time to time, for such type of exposures that are granted based on alternative data and/or income estimation models instead of income proof, the currency of the obligor’s income would be regarded as HKD if the data collected and/or the model-estimated income is in HKD. Similarly, if the data collected and/or the model-estimated income is in a currency other than HKD, the currency of the obligor’s income would be regarded as that other currency. Q73. If there is a CNY mortgage loan with joint borrowers where the salaries of the borrowers are in HKD but one of the borrowers receives CNY rental income and has been granted a CNY personal instalment loan, please clarify how an obligor-level
assessment of the applicability of the currency mismatch multiplier to the two loans should be conducted.

Chapter III STC approach Page 30 of 65
A73. In such case—
(a) the CNY rental income should be applied to the CNY personal instalment loan first; and (b) the CNY natural hedge available to the CNY mortgage loan is the CNY rental income less 90% of the instalment amount of the CNY personal instalment loan. Q74. In practice, an obligor may have two or more facilities in a currency set but the obligor’s income in the same currency set is only sufficient to cover partially the repayments of these facilities. In considering whether the facilities are hedged or unhedged, is it correct that the assessment should be based on facility level, in other words, whether the facilities are regarded as hedged if 90% or more of them are covered by income in the same currency set as that of the facilities? A74. When assessing whether the exposures are sufficiently hedged, the total of the instalment payment amounts of the exposures that are denominated in a particular currency should be compared to the amount of the borrower’s income that is also denominated in that currency. Each currency should be assessed separately. Q75. If there are 3 income based facilities and 2 asset based facilities, should the 3 income based facilities be considered as one group and the 2 asset based facilities as another group, and the currency mismatch and the total of the instalment payment amounts mentioned in the answer to Q74 be determined separately for each group? A75. (a) When assessing whether there is currency mismatch, only the currencies of the borrower’s source of income and the loans concerned are considered (also see Q66 and Q68 above for cases where there is insufficient or no readily available information to assess the currency of the borrower’s income, or where the borrower’s income is in multiple currencies). Grouping at this stage may not be necessary. For loans with currency mismatch (i.e. being deemed as such under §51A(2) or when the currency of the borrower’s source of income is different from the currency of the loan), an AI needs to further assess whether the loans are unhedged. For such purpose, the approach of assessing the 3 income based facilities as a group separately from the group of the 2 asset based facilities is acceptable, provided that the same financial hedge or natural hedge is not counted twice (e.g. if the borrower’s income is already counted in the assessment for the group of income based facilities, the same income must not be counted again in the assessment of the group of the asset based facilities). However, if any one of the facilities is a collateralized loan where the AI is not entitled to apply freely the proceeds from the disposal of the collateral concerned to settle other amounts owed to the AI by the borrower, the AI should treat the collateralized loan and the collateral as a group separated from other collateralized facilities. In other words, the collateral cannot be treated as the natural hedge for other collateralized facilities and unsecured facilities

Chapter III STC approach Page 31 of 65
granted to the borrower.
(b) Based on the grouping approach discussed above, in general— (i) all instalment amounts and bullet payments of the loans within the same group in a currency should be aggregated; (ii) similarly, all financial hedges and natural hedges for the loans referred to in subparagraph (i) in the same currency should be aggregated and compared with the aggregate amount obtained under that subparagraph. Q76. Clarification is sought on whether the following approach to maximizing natural hedge coverage is acceptable. Borrower’s monthly income: HKD 100, CNY 5 Acceptable financial asset A (free of encumbrance): HKD 500 Loan 1 Loan 2 Loan 3 Loan 4 Nature Unsecured Wealth based Collateralized Collateralized Tenor (in months) 6 12 Currency HKD CNY Outstanding amount (at origination) 500 1500 500 93 Recognized collateral B in HKD (after haircut)

• • 200
    Ineligible collateral C in
    HKD (acceptable financial asset and after haircut)
• • 400
    Natural hedge allocation
    HKD income allocated - 1200 - -
    CNY income allocated - - - 60
    Asset A and collateral C allocated 300 150 450 - Total 300 1350 450 60 Currency mismatch multiplier applicable (at origination)? Yes No No Yes A76. Grouping approach like the one described is acceptable as long as the grouping and allocation reflect correctly the reality (including differences in loan maturities/repayment structures and the contractual or legal rights of the AI in respect of the loans and natural hedges). For example, if, under the loan/collateral agreements for Loans 3 and 4, the AI is entitled to use the proceeds from the disposal of collateral C to offset the borrower’s other amounts owed to the AI (e.g. Loans 1 and 2), the AI may allocate part of the value of collateral C to cover Loan 1 or 2. Q77. Does “the total of the instalment payment amounts of the exposures” mentioned in the answer to Q74 exclude instalment payment amounts that are—

Chapter III STC approach Page 32 of 65
(a) financially hedged; or
(b) naturally hedged by other means?
A77. Please see paragraph (b) of the answer to Q75.
(c) Reassessment of existing exposures
Q78. Clarification is sought on whether AIs are expected to assess the applicability of the currency mismatch multiplier on facility level instead of on obligor level. For example, if a customer applies for a CNY mortgage then a HKD personal loan, should the subsequent income assessment (e.g. with updated income proof) only be used for that personal loan application (i.e. there should be no retrospective impact on the CNY mortgage loan as solid income proof was provided before)? A78. In the example cited, since the new personal loan is denominated in HKD, the HKMA considers that the status of the CNY mortgage should not be affected unless the AI is aware that (e.g. from the updated income proof obtained) the borrower’s CNY income has decreased to a level less than 90% of the monthly instalment amount of the CNY mortgage (i.e. the AI is now exposed to FX risk and therefore the currency mismatch multiplier should be applied to the CNY mortgage). Q79. If an existing HKD loan granted to an obligor (whose place of residence is Hong Kong) is classified as an unhedged credit exposure based on the income information (which indicates that the income is in CNY) obtained at loan origination and subsequently a new HKD loan is granted to the same obligor but there is no sufficient information on the currency of the obligor’s income, could an AI regard the new HKD loan or both loans as without currency mismatch based on §51A(4)(b) (i.e. the place of residence of the obligor)? A79. Since income information was obtained at the origination of the existing HKD loan and the AI does not have any new information at the time of the origination of the new HKD loan that supports a change in the currency of the borrower’s income, the AI should continue to regard CNY as the currency of the borrower’s income for the purpose of determining whether there is currency mismatch in respect of the new HKD loan. Q80. For an existing facility granted by an AI to an obligor, is it acceptable if the AI reassesses the applicability of the currency mismatch multiplier to the facility only when a credit assessment is conducted for consideration of granting a new facility to the obligor or increasing the credit limit of the existing facility, or when periodic credit review of the existing facility is performed? A80. The HKMA does not expect active monitoring or updating of borrowers’ income/net worth information outside the usual credit review cycle. Hence, “a change in the currency of the obligor’s source of income” mentioned in A65 and “recent information” mentioned in A66(a) would usually be new information obtained

Chapter III STC approach Page 33 of 65
during a recent credit review or a recent application for a new loan or limit increase. The same principle applies to the determination of the amount of natural hedge. Q81. As regards the answer to Q80, AIs may not necessarily require existing customers to provide new income proof for each new application (e.g. in the case of cross-selling) or for periodic credit reviews, would this situation be regarded as “does not have readily available or sufficient information” for the purposes of §51A(4)? A81. For a loan or facility granted on or after 1 January 2025, if an AI considers that the income proof obtained previously is still fit for the purpose of credit decision making, the AI may continue to use that income proof to determine whether the currency mismatch multiplier is applicable to the loan or facility. Q82. If reassessment at obligor level cannot be achieved due to operational difficulties such as having different originating systems with no linkage or different methods of income calculation for different retail lending facilities, is the AI concerned allowed to conduct the reassessment at account level, i.e. only when there is a credit review of the loan concerned? A82. The HKMA’s expectation is as follows:
(a) If an existing non-revolving loan of a borrower is currently classified as an unhedged credit exposure, no reassessment of the classification is required during the life of the loan. (b) If an existing non-revolving loan of a borrower is currently not classified as unhedged credit exposure, the high level principle is that any reliable information on the borrower’s income obtained from a new facility application must trigger reassessment of the classification of the existing non-revolving loan. However, exception to this principle is acceptable if the new facility is a revolving facility of a relatively small amount (i.e. a regulatory retail exposure). Assessment at account level may allow the same income being counted more than once (thus overstating the amount of natural hedge available). This may defeat the purpose of the currency mismatch multiplier requirement. If an AI’s information management system is not able to support assessment and reassessment at obligor level, the AI is expected to explore solutions in overcoming the operational challenges and enhancing the AI’s ability of having a holistic view of its exposures to, and the credit quality of, a single obligor. Q83. Credit facilities have been granted to a borrower with details as set out below:
Facility 1 Facility 2
Borrower’s HKD income at loan origination
(in HKD’000)
Unknown 90
Monthly instalment amount (in HKD’000) 100 100 Date of origination 1 Jan 2022 1 Mar 2025

Chapter III STC approach Page 34 of 65
Is the understanding correct that Facility 1 is not subject to the multiplier because §51A(4)(a) is applicable, and the multiplier is applicable to Facility 2 because the HKD income (after excluding 90% of the instalment amount of Facility 1) is insufficient to hedge the instalment amount of Facility 2? A83. When granting Facility 2, the currencies and amounts of the borrower’s income became known to the AI. Hence, the status of Facility 1 must be reassessed at the time of the origination of Facility 2 based on the new information obtained. Please also see Q65 and Q80. Q84. Clarification is sought on whether the conclusions (see the table below) regarding the applicability of the currency mismatch multiplier are correct, and whether the alternative treatment described below is acceptable. Borrower’s monthly income in HKD: 90,000 Facility A Facility B Facility C Scenario 1 Monthly instalment amount (in HKD’000) 70 50 30 Date of origination 1 Jan 202X 1 May 202X 1 Dec 202X Currency mismatch multiplier applicable on and from:
1 Jan 202X? No - -
1 May 202X? No Yes -
1 Dec 202X? No Yes No
Scenario 2
Monthly instalment amount
(in HKD’000)
70 10 30
Date of origination 1 Jan 202X 1 May 202X 1 Dec 202X Currency mismatch multiplier applicable on and from:
1 Jan 202X? No - -
1 May 202X? No No -
1 Dec 202X? No Yes No
Rationale
(a) In Scenario 1, at the time when Facility C is granted, since the income is sufficient to hedge 90% of the instalment amounts of Facilities A and C, the AI could apply the currency mismatch multiplier to Facility B only instead of both Facilities B and C. (b) In Scenario 2, at the time when Facility C is granted, since the income is sufficient to hedge 90% of the instalment amounts of Facilities A and C, the AI could apply the currency mismatch multiplier to Facility B instead of Facility C.

Chapter III STC approach Page 35 of 65
Alternative treatment
In both Scenarios 1 and 2, at the time when Facility C is granted, since the income is sufficient to hedge more than 90% of the instalment amounts of Facilities B and C, the AI could, at its discretion, apply the currency mismatch multiplier to Facility A only. A84. It is supposed that HKD is not the currency of the obligor’s source of income, otherwise none of the HKD facilities in the 2 scenarios should be classified as an unhedged credit exposure (see paragraph (c) of the definition of “unhedged credit exposure” in §51A(1)). In general, if the aggregate amount of hedges for a set of credit facilities is insufficient to cover 90% of the aggregate of the monthly instalment amounts of those facilities, but is sufficient to cover 90% of the aggregate of the monthly instalment amounts of some of the facilities (“covered facilities”), then the AI is not required to apply the currency mismatch multiplier to the covered facilities. Facilities other than the covered facilities must be subject to the currency mismatch multiplier. Also see Example 2 in Q89. Hence, the proposed treatments set out in the table and the alternative treatment are acceptable. Treatment of recognized collateral Q85. Does the currency mismatch multiplier apply to exposures that are secured by recognized collateral? A85. If the collateral or asset pledged is recognized collateral as defined in the BCR, the simple approach and comprehensive approach under the STC approach for handling recognized collateral already presume that the collateral or asset would be sold to repay the outstanding exposure in the event of default. More specifically— (a) under the simple approach, the currency mismatch multiplier does not apply to the risk-weight assigned to the credit protection covered portion of the exposure (i.e. the risk-weight applicable to the recognized collateral) but applies to the risk-weight assigned to the credit protection uncovered portion (i.e. the risk-weight applicable to the borrower); and (b) under the comprehensive approach, if there is any net credit exposure to the borrower after taking into account the haircut value of the collateral, the net credit exposure must be assigned a risk-weight that is equal to the product of the risk-weight applicable to the borrower and the currency mismatch multiplier.

Chapter III STC approach Page 36 of 65
Natural hedge
(a) Acceptable assets
Q86. What is the difference between “income” and “income sources” mentioned in paragraph (d)(i) of the definition of “unhedged credit exposure” in §51A(1)? A86. The wording “income sources” is intended to accommodate natural hedges in the form of financial assets held by a borrower. The term “income” refers to cash inflows such as salaries, rentals, dividends and interest incomes, while “income sources” are assets or operations that generate incomes, e.g. debt securities and employment. Q87. What assets are acceptable as natural hedge for wealth-based or collateralized lending? A87. Subject to any guidance that may be issued by the Basel Committee from time to time, a financial asset that meets all of the following requirements is acceptable (“acceptable financial asset”)— (a) the asset is free of encumbrance and denominated in the same currency as the exposure concerned; (b) its market price can be readily and objectively valued; and (c) there is an established liquid market in which it can be liquidated readily. In general, in a liquid market, a large volume of assets can be readily bought or sold with minimal effect on their prices. Accordingly, the HKMA expects that the assets counted as a natural hedge must be capable of being converted into cash quickly (within 20 business days) without significant discount in value/price. To avoid doubt, cash value under insurance policies (regardless of whether the cash value is guaranteed or not and the time needed to redeem the cash value), and asset or income that is merely declared by a customer without any proof provided to the AI, are not acceptable. Q88. Income sources are referred to in paragraph (d)(i) of the definition of “unhedged credit exposure” in §51A(1) as a means of hedging. Does this imply that tangible or intangible assets, other than the acceptable financial assets mentioned in the answer to Q87, that are able to generate incomes (e.g. real estate, and insurance policy that generates annuity income or policy dividend) can also be regarded as natural hedge in the case of wealth-based or collateralized lending? A88. The wording of paragraph (d)(i) of the definition of “unhedged credit exposure” is to provide for the flexibility to accommodate guidance to be provided by the HKMA on the acceptable types of natural hedge.

Chapter III STC approach Page 37 of 65
As mentioned in the answer to Q87, only liquid financial assets (e.g. equities, investment funds and debt securities) are acceptable, hence, the values of any intangible assets (e.g. copyright) and other assets (e.g. real estate) will not be taken into account in calculating the amount of available natural hedge. However, as natural hedge is described in CRE20.93 of the Basel Framework as the income received by the borrower from its normal operating procedures, AIs may regard regular incomes, and incomes with high certainty in terms of timing and amount, generated by those intangible and other assets (such as rental incomes and annuity payments made under annuity plans) as natural hedge. (b) Calculation of amount of natural hedge Q89. For acceptable financial assets that are regarded as natural hedge, how should an AI calculate the natural hedge value and determine if the 90% threshold mentioned in paragraph (d) of the definition of “unhedged credit exposure” in §51A(1) is met for wealth-based or collateralized lending? A89. (a) Wealth-based/net worth-based lending Subject to any guidance that may be issued by the Basel Committee from time to time and in the case of a loan (“subject loan”) in currency A, the AI should determine if the 90% threshold is met in a manner as illustrated by the numerical examples below:
In the following examples, the total market value of the unencumbered, acceptable financial assets of the borrower in currency A = $10,000k (“relevant asset amount”). (i) Example 1:
The subject loan is an instalment loan with 240 outstanding monthly instalments where each instalment is $30k and the borrower does not have any other in-scope exposures (see paragraph (b) of the definition of “unhedged credit exposure” in §51A(1)) in currency A. In such case, the value of the natural hedge = $10,000k / 240 = $41.7k. Hence, the currency mismatch multiplier does not apply to the subject loan. If the subject loan is a mortgage loan and the above calculation is conducted at origination of the loan, any down payment amount in currency A to be paid by the borrower should be deducted from the relevant asset amount. (ii) Example 2:
There are two subject loans: a 1-year bullet loan of $5,000k (“loan A”) and a loan to be repaid by 4 quarterly instalments (“loan B”) where the 4 instalment amounts are $1,000k, $1,500k, $2,000k and $2,500k respectively. In such case, the relevant asset amount is divided into 5 portions and each portion is regarded as a natural hedge against an instalment or bullet payment as illustrated below—

Chapter III STC approach Page 38 of 65
 1 st portion (loan A): $5,000k0.9 = $4,500k  2 nd portion (4 th instalment of loan B): $2,500k0.9 = $2,250k  3 rd portion (3 rd instalment of loan B): $2,000k0.9 = $1,800k  4 th portion (2 nd instalment of loan B): $1,500k0.9 = $1,350k  5 th portion (1 st instalment of loan B): $10,000k - 9,900k = $100k Since not all instalments of loan B are sufficiently hedged, loan B is subject to the currency mismatch multiplier while loan A is not. Alternatively, an AI may choose to allocate the relevant asset amount to the instalments of loan B first such that loan B is not subject to the currency mismatch multiplier but loan A is. (b) Collateralized lending Subject to any guidance that may be issued by the Basel Committee from time to time and in the case of a collateralized loan denominated in currency A (“subject loan”), the AI is expected to have specified a maximum loan-to￾value (“LTV”) ratio for each type of asset acceptable as collateral. For the purpose of determining the applicability of the currency mismatch multiplier— (i) the AI must not include recognized collateral as the natural hedge of the subject loan; (ii) if an acceptable financial asset in currency A other than recognized collateral (“subject collateral”) is the only collateral for the subject loan—  in cases where the subject loan is a bullet loan and the only loan secured by the subject collateral, the amount of the natural hedge is equal to the market value of the subject collateral times the applicable maximum LTV ratio (“haircut collateral value”);  in cases where the subject loan is an instalment loan similar to the one in Example 1 above and the only loan secured by the subject collateral, the value of the natural hedge for each instalment is calculated in the same way as Example 1, except that “relevant asset amount” should be replaced with the haircut collateral value of the subject collateral;  in cases where the subject collateral secures more than one loan/facility (including the subject loan) granted by the AI (whether denominated in currency A or not) and the subject collateral is the only collateral for those loans/facilities, the AI should determine if the 90% threshold is met in a manner as illustrated by Example 3 below.

Chapter III STC approach Page 39 of 65
(iii) Example 3:
Currency A is HKD and the haircut collateral value of the subject collateral is HKD10,000k. There are two loans: a subject loan to be repaid by 4 quarterly instalments (“loan C”) where the 4 instalment amounts are HKD500k, HKD1,000k, HKD1,500k and HKD1,500k respectively and a GBP overdraft facility whose credit limit is equivalent to HKD5,000k (“loan D”). The haircut collateral value should be reduced by the credit limit of loan D (i.e. the amount that can be used as natural hedges for loan C is only HKD5,000k). The natural hedge against each instalment of loan C is then calculated as follows—  1 st portion (4 th instalment of loan C): $1,500k0.9 = $1,350k  2 nd portion (3 rd instalment of loan C): $1,500k0.9 = $1,350k  3 rd portion (2 nd instalment of loan C): $1,000k0.9 = $900k  4 th portion (1 st instalment of loan C): $500k0.9 = $450k Since all instalments of loan C are sufficiently hedged, loan C is not subject to the currency mismatch multiplier. (c) In cases where the subject loan is secured by a pool of collateral that consists of the subject collateral and one or more than one other collateral in a currency other than currency A, the amount of the natural hedge should be calculated in a similar manner as described in paragraph (b)(i) and (ii) (first and second bullets) and the collaterals in currencies other than currency A should not be taken into account in the calculation. If the pool of collateral secures more than one loan/facility (including the subject loan), any loans/facilities that are denominated in currencies other than currency A can be ignored in the calculation described in paragraph (b)(ii) (third bullet).
10. Regulatory real estate exposures
§65(1) and (2) - criteria to be met for real estate exposure to be treated as regulatory real estate exposure (a) Type of properties Q90. If an exposure is granted for the acquisition of forest or agricultural land (“non￾residential real estate”) and secured by the said real estate where all criteria set out in §65(1) are met, is the exposure a regulatory commercial real estate exposure? A90. For the purposes of §65 to §65E, all immovable properties other than residential properties should be treated as commercial properties. Hence, a regulatory real estate exposure secured by forest or agricultural land, unless it falls within the definition of “land acquisition, development and construction exposure” in §51(1), should be classified as regulatory commercial real estate exposure.

Chapter III STC approach Page 40 of 65
Q91. Could guidance be provided on the meaning of “subsidized home ownership scheme” mentioned in §65(2)(c)(ii)(A)? A91. In general, “subsidized home ownership scheme” refers to schemes such as the Home Ownership Scheme (居者有其屋) and the Green Form Subsidized Home Ownership Scheme (綠置居) where the developers of the properties sold under such schemes include the Hong Kong Housing Authority or a domestic public sector entity (e.g. the Urban Renewal Authority) and the properties are sold at a discount in order to provide affordable housing to qualified buyers. Q92. For the requirement of written and reasoned legal advice set out in §65(2)(c)(ii)(B), is a blanket legal advice for each jurisdiction sufficient? A92. A blanket legal advice is acceptable. However, if the legal powers of the sovereign or public sector entity of the jurisdiction concerned are exercisable in relation to certain type(s) of residential properties development projects only, the AI still has the responsibility to verify that the development project in question is within the scope of the legal powers of the sovereign or public sector entity. (b) Loan purpose Q93. If an exposure is granted for cashing out the equity in a property, rather than for acquisition of the property, given all criteria set out in §65(1) are met, is the exposure a regulatory real estate exposure? A93. Cashing out equity is one of the eligible loan purposes set out in §65(1)(e). Hence, an exposure that is granted for such purpose and fulfils all other criteria set out in §65(1) is a regulatory real estate exposure. (c) Underwriting policies Q94. 13Would the HKMA issue any guidance on the requirements on underwriting policies set out in §65(1)(f)? A94. Apart from the guidance set out in the SPM modules on credit risk management (e.g. CR-G-1 on “General principles of credit risk management”) and the supervisory requirements issued by the HKMA from time to time (e.g. those on LTV ratios and debt servicing ratios in respect of mortgage loans), AIs should also implement underwriting policies and procedures that mirror the Principles for sound residential mortgage underwriting practices (April 2012) issued by the Financial Stability Board.

13 Reference: Basel Framework CRE20.73.

Chapter III STC approach Page 41 of 65
(d) Valuation
Q95. 14According to §65(1)(g), the mortgaged property should be valued in a manner consistent with the relevant guidance issued by the Monetary Authority. Is that guidance the SPM module CR-G-7 on “Collateral and Guarantees”? A95. Yes. Q96. Regarding the requirements on valuation of real estate per subsection 5.4 of SPM CR-G-7, clarification is sought on whether it is suggested for all real estate collaterals to be revalued on a quarterly basis, or only for the exposures that are regarded as “large or problem”? A96. The guidance in subsection 5.4 of SPM CR-G-7 on revaluation is related to risk management for the purpose of monitoring the effectiveness of credit protection. For the purposes of the BCR, the LTV ratio of an exposure must be calculated by using the value of the mortgaged property at origination. Revaluation is needed only in the scenarios mentioned in §65A(6) and (7). SPM CR-G-7 is only relevant to §65(1)(g) which is about the manner in which a valuation is arrived at. §65(4)(a) - dependence on cash flows generated from mortgaged property Q97. 15Regarding the assessment of whether an exposure is materially dependent on the cash flows generated by the property securing the exposure (“material dependence assessment”)— (a) are AIs required to update the assessment regularly? (b) is there any definition of “materially dependent on”? A97. (a) Having considered that mortgage loans are not usually subject to regular credit review, AIs may rely on the information obtained at loan origination for the material dependence assessment unless the AI is aware of any changes in the obligors’ sources of repayment. Nevertheless, if there is a more recent assessment conducted post loan drawdown (e.g. for considering a new loan application), AIs are expected to use the latest assessment result for the purposes of the BCR. For loans secured by properties that are subject to regular credit review, AIs are expected to update the materially dependence assessment during the credit review. (b) The primary source of the cash flows generated by the property would generally be lease or rental payments, or the sale of the property. Although exposures that depend materially on the cash flows generated by the properties securing the exposures are expected to be mainly exposures to corporates (including small businesses) or SPVs, other obligor types are also

14 Reference: Basel Framework CRE20.75(2).
15 Reference: Basel Framework CRE20.80.

Chapter III STC approach Page 42 of 65
possible (e.g. individuals, which are common in Hong Kong).
In general, it would be reasonable to consider an exposure as materially dependent if more than 50% of the obligor’s income used in the AI’s repayment ability assessment (e.g. in calculating the debt servicing ratio (“DSR”)) is from cash flows generated by the property securing the exposure. In calculating the percentage of the obligor’s total income contributed by cash flows generated by the property, if the exposure is subject to the DSR requirements promulgated by the HKMA, AIs may use the cash flows generated by the property that have been included as income for the DSR calculation as the numerator. In such case, the denominator would then be the same figure as the one used in calculating the DSR. §65(4)(b) and (c) - exceptions to §65(4)(a) Q98. If an obligor declares that there is more than one primary residence, is it correct that an AI may only count one property as the obligor’s primary residence and should determine which property is the primary residence based on the information provided by the obligor? A98. For the purposes of §65(4)(b) and (c), each obligor that is a natural person can only have one property as the primary residence. The AI should request the obligor to declare only one property as the primary residence. If there are legacy cases where the obligor has declared more than one property as the primary residence, the AI can only choose any one of them as the primary residence. Q99. When counting the total number of residential properties for the purpose of §65(4)(c)(ii)— (a) can multiple properties covered by a single title deed be counted as one property? (b) can residential properties with car parking space under the same title deed be counted as one residential property? (c) can self-use standalone car parking space be excluded? A99. (a) Residential properties covered by a single title deed can be counted as one residential property provided that the number of residential properties under the single title deed does not exceed three. However, if not all the properties are the primary residence of the individual who is the borrower or the owner of the borrower, the AI cannot rely on the ground of primary residence to claim that the exposure secured by the properties is not materially dependent on cash flows generated by the mortgaged properties. If there are more than three properties under the single title deed, each of the properties must be counted as one property. (b) Only the residential properties are counted, the car parking space will not be considered as a separate property and therefore will not be counted for the purpose of determining whether the residential real estate exposure concerned

Chapter III STC approach Page 43 of 65
can be regarded as not materially dependent on cash flows generated by the mortgaged property. (c) Self-use standalone car parking space, which is regarded as commercial property, is not counted. Q100. When counting the total number of residential properties for the purposes of §65(4)(c)(ii), should the properties pledged for exposures that fall within §65(4)(d) be counted? A100. When counting the number of properties for the purposes of §65(4)(c), only the residential property that falls within §65(4)(b)(i) or (ii) can be excluded. All other residential properties of the same obligor (including those owned by the obligor through one or more property-holding shell companies) that are provided to the AI as security must be counted, regardless of whether the exposures secured by those other residential properties fall within §65(4)(d). Q101. Multiple residential mortgage loans were granted to the same borrower on different days. The classification of the loans during the period in which the loans remained outstanding is summarized in the table below. Clarification is sought on the correct classification of the loans on Day 3 and whether the classification on other days is correct. Classification under §65(4)(a) Classification under §65(4)(b) or (c) Day 1 Day 2 Day 3 Day 4 Loan A (secured by primary residence) Non-IPRRE* Non￾IPRRE Loan B IPRRE Non￾IPRRE Non￾IPRRE ? - Loan C IPRRE - Non￾IPRRE ? Non￾IPRRE Loan D IPRRE - - ? Non￾IPRRE *Non-IPRRE denotes exposure that does not depend materially on the cash flows generated by the mortgaged property. IPRRE denotes exposure that is not non-IPRRE. Rationale:
(a) Day 1: Two loans were outstanding. Since the total number of mortgaged properties (after excluding the primary residence) was 1, Loan B could be treated as non-IPRRE under §65(4)(c). (b) Day 2: Loan C was granted on Day 2. The total number of mortgaged properties (after excluding the primary residence) increased to 2, both Loan B and Loan C could be treated as non-IPRRE under §65(4)(c).

