Climate scenarios: Where's the stress?

Event recap — Somerset House, London | 23 June 2026

On 23 June, Planetrics hosted a panel discussion at Somerset House as part of London Climate Action Week, bringing together a select group of banks, asset owners, asset managers and insurers to explore the future of climate scenario analysis. The session was led by Thomas Bremner Bligaard (Managing Director, Planetrics) and Ethan McCormac (Head of Advisory, Planetrics) and featured a panel of Roelof Coertze (Group Director for Environment, Prudential), Anna Moss (Sustainable Investment Manager, Aberdeen) and Billy Suid (Head of Climate and Sustainable Finance Risk, Barclays).

Why climate scenario analysis matters now

Demand for sophisticated climate scenario analysis is being driven by three converging forces: tightening regulatory expectations (including the Bank of England PRA's Supervisory Statement 5/25, the ISSB's IFRS S2 and the European Banking Authority’s guidance on environmental scenario analysis); rising stewardship and engagement obligations as clients and investees become more sophisticated; and the need to maintain a competitive edge in a rapidly changing operating environment.

Alongside these pressures, climate risk continues to materialise for corporations. European airlines, for example, are now facing significant and direct cost increases equivalent to 20-40%+ of earnings from the EU ETS and Sustainable Aviation Fuel mandates. Physical risks are making themselves felt too: Hyatt, the hotel chain, made headlines recently when it cut its full-year earnings guidance in December 2025, after Hurricane Melissa forced an extended suspension of its Jamaican resort operations, as an illustration of the sector’s growing climate risk exposure.

From today’s already-stressed environment, multiple futures could unfold. Most models see the 1.5°C global mean temperature threshold breached around 2030–2035, with outcomes diverging more sharply after 2035–2040 depending on the pace of global emissions reduction. Critically, incremental warming has a non-linear effect on physical climate risk, meaning even modest differences between scenarios can translate into materially larger financial impacts.

Against this backdrop, climate stress-testing should be uncovering significant vulnerabilities.

So, where's the stress?

Yet many regulatory stress-testing exercises, including the Bank of England's 2021 Climate Biennial Exploratory Scenario, the ECB's 2022 Climate Risk Stress Test and the Federal Reserve's 2024 pilot exercise, have struggled to yield material financial impacts. Our view is that this understates the true picture, for three reasons:

• Inappropriate scenarios: many climate scenarios don't focus on the most stressful, plausible events, and are hard to calibrate against history given how modest past extreme-weather losses have been. Event-based modelling is needed to capture tail-risk impacts that can be masked by average damage estimates: for example, UK flood damage that averages around 0.5% of asset value at the 99th spatial percentile can exceed 8% for a 1-in-100-year event at that same percentile.
• Missing impact channels: modelling often overlooks channels such as supply chain impacts, insurance feedback loops and macroeconomic dynamics. For instance, Planetrics analysis shows that the financial risk for a large technology company from climate's supply-chain climate impact can be more than 3x bigger than the direct impact of damages to its own assets and associated business interruption.
• Insufficiently granular data: physical risk is typically concentrated in specific locations, and regional averages can mask that concentration. For instance, Australian tropical cyclone damages are 4x larger in the 99th percentile of exposed locations than 90th percentile; so, knowing exactly where the risk is makes a significant difference in financial terms.

More realistic stress analysis would address these three limitations. For example, modelling a severe flood event with an approach that includes wider macroeconomic impacts, acknowledges that insurance coverage is only partial, and takes supply chain disruptions into account can lead to more specific and severe view of the resulting financial shock.

Panel discussion: how institutions are putting scenario analysis to work

The panel discussion illustrated how financial institutions apply scenario analysis differently, depending on the type of institution and the use case. Banks are exploring narrative-led, severe-but-plausible scenarios, and increasingly using reverse stress testing (asking what could ‘break the bank’) to understand where climate sits alongside other vulnerabilities. For asset owners and managers, scenario analysis is primarily an identification tool that feeds into portfolio construction, sector weightings and the timing of engagement. Insurers, in turn, need to understand the impact of climate risk on their business in an integrated way, including changes in client demand and expected losses.

Standard NGFS scenarios, while a useful starting point, were seen as too uniform for these applications. One asset manager had addressed this by building bespoke scenarios that draw on elements of existing scenarios and used two different integrated assessment models to capture more sectoral variation.

Panellists agreed that both top-down and bottom-up analysis need further development. Top-down analysis helps understand the overall shape of exposure across a bank or investor portfolio and test severe-but-plausible downside scenarios, recognising that the real risk often lies in multiple shocks occurring together (climate plus another factor). Bottom-up analysis is needed for understanding impacts on specific sectors and understand risk to concentrated assets, such as policy reversals in the energy sector.

Across the room, there was broad agreement that understanding what climate risk and opportunity is already priced into markets remains genuinely difficult, and that recent experience has underestimated the pace of technological change while overestimating the ambition of policy; a reminder that narrative judgement, not just modelling, should anchor scenario design.

There is also emerging interest in extending this thinking to the intersection of climate and nature-related risk, particularly in sectors such as mining and power, where EBITDA impacts from nature risks are already being observed.

Closing thought

A recurring theme was that the real value of scenario analysis lies as much in the journey as in the outputs; from scenario selection through transmission channels to financial impact. Communicating the insights from that journey clearly to stakeholders remains a shared challenge across institutions. Collaboration across the industry on how to develop and make use of best practices will continue to be a priority for building a more robust, decision-useful next generation of climate scenarios.