Rule · HKMA · 25 Sep 2015
Locally incorporated AIs must allocate private sector credit exposures to jurisdictions on an ultimate risk basis, using the immediate obligor's location unless the ultimate obligor's location differs under the MA(BS)21 completion instructions; if neither can be determined without disproportionate effort, exposures are allocated where booked. For CIS and securitisation exposures, the jurisdiction with at least 30% of underlying obligors is used, otherwise pro rata; IRB retail pools are divided into sub-pools by jurisdiction, or RWA split by EAD ratio. Exposures booked in jurisdictions the HKMA lists as lacking economic nexus and no Basel III CCyB framework must be allocated to Hong Kong unless the AI evidences genuine nexus.
Supervisory Policy Manual
CA-B-3
Countercyclical Capital Buffer (CCyB)
– Geographic Allocation of Private
Sector Credit Exposures
V.1 – 25.09.2015
This module should be read in conjunction with the Introduction and with the Glossary, which contains an explanation of abbreviations and other terms used in this Manual. If reading on-line, click on blue underlined headings to activate hyperlinks to the relevant module.
—————————
To provide guidance to AIs on determining the geographic allocation of private sector credit exposures1 for the purposes of implementing the Countercyclical Capital Buffer (CCyB) within the capital adequacy framework for AIs incorporated in Hong Kong.
A statutory guideline issued by the MA under the Banking Ordinance (BO), §7(3).
This is a new guideline.
To all locally incorporated AIs.
1. Introduction
1.1 Terminology
1.2 Background
2. Determining the RWA corresponding to an AI’s private sector credit exposures in a jurisdiction (RWAj)
[1] In this SPM module, the term “private sector credit exposure” has the same meaning as in BCR §3N.
Supervisory Policy Manual
CA-B-3
Countercyclical Capital Buffer (CCyB)
– Geographic Allocation of Private
Sector Credit Exposures
V.1 – 25.09.2015
2.1 Aggregating banking book and trading book private sector credit exposures for each geographic location
2.2 Banking book and trading book exposures for which RWA for credit risk is calculated under the BCR
2.3 Trading book exposures for which a market risk capital charge for specific risk is calculated under BCR Part 8
3. Determining the geographic location of obligors on an ultimate risk basis
1.1.1 Unless otherwise specified, abbreviations and terms used in this module follow those used in the Banking (Capital) Rules (“BCR”) and in the Banking (Disclosure) Rules (“BDR”). In this module, “AI” means “locally incorporated AI” and “BO” means “Banking Ordinance” unless otherwise specified.
1.2.1 The Basel III regulatory capital standards issued by the Basel Committee provide for the implementation of a CCyB beginning on 1 January 2016.
1.2.2 The MA has made the BCR under BO §97C and the BDR under BO §60A and has, by the Banking (Capital) (Amendment) Rules 2014 and the Banking (Disclosure) (Amendment) Rules 2014, incorporated provisions for the imposition of capital requirements arising from the operation of the CCyB into the BCR and for corresponding disclosures into the BDR respectively.
1.2.3 The MA has issued SPM module CA-B-1 to provide an overview of the CCyB framework in Hong Kong and
Supervisory Policy Manual
CA-B-3
Countercyclical Capital Buffer (CCyB)
– Geographic Allocation of Private
Sector Credit Exposures
V.1 – 25.09.2015
describe the MA’s approach to taking decisions with regard to the setting of the CCyB rates applicable to AIs.
1.2.4 As set out in BCR §3O(1) and explained in SPM CA-B-1 Section 2, an AI must determine its own AI-specific CCyB rate 2 as the weighted average of the applicable jurisdictional CCyB rates3 , effective at the date for which the determination is made, in respect of the jurisdictions (including Hong Kong) where the AI has private sector credit exposures.4 The weight to be attributed to a given jurisdiction’s applicable CCyB rate is the ratio of the AI’s aggregate risk-weighted amount for its private sector credit exposures (in both the banking book and the trading book) in that jurisdiction (RWAj) to the sum of the AI’s aggregate RWAj across all jurisdictions in which the AI has private sector credit exposure.
1.2.5 According to BCR §3O(2), the jurisdiction in which an AI is considered to have private sector credit exposures must be determined by the AI, where possible, on an ultimate risk basis. That is, private sector credit exposures must be allocated to the jurisdiction where the risk ultimately lies to the best of the AI’s knowledge and information (see BCR §3N). If it is not possible for an AI to determine the jurisdiction in which the AI has private sector credit exposures on an ultimate risk basis, such exposures are to be allocated to the jurisdiction where the exposures are booked (see BCR §3O(3)).