Chapter III STC approach Page 44 of 65
(c) Day 3: Loan D was granted on Day 3. It is unclear whether only Loan D should be treated as IPRRE (i.e. Loan B and Loan C would remain as non-IPRRE). (d) Day 4: Loan B was fully repaid, hence, the total number of mortgaged properties (after excluding the primary residence) decreased to 2. Loan C and Loan D could be treated as non-IPRRE under §65(4)(c). A101. The treatments on Day 1, Day 2 and Day 4 are in line with the requirements under
section 65(4)(c).
On Day 3, since the total number of mortgaged properties has exceeded 2, all the loans (except Loan A) must be treated as IPRRE. §65(4)(c) requires each loan to meet the conditions set out in §65(4)(c)(i) and (ii). Hence, Loan B and Loan C no longer met the condition set out in §65(4)(c)(ii) when the number of mortgaged properties exceeded 2. The correct classification is summarized below:
Loan A
(secured by primary residence)
Loan B Loan C Loan D
Day 1 Non-IPRRE Non-IPRRE - -
Day 2 Non-IPRRE -
Day 3 Non-IPRRE
Day 4 Non-IPRRE - Non-IPRRE Non-IPRRE
Q102. Customer A has in total 4 residential mortgage loans with details as shown below:
Borrower(s) Primary residence
RML 1 Customer A and Customer B as joint borrowers Customer B RML 2 Customer A as single borrower Customer A RML 3 Customer A as single borrower N/A RML 4 Customer A as single borrower N/A Which of the 4 loans would be regarded as an exposure that does not depend materially on the cash flows generated by the mortgaged property (“non-IPRE exposure”) under §65(4)(b) or (c)?

Chapter III STC approach Page 45 of 65
A102. The classification of the loans is as follows:
Non-IPRE exposure?
RML 1 Yes (based on §65(4)(b)(i))
RML 2 Yes (based on §65(4)(b)(i))
RML 3 Yes or No, subject to the assessment of the AI in accordance with §65(4)(a) RML 4 Yes or No, subject to the assessment of the AI in accordance with §65(4)(a) Since the total number of properties (excluding the primary residence) pledged by Customer A is 3 (i.e. the properties securing RML 1, RML 3 and RML 4), both RML 3 and RML 4 fail to meet the condition set out in §65(4)(c)(ii). As a result, RML 3 and RML 4 are not eligible for the preferential treatment provided for under §65(4)(c). Classification of an exposure secured by both residential and commercial properties (a) A collateral pool containing both residential and commercial properties Q103. If a regulatory real estate exposure is secured by a pool of residential and commercial properties, could the exposure be classified in the following ways— (a) as a regulatory residential real estate (“RRE”) exposure if only the values of the residential properties are taken into account in the LTV ratio calculation; or (b) as a regulatory commercial real estate (“CRE”) exposure if the values of both the commercial and residential properties are taken into account in the LTV ratio calculation? A103. As there should be separate valuations for the commercial properties and the residential properties, the AI is expected to divide the exposure into two portions (i.e. one regulatory RRE exposure and one regulatory CRE exposure) which respectively represents the proportion of the exposure covered by each type of property. The LTV ratio of the regulatory RRE exposure should be calculated based on the valuation of the residential properties, and the LTV ratio of the regulatory CRE exposure should be calculated based on the valuation of the commercial properties. To avoid doubt, the division of the exposure is only a mathematical operation, it is not meant to require execution of a separate set of loan documentation and mortgage deed for each portion of the exposure. Q104. If multiple regulatory real estate exposures granted to a borrower are secured by a pool of commercial and residential properties, could the exposures be classified in the following ways— (a) all the exposures are classified as regulatory RRE exposures if only the values of the residential properties are taken into account in the LTV ratio calculation; or

Chapter III STC approach Page 46 of 65
(b) all the exposures are classified as regulatory CRE exposures if the values of both the commercial and residential properties are taken into account in the LTV ratio calculation? A104. If one or more of the exposures default, the AI is entitled to demand repayments of all other exposures, take procession of all the properties in the pool (including any excess collateral, i.e. any property that is not financed by the exposures but is provided by the borrower to offer additional comfort to the AI) and exercise its power of sale in respect of those properties, the AI must treat all exposures for financing the residential properties in the pool as a single regulatory RRE exposure, and those exposures for financing the commercial properties in the pool as a single regulatory CRE exposure. In calculating the LTV ratio of the regulatory RRE exposure, only the aggregate value of the residential properties in the pool (including any excess collaterals that are residential properties) can be taken into account. Similarly, the LTV ratio of the regulatory CRE exposure must not take into account the aggregate value of any residential properties (including any excess collaterals that are residential properties) in the pool. As mentioned in A103, the division of the exposures is only a mathematical operation. (b) A property unit that can be used for both residential and commercial purposes Q105. If a regulatory real estate exposure is secured by a property unit that can be used for both commercial and residential purposes, can the exposure be classified as either regulatory RRE exposure or regulatory CRE exposure based on the use claimed by the borrower? A105. If the property unit can be used legally for both commercial and residential purposes, it is reasonable for the lending AI to categorize the loan granted for financing the acquisition of such property unit based on the intended use of the property unit declared by the borrower in the loan application. For any loan that was classified as a regulatory RRE exposure at loan origination, the AI is expected to require the borrower to obtain the AI’s prior consent if, after the loan is granted, the borrower wants to change the use of the property unit from residential use to commercial use. The AI should then reclassify the exposure as “regulatory CRE exposure”. (c) A whole block of composite (residential and commercial) building Q106. If a single mortgage loan is granted for the acquisition of a whole block of composite building (consisting of both commercial units and residential units) in one deed, given all requirements set out in §65(1) are met, whether the loan would be regarded as a regulatory RRE exposure? A106. If there are separate valuations for the commercial portion and the residential portion of the building, the AI is expected to divide the mortgage loan into two portions (i.e. one regulatory RRE exposure and one regulatory CRE exposure) which respectively represents the proportion of the loan
covered by each of the portions of the building. The LTV ratio of the regulatory RRE exposure should be calculated based on the valuation of the residential portion of the building, and the LTV ratio of the

Chapter III STC approach Page 47 of 65
regulatory CRE exposure should be calculated based on the valuation of the commercial portion of the building. As mentioned in A103, the division of the loan into two portions is only a mathematical operation. For existing loans or new loan applications where the surveyor has not provided separate valuations for the residential portion and the commercial portion of the building, the AI may divide the loan and the valuation into portions according to the respective floor areas of the residential portion and the commercial portion of the building. It should be noted that if the whole building is acquired by the borrower for the purposes of demolition and redevelopment, the loan should be classified as ADC exposure instead of regulatory real estate exposure. Q107. Would there be a transitional period for AIs to complete loan or asset class reclassification for mortgage loans secured by a whole block of composite building, instead of requiring AIs to perform a full review for reclassification upon the implementation date? A107. For a mortgage loan secured by a whole block of composite building granted before 1 January 2025 (“cut-off date”), the HKMA will provide for a transitional period of 1 year from the cut-off date for AIs to complete the reclassification of the loan. The classification of the loan under the BCR as in force immediately before the cut￾off date (“pre-amended BCR”) should be used for capital calculations during the transitional period. Since the composite building consists of commercial units, the mortgage loan is not a residential mortgage loan under the pre-amended BCR. As such, the whole loan must be classified as a commercial real estate exposure during the transitional period. All other requirements set out in the revised STC approach that are applicable to real estate exposures, such as LTV ratio calculation and eligibility criteria for regulatory real estate exposures and exposures that are not materially dependent on the cash flows generated by the mortgaged properties, still apply to the loan during the transitional period. LTV ratio (a) Calculation for multiple exposures secured by same collateral Q108. If a mortgage loan and a working capital loan are secured by the same property with details as follows:
(a) mortgage loan amount = $200
(b) working capital loan amount = $400
(c) property collateral value = $1000 what is the LTV ratio of the mortgage loan?

Chapter III STC approach Page 48 of 65
A108. According to §65A(3), both the mortgage loan amount and the working capital loan amount must be included in the LTV ratio calculation. That is, the LTV ratio for determining the risk-weight applicable to the mortgage loan will be 60% (i.e. (200+400)/1000). The treatment reflects the thinking that the working capital loan will reduce the amount of sale proceeds of the property available for repaying the mortgage loan, and therefore it should be included in the LTV ratio calculation. Otherwise, the risk of the mortgage loan is underestimated. The working capital loan will be considered as “other real estate” exposure whose risk-weight does not depend on LTV ratio. (b) Property value used in calculation Q109. For refinancing of a real estate exposure, if the current market value (“CMV”) of the property concerned at the time of refinancing is higher than the CMV at the time when the original exposure was granted, can the CMV at the time of refinancing be used for calculating the LTV ratio? A109. If the original loan is granted by another AI, the AI providing refinancing to that loan may use the valuation of the property at the time of origination of the refinancing loan (“CMV”) to calculate the LTV ratio. If both loans are granted by the same AI, the following general principles will apply:
(a) if the amount of the refinancing loan is larger than the outstanding amount of the original loan—CMV; (b) if the amount of the refinancing loan is the same as or smaller than the outstanding amount of the original loan— (i) CMV if it is lower than the market value of the property at the time of origination of the original loan (“OCMV”); (ii) in any other case—OCMV.
11. Other real estate exposures
Q110. If a general purpose loan granted to a borrower is secured by a basket of collaterals that include real estates as one of the collaterals, and the proportion of real estates in the collateral pool is small (e.g. <10%) or the real estate collateral is only considered as an additional comfort, should the exposure be classified as “other real estate exposure”? Could it be classified as retail/corporate exposure depending on the type of borrower? A110. As defined in §51(1), a real estate exposure is an exposure secured by immovable property. Immovable property may not be the only collateral afforded to the exposure. Hence, as long as the general purpose loan is partially or fully secured by immovable property, it should be classified as other real estate exposure (i.e. a real

Chapter III STC approach Page 49 of 65
estate exposure that is neither a regulatory real estate exposure nor an ADC exposure). Q111. There are scenarios that a general banking facility offered to a borrower group consists of a pool of term loans, revolving loans, overdraft, etc., and is secured by a basket of collaterals that include real estates, debt or equity securities, deposits, etc. The facility limit is shared by the members in the borrower group dynamically. Borrowers are free to replace the original collaterals by other collaterals which may be in different collateral types. There is no restriction in the composition of collateral types. Thus the loans or exposures are not always secured by real estate, and the proportion of real estate in the basket of collaterals could be really small. As long as the collateral value is sufficient, the original loans are not required to be settled or refinanced. In this connection, are AIs really expected to classify such exposures as other real estate exposures, even the proportion of real estate in the collateral pool is really small and could be replaced anytime? Any flexibility for AIs to classify this type of facility as retail/corporate exposure given the dynamic nature of collateral? If not, are AIs expected to proactively monitor the collaterals of these products and reclassify them as retail/corporate exposure if the underlying real estate collateral is replaced with other type of collateral? A111. As explained in the answer to Q110, any exposure that is fully or partially secured by one or more immovable properties should be classified as “real estate exposure” as required under the BCR. Such classification criterion also applies to a general banking facility with a dynamic pool of collaterals. Since an AI should be notified when the borrower requests changes in the composition of the collateral pool (otherwise the AI is unable to determine whether the collateral coverage is still sufficient) and such changes may involve increase or decrease of real estate collateral, the AI is expected to review the classification after the borrower’s request is accepted.
12. ADC exposures
Classification of exposures
Q112. Does financing of revitalization projects and renovation projects (e.g. revitalization of industrial buildings, renovation of hotel buildings, etc.) fall within the exposure class of “ADC exposure”? A112. Exposures arising from revitalization projects or renovation projects that do not involve development and construction of residential or commercial properties can be excluded from the exposure class of “ADC exposure”. Q113. After the construction of the property underlying an ADC exposure is complete, will the exposure be reclassified as “regulatory real estate exposure” (§65B or §65C) or “other real estate exposure” (§65D)?

Chapter III STC approach Page 50 of 65
A113. Whether an exposure is an ADC exposure is determined based on its loan purpose. If an ADC exposure is granted to finance or refinance the construction of a property, the exposure is still an ADC exposure after the property is fully completed and until the exposure is fully repaid. Q114. There are scenarios in which the obligor in respect of an ADC exposure will need refinancing after the property concerned is fully completed (e.g. the construction of a mall, which upon completion will not be sold but will be operated as an income producing property). In this situation, if the obligor seeks refinancing from the same bank, would the original ADC exposure be considered as fully repaid and therefore the refinancing facility could be classified as a real estate exposure (other than ADC exposure), given the property concerned is fully completed and the bank has conducted sufficient due diligence work for the refinancing facility as if it were a new loan with a new loan purpose? Also, are the terms and conditions of the refinancing facility required to be determined based on the nature and purpose of the refinancing facility (instead of, e.g., sticking to the terms and conditions of the original ADC loan)? A114. In order for the refinancing loan granted by the same bank secured by the completed property to be classified as real estate exposure (other than ADC exposure), the loan should be considered by the bank as a new loan (instead of, e.g. an extension of the maturity of the original loan) and is subject to all relevant due diligence processes that apply to new loans. Since the new loan is to be secured by a completed property, the credit considerations should be different from an ADC loan. Also, the economic and property market situations may have changed. AIs are expected to take into account these factors when determining the terms and conditions of the new loan. AIs are reminded that the loan purpose of refinancing an ADC loan is not one of the loan purposes specified in §65(1)(e) of the BCR. Accordingly, the refinancing loan should be risk-weighted in accordance with §65D of the BCR. ADC exposures eligible for 100% risk-weight Q115. Residential projects in Hong Kong usually involve development and construction of both residential units and other non-residential facilities (e.g. car parking spaces, club houses, etc.) and the latter are an integral part of the project. Hence, an AI’s financing of such a project will be based on the whole project. Deliberately separating the project into residential and commercial portions for valuation, loan records, etc. is not a general market practice and may lead to significant changes to the existing operational flow. As such, clarification is sought on whether the AI may treat the ADC exposure for financing the whole project as an ADC exposure in respect of residential property for the purpose of §65F(2). A115. If, according to the relevant land leases, the non-residential facilities (e.g. car parking
spaces and club houses) can only be used by the residents of the residential

Chapter III STC approach Page 51 of 65
units of the project and their bona fide guests and visitors, the facilities can be included as part of the residential properties concerned. For other non-residential facilities, such as shopping malls, under the same project that must be classified as commercial properties, the same principle mentioned in Q106 applies to the division of the ADC exposure into residential and commercial portions when separate valuations are unavailable. When determining whether the requirements under the HKMA’s circular of 28 February 2024 16 (and any revision that may be made by the HKMA subsequently to those requirements) are met, there is no need to make the apportionment mentioned in the previous paragraph. The apportionment is just for determining the amount of the exposure to be assigned 100% risk-weight. As mentioned in Q106, the apportionment is just a mathematical operation to facilitate capital calculations, which should be achievable without changing the general market practice and operational flow for valuation and loan documentation. Q116. If a property development project involves construction of a building that contains both residential part (majority) and commercial part (e.g. G/F to 3/F of the building is a shopping mall and 4/F and above are residential units), could AIs classify financing of this type of property development project as ADC exposures in respect of residential properties so as to be entitled to 100% risk￾weight? A116. The ADC exposure may be risk-weighted at 100%— (a) only to the extent of the portion that is for financing the development and construction of the residential units; and (b) only if the AI is satisfied that the repayment ability of the borrower in respect of the portion referred to in paragraph (a) does not depend on the completion of the commercial part of the project. If it is infeasible to identify the portion for residential units and/or the AI is uncertain about whether the condition in paragraph (b) is met, the whole ADC exposure must be risk-weighted at 150%. Q117. In determining whether an ADC exposure is eligible for a risk-weight of 100%— (a) which criteria set out in §65(1) are considered applicable when assessing whether §65F(2)(a) is met? (b) is it correct that “residential property” referred to in § 65F(2) only includes immovable property that falls within §65(2)(c)? A117. The policy intent is that only those criteria set out in §65(1) that are related to credit underwriting and applicable to ADC exposures should be met. Accordingly, AIs

16 https://www.hkma.gov.hk/media/eng/doc/key-information/guidelines-and-circular/2024/20240228e1.pdf

Chapter III STC approach Page 52 of 65
are expected to consider, as a minimum, the requirements set out in §65(1)(b), (c), (f), (g) and (h). Hence, AIs are not required to assess eligibility against §65(2)(c). Q118. For the purpose of determining whether a property developer in respect of an ADC exposure has substantial equity at risk (see §65F(2)(b)), could AIs rely on the financing caps referred to in page 3 of the circular issued by the HKMA on 28 February 2024? A118. If an AI has granted an ADC exposure in compliance with the relevant guidance on property construction finance set out in the circular issued by the HKMA on 28 February 2024 (and any revision that may be made by the HKMA subsequently to the requirements set out in that circular) and the following guidance, it is acceptable to see this as an indication of the borrower having substantial equity at risk— (a) if the property developer has weaker financial position (e.g. higher gearing ratios), the AI should consider adopting financing ratios that are lower than those set out in the above circular where appropriate; and (b) in determining the site value and the expected value of the completed properties, the AI should obtain valuations from at least two independent professional valuers, and use the lower valuation as the basis for calculating the financing ratios. For ADC exposures in respect of residential properties in jurisdictions outside Hong Kong, the treatment set out in §65F(3) applies if the conditions specified in that subsection are met, otherwise the exposures must be subject to the 150% risk-weight under §65F(1).
13. Equity exposures
Examples of equity exposures
Q119. 17 According to §54A(2), if an instrument represents direct or indirect ownership interests (whether voting or non-voting) in the assets and income of the issuer of the instrument or another entity, an exposure to such instrument is an equity exposure. Could some examples of indirect ownership interests be provided? A119. Indirect ownership interests include holdings of derivative contracts tied to equity interests, and holdings in corporations, partnerships, limited liability companies or other types of enterprises that issue ownership interests and are engaged principally in the business of investing in equity instruments. Equity investments in collective investment schemes are excluded. Q120. 18Could an example of an obligation falling within §54A(4)(b)(iii) be provided?

17 Reference: Basel Framework CRE20.54 footnote 20.
18 Reference: Basel Framework CRE20.55 footnote 21.

Chapter III STC approach Page 53 of 65
A120. For certain obligations that require or permit settlement by issuance of a variable number of the issuer’s equity shares, the change in the monetary value of the obligation is equal to the change in the fair value of a fixed number of equity shares multiplied by a specified factor. These obligations meet the conditions of §54A(4)(b)(iii) if both the specified factor and the referenced number of equity shares are fixed. For example, an issuer may be required to settle an obligation by issuing shares with a value equal to three times the appreciation in the fair value of 1,000 equity shares. That obligation is considered to be the same as an obligation that requires settlement by issuance of shares equal to the appreciation in the fair value of 3,000 equity shares. Q121. 19Could an example of an obligation falling within §54A(5) be provided? A121. Equities that are recorded as a loan but arise from a debt/equity swap made as part of the orderly realisation or restructuring of the debt are included in the definition of “equity exposures”. Speculative unlisted equity exposures Q122. According to paragraph (a) of the definition of “speculative unlisted equity exposure” in §65G(2), speculative unlisted equity exposures include unlisted equity investments that are invested for short-term resale purposes. Guidance is sought on the definition of “short-term” in this context. A122. AIs are expected to make their own judgement as to whether their intended holding period of the unlisted equities is “short-term”. Since unlisted equities are not traded in a liquid secondary market, it is less likely that a profitable resale deal could be closed within months after acquisition of the equities. As such, a holding period of 3 to 5 years could still be considered as “short-term” in the case of highly illiquid investments. Q123. 20Does “speculative unlisted equity exposures” defined in §65G(2) include the following investments/exposures? (a) investments in unlisted equities of corporate clients with which the AI has or intends to establish a long-term business relationship; (b) debt-equity swaps for corporate restructuring purposes; and (c) equity exposures incurred as part of the business facilitation (e.g. membership of payment systems and exchanges). A123. These investments/exposures are not regarded as “speculative unlisted equity exposures”.

19 Reference: Basel Framework CRE20.56 footnote 22.
20 Reference: Basel Framework CRE20.58 footnote 25.

Chapter III STC approach Page 54 of 65
14. CIS exposures - real estate investment trusts
Q124. For the purpose of risk-weighting an AI’s equity investments in REITs under the mandate-based approach, should the risk-weighting treatments of real estate exposures be applicable to the underlying assets of REITs? A124. If according to the mandate of the REIT concerned or the law to which the REIT is subject, the permitted real estate investments are holdings of immovable properties such as shopping malls, the underlying assets of the REIT are physical assets instead of financial assets and the applicable section is §66 (Other exposures that are not defaulted exposures) instead of §65 (What are regulatory real estate exposures). §65 is applicable only in cases where the REIT is allowed to hold mortgage loans. If the underlying assets held by the REIT are mortgage-backed securities (“MBSs”) or covered bonds, the following treatments apply— (a) MBSs, other than MBSs that fall within the definition of “securitization exposures” which are subject to Part 7 instead of Part 4 of the BCR, must be risk-weighted as exposures to the issuers of the MBSs; and (b) covered bonds secured by mortgage loans or similar real estate exposures must be risk-weighted in accordance with §59D if the covered bonds are eligible covered bonds defined in §2(1), otherwise the covered bonds must be risk￾weighted as exposures to the bond issuers.
15. Defaulted exposures
Defaulted real estate exposures
Q125. Clarification is sought on whether §67(1)(a) applies to a real estate exposure (other than a regulatory real estate exposure or an ADC exposure)— (a) that is defaulted; (b) that is secured by residential properties or a pool of residential and commercial properties; and (c) whose repayments do not materially depend on the cash flows generated by the properties securing the exposure. A125. In §67(1)(a), “a residential real estate exposure” includes any real estate exposure (other than ADC exposure) that is secured by one or more residential properties. The exposure does not need to be a regulatory residential real estate exposure in order to be eligible for the 100% risk-weight. If the exposure is secured by a pool that contains both residential properties and commercial properties, the exposure should be subject to §67(1)(b), i.e. it should be allocated a risk-weight of 150%.

