Resolution Planning – Liquidity and Funding in Resolution

Resolution Regime – Code of Practice LFIR-1 Resolution Planning – Liquidity and Funding in Resolution 22.07.2022 1 Purpose Section 196 of the Financial Institutions (Resolution) Ordinance (Cap. 628) (“FIRO”) empowers the Monetary Authority (“MA”) as resolution authority in relation to banking sector entities to issue a code of practice (“Code of Practice”) about any matter relating to the functions given to the MA as a resolution authority by the FIRO. This publication is a chapter of the Code of Practice. The purpose of this chapter is to provide guidance on the MA’s expectations in relation to the capabilities and arrangements of an authorized institution (“AI”) to support liquidity and funding in resolution (“LFIR”). This chapter of the Code of Practice should be read in conjunction with chapter RA-2, “The HKMA’s Approach to Resolution Planning” (“RA-2”)1 and chapter CI-1, “Resolution Planning – Core Information Requirements” (“CI-1”)2 of the Code of Practice. Application To all AIs. Structure

1. Introduction
2. Scope of entities and currencies
3. Methodology for estimating liquidity and funding needs in resolution
4. Capabilities for measuring, monitoring and reporting on liquidity and funding in resolution

1 https://www.hkma.gov.hk/media/eng/doc/key-functions/banking-stability/resolutions/RA￾2_The_HKMA_approach_to_resolution_planning.pdf 2 https://www.hkma.gov.hk/media/eng/doc/key-functions/banking-stability/resolutions/CI￾1_Resolution_Planning_Core_Information_Requirements.pdf

Resolution Regime – Code of Practice LFIR-1 Resolution Planning – Liquidity and Funding in Resolution 22.07.2022 2 5. Capabilities for identifying and mobilising collateral and for monitoring aggregate asset encumbrance levels 6. Governance 7. Testing and validation 8. The MA’s approach to implementation

Resolution Regime – Code of Practice LFIR-1 Resolution Planning – Liquidity and Funding in Resolution 22.07.2022 3

1. Introduction 1.1 A potential impediment to the orderly resolution of an AI relates to its inability to assess funding needs and access funding in resolution. In order to address this potential impediment to resolution, an AI should, in business as usual (“BAU”), have the capabilities and arrangements (“LFIR capabilities and arrangements”) such that, in the lead up to and during resolution, it may (i) reasonably anticipate its liquidity and funding needs in resolution; (ii) appropriately monitor, manage and report on, its liquidity position (including the amount and nature of available liquid assets 3 ); and (iii) effectively identify and mobilise collateral and monitor aggregate asset encumbrance levels. In turn, an AI’s ability to effectively perform these activities crucially depends on whether the AI has in place appropriate liquidity risk management capabilities and arrangements which address resolution-specific risks and factors. The Financial Stability Board (“FSB”) has identified liquidity and funding as important topics to be addressed in resolution planning and has issued related publications including its guidance on the Funding Strategy Elements of an Implementable Resolution Plan4 , complementing the Key Attributes of Effective Resolution Regimes for Financial Institutions 5 and the Guiding principles on the temporary funding needed to support the orderly resolution of a global systemically important bank6 . The MA’s expectations set out in this chapter are broadly in line with the FSB’s guidance as to banks’ capabilities. 1.2 As part of the bilateral resolution planning programme, the MA expects an AI to be able to demonstrate that it has in place the LFIR capabilities and arrangements. An AI’s LFIR capabilities and arrangements developed in line with this chapter are expected to take into account, as appropriate, the various phases including: (i) the lead up to resolution; (ii) the stabilization phase; and (iii) the post-stabilization restructuring phase. 1.3 The MA may consider an AI’s LFIR capabilities and arrangements

3 For the purpose of this chapter, these include the liquid and unencumbered assets identified by an AI to be appropriate for sale or for use as collateral in order to satisfy its funding and liquidity needs in resolution. 4 https://www.fsb.org/wp-content/uploads/P210618-3.pdf 5 https://www.fsb.org/wp-content/uploads/r_141015.pdf 6 https://www.fsb.org/wp-content/uploads/Guiding-principles-on-the-temporary-funding-needed-to-support￾the-orderly-resolution-of-a-global-systemically-important-bank-%E2%80%9CG-SIB%E2%80%9D.pdf

