Climate risk has emerged as one of the most frightening challenges of our time, affecting the economy, the financial system and society at large. From rare, catastrophic physical events to sudden changes in policy and consumer behavior, the uncertainty inherent in climate risks makes it extremely difficult to model accurately.
This post explores the complexities of modeling climate risks and focuses on both physical and transitional risks resulting from social and political change. Furthermore, it considers the impact on financial risk management and economic resource allocation.
System change and data issues
At the heart of physical climate risk modelling is the challenge of addressing the rapidly changing climate regime. Historically, risk models have relied on extensive datasets that account for past events. However, in climate change, evidence of future risk events is not yet present in historical records.
Furthermore, modeling the “left tail” of the probability distribution: the rare but catastrophic region of losses is challenging without the assumption of a change of government. By definition, extreme events are underestimated in historical data, but they are exactly the possible consequences.
For example, flood protection, urban planning, and agricultural investments could be based on historical climate patterns. However, as climate change changes weather patterns and increases the frequency and severity of extreme events, historical data becomes an unreliable guide to future risks.
Without accurate data from these new regimes, the model could underestimate the possibility and impact of such events, and could expose communities and financial institutions to unexpected shocks.
Butterfly effect
The inherent difficulty in modeling climate risk is further exacerbated by what meteorologist Edward Lorentz famously called the “butterfly effect.” This phenomenon highlights the extreme sensitivity of complex systems, such as the Earth's climate, to initial conditions. One minute error in input data can cause output to vary significantly. For example, small discrepancies in temperature, humidity, or wind speed inputs can lead to completely different climate predictions over decades in the future.
In fact, climate models that predict weather and climate trends for 2030 or 2040 must compete with high levels of uncertainty. The chaotic nature of climate systems means that even state-of-the-art models can lead to unreliable predictions given slightly incomplete data.
This “chaos” propagates into financial risk management, where the output of the climate model serves as an input to the financial model. As a result, uncertainty compounds can make the ultimate prediction of physical risk unworthy.
The complexity of migration risks
Physical risks stem from direct effects like extreme weather, but transition risk refers to the economic and financial impacts of a transition to a low-carbon economy. This includes a variety of factors, including political restrictions on emissions, changing consumer demand, technological changes, and even geopolitical tensions.
Migration risk is often characterized by a high degree of uncertainty driven by so-called “unknown unknowns.” In other words, I don't think these risks should be considered when modeling or making decisions.
For example, consider policies aimed at reducing carbon emissions. Although intentional, these policies can disrupt industries that rely heavily on fossil fuels. Companies in these sectors may be seeing a sudden drop in stock prices, and regions that rely on these industries may experience a recession.
Furthermore, consumer preferences are evolving rapidly, and market forces can accelerate or slow down the pace of transition in unpredictable ways. All these quadratic and cyclonic effects may not be clear on the start date of the policy.
Financial risk management traditionally relies on statistical models that function well under conditions of relative stability. However, when faced with transition risk, these models struggle because the future does not resemble the past. Events that promote transition risk are often unprecedented, and their effects can be both systemic and nonlinear.
In the area of transition risk, the advice of risk management thinkers like Nassim Nicholas Taleb is particularly relevant. Known for his work at the “Black Swan” event, Taleb argues that it is wiser to adopt strategies that explain extreme uncertainty when faced with unknown unknowns.
His approach suggests that instead of trying to accurately predict every possible outcome, risk managers should focus on building resilient systems that can absorb shocks. This includes:
- Diversification: Avoid excess concentrations in a single asset or sector.
- Redundancy: Build extra capacity or safety margins to handle unexpected events.
- Flexibility: Policy and financial product design that can be adapted to changing circumstances.
- Stress Test: Periodically simulate extreme scenarios to assess how the system responds under obsession.
Adopting these strategies can help mitigate the impact of transition risk, even when the underlying driver is difficult to predict.
The relevance of this approach has been highlighted in recent California wildfires. Given the increased temperature, drought conditions and rain patterns, the general trend towards more wildfires may have been predictable from a statistical perspective, but the timing, location and severity of the event were not.
As a risk manager, what you want to predict, not just wildfires, but what you want to predict is the severity of the event. So, financial institutions need to incorporate climate risk into their risk management frameworks, but complex uncertainties pose important challenges, leading to false pricing of risk and misallocation of capital.
What's next?
Data shortages and forecasting issues can be resolved to some extent. One promising tool for improving climate risk modeling is the integration of interdisciplinary insights. Advances in data science, machine learning, and complexity theories provide tools that can enhance the predictive capabilities of traditional climate and financial models.
For example, ensemble modeling, in which multiple models run in parallel to provide various results, can help capture uncertainties inherent to individual models.
Furthermore, incorporating real-time data from sensors, satellites, and IoT devices can provide more detailed input, potentially reducing some of the errors that lead to different outcomes in climate modeling. However, these technological advances must be strongly aware of the limitations and integrated.
As the model becomes more complex, so is the possibility of cascade errors if initial conditions are not captured accurately.
Policymakers and regulators are also addressing the impact of climate risks for financial stability. There is a growing consensus that stress testing and scenario analysis should incorporate climate-related risks, as well as traditional financial risks.
For example, the European Central Bank (ECB) and the US Federal Reserve have launched studies to assess the financial system's resilience to the climate shock.
These regulatory initiatives highlight the importance of a holistic approach to risk management that integrates climate science, financial modeling and policy analysis. As climate risks become central to global economic stability, cooperation between these sectors is essential to prevent both physical and transitional risks.
Important points
Modeling climate risk is one of the most challenging efforts in risk management today. The difficulty in predicting physical risks is attributed to the lack of accurate data in the world undergoing rapid regime change and the unpredictable nature of the butterfly effect. Transition risk combines these challenges by introducing layers of sociopolitical and economic uncertainty, with many unknowns.
As financial institutions and policymakers seek to mitigate these risks, integrating interdisciplinary insights and embracing new technologies offers hope for improving the predictive power of the model, but a careful and robust approach to risk management is a top priority.
