The influence of climate change on the economy is already noticeable and will grow substantially in the future. Even if we meet our climate targets and reduce greenhouse gas emissions to net zero, the climate change already set in motion will continue to have effects for decades driving up physical risk (extreme weather, flooding, crop failure), transition risk (stranded assets, policy shocks, shifting energy costs), and financial risk for the institutions exposed to them.
For governments, insurers, (central) banks, and companies, it is essential to be aware of these risks and take them into account. Even more important is understanding the factors that influence these risks, either enforcing or dampening, so that robust and adaptive policies can be pursued. Yet the models currently used to analyse these factors and risks fall short for several reasons, limiting their value in shaping policies that mitigates climate risk. Better use of data and empirical models, complemented by scenario thinking, produces a more realistic view on the risks and their determining factors. A view that can lead to genuinely robust and adaptive climate risk policies.
No “real” price
To study and influence economic development, macro econometric models are used. National planning bureaus, (central) banks, insurers, and asset managers all rely on them when either analysing or designing policies. To meaningfully support effective economic or financial policies, these models need to capture the most relevant and influenceable aspects of our economy. Yet the models in use often omit essential components, producing a distorted picture of the causes and impact of climate risk. Models such as SAFFIER II from the Dutch central planning bureau (CPB) typically include only labour and capital (i.e., investment) as inputs to production, leaving out the consumption of non-renewable resources such as fossil fuels or scarce metals and minerals. The societal impact of greenhouse gas emissions and other waste is likewise absent from these models.
Because these costs and damages fall outside the model, they cannot be analysed within it, the model is blind to them. As a result, the policies built on that model are blind to them too, which has contributed to the climate change we are now experiencing. It also creates an unfair comparison with renewable alternatives, holding back investment in them. A 2024 study by Bilal and Känzig found that once the economic effects of temperature variability are properly modelled, a 1°C rise in global temperature reduces world GDP by around 12% at its peak, roughly six times larger than earlier estimates. Central banks have begun to acknowledge the gap and the climate stress tests run by the European Central Bank and Dutch National Bank explicitly recognize that standard macroeconomic models understate climate-related risk. The tools policymakers rely on are catching up to the problem but still miss the true cost of resource use and emissions.
Fundamentally uncertain
A second, perhaps more important shortcoming is that the models in use assume a stationary economy moving toward equilibrium. This assumption underlies many macroeconomic models, including SAFFIER II and commercial macro econometric models. Yet empirical research shows this assumption does not hold. Because of continuous policy interventions, technological change, crises, and shifting consumer behaviour, the economy does not steadily move toward an equilibrium. It is not stationary but fundamentally uncertain.
As a result, policy built on an assumption of stability will fundamentally miss the mark as became painfully clear during the 2007–2008 financial crisis. More recently, the same lesson played out twice in quick succession: the COVID-19 shock and the inflation surge that followed caught most equilibrium-based macro models off guard. Also, the 2022 European energy crisis, triggered by Russia’s invasion of Ukraine, exposed how poorly these models capture sudden shifts in energy supply, price, and demand. Climate change will cause further economic disruption of the same kind, as consumer behaviour changes, energy production and consumption shift, and the raw materials we use and the ways we produce goods evolve. Dogmatically clinging to an incomplete economic theory will lead to ineffective policies for mitigating climate risk.
A diversified modelling approach leads to more robust policies
Every economic theory is based on certain simplifying assumptions and therefore incomplete, imperfect, and subject to change. The assumption of stability, for instance, grossly underestimates the complexity of the economy. Analyses built on a single theory as their sole starting point will therefore always be of limited value. Using data as empirical evidence allows us to ground our understanding of the economy more firmly. This approach helps surface the factors that matter for understanding, for example, the impact of climate change on the economy. By developing models that are informed by both data and theory, rather than massaging the data until it fits a theoretically “correct” but incomplete box, we arrive at practically useful insights and keep better sight of the changes ahead.
Because of the economy’s fundamental uncertainty, even comprehensive data can only offer insight into the past, while models make simplifications and therefore will always be limited in their ability to predict the future. As a consequence, data and models alone are not enough. However, they belong to the set of instruments that should be used to create a rich and diverse picture of uncertainty, which is the basis of good risk management. Using data and models to generate future scenarios helps us explore some of the ways the future might unfold. System dynamics can complement them by helping us understand how different factors interact over time, including feedback loops, delays, and effects that can reinforce or counteract one another. Scenario thinking can further help us imagine social, technological, ecological, or political shifts that fall outside the data or model forecasts. Together, these approaches allow us to explore not only what might happen, but also how different developments could interact and amplify or dampen one another. This produces a better-informed picture of possible unforeseen situations, allowing us to formulate, implement, evaluate, and adjust more effective climate policies.