Resolution Regime – Code of Practice LFIR-1 Resolution Planning – Liquidity and Funding in Resolution 22.07.2022 4 when developing resolution plans to support strategies for securing an orderly resolution of the AI (or of the AI’s holding company) 7 . The MA may also consider the LFIR capabilities and arrangements of an AI when conducting a resolvability assessment of the AI (or of the AI’s holding company) to help determine whether there are any impediments to its orderly resolution8 . 1.4 If an AI does not meet the expectations set out in this chapter, the MA may form the opinion that a significant impediment to the orderly resolution of the AI exists, and the MA may direct the AI, pursuant to section 14 of the FIRO, to take any measures in relation to its structure (including group structure), operations (including intra-group dependencies), assets, rights or liabilities that are, in the opinion of the MA, reasonably required to remove or mitigate the effect of this impediment. 1.5 Furthermore, it is expected that in the lead up to and during resolution, an AI may face strictures in terms of its sources of liquidity and funding along with a heightened rate of liquidity outflows, in part due to information asymmetries, which may impact the availability of funding especially from private sector sources until confidence in the AI is restored. There may be circumstances where temporary public backstop funding may be needed to support the orderly resolution of an AI, subject of course, to the lending criteria for such funding being met. 1.6 When resolving an AI under the FIRO, section 178(3) of the FIRO requires the MA to first consider the extent to which the AI’s own resources can be utilised before he may use the public money standing to the credit of the resolution funding account in applying a stabilization option to an AI. This is consistent with one of the resolution objectives in the FIRO relating to the protection of public money. An AI’s LFIR capabilities and arrangements as regards the nature and extent of the AI’s resources will therefore be useful in informing the MA’s approach to resolution funding arrangements pursuant to Part 12 of the FIRO. 1.7 In line with the MA’s proportionate and risk-based approach to resolution planning, an AI’s LFIR capabilities and arrangements for meeting the expectations outlined in this chapter are expected to be commensurate with the nature, scale and complexity of the AI’s

7 See section 13(1) of the FIRO. 8 See section 12(1) of the FIRO.

Resolution Regime – Code of Practice LFIR-1 Resolution Planning – Liquidity and Funding in Resolution 22.07.2022 5 business and liquidity risk profiles. Furthermore, in developing LFIR capabilities and arrangements to meet the expectations set out in this chapter, an AI may leverage its existing capabilities and arrangements developed for other purposes (e.g. compliance with supervisory requirements and standards for liquidity risk management, contingency funding planning, recovery planning, etc.). In such circumstances, the AI should be able to demonstrate to the MA that such capabilities and arrangements adequately address the resolution-specific liquidity risks and factors outlined in this chapter. 1.8 The rest of this chapter sets out the MA’s expectations regarding an AI’s LFIR capabilities and arrangements. Section 2 provides guidance on the scope of entities and currencies in respect of which the expectations outlined in this chapter apply. Section 3 outlines the expectation that an AI should develop and document a methodology to estimate ex ante the liquidity and funding needed to facilitate orderly resolution. Section 4 outlines expectations concerning an AI’s capabilities to measure, monitor and report on liquidity and funding needs as well as available liquid assets in a resolution scenario. Section 5 provides guidance on the MA’s expectations on the AI’s capabilities to assess the need for and availability of third-party funding, and for identifying and mobilising collateral and monitoring aggregate asset encumbrance levels. Section 6 addresses expectations on governance arrangements related to LFIR. Section 7 sets out expectations regarding an AI’s testing and validation activities. Section 8 sets out the MA’s approach to implementing this chapter.