1.2.6 This module sets out the MA’s expectations on how an AI should allocate private sector credit exposures, and the corresponding risk-weighted amount (RWA), to different jurisdictions on an ultimate risk basis in accordance with the provisions in the BCR mentioned above, in order to determine the AI’s aggregate RWAj for its private sector
[2] The term “AI-specific CCyB rate” corresponds to the term “CCyB ratio” as defined by formula 1A in BCR §3O(1).
[3] The terms “applicable jurisdictional CCyB rate” and “jurisdiction’s applicable CCyB rate” correspond to the term “applicable JCCyB ratio” as defined in BCR §3N.
[4] As defined in BCR §3N, “private sector credit exposures” exclude exposures to banks regardless of whether the latter are under public sector or private sector ownership.
Supervisory Policy Manual
CA-B-3
Countercyclical Capital Buffer (CCyB)
– Geographic Allocation of Private
Sector Credit Exposures
V.1 – 25.09.2015
credit exposures in both the banking book and the trading book in each jurisdiction.
2.1.1 As set out in BCR §3O(1), the aggregate RWAj of an AI’s private sector credit exposures in jurisdiction j for the purposes of calculating the weight to be attributed to that jurisdiction’s applicable CCyB rate (see para. 1.2.4 above) is the sum of the following two components:
(1) Banking book and trading book exposures for which RWA for credit risk is calculated under the BCR – this includes the following RWA corresponding to the AI’s private sector credit exposures in jurisdiction j –
i) the RWA for non-securitisation exposures calculated under the standardised (credit risk) (STC) approach (BCR Part 4), or under the basic (BSC) approach (BCR Part 5), or under the IRB approach (BCR Part 6), or for central counterparties (CCPs) under BCR Part 6A, Division 4; and
ii) the RWA for securitisation exposures calculated under BCR Part 7;
and
(2) Trading book exposures for which a market risk capital charge for specific risk is calculated under BCR Part 8 – this includes the RWA corresponding to the AI’s trading book private sector credit exposures in jurisdiction j derived by multiplying by 12.5 the aggregate of the market risk capital charge for specific risk for those exposures calculated in accordance with BCR Part 8.
Supervisory Policy Manual
CA-B-3
Countercyclical Capital Buffer (CCyB)
– Geographic Allocation of Private
Sector Credit Exposures
V.1 – 25.09.2015
2.1.2 AIs are expected to calculate the above two RWAj components as described, respectively, in Sub-sections 2.2 and 2.3 of this module.
2.2.1 General approach. An AI should first determine the geographic location of its banking book and trading book exposures for which credit risk RWA is calculated under the BCR (see para. 2.1.1(1) above) by identifying in which jurisdiction the obligor(s) 5 corresponding to each exposure is/are located (if possible on an ultimate risk basis as described in Section 3 below). In the case of collective investment schemes (CISs), securitisation exposures and pools of retail exposures under the IRB approach, if all obligors of the respective underlying exposures are located in the same jurisdiction, that jurisdiction should be used as the geographic location of the exposure. For each identified jurisdiction, the AI should then aggregate the RWA of all exposures whose obligors are located in that jurisdiction.
2.2.2 Special cases. The following approaches should be applied for determining the geographic location of exposures in the listed special cases:
(1) Exposure to a CIS:
(a) If the obligors of the underlying exposures of the CIS are located in multiple jurisdictions, the jurisdiction for which obligors represent in aggregate the highest proportion of the underlying exposures, as compared with the respective proportion for other jurisdictions, should be used as the geographic location of the exposure to the respective CIS, subject to the said proportion being at least 30%. In other
[5] The obligor in this case is the natural or legal person who is the AI’s counterparty to a credit exposure, or the issuer of a financial instrument not included in the trading book, or the counterparty to any other non-trading book exposure (see also definition of “obligor” in the BCR).
Supervisory Policy Manual
CA-B-3
Countercyclical Capital Buffer (CCyB)
– Geographic Allocation of Private
Sector Credit Exposures
V.1 – 25.09.2015
words, a form of “look through” approach should be applied to determine where the risk of the CIS is predominantly located.
(b) If however no jurisdiction reaches the above 30% threshold or it is not possible to identify without disproportionate effort a single jurisdiction for which obligors represent the highest proportion of the underlying exposures, the RWA corresponding to the exposure to the respective CIS should be allocated among jurisdictions in the same proportions as the aggregate of the AI’s other exposures have been directly allocated to identified geographic locations. So, for instance, if the AI’s other exposures have been directly allocated 20% to jurisdiction A, 30% to jurisdiction B and 50% to jurisdiction C, the exposure to the CIS will be allocated on the same 20 / 30 / 50 split to jurisdictions A, B and C.
(2) Securitisation exposure: The geographic location of a securitisation exposure in either the banking book or the trading book should be determined in the same way as for an exposure to a CIS (see above).
