Roberta Pierfederici explains why new climate scenarios are a crucial step forward for informing central banks’ and supervisors’ response to climate risk, but recommends improvements for the methodological framework.

On 24 June, the Network of Central Banks and Supervisors for Greening the Financial System (NGFS) published its first set of climate scenarios to explore the risks from both ‘physical’ impacts of climate change and the transition to a zero-carbon economy. The goal of the NGFS is to provide a common framework that will allow central banks, supervisors and financial firms to assess and manage future climate-related risks. This is an important step forward for understanding and managing the exposure of financial systems to climate-related risks. However, the use of scenarios by central banks and by companies requires caution, as they have many limitations that can hamper an accurate assessment of the risks and potentially harm financial decisions and climate risk management practices.

Why climate-related risks need to be integrated into financial monitoring and strategy

There is a growing understanding that climate-related risks should be incorporated into financial institutions’ balance sheets. Financial systems are exposed to a wide range of threats relating to the management of climate change, through a combination of physical and transition risks.

‘Physical’ risks arise from both ‘chronic’ impacts, such as sea level rise and desertification, and the increasing severity and frequency of ‘acute’ impacts, such as storms and floods.

Transition risks are associated with structural changes emerging as the economy becomes low- and zero-carbon. These changes come about with shifts in climate policy and regulation, as well as innovations that cause a decrease in the competitiveness of high-carbon technologies and infrastructures.

Both physical and transition risks can affect key macroeconomic variables and monetary policy decisions by central banks. Physical damages to infrastructure or crops, as well as the emergence of stranded assets due to the low-carbon transition, can impact greatly on GDP, productivity and interest rates.

Both types of risk can also affect the economy and financial system at the micro level, by disrupting businesses, impacting their profitability and asset values and ultimately generating financial risks for lenders, investors and insurers.

Further, the risks arising from a late and abrupt zero-carbon transition could potentially result in destabilising effects on the wider financial system.

Scenarios to assess the climate risk exposure of financial systems

Only a handful of central banks have, so far, embarked on climate scenario analysis to inform monetary policy and assess financial system risks. The different methodologies applied by central banks make use of these scenarios for stress testing exercises. The French, Dutch and Canadian central banks have all conducted quantitative top-down studies, and found a substantial potential risk. In contrast, the Bank of England is planning to incorporate bottom-up participation by financial firms for stress testing the UK financial system to climate risks.

Given the high uncertainty surrounding how policy, technology and socioeconomic factors might evolve in the future, scenario analysis, with its ‘what-if’ approach, seems better suited than standard risk modelling techniques to the exploration of different possible climate change impacts and low-carbon transition pathways.

What form do the new NGFS scenarios take?

The climate scenarios published by NGFS are based on existing research literature and include three representative scenarios exploring different future climate policy developments.

First is the ‘orderly scenario’, which assumes that stringent climate policies introduced from 2020 deliver a ‘smooth’ transition in line with the Paris Agreement’s target of limiting global warming to well below 2°C. Second, a ‘disorderly scenario’ assumes that climate policies are delayed until 2030, leading to a disruptive transition towards the Paris Agreement’s target. And third is a ‘hot house world scenario’, where only currently implemented policies are put in place, leading to warming of 3°C or more and severe physical risks.

In addition, the NGFS has produced five alternative scenarios to explore different assumptions about temperature targets, policy responses and technology development.

Employing multiple climate-economic models, for each scenario the NGFS provides a range of estimates about transition pathways, ‘chronic’ climate impacts and indicative economic impacts.

The set of scenarios will be updated during Phase II of the NGFS project, which will last until the end of 2020, to refine and expand the scope of the analysis. This might include exploring new sets of risks – including ‘acute’ climate impacts, improving the regional coverage and sectoral granularity, and expanding the set of macroeconomic outputs.

Limitations to the NGFS’s approach

Climate-economic scenario analysis is still developing and as a result has some important limitations that, if not recognised, could mislead central banks and financial firms about the likely outcomes of future financial decisions. 

The NGFS climate scenarios represent a crucial step forward for the central banks’ and supervisors’ responses to climate change. However, assessing a financial system’s resilience under different future scenarios is complicated by deep uncertainties around climate sensitivity (typically reported as the change in global mean surface temperature resulting from a doubling in the atmospheric concentration of carbon dioxide) and impacts, socioeconomic pathways and technological progress, as well as by the fundamental limitations of currently available modelling techniques. The NGFS needs to take account of the wide range of uncertainties underlying the scenarios. 

