Hong Kong Climate Stress Testing: Quantifying Transition Risk and Physical Risk for Financial Firms

The Hong Kong Monetary Authority (HKMA) published its second pilot climate risk stress test (CRST) in December 2024, covering 24 major authorised institutions. The results showed that a severe but plausible climate scenario could reduce the aggregate net profit of participating banks by 15% to 20% over a 30-year horizon. This is not a distant hypothetical. The HKMA has confirmed that by 2025, all authorised institutions must integrate climate scenario analysis into their regular ICAAP (Internal Capital Adequacy Assessment Process) submissions. Financial firms operating in Hong Kong — whether banks, securities brokers, or asset managers — now face a regulatory expectation to quantify both transition risk and physical risk. The Securities and Futures Commission (SFC) has also signalled that climate stress testing will be incorporated into its supervisory reviews for licensed corporations under the Fund Manager Code of Conduct. This article explains the two core risk categories, the methodology the HKMA expects, and how firms can prepare for the 2025-2026 compliance cycle.

The Regulatory Framework for Climate Stress Testing in Hong Kong

The HKMA’s Two-Pillar Approach

The HKMA’s supervisory policy manual (SPM) module CA-G-3, issued in July 2023, sets out the expectation that authorised institutions conduct climate stress testing as part of their risk management framework. The HKMA divides climate risk into two distinct pillars: transition risk and physical risk. Transition risk arises from the shift to a low-carbon economy — changes in policy, technology, and market sentiment. Physical risk stems from the direct impact of climate change, such as extreme weather events and chronic temperature shifts.

The second pilot CRST in 2024 required banks to apply a 30-year forward-looking scenario based on the Network for Greening the Financial System (NGFS) scenarios. The HKMA used two specific NGFS pathways: an “orderly transition” scenario (Net Zero 2050) and a “disorderly transition” scenario (Delayed Transition). The disorderly scenario assumed a sudden and disruptive policy shift after 2030, leading to a sharp increase in carbon prices to USD 200 per tonne by 2040.

The SFC’s Evolving Expectations for Licensed Corporations

The SFC has not yet mandated a standalone climate stress test for all licensed corporations. However, the SFC’s circular on climate risk management for fund managers, issued in August 2021, requires SFC-authorised funds to consider climate-related risks in their investment and risk management processes. The SFC’s 2024-2025 business plan explicitly states that the regulator will conduct thematic inspections on climate risk governance.

For licensed corporations that manage collective investment schemes, the SFC expects firms to demonstrate how they assess the impact of climate scenarios on portfolio value. The SFC has referenced the Task Force on Climate-related Financial Disclosures (TCFD) framework and the International Sustainability Standards Board (ISSB) standards. Firms should expect the SFC to request climate stress test results during routine inspections, particularly for firms with material exposure to carbon-intensive sectors.

Quantifying Transition Risk: Methodology and Data Requirements

Scenario Design and Carbon Price Pathways

Transition risk quantification begins with scenario selection. The HKMA’s pilot CRST used a 30-year projection period, from 2025 to 2055. The key variable is the carbon price trajectory. Under the disorderly transition scenario, carbon prices jump from approximately USD 20 per tonne in 2025 to USD 200 per tonne in 2040. This sharp increase directly impacts the earnings of firms with high Scope 1, Scope 2, and Scope 3 emissions.

For a financial firm, the calculation proceeds in three steps. Step 1: Map the firm’s loan book or investment portfolio to sectors with high transition risk — energy, utilities, heavy industry, and transportation. Step 2: Apply sector-specific emission factors to estimate financed emissions. Step 3: Model the impact of rising carbon costs on the borrower’s or investee’s profitability, and then translate that into credit risk metrics such as probability of default (PD) and loss given default (LGD).

The HKMA’s 2024 pilot results showed that the disorderly scenario increased the aggregate PD of the loan portfolio by 2.3 percentage points by 2050. The energy sector faced the largest increase, with PDs rising by 5.1 percentage points.

Data Gaps and Proxy Approaches

Data availability remains the most significant operational challenge. Companies in Hong Kong and across Asia often do not disclose Scope 3 emissions. The HKMA acknowledged in its 2024 report that only 40% of the corporate borrowers in the pilot had publicly available emission data. For the remaining 60%, banks used proxy methods — typically sector-average emission intensities from sources such as the International Energy Agency (IEA) or the Carbon Disclosure Project (CDP).

Firms should adopt a tiered data strategy. For large exposures, require borrowers to provide emission data directly through loan covenants or ESG questionnaires. For smaller exposures, use sector-level proxies and document the assumptions clearly. The HKMA expects firms to have a plan to improve data coverage over time, with a target of 80% direct data coverage by 2028.

Quantifying Physical Risk: Geographic Exposure and Damage Functions

Acute and Chronic Risk Modelling

Physical risk quantification requires a different methodological approach. The HKMA’s pilot CRST used two physical risk scenarios: a “current policies” scenario (SSP2-4.5, representing a 2.7°C warming by 2100) and a “high emissions” scenario (SSP5-8.5, representing a 4.4°C warming by 2100). The focus was on three acute perils: flooding, tropical cyclones, and heatwaves.