Chapter III STC approach Page 55 of 65
Definition of “defaulted exposure” - facility level vs obligor level Q126. If an individual has multiple real estate exposures and one of the real estate exposures is past due for more than 90 days, should the real estate exposures that are not past due be treated as defaulted exposures under the STC approach? A126. In such case, the real estate exposure that is past due for more than 90 days must be treated as a defaulted exposure (according to §67(2)(a)). For all other real estate exposures to the individual, whether they should also be treated as defaulted exposures will depend on whether the individual is assessed by the AI concerned as a defaulted borrower according to the criteria set out in §67(4) (§67(3) does not apply to real estate exposures). If the AI chooses to apply §67(4)(a) in a conservative manner by ignoring the word “material” in that section, then all other real estate exposures that are not past due will also be treated as defaulted exposures. Q127. Clarification is sought on whether §67(3) applies to account level or product level (e.g. cards level where multiple accounts are allowed). A127. If more than one credit card is issued by the AI to an individual, each credit card account is considered as a single retail exposure. Q128. If an AI has a credit card exposure and a real estate exposure to an individual, clarification is sought on the treatment of these exposures under the following scenarios:
(a) the credit card exposure is past due for more than 90 days but the real estate exposure is not past due; (b) the real estate exposure is past due for more than 90 days but the credit card exposure is not past due. A128. In scenario (a), the treatment of the real estate exposure will depend on whether the individual is assessed by the AI as a defaulted borrower according to the criteria set out in §67(4). If the AI chooses to apply §67(4)(a) in a conservative manner by ignoring the word “material” in that section, then the real estate exposure will also be treated as a defaulted exposure. In scenario (b), as provided for by §67(3)(b), the AI has the discretion to decide whether to treat the credit card exposure as a defaulted exposure. §67(4) - defaulted borrower Q129. Guidance is sought on what should be regarded as “material” in relation to the requirements set out in §67(4)(a), (b) and (d). A129. AIs are expected to develop their own criteria in determining the materiality of a credit obligation and the materiality of credit-related economic loss for the purposes of §67(4)(a), (b) and (d). The criteria should be prudent and applied consistently within the consolidation group of an AI and should not jeopardize the AI’s internal policies and procedures for problem credit management. If an AI’s parent bank is

Chapter III STC approach Page 56 of 65
incorporated outside Hong Kong and subject to capital standards and/or supervisory guidance published by the parent bank’s regulator that have specified levels for materiality thresholds or other criteria for determining materiality, the AI’s own criteria for determining materiality should not be less prudent than those materiality thresholds and other criteria, unless otherwise justified. To avoid doubt, it is acceptable for an AI to ignore the word “material” in §67(4)(a), (b) and (d).
16. Off-balance sheet exposures
Types of off-balance sheet exposure
Q130. Since most of the sale and repurchase agreements are in repo-style, what transactions would fall within Item 3 (Sale and repurchase agreements (excluding those that are repo-style transactions)) of the Table in Schedule 6? A130. Sale and repurchase agreements referred to in Item 3 of the Table are intended to capture transactions the underlying assets of which are not securities (e.g. loans or physical assets). Definition of “commitment” in §2 of Schedule 6 Q131. What is the treatment of off-balance sheet exposures such as conditional committed facilities that require mandatory approval from an independent regulatory or government body before drawdown? A131. Facilities of such nature also fall within the definition of “commitment” and should be subject to the same treatment applicable to commitments. Q132. For the purpose of the definition of “commitment”, is there any difference between “uncommitted credit facility” and “credit facility that can be unconditionally cancelled at any time”? A132. §2 of Schedule 6 defines “commitment” as any contractual arrangement that has been offered by an AI and accepted by its customer to extend credit, purchase assets or issue credit substitutes. The term “commitment” per se may lead to misunderstanding that “uncommitted” credit facilities are excluded. However, the meaning of “commitment” given by the BCR does not exclude “uncommitted” credit facilities. Hence, “uncommitted” credit facilities, which generally include, but not limited to, facilities that can be unconditionally cancelled (i.e. an AI may, at any time, with or without cause, cancel the facility or refuse to extend credit under the facility), would fall within any one of Items 10, 11 and 12 of the Table in
Schedule 6.
Q133. Would HKMA clarify the meaning of “automatic cancellation” in respect of a commitment referred to in the Table in Schedule 6 (Credit Conversion Factors) and provide an example of a product with such feature?

Chapter III STC approach Page 57 of 65
A133. “Automatic cancellation” is a term used in CRE20.100 of the Basel Framework. The HKMA believes that the Basel Committee is talking about bespoke banking facilities. There may be customised terms and conditions in the loan agreement that when a pre-specified event has occurred (e.g. certain financial indicators have exceeded certain levels), any unused credit facilities/limits will be cancelled automatically. Q134. If uncommitted credit facilities are granted to private wealth management clients (i.e. the agreements signed by the clients explicitly state that the credit facilities are granted on an uncommitted basis), would these facilities not be considered as commitment and in turn not be subject to 10% CCF? A134. Please see Q132. Credit facilities granted to private wealth management clients that fall within the definition of “commitment” must be assigned a CCF of 10% or 40%, as the case requires. Definition of “exempt commitment” in §2 of Schedule 6 Q135. Regarding the condition set out in paragraph (f) of the definition of “exempt commitment” in §2 of Schedule 6 in relation to the assessment of creditworthiness before each drawdown, would an AI be regarded as having met such condition if the AI relies on its routine credit assessment of the obligor, supplemented by a confirmation from a party independent of the function originating the commitment, that no material adverse information has been identified subsequent to the most recent credit assessment that would affect the obligor’s creditworthiness immediately prior to the drawdown? A135. According to the definition of “exempt commitment” in §2 of Schedule 6, the exemption is available to corporates (including small businesses as defined in §51(1)) that are closely monitored on an ongoing basis. As such, AIs should have in place systems and controls, in addition to the periodic credit reviews, to monitor the day-to-day performance of borrowers (e.g. delinquency, over limit or other early signs of credit deterioration) so as to identify promptly any material change in the creditworthiness of the borrowers subsequent to the most recent full scope credit assessment (see SPM modules CR-G-1, CR-G-2 and CR-G-3). The party primarily responsible for the monitoring should be independent of the front office. For the purpose of meeting the condition set out in paragraph (f) of the definition, the HKMA would allow AIs to rely on such day-to-day monitoring. Q136. For the purpose of meeting the condition set out in paragraph (f) of the definition of “exempt commitment”, whether an AI could set up some rule￾based approach, e.g. no delinquency, LTV level, etc., to fulfil the assessment requirement without the need to perform additional credit approval process? A136. There is no specific requirement on how the day-to-day monitoring and the assessment before each drawdown should be conducted as long as they are prudent and appropriate for their intended purposes. However, paragraph (f) of the definition
of “exempt commitment” in the BCR requires that an AI’s decision on the execution of each drawdown is only made after assessing the creditworthiness of the borrower

Chapter III STC approach Page 58 of 65
immediately prior to drawdown. As such, there should be an approval or confirmation for the drawdown which is well documented to substantiate that the drawdown is executed based on satisfactory day-to-day monitoring results (e.g. no indication of concerns about the borrower’s creditworthiness at the time of the drawdown). Q137. If an AI offers credit invitations to corporates or SMEs for granting a credit facility, which are neither included in any facility letter nor made to existing clients, is such offer an exempt commitment? A137. An AI must determine whether the credit invitations fall within the definitions of “commitment” and “exempt commitment” in §2 of Schedule 6. This will require examination of the terms and conditions of the credit invitations and the obligations and liabilities imposed on the AI and the customers. Determination of applicable CCFs Q138. 21Could examples be provided to illustrate how §71(1A) works? A138. (a) If— (i) an AI granted a credit line for a borrower to open letters of credit (“L/C”) and the credit line itself is a commitment that falls with Item 12 of the
Table in Schedule 6 (in other words, it is neither an exempt commitment
nor a commitment that falls within Item 11 of that Table); and (ii) only L/Cs that fall within the definition of “trade-related contingency” in §2(1) can be opened under the credit line, a 20% CCF will be applied (instead of a 40% CCF) to the credit line. (b) If an AI granted a credit line for issuing financial guarantee (i.e. a direct credit substitute as defined in §2(1)) and such credit line can be cancelled at any time unconditionally by the AI without prior notice (i.e. it falls within Item 11 of the Table in Schedule 6), a 10% CCF will be applied (instead of a 100% CCF) to the credit line. Q139. Since Part 2 of Schedule 1 mentioned in Item 13 of the Table in Schedule 6 is currently blank, what CCF should be applied to off-balance sheet exposures that fall within Item 13? A139. If an off-balance sheet exposure falls within item 13 of the Table— (a) if Part 2 of Schedule 1 is blank, the CCF applicable to the exposure is 100%; (b) if Part 2 of Schedule 1 is not blank and a CCF specified in that Part is applicable to the exposure, then that CCF should be applied to the exposure; and

21 Reference: Basel Framework CRE20.101 footnote 46.

Chapter III STC approach Page 59 of 65
(c) if Part 2 of Schedule 1 is not blank and none of the CCF(s) specified in that
Part is applicable to the exposure, then 100% CCF should be applied to the
exposure.
Q140. If a self-liquidating trade letter of credit— (a) has an original maturity equal to or more than one year; and (b) is associated with the movement of goods, should the off-balance sheet exposure arising from such letter of credit be classified as Item 12 of the Table in Schedule 6 (i.e. “Commitments that do not fall within any of Items 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 and 11”)? A140. Since the off-balance sheet exposure does not meet the definition of “trade-related contingency” in §2(1), the exposure will be classified as Item 10 “exempt commitment” if it falls within the definition of “exempt commitment” (see §2 of
Schedule 6), otherwise it will be classified as Item 12 (assuming that the exposure
does not fall within Item 11).
17. Recognized credit risk mitigation
Q141. If an AI has put in place and implemented proper policies, procedures and processes, including valuation, and validity and enforceability checks as applicable, in accordance with the requirements in the SPM module CR-G-7 in relation to residual risks (legal, operational, liquidity and market risks), may the AI conclude that the residual risks mentioned in §77(2)(g)(ii) (recognized collateral), §98(ca)(ii) (recognized guarantee) and §99(1)(f)(ii) (recognized credit derivative contracts) are mitigated? A141. It is unlikely to mitigate all residual risks completely with 100% certainty during the whole period in which the credit risk mitigation technique concerned is in force. Hence, having proper policies, procedures and processes to identify and manage the residual risks will not remove the need of assessing whether there is any material positive correlation between the credit quality of the obligor and the residual risks. In the assessment, an AI should consider, for example, how the relative liquidity of the collateral provided by an obligor is likely to change if the obligor is in default, whether the obligor has a considerable amount of holding in the same securities that are provided to the AI as collateral and therefore if the obligor needs to liquidate its holding of the securities to cure its financial difficulties, the price of the securities is likely to be driven down, etc. Recognized collateral (a) Resolution stay Q142. AIs subject to “Resolution Stay” rules implemented in different jurisdictions are not allowed, under those rules, to terminate a transaction immediately even the counterparty to that transaction commits an event of default. In such

Chapter III STC approach Page 60 of 65
situation, whether an AI would be considered as failing to meet the requirement stated in— (a) §77(2)(b), which requires the AI to have the right to take legal possession of the collateral in a timely manner in the event of default; or (b) §82(4)(h), which requires the AI to be able to terminate the transaction immediately in the event of default. A142. The existence of a resolution stay recognition clause will not result in the condition set out in §77(2)(b) or §82(4)(h), or any other similar requirements, say paragraph (h) of the definition of “valid bilateral netting agreement” in §2(1), not being met. For this purpose, resolution stay recognition clause means any term or condition of the contract that gives rise to the collateralized exposure to the effect that the parties to the contract agree to be bound by— (a) a suspension of the parties’ right to terminate the contract or to exercise rights over the collateral imposed by a resolution authority under the Financial Institutions (Resolution) Ordinance (Cap. 628); or (b) any similar suspension imposed by a non-Hong Kong resolution authority (within the meaning of section 2(1) of that Ordinance). (b) Main index Q143. Would the HKMA provide a list of “main index” or “main indices” for the purpose of recognizing collaterals under the simple and comprehensive approaches? A143. The definition of “main index” in §2(1) is clear. “Recognized exchange” referred to in that definition is defined in §2(1). The index products traded on the recognized exchanges can be easily identified from their websites. Hence, AIs should be able to determine whether an index is a main index. (c) Life insurance policy Q144. Could a life insurance policy pledged as collateral to an AI for premium financing/policy financing be treated as recognized collateral for the purposes of the STC approach? A144. Life insurance policy is not a type of collateral that may be recognized under the STC approach. In general, insurance policies may be treated as recognized guarantees if they satisfy all the eligibility requirements set out in §98. To avoid doubt, assignment of the rights and interests of the insured/policyholder (e.g. the borrower) under a life insurance policy to the lending AI is not regarded as a guarantee for the purpose of §98 because life insurance policy is not a type of credit insurance. Please also see Q146 and Q147.

Chapter III STC approach Page 61 of 65
Recognized guarantees and recognized credit derivative contracts (a) Applicable risk-weight of credit protection provider Q145. If an AI’s exposure that is risk-weighted by using the STC approach is covered by a recognized guarantee and the guarantor falls within an IRB adoption class for which an approval to use the IRB approach has been obtained by the AI, can the AI determine the risk-weight to be assigned to the protected amount of the exposure by using the IRB approach? A145. Under §100(2)(b)(i), the risk-weight of the protected amount of the exposure must be determined in accordance with Division 3 of Part 4 (i.e. the STC approach). (b) Recognized guarantee - eligibility requirements in §98 Q146. When assessing whether an insurance policy (e.g. mortgage insurance) meets the eligibility requirements set out in §98, is there anything that an AI should pay attention to? A146. A guarantee is considered as an instrument that allows an AI to transfer the credit risk of an exposure to the guarantor (e.g. the insurance company in the case of an insurance policy). The eligibility requirements in §98 should be assessed against the terms and conditions of the guarantee contract entered into by the guarantor and the AI. In this regard— (a) the word “guarantee” in §98 should be construed as “insurance policy” and “guarantor” should be construed as “insurer”; (b) §98(c) should be interpreted as requiring the exposure concerned (e.g. a mortgage loan) to be the subject matter of the guarantee. In other words, the exposure must be specified in the guarantee contract (or the insurance policy if the AI intends to recognize such policy for capital purposes); and (c) §98(d) is intended to ensure that the terms and conditions of the guarantee contract have provided that if the AI suffers a loss as a result of an event of default (as defined in the guarantee contract) relating to the guaranteed exposure, the guarantor is obliged to pay an amount as defined in the guarantee contract to the AI. Hence, in the case of an insurance policy, the AI should assess whether the circumstances specified in the insurance policy under which the insurance company is obliged to make payment to the AI (i.e. the risks insured or insurance coverage) include the default of the borrower relating to the guaranteed exposure. Q147. Insurance companies are generally able to cancel a policy if the policy holder has any fraudulent behavior (e.g. using fake ID, misstatement of health condition), would such a mechanism be regarded as a breach of §98(e) and cause an insurance policy to fail to be a recognized guarantee? A147. For insurance policies that may be recognized under the BCR (e.g. mortgage insurance), the policy holder is the AI itself. It is under the direct control of the AI to avoid any act that would result in the insurance policy being cancelled by the

Chapter III STC approach Page 62 of 65
insurer. Hence, the existence of a clause that allows the insurer to cancel the policy if the AI (as the policy holder) has any fraudulent behavior is not a breach of §98(e). (c) Guarantees provided by related companies Q148. According to §98(ca)(i), the credit quality of the obligor in respect of an exposure covered by a guarantee must not have material positive correlation with the credit quality of the guarantor concerned. Hence, it appears that guarantees provided by the obligor’s parent, subsidiaries or affiliate companies are not recognized for the purposes of the STC approach because the credit qualities of these parties are likely to be positively correlated with the credit quality of the obligor. Is this understanding correct? A148. The list of eligible credit protection provider set out in §99A does not exclude parent, subsidiaries and affiliate companies of the obligor. Guarantees provided by these entities are recognized provided that all the conditions set out in §98 are met, and for this purpose, there is no presumed material positive correlation between the obligor and its parent, subsidiary or affiliate company. AIs are expected to prudently assess whether material positive correlation exists (e.g. if the guarantee is provided by the obligor’s parent, whether the obligor is the major source of income of its parent) in order to determine whether the guarantee concerned can be recognized. (d) §99A - prudentially regulated financial institutions Q149. Could further guidance be provided on the definition of “prudentially regulated financial institution” in §99A(3)?

Chapter III STC approach Page 63 of 65
A149. The definition of “prudentially regulated financial institution” is formulated to take into consideration the classifications necessary for risk-weighting purposes under the STC approach. Features of entities Classification for risk￾weighting purposes Classification for credit risk mitigation purposes Non-bank entities engaged in regulated financial activities subject to regulatory and supervisory standards comparable to those for banks QNBFI Prudentially regulated financial institution Non-bank entities engaged in regulated financial activities subject to supervisory standards consistent with international norms but not comparable to those for banks Corporate Prudentially regulated financial institution Non-bank entities engaged in regulated financial activities subject to supervisory standards that are not consistent with international norms Corporate Corporate Non-bank entities engaged in financial activities that are not regulated financial activities Corporate Corporate Q150. What are the differences between the entities defined in paragraphs (a) and (b) of the definition of “prudentially regulated financial institution” in §99A(3)? A150. Paragraph (a) of the definition refers to entities (other than banks) subject to supervisory standards comparable to those applicable to banks, while paragraph (b) of the definition refers to entities (other than banks) subject to supervisory standards that are not comparable to those applicable to banks but are substantially consistent with international norms. Q151. How should AIs assess whether an entity is a major entity in a group for the purposes of paragraph (c) of the definition of “prudentially regulated financial institution” in §99A(3)? A151. There is no simple and single rule to determine whether an entity is a major entity. Although asset size, net profit, total income, etc. are usually the commonly used financial metrics, it may also be appropriate to take into consideration other relevant risk-based factors, such as size of risk exposures. Hence, AIs are expected to exercise expert judgement in deciding which metric(s) should be used after considering the specific circumstances of the entity and the group concerned.

Chapter III STC approach Page 64 of 65
Multiple recognized credit risk mitigation
Q152. If a defaulted exposure is covered by insurance protection and collateral, can the insurance covered portion be applied to the underwater portion of the collateral? A152. The treatments for multiple recognized credit risk mitigation are set out in §102 of the BCR and apply to all banking book exposures including defaulted exposures. If an exposure is partially secured by recognized collateral and partially covered by a recognized guarantee in the form of insurance, §102 does not prohibit allocating the insurance protection to the portion of the exposure that is not secured by the collateral if such allocation is consistent with the rights of the AI and the insurance coverage under the security documents and the insurance policy concerned. If the exposure concerned is a mortgage loan in negative equity with a mortgage insurance, §102 is not applicable because only the mortgage insurance may be considered as recognized guarantee (real estate is not a type of collateral that may be recognized under the STC approach). The credit protection covered portion in relation to the mortgage insurance must be determined in accordance with the terms and conditions of the mortgage insurance policy and based on the current outstanding loan amount. If an AI has classified the mortgage loan as a defaulted exposure according to the criteria set out in §67(2) and (4), the credit protection uncovered portion of the loan (i.e. the portion not covered by the mortgage insurance) should be risk-weighted according to §67(1)(a) or (b), as the case requires.

Chapter III STC approach Page 65 of 65
Annex 1
(For simplicity, charges, fees and interests are ignored in the example below) ABC 銀行 ABC Bank 樣本 信用卡戶口月結單 STATEMENT OF CREDIT CARD ACCOUNT SAMPLE 戶口號碼 Account Number XXXX Customer Name 信用限額 Credit Limit HKD XX,XXX Customer Address 總結欠 Statement Balance HKD 23,787.00 最低還款額 Minimum Payment HKD 238.00 結單日 Statement Date XX APR 2024 到期還款日 Due Date XX MAY 2024 交易日期 Trans Date 過帳日期 Post Date 交易說明 Description of Transaction 金額(港幣) Amount(HKD) 上期結餘 Previous Balance 7,118.00 19 Mar 19 Mar Merchant or Cash Advance Installment Plan #12/24 2,027.00 24 Mar 25 Mar Purchases 11,760.00 26 Mar 26 Mar Cash advance or Balance Transfer/Consolidation 10,000.00 12 Apr 13 Apr Payment 7,118.00 CR 今期結餘 Current Balance 23,787.00

• 財務費用概覽 Finance Charge Summary *
  類別 實際年利率 財務費用
  Type APR% Finance Charge
  購物簽帳 Retail Purchase 34.46% HKDXXX.XX
  現金透支 Cash Advance 35.61% HKDXXX.XX
  上期結餘 – 還款/ + 購物 + 現金透支 + 財務 + 逾期 + 其他 = 今期
  Previous 其他收入 簽帳 Cash 費用/ 結餘
  Balance Payment Retail Advance Financial Late Charge 調整 New & Credit Purchase Charge Fee/ Balance Adjustment
• 奬賞計劃 Rewards Programme *
  積分奬賞上期結餘 Reward Points Opening Balance
  本月賺取積分奬賞 Reward Points Earned
  本月行使積分奬賞 Reward Points Redeemed

---

Chapter IV IRB approach Page 1 of 37
IV. IRB Approach 1

1. Application and adoption
   Application for approval to use IRB approach Q1. Suppose an AI uses the foundation IRB approach for its exposures in the IRB adoption classes of corporate—other than specialized lending, sovereign and bank. Can the institution switch to using the STC approach for its sovereign exposures? If yes, is consent from the MA required for the switch? A1. Under §8(4)(a), when an AI is granted approval under §8(2)(a) to use the IRB approach for one or more than one IRB adoption class to calculate its credit risk for non-securitization exposures, the institution shall not, except with the prior consent of the MA, use any approach other than the IRB approach to calculate its credit risk for non-securitization exposures within the IRB adoption class for which an approval is granted to use the IRB approach. Accordingly, the switching to the STC approach must be justified (e.g. changes in group-wide regulatory reporting strategy against revisions of the requirements of the supervisory authority of the institution’s parent bank) and demonstrated not for regulatory capital arbitrage. The above principle also applies to the applications for switching from other IRB calculation approaches to the STC approach. Adoption of IRB approach Q2. What is the expected content of an implementation plan under §11(1)(a)? A2. The implementation plan under §11(1)(a) must specify to what extent and when an AI intends to use the IRB approach to calculate its credit risk for non-securitization exposure. It is thus expected that the plan in general includes— (a) the IRB adoption class(es) which is/are the subject of the application and the respective IRB calculation approach chosen to be used (e.g. foundation IRB approach or advanced IRB approach in case of the IRB adoption class of sovereign); (b) the expected commencement date for using the IRB approach for each IRB adoption class subject to the application; and (c) in case of a phased rollout within an IRB adoption class, the details of the planned rollout (see also A3 below).

1 This chapter is currently under revision, primarily to accommodate the review of the SPM module CA-G-4 on “Validating Risk Rating Systems under the IRB Approach”. Please refer to the HKMA’s letters to the HKAB and the DTC Association dated 29 November 2024, available on the HKMA’s public website (https://www.hkma.gov.hk/eng/regulatory-resources/consultations/) and the private website (http://www.stet.iclnet.hk/index.htm).

Chapter IV IRB approach Page 2 of 37
When submitting an implementation plan, the institution should also supplement comprehensive information useful for assessing its application. Such information is expected to be usually contained in its internal project plan, for example, but not limited to— (a) the organisational structure delineating the business lines, risk management and other functional units’ authorities and duties in the governance, developments and operations concerning the implementation of the IRB approach; (b) a gap analysis of the latest status of compliance with the applicable regulatory requirements; (c) a description of each rating system under the initial implementation or that is to be developed; (d) the details of intragroup arrangement (e.g. rating system development and validation, etc.) in cases where the institution is a subsidiary of a non-Hong Kong banking group and certain operations concerning the rating systems are performed by the associated entities of the banking group; and (e) any other information that the institution considers appropriate to be included in the plan. Q3. Is phased rollout for exposures falling within an IRB adoption class allowed when an AI applies the IRB approach to calculate its credit risk for non￾securitization exposures? A3. A phased rollout is generally not expected for exposures falling within an IRB adoption class, and an AI should ensure comprehensive readiness within that IRB adoption class (including a credible track record of at least 3 years using its rating system before the institution becomes qualified to use the relevant IRB approach). Phased rollout within an IRB adoption class may only be considered under exceptional circumstances with strong justifications, for example – (a) the formal adoption of rating systems to certain exposures within an IRB adoption class is subject to events not controlled by the institution (e.g. decisions made by home and host supervisors concerning a rating system for certain exposures within an IRB adoption class), (b) it is impracticable to roll out the IRB approach to a business unit of the institution located outside Hong Kong subject to specific local legal or regulatory restrictions (e.g. transfer of data concerning obligors); (c) the structure of the institution is complicated such that it is impracticable to roll out the IRB approach in one go. To avoid doubt, there is no supervisory expectation that AIs commence to use the IRB approach across IRB adoption classes simultaneously.

Chapter IV IRB approach Page 3 of 37
Exemption for exposures and revocation of such exemption Q4. What factors will the MA consider in determining an AI’s application for exempting a subset of exposures in an IRB adoption class or the exposures falling within a business unit from the use of the IRB approach? A4. In determining such an application from an AI, the MA will consider the following factors— (a) Practicality (re: §12(2)(a)(iii)) – whether the institution has genuine difficulty in applying the IRB approach to the exposures which are the subject of the application, due to practical reasons after taking into account the nature and complexity of the exposures (e.g. lack of data). The MA has no intention of exempting any exposures if the institution clearly has the ability to apply the IRB approach to such exposures without incurring significant cost or effort; and (b) Regulatory capital arbitrage (re: §12(2)(a)(iv)) – whether the exemption will materially prejudice the calculation of the institution’s regulatory capital for credit risk (e.g. because the requirement of the institution’s regulatory capital is artificially lowered by the institution selectively choosing a certain approach or method for certain of its exposures). In relation to the foregoing, the HKMA expects that in most cases, only immaterial exposures in terms of size and perceived risk profile within the IRB adoption class could justify that it is not practical for the institution to include such exposures in the calculation of credit risk under the IRB approach, and ensure that the exemption, if granted, would not materially prejudice the calculation of the institution’s regulatory capital for credit risk. In general, the aggregate risk-weighted amount of the non￾securitization exposures to which the exemption would relate is not expected to exceed 10% of an institution’s risk-weighted amount for credit risk (i.e. an indicative size reference level of 10%). Q5. Will the MA approve an AI to exclude from the IRB calculation only some exposures within a business unit? A5. No. As a general rule, the MA will only exempt exposures within a business unit from the IRB calculation in their entirety. When an AI uses the IRB approach in respect of a particular business unit, the institution should apply the IRB approach to all exposures within the IRB adoption classes approved to use the IRB approach that falls within that business unit. In this connection, when an AI applies for exempting exposures within a business unit, the institution should define the boundary of the concerned business unit in the application in a manner consistent with its business and management structure. Examples of a business unit may include a subsidiary, a branch outside Hong Kong, or a division of an AI. Q6. Is an exemption granted by the MA under §12 a permanent one?

Chapter IV IRB approach Page 4 of 37
A6. No. When the exemption granted in respect of a subset of exposures in an IRB adoption class or the exposures falling within a business unit under §12 becomes unsubstantiated (see A4 above for the factors relevant to the determination of a §12 application), the MA may take one or more of the actions set out in §13(2), including revocation of the exemption granted.
2. Classification of exposures
General
Q7. The reported figures of annual sales, annual revenue and total assets of a company or a group of companies of which a company concerned is a member (a "group") in its "latest annual financial statement" are required for classification and credit risk calculation under the IRB approach (see for example, §§143(3) and (3A) and 157A(3)). Must an AI use the audited financial statements to determine the value thresholds? In addition, companies often take time to prepare financial statements. What is the supervisory expectation regarding their timeliness? A7. Under §143(3A), the annual financial figures of a company or a group in a financial year (“financial figures”) must be ascertained through the audited financial report for the classification of large corporate exposures. The same applies to identifying large regulated financial institutions defined under §157A(3). Regarding the classification of small-and-medium sized corporate exposures and small business retail exposures, the HKMA has the same expectation as that provided in A28 to A31 of Chapter III STC approach for the classification of “small business” under the STC approach. Therefore, among others, an AI should ascertain the financial figures through the audited financial report to the extent feasible. Where the updated audited financial report is unavailable (e.g. sole proprietorships where statutory audit is not mandatory) or yet to be available, other reliable information for that year (e.g. the company’s internal report or transaction/payment data or those obtained through Commercial Data Interchange) may be used with any necessary adjustments (e.g. elimination of intra-group transactions). Q8. Must an AI use the $500 million threshold to classify its corporate exposures under the IRB subclass of small-and-medium sized corporates? What would be the case for the value thresholds for the classification of small business retail exposures and large corporates? A8. Under §143(3), an AI may classify an exposure to a corporate that falls within the description therein as a corporate exposure within the IRB subclass of small-and￾medium sized corporates. For this, the institution may opt for a threshold consistent with its risk management practice, which however must be below $500 million.