Resolution Regime – Code of Practice LFIR-1 Resolution Planning – Liquidity and Funding in Resolution 22.07.2022 6 2. Scope of entities and currencies 2.1 The MA expects an AI to identify the entities of the AI’s group which are considered material for liquidity and funding in resolution purposes (“material LFIR entities”). In this regard, the AI is expected to assess materiality in the context of the preferred resolution strategy determined by the MA and may consider the "material entities" identified pursuant to CI-1 as a starting point9 . When determining whether an entity is material for the purpose of this chapter, an AI is expected to consider, inter alia, the entity's role in terms of liquidity management responsibilities or as a significant liquidity provider or user within the group. 2.2 Additionally, the AI has the onus of identifying currencies that are material for LFIR purposes (“material currencies”). At a minimum, material currencies are expected to include all currencies which account for 5% or more of the total liabilities (including shareholders’ funds)10 of either a material LFIR entity or the AI’s group. 2.3 An AI is expected to be able to explain and justify to the MA the basis for the entities and currencies being included in, or excluded from, its scope of material LFIR entities and material currencies, respectively, and should be able to demonstrate that the liquidity risks arising from non-material LFIR entities and non-material currencies do not pose a significant risk to orderly resolution. Furthermore, an AI is expected to keep under review its scope of material LFIR entities and material currencies. 2.4 In determining the approach for consolidated computations or reports for LFIR purposes, an AI may make reference to its approach to consolidated reporting under the Banking (Liquidity) Rules (Cap. 155Q) as a starting point, and should account for resolution-specific considerations including the preferred resolution strategy determined by the MA. 2.5 For the purpose of meeting the expectations set out in this chapter, an

9 As stated in CI-1, “material entities” should, at a minimum, include (i) the AI itself; (ii) any holding company of the AI which is incorporated in Hong Kong; and (iii) any downstream subsidiary or any branch (both in Hong Kong and overseas) of the AI which represents a significant portion of its total balance sheet or business activities. 10 In addition to the minimum set of currencies identified as material by reference to total on-balance sheet liabilities, an AI is expected to consider whether additional currencies should be included in its scope of material currencies, taking into account, for instance, off-balance sheet items.

Resolution Regime – Code of Practice LFIR-1 Resolution Planning – Liquidity and Funding in Resolution 22.07.2022 7 AI is expected to have in place LFIR capabilities and arrangements, covering each material currency, at the level of each material LFIR entity and at the level of the AI’s group.

Resolution Regime – Code of Practice LFIR-1 Resolution Planning – Liquidity and Funding in Resolution 22.07.2022 8 3. Methodology for estimating liquidity and funding needs in resolution 3.1 An AI is expected to have in place a documented methodology for estimating ex ante the liquidity and funding needed 11..in order to facilitate an orderly resolution. The methodology should, among other things, involve the simulation of behavioural and contractual cash inflows and outflows associated with on-balance sheet assets and liabilities, and off-balance sheet items. Furthermore, an AI’s methodology should model and account for projected mismatches between cash inflows and outflows (including peak liquidity needs), whether on an intraday basis or over the period of the projected resolution scenario12 . 3.2 An AI’s methodology for estimating the liquidity and funding needed to facilitate orderly resolution should also account for, among other things, the amount and nature of the available liquid assets that will be required to satisfy operational needs in resolution, including intraday liquidity needs (e.g. cash held for clearing and settlement purposes), operating expenses (e.g. salaries and rent), and working capital needs (e.g. cash for ATMs). 3.3 To the extent relevant, an AI’s methodology for estimating liquidity and funding needs in resolution should model and account for, inter alia, the following factors influencing the key drivers which may impact the liquidity position of each material LFIR entity in the lead up to and during resolution: (a) the liquidity required to satisfy obligations related to payment, clearing and settlement activities, having regard to the liquidity effects of any risk management measures which may be adopted by a relevant financial market infrastructure (“FMI”) or FMI intermediary;

11 For the purposes of this chapter, reference to “liquidity and funding needs” means net outflows. 12 An AI may seek to satisfy the expectations in this chapter by leveraging capabilities and arrangements on the cash-flow approach to managing liquidity risk as well as on stress-testing and scenarios analysis which it already has in place for meeting the going-concern supervisory expectations placed on it (see sections 4 and 5 of the Supervisory Policy Manual (“SPM”) module LM-2 “Sound Systems and Controls for Liquidity Risk Management” (“LM-2”)). In such circumstances, for the purposes of assessing resolvability, the MA expects the AI to be able to demonstrate that the resolution-specific liquidity risks and factors outlined in this chapter are also adequately addressed.

