Skip to content
bazzi.ai

article

An emerging risk is managed from the day someone watches its number

By Maximilian Bazzi · Published 16 October 2026

  • risk appetite
  • operational risk
  • risk culture

An emerging risk becomes something an organisation manages on the day it has an indicator with a limit and a named owner. The quickest way to find that indicator is to start from the loss the organisation will not accept and work backwards, as European banking supervisors did with geopolitical risk this year.

Why the usual direction stalls

Emerging risks now come up on almost every executive agenda I see. Many of these discussions start from a headline, such as a trade dispute or a conflict, and guess at what it might do to the business. The result is often a register entry rated high, and a debate about whether it should be medium.

An entry of that kind cannot be breached, so it can never be escalated. Nobody can say when it worsened, and nobody is asked to notice.

What the supervisors asked for

In its 2026 reverse stress test, ECB Banking Supervision gave 110 directly supervised banks a fixed outcome: a fall of 300 basis points in their CET1 capital ratio. Each bank had to design the geopolitical scenario that would produce that fall (European Central Bank, 2026). The ECB found room for improvement in the granularity of risk assessments, in how consistently narratives translated into impacts, and in the realism of mitigating actions.

The European Banking Authority describes a reverse stress test as one that starts from a pre-defined outcome and explores the scenarios that might cause it, with the institution deciding the timing of its management actions through triggering events (European Banking Authority, 2018). A triggering event is an indicator with a limit. The method was built for capital, but nothing in its logic is specific to capital.

Three questions in reverse

First, which loss will the organisation not accept, stated in a measure the board already uses: capital, earnings, liquidity, or a critical service unavailable beyond its tolerance. Second, which plausible story would produce that loss. Third, which number would move first if that story began.

The third question carries most of the value: the number has to move before the loss, not with it. In its current supervisory priorities, the ECB observed that some early warning systems in lending to smaller firms were often solely dependent on ratings (European Central Bank, 2025). A rating tends to move once a borrower is already in difficulty, so it confirms a story rather than warning of it.

The number then needs a limit that triggers a conversation and a person expected to start it. For Swiss banks this is already the expectation for operational risk: FINMA expects, proportionately, risk and control indicators that identify relevant increases in risk in a timely manner, and a risk tolerance monitored through such indicators (FINMA, 2022).

The strongest objection

An emerging risk, a sceptic would say, is by definition one that nobody can yet narrate. Choosing a story usually means choosing the wrong one, and an indicator attached to the wrong story gives false comfort. And every headline adds an indicator, until the dashboard is as idle as the topic list.

Part of this should be conceded. Some losses arrive with no warning at all, and a wrong number watched carefully can be more dangerous than a topic everyone knows is still open.

The answer

Working backwards does not depend on picking the right story. It depends on finding the few places where many stories meet on their way to the same loss. A conflict, a tariff and a cyber attack on a supplier are different stories, yet they often reach capital through the same exposures. An indicator placed close to the loss moves whichever story is unfolding. The European Banking Authority warns that the pre-defined outcome can be produced by circumstances other than the one analysed (European Banking Authority, 2018), which argues for indicators near the loss rather than near the headline.

Where no number would move early enough, that is itself a finding, and the answer is a pre-agreed action and a tested recovery capability rather than another indicator. On the dashboard, the Financial Stability Board notes that the number of limits should balance comprehensiveness against monitoring cost, and that risks which resist quantification can still be limited through qualitative assessment (Financial Stability Board, 2013). One indicator for each unacceptable loss is enough to begin.

A test for your own organisation

Take the emerging risk at the top of your latest executive agenda and ask three things: which number would move first if it began, which value of that number would trigger a conversation, and whose name sits next to it. If any answer is missing, the risk is still a topic.

For when a number helps and when judgement does better, see qualitative or quantitative risk assessment.

References

Cookie preferences

Essential cookies keep the site working and cannot be switched off. Analytics is optional.

Always on. Required for the site to function.

Cookieless usage analytics (Vercel Analytics). No cross-site tracking.