Chapter IV IRB approach Page 5 of 37
Similar flexibility is also available to the institution with respect to the $10 million threshold set out in §144(2) for classifying small business retail exposures. Regarding the $5 billion threshold for classifying an AI’s exposures under large corporates pursuant to §143(3A), the institution must adopt this value threshold. It is also worth noting that under §143(4C), an AI must classify its exposures that fall within the description in §143(3C) as exposures falling within the IRB subclass of financial institutions treated as corporates regardless these exposures may also fall within the IRB subclass of small-and-medium sized corporates or large corporates. Corporate exposures – specialized lending Q9. Under which IRB subclass should an AI classify its specialized lending if the institution is able to estimate the credit risk components of such specialized lending for the purpose of using the foundation IRB approach or the advanced IRB approach? A9. An AI must classify all of its specialized lending under the five IRB subclasses of the specialized lending (project finance / object finance / commodities finance / income￾producing real estate / high-volatility commercial real estate), as the case requires, regardless of whether the institution uses the supervisory slotting criteria approach, the foundation IRB approach or the advanced IRB approach to calculate the risk￾weighted amount of its specialized lending. Q10. Is there any guidance on classifying exposures into each subclass of specialized lending in addition to paragraphs (a) to (e) of §143(1)? A10. The definition of “specialized lending” and the definitions of the five subclasses of specialized lending, namely project finance, object finance, commodities finance, income-producing real estate and high-volatility commercial real estate are set out in §139(1) and paragraphs (a) to (e) of §143(1) respectively. For further guidance, an AI should refer to (i) Q34 to Q40 in Chapter III STC approach and (ii) Basel Framework CRE30.8 to CRE30.16. Corporate exposures – small-and-medium sized corporates Q11. What circumstances are expected to be justifiable for an AI to substitute a corporate’s total assets for total annual sales for §143(4) concerning the classification of small-and-medium sized corporates? A11. These would be the cases where a corporate’s scale of business is not accurately reflected by the corporate’s total annual sales, such as the corporate’s annual sales including a significant amount generated by an exceptional, non-recurring transaction, or includes a significant amount of off-shore sales which has been booked through the corporate for tax planning purposes.

Chapter IV IRB approach Page 6 of 37
Corporate exposures – financial institutions treated as corporates Q12. Guidance is sought on the scope of exposures to be classified under the IRB subclass of financial institutions treated as corporates. A12. Any financial institutions as defined under §139(1) that do not fall within the five IRB subclasses of bank exposures (viz. banks (excluding covered bonds), qualifying non-bank financial institutions, public sector entities (excluding sovereign foreign public sector entities), unspecified multilateral bodies and covered bonds) should be classified under the IRB subclass of financial institutions treated as corporates. 2 Retail exposures Q13. Is there any limit on the size of an exposure qualified as a retail exposure? A13. Size limits are specified in §144(2) for the IRB subclass of small business retail exposures and §144(4) and (4A) for the IRB subclasses of qualifying revolving retail exposures (transactor) or qualifying revolving retail exposures (revolver). There is no size limit, however, for an exposure to qualify for inclusion in the other IRB subclasses for retail exposures. Q14. Where a retail borrower has a credit card facility and a revolving personal overdraft facility with an AI, does the institution need to add up both facilities to determine whether the aggregate amount of the two facilities is within the limit of $1 million as specified in §144(4) and/or (4A) such that both facilities can be classified under the IRB subclass of qualifying revolving retail exposures (transactor) or qualifying revolving retail exposures (revolver)? A14. No. In determining whether an exposure falls within the IRB subclass of qualifying revolving retail exposures (transactor) 3 or qualifying revolving retail exposures (revolver), an AI is only expected to apply the $1 million limit mentioned above on a facility basis, rather than on an obligor basis. To avoid doubt, the limit is intended to be applied at the facility level from the perspective of management, i.e. the exposures of several facilities of the same product type should be aggregated if the concerned facilities are managed as if they were one facility. Otherwise, it is acceptable to apply the limit to them individually. An indication of facilities being managed as one account, among others, is that they are subject to a shared limit. To check against the limit, the total exposure at the time of classification (e.g. the drawn portion of a performing credit line) could be used. In practice, the facility limit or the EAD of an exposure, according to an AI’s established policies, may be used as

2 Although such a case is not envisaged, if an exposure to a financial institution meets the criteria of specialized lending, it should be classified under one of the five specialized lending exposures. 3 Please refer to Q54 to Q57 of Chapter III STC approach for further guidance on identifying a transactor. The term carries the same meaning under the STC approach and the IRB approach.

Chapter IV IRB approach Page 7 of 37
long as the conditions under §144(2) and (4A) are met.
Q15. How should an AI classify a loan to an individual which is not managed on a pooled or portfolio basis? A15. As required under §144(6), such loans should be treated as corporate exposures. For an exposure to be eligible to be classified as a retail exposure, it must not be managed individually in a way comparable to corporate exposures, but rather as part of a portfolio segment or pool of exposures with similar risk characteristics for purposes of risk assessment and quantification. The requirement for exposures managed on a pooled or portfolio basis does not preclude retail exposures from being treated individually at some stages of the risk management process. The fact that an exposure is rated individually does not in itself preclude it from being eligible as a retail exposure. Other exposures Q16. Is gold bullion held on an unallocated basis for the institution by another person which is backed by gold bullion liabilities a cash item? A16. The item is required to be treated as an exposure to a counterparty and risk-weighted under the IRB approach accordingly.
3. IRB calculation approaches
Q17. Is there any expectation that an AI currently using the foundation IRB approach to calculate its credit risk for an IRB adoption class of corporate and sovereign exposures will migrate to the advanced IRB approach over time? A17. No, there is no such a supervisory expectation. Each AI should select an approach for an IRB adoption class which is appropriate for its exposures within that IRB adoption class and commensurate with the sophistication of its internal risk management functions. As such, an AI currently using the foundation IRB approach to calculate its credit risk for an IRB adoption class of corporate and sovereign exposures can choose to remain on that approach. Q18. Can an AI using the advanced IRB approach to calculate its credit risk for corporate and sovereign exposures switch to the foundation IRB approach? What if an AI applies to use the advanced IRB approach for some of its exposures and use the foundation IRB approach for the remaining exposures within an IRB adoption class? A18. Except for the cases specified in §147(3A) concerning the transitional arrangements, any switching of IRB calculation approach (regardless of whether the return involves

Chapter IV IRB approach Page 8 of 37
all exposures or only a portion of the exposures within an IRB class) requires the prior consent of the MA under §147(3). The switching from the advanced IRB approach to the foundation IRB approach for exposures within an IRB adoption class will be permitted only in exceptional circumstances (e.g. where an AI’s business has been downsized to a level which does not justify the institution maintaining a highly sophisticated risk management system, or an AI’s rating system is no longer able to reliably estimate one or more than one of the credit risk components 4 ). Q19. Can an AI choose to use the supervisory slotting criteria approach to calculate the risk-weighted amount of its specialized lending even though the institution is able to estimate the credit risk components of such lending as required in Part 6 of the BCR for corporate exposures? A19. No. According to §143(2), an AI can only use the supervisory slotting criteria approach to calculate the risk-weighted amount of its specialized lending when the institution is not able to estimate the credit risk components of such lending (by each of the five IRB subclasses) for the use of the foundation IRB approach or the advanced IRB approach. Q20. After 1 January 2025, which calculation approach should an AI use to calculate the credit risk of its equity exposures previously granted with the MA's approval to use the IRB approach? A20. Under §147(3C), if an AI used the IRB approach before 1 January 2025 to calculate its credit risk for equity exposures (within the meaning of the pre-amended Part 6 of the BCR), the institution must use the STC approach on and after that date to calculate its credit risk for equity exposures. It is worth noting that under §139(1) after the amendment, equity exposure means an exposure that falls within §54A, which excludes, among others, an exposure that is a CIS exposure.
4. Default of obligor
Q21. What is meant by “prescribed default criteria” in §149(9)?
A21. Under §149(9), “prescribed default criteria” means the criteria specified in §149(1). That section primarily sets out a default of the obligor in respect of an exposure of an AI has occurred if— (a) the institution considers that the obligor is unlikely to pay in full the obligor’s credit obligations to the institution (or to any member of the consolidation group of the institution) ("banking group") without recourse by the institution to realizing any collateral held by the institution or taking any other action in respect of the exposure; or

4 This may arise from the requirements from the Basel Committee or the home authority of an AI (in case where the institution is the subsidiary of a non-Hong Kong banking group) in estimation practices of credit risk components.

Chapter IV IRB approach Page 9 of 37
(b) subject to §149(2), (3) and (8), the obligor is past due for more than 90 days in respect of the payment of any material portion of all the obligor’s outstanding credit obligations to the banking group. The operation of other key provisions of §149 are outlined below for ease of reference. §149(1A) Presents a list of possible indications of “unlikely to pay” specified in §149(1). §149(2) Specifies that an AI may choose to apply the prescribed default criteria at the level of a particular exposure, rather than at the level of the obligor for retail exposures. As such, the default of a borrower on one obligation does not necessarily require an AI to treat all other obligations to the banking group as defaulted. In this connection, AIs are advised to set out, in their internal policies, under what circumstances all their exposures to the same obligor, including the retail exposure in question, would be treated as in default, and apply the policy consistently. §149(3) Clarifies the circumstances under which an overdraft provided by an AI to an obligor (being a borrower of the overdraft) should be regarded as past due. §149(4) and (5) Elaborate that an AI shall use the prescribed default criteria for its exposures under the IRB approach except for specific exposures where the institution is given with the MA's consent to use another set of default criteria set by the relevant banking supervisory authority of the parent bank of an institution, which is the local subsidiary of the parent bank ("alternative criteria"). It is currently envisaged that the use of alternative criteria is only justifiable for retail exposures and exposures to public sector entities, and the past due trigger within the alternative criteria is no more than 180 days. §149(5A), (5B), (5C) & (5D) Concern the default of obligors in a connected group:
 (5A) Requires an AI to treat its exposures to all individual obligors in a connected group as being in default if— (a) a default of an obligor (“defaulting obligor”) in the connected group has occurred; and (b) the defaulting obligor has been rated substantially on the basis of the economic or financial interdependence between the members in the connected group in accordance with the institution’s policy and practices referred to in §154(d).  (5B) Provides flexibility in relation to retail exposures resembling §149(2) in the context of the default of a connected group.  (5C) & (5D) Specify the circumstances under which an AI may disregard §149(5A).

Chapter IV IRB approach Page 10 of 37
§149(6) &
(7)
Set out the requirements for the keeping of records of default, the generation of estimates of credit risk components, and the use of internal or external data in relation to the definition of default while §149(8) reaffirms that the practice of re-ageing is not allowed. Q22. Guidance is sought on the “material” threshold in relation to the requirements set out in §149(1)(b), (1A)(a) and (c). A22. AIs are expected to develop their own criteria in determining the materiality of a credit obligation and the materiality of credit-related economic loss for the purposes of §149(1)(b), (1A)(a) and (c). The criteria should be prudent and applied consistently within the consolidation group of an institution and should not jeopardize its internal policies and procedures for problem credit management. If an AI’s parent bank is incorporated outside Hong Kong and subject to capital standards and/or supervisory guidance published by the parent bank’s regulator that have specified levels of materiality thresholds or other criteria for determining materiality, the institution’s own criteria for determining materiality should not be less prudent than those materiality thresholds and other criteria, unless otherwise justified. To avoid doubt, it is acceptable for an AI to ignore the word “material” in those paragraphs of §149(1) and (1A). Furthermore, an AI may, at its discretion, apply the flexibility for identifying defaults with respect to a group of exposures covering retail exposures of an obligor (or a connected group) set out in §149(2)(b) and (5B) in considering the relevant amount under its “materiality” criteria. Q23. Is an AI required to treat an “automatic” realization of an obligor’s collateral in respect of certain facility types (e.g. share margin financing) as an event of default where the realization of collateral is not due to the deterioration in the obligor’s creditworthiness but to a fall in the value of the collateral? A23. The definition of "default" may not apply in cases where the realization of collateral is not triggered by deterioration in an obligor’s creditworthiness but by a fall in the value of the obligor’s collateral (say, the shares pledged). In such cases, an AI will not be required to record a default of the obligor if the following two characteristics exist— (a) the facility is granted to finance the obligor’s position in a financial instrument which qualifies as recognized financial collateral under the IRB approach; and (b) the collateral is realized to restore an agreed collateral coverage ratio after a fall in the value of the obligor’s collateral, as a standard practice for such type of facility and where such practice has been disclosed to the obligor in writing at the inception of the facility. Q24. What is meant by a “connected group” as referred to in §149(5A), (5B), (5C) and (5D)? A24. The term “connected group” in these subsections should reflect the definition used by an AI for the purposes of §154(d)(ii). Where an
AI adopts a “group support” policy in rating assignment in accordance with §154(c) and (d), the institution is

Chapter IV IRB approach Page 11 of 37
required to determine and define (among other things) what constitutes a “connected group” of its obligors in that context. Please refer to A32 under the subject “Rating system design and operations” below for further guidance. Q25. Why is an AI required to treat its exposures to all individual obligors in a connected group as being in default in the circumstances described in §149(5A)? Are there exceptions to the rule? A25. To the extent that members of a connected group are treated on a group basis by an AI for the purposes of rating assignment pursuant to §154(c) and (d) and have, as a result, been assigned more favourable ratings (based on the available group support) than if they were rated on a standalone basis, it is prudent and logical that such group members be treated consistently on the same group basis for the purposes of the recognition of default within the group as provided for under §149(5A). Accordingly, AIs that adopt a group support policy in rating assignment should accept both:
(a) the benefit of more favourable ratings being assigned to members of a connected group on the strength of available group support pursuant to §154(c) and (d); and (b) the adverse impact on members’ ratings when §149(5A) becomes applicable. It would amount to cherry-picking if AIs were initially allowed to rate members of a connected group favourably on a group basis when there is no default among the members but, once the group support so recognized actually fails to prevent the default of a group member, were subsequently allowed to revert to rating other group members on a standalone basis. To do so would essentially ignore the interdependencies between the group members that had been recognized and relied upon pre-default. Recognizing however that the form and structure of conglomerates vary widely, the MA does not intend to mandate AIs to automatically regard the default of any one member of a connected group as a default of all the group members in all circumstances, and has therefore set out in §149(5B), (5C) and (5D) the circumstances under which §149(5A) will not apply.
5. Rating system design and operations
General
Q26. 5 To what extent should material and relevant information on climate-related financial risks be used when assigning ratings to obligors and facilities? Relatedly, to what extent do the requirements for rating criteria and rating assignment require consideration of climate-related financial risks? A26. AIs are advised to take into consideration material and relevant information on the impact of climate-related financial risks in rating assignment. The range of economic conditions or unexpected events that should be considered when making the

5 Reference: FAQs inserted to Basel Framework CRE36.26, CRE36.30 and CRE36.86.

Chapter IV IRB approach Page 12 of 37
assessment of a borrower’s ability to perform should include climate-related financial risks, both physical and transition risks, if these materialize as credit risks. For further guidance, please refer to each individual FAQ1 under Basel Framework CRE36.26, CRE36.30 and CRE36.86. The FAQs on climate-related financial risks published by the Basel Committee are primarily intended to encourage banks to continuously develop their measurement and mitigation of climate-related financial risks (given the challenges arising from methodological and data limitations cannot be fully resolved at this time) and is not to introduce changes to the Pillar 1 standards. Thus, the HKMA will adopt a proportionate approach in assessing an AI’s compliance and consider it pragmatic for an AI to focus on enhancing its framework and process initially, and to gradually improve its robustness. 6 Q27. 7 How can AIs reflect climate-related financial risks in the supervisory slotting criteria for specialized lending? A27. When performing the assessment of the category of the sub-factor components, AIs should analyse how climate-related financial risks could negatively impact the assignment into a category. This includes any potential impact on, but not limited to— (a) the financial strength (e.g. estimations of the future demand, economic assumption and stressed economic conditions used for stress analysis), (b) the political and legal environment (e.g. transition risk into “stability of legal and regulatory environment (risk of change in law)”), (c) physical risk into “force majeure risk (war, civil unrest, etc.)”, and (d) the asset characteristic in the case of object finance. AIs should also take into consideration whether climate-related financial risks have been adequately mitigated (e.g. through improving adaptation or taking insurance coverage against physical climate risks). Rating dimensions Q28. Is an AI required to assign to the same obligor grade, separate exposures which it has to the same obligor? A28. Yes, separate exposures to the same obligor should, in general, be assigned to the same obligor grade, irrespective of differences in the transaction-specific factors (e.g. collateral, seniority of repayment, tenor and product type) of those exposures. However, an AI may do otherwise if the institution demonstrates to the satisfaction of the MA that the risk of default of a particular obligor is different in respect of different exposures the institution has to that obligor (re: §150(3)(b)).

6 To avoid doubt, this paragraph is also applicable to Q27, Q39 and Q77. 7 Reference: FAQs inserted to Basel Framework CRE33.13.

Chapter IV IRB approach Page 13 of 37
Below are two typical examples where this might be the case— (a) To reflect country risk and transfer risk 8 , an AI may assign to different obligor grades, different exposures which it has to the same obligor, if some of the exposures are denominated in local currency and others are denominated in foreign currency. (b) Under the foundation or advanced IRB approach, an AI may reflect the credit risk mitigating effect of a recognized guarantee or recognized credit derivative contract in respect of an exposure through adjusting the PD of the obligor in respect of that exposure. Rating structure Q29. Is it possible for an AI to have more than one obligor grade or pool to which exposures to obligors who are in default can be assigned? A29. Yes, provided that the rating definitions and criteria of these obligor grades or pools are clear and specific. Q30. Under what circumstances will the MA regard an AI’s process for assigning its exposures to obligor grades, as leading to excessive concentration on a particular obligor grade? A30. Generally, if an AI’s process of assignment leads to more than 30% of its exposures being assigned to a particular obligor grade, this will be regarded as a sign of excessive concentration. Significant concentration on a particular obligor grade should be justified by convincing empirical evidence that the obligor grade concerned covers a reasonably narrow PD range and the default risk posed by all obligors in respect of exposures assigned to that grade falls within that PD range. Rating assignment horizon Q31. 9 What are the requirements concerning the drivers of migrations from one category of obligor grade to another? A31. As set out in §153(b), an AI shall ensure that the obligor grade to which an exposure is assigned accurately represents the institution’s assessment of the willingness and ability of an obligor in respect of the exposure to perform the obligor’s contractual obligations, after considering any potentially adverse economic conditions over an economic cycle within the industry or geographic region relevant to the obligor.

8 Country risk and transfer risk refer to the risks that the obligor may not be able to secure foreign currency to service its external credit obligations due to adverse changes in foreign exchange rates or when the jurisdiction in which it is operating suffers economic, political or social problems. 9 Reference: Basel Framework CRE36.30.

Chapter IV IRB approach Page 14 of 37
The range of economic conditions that are considered when making assessments must be consistent with current conditions and those that are likely to occur over a business cycle within the respective industry/geographic region. Rating systems should be designed in such a way that idiosyncratic or industry-specific changes are a driver of migrations from one category to another, and business cycle effects may also be a driver (re: Basel Framework CRE36.30). Rating coverage Q32. What are the key factors that an AI should consider if it wishes to apply the concept of “group support” in rating individual obligors in a connected group in accordance with §154(c) and (d)? A32. Some key considerations which AIs may take into account in formulating the policy referred to in §154(d) on the assignment of obligor grades to individual obligors in a connected group are set out below. 10

(a) AIs should define what constitutes a connected group for the purposes of §154(c) and (d), with strong justification and clear documentation of the grouping criteria. (b) AIs should establish and justify the “recognition” criteria for taking into account certain support provided by member(s) of a connected group, and the extent to which such support is reflected, in the determination of the obligor grades of individual obligors within the connected group, by assessing all relevant factors, which include, but are not limited to, the following— (i) the source, nature, form and availability of support to obligors in a connected group; (ii) the identification of, and justification for, those obligors within a connected group in respect of which the obligor grades will be adjusted to reflect the strength of support provided by the group; (iii) the extent to which the group support is actually available to individual obligors within the connected group; (iv) the willingness, ability and past behaviour of the support provider in honouring assurances to the relevant obligor or comparable commitments to similar beneficiaries, in both normal and stressed times; (v) any material specific wrong-way risk and interconnectedness between the obligor and the support provider; (vi) the potential obligations, whether contractual or not, of the “beneficiary” obligors in question to lend support, in turn, to other group members; and

10 Please note this list is not intended to be exhaustive, and AIs should take into account their specific circumstances in their effort to comply with §154(c) and (d).

Chapter IV IRB approach Page 15 of 37
(vii) the ability of, and the effectiveness with which, the AI is able to validate or benchmark its process, methodology and data for incorporating group support into the ratings of individual obligors in a connected group, and the resulting adjustments made to the stand-alone ratings of such obligors. (c) AIs may also draw reference to analogous requirements in the credit risk mitigation frameworks set out in the BCR (e.g. §77 and Division 10 of Part 6) or in SPM modules such as CR-G-7 on “Collateral and guarantees”. (d) In cases where the support provider and the beneficiary obligor fall under the purview of different regulators and/or are located in different jurisdictions, any cross-sector and cross-border restrictions and country risk (e.g. exchange controls, liquidity constraints, supervisory ring-fencing measures) that may hinder the availability of the support should be taken into account. (e) AIs should exercise prudence, conservatism and consistency in quantifying the extent of group support for the purposes of rating individual obligors, in order not to under-estimate the default risk arising from exposures to the individual obligors in the connected group. (f) There should not be any double-counting of the credit risk mitigating benefits incorporated into the internal ratings of obligors in a connected group pursuant to §154(c) and (d) and those recognized under the credit risk mitigation frameworks of the BCR. (g) As in the case of other established policies and rating systems of AIs, the group support framework should be subject to proper approval procedures, regular independent reviews and validations, and regular and timely updates. Integrity of rating process Q33. §155(a) requires that the rating process of an AI be “independent” of the staff and management responsible for originating such exposures. What are the factors for assessing whether this “independence” criterion is met? A33. The concept of “independence” is a fundamental risk management principle and is commonly deployed in numerous prudential supervisory standards and risk management literature. Consistent with its interpretation in a generic sense, the following scenarios would generally indicate that a member of the staff or management of an AI is “independent” of the credit origination process for an exposure for the purposes of §155(a):
(a) the person does not directly stand to benefit from the extension of credit (e.g. in the form of bonus or other type of monetary or non-monetary compensation the availability and size of which are primarily linked to the origination of credit exposures);

Chapter IV IRB approach Page 16 of 37
(b) the person is independent of the institution’s risk-taking functions, in terms of decision-making, reporting structure and resourcing (i.e. the risk-taking functions do not control the compensation package of the person concerned, or the budget or financing of the organizational unit to which that person belongs); and (c) the person is free from potential conflict of interest in relation to the credit origination process in general and the exposures being rated or reviewed in particular (e.g. that person is not a connected person (as defined by relevant regulatory and supervisory requirements applicable to the institution) in respect of the obligor of the exposure concerned). Q34. What is the supervisory expectation on the timeliness of rating review in response to new material information on an exposure in respect of audited financial statements of obligors (re: §155(c))? A34. In assessing whether an AI has timely incorporated new material financial information in respect of the audited financial statements of obligors into rating assignments under §155(c), the HKMA will consider, among others, the regulations and accounting standards in different jurisdictions, the utilization of information within a rating system, the characteristics of obligors, and an indicative benchmark of a 15-month time lag. Similar to the guidance provided in A7, an AI may supplement the updated audited financial report that is unavailable with other reliable information, subject to necessary adjustments. The HKMA considers that conservative adjustments to rating assignments could be used to compensate for the use of stale data, but these cannot replace an AI’s efforts to obtain new material information on exposures in order to ensure the integrity of its rating process as required by §155(c).
6. Estimation of credit risk components
General
Q35. 11 What are the regulatory requirements on the data for the estimation of credit risk components? A35. Internal estimates of PD, LGD, and EAD must incorporate all relevant, material and available data, information and methods. An AI may utilize internal data or data from external sources (including pooled data) or both provided— (a) the AI must demonstrate that its estimates are representative of long run experience (covering at least one economic cycle which captures a reasonable mix of high-default and low-default years) (re: §148(d)(i)).