(3) Pool of retail exposures under IRB: If the pool contains exposures located in more than one jurisdiction, then an AI is expected to divide exposures into sub-pools, one for each jurisdiction in which the respective obligors are located. The AI should then determine the RWA of each sub-pool based on the internal model(s) and methodologies for which the AI has the MA’s approval to calculate credit risk for its retail exposures using the IRB approach. If however it is not possible to follow this method without disproportionate effort, the AI should determine the RWA of each sub-pool by multiplying the total RWA of the pool by the following ratio: (sub-total of EAD in the sub-pool corresponding to exposures in the given jurisdiction) / (total EAD in the pool).
Supervisory Policy Manual
CA-B-3
Countercyclical Capital Buffer (CCyB)
– Geographic Allocation of Private
Sector Credit Exposures
V.1 – 25.09.2015
(4) Specialised lending: The geographic location of the exposure should be the location of the specific physical asset which generates the income that is the primary source of repayment of the obligation, except in the case of commodities and movable physical assets (e.g. ships and aircraft). In the latter case, the general approach of para. 2.2.1 (subject to Section 3) should prevail.
2.3.1 Determining the RWA corresponding to an AI’s trading book private sector credit exposures in a jurisdiction (see BCR §3O(1)). An AI should first identify in which jurisdiction the obligor(s) 6 is/are located (if possible on an ultimate risk basis as described in Section 3 below) in respect of the AI’s interest rate exposures (non-securitization and securitization 7 ) and equity exposures which are subject to a market risk capital charge for specific risk. Then, the AI should proceed as follows, depending on whether the AI calculates a market risk RWA based on the standardized (market risk) approach (STM approach) or on the internal models approach (IMM approach):
(1) STM approach: The RWA for specific risk calculated using the STM approach under the BCR for each exposure should be allocated to the jurisdiction in which the obligor(s) associated with the exposure is/are located.
(2) IMM approach: Solely for the purposes of calculating its AI-specific CCyB rate, the AI should divide its relevant exposures into sub-portfolios, one for each
[6] The obligor in this case is the natural or legal person who is the issuer of a financial instrument booked in the trading book, or the counterparty to any other trading book exposure (see also definition of “obligor” in the BCR).
[7] The geographic location of securitisation exposures in the trading book should be determined as set out in para. 2.2.2(2) above.
Supervisory Policy Manual
CA-B-3
Countercyclical Capital Buffer (CCyB)
– Geographic Allocation of Private
Sector Credit Exposures
V.1 – 25.09.2015
jurisdiction in which the respective obligors are located. The AI should then apply to each sub-portfolio the same IMM approach methodology which the AI uses to calculate its market risk capital charge for specific risk (including as applicable components corresponding to: VaR, stressed VaR, incremental risk charge (IRC), comprehensive risk charge (CRC) and supplemental capital charge (SCC) for a correlation trading portfolio), to compute a sub-portfolio market risk capital charge for specific risk for each jurisdiction. The RWA for specific risk (i.e. the market risk capital charge for specific risk multiplied by 12.5) calculated for all the relevant exposures of the AI under the IMM approach should then be allocated on a pro rata basis to each jurisdiction based on the sub-portfolio market risk capital charges for specific risk.
Notwithstanding the above, AIs which have been exempted by the MA from market risk calculation under BCR §22(1) will be considered as not having exposures subject to a market risk capital charge for specific risk for the above purposes (see BCR §3O(1)).
3.1.1 Ultimate risk basis. As set out in BCR §3O(2) and §3N (see para. 1.2.5 above), the jurisdiction in which an AI is considered to have private sector credit exposure should be determined by the AI, where possible, on an “ultimate risk basis”. “Ultimate risk basis” means the allocation of exposures to the jurisdiction where the risk ultimately lies, defined as the location where the “ultimate obligor” resides. AIs should apply the following guidelines to implement this principle.
3.1.2 Immediate obligor. The location of an immediate obligor in respect of an exposure is the jurisdiction where the counterparty in the corresponding contract or the issuer of the corresponding security is ordinarily resident (in the
Supervisory Policy Manual
CA-B-3
Countercyclical Capital Buffer (CCyB)
– Geographic Allocation of Private
Sector Credit Exposures
V.1 – 25.09.2015
case of a natural person), or has its registered office (or its actual centre of administration if that is in a different jurisdiction from its registered office) (in the case of a legal person). The location of the immediate obligor should be used as the location of the obligor for the purposes of Sub-sections 2.1 and 2.2 above unless, in the circumstances referred to in para. 3.1.3 below, an ultimate obligor’s location should be used instead. If neither the location of the ultimate obligor nor that of the immediate obligor can be determined without disproportionate effort, the exposure should be allocated to the jurisdiction where it is booked.