Resolution Regime – Code of Practice LFIR-1 Resolution Planning – Liquidity and Funding in Resolution 22.07.2022 9 (b) the liquidity required to secure the continued provision of essential services in order to support operational continuity in resolution13; (c) the liquidity effects of expected counterparty behaviour and requirements, such as increased initial or variation margin requirements; or the termination of contracts and the exercise of netting/set-off rights, giving due consideration to any temporary suspension of termination rights which may be imposed by resolution authorities, including the MA14 or a non-Hong Kong resolution authority, as the case may be; (d) any legal, regulatory or operational impediments to the transferability of liquidity resources amongst group entities (e.g. restrictions on intra-group liquidity flows); and (e) the impact of management mitigation actions, such as the implementation of a contingency funding plan or recovery plan15 . 3.4 An AI’s methodology is expected to estimate the liquidity and funding that would be needed under resolution scenarios involving: (i) different durations such as fast-moving scenarios leading to an immediate entry into resolution and slow-moving scenarios wherein entry into resolution occurs after an extended period of stress; and (ii) different types and severities (e.g. market-wide, idiosyncratic or a combination of stresses; and liquidity-driven and/or solvency-driven stresses). Additionally, an AI’s methodology is expected to account for cross-impacts between solvency and liquidity stresses, for example, where solvency deterioration is expected to impact the cost or availability of certain funding sources or where liquidity strictures require rapid disposal actions that exacerbate solvency deterioration. 3.5 An AI’s methodology is also expected to allow for the performance of sensitivity analyses, which should, among other things, enable the identification of key drivers of liquidity and funding needs in resolution (e.g. deposit outflows, drawdowns of committed credit facilities, FMI requirements, declines in value of available liquid assets, etc.) at the level of each material LFIR entity. Further, it is expected that an AI will

13 See Code of Practice chapter OCIR-1, “Resolution Planning - Operational Continuity in Resolution”. 14 Pursuant to section 90 of the FIRO in relation to qualifying contracts. 15 Details of the supervisory expectations on contingency funding plans and recovery planning may be found at section 12 of SPM module LM-2 and in SPM module RE-1 “Recovery Planning” (“RE-1”), respectively.

Resolution Regime – Code of Practice LFIR-1 Resolution Planning – Liquidity and Funding in Resolution 22.07.2022 10 set appropriate assumptions for the key drivers underlying its estimates of liquidity and funding needs in resolution, including during the stabilization phase and in the post-stabilization restructuring phase taking into account the preferred resolution strategy as determined by the MA - at a sufficiently granular level16 . 3.6 An AI is expected to document its justifications for the key assumptions used (e.g. haircuts, rollover rates, runoff rates, etc.) to derive its estimates for liquidity and funding needs in resolution. 3.7 Furthermore, an AI should ensure that its methodology for estimating liquidity and funding needs in resolution is subject to regular internal review and appropriate governance arrangements. See section 6 for further details on the MA’s expectations on governance for LFIR purposes.

16 When considering what constitutes an appropriate level of granularity for meeting this expectation, an AI may make reference to the list of individual cash flow items reflected in section 5.8 of SPM module LM-1 “Regulatory Framework for Supervision of Liquidity Risk” (“LM-1”). Although this aspect of SPM module LM￾1 applies only to category 1 institutions, any AI may, where appropriate, refer to this approach with such modifications as are proportionate to its own liquidity risk profile and liquidity risk management systems.

Resolution Regime – Code of Practice LFIR-1 Resolution Planning – Liquidity and Funding in Resolution 22.07.2022 11 4. Capabilities for measuring, monitoring and reporting on liquidity and funding in resolution 4.1 An AI is expected to have capabilities17 to measure, monitor and report on liquidity and funding needs and available liquid assets in resolution18 . These capabilities should apply the methodology developed for estimating liquidity and funding needs in resolution referred to in section 2 of this chapter. Specifically, an AI is expected to be able to, among other things: (a) project cash inflows and outflows over various time horizons with appropriately granular time bands, including at least daily time bands19 in the lead up to resolution and for at least 90 days from the point of entry into resolution, which may be followed by wider time bands (e.g. weekly or monthly); (b) generate projections of liquidity and funding needs in resolution on a T+1 basis20; (c) project how key balance sheet items as well as capital and liquidity metrics21may change over the projection period; (d) rapidly adjust the assumptions22 for the key drivers of liquidity and funding needs and the haircuts on available liquid assets in its projections for different periods along the timeline in the lead up to and during resolution; and