11 Reference: Basel Framework CRE36.66, CRE36.79, CRE36.82, CRE36.87, CRE36.88, CRE36.98 and CRE36.99.

Chapter IV IRB approach Page 17 of 37
(b) the AI must also demonstrate that the economic or market conditions that underlie the data are relevant to current and foreseeable conditions (re:
§148(d)(ii)).
(c) the population of exposures represented in the data used for estimation, and 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 the institution’s exposures and standards. (d) the number of exposures in the sample and the data period used for quantification must be sufficient to provide the AI with confidence in the accuracy and robustness of its estimates. (e) the criteria for identifying default of obligor in respect of an exposure comply with §149. In addition, the data should be in compliance with the specific requirements set out in— (a) §159(1)(d) for the estimation of PD under the foundation IRB approach or the advanced IRB approach (see also A40); (b) §161(1)(e) for the estimation of LGD under the advanced IRB approach; (c) §164(4)(f) for the estimation of EAD under the advanced IRB approach (see also A48); (d) §177(1)(e) and (2) for the estimation of PD under the retail IRB approach (see also A40); (e) §178(1)(g) for the estimation of LGD under the retail IRB approach; and (f) §180(3)(b) for the estimation of EAD under the retail IRB approach. Further to meeting the specific requirements for the minimum data observation period, if the available observation period spans a longer period for any source, and the data are relevant and material, this longer period must be used. In addition, as a general principle, the less data an AI has, the more conservative it must be in its estimation of credit risk components. Q36. How can an AI deal with the problem of limited default data in estimating the credit risk components of a low-default portfolio (“LDP”)? A36. An LDP is a portfolio of exposures which, for whatever reason, has a relatively low number of defaults. In practice, the following portfolios may be regarded as LDPs— (a) portfolios that historically have experienced low numbers of defaults and are generally considered to be low risk (e.g. exposures to sovereigns, banks, insurance companies and highly rated corporates);

Chapter IV IRB approach Page 18 of 37
(b) portfolios that are relatively small in size either for the banking sector as a whole or at an individual bank level (e.g. project finance and shipping loans); (c) portfolios for which a bank is a recent market entrant; and (d) portfolios that have not incurred recent losses but historical experience or other analysis suggests that there is a greater likelihood of losses than is captured in recent data. Historical incidents, particularly the global financial crisis of 2007/2008 and the subsequent European sovereign debt crisis, illustrate that although an LDP may hitherto have been characterized by its low number of defaults, this does not necessarily mean that it can inevitably also be characterized as low risk. There are a number of data-enhancing techniques and statistical or benchmarking tools an AI may wish to use in order to increase the reliability of the credit risk components relating to exposures falling within an LDP. For further details, see section 10 of the SPM module CA-G-4 on “Validating Risk Rating Systems under the IRB Approach”, and the explanatory guidance entitled “Validation of low-default portfolios in the Basel II Framework” issued by the Basel Committee in its Newsletter No. 6 (September 2005). Q37. 12 What shall an AI do in respect of the estimation of credit risk components if changes were made in lending practice or the process for pursuing recoveries over the observation period? A37. §148(c) sets out that an AI shall, among others, take into account all relevant data and information available to estimate credit risk components. Accordingly, the institution should consider any changes in lending practice or the process for pursuing recoveries over the observation period to make estimates of PD and, where relevant, LGD and EAD (re: Basel Framework CRE36.65). Q38. Generally, the estimation of PD, LGD and EAD requires the data source, among other things, to cover at least one economic cycle (see A35 above). What does an “economic cycle” mean? A38. There is no universally accepted definition of the term “economic cycle”. However, an “economic cycle” typically consists of a sequence of 4 distinct phases as described below— (a) economic downturn (or economic contraction) which depicts a trend of slowdown in the level of economic activity in terms of real GDP and other macroeconomic variables; (b) economic trough which describes the bottom of an economic cycle where an economic downturn turns into an economic upturn; (c) economic upturn (or economic expansion) which depicts a trend of

12 Reference: Basel Framework CRE36.65.

Chapter IV IRB approach Page 19 of 37
acceleration in the level of economic activity in terms of real GDP and other macroeconomic variables; and (d) economic peak which describes the peak of an economic cycle where an economic upturn turns into an economic downturn. Q39. 13 Should AIs add a margin of conservatism to estimates of credit risk components to account for the fact that historical data are less satisfactory to capture climate-related financial risks, increasing the likely range of errors? A39. When an AI’s credit portfolio is materially exposed to climate-related financial risks, the institution should strive primarily to consider these risks directly in its estimates. An AI should add a margin of conservatism due to data deficiencies, such as poor data quality or scarce climate-related data, and to other sources of additional uncertainties. For further guidance, please refer to FAQ1 under Basel Framework CRE36.67 and FAQ2 under Basel Framework CRE36.86. Probability of default (“PD”) Q40. 14 How shall an AI take appropriate account of the long run experience when estimating the average PD for each rating grade for its corporate, sovereign and bank exposures? A40. AIs shall use information, data and techniques that take appropriate account of the long run experience when estimating the average PD for each rating grade (re:
§159(2)(a)). AIs may use a primary technique and others as a point of comparison and potential adjustment, provided that the institution acts prudently in the comparison and adjustments, but mechanical application of a technique without supporting analysis is not acceptable (re: §159(2)(b)). AIs must recognize the importance of judgmental considerations in combining the results of techniques and in making adjustments for limitations of techniques and information. For example, it is understood that AIs may use one or more of the three specific techniques set out below: internal default experience, mapping to external data, and statistical default models. For all of them, institutions must estimate a PD for each rating grade based on the observed historical average one-year default rate that is a simple average based on the number of obligors (count weighted). Weighting approaches, such as EAD weighting, are not permitted. (a) An AI may use data on internal default experience for the estimation of PD. In this connection, the institution— (i) must demonstrate in its analysis that the estimates are reflective of underwriting standards and of any differences in the rating system that generated the data and the current rating system;

13 Reference: FAQs inserted to Basel Framework CRE36.67 and CRE36.86. 14 Reference: Basel Framework CRE36.78.

Chapter IV IRB approach Page 20 of 37
(ii) where only limited data are available, or where underwriting standards or rating systems have changed, must add a greater margin of conservatism in its estimate of PD;

(iii) in the case of using pooled data across institutions, must demonstrate that the internal rating systems and criteria of other institutions in the pool are comparable with its own. (b) AIs may associate or map their internal grades to the scale used by an external credit assessment institution or similar institution (“external institutions”) and then attribute the default rate observed for the external institution's grades to its grades. In this connection— (i) Mappings must be based on a comparison of internal rating criteria to the criteria used by the external institutions and on a comparison of the internal and external ratings of any common borrowers. (ii) Biases or inconsistencies in the mapping approach or underlying data must be avoided. (iii) External institutions’ criteria underlying the data used for quantification must be oriented to the risk of the borrower and not reflect transaction characteristics. (iv) The AI’s analysis must include a comparison of the default definitions used, subject to the requirements in §149. (v) The AI must document the basis for the mapping. (vi) The AI should consider whether the scale used by the external institutions reflects material climate-related financial risks. For details, please refer to FAQ1 under Basel framework CRE36.78. (c) An AI is allowed to use a simple average of default-probability estimates for individual borrowers in a given grade, where such estimates are drawn from statistical default prediction models. Q41. 15 What are the requirements on the length of the underlying historical observation period for estimating PD for corporate, sovereign, bank and retail exposures? A41. For PD estimation, irrespective of whether an AI is using external, internal, or pooled data sources, or a combination of the three, the length of the underlying historical observation period used must be at least five years for at least one source of data. The data should include a representative mix of good and bad years of the economic cycle relevant to the respective exposures (re: §159(1)(d) and §177(1)(e)). If the available observation period spans a longer period for any source, and this data are relevant and

15 Reference: Basel Framework CRE36.79 and CRE36.82.

Chapter IV IRB approach Page 21 of 37
material, the longer period must be used. Furthermore, the estimation of PD for retail exposures should be based on the observed historical average one-year default rate (re: §177(1)(ab)). Loss given default (“LGD”) Q42. 16 What is the definition of loss used in estimating LGD? A42. The definition of loss used in estimating LGD is economic loss. When measuring economic loss, all major factors should be taken into account, including the time value of money, the risk premium, and any direct and indirect costs associated with collection (re: §161(2)(b) and §178(2)(b)). AIs must not simply measure the loss recorded in accounting records, although they must be able to compare accounting and economic losses. The institution’s own workout and collection expertise significantly influences their recovery rates and must be reflected in their LGD estimates, but adjustments to estimates for such expertise must be conservative until the institution has sufficient internal empirical evidence of the impact of its expertise. Q43. 17 How can an AI ensure that the LGD estimate of each of its facility types (in the case of §161(1)(a)), or the LGD estimate of each pool of its retail exposures (in the case of §178(1)(a)) reflects economic downturn conditions? A43. An AI must 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. For certain types of exposures, loss severities may not exhibit such cyclical variability and LGD estimates may not differ materially from the long run default-weighted average. However, for other exposures, this cyclical variability in loss severities may be important and AIs will need to incorporate it into their LGD estimates. For this purpose, AIs may make reference to the 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. In addition, AIs should adhere to the principles set out in the explanatory guidance entitled “Guidance on Paragraph 468 of the Framework Document” (July 2005) issued by the Basel Committee in the process of identifying economic downturn conditions for incorporation into their LGD estimates where appropriate. These principles, as they relate to AIs, include the following— (a) an AI must have a rigorous and well-documented process for assessing the

16 Reference: Basel Framework CRE36.76.
17 Reference: Basel Framework CRE36.83.

Chapter IV IRB approach Page 22 of 37
effects, if any, of economic downturn conditions on recovery rates 18 and for producing LGD estimates consistent with these conditions. The process must consist of the following integrated components – (i) the identification of appropriate downturn conditions in each jurisdiction to which the institution’s recovery rates in respect of exposures within a particular IRB class are sensitive; (ii) the identification of adverse dependencies, if any, between default rates and recovery rates; and (iii) the incorporation of adverse dependencies, if identified, between default rates and recovery rates so as to produce LGD parameters for the institution’s exposures consistent with identified downturn conditions; (b) in discounting the cash flows used in LGD estimation, the measurement of recovery rates should reflect the cost of holding defaulted exposures over the workout period, including an appropriate risk premium; and (c) an AI should provide the HKMA with the long run default-weighted average loss rate given default for every relevant facility type unless the institution can demonstrate to the HKMA that:
(i) its estimate of loss rate given default under downturn conditions is consistent with items (a) and (b) above; and (ii) reporting a separate estimate of long run default-weighted average loss rate given default would not be practical. Q44. 19 Guidance is sought on using collateral’s estimated market value in LGD estimation. A44. The LGD estimates must be based on historical recovery rates of exposures but not solely be based on the estimated market value of collateral (re: §161(1)(c) and §178(1)(e)). This requirement recognizes the potential inability of AIs to gain both control of their collateral and liquidate it expeditiously. To the extent that LGD estimates take into account the existence of collateral, institutions must establish internal requirements for collateral management, operational procedures, legal certainty and risk management process that are generally consistent with those required for the foundation IRB approach. Q45. Are the LGD floor values in Table 19A intended to apply commonly to all recognized collaterals under the advanced IRB approach, disregarding whether they are recognized financial collateral and recognized IRB collateral? If any

18 Recovery rate means, for a defaulted exposure, the present discounted value at the default date of recoveries received net of material direct and indirect costs associated with collecting on the exposure, divided by the EAD of the exposure. 19 Reference: Basel Framework CRE36.85.

Chapter IV IRB approach Page 23 of 37
recognized collateral cannot be mapped to any of the four types of collateral in the table, what is the LGD floor value to be used for the purposes of §161? A45. In accordance with §139(1), a recognized collateral under the advanced IRB approach means any collateral which— (a) is recognized by an AI for credit risk mitigation in accordance with its policies and procedures, and (b) satisfies the requirements under §77(2). For the purposes of §161, an AI may map a recognized collateral that secures an exposure under the advanced IRB approach to the LGD floors by types of recognized collateral in Table 19A of §161. In this connection, an AI should have comprehensive policies and procedures in place on collateral recognition for capital calculation. Those recognized collaterals that an AI fails to map to the categories in Table 19A should be excluded from the LGD floor determination. Q46. 20 Guidance is sought on the LGD estimate assigned to a defaulted exposure. A46. Recognizing the principle that realized losses can at times systematically exceed expected levels, the LGD assigned to a defaulted exposure should reflect the possibility that an AI would have to recognize additional, unexpected losses during the recovery period (re: §161(1)(d) and §178(1)(f)). For each defaulted exposure, the AI must 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 exposure exceeds the AI’s best estimate of expected loss on the exposure represents the capital requirement for that exposure, and should be set by the institution on a risk-sensitive basis in accordance with §156(4) and §176(5) as appropriate. 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 must be justified. The details and the justification should be well documented for supervisory scrutiny upon request. Exposure at default (“EAD”) Q47. 21 Please give some examples of how EAD estimation is appropriate for an economic downturn referred to in §164(4)(c) apart from basing the estimate of the EAD on alternative measures of central tendency or only on the economic downturn data with specific requirements set out in §164(4)(ca)? A47. For AIs 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 institutions may have sufficient internal data to examine the impact of previous

20 Reference: Basel Framework CRE36.86.
21 Reference: Basel Framework CRE36.90.

Chapter IV IRB approach Page 24 of 37
recession(s). However, some institutions may only have the option of making conservative use of external data. Q48. 22 Guidance is sought on the reference data set for EAD estimation. A48. AIs are required to observe the below requirements concerning the reference data set for EAD estimation. (a) Fixed-horizon approach. AIs’ EAD estimates must be developed using a 12- month fixed-horizon approach, i.e. for each observation in the reference data set, default outcomes must be linked to relevant obligor and facility characteristics twelve months prior to default. (b) Homogeneity of reference data. AIs’ EAD estimates should be based on reference data that reflect the obligor, facility and bank management practice characteristics of the exposures to which the estimates are applied. Consistent with this principle, EAD estimates applied to particular exposures should not be based on data that comingle the effects of disparate characteristics or data from exposures that exhibit different characteristics (e.g. same broad product grouping but different customers that are managed differently by the institution). The estimates should be based on appropriately homogenous segments. Alternatively, the estimates should be based on an estimation approach that effectively disentangles the impact of the different characteristics exhibited within the relevant dataset. In this connection, Basel Framework CRE36.94 provides practices that generally do not comply with this principle. (c) Elements to be included. EAD reference data must not be capped to the principal amount outstanding or facility limits. Accrued interest, other due payments and limit excesses should be included in EAD reference data. (d) Concerning counterparty credit risk exposures. For transactions that expose AIs to counterparty credit risk, estimates of EAD must fulfil the requirements concerning counterparty credit risk set forth in Part 6A of the BCR and the relevant regulatory guidance. Q49. 23 Guidance is sought on the region of instability associated with facilities close to being fully drawn at the reference date when an AI estimates credit conversion factors (“CCFs”) with undrawn limit factor (“ULF”) approach. A49. A well-known feature of the commonly used ULF approach in estimating CCFs is the region of instability associated with facilities close to being fully drawn at reference date. AIs should ensure their EAD estimates are effectively quarantined from the potential effects of the region of instability by making reference to the practical guidance set out in Basel Framework CRE36.95.

22 Reference: Basel Framework CRE36.93, CRE36.94, CRE36.96 and CRE36.97. 23 Reference: Basel Framework CRE36.95.

Chapter IV IRB approach Page 25 of 37
Maturity under foundation IRB approach / advanced IRB approach Q50. §167(1)(c) and (2) allow an AI that uses the foundation IRB approach to give written notice to the MA within 7 days after commencing to calculate the maturity (“M”) of the institution’s corporate, sovereign and bank exposures in accordance with §168. What requirements should an AI follow for the purpose of this arrangement? A50. To strike a balance among flexibility, operational complexity and potential regulatory arbitrage, AIs are expected to— (a) switch the maturity treatment of all their corporate, sovereign and bank exposures under the foundation IRB approach to the advanced IRB approach, and such change should not be effected by phases (i.e. partial adoption is not allowed). 24 Furthermore, the maturity treatment under §168 should be adopted for any subsequent applications for switching the calculation approach of other corporate, sovereign and bank exposures from the STC approach to the foundation IRB approach; (b) establish rigorous internal processes and systems to capture the relevant data and calculate the maturity of exposures under the advanced IRB approach. Institutions should put in place adequate controls and monitoring to ensure the reliability and accuracy of the M used in regulatory capital calculation (see sections 4 to 6 of the SPM module CA-G-4 on “Validating Risk Rating Systems under the IRB Approach”; and (c) subject the related processes and systems to adequate assessment by a competent independent party with proper documentation for review by a third party. The “independent party” and the “third party” can be managed by the departments or units within an institution, provided that these parties are independent from developing the rating systems and related processes in the determination of M. Calculation of risk-weighted amount in respect of purchased receivables Q51. Are there any special considerations relevant to the use by AIs of the top-down approach to estimate PD and LGD (or, if applicable, EL) for the calculation of the risk-weighted amount for default risk of purchased receivables (as referred to in §§198 and 200)?

24 As a transitional arrangement, an AI that— (i) had obtained the MA's consent under the pre-amended §167(c) before 1 January 2025 to calculate the M under §168 for specific, but not all, corporate, sovereign or bank exposures, and (ii) encounters genuine difficulties in calculating the maturity for all relevant exposures in accordance with §168 by the commencement of the revised §167, may continue the prevailing calculation until the date set out in its plan to expand the treatment under §168 to all relevant corporate, sovereign or bank exposures agreed by the MA. Also note that, regardless of whether the previous consent covers all or specific corporate, sovereign or bank exposures, an AI must give written notice to the MA in the manner specified in §167(2), i.e. the previous consent has no bearing on the notification requirement.

Chapter IV IRB approach Page 26 of 37
A51. For the purposes of using the top-down approach to calculate the risk-weighted amount for default risk of purchased receivables, AIs are expected to be, and should ensure that they are operationally capable of managing various risks associated with the pool of purchased receivables and their advances against those receivables, as described in Basel Framework CRE36.114 to CRE36.121. The requisite systems, policies and controls are, in many aspects, akin to those applicable to the recognition of financial receivables for credit risk mitigation purposes under the IRB approach set out in §205(1) (re: Basel Framework CRE36.134 to CRE36.142), or are reflective of general credit risk management principles set out in the HKMA’s supervisory guidelines. The overarching objective is to ensure that AIs’ use of the top-down approach is supported by prudent risk management of the purchased receivables designed to safeguard their claims on those receivables from potential loss. Key elements of systems, policies and controls relevant to the risk management of purchased receivables are:
(a) Legal certainty: to ensure that, through the proper structuring of the contractual terms of the relevant facility and through the verification of payments where applicable, there is effective ownership and control of the purchased receivables and the associated cash receipts or remittances, including in cases where the seller or servicer of the receivables is in financial distress or bankruptcy (re: Basel Framework CRE36.116); (b) Effective monitoring and work-out systems: including measures to ensure the effective monitoring of both the quality of the purchased receivables and the financial condition of the relevant sellers and servicers. These would cover: (i) assessment of correlation between these two factors and safeguards against related contingencies; (ii) assessment of eligibility of the sellers and servicers and their credit risk management and collection systems; (iii) assessment and monitoring of the risk characteristics (including concentration risk) of the receivables; (iv) monitoring compliance with established policies, procedures and limits in respect of exposures to receivables; (v) monitoring and handling of problem credits; and (vi) related management reporting and documentation requirements (re: Basel Framework CRE36.117 to CRE36.118); (c) Effective controls over purchased receivables, credit availability and cash:
including having clear and effective policies and procedures to govern key aspects of the receivables purchase programme (“RPP”), including collateral requirements and controls, advancement of funds and receipt of cash (re: Basel Framework CRE36.119); and (d) Compliance with internal policies and procedures: including an effective internal process to assess compliance with critical policies and procedures through: (i) regular internal and/or external audits of all critical phases of the RPP; (ii) verification of separation of duties between business and risk management functions; and (iii) adequacy of back-office operations (re: Basel Framework CRE36.120 to CRE36.121). Credit risk mitigation

Chapter IV IRB approach Page 27 of 37
Q52. 25 What are supervisory expectations on the monitoring process of recognized financial receivables referred to in §205(1)(g)? A52. An AI must maintain a continuous monitoring process that is appropriate for the specific exposures (either immediate or contingent) attributable to the collateral to be utilized as a risk mitigant. This process may include, as appropriate and relevant, ageing reports, control of trade documents, borrowing base certificates, frequent audits of collateral, confirmation of accounts, control of the proceeds of accounts paid, analyses of dilution (credits given by the borrower to the issuers) and regular financial analysis of both the borrower and the issuers of the receivables, especially in the case when a small number of large-sized receivables are taken as collateral. Observance of the AI’s overall concentration limits should be monitored. Additionally, compliance with loan covenants, environmental restrictions, and other legal requirements should be reviewed on a regular basis. Q53. 26 Guidance is sought on the recognition of subsequent liens on, or subsequent charge over, the property collateral referred to in §206(c)(ii). A53. Subsequent liens or subsequent charges (“junior liens”) may be taken into account where there is no doubt that the claim for collateral is legally enforceable and constitutes an efficient credit risk mitigant. Where junior liens are recognized, an AI must first take the haircut value of the collateral, then reduce it by the sum of all exposures with liens that rank higher than the junior lien, the remaining value is the collateral that supports the exposure with the junior lien. In cases where liens are held by third parties that rank pari passu with the lien of the institution, only the proportion of the collateral (after the application of haircuts and reductions due to the value of exposures with liens that rank higher than the lien of the institution) that is attributable to the institution may be recognized. Q54. 27 Guidance is sought on the operational requirements to constitute commercial or residential real estate as recognized commercial real estate or recognized residential real estate. A54. Apart from observing paragraphs (a) to (l) of §206, AIs are expected to monitor on an ongoing basis the extent of any permissible prior claims (e.g. tax) on the property and appropriately monitor the risk of environmental liability arising in respect of the collateral, such as the presence of toxic material on a property. The assessment of the risk of environmental liability should have been a part of AIs’ collateral management, where a risk-based approach to the assessment and monitoring of it is deemed acceptable. Specifically, an AI’s collateral management system should specify possible follow-up actions for negative monitoring results, and the removal of a property’s eligibility as collateral for capital adequacy purposes should be one of the options where warranted. Such monitoring may
be carried out

25 Reference: Basel Framework CRE36.140.
26 Reference: Basel Framework CRE36.131(4).
27 Reference: Basel Framework CRE36.131 and CRE36.132.

Chapter IV IRB approach Page 28 of 37
on a best efforts basis, but should not rely on any single tool such as negative news checking. Furthermore, for the purposes of §206(f), among other measures, an AI should ensure any claim on collateral is properly filed on a timely basis. (re: Basel Framework CRE36.131(1), CRE36.132(3) and (4)). Q55. 28 Guidance is sought on the recognition of physical collateral for an exposure secured by a general security agreement (or an equivalent form of floating charges) under the foundation IRB approach. A55. Where an AI’s exposure is secured by a general security agreement (or an equivalent form of floating charge) over both recognized collateral and other types of collateral, the institution may only recognize the security interest over recognized collateral. Such recognition is conditional on the fulfilment of the relevant operational requirements set out in §207. Q56. 29 What should an AI observe regarding the periodic revaluation under §207(j) of “fashion-sensitive” physical collateral? A56. For the purposes of the periodic revaluation process, besides those requirements applicable to all physical collateral, an AI must pay particular attention to “fashion￾sensitive” collateral to ensure that valuations are appropriately adjusted downward of fashion, or model-year, obsolescence as well as physical obsolescence or deterioration. Q57. In general, what circumstances would be considered not practicable for an AI to conduct periodic inspection of physical collateral as required under §207(j)? A57. An AI that has not conducted periodic inspection of physical collateral on practicality grounds under §207(j) should, if requested by the HKMA, be able to explain, and substantiate with objective and reliable evidence, why it has not been possible or feasible for the institution to conduct a physical inspection. The institution’s justification will be assessed on a case-by-case basis, taking into account the circumstances of the institution at the material time. Physical inspection might, for example, be hindered by events such as— (a) the institution concerned was subject to some form of severe bank-wide distress or crisis, rendering it imprudent to divert resources to some routine operations, such as scheduled inspections of physical collateral; (b) the physical collateral to be inspected was contaminated (e.g. by chemical spills), rendering it hazardous for staff of the institution to conduct the inspection;

28 Reference: Basel Framework CRE36.145.
29 Reference: Basel Framework CRE36.144(4).

Chapter IV IRB approach Page 29 of 37
(c) the physical collateral to be inspected was located in an area where there was a severe natural disaster (e.g. earthquake). The above examples are provided for illustrative purposes only and it should be noted that strong justifications will be required to support claims of impracticability of inspection. The HKMA would not concur that it was not practicable for an AI to conduct periodic inspection of physical collateral as required under §207(j) if the institution clearly had the ability, and was in a position, to do so without incurring significant cost or effort. Therefore, a general principle is that if a “hindering” event is outside the control or influence of the institution concerned, the HKMA would be more inclined to accept it as an acceptable justification for the purposes of §207(j). Q58. Could an AI take into account the credit risk mitigating effect of a recognized guarantee or a recognized credit derivative contract if the institution does not use the IRB approach to calculate its credit risk for exposures to the guarantor or counterparty? A58. Yes. An AI is allowed to take into account the credit risk mitigating effect of a recognized guarantee or a recognized credit derivative contract even if the institution uses the STC approach to calculate its credit risk for exposures to the guarantor or counterparty provided the relevant requirements set out in §216 or 217, where applicable, are satisfied. Q59. What are the key requirements governing the adjustment of an AI's estimate of the PD or LGD of the underlying exposure for the purpose of taking into account the credit risk mitigating effect of a recognized guarantee or a recognized credit derivative contract under §217? A59. To take into account the credit risk mitigating effect, an AI must, according to §217(1), adjust the institution’s estimate of the PD or LGD of the underlying exposure. §217(3) further requires that if PD adjustment is chosen, such adjustment has to be made in accordance with §216, which implies that the discretionary LGD replacement under §216(3)(c) is also available. Correspondingly, an AI may reflect the concerned credit risk mitigating effect by adjusting the estimate of LGD exclusively. Regarding the criteria and processes for making adjustments to the PD and LGD estimates, §217(2) requires an AI to ensure that its criteria and processes satisfy, subject to some other conditions, the requirements set out in Part 6 of the BCR applicable to the institution for assigning exposures to obligor grades and facility grades. As such, the guarantor, the counterparty and the credit risk mitigant are subject to the same requirements on rating assignment and credit risk component estimation as the obligor and the underlying exposure. In addition, §217(2) is also intended to, among others, cover Basel Framework CRE32.27 that, whether the adjustments are made through PD or LGD, they must be made consistently for a given guarantee or credit derivative type.
7. Treatment of expected losses and eligible provisions

Chapter IV IRB approach Page 30 of 37
Q60. Why is an AI required to compare its total eligible provisions with its total EL amount as calculated under the IRB approach for the computation of its capital base? A60. The IRB approach is based on measures of unexpected losses and expected losses. For capital adequacy purposes, an AI should cover its expected losses by making adequate provisions and cover its unexpected losses by setting aside sufficient regulatory capital. The formulae used (e.g. Formula 16 of the BCR) to calculate the risk-weighted amount of an exposure produce a capital requirement for the exposure which covers unexpected loss only. Each AI is thus required to separately calculate the total EL amount of its exposures subject to the IRB approach and compare the amount so calculated with the total eligible provisions which are attributable to these exposures. Any excess of, or shortfall in, an AI’s eligible provisions should then be reflected in the institution’s capital base, as if the institution had reduced, or increased, its provisions to a level that would fully cover its expected losses. This rationale extends why requirements for an AI to calculate its EL amount for its other exposures under the specific risk-weight approach is not required. Q61. How should an AI apportion its total regulatory reserve for general banking risks and collective provisions for the purpose of §221 if the institution uses a combination of approaches, say the IRB approach and STC approach, to calculate its credit risk? A61. The method of apportionment is set out in §42(2)(a). In general, an AI should apportion its total regulatory reserve for general banking risks and collective provisions between the approaches it uses to calculate its credit risk (i.e. the STC approach, the IRB approach, SEC-IRBA, SEC-ERBA, SEC-SA and SEC-FBA) on a pro-rata basis. The apportionment should be made in accordance with the proportions of the institution’s risk-weighted amount for credit risk which have been calculated using the different approaches. However, if an AI has obtained the MA’s prior consent under §42(2)(b), the institution may use its own method to apportion its total regulatory reserve for general banking risks and collective provisions between the various approaches used. This would only be the case if the institution can justify that there is a valid reason for using such a method.
8. IRB use test
30
Q62. What is the rationale behind the IRB use test?
A62. The IRB use test is based on the concept that supervisors can take additional comfort in the credit risk components generated by a bank’s rating system where such components play an essential role in how the bank measures and manages risk in its businesses. If a bank were to use the credit risk components generated by its rating system solely for regulatory capital purposes, this could create an incentive for the

30 The requirements regarding the track records in using a rating system set out in §(1)(b)(v) and (vi) and 2(b) of
Schedule 2 are collectively referred to as the “IRB use test” herein.