The modelling process involves overlaying the firm’s portfolio exposure — by geographic location of collateral, operations, or counterparty — with hazard maps. For Hong Kong, the primary physical risk is flooding from extreme rainfall and storm surges. The Hong Kong Observatory has projected that by 2050, the mean sea level in Victoria Harbour will rise by 0.3 to 0.5 metres under the high emissions scenario. This directly affects property collateral values in low-lying areas such as Tuen Mun, Yuen Long, and parts of the Eastern District.

Damage Functions and Loss Estimation

Physical risk models use damage functions — statistical relationships between hazard intensity and economic loss. For example, a flood depth of 1 metre in a commercial building in Kwun Tong corresponds to a damage ratio of approximately 15% to 25% of the building’s replacement value. The HKMA’s pilot found that under the high emissions scenario, the aggregate credit losses from physical risk for the 24 participating banks reached HKD 8.2 billion by 2050, representing 0.4% of total gross loans.

Firms should prioritise geographic granularity. A Hong Kong-based firm with a loan book concentrated in the New Territories faces a different physical risk profile than a firm with exposure to Central and Wan Chai. The HKMA recommends that banks obtain property-level location data for all collateral. For firms without in-house climate modelling capability, third-party vendors such as RMS, AIR Worldwide, and Jupiter Intelligence provide hazard maps and damage functions for Hong Kong.

Integration into Risk Management and Regulatory Reporting

ICAAP and Pillar 2 Capital

The HKMA has made clear that climate stress testing is not a standalone exercise. It must be integrated into the ICAAP, which is the foundation for Pillar 2 capital requirements under the Banking (Capital) Rules (Cap. 155L). The HKMA expects authorised institutions to include climate risk in their risk appetite statements and to set materiality thresholds.

A practical approach is to start with a sensitivity analysis. Step 1: Identify the top 10 largest exposures by sector and geography. Step 2: Apply a uniform shock — for example, a 50% increase in carbon costs — and measure the impact on PD and LGD. Step 3: Compare the result against the firm’s capital buffer. If the impact exceeds 10% of Tier 1 capital, the firm should proceed to a full scenario analysis. The HKMA’s 2024 pilot showed that the aggregate impact on Tier 1 capital under the disorderly scenario was 3.2% for the median bank, but 8.7% for the most exposed bank.

Disclosure Obligations Under the HKEX Listing Rules

Listed firms in Hong Kong face additional requirements. The Hong Kong Exchanges and Clearing Limited (HKEX) amended its Listing Rules in April 2024 to require ESG report disclosure aligned with the ISSB standards. Appendix 27 of the Main Board Listing Rules now mandates disclosure of climate-related risks and opportunities, including the results of scenario analysis. Firms must disclose the scenarios used, the time horizons, and the key assumptions.

For financial firms that are also listed issuers, the climate stress test results will form part of the annual ESG report. The HKEX has stated that it will conduct a review of ESG report compliance in 2025. Firms that fail to disclose material climate risks may face enquiries from the Listing Division.

Practical Implementation Steps for 2025-2026

Build a Cross-Functional Climate Risk Team

Climate stress testing requires input from risk management, finance, credit, and sustainability teams. The HKMA expects a clear governance structure with board-level oversight. The board should approve the climate risk appetite and review the stress test results. The HKMA’s 2024 pilot found that 70% of participating banks had established a dedicated climate risk committee.

Select and Validate Scenarios

Do not attempt to build scenarios from scratch. Use the NGFS scenarios, which are freely available and accepted by the HKMA. The three recommended scenarios are: Net Zero 2050 (orderly), Delayed Transition (disorderly), and Current Policies (hot house world). Validate the scenarios against Hong Kong-specific data, such as the Hong Kong Observatory’s climate projections and the government’s Climate Action Plan 2050.

Start with a Materiality Assessment

A full climate stress test covering all exposures is resource-intensive. Start with a materiality assessment to identify the portfolios most exposed to transition and physical risk. The HKMA suggests focusing on sectors that constitute more than 5% of the total loan book and on geographic areas with high physical risk. This scoping exercise reduces the initial data collection burden by 60% to 70%.

Closing: Five Actionable Takeaways

1. By Q2 2025, all authorised institutions must submit climate stress test results as part of their ICAAP; start data collection now, particularly for Scope 3 emissions and property-level collateral locations.
2. Use the NGFS Net Zero 2050 and Delayed Transition scenarios as your baseline; the HKMA has confirmed these are the reference scenarios for the 2025 cycle.
3. For physical risk, obtain flood hazard maps from the Hong Kong Observatory and overlay them with your portfolio’s geographic data; this is the most cost-effective first step.
4. Licensed corporations that manage SFC-authorised funds should prepare to provide climate scenario analysis results during the SFC’s 2025 thematic inspections.
5. Listed firms must integrate climate stress test results into the ESG report under the HKEX’s ISSB-aligned Listing Rules, effective for financial years beginning on or after 1 January 2025.

This does not constitute legal advice. Consult a solicitor or regulatory consultant for your specific compliance obligations.