17 Including systems, processes and personnel. 18 An AI may seek to satisfy the expectations in this chapter by leveraging the liquidity measurement, monitoring and reporting capabilities and arrangements which it already has in place for meeting the going￾concern supervisory expectations placed on it (see section 3, 4 and 10 of SPM module LM-2 and also the Banking (Liquidity) Rules (Cap. 155Q) which call for an AI to have in place certain liquidity monitoring and reporting capabilities for going-concern supervisory purposes). In such circumstances, for the purposes of assessing resolvability, the MA expects the AI to be able to demonstrate that the resolution-specific liquidity risks and factors outlined in this chapter are also adequately addressed. 19 An AI’s capabilities should also be able to provide visibility into intraday cash inflows and outflows and related mismatches. 20 The “T+1 basis” refers to the timeframe in which an AI’s liquidity projections/reports can be refreshed to reflect the most current data on the AI’s liquidity position. An AI is expected to be able to refresh its liquidity projections/reports to reflect any changes in its liquidity position stemming from the previous day's activities. 21 Such as capital and liquidity metrics used for internal risk monitoring as well as regulatory reporting purposes. 22 Including various runoff, rollover and drawdown rates relating to the key drivers of liquidity and funding needs.

Resolution Regime – Code of Practice LFIR-1 Resolution Planning – Liquidity and Funding in Resolution 22.07.2022 12 (e) apply the capabilities at (d) above, to rapidly reflect the impact of actual crisis conditions into estimates of liquidity and funding needs in resolution. 4.2 An AI is expected to be able to assess potential future liquidity and funding shortfalls at the level of each material currency, each material LFIR entity and at the level of the AI’s group, taking into account the projections of liquidity and funding needs and available liquid assets in resolution. This should, in turn, inform effective strategies, which the AI is expected to develop, to address these potential shortfalls (including foreign currency mismatches). 4.3 An AI is expected to determine and be able to explain and justify what constitutes an appropriate frequency and format for internal reporting of liquidity metrics, including the estimates of liquidity and funding needs in resolution as well as the available liquid assets. Additionally, an AI is expected to be able to submit liquidity reports to the MA at an increased frequency, including on at least a daily basis, in the lead up to resolution and during the stabilization phase. 4.4 As part of the bilateral resolution planning programme, an AI is expected to be able to demonstrate to the MA its ability to produce liquidity reports in an appropriate format and timely manner to inform its internal stakeholders and/or the MA in the lead up to and during resolution.

Resolution Regime – Code of Practice LFIR-1 Resolution Planning – Liquidity and Funding in Resolution 22.07.2022 13 5. Capabilities for identifying and mobilising collateral and for monitoring aggregate asset encumbrance levels 5.1 An AI is expected to have in place, and be able to demonstrate, the capability to assess the potential need to rely on third-party funding providers in resolution. An AI is also expected to have in place, in BAU, capabilities to identify and mobilise (including to value, manage, monitor and deploy) the types of assets that the AI has assessed to be eligible, or likely to be eligible, as collateral for funding from potential third-party funding providers in resolution23 . The AI is expected to be able to demonstrate to the MA the effectiveness of such capabilities. 5.2 In support of its capabilities to mobilise assets for use as collateral24 , an AI is expected to be able to keep up-to-date, rapidly generate, and provide to the different potential third-party funding providers (including the HKMA and other central banks), the information which may be required to facilitate the funding providers’ risk assessments and funding decisions. In this regard, in BAU, an AI is expected to have in place capabilities to anticipate and monitor certain baseline information which potential third-party funding providers may require, including (as applicable) the following information with respect to assets that the AI has assessed to be eligible, or likely to be eligible, as collateral for third￾party funding in resolution: (a) asset type, value and currency; (b) maturity date associated with the asset as well as any call or conversion features; (c) asset quality/rating; (d) legal rights to the asset, including any rights of re-hypothecation, or whether the asset is pledged to or by a counterparty; (e) obligor-level information associated with the asset (e.g. in the context of a loan, the borrower location, the extent of any continuing/outstanding obligations under the loan agreement,

23 An AI may seek to satisfy the expectations in this chapter by leveraging the collateral management capabilities and arrangements which it already has in place for meeting the going-concern supervisory expectations placed on it (see section 11 of SPM module LM-2). In such circumstances, for the purposes of assessing resolvability, the MA expect the AI to be able to demonstrate that the resolution-specific liquidity risks and factors outlined in this chapter are also adequately addressed. 24 Including assets that do not qualify as High Quality Liquid Assets or as liquefiable assets, etc.