Chapter IV IRB approach Page 31 of 37
bank to minimize its capital requirements artificially, rather than produce an accurate measurement of those components. Moreover, a bank would have less incentive to keep the credit risk components accurate and up-to-date, whereas if those components are employed in the bank’s internal decision-making processes, this will automatically create an incentive for the bank to ensure the quality and robustness of the rating system generating such components. In such circumstances, the MA considers that the IRB use test plays a key role in ensuring and promoting the accuracy, robustness and timeliness of the credit risk components generated by an AI’s rating system, confirms the institution’s confidence in those components and allows the MA to place more reliance on the institution’s rating system and thus on the adequacy of its regulatory capital. Q63. For what period of time will the MA expect an AI to have been using its rating system prior to the institution adopting the IRB approach for regulatory capital purposes? A63. In general, any AI that seeks to use the IRB approach is required to have a credible track record of at least 3 years in using its rating system for the relevant exposures (which should be broadly consistent with the minimum requirements set out in the BCR relating to the use of the IRB approach) prior to the institution becoming qualified to use the relevant IRB approach. The MA will nevertheless take into account all relevant circumstances in deciding for what period of time (lengthier or shorter than 3 years) an AI should use its rating system, prior to it adopting the IRB approach. Q64. If an AI’s rating system has been developed by its parent bank and used at the group level for a certain period of time, will the institution be allowed to observe a shorter use test period than would otherwise be required? A64. An AI is required to satisfy the IRB use test in Hong Kong for a minimum period of 3 years. Hence, even if an AI’s rating system developed by its parent bank has been used at the group level for some time, the MA would still expect the institution to be able to meet the 3-year use test requirement in Hong Kong. Q65. If an AI refines or modifies its rating system during the use test period, does the use test period have to start again from the date of the refinement or modification? A65. Generally, refinements or modifications to an AI’s rating system will not render the institution non-compliant with the IRB use test. The use test period will usually only have to start again if the refinements or modifications involve a significant change in the design or operation of an AI’s rating system that substantially alters the ways the institution uses the internal ratings and default and loss estimates generated by the rating system. Q66. Where an AI maintains more than one rating system for the same portfolio of exposures (e.g. one for its regulatory capital calculation and another for benchmarking), how will the MA assess the
institution’s compliance with the IRB use test?

Chapter IV IRB approach Page 32 of 37
A66. In assessing whether or not an AI’s rating system has satisfied the IRB use test, the MA will consider the extent of the institution’s use of the rating system as a whole, rather than applying the use test to individual models separately. Q67. If an AI intends to start using its rating system for different portfolios (or segments) of exposures on different dates (e.g. phased implementation by business units), on what date does the use test period start? A67. The MA would consider it reasonable for an AI to treat the use test period for its rating system as starting on the date the rating system is used for a substantial portion (say, at least 50%) of the exposures in respect of which it intends to adopt the IRB approach. Q68. What is the meaning of the term “essential role” in §1(b)(v) and (vi) of Schedule 2? A68. “Essential role” means that the information generated by an AI’s rating system should be used in such a way as to exert a direct and observable influence on the institution’s internal decision-making processes. Where the internal ratings and default and loss estimates generated by the rating system are only used by an AI as auxiliary or reference information, the rating system will not normally be considered as playing an “essential role” for the purposes of §1(b)(v) and (vi) of Schedule 2. Q69. 31 What are the specific functions or areas in which an AI is expected to use the internal ratings and default and loss estimates generated by its rating system for internal decision-making purposes (see also A70 below)? A69. Internal ratings and default and loss estimates generated by the rating system of an AI using the IRB approach must play an essential role in the credit approval, risk management, internal capital allocation and corporate governance functions of the AI. Rating systems and estimates designed and implemented exclusively for the purpose of qualifying for the IRB approach and used only to provide IRB inputs are not acceptable. To elaborate on this principle-based requirement, the HKMA has, in
section 5.4.2 of the SPM module CA-G-4 on “Validating Risk Rating Systems under
the IRB Approach”, set out a list of specific areas or functions in which internal ratings and default and loss estimates generated by an AI’s rating system are expected to be used . These areas or functions include— (a) credit approval; (b) credit monitoring (e.g., more frequent rating review for riskier obligors); (c) analysis and reporting of credit risk information, including that used in the exercise of oversight by the board of directors and senior management; (d) pricing; (e) setting of limits for individual exposures and portfolios;

31 Reference: Basel Framework CRE36.60.

Chapter IV IRB approach Page 33 of 37
(f) determining provisioning;
(g) modelling and management of economic capital; (h) assessment of internal capital adequacy in respect of credit risk; (i) assessment of risk appetite; (j) formulating business strategies (e.g. acquisition strategy for new exposures and collection strategy in respect of problem loans); (k) setting of, and assessment against, profitability and performance targets; (l) determining performance-related remuneration (e.g. for staff responsible for rating assignment and approval); and (m) other aspects of risk management (e.g., information technology systems, skills and resources, and organisational structure). Q70. 32 Is an AI required to use its rating system in all the areas or functions specified in A69 above? A70. To satisfy the IRB use test, an AI is generally required to demonstrate that it has been using the internal ratings and default and loss estimates generated by its rating system for internal decision-making purposes for at least 3 years in the majority of the areas or functions set out in A69 above, which should include (a) credit approval, (b) credit monitoring, and (c) reporting of credit risk information (including to the institution’s board of directors and senior management). An AI using the advanced IRB approach must demonstrate that it has been estimating and employing LGDs and EADs in a manner that is broadly consistent with the minimum requirements for use of own estimates of LGD and EAD for at least the three years prior to qualification. Q71. Is an AI required to use exactly the same default and loss estimates generated by its rating system for both its regulatory capital calculation and all internal purposes? A71. Compliance with the IRB use test does not necessarily mean that an AI will have to use exactly the same default and loss estimates for both its regulatory capital calculation and all internal purposes. For example, pricing models are likely to use a PD relevant to the life of an asset, instead of using a PD with a 1-year horizon. Where such differences exist, the institution must document them and be prepared for demonstrating their reasonableness (e.g. to reflect legitimate risk management needs). Based on this guiding principle, an AI is expected to— (a) justify any differences in, and otherwise demonstrate consistency between, the internal ratings and default and loss estimates used for regulatory capital

32 Reference: Basel Framework CRE36.61.

Chapter IV IRB approach Page 34 of 37
calculation purposes and those used for the institution’s internal decision￾making purposes. Such comparison should cover both inputs (including rating criteria and risk factors) to, and outputs (such as ratings and risk estimates) from, the institution’s rating system; (b) provide qualitative and quantitative anslyses of the logic and rationale for the differences; and (c) have its credit risk control unit review, and its senior management approve, the justifications for the differences. Q72. What evidence will the MA require from AIs regarding the use of their rating systems? A72. AIs will need to demonstrate to the satisfaction of the MA that they satisfy the IRB use test. Whilst the use of internal ratings and default and loss estimates for internal decision-making purposes may vary from institution to institution and by portfolio type, the MA will normally expect an AI applying to use, or using, the IRB approach to have the following evidence to demonstrate that it satisfies the IRB use test— (a) the use of internal ratings and default and loss estimates should be articulated in the policies relating to given areas or functions as referred to in A69 and A70 above as approved by the institution’s board of directors or senior management; (b) for each area (or function) of use, there should be a clear indication that the information generated by the institution’s rating system plays an essential role in its internal decision-making process and that there is a clear relationship between the information generated by the rating system and the decisions made or actions taken (such indication should be able to facilitate the internal audit review as required in item (d) below); (c) users should be able to articulate how the information generated by the institution’s rating system is used, or the role played by the information, in the institution’s internal decision-making process; and (d) regular internal audit reviews should be conducted to verify whether or not the use of the information generated by the institution’s rating system complies with the institution’s approved policies referred to in item (a) above. Any documentation of internal challenges to the accuracy, robustness and timeliness of internal ratings and default and loss estimates generated by an AI’s rating system during the internal decision-making process, together with any follow-up actions taken, will also be regarded as evidence which demonstrates the institution’s commitment to the validity of its rating system for internal use purposes.
9. Stress-testing

Chapter IV IRB approach Page 35 of 37
Q73. What types of stress tests should be conducted by an AI for the purpose of §1(h) of Schedule 2? A73. An AI using the IRB approach is expected to conduct general stress tests which involve possible events or future changes in economic conditions that could have unfavourable effects on the institution’s credit exposures. Examples of stress scenarios that may be used include economic or industry downturns, market risk events (such as currency, stock or bond market crises) and liquidity squeezes. At a minimum, a specific stress test should be conducted to assess the effect of a mild recession on the AI’s estimates of credit risk components. In devising the stress scenario for this specific stress test, the institution may have regard to the conditions experienced in any 2 or more consecutive quarters of negative GDP growth occurring in Hong Kong during the period from 2001 to 2003 and/or occurring during other financial crises relevant to the institution, e.g. the global financial crisis in 2007/2008 or the subsequent European sovereign debt crisis. The impact of the stress scenario should be assessed based on a 1-year time horizon and take into account the lag effect of the recession on the institution’s credit exposures. The purpose of this specific stress test is to assess whether the assumptions and data used in the institution’s rating system are prudent enough to ensure that its regulatory capital calculated under the IRB approach is sufficient to cover any potential loss arising in a period of mild recession. The MA would expect to be consulted by the institution on the choice of the stress scenario to be used for this specific stress test. For more details about the use of stress tests, see section 12 of the SPM modules CA￾G-4 on “Validating Risk Rating Systems under the IRB Approach” and IC-5 on “Stress-testing”. Q74. 33 Is there any guidance on the sources of information for the stress tests for the purpose of §1(h) of Schedule 2? A74. Whatever method of stress-testing is used, an AI must include a consideration of the following sources of information. (a) The institution’s own data should allow estimation of the ratings migration of at least some of its exposures. (b) The institution should consider information about the impact of smaller deterioration in the credit environment on the institution’s ratings, giving some information on the likely effect of bigger, stress circumstances. (c) The institution should evaluate evidence of ratings migration in external ratings. This would include the institution broadly matching its buckets to rating categories.

33 Reference: Basel Framework CRE36.52.

Chapter IV IRB approach Page 36 of 37
Q75. How frequently should an AI conduct its stress tests for the purpose of §1(h) of
Schedule 2?
A75. Generally, an AI is expected to conduct its stress tests referred to in A73 above at least on an annual basis or more frequently if this is warranted by significant changes in the business strategies of the AI or in the external environment in which it operates. Q76. What would be the consequences for an AI which fails to address any shortfall in its regulatory capital identified by the specific stress test referred to in A73 above? A76. The two most likely consequences are that— (a) the MA may refuse to grant an approval to, or may withdraw an approval from, an AI for the use of the IRB approach if he is satisfied that the institution fails to operate its rating system in a prudent and consistently effective manner as required under §1(b)(iii) of Schedule 2; and (b) the MA may consider exercising his power under §97F of the Banking Ordinance to vary any capital requirement rule applicable to an AI, including by increasing all or any of the institution’s CET1 capital ratio, Tier 1 capital ratio and Total capital ratio (see item B6.2 in Annex B of the SPM module CA￾G-5 on “Supervisory review process”. Q77. 34 Should AIs consider climate-related risk drivers as possible events or future changes when performing stress tests used in the assessment of capital adequacy? A77. Climate-related financial risks may significantly impact an AI’s credit exposures within the assessment period. An AI should refer to FAQ1 attached to Basel Framework CRE36.50 for stress-testing purposes.
10. Parallel calculations
Q78. What is the period of time for which the MA requires an AI to carry out parallel calculations for the purpose of demonstrating the suitability and capability of its rating system for using the IRB approach? A78. The MA would normally expect an AI to carry out parallel calculations for a period of 4 consecutive calendar quarters (i.e. 1 year) before using the IRB approach for capital calculation. For example, an AI adopting the advanced IRB approach on 1 January 2018 would be required to carry out parallel calculations based on the STC approach, as the case may be, and the advanced IRB approach for the year 2017, covering the calendar quarter end dates of 31 March, 30 June, 30 September and 31 December.

34 FAQs inserted to Basel Framework CRE36.50.

Chapter IV IRB approach Page 37 of 37
The MA may, however, consider extending the period of an AI’s parallel calculations if the quality of the institution’s parallel calculations is not satisfactory, any subsequent slippage is identified in the institution’s implementation efforts, or any serious weaknesses are found in the institution’s rating system. Q79. Is an AI applying for switching the IRB calculation approach from the foundation IRB approach to the advanced IRB approach to calculate its credit risk for corporate and sovereign exposures of a particular IRB adoption class required to carry out parallel calculations? A79. Yes. The purpose of parallel calculations is to enable an AI to demonstrate to the MA’s satisfaction the suitability and capability of its rating system for the calculation of the institution’s credit risk and to familiarize itself with the use of its rating system prior to it implementing the IRB approach. As using the advanced IRB approach will require an AI to have a more sophisticated system for generating its own estimates of LGD and EAD for its corporate and sovereign exposures, it is both prudent and reasonable to require the institution to provide similar parallel calculations to the MA to prove its readiness to migrate to a more advanced approach. In such a situation, the parallel calculations will consist of one set of calculations using the foundation IRB approach (i.e. the current approach used by the institution) and the other using the advanced IRB approach (i.e. the approach the institution is seeking the MA’s prior consent to use). To be consistent with the time period referred to in A78 above, the MA would expect an AI to carry out parallel calculations for a period of 4 consecutive calendar quarters (i.e. 1 year) before migrating from the foundation IRB approach to the advanced IRB approach for regulatory capital calculation. Without limiting the considerations of the MA, a shorter timeframe (unlikely to be less than 2 consecutive calendar quarters) may be agreed by the MA after considering the relevant factors and evidence, for example, the robustness of the change management in respect of the relevant regulatory reporting. Q80. In what form, and using what type of data, should an AI provide its parallel calculations to the MA? A80. Generally, an AI should provide 2 sets of calculations to the MA, one based on the approach it currently adopts and the other based on the IRB approach it is applying to use, using the Return of Capital Adequacy Ratio of an Authorized Institution Incorporated in Hong Kong (MA(BS)3) (“the CAR return”). As regards the calculations derived from the IRB approach, the institution should complete Part IIIc of the CAR return and other relevant items relating to the use of that IRB approach under other parts of the CAR return (e.g. Division B of Part I for the calculation of the output floor and various items in Part IIId for the calculation of the risk-weighted amount for securitization exposures if SEC-IRBA is used).
If an AI encounters any practical difficulties in completing the CAR return for parallel calculation purposes, it should consult with the MA to discuss any alternative arrangement.

---

Chapter V Counterparty credit risk Page 1 of 23
V. Counterparty Credit Risk

1. General
   Scope of application
   Q1. Does Part 6A of the BCR apply to FX spot contracts?
   A1. FX spot contracts are outside the scope of Part 6A unless they fall within the definition of “long settlement transaction” in §2(1) of the BCR. AIs are reminded that in the case of a FX spot contract that is not a long settlement transaction and not yet due for settlement, the receivable arising from the contract falls within §65L(1)(b) or §114A(1)(b), or paragraph (g) of the definition of “cash items” in §139(1), of the BCR. If the settlement fails on the settlement date, the receivable will fall within §65L(3) or (4) or §114A(3) or (4), or paragraph (h), (i) or (j) of the definition of “cash items” in §139(1), as the case requires. Q2. Does Part 6A of the BCR apply to derivative contracts embedded in credit￾linked notes and currency linked deposits? A2. Credit default swaps embedded in credit-linked notes (“CLN”) are outside the scope of Part 6A, having considered that— (a) in cases where an AI invests in CLN (i.e. it is a protection seller), its credit exposure to the credit protection buyer (i.e. the issuer of the CLN) is an on￾balance sheet exposure with known amount; (b) in cases where an AI issues CLN (i.e. it is a protection buyer), the AI does not incur any default risk exposure to the protection sellers (i.e. the holders of the CLN) because they have already paid the principal upfront to the AI (which is equal to the maximum possible credit-event payment under the credit default swap embedded in the CLN). In the case of currency linked deposits, the HKMA would consider it acceptable for an AI to regard the put option embedded in the currency linked deposit bought by the AI as not having any default risk exposure if the seller of the put option (i.e. the depositor) has already delivered upfront to the AI the full settlement amount (being the amount that the seller is obliged to pay to the AI under the put option when the option is exercised). In such case, the bought put option may be regarded as being outside the scope of Part 6A. Q3. Certain bond transactions, such as primary issuance, would generally have settlement date longer than 5 business days after the trade date. Please clarify whether this kind of transactions is in scope of Part 6A of the BCR.

Chapter V Counterparty credit risk Page 2 of 23
A3. Bond transactions, including primary issuance, that have settlement date longer than 5 business days after the trade date fall within the definition of “long settlement transaction” in §2(1) of the BCR. They are within the scope of Divisions 1A, 2 and 2A of Part 6A of the BCR. Margin period of risk used in the SA-CCR approach and the IMM(CCR) approach Q4. If there is an illiquid transaction or collateral in a netting set, when should the higher supervisory floor under §226BZE(4) or §226M(3) be applied to the netting set? A4. The supervisory floor of 20 business days applies immediately once a netting set falls within §226BZE(4) or §226M(3). Q5. Under §226BZE and §226M, for future dates beyond the expected maturity of a transaction that leads to an increase in margin period of risk (e.g. an illiquid transaction falling within §226M(3)), should the margin period of risk used in respect of those future dates be reduced to the corresponding minimum set out in §226BZE(2) or §226M(1)? A5. The supervisory floors set out in §226BZE and §226M are minimum requirements. AIs should not mechanically apply the minimum requirements but should assess the market liquidity of the positions in question. The actual margin periods of risk that should be used in calculating the amounts of default risk exposures may be longer than the supervisory minima if the liquidity of the positions concerned warrants it. Q6. In the case of non-centrally cleared derivative contracts that are subject to the margin standards set out in SPM module CR-G-14 “Non-centrally Cleared OTC Derivatives Transactions – Margin and Other Risk Mitigation Standards”, what margin calls are to be taken into account for the purpose of counting the number of disputes in accordance with §226BZE(6) or §226M(7)? A6. In such case, it is acceptable for AIs to count variation margin call disputes only. Risk-weights applicable to default risk exposures to banks or qualifying non-bank financial institutions (“QNBFIs”) under STC approach Q7. Assuming that— (a) a set of nettable SFTs and derivative contracts are entered into with a bank or QNBFI;

Chapter V Counterparty credit risk Page 3 of 23
(b) some of the SFTs and derivative contracts have an original maturity of more than 3 months while the original maturity of the others is less than 3 months; and (c) the SFTs and derivative contracts are all under the same netting set, clarification is sought on whether the set of nettable SFTs and derivative contracts are required to be assigned risk-weights separately in accordance with the original maturity under §59(2) of the BCR. A7. As cross-product netting is not recognised except for cases where the IMM(CCR) approach is used for both SFTs and derivative contracts, the amount of the default risk exposure of the SFTs and the amount of the default risk exposure of the derivative contracts must be calculated separately even though they are within the same netting set. When some of the SFTs and/or derivative contracts within the same netting set have an original maturity of more than 3 months, all SFTs and derivative contracts in the same netting set should be treated as general bank exposures under §59 of the BCR. In other words, both the default risk exposure of the SFTs and the default risk exposure of the derivative contracts are general bank exposures to the bank or QNBFI. Collateral posted outside netting set Q8. Could the HKMA provide examples to illustrate how §78(1A) to (1C) work? A8. (a) The amount of the default risk exposure in respect of an SFT of an AI referred to in §78(1A)(a) is equal to the principal amount of the securities or money provided by the AI under the SFT to the counterparty concerned. Unlike the calculations under §226MK and §226ML, the securities or money received by the AI from the counterparty is not included as part of the default risk exposure. The credit risk mitigation effect of the securities or money received must be taken into account under Part 4 in accordance with— (i) if the simple approach is used⸺§85; or (ii) if the comprehensive approach is used⸺§88. (b) §78(1A)(b) is primarily intended to cater for derivative contracts entered into by an AI with a commercial end-user under a general banking facility where— (i) the commercial end-user is usually not a covered entity as defined in SPM module CR-G-14 “Non-centrally Cleared OTC Derivatives Transactions – Margin and Other Risk Mitigation Standards”; (ii) the general banking facility consists of multiple credit lines for various purposes (e.g. overdraft, letter of credit / trust receipt for importing goods

Chapter V Counterparty credit risk Page 4 of 23
from overseas exporters, FX derivative contracts for hedging the FX risk associated with payments in foreign currencies, etc.) and at least one of the credit lines is for derivative transactions (Remarks: A general banking facility under which all credit lines are for entering into derivative contracts will not be considered as meeting the condition specified in §78(1B)(a). In such case, collateral can be taken into account only in the manner as specified in Division 1A or 2 of Part 6A); (iii)the credit lines are secured by the same recognized collateral. This includes cases where the credit lines are secured by a single pool of different assets pledged (e.g. cash deposits and listed shares); and (iv) in case, after default of the commercial end-user, the proceeds from liquidation of the assets pledged are insufficient to offset all the losses incurred by the AI under the facility, the AI would have discretion not to use any of the proceeds to offset losses on derivative contracts. A numerical example is provided below to illustrate how the total risk￾weighted amount of a general banking facility should be calculated under Part 4:
 A general banking facility of HK$ 2 million is granted to an unrated local corporate (applicable risk-weight (“RW”) is 100%).  The facility is secured by cash deposits of HK$0.5 million and double-A rated US$ debt securities (residual maturity is 8 years) issued by a US corporate with a current market value of HK$0.8 million.  The facility consists of two credit lines with the following sub-limits and outstanding balances:
Type of credit line
Sub-limit
(HK$ million)
Drawn amount
(HK$ million)
Undrawn amount
(HK$ million)
Exposure amount
(HK$ million)
Overdraft 1.5 1 0 1
FX derivatives 2* 1 0 0.08

• Expressed in notional amount.
  Representing default risk exposure amount calculated under the SA-CCR approach without taking into account any collateral received by the AI.  By using the comprehensive approach, the risk-weighted amount (“RWA”) of the facility = 𝑚𝑎𝑥 0; 𝐸 − 𝐶𝑐𝑎𝑠ℎ − 𝐶𝑑𝑒𝑏𝑡𝑠 ∙ 1 − 𝐻𝑑𝑒𝑏𝑡𝑠 − 𝐻𝑓𝑥 ∙ 𝑅𝑊 = 𝑚𝑎𝑥 0; 1.08 − 0.5 − 0.8 ∙ 1 − 0.06 − 0.08 ∙ 100% = 0 (c) §78(1C) is relevant to an AI if—

Chapter V Counterparty credit risk Page 5 of 23
(i) a general banking facility granted by the AI consists of multiple credit lines and at least one of the credit lines is for derivative transactions; (ii) the credit lines under the facility are secured by the same recognized collateral; and (iii) all or part of the recognized collateral has been designated solely for offsetting losses on derivative contracts (in other words, the designated amount cannot be used to offset losses on non-default risk exposures such as overdraft or term loan). The same example in paragraph (b) is used for illustration except that out of the cash collateral of HK$0.5 million, HK$0.2 million can only be used to offset losses on default risk exposures. Also, it is assumed that there is only one unmargined FX forward contract outstanding with a remaining maturity of 6 months. The amount of the default risk exposure of the FX forward contract is calculated by using the SA-CCR approach as follows:
(HK$’000) V C RC MF AddOn PFE Default risk exposure amount FX forward 30 200.0 0.0 0.707 28.28 2.55 3.57 By using the comprehensive approach⸺  the RWA of the overdraft (in HK$’000) = 𝑚𝑎𝑥 0; 𝐸 − 𝐶𝑐𝑎𝑠ℎ − 𝐶𝑑𝑒𝑏𝑡𝑠 ∙ 1 − 𝐻𝑑𝑒𝑏𝑡𝑠 − 𝐻𝑓𝑥 ∙ 𝑅𝑊 = 𝑚𝑎𝑥 0; 1000 − 300 − 800 ∙ 1 − 0.06 − 0.08 ∙ 100% = 12  the RWA of the FX forward contract (in HK$’000) = 3.57 * 100% = 3.57  total RWA of the facility (in HK$’000) = 15.57 (d) In general, for the purposes of §78(1B) and §226BJ(7), collateral is considered to be designated solely for offsetting losses on default risk exposures if the designation is given effect to through contractual arrangements. (e) Although the manner in which collateral is taken into account in RWA calculations under the BSC approach and IRB approach is different from that under the STC approach, the policy intent explained above also applies to §126(1A), (1B) and (1C) and §204(2), (3) and (4).

