Management Summary
- Purpose / Background: The HKMA is launching "Granular Data Reporting (GDR) 3.0" under the Fintech Promotion Blueprint to transition from legacy, aggregated, template-based reporting to a modern, data-point-centric model supported by the "DataGRID" framework.
- One-line conclusion: AIs must shift from manual template submissions to automated, granular data pipelines; preparation requires immediate data inventory stock-taking and participation in HKMA co-design workshops.
- Key Changes:
- Shift from aggregated templates to direct granular data-point submissions.
- Implementation of the "Report Once, Use Multiple" principle to eliminate duplicative reporting.
- Adoption of an AI-ready data dictionary to facilitate straight-through processing.
- Streamlining/discontinuation of over 10 existing surveys as a first step.
- Future-proofing for "on-demand" data reporting capabilities.
- Key Dates / Deadlines:
- Q2–Q3 2026: Industry engagement and co-design activity period.
- 2026 (Full Year): Finalization of the GDR 3.0 implementation plan.
- Applicability / Impact scope: All Authorized Institutions (AIs) in Hong Kong.
- Recommended management actions:
- Conduct an immediate internal stock-take of existing risk data inventories.
- Evaluate current data infrastructure readiness for migration to granular reporting.
- Participate in the HKMA’s "co-design" engagement series starting Q2 2026.
- Review provisional implementation parameters (Annex) and provide formal feedback to the HKMA.
- Appoint a dedicated project lead to align internal reporting architecture with DataGRID standards.
Detailed Summary
1) Document overview
The circular introduces GDR 3.0, a fundamental shift in the HKMA’s supervisory data collection strategy. It aims to enhance supervisory agility by replacing fragmented, template-based returns with a unified, data-point-centric approach under the "DataGRID" framework.
2) Main requirements
- Migration to Granular Data: AIs must transition from providing aggregated template data to submitting raw, granular data directly at the source.
- Infrastructure Readiness: Institutions are required to assess their internal data management systems to ensure they can support AI-ready, straight-through processing.
- Feedback & Engagement: AIs are expected to actively provide feedback on the provisional implementation parameters provided by the HKMA.
3) Key changes
- Efficiency: Adoption of "Report Once, Use Multiple" to minimize redundant supervisory requests.
- Automation: Phasing out manual template-based surveys in favor of AI-ready data dictionaries.
- Streamlining: Over 10 legacy surveys have already been identified for removal or overhaul.
4) Important dates & transition
- Q2–Q3 2026: Industry co-design phase and consultation.
- End of 2026: Target date to finalize the GDR 3.0 implementation roadmap.
5) Impact and risks
- Operational: Significant change management required as manual reporting workflows are replaced by automated data pipelines.
- Technical: Potential need for IT system upgrades to meet new granular data standards and submission protocols.
- Strategic: Increased visibility into risk data, requiring higher data quality and governance standards within AIs.
6) Compliance action checklist
- [ ] Perform a comprehensive risk data inventory stock-take.
- [ ] Assess the capability of existing systems to support granular, point-centric data extractions.
- [ ] Review the Annex provided with the circular regarding provisional parameters.
- [ ] Designate a representative contact for the HKMA engagement activities scheduled for Q2/Q3 2026.
- [ ] Submit institutional feedback to
gdr@hkma.gov.hk.
7) Appendices/attachments summary
- Annex (Provisional Implementation Parameters): Details the preliminary technical standards and reporting frameworks proposed for GDR 3.0. AIs are required to review this document to provide feedback on feasibility and transition impacts during the consultation period.