Resolution Regime – Code of Practice LFIR-1 Resolution Planning – Liquidity and Funding in Resolution 22.07.2022 14 the delinquency history of the obligor, and information on the collateral in the case of a secured loan); (f) location of the asset (e.g. the entity holding the asset, physical location, custodian location); and (g) governing law of the asset. 5.3 An AI is expected to identify and address any legal, regulatory or operational obstacles that could affect the mobilisation of particular types of assets, including less-liquid assets (e.g. loans), that have been assessed to be eligible, or likely to be eligible, as collateral for third￾party funding in resolution. In preparation for mobilising assets as collateral for third-party funding in resolution, the AI is expected to obtain assurance on, and be able to assess and evaluate, the transferability of these assets, and that its intended collateral arrangements are valid and enforceable, including on a cross-border and intra-group basis where relevant. 5.4 An AI is expected to be able to estimate the time required to mobilise and monetise different types of assets as collateral including assets which are located outside Hong Kong and in different time zones. The AI is expected to document, such as through the development of a playbook, the steps required and expected timeframes for mobilising and monetising these assets. 5.5 To support readiness for mobilising these assets, an AI is expected to maintain, in BAU, up-to-date asset valuations. Furthermore, an AI’s capabilities are expected to enable timely updates to asset valuations, taking appropriate account of changing conditions in different resolution scenarios. 5.6 For the purpose of informing, inter alia, any post-stabilization restructuring action to be taken, an AI is expected to have capabilities to measure, monitor and report on the aggregate level of asset encumbrance and to project changes in the amount of unencumbered assets (including assets dedicated to satisfying intraday liquidity needs) available at different stages in resolution.

Resolution Regime – Code of Practice LFIR-1 Resolution Planning – Liquidity and Funding in Resolution 22.07.2022 15 6. Governance 6.1 An AI is expected to have in place appropriate governance arrangements, including oversight, decision-making and organisational arrangements, to support the ability to meet liquidity and funding needs in resolution. The governance arrangements and processes should be embedded into the AI’s overarching governance and organisational arrangements in place to support the HKMA’s resolution planning process as explained in paragraph 8.1 of RA-2. 6.2 As part of this, an AI is expected to maintain effective governance arrangements over its activities for building up its LFIR capabilities and arrangements, and for ensuring that the expectations outlined in the chapter are met on an ongoing basis. For this purpose, an AI is expected to assign an officer of appropriate seniority with accountability for overseeing the work to ensure the effectiveness of the AI’s LFIR capabilities and arrangements. Furthermore, the board of directors of the AI should be kept adequately informed of the LFIR capabilities and arrangements which the AI puts in place to meet the expectations outlined in this chapter25 . 6.3 Consistent with the expectation that an AI should have the LFIR capabilities and arrangements in place in BAU, an AI is expected to fully embed its LFIR capabilities and arrangements into its existing liquidity risk management framework, including its processes for risk monitoring, reporting, review and tolerance setting26 . 6.4 The relevant governance arrangements and processes are expected to support effective decision-making and timely action by senior management, the board of directors or relevant authorities in the lead up to and during resolution, including at the post-stabilization restructuring phase. These governance arrangements and processes are expected to include, but are not limited to, early warning triggers, escalation procedures and clear lines of responsibility.

25 This would include any procedures, playbooks or other documentation developed consistent with this chapter and the role of the board of directors as set out in such procedures, playbooks, or other documentation. 26 For details of the supervisory expectations on the going-concern liquidity risk governance and the governance framework to support an actionable recovery plan, see section 2 of SPM module LM-2 and section 2 of SPM module RE-1, respectively.

