Climate scenario models in financial services significantly underestimate climate risk
The financial services sector must act urgently to effectively consider and embed the impact of climate change into its risk management.
A new report from the Institute and Faculty of Actuaries (IFoA) created in partnership with the University of Exeter has found that economic models underpinning climate scenario modelling in financial services do not always reflect the threat climate change poses to our planet and society.
‘The Emperor’s New Climate Scenarios – a warning for financial services’ points to a disconnect between climate scientists, economists, those building models and model users in financial services.
Some current scenarios could have limited use as they do not adequately communicate the level of risk we are likely to face if we fail to decarbonise quickly enough.
Techniques being used now exclude many of the most severe impacts of climate change such as sea-level rise, heat stress or climate tipping points, where a change in the climate system becomes self-perpetuating such as the loss of Arctic sea ice or the Greenland ice sheet.
They also exclude second order impacts for human society such as civil unrest and involuntary mass migration which could cause significant economic impacts.
The report lays out how the results emerging from current models are far too benign and, in some cases, implausible, using actuarial techniques to examine the limitations and assumptions of models used.
This severely limits the usefulness of the models to business leaders and policy makers, who may reasonably believe these models effectively capture risk levels, unaware that many of the most severe climate impacts have not been considered.
The report also highlights the uncertainty in carbon budgets, where there is a wide error margin, meaning there is a risk that ‘net zero’ carbon budgets may already be exhausted.
The report proposes a way forward to make a more realistic assessment of climate risk, which would show significant economic damage above 2°C of warming.
Professor Tim Lenton, from the University of Exeter, said: “Some economists have predicted relatively low economic damage – even from extreme levels of climate change. It is concerning to see these same economic models being used to underpin climate-change scenario analysis in financial services. It is essential that financial services institutions and regulators move towards realistic climate scenarios that recognise the potentially catastrophic risks posed by climate change.
“We have left it too late to tackle climate change incrementally. It now requires transformational change and a dramatic acceleration of progress. We have identified a variety of positive tipping points in human societies that can propel rapid decarbonisation. We need the support of the capital and insurance markets to achieve this, and actuaries have an important contribution to make. In addition to their role in the insurance markets, their work in pensions means they can impact capital allocation in long-term savings in a way few other professions can.”
Sandy Trust, Lead author and Past-Chair, IFoA Sustainability Board, said: “In the context of climate change, it is as if we are modelling the scenario of the Titanic hitting an iceberg but excluding from the impacts the possibility that the ship could sink. It is critical we develop realistic downside scenarios that reflect the level of risk we face – as this will inform the level of effort we put into decarbonising to mitigate that risk. It is crucial that model users understand the limitations and assumptions of models, take action to break down silos and develop techniques to understand how different combinations of risks will impact future solvency and what actions can help to mitigate this. It is deeply concerning that what we see happening in the real world is largely excluded from models widely used across the financial sector.”
As well as providing detailed analysis of these challenges around climate scenario modelling, the report recommends ways to move forward:
- Education on the assumptions underpinning the models and their limitations
- Development of realistic qualitative and quantitative climate scenarios
- Model development required to better capture risk drivers, uncertainties and impacts
Sandy Trust continued: “A fact still poorly understood in financial services is that there is considerable uncertainty in Earth system modelling, which has profound implications. Carbon budgets have high error margins and could now be negative for a temperature goal of 1.5°C. All of which reinforces the need to urgently reduce emissions by accelerating socio-economic tipping points, remove greenhouse gases from the atmosphere and repair broken parts of the climate system.”
Nigel Topping, UK Climate Action Champion for COP26 and Honorary Professor at the University of Exeter Business School, said: “Everyone who cares about the stability of our financial system should read this paper. Failing to include known non-linear effects in strategic thinking about climate change will lead to complacency, heightened risk and missed opportunities. So the scenarios that are used as part of TCFD processes really matter – both because economic damage will grow much faster and because the transition to clean technologies will happen much faster than conventional economic modelling suggests.”
Katie Blacklock, Non-executive director of Edmond de Rothschild, member of M&G’s With Profits Advisory Board and Governor of the Health Foundation, said: “Those in governance positions have a clear fiduciary duty to understand and mitigate the risks posed to clients’ financial assets. To do so, we need to improve our collective climate literacy. Accepting the output of climate-scenario modelling at face value is at best inadequate and at worst dangerous – not just for the price of financial assets but for the planet.”