Chapter V Counterparty credit risk Page 6 of 23
Cash collateral posted
Q9. We would like the HKMA to confirm our understanding that the credit risk or market risk of the posted collateral itself is not to be considered when determining capital requirements under Part 4, 5, 6, 7 or 8 if the posted collateral is “cash” (in any currency). A9. If the cash, before it is posted as collateral, is not subject to any capital charge for credit risk or market risk under the BCR, this treatment will remain unchanged after it has been posted as collateral.
2. SA-CCR approach
Classification of derivative contracts
Q10. Will there be further guidance on the determination of primary risk drivers of specific products that are common in the market. For example, where a cross currency interest rate swap without any principal exchange is mainly driven by interest rate risk, should it be classified as interest rate contract instead of exchange rate contract? A10. For contracts that are commonly traded in the market, there is usually only one primary risk driver (i.e. the market risk factor that most significantly affects the mark-to-market value of a contract). In general, if a derivative contract is mainly driven by interest rate risk, unless the contract has another equally important risk driver, it would be classified as an interest rate contract for the purposes of the SA￾CCR approach. AIs are expected to have the capability to identify the primary risk factors of their derivative contracts, including the ability to assess how sensitivities and volatilities of an underlying exposure drive the market value or payoff of the derivative contract concerned. Otherwise, it may call into question whether an AI’s risk management framework is commensurate with the size and complexity of its derivative activities. Q11. Should CNH (offshore) and CNY (onshore) be considered as two different currencies for the purpose of determining hedging sets? Similarly, should shares of the same company listed in multiple markets (e.g. H shares and A shares) be considered as shares issued by the same company? A11. For the purposes of the SA-CCR approach, if rate or price differentials persistently exist between the onshore exchange rate and offshore exchange rate of a currency, or between the equities of a company listed in multiple exchanges, the two exchange rates must be treated as if they were exchange rates of two different currencies and derivative contracts referencing the equities must be treated as if they were referencing equities issued by different companies. Hence, CNH and CNY must be regarded as two different currencies, and H shares and A shares of the same company must be regarded as equities issued by two different companies, due to the concerns

Chapter V Counterparty credit risk Page 7 of 23
over basis risk.
Q12. Could more explanation be given on what AIs are expected to do under §226BQ(2), (3) and (4)? A12. Under the SA-CCR approach, there are four pre-specified hedging sets for commodities (i.e. Energy, Metals, Agricultural, and Other commodities). §226BQ(2) and (3) require an AI to further classify contracts falling within each hedging set into subsets defined by the AI based on commodity types. For example, for the hedging set “Metals”, an AI may want to introduce subsets such as “precious metals” and “base metals” if this would be more reflective of the basis risk to which the AI is exposed. The subsets may need to be redefined from time to time in light of any changes in the risk profile of the AI’s commodity-related derivative contracts. §226BQ(4) empowers the HKMA to require an AI to use more refined definitions of commodity types for the purposes of setting up subsets. For example, crude oil could be a commodity type, but more refined definitions of commodity type such as Brent and West Texas Intermediate may also be used. Such power will be exercised only when the HKMA identifies, during its usual supervisory process, that some products which are grouped by the AI into the same hedging set or subset pose significant basis risk to the AI. Q13. Does one-way margin agreement (where only the AI posts variation margin) fall within the definition of “variation margin agreement” in §226BA? A13. No. As a result, contracts subject to an one-way margin agreement must be treated as unmargined contracts for the purposes of the SA-CCR approach. Q14. For long settlement transactions that are generated from buy and sell trades with the same underlying securities and the same settlement date, please clarify whether they are subject to default risk exposure calculation as interest rate exposures under the SA-CCR approach. A14. In cases where buy and sell trades with the same underlying securities and the same settlement date fall within the definition of “long settlement transaction” in §2(1) of the BCR, the following principles apply for the purposes of calculating the amounts of the default risk exposures of these trades under the SA-CCR approach— (a) if the underlying securities are equities, the trades should be treated as if they were equity-related derivative contracts; (b) if the underlying securities are debt securities, the trades should be treated as if they were interest rate contracts or credit-related derivative contracts, depending on the AI’s own assessment of the primary risk factor that drives changes in the market values of the debt securities.

Chapter V Counterparty credit risk Page 8 of 23
Q15. Please clarify whether bond transactions that are long settlement transactions should be classified as interest rate exposures under the SA-CCR approach. A15. Under the SA-CCR approach, bond transactions that are long settlement transactions could be treated as if they were interest rate contracts or credit-related derivative contracts, depending on the AI’s own assessment of the primary risk factor that drives changes in the market values of the bonds. Treatment of multiple netting sets subject to a single variation margin agreement Q16. If there is more than one netting set covered by the same variation margin agreement, how should the multiplier for each of the netting sets be calculated? A16. In order to calculate the multiplier applicable to each of the individual netting sets covered by a single variation margin agreement or collateral amount, the haircut value of net collateral held (“C”) for the netting sets as calculated under §226BJ should be allocated to each of the netting sets as follows:
(a) If the AI concerned is a net receiver of collateral (C>0), all of the individual amounts allocated to the individual netting sets must also be positive or zero. Netting sets with positive current mark-to-market (“MTM”) values must first be allocated collateral up to the amount of those MTM values. Only after all positive MTM values have been compensated may surplus collateral be attributed freely among all netting sets. (b) If the AI concerned is a net provider of collateral (C<0), all of the individual amounts allocated to the individual netting sets must also be negative or zero. Netting sets with negative MTM values must first be allocated collateral up to the amount of those MTM values. If the collateral provided is larger than the sum of the negative MTM values (e.g. where C = −17 and sum of –ve MTM values = −15), then all multipliers must be set equal to 1 and no allocation is necessary. (c) The sum of the allocated parts must be equal to C. Apart from the above limitations, AIs may allocate collateral at their discretion. The multiplier is then calculated per netting set by using Formula 23AN with C in the formula set equal to the allocated amount of collateral. Effective notional amount Q17. Is it correct that §226BZC(5)(b) does not cover derivative contracts where the notional amount varies due to price changes (typically, FX, equity and commodity derivative contracts)?

Chapter V Counterparty credit risk Page 9 of 23
A17. §226BZC(5)(b) is intended to cover interest rate contracts and credit-related derivative contracts with variable notional amounts specified in the contracts (e.g. amortizing and accreting swaps). §226BZC(5)(b) does not apply to derivative contracts where— (a) the notional amount is denominated in a foreign currency and is variable solely because of the changes in the exchange rate between the foreign currency and Hong Kong dollars; or (b) the notional amount is expressed in units of an underlying exposure (e.g. equity or commodity) and is variable solely because of the changes in the unit price of the underlying exposure. Q18. §226BZC(5)(b) requires AIs to use time-weighted average notional amount over the remaining life of a variable notional swap as the notional amount of the swap. Confirmation is sought on the interpretation of time-weighted average notional as below:
Remaining maturity (year)
Notional
(a)
Duration
(b)
Weighted notional
(a) × (b)
1 10,000 0.25 2,500
0.75 7,500 0.25 1,875
0.5 5,000 0.25 1,250
0.25 2,500 0.25 625
Average notional 6,250
A18. The interpretation is consistent with the requirement set out in §226BZC(5)(b), presuming that the word “Duration” in column (b) of the table carries its ordinary dictionary meaning (as opposed to it meaning either the effective, modified or Macaulay duration, or any other similar duration measure for the swap). Q19. Is it correct that only non-linear products can be decomposed? A19. Yes, linear products whose prices depend linearly on one or more underlying variables, such as ordinary interest rate swaps, must not be decomposed. Q20. How should the effective notional amounts be calculated for options? A20. (a) For European, Asian, American and Bermudan put and call options, the effective notional amount of each of these options must be calculated in accordance with §226BZA(1), with the supervisory delta adjustment calculated by using the simplified Black-Scholes formula set out in §226BZB(2) and (3). In particular—

Chapter V Counterparty credit risk Page 10 of 23
(i) in the case of Asian options, the price of the underlying exposure (“P”) must be set equal to the current value of the average used in the payoff (see §226BZB(4)); and (ii) in the case of American and Bermudan options, the latest allowed exercise date must be used as the exercise date (“T”) (see §226BZB(3)(b)(v)). (b) In the case of Bermudan swaptions, the supervisory duration used in calculating the adjusted notional of the swaption must be computed in accordance with §226BZC(2) and the start date (“S”) in Formula 23AZ must be equal to the earliest allowed exercise date of the swaption, while the end date (“E”) in the formula must be equal to the end date of the underlying interest rate swap. (c) In the case of a complex option to which §226BZA(5) applies, the option can be decomposed into vanilla options in a manner consistent with the approaches below:
(i) For a digital option, the payoff of the option (bought or sold) with strike price K must be approximated via a “collar” combination of bought and sold European options of the same type (call or put). The strike prices of the European options must be set equal to 0.95∙K and 1.05∙K. The size of the position in the collar components must be such that the payoff of the digital option is reproduced exactly outside the region between the two strike prices. An effective notional amount is then computed for each of the bought and sold European options separately in accordance with §226BZA(1) with the supervisory delta adjustment calculated in accordance with §226BZB(2) and (3) (T and P in the formula in §226BZB(3) must be the exercise date of the digital option and the current price of the underlying exposure of the digital option respectively). The absolute value of the effective notional amount of the digital option is capped by the ratio of the digital payoff to the relevant supervisory factor. (ii) For a derivative contract whose payoff can be represented as a combination of European option payoffs (e.g. collar, butterfly/calendar spread, straddle, and strangle), each European option component must be treated as a separate contract. (iii) For a derivative contract that is a multiple-payment option (e.g. interest rate caps and floors), the contract may be represented as a combination of single-payment options. In particular, interest rate caps/floors may be represented as a portfolio of individual caplets/floorlets, each of which is a European option on a floating interest rate over a specific coupon period. For each caplet/floorlet, S and T are the time periods starting from the current date to the start of the coupon period, while E is the time period starting from the current date to the end of the coupon period. Q21. We understand that the HKMA prefers calculation of the effective notional amounts of target redemption forwards (“TRF”) through approximation by a set of (m×n) options, where m is the number of remaining fixing and n is the

Chapter V Counterparty credit risk Page 11 of 23
number of options decomposed in each fixing. However, such approach would require significant enhancement in AIs’ system. Processing massive decomposition transactions would also impose daily computational burdens on AIs. We therefore suggest approximating a TRF by a set of m leveraged forward, where m is the number of remaining fixing. Each of the leveraged notional amounts (equals to the unleveraged notional amount multiplied by the leveraging factor) reflects the default exposure when the TRF is out-of-the￾money, and the effective notional amount of the TRF is the aggregate effective notional amount for the whole set of leveraged forwards. Would the HKMA consider this alternative approach acceptable? A21. The HKMA believes that approximation by options is more in line with the guiding principles for decomposition established in Chapter CRE52 of the Consolidated Basel Framework. However, if AIs consider their suggested approach is preferable because it is easier to implement, the HKMA would not have objection provided that— (a) TRFs and other similar derivative contracts referencing the same currency pair are grouped together to form a separate hedging set (it is also acceptable if the AI treats each such contract as a separate hedging set); and (b) the method chosen by an AI for a particular product type is applied consistently across all existing and future outstanding contracts in that product type. Supervisory delta adjustment Q22. How should the supervisory delta adjustment for options be calculated when the term P/K is zero or negative (e.g. as may be the case in a negative interest rate environment)? A22. The supervisory delta adjustment should be calculated in accordance with §226BZB(2) and (3) by assigning a non-zero value to the parameter λ to incorporate a shift in the price of the underlying exposure and the strike price. The same value of λ must be used consistently for all interest rate options in the same currency. If the relevant supervisory authority in a jurisdiction has recommended an appropriate value of λ for the jurisdiction’s local currency, AIs are encouraged to adopt the recommended value in calculating the supervisory delta adjustments for interest rate options in that currency. Nevertheless, AIs may use lower values if it suits their portfolios. Q23. The calculation of supervisory delta adjustments for foreign exchange options depends on the convention taken with respect to the ordering of the respective currency pair. For example, a call option on EUR/USD is economically identical to a put option in USD/EUR. Nevertheless, the calculation of the

Chapter V Counterparty credit risk Page 12 of 23
supervisory delta adjustment leads to different results in the two cases. Which convention should AIs select for each currency pair? A23. For each currency pair, the same ordering convention must be used consistently across the AI’s portfolios and over time. The convention is to be chosen in such a way that it corresponds best to the market practice for how derivative contracts in the respective currency pair are usually quoted and traded. Maturity Q24. If a cross currency swap requires settlement of exchange differences on notional on a reset date, but despite substantial portion of the fair value of the transaction being settled on the reset date, the fair value would strictly not be zero because of discounting and the basis swap, can the reset date be used as the remaining maturity for exposure calculation? A24. §226BZD(3), which reflects the requirement in paragraph 52.37(5) of Chapter CRE52 of the Consolidated Basel Framework, requires that the terms of the contract must be reset so that the fair value of the contract is zero. Hence, for a contract whose fair value after reset is not zero, the reset date cannot be taken as the remaining maturity of the contract for the purposes of the SA-CCR approach. Q25. Is an AI required to use the SA-CCR approach to calculate amounts of its default risk exposures in respect of matured derivative contracts that are pending for settlement? A25. Yes. The amount of default risk exposure calculated by the SA-CCR approach is intended to measure the risk that the counterparty could default before the final settlement of the derivative contracts’ cash flows (see the definition of “counterparty credit risk” in §2(1) of the Banking (Capital) Rules). Hence, the AI’s default risk exposures in respect of matured derivative contracts that are pending for settlement should also cover the period between maturity dates of the contracts and their final settlement dates. Q26. If the answer to Q25 is yes, is the AI still required to calculate the potential future exposures of those matured derivative contracts (given their delivery amounts are fixed, i.e. there will not be any further change in the values of the contracts)? A26. Yes. The AI should calculate the potential future exposure according to subdivision 4 of Division 1A of Part 6A subdivision 4 of the BCR.

Chapter V Counterparty credit risk Page 13 of 23
Conversion between years and days
Q27. In §226BZE(1), should the denominator (i.e. 1 year) in Formula 23AZB be converted into, say, 250 business days, considering that the numerator (i.e. MPORi) is often expressed in days? Similarly, in other sections where time is expressed in years but subject to floors expressed in business days, should the floors be converted into years in the same way, i.e. dividing the floor concerned by 250 (e.g. a floor of 10 business days is equal to 10/250 year)? A27. If there is a need to convert the unit of time from business days into years or vice versa, the conversion must be made by using the standard market convention applicable to the derivative contracts and the financial markets concerned. For example, 1 year may be converted into 250 business days. Similarly, 10 business days can be converted into years by dividing it by 250. Haircut value of net collateral held Q28. If there is a legally enforceable binding agreement to link a CLN issued to a designated portfolio of OTC derivative contracts with a counterparty such that any default loss in respect of the portfolio will be borne by the holder of the CLN once the counterparty is in default, can the CLN be classified as a recognized credit risk mitigation (“CRM”) and captured in SA-CCR calculation? A28. Since the CRM in question is considered a credit derivative contract instead of collateral under the BCR, the CLN concerned must not be included in the calculation of the haircut value of net collateral held under the SA-CCR approach. However, an AI may still recognise the CRM effect of the CLN in accordance with the provisions applicable to recognized credit derivative contracts in Part 4, 5 or 6, as the case requires, if the credit derivative contract embedded in the CLN meets all the applicable recognition criteria set out in the BCR (e.g. §99). More specifically, when an AI calculates the risk-weighted amount of the default risk exposure in respect of the designated portfolio of OTC derivative contracts under Part 4 of the BCR, the AI may determine the credit protection covered portion of the default risk exposure in accordance with §101(8) of the BCR.
3. IMM(CCR) approach
Calculation of current exposures
Q29. §3(e)(i) of Schedule 2A requires an AI to compute current exposures using current market data. Clarification is sought as to whether current market data include market implied data. A29. In Schedule 2A §3(e)(i), “current market data” means any directly observed market

Chapter V Counterparty credit risk Page 14 of 23
data (e.g. interest rates, equity prices, etc.), or data implied (e.g. option implied volatility) by other observable prices, as of the valuation date. In other words, for the purpose of computing current exposures, “market implied data” is interpreted more narrowly, i.e. it only means data implied by current (as opposed to past) market data. Stressed effective EPE Q30. Under §226D(1)(b), an AI is required to use a stress calibration as set out in §3(f) of Schedule 2A to calculate a stressed IMM(CCR) risk-weighted amount. Clarification is sought on the length of the stress period that should be used. A30. §226D(1)(b) requires an AI to calculate a stressed IMM(CCR) risk-weighted amount using a stress calibration which must include a period of stress to the credit default spreads of the AI’s counterparties. The length of such period is not specified. The AI should select the stress period based on its specific circumstances and the characteristics/profile of its CCR exposures. As required by §3(g) of Schedule 2A, the AI must assess the soundness and adequacy of the stress calibration regularly (at least quarterly). The period of stress selected is expected to be one of the items covered by this regular assessment. The assessment procedures and results are subject to review by the HKMA as part of its on-going supervisory process. Moreover, the HKMA may require an AI to adjust the stress calibration if the comparison conducted by the AI as required by §3(g)(iii) of Schedule 2A shows that the exposures of the benchmark portfolios deviate from each other substantially. Q31. For the purposes of §226D(1)(b), should the credit spread stress period be at the centre of the 3-year period mentioned in §3(f) of Schedule 2A (i.e. there will be an equal length of time before and after the credit spread stress period)? A31. There is no such requirement. When applying to the HKMA for approval to use the IMM(CCR) approach, an AI should discuss and agree with the HKMA the approach / methodology for determining and reviewing the stress period. Q32. For the purposes of §226D, how frequent should the Effective EPE calculated using current market data be compared with the Effective EPE calculated using a stress calibration? A32. When applying to the HKMA for approval to use the IMM(CCR) approach, an AI should discuss and agree with the HKMA the frequency at which the comparison required by §226D(2) should be conducted. Generally, the AI should expect the frequency of comparison to be at least quarterly. The HKMA may require the AI to increase the agreed frequency if the HKMA considers that such frequency is no longer adequate because of, for example, material changes in the level or nature of the AI’s derivatives activities or significant increase in market volatilities.

Chapter V Counterparty credit risk Page 15 of 23
Collateral
Q33. Under §226K(3), an AI may take into account the effect of collateral that is not cash of the same currency as the default risk exposure concerned if the AI applies standard supervisory haircuts to the collateral. Clarification is sought on how the haircut for currency mismatch should be applied to mixed currency exposures. A33. For the purposes of §226K(3)(b), the standard supervisory haircut applicable in consequence of a currency mismatch (i.e. 8%) should be applied to each element of the collateral that is provided in a currency different from that of the exposure. For example, if cash in US dollars is provided by a counterparty as collateral in respect of performance under a derivative contract, and the default risk exposure to the counterparty of the contract is partly denominated in Euro and partly denominated in Japanese Yen, the currency mismatch haircut should be applied to that portion of the collateral covering the Euro denominated exposure, and likewise for the portion of the collateral covering the Yen denominated exposure. Q34. If an AI uses both the IMM(CCR) approach and the SA-CCR approach to calculate the amounts of its default risk exposures to a counterparty (this may happen if the AI’s IMM(CCR) approval only covers a certain category of transactions or the AI is permitted, under §10B(5) or (7), to use the SA-CCR approach for certain transactions), how should the collateral posted by the counterparty be allocated across the different calculation methods? A34. The AI has to split the original netting set into two new netting sets, one that is subject to the IMM(CCR) approach and the other that is subject to the SA-CCR approach. The AI is free to decide how the collateral posted by the counterparty should be allocated between the two netting sets. However, no double-counting of the collateral is allowed.
4. Current exposure method
Q35. How should the credit conversion factor (“CCF”) applicable to a debt security contract (i.e. a derivative contract the value of which is determined by reference to the value of, or any fluctuation in the value of, one or more than one underlying debt security or underlying debt security index) be determined under the current exposure method set out in Division 2A of Part 6A? A35. An AI should determine the primary risk factor of the contract and classify the contract into one of the following types based on the primary risk factor so determined:
(a) Interest rate contract;
(b) Credit-related derivative contract;

Chapter V Counterparty credit risk Page 16 of 23
(c) Exchange rate contract;
(d) Equity-related derivative contract;
(e) Commodity-related derivative contract;
(f) Derivative contract other than the above.
A debt security contract is usually an interest rate contract or a credit-related derivative contract. The CCF applicable to the contract is then determined in accordance with Table 23AI in §226MD. Q36. How recognized netting is taken into account under the current exposure method? A36. For simplicity, the modified current exposure method (“CEM”) set out in Division 2A of Part 6A of the BCR does not take into account recognized netting in the calculation. As a result, each derivative contract covered by a valid bilateral netting agreement should be treated under the CEM as if it were not covered by such agreement.
5. Securities financing transactions (SFTs)
Calculation of amount of default risk exposure Q37. Clarification is sought as to whether an SFT arranged by an AI as agent is subject to capital charge. A37. Where an AI, acting as an agent, arranges an SFT between a customer and a third party and provides a guarantee to the customer that the third party will perform on its obligations, then the risk to the AI is the same as if the AI had entered into the transaction as a principal. In such case, the AI must calculate capital requirement for the SFT as if it were itself the principal. Q38. Is it correct that if an AI uses the STC approach and the comprehensive approach in its treatment of recognized collateral, the amount of the default risk exposure of the AI’s nettable SFTs can be calculated in accordance with §226MK? A38. Yes. The understanding is correct.

Chapter V Counterparty credit risk Page 17 of 23
Haircut floors for SFTs 1
Q39. If there is a non-zero minimum transfer amount (“MTA”) agreed between an AI and its counterparty for SFTs, does the amount of “collateral that is called” referred to in §226ZED(5) and §226ZEE(5) include any amount that is below the MTA and has not been called? A39. “Collateral that is called” only includes the amount that is actually called by the counterparty or the AI. If an amount has not been called, such amount should not be counted in “collateral that is called”. Q40. If an amount below the MTA is notified to the counterparty but there is no actual transfer of collateral taking place, could the amount “notified” be considered as “collateral that is called”? A40. The amount “notified” cannot be considered as “collateral that is called” because the notification will not be followed by a transfer of the amount “notified” from the counterparty to the AI. Only collateral calls that will result in a transfer of the collateral called to the AI are taken into consideration in assessing whether the haircut floor is complied with.
6. Exposures to CCPs
Supervisory approval
Q41. If an AI has been granted an approval to use the IMM(CCR) approach for a specific product, does the AI need to obtain further approval from the HKMA to use the IMM(CCR) approach, as the case may be, for the centrally cleared version of the product? A41. Under §10B(9), an AI must obtain the prior consent of the MA before making any significant change to any approved internal model. Hence, further approval is needed if the inclusion of the centrally cleared version of the product would require significant change to the approved internal models concerned. Determination of a CCP’s status Q42. Who will determine whether a CCP is qualifying? A42. It is the primary responsibility of AIs to determine whether a CCP is qualifying. If a CCP regulator has provided a public statement on whether a CCP is qualifying

1 Division 5 of Part 6A of the BCR on haircut floors for SFTs has not yet commenced operation.

Chapter V Counterparty credit risk Page 18 of 23
or non-qualifying, then AIs may rely on the statement to determine the appropriate capital treatments for their exposures to the CCP. Otherwise, AIs should determine whether a CCP is qualifying based on the criteria set out in paragraph (a) of the definition of “qualifying CCP” (“QCCP”) in §226V(1). AIs should be prepared to provide the HKMA with a list of CCPs to which they have exposures, including the AIs’ evaluation of the relevant criteria in respect of each such CCP. If a CCP ceases to be a QCCP because it no longer meets all the criteria set out in paragraph (a) of the definition of “qualifying CCP” in §226V(1), a 3-month grace period is available during which AIs may calculate the capital requirements for their exposures to the CCP as if the CCP were a QCCP (see §226ZC(1)). If a CCP in a jurisdiction outside Hong Kong calculates its counterparty credit risk exposures to its clearing members using methods other than a method that is consistent with the SA-CCR published by the Basel Committee (thus failing to meet the description in paragraph (a)(iii) and (iv) of the definition of “qualifying CCP” in §226V(1)), an AI may deem such CCP as a QCCP under §1(1) of Schedule 16 provided that all the conditions set out in §1(2) of that Schedule are met. Q43. If a jurisdiction outside Hong Kong has published on or before 30 June 2021 its SA-CCR rule but the mandatory compliance date has not yet been announced, is a CCP in such jurisdiction eligible for the transitional arrangement provided for under Schedule 16? A43. Yes. In such case, the end date of the transition period (i.e. the period during which the CCP can be regarded as a QCCP) will be known once the mandatory compliance date is announced by the jurisdiction concerned. Q44. If the information mentioned in paragraph (a)(i) and (ii) of the definition of “qualifying CCP” in §226V(1) in respect of a CCP is not publicly disclosed, how should an AI assess whether the CCP is a qualifying CCP? A44. If such information is not made available to market participants by the relevant regulatory authority of the jurisdiction concerned or is not otherwise made available by the CCP, this may call into question about the adequacy and transparency of the jurisdiction’s CCP regulatory regime. The HKMA is prepared to provide assistance on a case-by-case basis if an AI has difficulties in determining whether a CCP meets any one or more of the criteria set out in the definition of “qualifying CCP”. Default fund exposures (applicable to all QCCPs) Q45. Is collateral posted as default fund contributions to a QCCP subject to standard supervisory haircuts in the computation of KAI?