Resolution Regime – Code of Practice LFIR-1 Resolution Planning – Liquidity and Funding in Resolution 22.07.2022 16 7. Testing and validation 7.1 An AI should ensure and demonstrate that its LFIR capabilities and arrangements are fit for purpose through regular testing and validation. The scope for testing and validation should cover the AI’s methodology, capabilities and governance for LFIR, including the decision-making processes, related playbooks or other documentation. 7.2 As part of its testing and validation, an AI is expected to project, measure, monitor and report on liquidity and funding needs, and assess its ability to mobilise collateral for third-party funding in an assumed resolution scenario reflecting a sufficiently severe stress, taking into account the preferred resolution strategy determined by the MA. In addition to an AI’s internally developed resolution scenarios, the MA may set one or more resolution scenarios to be included in an AI’s testing. 7.3 Testing and validation activities may take different forms, such as internal testing or dry-runs, internal audit review, or validation by an independent third-party. 7.4 An AI is expected to justify to the MA its approach to and plans for testing and validating its LFIR capabilities and arrangements, and discuss with the MA the outcomes of the activities, including any lessons learned and resulting work to further enhance the resolvability of the AI. An AI is expected to reflect relevant lessons learned into updates to its documented decision-making processes such as playbooks and procedures, as the case may be. 7.5 The appropriateness of the testing and validation activities, including their nature and frequency, should be kept under review by the AI.

Resolution Regime – Code of Practice LFIR-1 Resolution Planning – Liquidity and Funding in Resolution 22.07.2022 17 8. The MA’s approach to implementation 8.1 In line with the MA’s proportionate and risk-based approach to resolution planning, all Domestic Systemically Important Authorized Institutions and other locally incorporated AIs with total consolidated assets of more than HKD 150 billion will be prioritised by the MA for resolution planning. Work towards satisfying the expectations outlined in this chapter will be an integral part of the MA’s bilateral resolution planning programmes with these AIs. The MA would typically expect to engage an AI in the resolution planning process27 (if not already engaged) before requesting the AI to demonstrate the adequacy of its LFIR capabilities and arrangements. 8.2 As part of a resolvability assessment, the MA would expect an AI to be able to demonstrate the relevant LFIR capabilities and arrangements required to meet the expectations outlined in this chapter. The development and putting in place of adequate LFIR capabilities and arrangements by an AI will be an iterative process between the MA and the AI in the course of bilateral resolution planning. As mentioned in section 2 above, as a starting point, an AI is expected to explain and justify to the MA, its proposed scope of material LFIR entities and material currencies. An AI is also expected to self-assess its existing capabilities and arrangements against the expectations set out in this chapter on an ongoing basis. An AI is expected to submit to the MA, a work plan which is informed by, inter alia, the outcomes of its self￾assessment. The AI’s work plan should include: (a) its proposed timeline for, and approach towards, meeting the expectations outlined in this chapter; (b) key milestones towards satisfying the expectations outlined in this chapter; (c) details of its governance arrangements for meeting the expectations outlined in this chapter; and (d) details of its testing and validation approach. 8.3 The MA will assess an AI’s work plan and the effectiveness of the LFIR capabilities and arrangements as well as any strategies to address potential liquidity and funding shortfalls (including foreign currency

27 See Part 8 of RA-2.

Resolution Regime – Code of Practice LFIR-1 Resolution Planning – Liquidity and Funding in Resolution 22.07.2022 18 mismatches) in the context of the AI’s preferred resolution strategy as determined by the MA. The MA may require submission of information, records or documents in relation to these capabilities and arrangements28 . 8.4 Where an AI is part of a cross-border group, the MA works closely with non-Hong Kong resolution authorities on resolution planning through Crisis Management Groups and other cross-border resolution planning fora. The MA may take account of information from its interactions with a non-Hong Kong resolution authority when assessing the effectiveness of an AI’s LFIR capabilities and arrangements and in considering whether a significant impediment exists to the orderly resolution of the AI, in accordance with a non-Hong Kong resolution plan (to the extent that the non-Hong Kong resolution plan has been adopted by the MA). Nevertheless, the expectations in this chapter apply to the AI regardless of whether the group is subject to, and meets, similar LFIR standards in other jurisdictions. 8.5 It is acknowledged that due to the interrelationships between some of the expectations in this chapter and some existing going-concern regulatory requirements/supervisory expectations, an AI may already have in place capabilities and arrangements which could be suitably leveraged for the purposes of meeting expectations outlined in this chapter. In such instances, the MA expects an AI to demonstrate that the relevant existing capabilities or arrangements adequately address the resolution-specific liquidity risks and factors outlined in this chapter.

28 The MA may impose such a requirement pursuant to section 158 of the FIRO.