For instance, the estimates it provides are based on central assumptions about physical and transition risks and do not include the lower probability and higher impact consequences of climate change and its mitigation. Estimates of physical risks in a ‘hot house world’ are estimated for 3°C global warming. Given the high uncertainty about climate sensitivity, it is important to consider that temperature increases could be higher with much more severe physical risks.

Preliminary recommendations for the methodological framework

Based on a preliminary assessment of the NGFS scenarios, there are a number of recommendations for the methodological framework.

Consider tail risks: As highlighted in a response by researchers to the Bank of England’s ‘2021 biennial exploratory scenario on the financial risks from climate change’, it is crucial that financial institutions do not to rely only on the central case for physical and transition risks to effectively use climate scenarios in stress testing. A stress test that does not include the low probability, high impact consequences of climate change is not strictly a stress test. And a framework that does not assess resilience to tail risks could be counterproductive by hiding potential exposures and vulnerabilities of the financial system.

Recognise uncertainty in climate sensitivity: The NGFS scenarios framework should recognise climate sensitivity as an important source of uncertainty, and include in the analysis also the consequences of breaching critical thresholds in the climate system (tipping points), such as destabilisation of the land-based polar ice sheets, which would have serious implications for the magnitude and nature of physical risks. This would enable firms to consider the true extent of the physical risks they may face.

Extend the set of physical risks: The proposed scenarios might underestimate physical risks, given that a wide range of impacts – e.g. sea level rise, heatwaves, cyclones, floods – are currently not included in the analysis. Incorporating a wider set of physical impacts – including acute impacts and climate thresholds – would significantly improve the estimates.

Integrate transition and physical risks: In the NGFS scenarios, impacts on GDP from transition and physical risks are modelled separately, meaning that the overall economic impacts resulting from the combination and feedback effects of both types of risk are currently unknown. As the financial system will experience them at the same time, integrating transition and physical risks is therefore of fundamental importance.

Improve the representation of technological change: There are also some important aspects of mitigation options that are not represented in the models used to develop the climate-economic scenarios. For instance, many models fail to represent the innovation dynamics underlying technological change. This has implications for the estimates of the cost and speed of the transition, as events such as technological breakthroughs are hard to model but can reduce economy‐wide transition costs, create new opportunities for investment, and result in significant financial losses in carbon-intensive sectors. By improving the representation of technological change, the NGFS scenarios would be better able to capture key transition risks related to the emergence of disruptive innovations, rapid technological change and falling costs that can swiftly destroy the competitiveness of old high-carbon technologies.

Improve geographical coverage and sectoral granularity: Most modelling approaches used for producing mitigation pathways lack geographical and sectoral granularity. For example, many integrated assessment and energy system models cannot identify the technological transitions associated with low-carbon pathways at the level of individual business and industrial sectors. It is important that the NGFS recognises this shortcoming.

Consider alternative policy developments: The NGFS should consider that the way climate policy will unfold in the future is unclear and that there are some fragmented developments in climate policy already happening across the world that do not fall within the schematic representations described by the three scenarios.

Overall, it is critical to appreciate that there are currently gaps in the literature which prevent the development of a framework that would allow all of these recommendations to be implemented. Current models, for instance, fail to comprehensively capture climate thresholds, as well as the role of feedbacks in physical and transition risks or societal responses to climate change, such as mass migration and conflicts.

Nevertheless, the NGFS scenarios should explicitly acknowledge these gaps and their consequences for climate risks assessment. Limitations in the current NGFS modelling approach should be addressed in Phase II, along with the development of a coherent framework to link climate scenarios with macro/micro-economic modelling and financial tools that will allow central banks and firms to assess their exposure to climate-related risks.

An analytical tool with plenty of potential

The new NGFS scenarios should not be interpreted as the endpoint for climate-risk analysis in the financial system. Rather, they should be continuously revised in light of new scientific findings and the development of better methodologies. By taking into account the wide range of uncertainties in the scenarios’ estimates, it will be possible to significantly improve the reliability of decisions based on the models’ results.

By improving the resilience of the financial system to climate risks, climate scenarios and associated stress tests can accelerate the reallocation of investments towards industries and business models that are more consistent with the zero-carbon transition. The stress tests are also an important instrument for driving systemic changes that can promote both the transition towards a net-zero economy and the alignment of the financial system with the Paris Agreement.

Keep in touch with the Grantham Research Institute at LSE
Sign up to our newsletters and get the latest analysis, research, commentary and details of upcoming events.