Chapter V Counterparty credit risk Page 19 of 23
A45. No. When using Formula 23K in §226X(4) to calculate the capital requirement (KAI) for default fund contributions made by an AI, there is no need to apply haircuts to the value of any default fund contribution made by the AI in the form of collateral posted. Q46. If the default fund contributions from clearing members of a QCCP are segregated by product types such that default fund contributions for a particular product type are accessible only for that particular product type, should the KAI in Formula 23K be calculated separately for each product type? A46. In this case, KAI in Formula 23K should be calculated separately for each product type. For this purpose, the AI should seek to ascertain whether data provided by the QCCP concerned, the QCCP’s regulator or other bodies enable calculation of KAI on such a basis. Default fund exposures (applicable to QCCPs falling within paragraph (a) of the definition of “qualifying CCP” in §226V(1)) Q47. What if a QCCP, though being informed by its AI clearing members about an increase in risk-weight under §226X(5), fails to provide Kccp calculated based on the increased risk-weight? A47. If the QCCP has not adopted the new risk-weight for Kccp calculation after the lapse of the grace period provided for under §226X(6), an AI may continue to use the Kccp provided by the QCCP for the purposes of §226X(4) provided that the AI scales up the Kccp in a linear way by a factor corresponding to the increase in the risk-weight required under §226X(5), e.g. if the risk-weight is to increase from 20% to 50%, the factor is 2.5. Default fund exposures (applicable to QCCPs falling within paragraph (b) of the definition of “qualifying CCP” in §226V(1)) Q48. Under §226X(4) of the pre-amended Rules (as defined in Schedule 16), there are two methods that an AI can use to calculate the capital requirements for default fund exposures to QCCPs. Is it acceptable for an AI to apply one method to certain QCCPs and at the same time apply another method to other QCCPs? A48. Yes. AIs may select the appropriate method to use separately for each QCCP. Moreover, the selection is not a one-off process. An AI may at any time reconsider its decision and change the method applied to a QCCP.

Chapter V Counterparty credit risk Page 20 of 23
Q49. Under §226X(6) of the pre-amended Rules (as defined in schedule 16), should the calculation of RWA (TE+DF) be performed for each CCP separately? Or should it be performed for all CCP exposures combined? A49. The calculation should be performed for each CCP separately. However, if the default funds of a QCCP are segregated by product types such that the default fund for a particular product type is accessible only for that particular product type, the calculation should be performed for each segregated default fund separately. Portability of trades Q50. §226ZA(6)(c) states that relevant laws, regulations, rules, contractual or administrative arrangements provide that the offsetting transaction between the CCP and the clearing member is highly likely to continue to be indirectly transacted through the CCP….”. Without further guidance, it is difficult to determine what “highly likely” would mean in practice. A50. If there is a clear precedent for transactions being carried over and continued at a CCP and industry intent for this practice to continue, then these factors should be considered when assessing if trades are highly likely to continue to be transacted for the purposes of §226ZA(6)(c). The fact that CCP documentation does not prohibit client trades from being carried over and continued is not sufficient for saying they are highly likely to be carried over and continued. Other evidence such as the criteria in §226ZA(6)(c) is necessary to make this claim. Segregation of collateral Q51. Further explanation is sought as to the meaning of the part of §226ZA(6)(a) which requires collateral to be held under arrangements that prevent any losses to the institution due to default or insolvency of the clearing member and/or any of the clearing member’s other direct clients. A51. The requirement set out in §226ZA(6)(a) essentially means that upon the insolvency of the clearing member, there is no legal impediment (other than the need to obtain a court order to which the AI is entitled) to the transfer of the collateral belonging to the AI to the CCP, to one or more of the other surviving clearing members or to the AI or the AI’s nominee. Hence, AIs should look at the collateral segregation arrangements adopted by CCPs in respect of collateral posted by clearing members and their direct clients, and demonstrate to the satisfaction of the HKMA, that the arrangements can achieve the level of protection required by §226ZA(6)(a) if they want to benefit from the preferential risk-weight of 2% or 4%.

Chapter V Counterparty credit risk Page 21 of 23
Exposures of indirect clients within a multi-level client structure Q52. What necessary modifications to §226ZA(6) are needed for the purposes of §226ZBA(5)(a) and (b)? A52. Because indirect clearing was a concept introduced in the European Markets Infrastructure Regulation (EMIR) and it is less likely that locally incorporated AIs would be part of an indirect clearing arrangement that involves more than four parties (i.e. a QCCP, a clearing member, a direct client of the clearing member and the end client), the discussion below assumes all transactions are cleared according to a “principal-to-principal” clearing model and focuses on the modifications to §226ZA(6) necessary for an AI, as an end client, to determine whether its default risk exposure to the direct client could be risk-weighted as if it were a default risk exposure to the QCCP. §226ZBA(5)(a) requires the conditions in §226ZA(6), with all necessary modifications, to be met for arrangements among the QCCP, clearing member, all clients at levels higher than the AI within the multi-level client structure, and the AI. To this end— (a) in evaluating the arrangements among the QCCP, clearing member and direct client against §226ZA(6)— (i) any reference to “institution” in §226ZA(6) would be construed as a reference to the direct client who provides clearing services to the AI; (ii) §226ZA(6)(a) would be construed in the context of the collateral posted by the direct client in respect of the offsetting transaction related to its CCP-related transaction with the AI. The following illustrates modifications that would generally be needed:
“the offsetting transaction with the CCP for the relevant transaction is identified by the CCP as a clearing an indirect client transaction and the collateral for supporting the offsetting transaction is held by the CCP or the clearing member, or both, as applicable, under arrangements that prevent any losses to the institution direct client due to— (i) the default or insolvency of the clearing member; (ii) the default or insolvency of the clearing member’s other direct clients; and (iii) the joint default or joint insolvency of the clearing member and any of its other direct clients;”; and (iii) §226ZA(6)(c) would be construed to refer to the transfer of the direct client’s positions and assets relating to the AI to a back-up clearing member; and (b) in evaluating the arrangements between the direct client and the AI against §226ZA(6)—

Chapter V Counterparty credit risk Page 22 of 23
(i) §226ZA(6)(a) would be construed in the context of the collateral posted by the AI to the direct client. The following illustrates modifications that would generally be needed:
“the offsetting transaction with the CCP clearing member for the relevant transaction is identified by the CCP clearing member as a clearing an indirect client transaction and the collateral for supporting the offsetting transaction is held by the CCP, or the clearing member or the direct client, or both all or any two of them, as applicable, under arrangements that prevent any losses to the institution due to— (i) the default or insolvency of the clearing member direct client; (ii) the default or insolvency of the clearing member direct client’s other direct clearing clients; and (iii) the joint default or joint insolvency of the clearing member direct client and any of its other direct clearing clients;”; and (ii) §226ZA(6)(c) would be construed to refer to a scenario where the direct client defaults or becomes insolvent. The AI is required to assess whether it is highly likely that the assets and positions held by the direct client for the account of the AI will be transferred to another direct client or another clearing member. The following illustrates modifications that would generally be needed:
“relevant laws, regulations, rules, contractual or administrative arrangements provide that the offsetting transaction between the CCP clearing member and the clearing member direct client is highly likely to continue to be indirectly transacted through the CCP clearing member or another clearing member, or by the CCP clearing member or another clearing member, if the clearing member direct client defaults or becomes insolvent, and in such circumstances, the institution’s positions and collateral with the CCP will be transferred at market value unless the institution requests to close out the positions at market value”. Similar adaptation also applies for the purposes of §100(7)(a)(iii) and (b)(ii), §134(6)(a)(iii) and (b)(ii), §216(3B)(a)(iii) and (b)(ii), and §217(5)(a)(iii) and (b)(ii). Any modification made to the conditions set out in §226ZA(6) should not result in modified conditions that are less stringent than what the Basel Committee has intended. Collateral posted Q53. What treatment must a clearing member apply to collateral that is collected from its direct client and posted to a CCP, but that is not held in a bankruptcy￾remote manner?

Chapter V Counterparty credit risk Page 23 of 23
A53. If the clearing member is not obligated to reimburse the direct client for any loss of such posted collateral in the event that the CCP defaults, the clearing member is not subject to capital requirements for the posted collateral. If the clearing member is obligated to reimburse the direct client for any loss of posted collateral in the event the CCP defaults, the clearing member should compute the capital requirement for the posted collateral held by the CCP as an exposure to the CCP. Q54. Clarification is sought on the interactions among §226ZE, Divisions 1A, 2, 2A and 2B of Part 6A. A54. Unsegregated collateral posted by an AI for securing counterparty credit risk arising from derivative contracts or SFTs should have been included in the calculations conducted under Division 1A, 2, 2A or 2B of Part 6A. Hence, §226ZE(1) and (2) only apply to unsegregated collateral posted by the AI for other purposes (see §226ZE(6A)) and §226ZE(5) and (6) were repealed to avoid duplication of the requirements in §226ZA (see §226ZE(7)(a)). Unless otherwise specified in the BCR, the amounts of default risk exposures calculated under Divisions 1A, 2 and 2A of Part 6A do not include segregated collateral posted by an AI. §226ZE(3) and (4) are intended to confirm similar capital treatment for segregated collateral posted by an AI in relation to its centrally cleared transactions.

---

1/10
VI. Securitization

1. Meaning of “securitization transaction” for the purposes of Part 7 of the BCR
   Q1. How should one determine whether a transaction involving securitization of assets is a securitization transaction for the purposes of the BCR? A1. A transaction involving securitization of assets is a securitization transaction for the purposes of the BCR if there are at least two different tranches of credit risk associated with the assets securitized, where the junior tranche absorbs credit losses first in order to reduce the likelihood of credit losses being allocated to the more senior tranche(s). If the differences between different tranches or different classes are solely due to different coupon rates, yields or maturities, and any credit losses will be allocated to all tranches or classes on a pro-rata basis (i.e. the tranches or classes rank equally with each other in terms of repayment priority), the transaction is not a securitization transaction for the purposes of the BCR. Q2. What is the capital treatment for an exposure held by an authorized institution (“AI”) in its banking book in respect of a transaction involving securitization of assets that is not a securitization transaction for the purposes of the BCR? A2. Capital treatment for exposures (e.g. debt securities) to the transaction held by an AI in its banking book The exposure is a non-securitization exposure and therefore should be risk-weighted by using the BSC approach, STC approach or IRB approach, as the case requires. Capital treatment for assets securitized by an AI through the transaction If the credit risk of the assets is transferred to other parties through the use of credit risk mitigation (“CRM”) (e.g. credit derivative contracts), the AI may take into account the CRM when calculating the risk-weighted amount (“RWA”) of the assets under the BSC approach, STC approach or IRB approach, as the case requires. However, the CRM must be recognized credit risk mitigation under the approach used. If the securitization of the assets is through selling the assets to a special purpose entity (“SPE”) and the assets have been derecognized from the AI’s financial statements/accounts in accordance with prevailing accounting standards, the AI may exclude the assets from its calculation of the RWA for credit risk under the BCR.

2/10
2. Recognition of Credit Risk Transfer under Securitization Transaction
Q3. Please briefly explain the procedures of recognition of credit risk transfer under a securitization transaction. A3. Under the BCR, if an originating institution (“originating AI”) of a securitization transaction intends to obtain capital relief for the underlying exposures of the transaction (i.e. to apply the treatment set out in §230(1), (2) or (2A) of the BCR to the underlying exposures), the AI must have conducted a prudent assessment against the criteria set out in §229(1)(a) or (b) of the BCR, as the case requires, and ascertained that the transaction is an eligible securitization transaction. The AI should then make a notification to the HKMA in the manner prescribed in §230(3) and (4) of the BCR (also see Q6 and Q7 below). The HKMA may, where necessary, require the AI to provide further information to support its claim that the transaction is an eligible securitization transaction. The HKMA will issue an acknowledgement letter to the AI when it receives the notification made by the AI under §230(3) of the BCR. The AI could presume the HKMA has no objection to its application of the treatment under §230(1), (2) or (2A), as the case may be, to the underlying exposures if it does not receive a notice issued by the Monetary Authority (“MA”) under §230(5) of the BCR on or before a date specified in the acknowledgement letter. In general, if a notice is intended to be issued under §230(5), the HKMA will indicate such intention to the AI and will provide an opportunity to the AI to make representations to the HKMA. After representations are made, the HKMA will take them into account in deciding whether to serve the notice concerned. Q4. Can an originating AI of a securitization transaction simply calculate the RWA of the underlying exposures of the transaction as if they had not been securitized (i.e. not applying the treatment under §230(1), (2) or (2A) of the BCR to the underlying exposures although the transaction is able to meet the criteria for being regarded as an eligible securitization transaction)? A4. Yes. This is entirely at the discretion of the AI. If the AI does not act in accordance with §230(3) and (4) of the BCR for the transaction, the AI must treat the transaction as a non-eligible securitization transaction and determine the capital requirement of the underlying exposures of the transaction in accordance with §233 of the BCR. Q5. §230(4)(b)(ii) of the BCR requires a written confirmation provided by a person who has sufficient seniority and authority. What persons are considered by the HKMA as having sufficient seniority and authority? A5. The persons below are considered by the HKMA as having sufficient seniority and authority for this purpose— (a) a qualified external or in-house legal counsel; or

3/10
(b) a manager (within the meaning of §2(1) of the Banking Ordinance (“BO”)) of the originating AI who— (i) is principally responsible for the affairs of the AI specified in §4 (relating to risk management), §7 (relating to internal audits and inspections) or §8 (relating to compliance) of the Fourteenth Schedule to the BO; and (ii) has the necessary knowledge and experience to understand the assessment for supporting the issuance of the confirmation. Q6. In general, what information or supporting documents would the HKMA expect an originating AI to provide for the purpose of §230(4)(b)(i) of the BCR? A6. The AI should provide sufficient information to help the HKMA understand the risk profile of the securitization transaction concerned and the nature and size of the risks retained by the AI. The information generally includes— (a) a summary of the self-assessment of compliance with the criteria set out in §229(1)(a) or (b), as the case requires, including the values of the two ratios mentioned in Q7 below calculated for the transaction; (b) the terms and structure of— (i) the securitization transaction, including any special features (e.g. amortization schemes, excess spread features, call options, materiality thresholds, replenishment mechanisms, loss allocation mechanisms, workout agreements, and, if applicable, the cost and the terms (such as credit events and early termination) of credit protection); and (ii) any recourse, warranty, credit enhancement, credit protection, liquidity facility or other commitments (e.g. to repurchase certain underlying exposures) provided by the AI; (c) details of the underlying exposures of the transaction, including exposure size, composition, and credit quality; (d) the identity of the parties that provide credit enhancement or credit protection to the transaction and their respective obligations and rights, and evidence that the parties (including key investors) to which significant credit risk is transferred are not connected to the AI; and (e) the particulars (e.g. size, nature, terms and conditions) of the credit risk to the underlying exposures retained by the AI, such as first loss positions retained (including partial first loss positions), debt securities held, credit enhancement, liquidity facilities or credit protection granted, and any other form of support provided, by the AI. Q7. Any minimum standards the self-assessment referred to in the answer to Q6 is expected to meet?

4/10
A7. The self-assessment must be well-documented and conducted by a party (whether inside or outside the originating institution) that is independent of the parties responsible for originating—  the securitization transaction concerned; and/or  the underlying exposures of the transaction. The self-assessment must consider each relevant requirement set out in §229(2) of the BCR and in Schedule 9 or 10 to the BCR individually and clearly set out references to the corresponding supporting documents and legal or accounting opinions. The assessment should be updated regularly to ensure compliance on a continuous basis (see Q8). In assessing whether there is significant credit risk transfer under a securitization transaction (see paragraph (a) of Schedule 9 or §1(a) of Schedule 10 to the BCR), the self-assessment must include analyses to—  measure and compare the level of reduction in regulatory capital as a result of applying the treatment under §230(1), (2) or (2A) to the underlying exposures relative to the level of credit risk retained by the originating AI and substantiate that the reduction is justified by a commensurate transfer of credit risk to third parties; and  in the case of a synthetic securitization transaction, substantiate that the credit protection through which the credit risk is transferred does not exhibit any characteristics of a “high-cost credit protection” described in the circular on “Basel Guidance on High-cost Credit Protection” issued by the Monetary Authority on 21 February 2012 and Annex G of the Supervisory Policy Manual module CA-G-5 “Supervisory Review Process”. The AI must identify and measure the risks it retains based on substance rather than form. The AI’s self-assessment should address whether the AI is actually exposed to substantial credit risk even though the transaction might appear to have isolated the AI from the credit risk associated with the underlying exposures. Since different originating AIs may use different methodologies to measure the degree of credit risk transfer achieved by a securitization transaction, the HKMA also expects an originating AI to, besides the figures calculated by using its own methodologies, provide the values of the following two ratios calculated for the transaction to the HKMA for its reference— (a) the ratio of the AI’s aggregate holding of non-senior securitization issues of the transaction (including funding provided by the AI to third parties for their acquisition of the issues) to the total non-senior securitization issues of the transaction; and (b) the ratio of the AI’s aggregate holding of non-senior securitization exposures to the transaction (including not only the holding of non-senior securitization issues, but also other forms of non-senior securitization exposures such as first loss position held or provision of liquidity facilities or credit enhancement to

5/10 the transaction) to the total non-senior securitization exposures to the transaction. For the purpose of the calculation of the two ratios— (a) “AI” includes the AI and any member of its group (i.e. a set of entities whose positions are required to be consolidated with those of the AI under the accounting standards applicable to it); (b) the numerator and the denominator of each ratio must be measured at the RWAs of the securitization issues or securitization exposures calculated in accordance with Part 7 of the BCR; and (c) securitization issues held by the AI and any member of its group resulting from underwriting of the securitization issues can be excluded from the calculation only if— (i) the underwriting has been done at an arm’s length basis; and (ii) there is a timetable for disposal of the holding within a reasonable period. Q8. How will the change in eligibility of a securitization transaction affect the capital treatment of it? A8. The requirements set out in §229(1) of the BCR must be satisfied on a continuing basis throughout the life of the securitization transaction concerned. Hence, the HKMA expects the AI concerned to continue to monitor risks related to the transaction to which it may still be exposed and review its assessment (including any legal advice) against the requirements regularly and whenever there are any changes (e.g. changes in law or regulation, amendments of any provisions of the securitization documentation, or triggering of any call options or early amortization provisions) that may impact the extent to which the requirements are complied with. If the review indicates that the transaction is no longer an eligible securitization transaction, the AI must stop applying the treatment under §230(1), (2) or (2A) to the underlying exposures concerned. However, there might be cases where the change in the eligibility of the transaction is only temporary. In these cases, if the AI wants to reapply the treatment under §230(1), (2) or (2A) after the transaction becomes an eligible securitization transaction again, the AI has to notify the MA such intent in accordance with §230(3) and (4). Q9. One of the requirements that must be met in order for a traditional securitization transaction to be regarded as an eligible securitization transaction is that the originating AI does not maintain effective control, directly or indirectly, over the underlying exposures in the transaction (paragraph (b) of Schedule 9 to the BCR). Please give examples of circumstances where an originating AI will be considered as failing to meet this requirement.

6/10
A9. An originating AI will be considered to have maintained effective control over the underlying exposures being transferred to a third party under a traditional securitization transaction if the AI— (a) has the right to repurchase the underlying exposures, or is otherwise in a position to bring about the repurchase of the underlying exposures, from the third party without the repurchase being made on an arm’s length basis in order to realize benefits from the underlying exposures; or (b) is obligated to retain the risk of the underlying exposures to an extent that exceeds the level necessary to comply with the risk retention requirements applicable to the transaction imposed by regulatory authorities. To avoid doubt, servicing rights retained by the originating AI in respect of the underlying exposures does not of itself constitute effective control of the underlying exposures.
3. Implicit support
Q10. What factors will be considered when determining whether there is “implicit support” for the purpose of §234 of the BCR? A10. The HKMA will determine whether there is implicit support based on the specific circumstances of each case. The following situations, if they are beyond the predetermined contractual obligations of an AI under the securitization transaction concerned, are examples that may be considered as implicit support provided by the AI to the transaction— (a) purchase of the securitization issues of the transaction or repurchase of the underlying exposures of the transaction by the AI— (i) at a price above their fair market value; or (ii) that does not appear to be supported by any legitimate or commercially sound reason (Note 1); (b) sale by the AI of credit exposures at discounted prices into the pool of underlying exposures of the transaction; (c) increase in the AI’s holding of the first loss tranche of the transaction when there is deterioration in the credit quality of the underlying exposures; (d) substitution or replenishment of underlying exposures, or addition of underlying exposures or other assets (including cash), by the AI that will improve the credit quality of the pool of underlying exposures; (e) waiver by the AI of its contractual rights to receive future margin income generated from the transaction;

7/10
(f) renegotiation of any annual fee for any liquidity facility provided by the AI at below market rates; or (g) payments made by the AI under the guise of fulfilling its contractual obligations under the transaction (e.g. representations and warranties) where the amount of the payments exceeds the amount that the AI is contractually obliged to make. The above list of examples is by no means exhaustive. It should also be noted that implicit support may be provided indirectly through other means (e.g. the use of SPEs). Note 1:
The following are examples of situations which may be considered as legitimate or sound commercial reasons– (a) purchase of securitization issues in order not to breach the risk retention requirements applicable to the securitization transaction imposed by regulatory authorities; (b) an obligation of an originating AI to repurchase underlying exposures due to breach of a representation or warranty given by the AI (i.e. the situation allowed by paragraph (g)(ii)(A) of Schedule 9 to the BCR); (c) an obligation of an originating AI to repurchase an underlying exposure to a borrower in order to enable the AI to provide further advances to the borrower and the repurchase does not expose the AI to excessive credit risk (a case that falls within paragraph (g)(ii)(B) of Schedule 9); or (d) there is a case to facilitate a more cost-effective workout of those underlying exposures that have defaulted. Q11. What does “predetermined contractual obligations” mean in the definition of “implicit support” in the BCR? A11. In the context of credit support provided by an AI under a securitization transaction, “predetermined contractual obligations” means— (a) the support that may be given by the AI is expressly set out in the contractual and marketing documents for the transaction; (b) the nature of the support that may be given by the AI is precisely described in the documents referred to in paragraph (a) above; and (c) the maximum degree of support that may be given by the AI can be ascertained at the time of the securitization both by the AI or by any person based on the information given in the marketing documents for the transaction. The credit support that may be provided by the AI within its predetermined contractual obligations must be—

8/10
(a) if the AI is the originating institution of the transaction concerned—taken into account in the assessment of significant credit risk transfer for the purpose of determining whether the transaction is an eligible securitization transaction under §229 of the BCR. The assessment should be made on the basis that the AI will provide support to the maximum degree possible under the contractual documents for the transaction; and (b) if the AI is the originating institution of the transaction and the transaction is an eligible securitization transaction or the AI is an investing institution— included in the calculation of the AI’s RWA for credit risk, assuming that the AI has provided support to the maximum degree possible under the contractual documents for the transaction (unless otherwise stated in the BCR, e.g. in the case of servicer cash advance facilities).
4. Classification of underlying exposures of a securitization transaction
Q12. Under what circumstances the MA may, under §16(5) of the BCR, require an AI to classify a pool of underlying exposures of a securitization transaction as an SA pool. A12. The circumstances include securitization transactions with highly complex loss allocations, tranches of which credit enhancement could be eroded for reasons other than pool losses, and tranches backed by a pool with high correlations within the pool (e.g. high concentration to a single sector or a geographical region). Incidentally, if an AI uses the IRB approach to calculate the credit risk of non￾securitization exposures and the AI holds a securitization exposure backed by a pool of underlying exposures that contains only IRB underlying exposures (see the definition in §16(7) of the BCR), the HKMA expects the AI to be able to obtain sufficient information to estimate the capital requirements for the IRB underlying exposures by using the IRB approach. If the AI fails to do so for all the IRB underlying exposures, this may cast doubt on whether the AI meets the due diligence requirements in §15A. The AI should therefore inform the HKMA of the situation and demonstrate to the HKMA why it cannot do so. Depending on the explanation given by the AI, the HKMA may consider requiring the AI to take necessary actions to enhance its credit risk management system.
5. SEC-IRBA and SEC-SA
Q13. In Part 7 of the BCR there are provisions that require an AI to verify or assess prudently certain matters (e.g. §251(5) and §256(2)). What is the HKMA’s supervisory approach in this regard? A13. The verification or assessment performed by the AI must be well-documented and supported by detailed analysis accompanied by proper supporting documents. The documentation must be made available to the HKMA for review upon request.

9/10
For example, if an AI intends to exclude an SPE’s exposure, which is resulted from an interest rate swap entered into by the SPE, from the pool of underlying exposures under §256(2) on the ground that the risk of the SPE’s exposure has been mitigated, the AI should ascertain that the SPE has adopted measures to eliminate or at least significantly reduce the potential risk from a default of the swap provider. Examples of such measures could be—  cash collateralization of the market value of the swap, coupled with a margin agreement that provides for prompt additional payments when the market value of the swap increases; and  minimum required credit quality of the swap provider with an obligation to post collateral or present an alternative swap provider without any costs for the SPE in the event of a credit deterioration on the part of the original swap provider. The AI should also assess the extent of risk posed by the SPE’s exposure to the securitisation exposure held by the AI.
6. Credit enhancement
Q14. Please provide examples of credit enhancement.
A14. Credit enhancement can be in the form of subordination, excess spread, over￾collateralization, guarantee or insurance, etc. Q15. If the amount of credit enhancement provided by an AI to a securitization transaction cannot be readily ascertained (e.g. the exposure is not limited to a specified amount), how should the AI calculate the RWA of the credit enhancement? A15. In such a case, the AI should calculate the RWA of the credit enhancement as if the credit enhancement covered the full value of the securitization exposures that are the subject of the credit enhancement.
7. Gain-on-sale
Q16. Should the amount of capital deduction in respect of gain-on-sale arising from a securitization transaction be the “accumulated” amount recorded in the originating AI’s books (i.e. including the amount of gain-on-sale recognized for the current year and the amounts recognized in previous years)? A16. The amount of deduction should be the “accumulated” amount of the gain-on-sale recorded in the originating AI’s books because such an amount represents the total amount of increase in the AI’s CET1 capital resulted from the sale of the underlying exposures concerned.

10/10
8. Recognition of External Credit Assessment Institutions
Q17. Is the HKMA’s policy paper of “Recognition of External Credit Assessment Institutions” (Sep 2013) relevant to the SEC-ERBA under the revised securitization framework? A17. Yes. An AI may use the SEC-ERBA to risk-weight its securitization exposures booked in the banking book only if the credit ratings concerned are issued by an external credit assessment institution recognized by the HKMA in accordance with the criteria set out in the policy paper. For this purpose— (a) paragraph 3, second bullet, subparagraph (ii) of the policy paper should be read as “the securitization external ratings-based approach (“SEC-ERBA”) for securitization exposures booked in the banking book – Section III”; and (b) any references in the policy paper to “STC(S) approach” or “RBM” should be construed to mean the SEC-ERBA.