3.1.3 Ultimate obligor. The location of the ultimate obligor should be used for the purposes of Sub-sections 2.1 and 2.2 above when it differs from that of the immediate obligor in the circumstances specified in the completion instructions for MA(BS)21 “Return of International Banking Statistics”. 8 In addition to the cases contemplated in the completion instructions for MA(BS)21, AIs are expected to apply the following guidelines in determining the ultimate obligor or the location of the ultimate obligor: ,9
(1) In line with the Completion Instructions for MA(BS)21, to the extent that credit risk has been mitigated by means of a recognized guarantee or a recognized credit derivative contract, the ultimate obligor of the credit protection covered portion of the exposure is the credit protection provider under the recognized guarantee or the recognized credit derivative contract. The RWA of the credit protection covered portion is then allocated to the jurisdiction where the ultimate obligor is located. The credit protection covered portion of the exposure is a private sector credit
[8] As of the date of issue of this SPM module, the relevant text is contained in para. 26 of the completion instructions for MA(BS)21.
[9] Future revisions to the completion instructions for MA(BS)21 will have precedence over the following provisions in this SPM module in case of any contradiction.
Supervisory Policy Manual
CA-B-3
Countercyclical Capital Buffer (CCyB)
– Geographic Allocation of Private
Sector Credit Exposures
V.1 – 25.09.2015
exposure only if the credit protection provider is a private obligor as defined in BCR §3N.
(2) To the extent that credit risk has been mitigated by means of the posting of recognized collateral, the following cases should be considered in determining the geographic location of the ultimate risk:
(a) For the simple approach under the STC approach (as provided for under BCR Part 4, Division 6) or under the BSC approach (as provided for under BCR Part 5, Division 5), AIs should allocate the RWA of the credit protection covered portion to the jurisdiction where the recognized collateral, or the issuer / obligor of the recognized collateral, is located, subject to (c), (d), (e) and (f) below.
(b) For the comprehensive approach under the STC approach and for the IRB approach, the location of the collateral or of the issuer / obligor of the collateral (or of the underlying exposures of the collateral) is deemed irrelevant for purposes of the CCyB ratio calculation.
(c) Where the recognized collateral is in the form of real property, the geographic location of the collateral is the jurisdiction where the collateral is physically located.
(d) Where the recognized collateral is in the form of securities which are neither credit-linked notes nor securitization issues, (subject to (f)) the geographic location of the collateral is the jurisdiction where the issuer of the securities resides.
(e) Where the recognized collateral is in the form of shares or units in a CIS or securities in a securitisation issue, the geographic location of the collateral is determined as per para. 2.2.2(1) above.
(f) Where the recognized collateral is in the form of cash on deposit held at a bank, or where the
Supervisory Policy Manual
CA-B-3
Countercyclical Capital Buffer (CCyB)
– Geographic Allocation of Private
Sector Credit Exposures
V.1 – 25.09.2015
collateral is in the form of securities and the issuer / obligor of the securities is not a private obligor as defined in BCR §3N, the credit protection covered portion of the exposure is not a private sector credit exposure for CCyB purposes and therefore its RWA should not be considered as part of RWAj for the purposes of BCR §3O(1).
(3) In circumstances where, in the HKMA’s judgement, exposures booked in a jurisdiction and/or to obligors residing in that jurisdiction (whether an off-shore financial centre or otherwise) typically do not appear to have an economic nexus with that jurisdiction or it appears unlikely that much of the proceeds will actually be used in that jurisdiction, and where the jurisdiction in question has not implemented, and does not operate, a Basel III countercyclical capital buffer framework, the HKMA may notify AIs and post on its website a specific list of such jurisdictions. If the MA does so, an AI’s private sector credit exposures booked in those jurisdictions and/or to obligors residing in those jurisdictions should be allocated to Hong Kong for the purposes of calculating the AI’s specific CCyB rate, unless on a case-by-case basis the AI can present evidence acceptable to the HKMA that an exposure indeed has a genuine nexus with such jurisdictions.
—————————
Contents
Glossary
Home
Introduction
Note from RegAlert. AI assistants can read this document in full, and search 70,000+ more, through the RegAlert MCP connector (https://mcp.regalert.today/mcp). Free with an account. How to connect ChatGPT, Claude or Cursor.
RegAlert holds no document that amends or replaces this one. This is not a check that it is in force: confirm with the regulator.
“an AI should make reference to the guidance provided in section 3 of the HKMA’s Supervisory Policy Manual module CA-B-3 Countercyclical Capital Buffer”
“guidance provided in the HKMA’s Supervisory Policy Manual module CA-B-3 Countercyclical Capital Buffer (CCyB)”
More like this from HKMA
HKMA published 5 documents in the last 30 days. We email you each new one the day it's published.
Source: Hong Kong Monetary Authority — original document · Summary generated with machine assistance and checked by an AI reviewer before publication; the authoritative text is the regulator's original document. How RegAlert works
In RegAlert: added