Data-Driven Supervision and the Reduction of Supervisory Burden A Conceptual and Operational Framework for National Competent Authorities

Financial supervisors are doing more with less. Over the past fifteen years, mandates have multiplied — covering prudential soundness, conduct, market integrity, operational resilience, and systemic risk — while supervised populations have grown and budgets have not kept pace. The resulting capacity gap is structural, not cyclical. This paper proposes data-driven supervision (DDS) as a scalable response. DDS converts regulatory reporting data into reproducible risk indicators, composite scores, and automated triage pathways, concentrating expert judgment where it adds most value. Because marginal costs per entity are low and infrastructure is reusable across supervisory missions, DDS generates economies of scale that traditional inspection-based approaches cannot. The paper develops formal frameworks for indicator design and triage, documents how regulatory reporting enables population-scale risk detection, and shows how a shared data infrastructure can serve multiple supervisory objectives simultaneously. Governance challenges — model risk, false positives, preservation of human judgment — are addressed directly, and a phased implementation roadmap for National Competent Authorities is presented. The analysis is candid about limits: DDS depends on data quality, complements rather than replaces traditional supervision, and requires active management of implementation risk.