The Quest for Supervisory Effectiveness

In a speech delivered last week, Pablo Hernández de Cos, General Manager of the Bank for International Settlements (BIS), argued that supervisory effectiveness depends less on implementing new rules than on addressing the frictions that prevent authorities from acting early.

Hernández de Cos opened with a striking observation from the March 2023 banking turmoil: banks failed despite meeting regulatory requirements. The culprits were not inadequate rules, but “poor governance, weak risk management and unsustainable business models,” qualitative weaknesses that supervisors identified too late, or too timidly, and which eventually manifested in liquidity runs and market corrections. A purely rules-based approach, he argued, is unlikely to identify and mitigate such risks in time. Rather, qualitative tools, applied early, can address root causes before financial risks materialize, Hernández de Cos noted.

He traced the problem across the full supervisory cycle. Risk scoping is often too narrow, he argued. As such, if key risk areas are not prioritized at the outset, supervisory findings in those areas will also be limited. Risk assessments are distorted by strong short-term financial performance, he added, obscuring weak governance and unsustainable business models. And when weaknesses are found, supervisors hesitate to escalate, leaning instead on informal tools that "lack the urgency and clarity needed to drive meaningful change.” This allows banks to treat concerns as recommendations rather than matters requiring corrective action, he said.

Hernández de Cos was direct about the role of institutional culture and risk appetite in enabling, or inhibiting, early action. Every supervisory authority must accept a certain level of supervisory risk, he argued: the risk that, despite its best efforts, it fails to identify or mitigate prudential risks in time. Attempts to eliminate that risk entirely — through a “zero bank failure” policy, for instance — are neither realistic nor desirable, he said, as they would undermine banks’ ability to take informed risks and support the real economy. The task, therefore, is to manage supervisory risk rather than eradicate it, he asserted.

The remedy lies not in more rules but in a clearer structure around judgment. Supervisory risk appetite frameworks, Hernández de Cos argued, can help authorities articulate their tolerance for supervisory risk, make trade-offs explicit, and institutionalize proportionality. By linking an authority’s risk tolerance to tangible guidance, such frameworks provide structure and consistency to supervisory decision-making while preserving the flexibility needed to exercise judgment effectively. “Judgment should in no way be confused with unconstrained discretion,” he cautioned. “The real challenge is to put in place an effective structure to properly frame judgment.”

He closed with a call for international coordination. “[D]eveloping further international guidance to improve the quality of supervision worldwide is particularly important,” he concluded. “Such efforts are also key to mitigate unwarranted international heterogeneity in supervisory practices.”